EPA
625/6-86-014
United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park, NC 27711
Technology Transfer
EPA/625/6-86/014'
v>EPA Handbook
Control Technologies for
Hazardous Air Pollutants
cNVIRONMENTAI
PROTECTION
AGENCY
DALLAS, TEXAp
UBUfif
-------
EPA/625/6-86/014
September 1986
Handbook
Control Technologies for
Hazardous Air Pollutants
Air and Energy Engineering Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Center for Environmental Research Information
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
Notice
This document has been reviewed in accordance with the U.S. Environmen-
tal Protection Agency's peer and administrative review policies and ap-
proved for publication. Mention of trade names or commercial products does
not constitute endorsement or recommendation for use.
-------
Foreword
Today's rapidly changing industrial technologies, products, and practices
frequently carry with them an increasing generation of hazardous air pollu-
tants (HAPs). When energy and material resources are extracted, processed,
converted, and used, the consequent impacts on health and the environment
require that efficient pollution control methods be used.
/
The 1970 amendments to the Clean Air Act require the U.S. Environmental
Protection Agency (EPA) to set National Ambient Air Quality Standards for
criteria air pollutants found throughout the country. The Clean Air Act
Amendments also require EPA to review and regulate hazardous air pollu-
tants, defined as those air pollutants that can contribute to increased mortal-
ity or serious illness but which are not already regulated as criteria pollutants.
Since the definition of a HAP is very broad and encompasses thousands of
specific compounds, it is not practical to develop an all-inclusive list of HAP
compounds and compound-specific control techniques. However, the num-
ber 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.
The purpose of this handbook is to help EPA regional. State, and local air
pollution control agency technical personnel select, evaluate, and cost air
pollution control techniques for reducing or eliminating the emission of
potentially hazardous air pollutants from industrial/commercial sources. The
information provided by this document will be useful for reviewing permit
applications or for informing interested parties of the type, basic design, and
cost of available HAP control systems.
-------
Acknowledgments
This handbook has been adapted from a two-volume EPA report titled Evalu-
ation of Control Technologies for Hazardous Air Pollutants. The report was
written by Robert Y. Purcell, Pacific Environmental Services Inc., Durham,
North Carolina, and Gunseli Sagun Shareef, Radian Corporation, Research
Triangle Park, North Carolina. They were assisted by: Vishnu S. Katari, Karin
C. C. Gschwandtner, Michael K. Sink, and Charlotte R. Clark of Pacific Envi-
ronmental Services Inc.; and Andrew J. Miles, D. Blake Bath, and Glynda E.
Wilkins of the Radian Corporation. Dr. Bruce A. Tichenor of EPA's Air and
Energy Engineering Research Laboratory, Research Triangle Park, North
Carolina, was the EPA project officer for the report.
This handbook was adapted and produced by JACA Corporation, Fort Wash-
ington, Pennsylvania. Norman Kulujian of EPA's Center for Environmental
Research Information, Cincinnati, Ohio, was the EPA project officer for this
publication.
If you wish to obtain the more detailed two volume report, is is available from
the National Technical Information Service (NTIS), 5285 Port Royal Road,
Springfield, Virginia 22161; (703) 487-4650. The order numbers are:
Volume I. Technical Report; PB 86-167 020; EPA/600/7-86/009a
Volume II. Appendices; PB 86-1677 038; EPA/600/7-86/009b
IV
-------
Contents
Nomenclature x
1 Introduction 1
1.1 Objective 1
1.2 How to Use the Handbook 2
1.3 References 2
2 HAP Emissions and Their Key Physical Properties 5
2.1 Background 5
2.2 Identification of Potential HAP's and Emission Sources • 5
2.3 Identification of Key Emission Stream Properties 18
2.4 References 19
3 Control Device Selection 23
3.1 Background 23
3.2 Vapor Emissions Control 23
3.3 Particulate Emissions Control 33
3.4 References 45
4 HAP Control Techniques 47
4.1 Background 47
4.2 Thermal Incineration 47
4.3 Catalytic Incineration 53
4.4 Flares 60
4.5 Boilers/Process Heaters 63
4.6 Carbon Adsorption 63
4.7 Absorption 69
4.8 Condensation 75
4.9 Fabric Filters 80
4.10 Electrostatic Precipitators 87
4.11 Venturi Scrubbers 90
4.12 References 94
5 Cost Estimation Procedure 97
5.1 Objective 97
5.2 Total Capital Cost 97
5.3 Annualized Operating Costs 107
5.4 References 111
Appendices
A.1 Potential HAP's for Solvent Usage Operations 113
B.1 Gas Stream Parameters Calculations 115
B.2 Dilution Air Requirements Calculations 119
C.1 HAP Emission Stream Data Form 121
C.2 Calculation Sheet for Dilution Air Requirements 123
C.3 Calculation Sheet for Thermal Incineration 124
C.4 Calculation Sheet for Catalytic Incineration 130
C.5 Calculation Sheet for Flares 137
C.6 Calculation Sheet for Carbon Adsorption 141
C.7 Calculation Sheet for Absorption 147
C.8 Calculation Sheet for Condensation 154
C.9 Calculation Sheet for Fabric Filters 159
C.10 Calculation Sheet for Electrostatic Precipitators 162
C.11 Calculation Sheet for Venturi Scrubbers 164
C.12 Capital and Annualized Cost Calculation Worksheet 167
-------
Tables
2-1 Potential HAP's and Emission Sources for Solvent Usage
Operations 7
2-2 Potential HAP's and Emission Sources for Metallurgical
Industries 9
2-3 Emission Sources for the SOCMI 10
2-4 Potential HAP's and Emission Sources for the Inorganic
Chemical Manufacturing Industry 12
2-5 Potential HAP's and Emission Sources for the Chemical
Products Industry 14
2-6 Potential HAP's and Emission Sources for the Mineral
Products Industry 15
2-7 Potential HAP's and Emission Sources for the Wood
Products Industry 16
2-8 Potential HAP's for Petroleum Related Industries 16
2-9 Potential HAP's for Petroleum Refining Industries 17
2-10 Emission Sources for Petroleum Related Industries 17
2-11 Potential HAP's and Emission Sources for Combustion
Sources 18
2-12 Key Properties for Organic Vapor Emissions 19
2-13 Key Properties for Inorganic Vapor Emissions 19
2-14 Key Properties for Particulate Emissions 19
3-1 Key Emission Stream and HAP Characteristics for Selecting
Control Techniques for Organic Vapors from Point Sources 24
3-2 Current Control Methods for Various Inorganic Vapors 29
3-3 Range of Capture Velocities 30
3-4 Summary of Control Effectiveness for Controlling
Organic Area Fugitive Emission Sources 31
3-5 Key Characteristics for Particulate Emission Streams 35
3-6 Advantages and Disadvantages of Particulate Control
Devices 40
3-7 Control Technology Applications for Transfer
and Conveying Sources 42
3-8 Control Technology Applications for Loading
and Unloading Operations 43
3-9 Control Technology Applications for Plant Roads 43
3-10 Control Technology Applications for Open Storage Piles 44
3-11 Control Technology Applications for Waste Disposal Sites 45
4-1 Thermal Incinerator System Design Variables 49
4-2 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for Thermal Incineration 53
4-3 Catalytic Incinerator System Design Variables 55
4-4 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for Catalytic Incineration 59
4-5 Flare Gas Exit Velocities 62
4-6 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for Flares 63
4-7 Carbon Adsorber System Design Variables 65
4-8 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for Carbon Adsorption 69
4-9 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for Absorption 75
4-10 Coolant Selection 77
vi
-------
Tables (continued)
4-11 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for Condensation 79
4-12 Characteristics of Several Fibers Used in Fabric Filtration 82
4-13 Comparisons of Fabric Filter Bag Cleaning Methods 83
4-14 Recommended Air-to-Cloth Ratios for Various Dusts
and Fumes by Cleaning Method 85
4-15 Factors to Obtain Gross Cloth Area from Net Cloth Area 86
4-16 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for Fabric Filters 86
4-17 Typical Values for Drift Velocity for Various Particulate
Matter Applications 88
4-18 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for ESP's 89
4-19 Pressure Drops for Typical Venturi Scrubber Applications 92
4-20 Construction Materials for Typical Venturi Scrubber
Applications 93
4-21 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for Venturi Scrubbers 93
5-1 Identification of Design Parameters and Cost Curves
for Major Equipment 97
5-2 Chemical Engineering Fabricated Equipment Cost Indices 98
5-3 Unit Costs for Various Materials 99
5-4 Price of Packing for Absorber Systems 100
5-5 Bag Prices 101
5-6 Identification of Design Parameters and Cost Curves
for Auxiliary Equipment 103
5-7 Assumed Pressure Drops Across Various Components 105
5-8 Capital Cost Elements and Factors 107
5-9 Unit Costs to Calculate Annualized Cost 108
5-10 Utility/Replacement Operating Costs for HAP Control
Techniques 109
5-11 Additional Utility Requirements 110
5-12 Estimated Labor Hours Per Shift and Average Equipment
Life 110
-------
Figures
1-1 Steps Used When Responding to Inquiries 3
1-2 Steps Used When Reviewing Permits 4
2-1 A Partially Completed HAP Emission Stream Data Form
for One of Six HAP Emission Streams (#1) Generated at
a Fictitious Company 6
2-2 Potential Emission Points for a Vacuum Distillation Column
Using Steam Jet Ejectors with Barometric Condenser 10
3-1 Percent Reduction Ranges for Add-on Control Devices 24
3-2 Effluent Characteristics for Emission Stream #1 33
3-3 Effluent Characteristics for Emission Stream #2 34
3-4 Effluent Characteristics for Emission Stream #3 35
3-5 Effluent Characteristics for Emission Stream #4 36
3-6 Effluent Characteristics for Emission Stream #5 37
3-7 Effluent Characteristics for Emission Stream #6 39
3-8 Effluent Characteristics for a Municipal Incinerator Emission
Stream 41
4-1 Schematic Diagram of a Thermal Incinerator System 48
4-2 Supplementary Heat Requirement vs. Emission Stream Heat
Content (Dilute Stream/No Combustion Air) 50
4-3 Supplementary Heat Requirement vs. Emission Stream
Heat Content (No Oxygen in Emission Stream/Maximum
Combustion Air) 50
4-4 Heat Exchanger Size vs. Emission Stream Flow Rate
(Dilute Stream/No Combustion Air) 52
4-5 Heat Exchanger Size vs. Emission Stream Heat Content (No
Oxygen in Emission Stream/Maximum Combustion Air) 52
4-6 Schematic Diagram of a Catalytic Incinerator System 53
4-7 Supplementary Heat Requirement vs. Emission Stream Heat
Content (Dilute Stream/No Combustion Air) 56
4-8 Supplementary Heat Requirement vs. Emission Stream
Heat Content (No Oxygen in Emission Stream/Maximum
Combustion Air) 57
4-9 Heat Exchanger Size vs. Emission Stream Heat Content 59
4-10 A Typical Steam-Assisted Flare System 60
4-11 Adsorption Isotherms for Toluene/Activated Carbon System 63
4-12 A Typical Fixed-Bed Carbon Adsorption System 64
4-13 Carbon Requirement vs. HAP Inlet Concentration 67
4-14 Steam Requirement vs. Carbon Requirement 68
4-15 A Typical Countercurrent Packed Column Absorber System 70
4-16 Correlation for Flooding Rate in Randomly Packed Towers 72
4-17 NOG for Absorption Columns with Constant Absorption
Factor AF 73
4-18 Flow Diagram for a Typical Condensation System with
Refrigeration 75
4-19 Vapor Pressure-Temperature Relationship 76
4-20 Venturi Scrubber Collection Efficiencies 91
4-21 Psychrometric Chart, Temperature Range 0-500°F,
29.92 in Hg Pressure 94
5-1 Prices for Thermal Incinerators, Including Fan and Motor, and
Instrumentation and Control Costs 98
-------
Figures (continued)
5-3 Prices for Catalytic Incinerators, Less Catalyst 99
5-4 Prices for Carbon Adsorber Packages 99
5-5 Prices for Custom Carbon Adsorbers, Less Carbon 100
5-6 Prices for Absorber Columns 100
5-7 Prices for Adsorber Platforms and Ladders 100
5-8 Total Capital Costs for Cold Water Condenser Systems 101
5-9 Additional Capital Cost for Refrigerant Condenser Systems 101
5-10 Prices for Negative Pressure, Insulated Fabric Filter Systems,
Less Bags 101
5-11 Prices for Insulated Electrostatic Precipitators 102
5-12 Prices for Venturi Scrubbers 102
5-13 Required Steel Thicknesses for Venturi Scrubbers 102
5-14 Price Adjustment Factors for Venturi Scrubbers 102
5-15 Carbon Steel Straight Duct Fabrication Price
at Various Thicknesses 103
5-16 Stainless Steel Straight Duct Fabrication Price
at Various Thicknesses 103
5-17 Fan Prices 104
5-18 Carbon Steel Stack Fabrication Price for W Plate 105
5-19 Carbon Steel Stack Fabrication Price for 5/16" and %" Plate 105
IX
-------
Nomenclature*
a = packing constant
A = heat exchanger surface area, ft2
Abed = carbon bed cross sectional area, ft2
Acoiumn = absorber column cross sectional area, ft2
ACOn = condenser surface area, ft2
Anc = net cloth area, ft2
Ap = collection plate area, ft2
A, = venturi scrubber throat area, ft2
Atc = total cloth area, ft2
ABS = abscissa (Figure 4.7-2)
AC = adsorption capacity of carbon bed, Ib HAP/100 Ib carbon
A/C = air to cloth ratio for baghouse, acfm/ft2
AF = absorption factor
b = packing constant
c = packing constant
C = annual credits, $/yr
Creq = amount of carbon required, Ib
Cpair = average specific heat of air, Btu/scf-°F
Cpair = average specific heat of air, Btu/lb-mole-°F
= average specific heat of combined gas stream, Btu/scf-°F
. = average specific heat of coolant, Btu/lb-°F
Cpe = average specific heat of emission stream, Btu/scf-°F
Cpe = average specific heat of emission stream, Btu/ lb-°F
Cpf = average specific heat of supplementary fuel (natural gas), Btu/lb-°F
Cptg = average specific heat of flue gas, Btu/scf-°F
Cpfig = average specific heat of flare gas, Btu/lb-°F
Cpw = average specific heat of water, Btu/lb-°F
CPHAP = average specific heat of HAP, Btu/lb-mole-°F
CE = collection efficiency (based on mass), %
CRF = capital recovery factor
CRFW = weighted average capital recovery factor
d = packing constant
D = annual direct labor costs, $/yr
Dbed = carbon bed diameter, ft
Dcoiumn = absorber column diameter, ft
Dduct = duct diameter, in
Dp = mean particle diameter, /urn
D, = venturi scrubber throat diameter, ft
Dtip = flare tip diameter, in
DG = diffusivity in gas stream, ft2/hr
aEnglish units are used throughout this report. Many engineering handbooks provide
conversion factors for English to metric units.
-------
DL
Di
D2
DE
L't reported
DP
ex
f
FE
PER
9
9c
G
^area
Garea.f
Gmol
hd
he
hf
hflg
AH
'I con
Hf
H|0ad
rlnoncon
"uncon
HG
HL
HOG
Htcolumn
Httotal
HAPcon
HAPe
HAPe,m
HAP0
HAP0.m
HP
HR
HRS
L
L"
Lgal
U/Qe,a
LEL
m
M
Me
diffusivity in liquid, ft2/hr
annual operating labor cost, $/yr
annual supervision labor cost, $/yr
destruction efficiency, %
reported destruction efficiency, %
stream dew point, °F
excess air, % (volume)
fraction
fabricated equipment cost index
fan electricity requirement, kWh
packing constant
gravitational constant, = 32.2 ft/sec2
gas (emission stream) flow rate, Ib/hr
gas (emission stream) flow rate based on column cross sectional area, Ib/sec-ft2
gas (emission stream) flow rate at flooding conditions based on column cross sectional
area, Ib/sec-ft2
gas (emission stream) flow rate, Ib-mole/hr
heat content of emission stream after dilution, Btu/scf
heat content of emission stream, Btu/scf
lower heating value of supplementary fuel (natural gas), Btu/scf
flare gas heat content, Btu/scf
heat of vaporization of HAP, Btu/lb-mole
enthalpy change associated with condensed HAP, Btu/min
supplementary heat requirement (heat supplied by the supplementary fuel), Btu/min
condenser heat load, Btu/hr
enthalpy change associated with noncondensible vapors, Btu/min
enthalpy change associated with uncondensed HAP, Btu/min
height of a gas transfer unit, ft
height of a liquid transfer unit, ft
height of a gas transfer unit (based on overall gas film coefficients), ft
absorber column packed height, ft
absorber column total height, ft
quantity of HAP condensed, Ib-mole/min
inlet HAP concentration, ppmv
quantity of HAP in the emission stream entering the condenser, Ib-mole/min
outlet HAP concentration, ppmv
quantity of HAP in the emission stream exiting the condenser, Ib-mole/min
fan power requirement, hp (horsepower)
heat recovery in the heat exchanger, %
number of hours of operation per year
solvent flow rate, Ib/hr
solvent flow rate based on absorber column cross sectional area, Ib/hr-ft2
solvent flow rate, gal/min
solvent flow rate, Ib-mole/hr
liquid flow rate in venturi scrubber, gal/min
liquid to gas ratio, gal/103 acf
lower explosive limit, % (volume)
slope of the equilibrium curve
annual maintenance costs, $/yr
moisture content of emission stream, %(volume)
XI
-------
M2
M3
MWavg
MWe
MWflg
MWso|Vent
MWHAP
N
NOG
02
ORD
AP
APa
Pe
Ppartial
"vapor
AP.otal
APV
PC
Qa
Qc
'-'co
olant
Qe,a
Qe,s
Of
Qfg
Qfg,a
Qfig
Qflg,a
Qrec
Qs
Qw
r
R
""•hum
Ref
RE
S
S
ScG
ScL
St
SV
annual maintenance labor cost, $/yr
annual maintenance supervision cost, $/yr
annual maintenance materials cost, $/yr
average molecular weight of a mixture of components, Ib/lb-mole
average molecular weight of emission stream, Ib/lb-mole
average molecular weight of flare gas, Ib/lb-mole
molecular weight of solvent, Ib/lb-mole
molecular weight of HAP (average molecular weight if a mixture of HAPs is present),
Ib/lb-mole
number of carbon beds
number of gas transfer units (based on overall gas film coefficients )
oxygen content of emission stream, % (volume)
ordinate (Figure 4.7-2)
total pressure drop for the control system, in H20
absorber column pressure drop, Ib/ft2-ft
emission stream pressure, mm Hg
partial pressure of HAP in emission stream, mm Hg
vapor pressure of HAP in emission stream, mm Hg
absorber column total pressure drop, in H20
pressure drop across venturi, in H20
purchased equipment cost, $
flow rate of gas stream at actual conditions, acfm
combustion air flow rate, scfm
flow rate of combined gas stream entering the catalyst bed, scfm
coolant flow rate, Ib/hr
cooling water flow rate, Ib/min
emission stream flow rate at actual conditions, acfm
saturated emission stream flow rate, acfm
supplementary fuel ( natural gas ) flow rate, scfm
flue gas flow rate, scfm
flue gas flow rate at actual conditions, acfm
flare gas flow rate, scfm
flare gas flow rate at actual conditions, acfm
quantity of HAP recovered, Ib/hr
steam flow rate, Ib/min
cooling water flow rate, gal/min
packing constant
gas constant, = 0.73 ft3-atm/lb-mole °R;
relative humidity, %
refrigeration capacity, tons
removal efficiency, %
reported removal efficiency, %
packing constant
annual cost of operating supplies, $/yr
Schmidt number for HAP/emission stream
Schmidt number for HAP/solvent system
steam ratio, Ib steam/lb carbon
space velocity, hr"1
cleaning interval, min
= 1.987cal/g-mole°K
-------
tr
T
To
Tci
1 con
' cool.i
'cool.o
T6
'fig
The
Tr
Tsti
Tsto
1 wo
ATLM
' ncolumn
U
Uduct
ue
••e.s
fig
U
'-'max
Ut
vc
"carbon
Vbed
"packing
w
*Vtco|urrln
x
X
y
Y
Y
Zbed
e
X
•n
Pbed
PC
PG
PL
9ads
Greg
"dry-cool
M-L
residence time, sec
temperature, °F
combustion temperature, °F
temperature of combined gas stream entering the catalyst bed, °F
temperature of flue gas leaving the catalyst bed, °F
condensation temperature, °F
inlet temperature of coolant, °F
outlet temperature of coolant, °F
emission stream temperature, °F
temperature of saturated emission stream, °F
flare gas temperature, °F
emission stream temperature after heat exchanger, °F
reference temperature, = 70 °F
inlet steam temperature, °F
condensed steam outlet temperature, °F
inlet cooling water temperature, °F
outlet cooling water temperature, °F
logarithmic mean temperature difference, °F
absorber column thickness, ft
overall heat transfer coefficient, Btu/hr-ft2- °F
drift velocity of particles, ft/sec
velocity of gas stream in the duct, ft/min
emission stream velocity through carbon bed, ft/min
throat velocity of saturated emission stream, ft/sec
flare gas exit velocity, ft/sec
maximum flare gas velocity, ft/sec
annual utility costs, $/yr
combustion chamber volume, ft3
volume of carbon bed, ft3
catalyst bed requirement, ft3
absorber column packing volume, ft3
particle grain loading, gr/acf
absorber column weight, Ib
mole fraction of solute in solvent, moles solute/(moles solute + moles solvent)
mole fraction of gaseous component in liquid, moles solute/ moles solvent
mole fraction of solute in air, moles solute/(moles solute + moles air)
packing constant
mole fraction of solute in air, moles solute/moles air
carbon bed depth, ft
packing constant
latent heat of vaporization for steam, Btu/lb
fan efficiency, percent
density of carbon bed, Ib/ft3
density of carbon steel plate, Ib/ft3
density of gas (emission stream), Ib/ft3
density of solvent, Ib/ft3
cycle time for adsorption, hr
cycle time for regeneration, hr
cycle time for drying and cooling the bed, hr
viscosity of solvent, centipoise
viscosity of solvent, Ib/ft-hr
-------
Chapter 1
Introduction
1.1 Objective
The objective of this handbook is to present a
methodology for determining the performance and
cost of air pollution control techniques for reducing
or eliminating the emission of potentially hazard-
ous air pollutants (HAP's) from industrial/commer-
cial sources. (Note: The term "hazardous" in this
document is very broad. It is not limited to the
specific compounds listed under current regula-
tions [i.e., the Clean Air Act, the Resource Conser-
vation and Recovery Act, and the Toxic Substances
Control Act].) This handbook is to be used by EPA
regional, State, and local air pollution control agen-
cy technical personnel for two basic purposes: (1)
to respond to inquiries from interested parties (e.g.,
prospective permit applicants) regarding the HAP
control requirements that would be needed at a
specified process or facility, and (2) to evaluate/
review permit applications for sources with the po-
tential to emit HAP's. It should be noted that this
document provides general technical guidance on
controls and does not provide guidance for compli-
ance with specific regulatory requirements for haz-
ardous air pollutants. Specifically, it does not speci-
fy design requirements necessary to achieve
compliance with standards established under spe-
cific programs such as Section 112 of the Clean Air
Act or standards established under the Resource
Conservation and Recovery Act. Such require-
ments vary with the hazardous air pollutant emit-
ted and with the emission source; thus, regulatory-
specific detailed specifications are beyond the
scope of this handbook.
Section 1.2 discusses the use of this handbook.
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
presents simple step-by-step procedures to deter-
mine basic design parameters of the specific con-
trol devices and auxiliary equipment. Chapter 5
provides the necessary data and procedures to de-
termine order-of-magnitude estimates (-60 to
+ 30 percent) for the capital and annualized costs of
each control system. Appendices A and B present
supplementary data and calculation procedures.
Appendix C contains blank worksheets to be used
while performing the functions described in this
handbook. These worksheets are masters from
which to make copies.
Additional appendices can be found in Evaluation
of Control Technologies for Hazardous Air Pollu-
tants—Appendices (EPA 600/7-86-009b; NTIS order
no. PB 86-167/038/AS; $22.95, price subject to
change). The additional appendices further clarify
and expand the text and give derivations of equa-
tions, calculation procedures, and unit conversion
techniques. They are referenced in this handbook.
The volume can be ordered from the National
Technical Information Service (NTIS), 5285 Port
Royal Road, Springfield, Virginia 22161; (703)
487-4650.
A good source of current information pertaining to
HAP's is the "Air Toxics Information Clearing-
house," 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 Pollution Control Offi-
cials (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: regulatory program de-
scriptions, 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 agen-
cies, and references for ongoing EPA air toxic pro-
jects. It also publishes a quarterly newsletter with
articles on current air toxics concerns. Finally, the
Clearinghouse periodically publishes various spe-
cial reports on topics of interest to users. For fur-
-------
ther information regarding the "Air Toxics Informa-
tion Clearinghouse," contact the appropriate EPA
regional office air toxics contact, or EPA/OAQPS,
Pollutant Assessment Branch, MD-12, Research Tri-
angle Park, North Carolina 27711; (919) 541-5645 or
FTS 629-5645.
1.2 How to Use the Handbook
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 per-
mits. As shown by these figures, these two func-
tions are basically the same; the only substantive
difference is that the review process also compares
the determined/calculated parameters with the cor-
responding parameters stated in the permit appli-
cation to ensure that the control system(s) pro-
posed 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 cate-
gory in question (Section 2.2). The HAP's are cate-
gorized under four headings: organic vapor, organ-
ic particulate, inorganic vapor, and inorganic
particulate. (Note: 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-inclu-
sive nor a declaration that the compounds present-
ed are hazardous.) Next, identify the potential
emission sources for each HAP group (Section 2.2).
The HAP emission sources are listed under one of
three classifications: process point sources, pro-
cess fugitive sources, and area fugitive sources.
(Note: See Section 2.2 for classification defini-
tions.) After each emission source is determined,
identify the key HAP emission stream characteris-
tics (e.g., HAP concentration, temperature, flow
rate, heat content, particle size) needed to select
the appropriate control technique(s) (Section 3.2).
Obtain the actual values for these characteristics
from the owner/operator or from available litera-
ture if the owner/operator cannot provide the nec-
essary data. If two or more emission streams are
combined prior to entry into an air pollution control
system, determine the characteristics of the com-
bined emission stream ( Appendix B.1).
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 im-
posed, and (4) a specific work practice or "other"
related practice may be required. The regulation
format will define the steps that lead to the selec-
tion of the appropriate control technique(s). The
"control device" and "other" formats specify the
appropriate control technique(s). A "numerical lim-
it" format requires the determination of the HAP
removal efficiency before the appropriate control
technique(s) can be identified. Lastly, the "technol-
ogy 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 con-
junction with the limitations imposed by the appli-
cable regulations, are used to select the appropri-
ate 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, parti-
cle size, flow rate). Basic design parameters are
then determined to provide general design condi-
tions that should be met or exceeded for each se-
lected control technique to achieve the specified
HAP removal efficiency (Chapter 4). This exercise
also identifies which of the selected control tech-
niques 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 inte-
gral part of the HAP control system selection pro-
cess. After completing the above process, a HAP
control program can be recommended or evaluat-
ed. (For an example of a State HAP program, see
Appendix A.1, reference 1.)
Example Case
To guide the user through the steps and calcula-
tions described in this handbook, examples are
provided throughout the text. As shown here,
each example is boxed. The primary example
case pertains to a hypothetical plant owner re-
questing assistance in determining the type of
control system that should be used on an emis-
sion stream generated by a paper coating drying
oven. This example is carried through the entire
handbook. Additional example emission
streams are introduced in Chapter 3 to illustrate
fully the control technique selection process and
to clarify the design procedures of Chapter 4.
1.3 References
1. U.S. EPA. Evaluation of Control Technologies
for Hazardous Air Pollutants — Appendices.
EPA-600/7-86-009b (NTIS PB 86-167/038/AS). Oc-
tober 1985.
-------
Figure 1-1. Steps used when responding to inquiries.
Manual Location to
Perform Step
1Section 2.2
2Section 2.3
3None
""Appendix B.1
5None
6Chapter 3
7Chapter 4
8Chapter 5
Information Requested
on HAP Control for
a Specific Facility
Obtain Available
Plant-Specific Data
I
Define HAP's1
_L
Define Emission Sources
Generating HAP's1
!
Define Characteristics
Needed for Each HAP
Emission Stream2
±
Combine HAP Streams3
M
Inquirer Assistance
Define Characteristics
Of Combined Streams4
±
(Each Single/Combined Y^_
HAP Emission Stream JT~
~ I
Define HAP Control
Requirements6
Control
Device
Numerical
Limit
Select Appropriate
Control Techniques(s)6
±
Determine Required
Control Efficiency5
Are
Basic Design Parameters
Requested /
Determine Basic Design
Parameters for
Control System(s)7
Technology
Forcing
Cost of Control a
Decision Factor
Other
Requirement
Regulation Agency
Policy Decision
Agency Determination
of Cost Constraints5
Select Appropriate
Control Technique(s)6
Select Appropriate
Control Technique(s)6
±
Determine Basic Design
Parameters and Cost for
Control System(s)7'8
t
Select Control System
Having Most Stringent
Level of Control Within
Given Cost Constraints
A.
Recommend Appropriate
Control Technique(s)
_L
J_
Determine Basic Design
Parameters for
Control System(s)7
±.
Select Control System
Having Most Stringent
Level of Control
Last
HAP Emission
Stream
No
\
Yes
Recommend HAP
Control Program
-------
Figure 1-2. Steps used when reviewing permits.
Manual Location to
Perform Step
1Section 2.2
2Section 2.3
3None
"Appendix B.1
6None
6Chapter3
'Chapter 4
8Chapter 5
Permit Application
for Review/Approval
/ Are All HAP's \
\ Addressed1 ? /
Yes
/ Are All HAP Emission \
\ Sources Addressed1 ? /
Yes
i
i
No
No
Obtain Additional Data
from Applicant
Obtain Additional Data
from Applicant
Are All HAP Emission
Stream Characteristics
Provided2 ?
No
Obtain Additional Data
from Applicant
Yes
Are Any HAP
Emission Streams
Combined3 ?
LYes
Are Combined Stream
Characteristics
Correct4 ?
No
Yes
Each Single/Combined
HAP Emission Stream
1
Define HAP Control
Requirements5
Obtain Additional Data
from Applicant
Control
Device ,
Select Appropriate
Control Techniques(s)6
Numerical
Limit ,
Determine Required
Control Efficiency5
Other
Requirement
Regulatory Agency
Policy Decision5
Technology
Forcing
Yes/ Is Cost of Control a \
r\ Decision Factor ? /
Determine Basic Design
Parameters for
Control System(s)7
/ Is
Permit Control System
Appropriate
Recommend Appropriate
Control Technique(s)
Yes
Permit System Design
Appropriate
Recommend Appropriate
Basic Design Parameters
Yes
Agency Determination
of Cost Constraints5
1
No
Select Appropriate
Control Technique(s)6
Select Appropriate
Control Technique(s)6
Determine Basic Design
Parameters and Cost for
Control System(s)7'8
Determine Basic Design
Parameters for
Control System(s)7
Select Control System
Having Most Stringent
Level of Control Within
Given Cost Constraints
J_
Select Control System
Having Most Stringent
Level of Control
Last HAP Emission
Stream ?
No
1
Yes
Permit Approval or
Provide Recommendations
-------
Chapter 2
HAP Emissions and Their Key Physical Properties
2.1 Background
This chapter's primary goal is to identify the follow-
ing: (1) potential HAP'sfor a given source category
and the specific sources that may emit the potential
HAP's—Section 2.2, and (2) key emission stream
physical properties needed to select appropriate
control strategies and size control devices for the
HAP emission sources Section 2.3. Specific source
categories are divided into nine general classifica-
tions in this manual. (Note: The general classifica-
tion 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 Clearing-
house—A Compilation of Control Technology De-
terminations, April 1983.) Every possible source
category cannot be listed; however, similarities ex-
ist 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
handbook. (For a listing of chemical hazard infor-
mation profiles [CHIP's] and CAS numbers, see ap-
pendix A.2, reference 1.)
Individual source categories have been classified
based on the manufacturing process associated
with emissions of potential HAP's. The Solvent Us-
age Operations classification includes processes
dependent on solvents, such as surface coating
and dry cleaning operations. Metallurgical Indus-
tries include processes associated with the manu-
facture of metals, such as primary aluminum pro-
duction. Processes and operations associated with
the manufacture of organic and inorganic chemi-
cals have been grouped into the Synthetic Organic
and Inorganic Chemical Manufacturing classifica-
tions, respectively. Industries using chemicals in
the formulation of products are classified as
Chemical Products Industries. The Mineral and
Wood Products Industries classifications include
operations such as asphalt batch plants and kraft
pulp mills, respectively. The Petroleum Related In-
dustries classification is defined as oil and gas pro-
duction, petroleum refining, and basic petrochemi-
cals production. Combustion Sources are utility,
industrial, and residential combustion sources us-
ing coal, oil, gas, wood, or waste-derived fuels.
To assist the user in recording the pertinent infor-
mation, a worksheet has been provided. A copy of
this worksheet, the HAP Emission Stream Data
Form, is presented in Appendix C.1. An example of
a partially completed worksheet is shown in Figure
2-1. This worksheet is designed to record informa-
tion pertaining to one emission stream, be it a sin-
gle stream or a combined stream consisting of sev-
eral single streams.
Example Case
Information has been requested by a paper coat-
ing plant owner regarding the control of an
emission stream from his facility's drying oper-
ations. The most likely generic classification to
include a paper coating plant (one of many sur-
face coating industries) would be Solvent Usage
Operations. To determine if a paper coating fa-
cility is listed within this category, see Section
2.2.1 and Table 2-1, which indicates that the ini-
tial choice was correct (i.e., paper coating is list-
ed under SC-Paper, Tapes, Labels) and the infor-
mation retrieval process can begin.
2.2 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 in-
cludes the names of specific compounds, the clas-
sification of the compounds (i.e., organic or inor-
ganic), and the form in which these compounds
would be emitted (i.e., vapor or particulate). Owing
to process variations, actual emissions from specif-
ic facilities may differ from the general information
presented. Complete identification of HAP emis-
sions is best accomplished with assistance from
the owner/operator of the facility. (See Appendix
A.3, reference 1, for a listing of trade names and
common synonyms for HAP's.)
This section also presents information pertaining
to the sources (e.g., processes) within each specific
source category that have the potential to emit
-------
Figure 2-1. A partially completed HAP Emission Stream Data Form for one of six emission streams (#1) generated at a fictitious
company.
HAP EMISSION STREAM DATA FORM*
Cor
Loc
A.
B.
C.
D.
E.
F.
G.
H.
1.
J.
K.
L.
M.
N.
O.
npany Glaze Chemical Company
atjnn ( nf Fmissinn Streams Under Review
ant Identification *1 /
(a) paper coating oven
(a) process point
,a) toluene
(a) organic vapor
(a) 73 ppmv
(a) 28.4 mm Hg (3 77°F
(a) insoluble in water
|a\ provided
(a) 92 Ib/lb-mole
2% vol.
12C°F
15,000 scfm (max)
atmospheric
none / none
#3 Oven Exhaust
(b)
(b)
(b)
(h)
(h)
(h)
(b»
P. Organic Content (1)***
Q. Heat/02 Content (1)
R. Paniculate Content (3)
S Particle Mean Diam (3)
T. DriftVelocitv/SO-,(3)
, >
(r)
(c)
(c)
(r)
(n)
(r)
(r.\
44 ppmv CH/j, 4 pomv ot
0.4 Btu/scf / 20.6% vol.
/
U. Applicable Regulation(s)
V. Required Control Level
W. Selected Control Methods
*The data presented are for an emission stream (single or combined streams) prior to entry into the selected control
method(s). Use extra forms if additional space is necessary (e.g., more than three HAP's). and note this need.
**The numbers in parentheses denote what data should be supplied depending on the data in steps "C" and "E":
1 = organic vapor process emission
2 = inorganic vapor process emission
3 = participate process emission
l»* Organic emission stream combustibles less HAP combustibles shown on Lines D and F.
HAP's. In this handbook, 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, condens-
ers, 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 from 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 de-
vices once the emissions are captured by hooding,
enclosures, or closed vent systems and then trans-
ferred to a control device.
Area fugitive sources are characterized by large
6
surface areas from which emissions occur. In addi-
tion, process equipment such as pumps, valves,
and compressors are considered area fugitive
sources; 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 handbook generate emis-
sions; however, a definitive statement as to wheth-
er they emit a HAP cannot be made. As in the case
of identifying the potential HAP's emitted at a spe-
cific facility, communication with the owner/
operator is useful in identifying each source that
emits a HAP. The listings found in this section are
not 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.
-------
Table 2-1. Potential HAP's and Emission Sources for Solvent Usage Operations
Potential HAP's8
Source Key
A — bath evaporation
B —solvent transfer
C —ventilation
D —waste solvent disposal
E —washer
F —drying
G —still, filtration
H — cooker
Potential Emission Sources
Source Category
Solvent Degreasing
Dry Cleaning
Graphic Artsb
Waste Solvent Reclaiming
SCc-Flatwood Panelingd
SC-Machinerye
SC-Appliances'
SC-Metal Furniture
SC-Auto/Truck9
SC-Fabrics
SC-Cansh
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
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
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
UP
Q,R
Q,R
Q,R
S,R
D,K,Q,R,T
Q,U
B,I,Q,T
I,Q,T
Q
UP
S,R,T
S,R
I,V,W
Area
Fugitive
K
J,K
J,K
Q
Q
I — solvent storage
J — pipes, flanges, pumps
K —transfer areas
L — rollers
M— ink fountains
N — condenser
O — oven
P —coaters
Q — application area
R —flashoff area
S — spray booth
T — solvent/coating mixing
U —quench area
V — green tire spraying
W— sidewall/tread end/undertread
cementing
"References 2, 3, 4, 5, 6, 7, 8, 9, 10, 11,13, 14, 15,18, 31, 32.
blncludesflexography, lithography, offset printing, and textile printing.
CSC: surface coating.
dlncludes coating of other flat stock.
"Includes coating of misc. metal parts, machinery, and equipment.
' Includes all categories of appliances, large and small.
9lncludes coating of automobiles and light-duty trucks.
hlncludes surface coating of coils, cans, containers, and closures.
' Includes coating and maintenance of marine vessels.
' Includes vinyl, acrylic, and nitrocellulose coatings.
"Includes coating of trucks, buses, railroad cars, airplanes, etc.
2.2.1 Solvent Usage Operations
Solvent usage operations are defined as manufac-
turing processes that use solvents, including such
processes as surface coating operations, dry clean-
ing, solvent degreasing, waste solvent reclaiming,
and graphic arts. Table 2-1 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 2-1, all solvent usage operations
generate organic vapor emissions. (Note: Some of
the emission sources generate aerosols [i.e., or-
ganic particulate]; however, the aerosols evaporate
in a short time and the emissions normally are
controlled as a vapor. Therefore, Table 2-1 does not
indicate the presence of organic particulates.) Due
to the large number of potential HAP's associated
with these types of operations, the format of Table
2-1 prohibits the inclusion of compound-specific
data. Potential HAP's that may be emitted by
sources in Table 2-1 are summarized in Appendix
A.1; this appendix lists both specific compounds
and classes of compounds that may be emitted by
sources within the category. Appendix A.1 can be
used to determine whether a particular solvent us-
age operation may emit a specific potential HAP or
group of potential HAP's, as well as to determine all
solvent use operations that may emit a particular
potential HAP. Table 2-1 presents the emission
sources that may emit potential HAP's.
2.2.2 Metallurgical Industries
The metallurgical industries can be broadly divided
into primary, secondary, and miscellaneous metal
production operations. The majority of this indus-
try is covered under SIC Codes 331, 332, 333, 334
-------
Example Case
As directed by Section 2.2.1, Appendix A.1 is
used to determine the potential HAP's. The po-
tential 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 pres-
ent in the solvent being evaporated by the
ovens. Table 2-1 indicates that toluene is an or-
ganic compound, and it would be emitted as a
vapor. This information is then listed on the HAP
Emission Stream Data Form provided in Appen-
dix 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-
1 indicates that the remaining sources include
solvent transfer, solvent storage, application
areas, and solvent/coating mixing.
and 336. The term primary metals refers to produc-
tion 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 miscellaneous subdivision includes in-
dustries with operations that produce or use metals
for final products. Table 2-2 presents the potential
HAP's for these industries and the industry-specific
emission sources.
2.2.3 Synthetic Organic Chemical
Manufacturing Industry (SOCMI)
The SOCMI is a large and diverse industry produc-
ing several thousand intermediate and end-product
chemicals from a small number of basic chemicals.
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 sec-
tion to describe generic emission sources and spe-
cific emission source types. This approach is identi-
cal 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 condi-
tions). The emissions typically contain raw materi-
als (including impurities) used in and intermediate
and final products formed during the manufactur-
ing process. Many of these emission streams may
contain HAP's due to the great number of com-
pounds manufactured in the SOCMI. (See Appen-
dix A.5, reference 1, for further information on spe-
cific emissions for different SOCMI processes.)
Potential emissions from this industry can be de-
scribed generically as follows:
(a) Storage and handling emissions
(b) Reactor process emissions
(c) Separation process emissions
(d) Fugitive emissions
(e) Secondary emissions (e.g., from waste treat-
ment).
Emissions can potentially occur from raw materials
and product storage tanks as working and breath-
ing losses through vents. Emissions from handling
result during transportation or transfer of the vola-
tile organic liquids. Reactor processes and separa-
tion processes are the two broad types of pro-
cesses used in manufacturing organic chemicals.
Reactor processes involve chemical reactions that
alter the molecular structure of chemical com-
pounds. 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 economical-
ly. Typical emission sources in reactor processes
include point sources (e.g., vents on reactors and
product recovery devices), process fugitive 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 compres-
sors).
Separation processes often follow reactor pro-
cesses and divide chemical product mixtures into
distinct fractions. Emissions from separation pro-
cesses are associated primarily with absorption,
scrubbing, and distillation operations. Other sepa-
8
-------
Table 2-2. Potential HAP's and 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
Potential
Organic
Vapor Paniculate
3,8, 13, 19, 18
21,23,26,
27,28,30
18
18
18
18
18
18
18
Secondary Aluminum Operations
Secondary Copper Operations
(Brass and Bronze Production)
Gray Iron Foundries
Secondary Lead Smelting
Steel Foundries
Secondary Zinc Processing
Lead Acid Battery Production
18
2,3,13, 18
19,21,23
18
Cadmium-Nickel Battery Production
Dry Battery Production
Misc. Lead Products
Pollutant Key
7 — arsenic
2— acrolein
3— acetaldehyde
4 — ammonia
5 — antimony
6— barium
7 — beryllium
8 — benzene
9 — cadmium
70 — chromium
7 7 — copper
72 — fluoride
73 — formaldehyde
74— lead
75 — mercury
16 — manganese
17— nickel
18— polycyclic organic
matter (POM)
75 — phenol
20— selenium
27— toluene
22 — vanadium
23— xylene
24 — zinc
25 — iron
26— cresols
27— cyanides
28— pyridine
29 — hydrogen sulfide
30 — methyl mercaptan
HAP's0 Potential Emission Sources
Inorganic prorssK
Vapor Paniculate Point
12 72 A,I,J,
M,N,R
9 J,E
4,29 7,5,6,7,9 B
74,75,76
17,20,22
1, 12 1,5,9, 1 1 F,J,T
74,75,20,24
9, 10, 11,14 J
76,77,22,24
72 6,9,70,77 B,J,V
74,76,77,22,24
7, 12 1,5,9, 1 1 J,V
74,75,20
7,72 7,9,77 E,J,T,S
74, 75, 20,24
76 J
7,72 7,9,74 A,I,J,M,T
77,20,24
12 12, 17 J
24 9,77,74 J
77,20,24
7,6,7,9,70,77 J,Y
74,75,76,77
22,24,25
7,74,76,20 J
7,70,7,76,77,25 J,Y
24 9,75,77,20,24 J,E,S
74 74
9,74 V
76
74 5, 74
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
handling
Process Area
Fugitive Fugitive
H,K,D N,Q,U,Z
0,P N,Z
C,0,X N,D,Q,U
G,H,K,0,P,X N,Q,U,W,Z
H,K,0,P N,Q,W
C,H,K,0,X D,N,Q,U,W,Z
H,K,O,P N,Q,U,W,Z
0 N,Q,U,W,Z
H,K,M,P N,Q,Z
P N,Q.Z
H,K,P U
H,K,P U
H,K,G,P U
H,K,P U,Q
G,H,K,P U
H,K,L,P U
0,P
N,0
M,N,0
G,0,P
0 — material preparation
P — metal casting
Q — outdoor storage pile
R — reduction cell
S — retort
T — roaster
U — service road
V — sintering machine
W — slag dumping
X — vessel leakage
Y — foundry mold and core
decomposition
Z — mining operations
a References 6,16, 17, 19, 20, 21, 22, 23, 24, 25, 26, 28, 29, 30, 33, 34, 35.
9
-------
ration processes that may contribute to emissions
include drying, filtration, extraction, settling, crys-
talization, quenching, evaporation, ion exchange,
dilution, and mixing/blending. One of the more
commonly employed separation techniques is dis-
tillation. Depending on the type of distillation sys-
tem used (i.e., vacuum or nonvacuum), typical
emission points can include condensers, accumu-
lators, 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, accumu-
lators, and process drains from reactors, product
recovery devices, and separation equipment.
Table 2-3 presents information on specific emis-
sion points and emission source types for each of
Table 2-3. Emission Sources for the SOCMI"
Generic Source Category
Potential Emission Sources (Specific)
Process Process Area
Point Fugitive Fugitive15
Storage and Handling
Reactor Processes
Separation Processes
Fugitives
E,F
F,L
A
G
G,M,N
G,M,N
B,C,D
C,D,H,I,J,K
K
B,C,D,H,I
J,K,M,N,0
Source Key
A — storage, transfer, and I —
handling J —
B—spills K —
C —valves L —
D —flanges
E — reactors
F — product recovery devices
(absorber, adsorber, M —
condenser) N —
G — process drains 0 —
H — pumps
compressors
sampling lines
pressure relief devices
separation devices
(distillation column,
absorber, crystalizer,
dryer, etc.)
hotwell
accumulator
cooling tower
"References 12, 36, 37, 38, 39, 40, 41.
bGroups of small point sources (e.g., valves, compressors,
pumps, etc.) at a SOCMI plant are considered as area fugitive
sources in this handbook.
Figure 2-2. Potential emission points (shaded) for a vacuum distillation column using steam jet ejectors with barometric
condenser.(40)
Steam
Steam
I Ejector
Hotwell
10
-------
the generic emission source groups. Using this in-
formation, the user can identify the potential emis-
sion sources pertaining to his specific situation.
(See Appendix A.5, reference 1, for an illustration
of the approach outlined above and showing the
emission sources and the HAP's potentially emitted
from a SOCMI process.)
2.2.4 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 pro-
cesses may be high, but because of economic ne-
cessity they are usually recovered. In some cases,
the manufacturing operation is run as a closed sys-
tem, allowing little or no emissions to escape to the
atmosphere. Table 2-4 presents the potential HAP's
and industry-specific emission sources for these
industries.
2.2.5 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 man-
ufacture of finished chemical products for ultimate
consumption such as Pharmaceuticals, 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 eco-
nomic necessity they are usually recovered. Table
2-5 presents the potential HAP's and the industry-
specific emission sources for this industry.
2.2.6 Mineral Products Industry
This industry involves the processing and produc-
tion of various nonmetallic minerals. The industry
includes cement production, coal cleaning and con-
version, glass and glass fiber manufacture, lime
manufacture, phosphate rock and taconite ore pro-
cessing, as well as various other manufacturing
processes. Most of this industry falls under SIC
Codes 142, 144, 145, 147, 148, 149, 321, 322, 323,
324, 325, and 327. Table 2-6 presents the potential
HAP's and the industry-specific emission sources
industries.
2.2.7 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 kraft, sul-
fite, and neutral sulfite. Plywood production in-
volves manufacturing 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 fungistatic and insecticidal
properties or impart fire resistance. Table 2-7 pre-
sents the potential HAP's and the industry specific
emission sources for these industries.
2.2.8 Petroleum Related Industries
In this handbook, the petroleum related industries
source category includes the oil and gas produc-
tion industry, the petroleum refining industry, and
the basic petrochemicals industry; these industries
fall under SIC Codes 13 and 29.
The oil and gas production industry includes the
following processes: exploration and site prepara-
tion, drilling, crude processing, natural gas pro-
cessing, and secondary or tertiary recovery. The
principal 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 vari-
ety of fuel oils, lubricating oils, and feedstocks for
the petrochemicals industry. The different pro-
cesses involved in the petroleum refining industry
are crude separation, light hydrocarbon process-
ing, middle and heavy distillate processing, and
residual hydrocarbon processing.
In the basic petrochemicals industry, hydrocarbon
streams from the oil and gas production and petro-
leum refining industries are converted into feed-
stocks 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, purifi-
cation, and chemical conversion processes. (See
Appendix A.6, reference 1, for a breakdown of typi-
cal processes involved in each of the three indus-
tries.)
Table 2-8 presents the potential HAP's that may be
emitted from these industries. Table 2-9 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 or-
ganic vapor emissions. This is due to the chemical
77
-------
Table 2-4. Potential HAP's and Emission Sources for the Inorganic Chemical Manufacturing Industry
Potential HAP's Potential Emission Sources
Source Category
Aluminum Chloride
Aluminum Fluoride
Ammonia
Ammonium Acetate
Ammonium — nitrate, sulfate
thiocyanate, formate, tartrate
Ammonium Phosphate
Antimony Oxide
Arsenic — disulfide, iodide
pentafluoride, 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
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
Nickel Sulfate
Nitric Acid
Phosphoric Acid
Wet process
Thermal process
Phosphorus
Phosphorus Oxychloride
Inorganic
Vapor
4,W
17
1
1
1
1,17
5
2
8,10
3,17
10
19,24
12
14
17
1,39
19,20
17
10
10,20
40
3
22
24
27
28
10,17,2,30,18
17
10
Particulate
2
6
7
9
15
2
25
11,12
11
11
13
20
38
20
2,21
21
23
23
25
26
26
28
30
Process Process
Point Fugitive
X X
X X
B,D,E K
X X
C,F,I,L Q
X X
X X
H,U K,Q,T
C,E,G,I,L,U N,P,Q,T
X X
X X
X X
X X
H K,P
H,C K,R
X X
H K,N,0,Q
X X
X X
X X
X X
X X
X X
B
B,G K,R
X X
X X
X X
G,L P,Q
G,R P,Q
G,R P,Q
G,L Q,P,T
G,L Q,P,T
X X
P,Q
L Q,T
B,H K,N,R
H,C,W K,N,P,T
B,G K,N,R,T
X X
X X
Area
Fugitive
J,S
J,S
J
J,S
J,S
J,S
J,S
-------
Potential HAP's3
Potential Emission Sources
Source Category
Inorganic
Vapor
Paniculate
Process
Point
Process
Fugitive
Area
Fugitive
Phosphorus Pentasulfide
Phosphorus Trichloride
Potassium — bichromate,
chromate
Potassium Hydroxide
Sodium Arsenate
Sodium Carbonate
Sodium Chlorate
Sodium Chromate —
dichromate
Sodium Hydrosulfide
Sodium— silicofluoride,
fluoride
Sulfuric Acid
Sulfur Monochloride —
dichloride
Pollutant Key
7— ammonia
2— arsenic
3— arsenic trioxide
4—aluminum chloride
5— antimony trioxide
6— barium salts
7— beryllium
8— bromine
9— boron salts
10— chlorine
77— chromium salts
12— chromic acid mist
13— cobalt metal fumes
14— copper sulfate
15—cadmium salts
76—chromates (chromium)
77—fluorine
18— hydrogen sulfide
79— hydrogen chloride
20— hydrochloric acid
21— lead
22— lead chromate
23— manganese salts
24— manganese dioxide
25— mercury
26— nickel
29,31
32,10,29
16
10
1
10
16
18
17
33,34
10
29
29
16
25
2
16
33
X
X
I
X
H
I,L,V
X
G,I,L,M
X
X
A,B,C,H
X
X
X
X
K,P
P
X
P,Q
X
X
K,R
X
J,S
Zinc Chloride
Zinc Chromate (pigment)
Zinc Oxide (pigment)
36,27
35
37
21
X
X
X
X
X
X
27— nickel sulfate
28— nitric acid mist
29—phosphorus
30— phosphoric acid mist
31— phosphorus pentasulfide
32— phosphorus trichloride
33— sulfuric acid mist
34—sulfur trioxide
35— zinc chromate
36— zinc chloride fume
37— zinc oxide fume
38—iodine
39— hydrazine
40— iron oxide
Source Key
A —converter
B — absorption tower
C — concentrator
D —desulfurizer
E — reformer
F — neutralizer
G—kiln
H — reactor
I —crystallizer
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 well
X — no information
a References 6, 13, 20, 22, 23, 24, 25, 27, 28, 29, 30, 33, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51.
13
-------
Table 2-5. Potential HAP's and Emission Sources for the Chemical Products Industry
Source Category
Carbon Black
Charcoal
Explosives
Fertilizers
Paint and Varnish
Pharmaceutical
Plastics
Printing Ink
Pesticides
Soap and Detergents
Synthetic Fibers
Synthetic Rubber
Pollutant Key
1 — arsenic
2 — acrolein
3 — acrylonitrile
4 — acetic acid
5 — boron
6 — barium
7— beryllium
8 — benzene
9 — cresols
10 — cadmium
11 — chromium
12 — chloroprene
13 — caprolactum
14 — carbon disulfide
15 — carbonyl sulfide
16 — carbon tetrachloride
77— chloroform
18 — dichlorobenzene
19 — dimethylformamide
20 — dimethylamine
21— ethylene
22— ethylene dichloride
23— formaldehyde
24 — hydrogen sulfide
25 — hexachlorocyclopentadiene
26— hydrogen fluoride
27— ketones
28 — mercury
29 — manganese
30— methanol
Potential HAP's3
Organic Inorganic
Vapor Particulate Vapor Particulate
14,15,21 41 1 1,7,10,11
16,24 28,29,37
4,23,30 41
23
23,26,44 49
16,22, 6,28,43,48
31,46
17,18,31, 28
34,46
23,33,35,
39,42,50,
51,52
2, 16,27,
42,46,45
9,16,17,18, 28
20,25,32,
36,39,47
8,34 41 1 1,5
3,8,13,14
19,23,32,
38,40,42,
46,24
3,12,18,20, 41
22,33,34,
35,36,
46,49
31 — methyl chloroform
(1,1,1-trichloroethane)
32 — maleic anhydride
33— 1,3- butadiene
34 — morpholine
35 — metnylene chloride
36 — nitrosomines
37— nickel
38 — perchloroethylene
39 — phosgene
40— phthalic anhydride
41 — polycyclic organic matter
42 — phenol
43— selenium
44 — silicontetrafluoride
45 — terpenes
46— toluene
47 — xylene
48— zinc
49 — ammonia
50 — vinyl chloride
51 — toluene diisocyanate
52 — pyridine
Potential Emission Sources
Process Process Area
Point Fugitive Fugitive
B,H G,K,L 1
E
A,C,H K
D,H,R,S,V K,T
N,0 L
A,H,U,W G,L F
A,P,V K,L F,l
Q
A,H,O,X G F,l
M,N,0 K,L
A,H,J,O,U, G,K 1
v,x,z
A,H,0,P,X,Z Y F
Source Key
A — reactor
B — furnace
C — concentrator
D — neutralizer
E —kiln
F — compressor and pump seals; valves,
flanges, open ended lines, sampling lines
G — storage tank vents
H — dryer
1 — spills
J — spin cell or bath
K — product handling, finishing.
and packaging
L — raw material transport and unloading
M — spray dryer
N — kettle
0 — mixing tank (blend tank)
P — polymerization vessel
Q — cooking vessel
R — prill tower
S — granulator
T — screen
U —distillation
V — cooler (condenser)
W — crystallizer
X— filter
Y — milling/blending/compounding
Z —flash tank
References 6, 13, 21, 22, 23, 33, 52, 53, 54, 55, 56, 57, 58, 59, 60.
14
-------
Table 2-6. Potential HAP's and Emission Sources for the Mineral Products Industry
Potential HAP's"
Potential Emission Sources
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
Organic Inorganic
Vapor Paniculate Vapor
2,8, 13 18
10,12
21,26
10,12
18 10,12
17,23
8, 19 18 4,23
27,28
13,19 19 6,20,22
12
1,4, 12,
14,26
15
15
13,19 12,23
12
6,20,21
Particulate
3,10,17
7,10,12
10,12
7,9, 10, 14
15,17,
24,25
22
1,5,6,7,9,
10,11,14,
16,17,20,
21,24
1,5,7,9,14
16,17,
20,29
12
1,5,6,12,
14,20,21
15
15
12
6,20,21
3
Process
Point
B
B,E,C
B,E
E
B,C
B,H
C,0
B,C
C
E,T
C
C,0
B,C
A,B,Q
C,Q
Process Area
Fugitive Fugitive
D,N I,L
F,J,M 1
D,F,IM I,L
D,F,N 1
F,G,N,S I,L
M,N,R I,L
M,N I,L
F,G,M,N I,L
D,F,G,N,P 1
S I,L
D,F,M,N 1
G,R,S I,L
G,N I,L
D,G,P I,L
G,M,N,S I,L
F,M,N,R I,L
F,M,N,R I,L
Pollutant Key
1—arsenic
2—aldehydes
3— asbestos
4— ammonia
5— antimony
6— barium
7— beryllium
8— benzene
9—cadmium
10—chromium
11—copper
12— fluoride
13— formaldehyde
14— lead
15— mercury
16— manganese
77—nickel
18— polycyclic organic
matter (POM)
19— phenol
20—selenium
21— boron
22—coal dust
23— hydrogen sulfide
24—zinc
25— iron
26— chlorine
27— cresols
28—toluene
29— phosphorus
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
a References 6, 13, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 45, 47, 64, 65, 66, 67, 68, 69, 70, 71, 73.
75
-------
Table 2-7. Potential HAP's and Emission Sources for the Wood Products Industry
Potential HAP's3
Potential Emission Sources
Source Category
Chemical Wood Pulping
Kraft pulp mill
Sulfite pulp mill
Neutral sulfite pulp mill
Plywood, Particleboard, Hardboard
Wood Preservative
Organic
Vapor Particulate
i e
h e
e
h,l,o,p
j,g,m,n
Inorganic
Vapor Particulate
k
f,k
k
a,b,c,d
a,b,c,d
a,c,d
Process
Point
A,B,C,D
A,B,C,
A,C,E
G
Process Area
Fugitive Fugitive
F
F
Pollutant Key
a —arsenic
b —asbestos
c —chromium
d — mercury
e —polycyclic organic matter (POM)
f —chlorine
g —chlorobenzene
h —formaldehyde
i —methyl mercaptan
j —dioxin
k — hydrogen sulfide
I —phenol
m — pentachlorophenol
n —cresols
o — abietic acid
p — pinene
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
"References 4, 13, 21, 22, 23, 33, 56, 73.
Table 2-8. Potential HAP's for Petroleum Related Industries"
(General Listing for Entire Source Category)
Potential HAP's
Organic
Inorganic
Vapor
Particulate
Vapor
Particulate
Parafins (CrC10) Coke fines Sulfides
Cycloparafins
Aromatics (e.g.,
benzene, toluene
xylene)
Phenols
Sulfur containing
compounds (e.g.,
mercaptans,
thiophenes)
(e.g., hydrogen
sulfide, carbon
disulfide,
carbonyl sulfide)
Ammonia
Catalyst
fines
"References 26, 75, 76, 77, 78, 79, 80.
composition of the two starting materials used in
these industries—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 contain-
ing sulfur, nitrogen, and oxygen. Natural gas is
largely saturated hydrocarbons (mainly methane).
The remainder may include nitrogen, carbon diox-
ide, hydrogen sulfide, and helium. Organic and in-
organic particulate emissions, such as coke fines or
catalyst fines, may be generated in some pro-
cesses.
The emission sources within each of the petroleum
related industries are given in Table 2-10. Sources
of potential HAP emissions from the oil and gas
production industry include blowouts during drill-
ing operations; storage tank breathing and filling
losses; wastewater treatment processes; and fugi-
tive 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 combus-
tion units (e.g., process heaters). Fugitive emis-
sions are a major source of emissions in this indus-
try. Emission sources in the basic petrochemicals
industry are similar to those from the petroleum
refining industry and the SOCMI (see Section
2.2.3).
2.2.9 Combustion Sources
The fuel combustion industry encompasses a large
number of combustion units generally used to pro-
duce 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, configura-
tion, and type of fuel burned. Coal, fuel oil, and
natural gas are the major fossil fuels burned, al-
though 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 in-
ternal combustion units such as generators,
pumps, and well-drilling equipment are also in-
cluded in this category.
The waste incineration category includes combus-
tion processes whereby municipal solid wastes or
sewage treatment sludges are disposed. Table 2-11
presents the potential HAP's and the facility specif-
ic emission sources for the above combustion
sources.
16
-------
Table 2-9.
Process
Potential HAP's for Petroleum Refining Industries (26)
(Specific Listing for Petroleum Refining Segment)
Potential HAP's
Organic
Inorganic
Vapor
Paniculate
Vapor
Pollutant Key
a —maleic anhydride
b — benzoic acid
c —chlorides
d — ketones
e —aldehydes
f —heterocyclic compounds
(e.g., pyridines)
g — benzene
h —toluene
i —xylene
j — phenols
k —organic compounds containing
sulfur (sulfonates, sulfones)
I — cresol
m — inorganic sulfides
n — mercaptans
o — polynuclear compounds (benzo
pyrene, anthracene, etc.)
p — vanadium
q — nickel
r — lead
s — sulfuric acid
t — hydrogen sulfide
u — ammonia
v —carbon disulfide
x — carbonyl sulfide
y — cyanides
z — chromates
A — acetic acid
B —formic acid
C —methylethylamine
D —diethylamine
E —thiosulfide
F — methyl mercaptan
Particulate
Crude Separation
Light Hydrocarbon Processing
Middle and Heavy
Distillate Processing
Residual Hydrocarbon Processing
Auxiliary Processes
a,b,d,e,f,g,h,i,j,k,l,m,o,
A,B,C,D,E,F,J
g,h,i,n,N,0,P
a,d,e,f,g,h,i,j, k,l,
F,J,K,0,P,S,T
a,d,e,f,g,h,i,j,k,l,n,
F,J,M,N,P,S,T
a,b,d,e,f,g,h,i,j,k,l,n,
A,B,C,D,J,K,M,T
0
R
o,R
o,R
o,R
c,m,t,u,
v,x,y,L
t,v
m,t,u,v,
x,y,L
m,s,t,u,
v,x,y,L
c,m,s,u,
y,L
P,I,Q,R
G,H,Q
P,q,G,H,
I,Q,U
P,q,G,H,
I,Q,U
P,q,r,z,
1
G —cobalt
H — molybdenum
I —zinc
J —cresylic acid
K —xylenols
L —thiophenes
M — thiophenol
N — nickel carbonyl
0 — tetraethyl lead
P —cobalt carbonyl
Q —catalyst fines
R —coke fines
S —formaldehyde
T —aromatic amines
U —copper
Table 2-10. Emission Sources for Petroleum Related Industries
Source Category
Potential HAP Emission Sources
Process
Point
Process
Fugitive
Area
Fugitive
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
A
G
G,J,K
G
G,J,L
O,G
G,O,P,R
B,G,K,0,R
G
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
C
F,H
H
F,H,M,N
F,H
F,H
H
F,H
D,E
I
Q
I
I
Basic Petrochemicals Industry
Olefins Production
Butadiene Production
Benzene/Toluene/Xylene (BTX) Production
Naphthalene Production
Cresol/Cresylic Acids Production
Normal Paraffin Production
G,K,0
G,J,L,0,R
G,K,O,R
G,L,0
G,L
G,0
F,H
F,H,N
F,Q
F,H
F,H
F,H
1
1
1
1
1
L —distillation/fractionation
M — hotwells
N — steam ejectors
0 —catalyst regeneration
P —evaporation
Q — catalytic cracker
R —stripper
77
-------
Table 2-11. Potential HAP's and Emission Sources 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 Sludge Incineration
PCB Incineration
Potential
Organic
Vapor
14,25,28
14
14
12,14
12
3,4, 12,
74,25
7,12,21,
23,26
12
12
12,21
Particulate
19
19
19
12,19
12,19
12,19
12,19
12,19
12,19
12,19
HAP's
Inorganic
Vapor Particulate
1,2,8,9,13 1,2,5,6,8,9
17,27 10,11,15,
16, 18,20,
22,24
13,17,27 1,2,5,6,8,9,
10,11,15,
16,18,22,
24,29
17 15
6,18
27 16,20
6,8,9,
15,18
77,27 6,8,9,11,
15,16,18
17 1,6,8,9,
15,16,18
30
Potential
Process
Point
A,B
A,B,E
A,B,E,F,
G
G
A,B,C
A,B,D
D
D
D,B
Emission Sources
Process Area
Fugitive Fugitive
I H
Pollutant Key
7 — arsenic
2— antimony
3—acetaldehyde
4— acetic acid
5— barium
6— beryllium
7— benzene
8—cadmium
9—chromium
10—cobalt
77 — copper
72—dioxin
73—fluoride
14—formaldehyde
75—lead
16— manganese
77— mercury
18— nickel
19— polycyclic organic matter (POM)
20—phosphorus
27 — polychlorinated biphenyls (PCB)
22— radionuclides
23—trichloroethy lene
24—zinc
25— phenol
26— ethyl benzene
27— chlorine
28— pyridine
29—vanadium
30—dibenzofuran
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
a References 6, 13, 20, 21, 23, 24, 25, 26, 27, 28, 29, 30, 33, 34, 35, 53, 72, 81, 82, 83, 84 .
2.3 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 meth-
odologies are outside the scope of this handbook;
however, control techniques for vapor emissions
and particulate emissions from area fugitive
sources are discussed in Sections 3.2.4 and 3.3.2,
respectively.
The actual/estimated values for the process emis-
sion 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 2-12 lists
the required information for organic vapor emis-
sions, Table 2-13 for inorganic vapor emissions,
and Table 2-14 for particulate emissions. After ob-
taining the values for the key physical properties
for each HAP emission stream, record the data on
the HAP Emission Stream Data Form found in Ap-
pendix C.1 (see Figure 2-1).
There will be occasions when it would be prudent
for the owner/operator to combine similar emis-
sion streams. For example, if two or more emission
streams require the use of the same control tech-
nique, it will normally be more cost effective to
combine the streams and use just one control de-
vice as opposed to using a control device for each
separate emission stream. If the owner/operator
18
-------
decides to combine emission streams, Appendix
B.1 provides calculation procedures to determine
the key effluent properties of combined emission
streams.
Table 2-13. Key Properties for Inorganic Vapor Emissions
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 2-12. 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 pre-
sented in Appendix C.1. The source test pro-
vided the following data (see Figure 2-1):
HAP content
Organic content
Moisture content
Halogen content
Metal content
Temperature
Pressure
Flow rate
Heat content
Oxygen content
HAP molecular weight
HAP vapor pressure
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
Emission Stream Properties
(Preferred units of measure)
HAP Properties3
HAP Contentb (ppm by volume)
Moisture Content (% by volume)
Halogen/Metal Content (yes or no)
Flow Rate (scfm)
Temperature (°F)
Pressure (mm Hg)
Molecular Weight
Vapor Pressure
Solubility (graph)
Adsorptive Properties
(isotherm plot)
"These properties pertain to the specific HAP or mixture of
HAP's in the emission stream.
bPrimary properties that affect control technique selection.
Table 2-14. Key Properties for Paniculate Emissions
Emission Stream Properties HAP Properties3
(Preferred units of measure)
HAP Content (% by mass)
Paniculate Contentb (Ib/acf)
Moisture Content (% by volume)
S03 Content (ppm by volume)
Flow Rate (acfm)
Temperature (°F)
Particle Mean Diameter0 (
Drift Velocity0 (ft/sec)
None
"These properties pertain to the specific HAP or mixture of
HAP's in the emission stream.
bData include total particulate loading and principle paniculate
constituent.
These properties are necessary only for specific control tech-
niques.
Table 2-12. Key Properties for Organic Vapor Emissions
Emission Stream Properties HAP Properties3
(Preferred units of measure)
HAP Content (ppm by volume)
Organic Content6 (ppm by volume)
Heat Content0 (Btu/scf)
Oxygen Content (% by volume)
Moisture Content (% by volume)
Halogen/Metal Content (yes or no)
Flow Rate (scfm)
Temperature (°F)
Pressure (mm Hg)
Molecular Weight
Vapor Pressure
Solubility (graph)
Adsorptive Properties
(isotherm plot)
3These properties pertain to the specific HAP or mixture of
HAP's in the emission stream.
bPrimary properties that affect control technique selection. Or-
ganic content is defined as organic emission stream combusti-
bles less HAP emission stream combustibles.
°Heat content is determined from HAP/Organic Content (see
Appendix B.1 for calculation procedures).
2.4 References
1. U.S. EPA. Evaluation of Control Technologies
for Hazardous Air Pollutants. EPA-600/7-86-
009 (NTIS PB 86-167/038/AS). October 1985.
2. National Paint and Coatings Association. Sec-
tion III: Paint and Coatings Markets. Table A-6,
Estimated Consumption of Solvents in Paints
and Coatings, by Market - 1981. pp. 208-209.
(no date).
3. U.S. EPA. Organic Solvent Cleaners - Back-
ground Information for Proposed Standard
(Draft). EPA-450/2-78-045a. October 1979.
4. U.S. EPA. End Use of Solvents Containing
Volatile Organic Compounds. EPA-450/3-79-
032. May 1979.
5. U.S. EPA. Source Assessment: Solvent Evapo-
ration - Degreasing Operations. EPA-600/2-79-
019f. August 1979.
19
-------
6. U.S. EPA. Compilation of Air Pollutant Emis-
sion Sources. Third Edition: Supplements 1-15.
AP-42. January 1984.
7. U.S. EPA. Guidance for Lowest Achievable
Emission Rates for 18 Major Stationary
Sources of Particulates, Nitrogen Oxides, Sul-
fur Dioxide, or Volatile Organic Compounds.
EPA-450/3-79-024. April 1979.
8. U.S. EPA. Control of Volatile Organic Emis-
sions from Existing Stationary Sources - Vol-
ume VI: Surface Coating of Miscellaneous Met-
al Parts and Products. EPA-450/2-78-015. June
1978.
9. U.S. EPA. Control of Volatile Organic Emis-
sions from Existing Stationary Sources - Vol-
ume II: Surface Coating of Cans, Coils, Paper,
Fabrics, Automobiles, and Light Duty Trucks.
EPA-450/2-77-008. May 1977.
10. U.S. EPA. Control of Volatile Organic Com-
pound Emissions from Large Petroleum Dry
Cleaners. EPA-450/3-82-009. September 1982.
11. U.S. EPA. Pressure Sensitive Tape and Label
Surface Coating Industry - Background Infor-
mation for Proposed Standards. EPA-450/3-80-
003a. September 1980.
12. U.S. EPA. Hazardous/Toxic Air Pollutant Con-
trol Technology: A Literature Review. EPA-
600/2-84-194. December 1984.
13. U.S. EPA. Nonindustrial Sources of Toxic Sub-
stance Emissions and Their Applicability to
Source Receptor Modeling. Draft Report, EPA
Contract No. 68-02-3509, Task No. 42. July 27,
1983.
14. U.S. EPA. Control Technique Guidelines for the
Control of Volatile Organic Emissions from
Wood Furniture Coating (Draft). April 1979.
15. U.S. EPA. Flexible Vinyl Coating and Printing
Operations - Background Information for Pro-
posed Standards. EPA-450/3-81-016a. January
1983.
16. U.S. EPA. Background Information for New
Source Performance Standards: Primary Cop-
per, Zinc, and Lead Smelters - Volume 1: Pro-
posed Standards. EPA-450/2-74-002a. October
1974.
17. U.S. EPA. Background Information for Stan-
dards of Performance: Electric Submerged Arc
Furnaces for Production of Ferroalloys - Vol-
ume 1: Proposed Standards. EPA-450/2-74-
018a. October 1974.
18. U.S. EPA. Control Techniques for Volatile Or-
ganic Compound Emissions from Stationary
Sources- Third Edition (Draft). April 1985.
19. 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.
20. U.S. EPA. Locating and Estimating Air Emis-
sions from Sources of Chromium. EPA-450/4-
84-007g. July 1984.
21. U.S. EPA. A Survey of Emissions and Controls
for Hazardous and Other Pollutants. EPA-R4-
73-021. February 1973.
22. U.S. EPA. Industrial Sources of Hazardous Air
Pollutants (Draft). September 1983.
23. U.S. EPA. Source Assessment: Noncriteria Pol-
lutant Emissions (1978 Update). EPA-600/2-78-
004T. July 1978.
24. U.S. EPA. Locating and Estimating Air Emis-
sions from Sources of Manganese (Draft). Sep-
tember 1984.
25. U.S. EPA. Locating and Estimating Air Emis-
sions from Sources of Nickel. EPA-450/4-84-
007F. March 1984.
26. U.S. EPA. Potentially Hazardous Emissions
from the Extraction and Processing of Coal and
Oil. EPA-650/2-75-038. April 1975.
27. U.S. EPA. Review of National Emission Stan-
dards for Mercury. EPA-450/3-84-01. December
1984.
28. U.S. EPA. Status Assessment of Toxic Chemi-
cals: Lead. EPA-600/2-79-210h. December
1979.
29. U.S. EPA. Status Assessment of Toxic Chemi-
cals: Mercury. EPA-600/2-79-210J. December
1979.
30. U.S. EPA. Sources of Copper Air Emissions.
EPA-600/2-85-046. April 1985.
31. 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.
32. U.S. EPA. Rubber Tire Manufacturing Industry -
Background Information for Proposed Stan-
dards. EPA-450/3-81-008a. July 1981.
33. U.S. EPA. Human Exposure to Atmospheric
Concentrations of Selected Chemicals. EPA
Contract No. 68-02-3066. February 1982.
34. U.S. EPA. Survey of Cadmium Emission
Sources. EPA-450/3-81-013. September 1981.
35. U.S. EPA. Source Category Survey: Secondary
Zinc Smelting and Refinery Industry. EPA-
450/3-80-012. May 1980.
20
-------
36. U.S. EPA.XWr Oxidation Processes in Synthetic
Organic Chemical Manufacturing Industry -
Background Information for Proposed Stan-
dards. EPA-450/3-82-001a. October 1983.
37. U.S. EPA. Reactor Processes in Synthetic Or-
ganic Chemical Manufacturing - Background
Information for Proposed Standards (Draft).
October 1984.
38. U.S. EPA. VOC Emissions from Volatile Organic
Liquid Storage Tanks — Background Informa-
tion for Proposed Standards (Draft). EPA-450/3-
81-003. July 1984.
39. U.S. EPA. VOC Fugitive Emissions in Synthetic
Organic Chemicals Manufacturing Industry -
Background Information for Promulgated Stan-
dards. EPA-450/3-80-033b. June 1982.
40. U.S. EPA. Distillation Operations in Synthetic
Organic Chemical Manufacturing - Background
Information for Proposed Standards. EPA-
450/3-83-005a. December 1983.
41. U.S. EPA. Organic Chemical Manufacturing
Volume 6: Selected Processes. EPA-450/3-80-
028a. December 1980.
42. U.S. EPA. Source Category: Ammonia Manu-
facturing Industry. EPA-450/3-80-014. August
1980.
43. U.S. EPA. Source Assessment: Ammonium Ni-
trate Production. EPA-600/2-77-1071. Septem-
ber 1977.
44. U.S. EPA. Ammonium Sulfate Manufacture -
Background Information for Proposed Stan-
dards. EPA-450/3-79-034a. December 1979.
45. U.S. EPA. Preliminary Study of Sources of Inor-
ganic Arsenic. EPA-450/5-82-005. August 1982.
46. U.S. EPA. Source Assessment: Major Barium
Chemicals. EPA-600/2-78-004b. March 1978.
47. U.S. EPA. Emission Factors for Trace Sub-
stances. EPA-450/2-73-001. December 1973.
48. U.S. EPA. Review of New Source Performance
Standards for Nitric Acid Plants. EPA-450/3-84-
011. April 1984.
49. U.S. EPA. Sodium Carbonate Industry - Back-
ground Information for Proposed Standards.
EPA-450/3-80-029a. August 1980.
50. U.S. EPA. Industrial Process Profiles for Envi-
ronmental Use: Sulfur, Sulfur Oxides and Sul-
furicAcid. EPA-600/2-77-023w. February 1977.
51. U.S. EPA. Final Guideline Document: Control
of Sulfuric Acid Mist Emissions from Sulfuric
Acid Production Plants. EPA-450/2-77-019. Sep-
tember 1977.
52. U.S. EPA. Source Assessment: Charcoal Manu-
facturing. EPA-600/2-78-004z. December 1978.
53. U.S. EPA. Locating and Estimating Air Emis-
sions from Sources of Formaldehyde. EPA-
450/4-84-007e. March 1984.
54. U.S. EPA. Locating and Estimating Air Emis-
sions from Sources of Chloroform. EPA-450/4-
84-007c. March 1984.
55. U.S. EPA. Locating and Estimating Air Emis-
sions from Sources of Carbon Tetrachloride.
EPA-450/4-84-007b. March 1984.
56. U.S. EPA. Locating and Estimating Air Emis-
sions from Sources of Chlorobenzenes (Draft).
September 1984.
57. U.S. EPA. Plastics and Resins Industry - Indus-
trial Process Profiles for Environmental Use.
EPA-600/2-77-023J. February 1977.
58. U.S. EPA. Locating and Estimating Air Emis-
sions from Sources of Phosgene (Draft). Sep-
tember 1984.
59. U.S. EPA. Locating and Estimating Air Emis-
sions from Sources ofAcrylonitrile. EPA-450/4-
84-007a. March 1984.
60. U.S. EPA. Locating and Estimating Air Emis-
sions from Sources of Ethylene Dichloride.
EPA-450/4-84-007d. March 1984.
61. U.S. EPA. Asphalt Roofing Manufacturing In-
dustry - Background Information for Proposed
Standards (Draft). EPA-450/3-80-021a. June
1980.
62. U.S. EPA. Trace Pollutant Emissions from the
Processing of Nonmetallic Ores. EPA-650/2-74-
122. November 1974.
63. U.S. EPA. Source Category Survey: Refractory
Industry. EPA-450/3-80-006. March 1980.
64. U.S. EPA. A Review of Standards of Perfor-
mance for New Stationary Sources - Portland
Cement Industry. EPA-450/3-79-012. March
1979.
65. U.S. EPA. Background Information for Stan-
dards of Performance: Coal Preparation Plants
Volume I: Proposed Standards. EPA-450/2-74-
021 a. October 1974.
66. U.S. EPA. Glass Manufacturing Plants, Back-
ground Information: Proposed Standards of
Performance (Draft). EPA-450/3-79-005a. June
1979.
67. U.S. EPA. Wool Fiberglass Insulation Manufac-
turing Industry - Background Information for
Proposed Standards (Draft). EPA-450/3-83-
002A. December 1983.
21
-------
68. U.S. EPA. Standards Support and Environmen-
tal Impact Statement Volume I: Proposed Stan-
dards of Performance for Lime Manufacturing
Plants. EPA-450/2-77-007a. April 1977.
69. U.S. EPA. Final Standards Support and Envi-
ronmental Impact Statement Volume II: Pro-
mulgated Standards of Performance for Lime
Manufacturing Plants. EPA-450/2-77-007b. Oc-
tober 1977.
70. U.S. EPA. Source Category Survey: Mineral
Wool Manufacturing Industry. EPA-450/3-80-
016. March 1980.
71. U.S. EPA. Source Category Survey: Perlite In-
dustry. EPA-450/3-80-005. May 1980.
72. U.S. EPA. Radionuclides - Background Informa-
tion Document for Final Rules. Volume I. EPA-
520/1-84-022-1. October 1984.
73. Standards Support and Environmental Impact
Statement for the Iron Ore Benefication Indus-
try (Draft). Battelle Columbus Laboratories. De-
cember 1976.
74. U.S. EPA. Kraft Pulping - Control of TRS Emis-
sions from Existing Mills. EPA-450/2-78-003b.
March 1979.
75. U.S. EPA. Industrial Process Profiles for Envi-
ronmental Use: Chapter 2. Oil and Gas Produc-
tion Industry. EPA-600/2-77-023b. February
1977.
76. U.S. EPA. Industrial Process Profiles for Envi-
ronmental Use: Chapter 3. Petroleum Refining
Industry. EPA-600/2-77-023c. January 1977.
77. U.S. EPA. Industrial Process Profiles for Envi-
ronmental Use: Chapter 5. Basic Petrochemi-
cals Industry. EPA-600/2-77-023e. January
1977.
78. U.S. EPA. VOC Fugitive Emissions in Petroleum
Refining Industry - Background Information for
Proposed Standards. EPA-450/3-81-015a. No-
vember 1982.
79. U.S. EPA. VOC Species Data Manual, Second
Edition. EPA-450/4-80-115. July 1980.
80. U.S. EPA. Bulk Gasoline Terminals - Back-
ground Information for Proposed Standards
(Draft). EPA-450/3-80-038a. December 1980.
81. U.S. EPA. Air Toxics Emission Patterns and
Trends - Final Report. EPA Contract No. 68-02-
3513, Task 46. July 1984.
82. U.S. EPA. Hazardous Emission Characteriza-
tion of Utility Boilers. EPA-600/2-75-066. July
1975.
83. U.S. EPA. Thermal Conversion of Municipal
Wastewater Sludge Phase II: Study of Heavy
Metal Emissions. EPA-600/2-81-203. Septem-
ber 1981.
84. U.S. EPA. Locating and Estimating Air Emis-
sions from Sources of Polychlorinated Biphen-
yls (Draft). November 1984.
22
-------
Chapter 3
Control Device Selection
3.1 Background
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 techniques
that can be applied to control HAP emissions from
a specific emission source will depend on the emis-
sion source characteristics and HAP characteristics.
Therefore, Section 3.2, Vapor Emissions Control,
and Section 3.3, Particulate Emissions Control,
each pertain to specific HAP groups. The discus-
sion of control technique selection within each sec-
tion is according to type of HAP (organic or inor-
ganic) 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 consid-
erations that are important in control device selec-
tion are described in detail.
Work practices, including equipment modifica-
tions, 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 hood-
ing or enclosure or collected by closed vent sys-
tems and then transferred to a control device. Note
that the overall performance of the control system
will then be dependent on both the capture efficien-
cy 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 Chap-
ter 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).
3.2 Vapor Emissions Control
3.2.1 Control Techniques for Organic Vapor
Emissions from Point Sources
The most frequent approach to point source con-
trol 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 ab-
sorbers. The combustion devices are the more
commonly applied control devices, since they are
capable of high removal (i.e., destruction) efficien-
cies 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 con-
sideration than on the particular source category
(e.g., degreasing vs. surface coating in solvent us-
age 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 applica-
bility of each control technique and presents limit-
ing 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 applica-
ble control techniques will then be narrowed fur-
ther depending on the capability of the applicable
control devices to achieve the required perfor-
mance 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
23
-------
Table 3-1. Key Emission Stream and HAP Characteristics for Selecting Control Techniques for Organic Vapors from Point Sources
Emission Stream Characteristics HAP Characteristics3
Control Device
Thermal
Incinerator
Catalytic
Incinerator
Flare
Boiler/
Process Heater9
Carbon
Adsorber
Absorber
Condenser
HAP/Organics
Contents6
(ppmv)
>20;
(<25% of LELC)
50-10,000;
(<25% of LELC)
1,000-10,000
(<25% of LELC)
250-10,000
>5,000
Heat Moisture
Content Content Flow Rate
(Btu/scf) (%) (scfm)
<100,000d
< 100,000
>300e <2,000,000f
>150h Steady
50%' 300-100,000
1,000-100,000
<2,000
Molecular
Weight Vapor
Temp. (Ib/lb- Pressure Adsorptive
(°F) mole) Solubility (mm Hg) Properties
Must be able
100-200 45-130 to adsorb on/
desorb from
available
adsorbents
Must be readily
soluble in water
or other solvents
>10(at
room temp-
erature)
"Refers to the characteristics of the individual HAP if a single HAP is present and to that of the HAP mixture if a mixture of HAP's is
present.
bDetermined from HAP/hydrocarbon content.
°For emission streams that are mixtures of air and VOC; in some cases, the LELcan be increased to 40 to 50% with proper monitoring
and control (see Section 4.2 for definition of LEL).
dFor packaged units; multiple-package or custom-made units can handle larger flows.
6 Based on EPA's guidelines for 98% destruction efficiency.
'Units: Ib/hr. Source: Reference 14.
9 Applicable if such a unit is already available on site.
hTotal heat content.
'Relative humidity.
Figure 3-1. Percent reduction ranges for add-on control devices.
Thermal Incineration
T »95% T
T T
Catalytic Incineration
T »90% T
T * T
Carbon Adsorption
Absorption
Condensation
i i i
99%
95%
T 50%
T
T »90%T
T " T
I,,,
i i i
T
T 95% _ 99% _
T T * T
95% ^_ 98% T
T " T
50% 80% _ 95%
T * T *
10
20
50 100 200 300 500 1,000 2,000 3,000 5,000 10,000 20,000
Inlet Concentration, HAPe (ppmv)
24
-------
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 com-
pound (VOC) removal efficiency.
3.2.1.1 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 emis-
sion stream characteristics. Destruction efficiencies
up to 99+ percent are achievable with thermal in-
cineration. 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 mix-
ing during increased flow conditions decrease the
completeness of combustion. This causes the com-
bustion chamber temperature to fall, thus decreas-
ing the destruction efficiency.
Thermal incineration is typically applied to emis-
sion 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 (low-
er explosive limit) for the VOC in question (see
Section 4.2.2 for more details). Thus, if the VOC
concentration is high, dilution may be required.
When emission streams treated by thermal inciner-
ation are dilute (i.e., low heat content), supplemen-
tary fuel is required to maintain the desired com-
bustion 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 us-
ing the emission stream as fuel gas should be con-
sidered.
Packaged single unit thermal incinerators are avail-
able in many sizes to control emission streams with
flow rates from a few hundred up to about 100,000
scfm.
3.2.1.2 Catalytic Incinerators
Catalytic incinerators are similar to thermal inciner-
ators in design and operation except that the for-
mer employ a catalyst to enhance the reaction rate.
Since the catalyst allows the reaction to take place
at lower temperatures, 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 charac-
teristics and process conditions than is thermal in-
cinerator performance. Materials such as phospho-
rus, bismuth, lead, arsenic, antimony, mercury,
iron oxide, tin, zinc, sulfur, and halogens in the
emission stream can poison the catalyst and se-
verely affect its performance. (Note: Some cata-
lysts can handle emission streams containing halo-
genated 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. Cata-
lyst 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 main-
tained for 3 to 5 years before replacement of the
catalyst is necessary.
Catalytic incineration is generally less expensive
than thermal incineration 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 cata-
lyst.
Destruction efficiencies of up to 95 percent of
HAP's are generally achieved with catalytic inciner-
ation. Higher destruction efficiencies (99 percent)
are also achievable, but require larger catalyst vol-
umes and/or higher temperatures.
Catalytic incinerators have been applied to continu-
ous emission streams with flow rates up to about
100,000 scfm.
3.2.1.3 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 emis-
sion streams.
Flares can be used for controlling almost any VOC
emission stream. They can be designed and oper-
ated to handle fluctuations in emission VOC con-
tent, inerts content, and flow rate. There are several
different types of flares including steam-assisted,
air-assisted, and pressure head flares. Steam-as-
sisted flares are very common and typically em-
ployed in cases where large volumes of waste gas-
es 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 in-
creased as the gas flow increases.
Flaring is generally considered a control option
when the heating value of the emission stream
25
-------
cannot be recovered because of uncertain or inter-
mittent 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. De-
pending 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 million Ib/hr or more
for elevated flares. The capacity of an array of pres-
sure head flares depends on the number of flares in
the array.
3.2.1.4 Boilers/Process Heaters
Existing boilers or process heaters can be used to
control emission streams containing organic com-
pounds. These are currently used as control de-
vices for emission streams from several industries
(e.g., refinery operations, SOCMI reactor processes
and distillation operations, etc.)
Typically, emission streams are controlled in boil-
ers 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 manu-
facturing). Note that emission streams with low
heat content can also be burned in boilers or pro-
cess 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 efficien-
cies 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 de-
vices. 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 perfor-
mance of a boiler or process heater. By lowering
furnace temperatures, emission streams with large
flow rates and low heating values can cause incom-
plete combustion and reduce heat output. The per-
formance and reliability of the process heater or
boiler may also be affected by the presence of cor-
rosive compounds in the emission stream; such
streams are usually not destroyed in these devices.
26
3.2.1.5 Carbon Adsorbers
Carbon adsorption is commonly employed as a
pollution control and/or a solvent recovery tech-
nique. 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 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 charac-
terized by low volatility are strongly adsorbed on
carbon. The affinity of carbon for these compounds
makes it difficult to remove them during regenera-
tion 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 emis-
sion stream conditions. The presence of liquid or
solid particles, high boiling organics, or polymeri-
zable substances may require pretreatment proce-
dures such as filtration. Dehumidification is neces-
sary if the emission stream has a high humidity
(relative humidity > 50 percent) and cooling may
be required if the emission stream temperature ex-
ceeds 120° - 1 SOT.
To prevent excessive bed temperatures resulting
from the exothermic adsorption process and oxida-
tion reactions in the bed, concentrations higher
than 10,000 ppmv must frequently be reduced. This
is usually done by condensation or dilution of the
emission stream ahead of the adsorption step. Ex-
othermic reactions may also occur if incompatible
solvents are mixed in the bed, leading to polymer-
ization. If flammable vapors are present, the VOC
concentrations may be limited by insurance com-
panies to less than 25 percent of the LEL. If proper
controls and monitors are used, LEL levels up to 40
to 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.
-------
3.2.1.6 Absorbers (Scrubbers)
Absorption is widely used as a raw material and/or
a product recovery technique in separation and pu-
rification of gaseous streams containing high con-
centrations of VOC's. As an emission control tech-
nique, 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 prob-
lems 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 dis-
posed of in an environmentally acceptable manner.
Another factor that affects the suitability of absorp-
tion for organic vapor emissions control is the
availability of vapor/liquid equilibrium 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 avail-
able.
Another consideration involved in the application
of absorption as a control technique is disposal of
the absorber effluent (i.e., used solvent). If the ab-
sorber effluent containing the organic compounds
is discharged to the sewer, pond, etc., the air pollu-
tion problem is merely being transformed into a
water pollution problem. Hence, this question
should be addressed (e.g., are there chemical/phy-
sical/biological means for treating the specific efflu-
ent under consideration?). In solvent recovery,
used organic solvents are typically stripped (re-
verse 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 con-
centrations 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 out-
let concentrations will typically be required. Trying
to meet such requirements with absorption 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 combi-
nation with other control devices such as inciner-
ators.
Removal efficiencies in excess of 99 percent can be
achieved with absorption.
3.2.1.7 Condensers
Condensers are widely used as raw material and/or
product recovery devices. They are frequently ap-
plied as preliminary air pollution control devices
for removing VOC contaminants from emission
streams prior to other control devices such as in-
cinerators, adsorbers, or absorbers.
Condensers are also used by themselves for con-
trolling 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 rela-
tionship) 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 com-
monly 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 temperature of the cool-
ing water. Therefore, it is not possible for conden-
sation 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 tem-
peratures can be obtained with coolants such as
chilled water, brine solutions, or chlorofluorocar-
bons. However, for extremely low outlet HAP con-
centrations, condensation will usually be economi-
cally infeasible.
Depending on the type of condenser used, there
may be potential problems associated with the dis-
posal of the spent coolant. Therefore, using contact
condensers that generate such effluents for con-
trolling HAP emissions is not recommended.
Flow rates up to about 2,000 scfm can be consid-
ered as representative of the typical range for con-
densers used as emission control devices. Con-
densers for emission streams with flow rates above
2,000 scfm and containing high concentrations of
noncondensibles will require prohibitively large
heat transfer areas.
27
-------
3.2.2 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. Poten-
tial sources of the various inorganic vapors found
in the atmosphere are discussed in Chapter 2. Inor-
ganic HAP vapors typically include gases such as
ammonia, hydrogen sulfide, carbonyl sulfide, car-
bon 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 atmo-
sphere, these particulates are typically controlled
by methods that will be discussed in Section 3.3. In
this section, the discussion will be based on control
techniques for HAP's that are emitted as vapors to
the atmosphere.
Only a limited number of control methods are ap-
plicable to inorganic vapor emissions from point
sources. The two most commonly used control
methods are absorption (scrubbing) and adsorp-
tion. Absorption is the most widely used and ac-
cepted method for inorganic vapor control. Al-
though combustion can be used for some
inorganic HAP's (e.g., hydrogen sulfide, carbonyl
sulfide, nickel carbonyl), typical combustion meth-
ods 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.
Applicability of absorption and adsorption as con-
trol 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 nec-
essarily 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 inor-
ganic vapor emissions will be discussed.
3.2.2.1 Absorbers (Scrubbers)
Absorption is the most widely used recovery tech-
nique for separation and purification of inorganic
vapor emissions. The removal efficiency achiev-
able with absorbers can be greater than 99 percent.
It will typically be determined by the actual concen-
trations of the specific HAP in gas and liquid
streams and the corresponding equilibrium con-
centrations. Table 3-2 summarizes the reported ef-
ficiencies for various inorganic vapors employing
absorption as the control method.
28
As discussed in Section 3.2.1.6 for organic vapors,
the suitability of absorption for controlling inorgan-
ic vapors in gaseous emission streams is depen-
dent on several factors. The most important factor
is the solubility of the pollutant vapor in the sol-
vent. The ideal solvent should be nonvolatile, non-
corrosive, nonflammable, nontoxic, chemically sta-
ble, readily available, and inexpensive. Typical
solvents used by industry for inorganic vapor con-
trol 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
3.2.1.6).
Water is the ideal solvent for inorganic vapor con-
trol 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.
3.2.2.2 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 ad-
sorbing various inorganic vapors and gases. The
degree of adsorption is dependent not only on the
waste stream characteristics, but also on the differ-
ent characteristics of the adsorbents.
Carbon adsorption, using conventional and chemi-
cally impregnated carbons, is widely used for con-
trolling 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 sul-
fide. 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 3.2.1.5. Some of these
factors include the amount of adsorbent needed,
temperature rise of the gas stream due to adsorp-
tion, ease of regeneration, and the useful life of the
adsorbent. Most of the reported removal efficien-
cies for inorganic vapors are for activated carbon
and impregnated activated carbon, and range from
90 to 100 percent. Table 3-2 summarizes removal
efficiencies reported for various inorganic vapors
controlled by adsorption.
-------
Activated carbons are the most widely used adsor-
bents for inorganic vapor control. In several cases,
they must be treated (i.e., impregnated with chemi-
cals) for effective application. Since activated car-
bons 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 re-
quired depending on the emission stream condi-
tions. Filtration is used to prevent plugging of the
adsorber bed by any solids or particles which may
be in the emission stream. Ideal adsorption condi-
tions for impregnated activated carbons are rela-
tive humidities less than 50 percent and gas stream
temperatures below 130°F. Inorganic vapor con-
centrations are not recommended to exceed 1,000
ppmv (preferably, less than 500 ppmv) when acti-
vated carbon is used as an adsorbent.
3.2.3 Control Techniques for Organic/Inorganic
Vapor Emissions From Process Fugitive
Sources
Process fugitive emissions are defined in this hand-
book as emissions from a process or piece of
equipment that are being emitted at locations other
than the main vent or process stack. Process fugi-
tive emissions include fumes or gases which es-
cape 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 enclo-
sures containing an emissions source. An example
would be a vent fan on a perchloroethylene dry
cleaner or the vent fan on a press room. Other
examples of process fugitive sources include cool-
ing 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 ac-
cess or maintenance), the opening through which
emissions escape cannot be totally enclosed or
blanked off. Operators have to access the equip-
ment 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 knowl-
edge of the process or operation so that the most
effective hood or enclosure can be installed to pro-
vide minimum exhaust volumes for effective conta-
minant 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 de-
signed using the capture velocity principle which
involves creation of an air flow past the source of
containment sufficient to remove the highly con-
taminated 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 dis-
tances (on the order of inches) when thrown or
emitted from a source and therefore can be as-
sumed 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 con-
sidered is the capture velocity. Standard design
values of capture velocity are available from the
American Conference of Government Industrial Hy-
Table 3-2. Current Control Methods for Various Inorganic Vapors (1)
Absorption
Calcium Fluoride (CaF2)
Silicon Tetrafluoride (SiF4)
Hydrogen Fluoride (HF)
Hydrogen Bromide (HBr)
Titanium Tetrachloride (TiCI4)
Chlorine (CI2)
Hydrogen Cyanide (HCN)
Adsorption
Inorganic Vapor
Mercury (Hg)
Hydrogen Chloride (HCI)
Hydrogen Sulfide (H2S)
Reported
Removal
Efficiency (%)
95
95
98
Solvent
Brine/Hypochlorite
solution
Water
Sodium carbonate/Water
Reported
Removal
Efficiency (%)
90
100
Adsorbent
Sulfur-impregnated
activated carbon
Ammonia-impregnated
95
95
85-95
99.95
99
90
Water
Water
Water
Water
Water
Alkali solution
99
activated carbon
Calcined alumina
Ammonia-impregnated
activated carbon
29
-------
gienists in the Industrial Ventilation Manual (see
Table 3-3).(2) Once a capture velocity has been de-
termined, the volume of air required should be
based on maintaining this capture velocity at the
emissions point furthest from the hood. This cap-
ture velocity should be sufficient to overcome any
opposing air currents. (For additional information
on hood design guidelines for several industries,
see Appendix A.7, reference 3, or reference 2.)
Very few measurements of hood capture efficiency
have been conducted.(4,5,6) Hood capture efficien-
cies of between 90 and 100 percent are possible
depending on the situation and the particular pro-
cess fugitive sources being controlled. For sources
where operator access is not needed and where
inspection doors can be provided, efficiencies to-
ward the upper end of the range are achievable. For
sources where emissions are more diffused, for
example, from printing presses, capture efficien-
cies 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.(4) In the publication rotogravure industry,
capture efficiencies of 93 to 97 percent have been
demonstrated based on material balances.(B)
Once the process fugitive emissions are captured,
the selection of the control device will be depen-
dent on the emission stream characteristics, HAP
characteristics, and the required overall perfor-
mance levels (e.g., removal efficiency). Note that
the required performance level for the control de-
vice will be determined by the capture efficiency.
The factors that affect the control device selection
process are the same as for point sources; there-
fore, refer to Sections 3.2.1 and 3.2.2.
For process fugitive emission sources such as pro-
cess drains, the control alternatives involve a clo-
sure 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 com-
pletely 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, diame-
ter of the drain, and ambient atmospheric condi-
tions. Emission reductions of 40 to 50 percent may
be achieved with water-sealed drains.(7) In a com-
pletely closed drain system, the system may be
pressured and purged to a control device to effec-
tively capture all emissions. The control efficiency
would then depend on the efficiency of the control
device; 95 percent should be achievable.(7)
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.(8) Prob-
ably the best control technique currently available
is close monitoring of heat exchangers and other
equipment to detect small leaks as they occur.
3.2.4 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 exten-
sively studied than inorganic fugitive emissions.
Therefore, the following discussion will be primar-
ily based on organic vapors. However, control tech-
niques for inorganic vapor emissions will also be
discussed.
Table 3-3. Range of Capture Velocities (2)
Condition of
Contaminant Dispersion
Released with practically no velocity into quiet air
Released at low velocity into moderately still air
Examples
Evaporation from tanks; degreasmg, etc.
Spray booths; intermittent container filling;
Capture
Velocity (fpma)
50-100
100-200
Active generation into zone of rapid air motion
Released at high initial velocity into zone of very rapid
air motion
low speed conveyor transfers; welding;
plating; pickling
Spray painting in shallow booths; barrel fill-
ing; booths; barrel filling; conveyor load-
ing; crushers
Grinding; abrasive blasting; tumbling
200-500
500-2,000
In each category above, a range of capture velocity is shown
Lower End of Range:
- Room air currents minimal or favorable to capture.
- Contaminants of low toxicity or of nuisance value only.
- Intermittent, low production.
- Large hood-large air mass in motion.
Upper End of Range:
- Disturbing room air currents.
- Contaminants of high toxicity.
- High production, heavy use.
- Small hood-local control only.
The proper choice of values depends on several factors:
afpm = feet per minute.
30
-------
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 environment. Control techniques for
equipment leaks include leak detection and repair
programs and equipment installation or configura-
tion. The following sections contain information
about control techniques for common types of pro-
cessing equipment found in plants processing or-
ganic materials. Control techniques and control ef-
ficiencies for common types of processing
equipment are summarized in Table 3-4.
Table 3-4. Summary of Control Effectiveness for Controlling
Organic Area Fugitive Emission Sources (9)
Control Technique Control
Emission Source Equipment Effectiveness (8)
Modification (%)
Pumps Monthly leak detection
and repair 61
Sealless pumps 100
Dual mechanical seals 100
Closed vent system8 100
Valves
- Gas Monthly leak detection
and repair 73
Diaphragm valves 100
- Light liquid Monthly leak detection
and repair 46
Diaphragm valves 100
Pressure Relief Rupture disk 100
Valves Closed vent system3 100
Open-ended Lines Caps, plugs, blinds 100
Compressors Mechanical seals with
vented degassing
reservoirs 100
Closed vent system8 100
Sampling Connections Closed purge sampling 100
"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.
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 applica-
tion.
Sophisticated pump seals can also be used to cap-
ture or eliminate fugitive emissions. Dual seal sys-
tems 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 virtu-
ally 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 de-
tection and repair programs. A leak detection and
repair system modeled after the one EPA devel-
oped for the New Source Performance Standards
(NSPS) of the synthetic organic chemical industry
(SOCMI) should achieve about 60 percent control
efficiency.(9) 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 tenden-
cy to leak. These factors and their effect on control
efficiency have been studied and discussed in re-
ferences 9, 10, 11 and 12. Also, models are avail-
able for calculating the effectiveness of leak detec-
tion and repair programs (e.g., see reference 13).
Valves—As with pumps, control of fugitive emis-
sions from valves may be accomplished by install-
ing equipment designed to isolate the process fluid
from the environment. But also as with pumps,
leakless valves such as diaphragm valves are limit-
ed in their application.
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 ten-
dency to leak, and other factors. A leak detection
and repair program modeled after the one devel-
oped 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 for
valves in light liquid service.(9)
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 effi-
ciency, then, depends on the destruction efficiency
of the flare. If flares are operated in accordance
with flare requirements recently established 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.
31
-------
Compressors—Fugitive emissions from compres-
sor seals can be controlled by venting the seal area
to a flare or other control device. Barrier fluid sys-
tems can also be used to purge the seal area and
convey leakage to a control device. Capture effi-
ciency should be 100 percent and, therefore, the
overall control efficiency would depend on the effi-
ciency of the control device.
Sampling Connections—Fugitive emissions from
sampling connections can be controlled by return-
ing the purged material to the process or by dispos-
ing of it in a control device. The practice of return-
ing purged material to the process in a closed
system should achieve almost 100 percent control
efficiency. The control efficiency achieved by di-
verting the collected purge material to a control
device depends on the efficiency of the device.
Other area fugitive emission sources include la-
goons and ponds where liquid waste streams con-
taining organic compounds are disposed of. Emis-
sions and emission rates of organic vapors from
such sources are not well documented, and such
sources are not easily controlled. The best method
currently available for reducing emissions from la-
goons and ponds is enhancing upstream treatment
processes, thereby minimizing the amount of or-
ganic material reaching 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 process-
ing organic chemicals. Therefore, plants process-
ing highly volatile compounds such as hydrogen
chloride or ammonia would be expected to benefit
by the same control techniques 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 tech-
niques that employ a control device to treat collect-
ed vapors, the control device will probably differ.
Instead of a combustion device, an absorber, con-
denser, 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 de-
veloped 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 com-
pounds 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 treat-
ment processes.
3.2.5 Control Device Selection for a Hypothetical
Facility
This subsection illustrates the control device selec-
tion process discussed in the previous sections for
a hypothetical facility with several emission
streams. Assume that the owner/operator of this
facility has requested assistance regarding the con-
trol 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 is 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.1-1); hence, the concentration limit indicat-
ed 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 Ta-
ble 3-1.
Emission Stream 2 (see Figure 3-3)—Assume the
HAP control requirement for Emission Stream 2 is
95 percent reduction. For this level of performance,
the applicable control techniques for inlet concen-
trations of -500 ppmv are thermal incineration,
catalytic incineration, and absorption. If either of
the incineration techniques are applied, the con-
centration 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.1-1).
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 con-
trol technique should also be based on design crite-
ria (Chapter 4) and costs (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 HAP
concentration falls outside the range indicated in
Figure 3-1; therefore, none of the control devices in
this figure are applicable. According to Section
3.2.1.3, 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 pro-
cess 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 is
95 percent reduction. For this level of performance,
the applicable control techniques for inlet concen-
32
-------
Figure 3-2. Effluent characteristics for emission stream #1.
HAP EMISSION STREAM DATA FORM*
Cot
Loc
A.
B.
C.
D.
E.
F.
G.
H.
I.
J.
K.
L.
M.
N.
O.
U.
V.
W.
-npany Glaze Chemical Company
•atinn (StmPt) 87 Octane Drive
inty) Somewhere
(State, 7ip)
Emission Stream Number/Pi
HAP Emission Source
Source Classification
Emission Stream HAP's
HAP Class and Form
HAP Content (1,2,3)**
HAP Vapor Pressure (1,2)
HAP Solubility (1,2)
HAP Adsorptive Prop. (1,2)
HAP Molecular Weight (1,2)
Moisture Content (1,2,3)
Temperature (1,P,3)
Flow Rate (1,2r3)
Pressure (1,2)
Halogen/Metals (1 7)
Applicable Regulation(s)
RpquirfH r.nntrnl LPVP!
Selected Control Methods
Plant Contact Mr- J°nn
Leake
(999) 555-5024
Anpnrw Tnntar.t Mr. Efrem Johnson
ant Identification #1 /
(a) paper coating oven
(a) process point
(a) toluene
(a) orqanic vapor
(a) 73 ppmv
(a) 28.4 mm Hg 3 77°F
(a) insoluble in water
(a) provided
(a) 92 Ib/lb-mole
2% vol.
12C°F
15,000 scfm (max)
atmospheric
none / none
No.
#3
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
P.
Q.
R.
S
T.
. of Emission Streams Ur
Oven Fxhaust
-
-
-
-
-
-
-
-
-
Organic Content (1)***
Heat/02 Content (1)
Particulate Content (3)
Particle Mean Diam (3)
Drift Velocity/S03 (3)
iHpr Rpview
(r)
(r-)
(c)
(c)
(<•-)
(c)
(^)
(c-)
(c)
44 ppmv CH4, 4 comv ot
0.4 Btu/scf / 20.6% vol.
/
*The data presented are for an emission stream (single or combined streams) prior to entry into the selected control
method(s). Use extra forms if additional space is necessary (e.g., more than three HAP's). and note this need.
**The numbers in parentheses denote what data should be supplied depending on the data in steps "C" and "E":
1 = organic vapor process emission
2 = inorganic vapor process emission
3 = particulate process emission
""Organic emission stream combustibles less HAP combustibles shown on Lines D and F.
trations of -1,000 ppmv are thermal incineration,
catalytic incineration, carbon adsorption, and ab-
sorption. If either of the incineration methods or
carbon adsorption is 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. 1-1). The flow rate of Emis-
sion Stream 4 falls in the range indicated as appli-
cable in Table 3-1 for these control techniques. The
final selection of the control technique should also
be based on design criteria (Chapter 4) and costs
(Chapter 5).
Emission Stream 5 (see Figure 3-6)—Assume the
HAP control requirement for Emission Stream 5 is
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 absorp-
tion. 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 is
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 concen-
trations of —13,000 ppmv are absorption and con-
densation. The final selection among these tech-
niques should also be based on design criteria
(Chapter 4) and costs (Chapter 5).
3.3 Particulate Emissions Control
Section 3.3.1 discusses add-on paniculate control
devices and presents guidelines that are used to
determine the applicability of each control device.
Section 3.3.2 discusses control techniques that are
used to reduce fugitive particulate emissions.
3.3.1 Control Techniques for Particulate Emissions
from Point Sources
Three types of control devices applicable to parti-
culate-laden emission streams from point sources
are discussed below: fabric filters (baghouses),
electrostatic precipitators (ESP), and venturi scrub-
bers. The control efficiencies and applicability of
these devices depend on the physical and chemi-
33
-------
Figure 3-3. Effluent characteristics for emission stream #2.
HAP EMISSION STREAM DATA FORM*
Cor
Loc
A
B.
C.
D.
E.
F.
G.
H.
1.
J.
K.
L.
M.
N.
0.
U.
V.
W.
npany Glaze Chemical
Company
atmn(StrPPt) 87 Octane Drive
(Pity) Somewhere
(State, 7ip)
Emission Stream Number/P
HAP Emission Source
Source Classification
Emission Stream HAP's
HAP Class and Form
HAP Content (1,2,3)**
HAP Vapor Pressure (1,2)
HAP Solubility (1,2)
HAP Adsorptive Prop. (1,2)
HAP Molecular Weight (1,2)
Moisture Content (1,2,3)
Temperature (1,2,3)
Flow Rate (1,2,3)
Pressure (1,?)
Halogen/Metals (1,7)
Applicahlp Rpgnlation(s)
Rpquirpri Control I evpl
Selected Control Methods
Plant Contact M>". John
Leake
To|oph^ne M^ (999) 555-5024
Anpnrv Contact Mr- Efrem
Johnson
No nf Fmksinn Strpamc; I Inripr Rpvipvu
lant
(a)
(a)
(a)
(a)
(a)
(a)
(a)
(a)
(a)
Identification #2 /
metal coating oven
process point
toluene
organic vapor
500 ppmv
28.4 mm Hg @ 77° F
insoluble in water
provided
92 Ib/lb-mole
2% vol.
120°F
20,000 scfm (max)
atmospheric
none / none
assume 95% removal
thermal incineration,
#1 Oven Exhaust
(h)
(b)
(b)
(b)
(h)
_. (b)
(h)
(h)
(b)
P. Organic Content (1)***
Q. Heat/02 Content (1)
R Partir.nlatP TontPnt (-^)
S. Particle Mean Diam. (3)
T Drift Velnfity'^n3 (i)
H
(c)
(c*
(r.\
(c)
(c)
(c)
none
2.1 Btu/scf / 20.6% vol.
1
catalytic incineration, absorption
*The data presented are for an emission stream (single or combined streams) prior to entry into the selected control
method(s). Use extra forms if additional space is necessary (e.g., more than three HAP's). and note this need.
**The numbers in parentheses denote what data should be supplied depending on the data in steps "C" and "E":
1 =organic vapor process emission
2 = inorganic vapor process emission
3 = paniculate process emission
**Organic emission stream combustibles less HAP combustibles shown on Lines D and F.
cal/electrical properties of the airborne participate
matter under consideration. Brief descriptions of
each of these control devices appear in the subsec-
tions that follow.
Selection of the control devices themselves de-
pends on the specific stream characteristics and
the parameters (e.g., required collection efficiency)
that affect the applicability of each control device.
Table 3-5 identifies some key emission stream
characteristics that affect the applicability of each
device. Matching the characteristics of the emis-
sion stream under consideration with the corre-
sponding information presented in Table 3-5 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 3-5 are
given as typical guidelines and should not be taken
as absolute, definitive values. Gas stream pretreat-
ment equipment can be installed upstream of the
control device (i.e., cyclones, precoolers, pre-
heaters) which enables the emission stream to fall
within the parameters outlined in Table 3-5.
The temperature of the emission stream should be
within 50 to 100°F above its dew point if the emis-
sion stream is to be treated (i.e., particulate matter
collected) by an ESP or a fabric filter. If the emis-
sion stream temperature is below this range, con-
densation can occur; condensation can lead to cor-
rosion of metal surfaces, blinding and/or
deterioration of fabric filter bags, etc. If the emis-
sion stream is above this range, optimal HAP col-
lection 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.1. (For discussions of
gas stream pretreatment equipment, see reference
15 and Appendix B.11 of reference 3.)
Table 3-6 identifies general advantages and disad-
vantages for each particular control device. Table
3-6 is used to provide additional information on
other considerations that, while not necessarily af-
fecting the technical feasibility of the control device
for the stream, may affect the overall desirability of
34
-------
Figure 3-4. Effluent characteristics for emission stream #3.
HAP EMISSION STREAM DATA FORM*
Coi
Loc
A.
B.
C.
D.
E.
F.
G.
H.
1.
J.
K.
L.
M.
N.
0.
mpany Glaze Chemical Company
-atinn (Street) 87 Octane Drive
(Pity) Somewhere
(State 7ip)
Emission Stream Number/Pi
HAP Emission Source
Source Classification
Emission Stream HAP's
HAP Class and Form
HAP Content (1,2,3)**
HAP Vapor Pressure (1,2)
HAP Solubility (1,2)
HAP Adsorptive Prop. (1,2)
HAP Molecular Weight (1,2)
Moisture Content (1,2,3)
Temperature (1,7,3)
Flow Rate (1.2.3)
Pressure (1,2)
Halogen/Metals (1 7)
Plant Contact Mr. John Leake
Toi-nh^M,, (999) 555-5024
Anenr-u Pnntart Mr. Efrem Johnson
lant Identification #1 /
(a) absorber vent
(a) process point
(a) methylene chloride
;ai organic vapor
(a) 44.000 ppmv
(a) 436 mm Hq @ 77°F
(a) insoluble in water
(a) not given
(a) 85 Ib/lb-mole
none
100°F
30,000 scfm (expected)
atmospheric
none / none
No. of Emission Streams U
nder Review
6
Arptaldphyde Manufacturi na Absorber Vent
-- (b)
(b)
(hi
(b)
(h)
(b)
(h)
(h)
(b)
P. Organic Content (1)***
Q. Heat/02 Content (1)
R. Particulate Content (3)
S. Particle Mean Diam (3)
T. Drift Velocitv/SO-, (3)
(c)
(c)
(c)
(c)
(c)
(f)
(c)
17.8% vol. C
180 Btu/scf
-
-
-
-
-
-
-
-
-
Hd
/ none
/
U. Applicable Regulation(s)
V. Required Control Level —
W. Selected Control Methods
assume 98% removal
flare, boiler, process heater
*The data presented are for an emission stream (single or combined streams) prior to entry into the selected control
method(s). Use extra forms if additional space is necessary (e.g., more than three HAP's). and note this need.
**The numbers in parentheses denote what data should be supplied depending on the data in steps "C" and "E":
1 =organic vapor process emission
2 = inorganic vapor process emission
3 = paniculate process emission
'"Organic emission stream combustibles less HAP combustibles shown on Lines D and F.
Table 3-5. Key Characteristics for Particulate Emission Streams
Control
Device
Baghouse
ESP
Achievable
Efficiency
Range
Up to 99 + %
Up to 99 + %
Particle
Size
Limitation
Least efficient with
particles 0.1 \i.m to
0.3 (xm diameter.
Generally least
efficient with
Temperature
Dependent
on fiber type
but not
exceeding
550°F
without a
precooler
Generally up
to 1,000°F
Corrosiveness/
Resistivity
Special fiber types necessary to
resist corrosion.
Corrosion resistant materials
required. May require
Moisture
Content
Poor efficiency with
emission streams of high
moisture content, very
sensitive to changes in
moisture content of an
emission stream.
Can control streams with
relatively high moisture
Venturi
Up to 99 + %
particles ranging in
size from 0.2 jj.m to
0.5 (Am diameter.
Generally operates
best with particles
>0.5 |j,m diameter.
No general
limitations
conditioning agents for highly
resistive particles. Additionally,
ESP's are not used to control
organic matter since this
constitutes a fire hazard.
Special construction may be
required for corrosive emission
streams.
content (i.e., 34%vol) if so
designed, but sensitive to
moisture changes of an
emission stream.
Not sensitive to changes in
moisture content of
emissions stream.
35
-------
Figure 3-5. Effluent characteristics for emission stream #4.
HAP EMISSION STREAM DATA FORM*
Cor
Loc
A.
B.
C.
D.
E.
F.
G.
H.
1.
J.
K.
L.
M.
N.
O.
U.
V.
W.
npany Glaze Chemical Company
•atinn fStrPPt) 87 Octane Drive
(Pity) Somewhere
(Rtatp 7ip)
Emission Stream Number/Pi
HAP Emission Source
Source Classification
Emission Stream HAP's
HAP Class and Form
HAP Content (1,2,3)**
HAP Vapor Pressure (1,2)
HAP Solubility (1,2)
HAP Adsorptive Prop. (1,2)
HAP Molecular Weight (1,2)
Moisture Content (1,2,3)
Tp.mppratnrp (1,2,3)
Flnw Ratfi (1,2,3)
Prpccurp (1 2)
Hq|ngpn/MRtals(1,9)
Applicable Regulation(s)
Required Control I PVP!
Selected Control Methods
Plant Contact Mr- John
Leake
T0,o.h.np Nn (999) 555-5024
Agpnr.y Contact Mr. Efrem Johnson
Nn nf Fmissinn Strpams I InHpr Rpvipuv
ant IHpntifiratinn #4 i
(a) printing press
(a) process point
(a) toluene
la^ organic vapor
la) 1,000 DDmv
(a) 28.4 mm Hg @ 77"F
(a) insoluble in water
(a) provided
(a) 92 Ib/lb-mole
40% rel. humiditv
90° F
15,000 scfm (max)
atmospheric
none / none
assump 9K rpmnval
thermal incineration,
' #1 Printing Press Exhaust
(b)
(h)
(b)
(b)
(h)
(h)
(h)
(h)
(h)
P. Organic Content (1)***
Q. Heat/02 Content (1) 4'
R Partir.iilatfi T.nntpnt (3)
S Partirle Mean Piam (3)
T Drift Velocity/S03 (3)
catalytic incineration, carbon
(r)
(0
(c)
(c)
(r.)
(^)
(r)
(<0
(c)
none
2 Btu/scf / 20.6% vol.
/
adsorption, absorption
'The data presented are for an emission stream (single or combined streams) prior to entry into the selected control
method(s). Use extra forms if additional space is necessary (e.g., more than three HAP's). and note this need.
'The numbers in parentheses denote what data should be supplied depending on the data in steps "C" and "E":
1 = organic vapor process emission
2 = inorganic vapor process emission
3 = paniculate process emission
'Organic emission stream combustibles less HAP combustibles shown on Lines D and F.
using the device for a given emission stream. Thus,
Tables 3-5 and 3-6 used in conjunction provide
guidelines to determine if a particular control de-
vice 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 com-
plete technical evaluation of the applicability of
these devices to an emission stream.
3.3.1.1 Fabric Filters
Fabric filters, or baghouses, are an efficient means
of separating particulate matter entrained in a gas-
eous stream. A fabric filter is typically least efficient
collecting particles in the range of 0.1 to 0.3 (im
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 4.9, pulse-jet cleaned filters
are not as efficient as those cleaned by other meth-
ods, and emissions are not as constant over a filtra-
tion cycle as those from filters using the other two
cleaning methods. However, pulse-jet cleaning is
widely used in general industrial fabric filter appli-
cations and, therefore, Section 4.9 does include
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 baghouses.
Fabric filters using mechanical shaking or reverse
air cleaning are fundamentally different from ESP's
and venturi scrubbers in that they are not "efficien-
cy" devices. A properly designed and operated fab-
ric filter using one of these two cleaning methods
will yield a relatively constant outlet particle con-
centration, regardless of inlet load changes. The
typical outlet particle concentration range is be-
tween 0.003 to 0.01 grains/scf (gr/scf), averaging
36
-------
Figure 3-6. Effluent characteristics for emission stream #5.
HAP EMISSION STREAM DATA FORM'
Cor
Loc
A.
B.
C.
D.
E.
F.
G.
H.
1.
J.
K.
L.
M.
N.
0.
U.
V.
W.
-npany Glaze Chemical Company
-atinn (StmPt) 87 Octane Drive
ir.ity) Somewhere
(State, 7ip)
Emission Stream Number/Pi
HAP Emission Source
Source Classification
Emission Stream HAP's
HAP Class and Form
HAP Content (1,2,3)**
HAP Vapor Pressure (1,2)
HAP Solubility (1,2)
HAP Adsorptive Prop. (1,2)
HAP Molecular Weight (1,2)
Moisture Content (1,2,3)
Tpmppraturp (1,7,3)
Flow Rate (1.2.3)
Prpssnrp (1,2)
Halngpn/Mptals(1 9)
Applicable Regulation(s)
Required Control Level
Selected Control Methods
Plant Contact Mr- J°hn
Leake
Telephone No <999> 555-5024
Anpnrv Tnntar.t Mr. Efrem
Johnson
Nn of Fmissinn Strpamc I Indpr Rpuipuu
lant IHpntifir.atinn #5 /
(a) evaporator off-qas
(a) process point
(a) ammonia
(.,) inorganic vapor
(a) 20,000 ppmv
(a) 8.46 atm. @ 68°F
(a) provided
(a) not given
la) 17 Ib/lb-mole
2% vol.
85° F
3,000 scfm (max)
atmospheric
none / none
assume 98% removal
absorption
Urea Fvapnratnr Off-ga^ Fxhaust.
(h)
(h)
(b)
(b)
(h)
(b)
(h)
(h)
(b)
P. Organic Content (1)***
R. Particulate Content (3)
S Particle Mean Diam (3)
T Drift Ve|nrity/Rn3 (3)
(r)
(c)
(c)
(p)
(r)
(^)
(c)
/
/
*The data presented are for an emission stream (single or combined streams) prior to entry into the selected control
method(s). Use extra forms if additional space is necessary (e.g., more than three HAP's). and note this need.
**The numbers in parentheses denote what data should be supplied depending on the data in steps "C" and "E":
1 = organic vapor process emission
2 = inorganic vapor process emission
3 = paniculate process emission
** Organic emission stream combustibles less HAP combustibles shown on Lines D and F.
around 0.005 grains/scf. These numbers can be
used to ascertain an expected performance level
(see the following Example Case). 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 perfor-
mance (i.e., air-to-cloth ratio, cleaning mechanism,
fabric type) are discussed in detail in Section 4.9.
Fabric filters are sensitive to emission stream tem-
perature and a precooler or preheater may be re-
quired, as discussed previously. Fabric filters oper-
ate 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.
3.3.1.2 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 fac-
tors that affect the maximum electrical power (volt-
age) 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 veloc-
ity which will decrease the overall collection effi-
ciency. Electrostatic precipitators are discussed fur-
ther in Section 4.10.
3.3.1.3 Venturi Scrubbers
Venturi scrubbers use an aqueous stream to re-
move particulate matter from an emissions stream.
The performance of a venturi scrubber is not affect-
ed by sticky, flammable, or corrosive particles. Ven-
turi scrubbers are more sensitive to particle size
distribution than either ESP or fabric filters. In gen-
eral, venturi scrubbers perform most efficiently for
particles above 0.5 |xm in diameter (see Section
4.11 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 collec-
tion efficiencies contributes to high operating
costs.
37
-------
Example Case
Assume a facility is required to achieve an emis-
sion limit for particulate emissions from a mu-
nicipal waste incinerator. The emissions stream
particles consist primarily of fly ash; however,
10 percent of these particles is a HAP: cadmium.
The characteristics of the emission stream after
exiting a heat exchanger are shown in Figure 3-
8. From Figure B.1-1 (Appendix B), the dew point
of an emission stream containing 200 ppmv S03
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 concen-
tration:
0.005 gr v 110,000 acf
530°R-scf
x
scf
60 min-lb
mm
(400 + 460)°R-acf
I b
hr
7,000 gr-hr
To calculate the HAP outlet emission rate:
2.9 -x 0.10 = 0.29
hr
lbHAP
hr
Ib
110,000 acf 3.2 gr
min acf 7,000 gr
- 3'017 HF
60 min
hr
3,017 -j^x (1 - 0.999) = 3.017 -]£-
Since the HAP constitutes 10 percent of the total
particulate matter, the outlet concentration of
the HAP is:
0.10 x 3.017 -= 0.3017
lb
This value assumes that the HAP is in particulate
form and that it is collected as efficiently 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:
This value again assumes that the HAP is in
particulate form and that it is collected as effi-
ciently 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, pro-
vided the particle resistivity is not "high." The
drift velocity of the particles (0.30 ft/sec) is indic-
ative of particles with an "average" resistivity;
therefore, an ESP can probably be used to con-
trol this stream and, thus, it also is an appropri-
ate control technique for this emission stream.
A venturi scrubber has difficulty controlling par-
ticles below 0.5 |xm 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 appropriate control techniques for
this emission stream. To determine the basic
design parameters and actual applicability of
each control device, Section 4.9 (Fabric Filter),
Section 4.10 (ESP's), and Section 4.11 (Venturi
Scrubbers) must be examined.
3.3.2 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 2.2. The meth-
ods used to control process sources of fugitive par-
ticulate emissions are generally different from
those applied to area sources. Basically, process
fugitive sources can employ conventional mea-
sures (i.e., capture techniques and add-on control
devices) while area fugitive sources either cannot
use conventional measures or the use of conven-
tional measures is precluded due to cost. For exam-
ple, 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 tech-
niques rather than capture/control techniques.
Section 3.2.3 discusses methods of hooding and
capture of process emissions. The remaining fugi-
tive particulate emission control methodologies
(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 tech-
niques that can be applied to general process and
area fugitive particulate emission sources (i.e.,
sources common to many industries). (See Appen-
dix A.8, reference 3, for industry-specific and
source-specific information for fugitive particulate
emission control and Appendix A.9, reference 3, for
information on chemical dust suppressants.)
38
-------
Figure 3-7. Effluent characteristics for emission stream #6.
HAP EMISSION STREAM DATA FORM*
Cor
Loc
A
<"-\.
B.
C.
D.
E.
F.
G.
H.
1.
J.
K.
L.
M.
N.
0.
npany Glaze Chemical Company
:atmn (Street) 87 Octane Drive
irityl Somewhere
(State, 7ip)
Emission Stream Number/Pi
HAP Emission Source
Source Classification
Emission Stream HAP's
HAP Class and Form
HAP Content (1,2,3)**
HAP Vapor Pressure (1,2)
HAP Solubility (1,2)
HAP Adsorptive Prop. (1,2)
HAP Molecular Weight (1,2)
Moisture Content (1,2,3)
Temperature (1,7, 3)
Flow Rate (1,2,3)
Pressure (1.2)
Halogen/Metals (1 7)
Plant Contact Mr- John
Leake
(999) 555-5024
Agenry Cnntar.t Mr. Efrem Johnson
Nn nf Fmkcinn Streams I Inrier Review;
lant
(a)
(a)
la)
(a)
(a)
(a)
(a)
(a)
(a)
Identification #6 /
condenser vent
process point
styrene
organic vapor
13,000 ppmv
provided
insoluble in water
not given
104 Ib/lb-mole
negl igible
90°F
2,000 scfm (max)
atmospheric
none / none
Styrene Recovery Condenser Vent
(b)
(h)
(hi
(b)
(h)
(h)
(h)
(h)
(h)
P. Organic Content (1)***
Q. Heat/02 Content (1) 61>
R. Particulate Content (3)
s Partible Mean Diem (3)
T. Drift Velocitv/SOn (3)
If)
(<••)
(r\
(c)
-(f-)
- (c)
(f)
(r)
(c)
none
5 Btu/scf / 20.7% vol.
/
U. Applicable Regulation(s)
V. Required Control Level
W. Selected Control Methods
assume 90% removal
absorption, condensation
*The data presented are for an emission stream (single or combined streams) prior to entry into the selected control
method(s) Use extra forms if additional space is necessary (e.g., more than three HAP's). and note this need.
**The numbers in parentheses denote what data should be supplied depending on the data in steps "C" and "E":
1 = organic vapor process emission
2 = inorganic vapor process emission
3 = paniculate process emission
**0rganic emission stream combustibles less HAP combustibles shown on Lines D and F.
An extensive review of available literature on fugi-
tive emissions revealed that one reference included
almost all necessary information pertinent to the
scope of this handbook. Consequently, most of the
following subsections are taken directly from the
following document: Technical Guidance for Con-
trol of Industrial Process Fugitive Particulate Emis-
s/"ons.(16) An April 1985 draft final document from
EPA on fugitive emissions (17) 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.
Throughout the discussions, control efficiencies
are stated for many of the control techniques. It is
important to note that the efficiency values are esti-
mates. The ability to quantify accurately the emis-
sion rates from a fugitive emission source has not
yet been fully realized.
3.3.2.1 Process Fugitive Particulate Emission
Control
Control of HAP process fugitive emissions may be
accomplished by capturing the paniculate 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 eliminat-
ed 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 oper-
ational procedures, movable hoods may be a via-
ble alternative. Movable hoods can be placed over
the fugitive emissions source as the production
cycle permits. For example, movable hoods can be
39
-------
Table 3-6. Advantages and Disadvantages of Particulate Control Devices
Advantages
Disadvantages
Baghouse —Very efficient at removing fine participate matter from
a gaseous stream; control efficiency can exceed 99%
for most applications.
—Lower pressure drop than venturi scrubber when
controlling fine particulates; i.e., 2" to 6" H20 compared
with a40" H20.
—Can collect electrically resistive particles.
—With mechanical shaking or reverse air cleaning,
control efficiency is generally independent of inlet
loading.
—Simple to operate.
ESP —Can control very small (<0.1 (im) particles with high
efficiency.
—Low operating costs with very low pressure drop
(0.5" H20).
—Can collect corrosive or tar mists.
—Power requirements for continuous operation are low.
—Wet ESP's can collect gaseous pollutants.
Venturi —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 generally required upstream if
significant amounts of large particulates (>20 n-m) are
present.
—Needs special or selected fabrics to control corrosive
streams.
—Least efficient with particles between 0.1 n.m to 0.3 |xm
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" H2O
or greater), particularly for smaller (<1 n-m) particles.
—Has wastewater and cleaning/disposal costs.
—Least efficient with particles less than 0.5 jim diameter.
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, where-
by a movable hood follows quench cars during
coke pushing. Another alternative 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 emis-
sion 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.
3.3.2.2 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 particu-
late emissions are generally from spillage and me-
chanical agitation of the material at uncovered
transfer points. However, emissions from inad-
equately enclosed systems can be quite extensive.
Table 3-7 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 con-
veyor feed and discharge points, with some appli-
cations at conveyor transfer points. Wet suppres-
sion with water only is a relatively inexpensive
technique; however, it has the inherent disadvan-
tage 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 50 |xm diameter)
break the surface of the bubbles in the foam when
they come in contact, thereby wetting the particles.
Particles larger than 50 |xm 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 in-
jected into free-falling aggregate at a transfer point,
the mechanical motion provides the required parti-
cle 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, al-
lowing 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
40
-------
Figure 3-8. Effluent characteristics for a municipal incinerator emission stream.
HAP EMISSION STREAM DATA FORM*
Cor
Loc
A.
B.
C.
D.
E.
F.
G.
H.
1.
J.
K.
L
M.
N.
0.
-npany Incineration, Inc.
•af,nn'(StrPPt! 123 Main Street
irityl Somewhere
(Rtatp, 7ip)
Emission Stream Number/Pi
HAP Emission Source
Source Classification
Emission Stream HAP's
HAP Class and Form
HAP Content (1,2,3)**
HAP Vapor Pressure (1,2)
HAP Solubility (1,2)
HAP Adsorptive Prop. (1,2)
HAP Molecular Weight (1,2)
Mnisttirp Content (1 7 3)
Temperature (1,? 3)
Flow Rate (1.2.3)
Prpssum (1,2)
Haloqen/Metals(1,2)
Plant Cnntar.t Mr. Phil
Brothers
TplPnhoneNo. (999) 555-5624
Anonrv, Pnntart Mr. Ben Hold
No. of Emission Streams Ur
iripr Rpvifiw
1
(ant |Hpntifi<-ptinn #1 / Incineration Exhaust
(a) municipal incinerator
(a) process point
(al cadmium
/ 1 inorganic particulate
(a) 10%
(a)
(a)
(a)
(a)
5% vol.
400° F
110,000 scfm
atmospheric
none / none
(b)
(b)
(b)
(hj
(h)
(h)
(h)
P. Organic Content (1)***
Q. Heat/02 Content (1)
R. Particulate Content (3)
S Particle Mean Diam (3).
T. Drift Velocity/So, (3)
(r)
(c)
(c)
(r.)
(c.)
(r)
(0
3.2 qrains/acf.
1.0 urn
0.30 ft/sec /
-
-
-
-
_
-
-
-
-
flvash
200 ppmv
u.
V. Required Control Level
W. Selected Control Methods .
assume 99.9% removal
fabric filter, electrostatic precipitator, venturi scrubber
•The data presented are for an emission stream (single or combined streams) prior to entry into the selected control
method(s). Use extra forms if additional space is necessary (e.g., more than three HAP's). and note this need.
**The numbers in parentheses denote what data should be supplied depending on the data in steps "C" and "E":
1 = organic vapor process emission
2 = inorganic vapor process emission
3 = particulate process emission
***0rgamc emission stream combustibles less HAP combustibles shown on Lines D and F.
spray should be applied at each point where the
particles might be fractured, allowed to free fall, or
subjected to strong air currents.
3.3.2.3 Area Fugitive Emission Control From
Loading and Unloading
Loading and unloading bulk material is common to
many processing industries. Loading and unload-
ing operations can be either for external transpor-
tation of material to or from a facility or for internal
transportation within a facility (for example, inter-
nal transportation might consist of loading of a
mining haul truck with ore via a front-end loader for
subsequent unloading to a crushing process). (See
reference 16 or Appendix A.8, reference 3, for in-
dustry-specific information on loading and unload-
ing for internal transportation.)
Various control technology applications for loading
and unloading operations are presented in Table 3-
8. These techniques can be used alone or at times
in various combinations. Generally, the simulta-
neous 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 tech-
nique 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 es-
sentially eliminates the free-fall distance of the ma-
terial 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 ef-
fective in eliminating emissions since the surface of
41
-------
Table 3-7. Control Technology Applications for Transfer and
Conveying Sources (16)
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-collectors)
Wet suppression (water, chemical, foam)
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 sys-
tem can be placed over the filling hatch in some
types of trucks and railcars during 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 air-
flow mounted around this opening could capture
most of the dust generated.
Wet suppression techniques, when applied to load-
ing 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 en-
trained. 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, es-
pecially 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, dis-
placed 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 collect-
ed material being returned to the hold. Such a sys-
tem can practically eliminate loading emissions if
carefully maintained and properly operated. The
use of a canopy hood and exhaust system over the
loading boom is less effective than a totally en-
closed system, but can still reduce emissions and is
a viable alternative for open barges. Effective utili-
zation of this technique requires some type of wind
break to increase the hood capture efficiency.
Choke feed and telescopic chutes or spouts as pre-
viously described can also be used for loading both
enclosed and open ships or barges. Wet suppres-
sion techniques may also help reduce airborne
emissions if the product specifications do not pro-
hibit use of this technique.
Rail car and truck unloading — Many of the unload-
ing 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 device will effec-
tively reduce emissions. By causing air to flow
down through the grating, dust emissions are con-
tained. The face velocity of air through the grating
is a critical design parameter in this technique. Un-
loading cars with a screw conveyor causes less
distribution of the material and thereby less dust.
Problems of material handling and time require-
ments limit the application of this technique. Pneu-
matic unloading of very fine materials is an effec-
tive and widely used technique that practically
eliminates dust emissions. With this system, care-
ful maintenance of hose fittings and the fabric filter
through which the conveying air exhausts is re-
quired.
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 observa-
tion 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 possi-
bly wet suppression.
3.3.2.4 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, deposi-
tion of biological debris, wear from tires and brake
linings, and wear of anti-skid compounds. This ma-
terial is reentrained by contact with tires and by the
air turbulence created by passing vehicles.
42
-------
Table 3-8. Control Technology Applications for Loading and Unloading Operations (16)
Control Procedures
Emission Points
Loading Operations
Unloading Operations
Railcar, Truck
Barge, 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
Mm it free-fall distance
-For tanker types, use of gravity filler spouts with
concentric outer exhaust duct to control
equipment
-Wet suppression (water, chemicals)
— 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-feed to receiving pit (hopper car and
hopper truck)
—Unloading with screw conveyor (box car)
—Wet suppression (water, chemicals)
—Use of 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
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 con-
tinuing mechanical breakdown of large particles on
the road surface, thus providing new material in
the suspended particulate size range. Available
procedures for reducing emissions from plant
roads and their estimated efficiencies are present-
ed in Table 3-9.
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 vacu-
um-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 fre-
quency of cleaning and the removal efficiency of
the equipment.
For plants with small amounts of paved roads, in-
dustrial 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 stand-
point in that they redistribute material into the ac-
tive traffic lanes of the streets and they remove
almost none of the fine material (less than 43 |xm)
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. Howev-
er, the smaller industrial sweepers are usually de-
signed for use in warehouse and storage areas that
are not curbed. A factor which might limit the appli-
cability 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 re-
moval of spills on roadways and at conveyor trans-
fer 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.
Table 3-9. Control Technology Applications for
Plant Roads (16)
Emission Points
Control Procedure
Efficiency
Paved Streets
Unpaved
Roads
Road
Shoulders
Street cleaning No estimate
Housecleaning programs to
reduce deposition of material
on streets No estimate
Vacuum street sweeping
(daily) (2) 25% (17)
Speed reduction Variable
Paving 85%
Chemical stabilization 50%
Watering 50%
Speed reduction Variable
Oiling and double chip surface 85%
Stabilization 80%
43
-------
The paving of unpaved roadways is the most per-
manent of the various types of controls. However,
the degree of effectiveness of this technique is
highly dependent on prevention of excessive sur-
face dust loading.
Watering of unpaved roads is effective only when
carried out on a regular basis. The schedule de-
pends on climate, type of surface material, vehicle
use, and type of vehicles.
Oiling unpaved roads is more effective than water-
ing and needs to be applied less often. However,
special precautions must always be taken so as not
to add to surface water runoff problems.
3.3.2.5 Area Fugitive Emission Control From
Storage Piles
Most dust arises from stockpile areas as the materi-
al is dumped from the conveyor or chute onto the
pile, and as bulldozers move the pile. During per-
iods with high wind speeds (greater than about 6
m/sec [13 mph]) or low moisture, wind erosion of a
nonweathered surface may also cause emissions.
Applicable control techniques for open storage
piles are presented in Table 3-10.
Enclosing materials in storage is generally the
most effective means of reducing emissions from
this source category because it allows the emis-
sions to be captured. However, storage bins or si-
los may be very expensive. Storage buildings must
be designed to withstand wind and snow loads and
to meet requirements for interior working condi-
tions. One alternative to enclosure of all material is
Table 3-10. Control Technology Applications for
Open Storage Piles (16)
Emission Points Control Procedure
Loading onto Enclosure
Piles Chemical wetting agents or
foam
Adjustable chutes
Movement of Enclosure
Pile Chemical wetting agents
Watering
Traveling booms to distribute
material
Efficiency
70-99%
80-90%
75%
95-99%
90%
50%
No estimate
Wind Erosion Enclosure 95-99%
Wind screens Very low
Chemical wetting agents or
foam 90%
Screening of material prior to
storage, with fines sent
directly to processing or to a
storage silo No estimate
Loadout Water spraying 50%
Gravity feed onto conveyor 80%
Stacker/reclaimer 25-50%
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 emis-
sions. Earthen berms, vegetation, or existing struc-
tures 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 exten-
sions for piles with high material flow rates may
require closer control of operations because of the
possibility of jamming. Traveling or adjustable
booms can handle high flow rates, but have greater
operating costs.
Wetting agents or foams that are sprayed onto the
material during processing or at transfer points re-
tain their effectiveness in subsequent storage oper-
ations. 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 effec-
tive dust suppressant for some short-term storage
operations.
3.3.2.6 Area Fugitive Emission Control From Waste
Disposal Sites
Fugitive dust can occur anywhere dusty waste ma-
terial is dumped for disposal. This includes over-
burden piles, mining spoils, tailings, fly ash, bot-
tom 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. How-
ever, there may be emissions from transporting the
waste material on-site (if it is dry when it is pro-
duced) or from a reclamation process such as land-
fill covering associated with the waste disposal op-
eration. If the surface of the waste material does
not include a compound that provides cementation
upon weathering, or if the surface is not compact-
ed, 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 3-11 presents control
techniques for waste disposal sites.
44
-------
Table 3-11. Control Technology Applications for Waste
Disposal Sites (16)
Emission Points
Control Procedure
Efficiency
Handling Keep material wet 100%
Cover or enclose hauling No estimate
Minimize free fall of material No estimate
Dumping Spray bar at dump area 50%
Minimal free fall of material No estimate
Semi-enclose bin No estimate
Wind Erosion
Grading
Cover with dirt or stable
material
Chemically stabilize
Revegetate
Rapidly reclaim newly
filled areas
Water
100%
80%
25% -100%
No estimate
50%
3.4 References
1. Control Technology for Toxic and Hazardous
Air Pollutants. McFarland, A. R., ed. Illinois In-
stitute for Environmental Quality. Chicago, Illi-
nois. 1975.
2. Committee on Industrial Ventilation. Industrial
Ventilation: A Manual of Recommended Prac-
tice. 17th Edition. Lansing, Michigan. 1982.
3. U.S. EPA. Evaluation of Control Techniques for
Hazardous Air Pollutants - Appendices. EPA-
600/7-86-009D (NTIS PB86-167/038/AS.)October
1985.
4. U.S. EPA. Flexible Vinyl Coating and Printing
Operations — Background Information for Pro-
posed Standards. EPA-450/3-81-016a. January
1983.
5. U.S. EPA. Publication Rotogravure Printing —
Background Information for Proposed Stan-
dards. EPA-450/3-80-031a. October 1980.
6. U.S. EPA. Measurement of Process Capture Ef-
ficiency. Draft Report of Laboratory Testing.
EPA Contract No. 68-03-3038. January 1984.
7. U.S. EPA. VOC Emissions from Petroleum Re-
finery Wastewater Systems — Background In-
formation for Proposed Standards. Research
Triangle Park, North Carolina. July 1984.
8. U.S. EPA. Assessment of Atmospheric Emis-
sions from Petroleum Refining — Appendix B:
Detailed Results. EPA-600/2-80-075c. April
1980.
9. U.S. EPA. VOC Fugitive Emissions in Synthetic
Organic Chemicals Manufacturing Industry —
Background Information for Promulgated Stan-
dards. EPA-450/3-80-033b. June 1982.
10. U.S. EPA. Fugitive Emission Sources of Organ-
ic Compounds — Additional Information on
Emissions, Emission Reductions, and Costs.
EPA-450/3-82-01 O.April 1982.
11. Wilkins, G. E., J. H. E. Stelling, and S. A. Shar-
eef. Monitoring and Maintenance Programs for
Pumps and Valves in Petroleum and Chemical
Processing Plants: Costs and Effects on Fugi-
tive Emissions. Presented at 6th World Con-
gress on Air Quality, IUAPPA, Paris. May 1983.
12. Wilkins, G. E., and J. H. E. Stelling. Monitoring
and Maintenance Programs for Control of Fugi-
tive Emissions from Pumps and Valves in Pe-
troleum and Chemical Processing Plants. Pre-
sented at 1984 Industrial Pollution Control
Symposium, ASME. New Orleans, Louisiana.
February 1984.
13. U.S. EPA. VOC Fugitive Emission Predictive
Model — User's Guide. EPA-600/8-83-029. Oc-
tober 1983.
14. U.S. EPA. Evaluation of the Efficiency of Indus-
trial Flares. Background — Experimental De-
sign — Facility. EPA-600/2-83-070. August
1983.
15. Environmental Engineers' Handbook, Volume
II: Air Pollution. Liptak, B. G., ed. Chilton Book
Company. Radnor, Pennsylvania. 1974.
16. U.S. EPA. Technical Guidance for Control of
Industrial Process Fugitive Paniculate Emis-
sions. EPA-450/3-77-010. March 1977.
17. U.S. EPA. Identification, Assessment, and Con-
trol of Fugitive Paniculate Emissions. Final re-
port. EPA Contract No. 68-02-3922. May 1986. Con-
tact Mr. Dale Harmon of EPA at (919) 541-2429.
45
-------
Chapter 4
HAP Control Techniques
4.1 Background
This section describes and illustrates the proce-
dures used to calculate the basic design and oper-
ating variables of HAP control techniques in terms
of commonly employed design principles and val-
ues. For each technique, the handbook 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 handbook will result in conservatively de-
signed control systems. In instances in which less
conservatively designed control systems might
achieve the target control level, more detailed cal-
culation procedures requiring compound-specific
data would be needed. This level of specificity is
beyond the scope of this handbook.
The data for the HAP emission stream to be con-
trolled are taken from the HAP Emission Stream
Data Form 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 refer-
ring to Chapters 2 and 3.
The step-by-step calculation procedures are illus-
trated 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 applica-
tion to determine the adequacy of the applicant's
proposed design. Appendices C.2 through C.11
contain blank calculation sheets to use in applying
the calculations described for each control tech-
nique. If control systems costs are required, see
Chapter 5.
4.2 Thermal Incineration
Thermal incineration (Figure 4-1) is a widely used
air pollution control technique whereby organic va-
pors are oxidized at high temperatures. The most
important variables to consider in thermal inciner-
ator design are the combustion temperature and
residence time because these design variables de-
termine the incinerator's destruction efficiency.
Further, at a given combustion temperature and
residence time, destruction efficiency is also affect-
ed 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 then unsubstituted organ-
ics; hence, the presence of halogenated com-
pounds in the emission stream requires higher
temperature and longer residence times for com-
plete oxidation. Thermal incinerators can achieve a
wide range of destruction efficiencies. This discus-
sion focuses 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 inciner-
ator applications, the available enthalpy in the flue
gases is used for preheating the emission stream.
This discussion will be based on a thermal inciner-
ation system where the emission stream is pre-
heated.
The incineration of emission streams containing
organic vapors with halogen or sulfur components
may create additional control requirements. For ex-
ample, if sulfur and/or chlorine are present in the
emission stream, the resulting flue gas will contain
sulfur dioxide (S02) and/or hydrogen chloride
(HCI). Depending upon the concentrations of these
compounds in the flue gas and the applicable regu-
lations, scrubbing may be required to reduce the
concentrations of these compounds. The selection
and design of scrubbing systems are discussed in
Section 4.7.
In this subsection, the calculation procedure will be
illustrated using Emission Stream 1 described in
Chapter 3. Appendix C.3 provides worksheets for
calculations.
47
-------
Figure 4-1. Schematic diagram of a thermal incinerator system.
Emission Source
Scrubber*
Combustion Air*
Supplementary Fuel
-Stack
"Required for specific situations.
4.2.1 Data Required
The data necessary to perform the calculations
consist of HAP emission stream characteristics pre-
viously compiled on the HAP Emission Stream
Data Form and the required HAP control as deter-
mined by the applicable regulations.
Example Case
Maximum flow rate, Qe = 15,000 scfm
Temperature, Te = 120°F
Heat content, he = 0.4 Btu/scf
Oxygen content, O2 = 20.6%
Moisture content, Me = 2%
Halogenated organics: Yes
Based on the control requirements for the emis-
sion 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 inciner-
ator, 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):
Reported destruction efficiency, DEreported, %
Temperature of the emission stream entering the
incinerator, Te, °F (if no heat recovery);
The, °F (if a heat exchanger is employed)
Combustion temperature, Tc °F
Residence time, tr, sec
Maximum emission stream flow rate, Qe, scfm
Excess air, ex, %
Fuel heating value (assume natural gas), hf, Btu/scf
Supplementary heat requirement, Hf, Btu/min
Combustion chamber volume, Vc, ft
Flue gas flow rate, Qfg, scfm
Heat exchanger surface area (if a heat exchanger is
employed), A, ft2
4.2.2 Pretreatment of the Emission Stream:
Dilution Air Requirements
In HAP emission streams containing oxygen/air
and flammable vapors, the concentration of flam-
mable vapors is generally limited to less than 25
percent of the lower explosive limit (LED to satisfy
safety requirements imposed by insurance com-
panies. (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 mixtures 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 concentra-
tions of approximately 25 percent of LEL), the cal-
culation procedure in this handbook assumes that
dilution air is required. (See Appendix B.2 for calcu-
lation of dilution air requirements and Appendix
C.2 for a calculation worksheet.)
48
-------
Example Case
Since 02 = 20.6% and he = 0.4 Btu/scf, no dilu-
tion air is required.
4.2.3 Thermal Incinerator System Design
Variables
Table 4-1 presents suggested combustion tempera-
ture and residence time values for thermal inciner-
ators 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 combus-
tion 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 ef-
ficiency, it may be possible to incinerate HAP emis-
sion streams at lower temperatures with longer
residence times.
Based on the required destruction efficiency (DE),
select appropriate valuesforTc and tr from Table 4-1.
Table 4-1. Thermal Incinerator System Design
Variables (1,2)
Required Nonhalogenated Stream Halogenated Stream
Destruction Combustion Residence Combustion Residence
Efficiency Temperature Time Temperature Time
DE(%) TC(°F) tr(sec) Tc (°F) tf (sec)
98
99
1,600
1,800
0.75
0.75
2,000
2,200
1.0
1.0
Example Case
The required destruction efficiency is 99% and
the HAP emission stream is nonhalogenated,
therefore:
Tc = 1,SOOT (Table 4-1}
tr = 0.75 sec (Table 4-1)
In a permit evaluation, if the reported values for Tc
and tr are sufficient to achieve the required DE
(compare the applicant's values with the values
from Table 4-1), proceed with the calculations. If
the reported values for Tc and tr are not sufficient,
the applicant's design is unacceptable. The review-
er may then wish to use the values for Tc and tr
from Table 4-1. (Note: If DE is less than 98 percent,
obtain information from literature and incinerator
vendors to determine appropriate values for Tc and
4.2.4 Determination of Incinerator Operating
Variables
4.2.4.1 Supplementary Heat Requirements
Supplementary fuel is added to the thermal inciner-
ator to attain the desired combustion temperature
(Tc). For a given combustion temperature, the
quantity of heat needed to maintain the combus-
tion temperature in the thermal incinerator is pro-
vided by: (a) the heat supplied from the combus-
tion of supplementary fuel, (b) the heat generated
from the combustion of hydrocarbons in the emis-
sion 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 exchange 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 oxy-
gen concentration of the emission stream exceeds
16 percent. Depending on the heat content of the
emission stream and the desired combustion tem-
perature, combustion air requirements may be zero
even when the oxygen concentration is below 16
percent. Hence, this cut-off value will lead to a con-
servative design. Use the following simplified
equation for dilute streams to calculate supplemen-
tary heat requirements (based on natural gas):
Hf = 1.1 hf
f 0.002 Me) [CPair(Tc-Tr) - Cpair(The-Tr) - he]
hf-1.4Cpair(Tc-Tr)
(4.2-1)
where:
Hf =
hf =
Qe =
Me =
supplementary heat requirement, Btu/min
heating value of natural gas, Btu/scf
maximum emission stream flow rate, scfm
moisture content of the emission stream, %
Cpair = average specific heat of air over a given
temperature interval (Tr to T), Btu/scf-°F
Tc = combustion temperature, °F
Tr = reference temperature, = 70°F
The = emission stream temperature after heat
recovery, °F
he = heat content of the emission stream,
Btu/scf
Calculate The using the following expression if the
value for The is not specified:
The = (HR/100)TC + [1 - (HR/100)] Te
where:
HR = heat recovery in the exchanger, %
Assume a value of 50 percent if no other informa-
tion is available.
49
-------
The factor 1.1 in Equation 4.2-1 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 handbook,
it is assumed that the minimum supplementary
heat requirement is 5 Btu/min per scfm of emission
stream.
A graph of Equation 4.2-1 is shown in Figure 4-2,
where the ratio Hf/Qe is plotted against the emis-
sion stream heat content (he) for four different com-
bustion temperatures (Tc). Instead of evaluating
Equation 4.2-1, Figure 4-2 can be used directly to
determine supplementary heat requirements. This
graph is based on the following assumptions: (1)
temperature of the emission stream (Te) is 100°F,
(2) moisture content of the emission stream (Me) is
2 percent, (3) preheat temperature of the emission
stream (The) 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).
Figure 4-2. Supplementary heat requirement vs. emission
stream heat content (dilute stream/no
combustion air).
Figure 4-3.
100
^_ 90
II
ffi~» 80
Supplementary heat requirement vs. emission
stream heat content (no oxygen in emission
stream/maximum combustion air).
s£
1 E
I"
E o
70
60
50
40
30
20
• Tc = 2,200°F
O Tc = 2,000°F
• Tc = 1,800°F
D Tc = 1,600°F
1234
Emission Stream Heat Content, he (Btu/scf)
50
I
O 40
I
C 30
E
• ro
o
« •«
Q. 10
§E
05 W
20
10
• Tc = 2,200°F
O Tc = 2,000°F
• Tc = 1,800°F
D Tc = 1,600°F
1234
Emission Stream Heat Content, he (Btu/scf)
For emission streams that do not contain sufficient
quantities of oxygen to satisfy the combustion air
requirements (e.g., process emissions), refer to Fig-
ure 4-3 which shows a plot of Hf/Qe versus he to
obtain a conservative estimate for Hf. Figure 4-3 is
based on the same assumptions as those stated for
Figure 4-2. In addition, the oxygen content of the
emission stream (02) is assumed as zero; this cor-
responds to maximum combustion air require-
ments for the thermal incinerator system. If the
oxygen content of a particular emission stream
falls between zero and 16 percent, use Figure 4-3 to
obtain a conservative estimate of Hf/Qe.
Example Case
Using Equation 4-1:
Since the emission stream is very dilute and has
an oxygen content greater than 16%, Equation
4.2-1 is applicable. The values to be inserted in
the equation are:
= 15,000 scfm
= 2%
= 0.4 Btu/scf
= 1,800°F (Table 4-1)
= 70°F
= 960°F (based on heat recovery of 50%)
= 0.0196 Btu/scf-°F for the interval
70° - 1,800°F (reference 3)
Cpair = 0.0187 Btu/scf-°F for the interval
70° - 960°F (reference 3)
hf = 882 Btu/scf
Me
he
To
Tr
The
Hf = 1.1x882
"[15,000(33.91 -16.64-0
882 - 47.5
•4H1
Hf = 295,000 Btu/min
(Note: Hf is greater than the minimum supple-
mentary heat requirement assumed in this hand-
book.)
Using Figure 4-2:
For he = 0.4 Btu/scf and Tc = 1,800°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.
50
-------
4.2.4.2 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 (4.2-2)
where:
Qfg = flue gas flow rate, scfm
Qf = natural gas flow rate, scfm
Qc = combustion air requirement, scfm
Calculate Qf from the following equation:
Of = Hf/hf (4.2-3)
As indicated earlier, Qc will typically be zero for
dilute emission streams with oxygen contents (02)
greater than 16 percent. If 02 is less than 16 per-
cent, then:
Qc = (0.01 He + 9.4 Qf) (1 + 0.01 ex) - 0.0476 02 Qe
where:
He = heat generated due to the combustion of
hydrocarbons in the emission stream
He = Qehe
Example Case
Using Equation 4.2-3:
Hf = 295,000 Btu/min
h, = 882 Btu/scf
Qf = 295,000/882 = 330 scfm
Using Equation 4.2-2:
Qe = 15,000 scfm
Qc = 0 (since 02 is greater than 16%)
Qfg = 15,000 + 330 + 0
Qfg = 15,330 scfm
4.2.5 Combustion Chamber Volume
The flue gas flow rate (Qfg) determined by Equation
4.2-2 is expressed at standard conditions. In order
to calculate the combustion chamber size, Qfg must
be expressed at actual conditions, i.e., temperature
effects must be considered (assume pressure ef-
fects are negligible). Use the following equation to
convert Qfg from "scfm" to "acfm":
Qfg,a = Qfg KTC + 460)7530] (4.2-4)
where:
Qfg.a = flue gas flow rate at actual conditions, acfm
The volume of the combustion chamber (Vc) is de-
termined from the residence time (tr) and flue gas
flow rate at actual conditions (Qfg,a) according to
the following equation:
The factor 1.05 is used in Equation 4.2-5 to increase
the chamber volume by 5 percent. This technique is
an accepted industry practice and allows for fluctu-
ations in the operating conditions (e.g., flowrate,
temperature, etc.). The smallest commercially
available incinerator has a combustion chamber
volume of about 36 ft3 (1 m3). If the calculated Vc is
less than 36 ft3, define Vc as 36 ft3.
Example Case
Using Equation 4.2-4:
Tc = 1,800°F
Qfg = 15,330 scfm
Qfg,a = 15,330 [(1,800 + 460)/530]
Qfg'a = 65,370 acfm
Using Equation 4.2-5:
tr = 0.75 sec (from Table 4-1)
Vc = [(65,370/60)0.75] 1.05
Vr = 860 ft3
4.2.6 Heat Exchanger Size
The size of the heat exchanger required for pre-
heating the emission stream to The before it enters
the thermal incinerator is based on the heat ex-
changer design. Use the following expression to
calculate the required size, i.e., surface area, of the
heat exchanger:
A =
[60Qe(1 + 0.002 Me) Cpair (The - Te)]
UAT,
(4.2-6)
LM
heat exchanger surface area, ft2
emission stream temperature, °F
overall heat transfer coefficient, Btu/hr-ft2-°F
logarithmic mean temperature difference, °F
Vc= [(Q,g,a/60)tr]x1.05
(4.2-5)
where:
A
Te =
U
ATLM =
and:
ATUM = [(Tc - The) - (Thc - Te)] In [(Tc - The)/(Thc - Te)]
For dilute emission streams that do not require
additional combustion air, ATLM can be approxi-
mated by:
ATLM = Tc - The
For a recuperative heat exchanger where the heat
transfer takes place between two gas streams, the
overall heat transfer coefficient (U) ranges from 2
to 8 Btu/hr-ft°F, generally depending on the heat
exchanger configuration and properties of the gas
streams.
Equation 4.2-6 has been evaluated for dilute emis-
sion streams that require no additional combustion
air, as shown in Figure 4-4. In the figure, heat ex-
changer surface area (A) is plotted against the
emission stream flow rate (Qe). The assumptions
inherent in this figure are the same as those de-
scribed for Figure 4-2. The overall heat transfer
coefficient is assumed as 4 Btu/hr-ft2-°F.
57
-------
Example Case
Using Equation 4.2-6:
Qe = 15,000 scfm
M
= 2%
he
960°F (based on heat recovery of 50%)
120°F (input data)
0.0187 Btu/scf-°F for the interval 120° -
960°F (reference 3)
4 Btu/hr-ft2-°F
Tc-The
1,800°F
ATLM= 1,800°-960° = 840°F
Substituting in Equation 4.2-6:
[60 x 15,000 x 1.004 x 0.0187 (960 - 120)]
A =
U
ATLM =
Tr =
A = 4,200ft2
(4 x 840)
Using Figure 4-4:
For Qe = 15,000, the value for A from the figure
is about 4,000 ft2.
For emission streams that are not dilute and re-
quire additional combustion air, use Figure 4-5 to
obtain an estimate of the heat exchanger surface
area. In Figure 4-5, the ratio A/Qe is plotted against
the emission stream content (he) for four different
combustion temperature (Tc). The assumptions in-
herent in this figure are the same as those stated
for Figure 4-4; in addition, maximum combustion
air requirements are assumed (i.e., 02 = 0). If the
conditions represented in Figures 4-4 and 4-5 are
not directly applicable for a particular emission
stream, use Figure 4-4 to obtain a conservative
estimate.
If the calculated values and the reported values are
not different, then the design and operation of the
proposed thermal incinerator system may be con-
sidered appropriate based on the assumptions
used in this handbook.
Figure 4-4. Heat exchanger size vs. emission stream flow rate
(dilute stream/no combustion air).
15,000
12,500
N""
C
<
-------
Table 4-2. Comparison of Calculated Values and Values
Supplied by the Permit Applicant for Thermal
Incineration
Calculated
Value Reported
(Example Case)a Value
Supplementary heat
requirement, Hf
Supplementary fuel flow
rate, Q,
Flue gas flow rate, Qfg
Combustion chamber size, Vc
Heat exchanger surface area, A
295,000
330 scfm
1 5,330 scfm
860ft3
4,200 ft2
aBased on Emission Stream 1.
4.3 Catalytic Incineration
Catalytic incineration (Figure 4-6) 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 ap-
preciably changed during the reaction. Catalysts
typically used for VOC incineration include plati-
num 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 ce-
ramic matrix structure designed to maximize cata-
lyst 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 affect-
ed by several factors including: (a) operating tem-
perature, (b) space velocity (reciprocal of residence
time), (c) VOC composition and concentration, (d)
catalyst properties, and (e) presence of poisons/in-
hibitors in the emission stream. In catalytic inciner-
ator design, the important variables are the operat-
ing 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) en-
tering 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 veloc-
ity, increasing the operating temperature at the in-
let 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 velocities in the range 30,000 to
100,000 hr1.(4,5,6) However, the greater catalyst
volumes and/or higher temperatures required for
higher destruction 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 sensi-
tive to pollutant characteristics and process condi-
tions. In the following discussion, it is assumed that
the emission stream is free from poisons/inhibitors
such as phosphorus, lead, bismuth, arsenic, anti-
mony, mercury, iron oxide, tin, zinc, sulfur, and
halogens. (Note: Some catalysts can handle emis-
sion streams containing halogenated compounds.)
Figure 4-6. Schematic diagram of a catalytic incinerator system.
Emission Source
O
Combustion Air*
Supplementary Fuel
Catalytic Incinerator
Preheater
I
»f V
**N
\A
M Vv
v^^
\-Catalyst Bed
1
*• otacK
Heat Exchanger
(Optional)
-Dilution Air*
"Required for specific situations.
53
-------
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 pro-
cess, producing hot water or steam, etc.). In recu-
perative 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 over-
heat 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., preheat-
ing the emission stream). The calculation proce-
dure will be illustrated using Emission Stream 2
described in Chapter 3. Appendix C.4 provides
worksheets for calculations.
4.3.1 Data Required
The data necessary to perform the calculations
consist of HAP emission stream characteristics pre-
viously compiled on the HAP Emission Stream
Data Form and the required HAP control as deter-
mined by the applicable regulations.
Example Case
Maximum flow rate, Qe = 20,000 scfm
Temperature, Te = 120°F
Heat content, he = 2.1 Btu/scf
Oxygen content, 02 = 20.6%
Moisture content, Me = 2%
Based on the control requirements for the emis-
sion 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 inciner-
ator, 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):
Reported destruction efficiency, DEreportedi %
Temperature of the emission stream entering the
incinerator, Te, °F (if no heat recovery); The, °F (if
emission stream is preheated)
Temperature of flue gas leaving the catalyst bed,
T °F
I co» r
Temperature of combined gas stream (emission
stream + supplementary fuel combustion prod-
ucts) entering the catalyst bed,3 TCI, °F
Space velocity, SV, hr"1
Supplementary heat requirement, Hf, Btu/min
Flow rate of combined gas stream entering the
catalyst bed,3 Qcom, scfm
Combustion air flow rate, Qc, scfm
Excess air, ex, %
Catalyst bed requirement, Vbed, ft3
Fuel heating value, hf, Btu/scf
Heat exchanger surface area (if a heat exchanger is
employed), A, ft2
4.3.2 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 to deactivate 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 handbook that catalytic incineration is appli-
cable 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 re-
quired 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 10
"If no supplementary fuel is used, the value for this variable will be the
same as that for the emission stream.
54
-------
Btu/scf cut-off value for determining dilution air
requirements. See Appendix B.2 for calculating di-
lution air requirements.
Based on the required destruction efficiency (DE),
specify the appropriate ranges for Tci, Tc
select the value for SV from Table 4-3.
and
Example Case
Since the heat content of the emission stream
(he) is 2.1 Btu/scf, no dilution is necessary.
4.3.3 Catalytic Incinerator System Design Variables
Table 4-3 presents suggested values and limits for
the design variables of a fixed bed catalytic inciner-
ator system to achieve a given destruction efficien-
cy. For specific applications, other temperatures
and space velocities may be appropriate depend-
ing on the type of catalyst employed and the emis-
sion 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. (See reference 7 or Appen-
dix B.6, reference 8, for data on temperatures typi-
cally required for specific destruction efficiency lev-
els for several compounds.)
Table 4-3. Catalytic Incinerator System Design
Variables (1,5,6)
Required
Destruction
Efficiency
DE(%)
90
95
Temperature
at the Catalyst
Bed Inlet3
Tc, (°F)
600
600
Temperature
at the Catalyst
Bed Outletb
Tco (°F)
1,000-1,200
1,000-1,200
Space
Velocity
SV(hr1)
40,000C
30,000d
a Minimum temperature of combined gas stream (emission
stream + supplementary fuel combustion products) entering
the catalyst bed is designated as 600°Fto ensure an adequate
initial reaction rate.
bMinimum 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 destruc-
tion level. Maximum temperature of flue gas leaving the cata-
lyst bed is limited to 1,200°F to prevent catalyst deactivation by
overheating.
•"Corresponds to 1.5 ft3 of catalyst per 1,000 scfm of combined
gas stream.
dCorresponds to 2.0 ft3 of catalyst per 1,000 scfm of combined
gas stream.
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.(4) The destruc-
tion efficiency for a given compound in different
VOC mixtures may also vary with mixture composi-
tion. (See reference 4 or Appendix B.6, reference 8,
for compound-specific destruction efficiency data
for two different VOC mixtures.)
Example Case
The required destruction efficiency is 95%;
therefore:
Tci (minimum) = 600°F
Tco (minimum) = 1,000°F
Tco (maximum) = 1,200°F
SV = 30,000 hr1
In a permit evaluation, determine if the reported
values for Tci, Tco, and SV are appropriate to
achieve the required destruction efficiency by com-
paring the applicant's values with the values in
Table 4-3. The reported value for Tci should equal
or exceed SOOT 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 Tco falls in the interval
1,000° - 1,200°F. Note that 1,000°F is a conservative
value. Then check if the reported value for SV is
equal to or less than the value in Table 4-3. 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 low-
er temperature level. Otherwise, the applicant's de-
sign is considered unacceptable. In such a case, the
reviewer may then wish to use the values in Table
4-3.
4.3.4 Determination of Incinerator System
Variables
4.3.4.1 Supplementary Heat Requirements
Supplementary fuel is added to the catalytic incin-
erator system to provide the heat necessary to
bring the emission stream up to the required cata-
lytic oxidation temperature (TCI) for the desired lev-
el of destruction efficiency. For a given TCI, 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
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
(O2) is greater than or equal to 16 percent.
55
-------
Before calculating the supplementary heat require-
ments, the temperature of the flue gas leaving the
catalyst bed (Tco) 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 Tco falls in the interval 1,000° -
1,200°F. Use the following expression to calculate
Tco, taking into consideration the temperature rise
across the catalyst bed due to heat generation from
combustion of VOC in the emission stream:
= Tci + 50 he
(4.3-1)
where:
he = heat content of the emission stream, Btu/scf
In this expression, it is assumed that the heat con-
tent of the emission stream and the combined gas
stream is the same. Inserting Tci = 600°F, if Tco is
in the range 1,000° - 1,200°F, then Tci = 600°F is
satisfactory. If Tco is less than 1,000°F, use the
following equation to determine an appropriate
value for Tci (above 600°F) and use this value in the
following calculations:
Tci = 1,000-50he
(4.3-2)
(Note: Emission streams with high heat contents
will be diluted based on the requirements dis-
cussed in Section 4.3.2. Therefore, values for Tco
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:
Hf = 1.1 hfxQe(1 + 0.002 Me)
[Cpair (Tci-Tr) - Cpair (The - Tr)
ir (Tci - Tr)
(4.3)
where:
h,f =
0-e =
MP =
supplementary heat requirement, Btu/min
heating value of natural gas, Btu/scf
maximum emission stream flow rate, scfm
moisture content of the emission stream,
percent
Cpair = average specific heat of air over a given
temperature interval, (Tr -T) Btu/scf-°F
Tci = temperature of combined gas stream
entering the catalyst bed, °F
Tr = reference temperature, = 70°F
The = emission stream temperature after heat
recovery, °F
Note that for the case of no heat recovery, The =
Te. The factor 1.1 accounts for an estimated heat
loss of 10 percent in the incinerator. Supplemen-
tary heat requirements are based on maximum
emission flow rate, and hence will lead to a conser-
vative design. In contrast to thermal incineration,
there is no minimum supplementary heat require-
ment specified for catalytic incineration since no
fuel is needed for flame stabilization. Depending on
the VOC concentration, emission stream tempera-
ture, and level of heat recovery, supplementary
heat requirements may be zero when heat recovery
is practiced.
A plot of Equation 4.3-3 is shown in Figure 4-7
where the ratio of Hf/Qe is plotted against the emis-
sion stream heat content for two levels of heat
recovery (zero and 50 percent). As an alternative to
Equation 4.3-3, use Figure 4.7 directly to determine
Hf. The figure is based on the following assump-
tions: (1) moisture content of the emission stream
(Me) is 2 percent, (2) emission stream temperature
(Te) is 100°F, (3) preheat temperature of the emis-
sion stream (The) is based on 50 percent heat re-
covery in the heat exchanger, and (4) net heating
value of supplementary fuel (natural gas) is 882
Btu/scf.
Figure 4-7.
28
26 -
O 24
Supplementary heat requirement vs. emission
stream heat content (dilute stream/no
combustion air).
CO
tr
I
U_
E
co
CD
•*-*
C/3
O
'
22
20
18
16
c 14
E
m 12
x"
I 4
a
> No heat recovery
I 50% heat recovery
12345
Emission Stream Heat Content, he (Btu/scf)
56
-------
Example Case
Using Equations 4.3-1, -2, and -3:
Since the emission stream is dilute (he = 2.1
Btu/scf) and has an oxygen concentration great-
er than 16% (O2 = 20.6%), these equations are
applicable.
a.
Determine if Tco
1,200°F:
falls in the range 1,000° -
he
Tco
= 600°F
= 2.1 Btu/scf (input data)
= 600 + (50x2.1) = 705°F
Since Tco is less than 1,000°F, use Equation
4.3-2 to calculate an appropriate value for Tci:
Tci = 1,000-(50x2.1) = 895°F
b. Determine Hf (assume recuperative heat
recovery will be employed):
Qe
Me
Tr
The
Cpair
Hf =
(882 - 21.60)
Hf = 150,500 Btu/min
Using Figure 4-7:
For he = 2.1 Btu/scf and using the curve for
50% heat recovery, Hf/Qe from the figure is
about 7.5 Btu/min/scfm. Multiplying 7.5 by Qe,
(7.5 x 20,000), yields an approximate value of
150,000 Btu/min for Hf.
20,000 scfm
2%
70°F
550°F (based on heat recovery of 50%)
0.0187 Btu/scf-°F for the interval
70° - 895°F (reference 3)
Cpair = 0.0183 Btu/scf-°F for the interval
70° - 550°F (reference 3)
1.1 x 882 x 20,000 x 1.004
(15.43 - 8.78)
o
of
ro
CC
^
u_
e
CO
a
a
CO
Figure 4-8. Supplementary heat requirement vs. emission
stream heat content (no oxygen in emission
stream/maximum combustion air).
42
40
38
36
34
32
30
28
26
24
22
18
16
14
12
10
8
6
4
2
12
1 No heat recovery
Heat recovery
(The = 550°F)
1234
Emission Stream Heat Content, he (Btu/scf)
For emission streams that do not contain sufficient
quantities of oxygen to satisfy combustion air re-
quirements (e.g., process emissions), refer to Fig-
ure 4-8 which shows a plot of Hf/Qe versus he for
two levels of heat recovery (i.e., no heat recovery
and where The is 550°F). In this figure, the oxygen
content of the emission stream (O2) is assumed as
zero; this corresponds to maximum combustion air
requirements. The emission stream moisture con-
tent (Me), emission stream temperature (Te) and
the fuel heating value (hf) are as specified for Fig-
ure 4-7. The preheat temperature of the emission
stream (The) is 550°F for the heat recovery case. If
02 for a particular emission stream is between 0
and 16 percent, use Figure 4-8.
4.3.4.2 Flow Rate of Combined Gas Stream
Entering the Catalyst Bed
In order to calculate the quantity of catalyst re-
quired, the flow rate of the combined gas stream
(emission stream + supplementary fuel combus-
tion products) at the inlet to the catalyst bed has to
be determined. Use the following equation:
Qf + Qc
(4.3-4)
where:
QCom = flow rate of the combined gas stream, scfm
Qf = natural gas flow rate, scfm
Qc = combustion air requirement, scfm
57
-------
Calculate Qf from the following expression:
Qf = Hf/hf (4.3-5)
As indicated earlier, Qc will typically be zero for
dilute emission streams with oxygen contents (02)
greater than 16 percent. If 02 is less than 16 per-
cent, then:
Qc = (0.01 He + 9.4 Qf) (1 + 0.01 ex) - 0.476 02Qe
where:
He = heat generated due to the combustion of
hydrocarbons in the emission stream
He = Qehe
Example Case
Using Equation 4.3-4:
Qe = 20,000 scfm
Qc =0 (since 02 is greater than 16%)
Qcom = 20,000 + 170 + 0
Qcom = 20,170 scfm
Using Equation 4.3-5:
Hf = 150,500 Btu/min
hf = 882 Btu/scf
Qf =170 scfm
4.3.4.3 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 must 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 combus-
tion of the VOC in the mixed gas stream is small,
especially when dilute emission streams are treat-
ed. Therefore,
Qfg = Qcom
where:
Qfg = flow rate of the flue gas leaving the catalyst
bed, scfm
While figuring costs, assume that catalytic inciner-
ators are designed for a minimum Qfg of 500 scfm.
Therefore, if Qfg is less than 500 scfm, define Qfg
as 500 scfm.
In order to determine operating costs, the flue gas
flow rate (Qfg) has to be expressed at actual condi-
tions. Use the following equation to convert Qfg
from "scfm" to "acfm":
Qfg,a =
460)/530]
58
(4.3-6)
where:
Qfg;a is the flue gas flow rate at actual conditions
(acfm)
Example Case
Using Equation 4.3-6:
Qfg = Qcom = 20,170 SCfm
Tco - 1,000°F
Qfg,a = [20,170(1,000 + 460)75301
Qfg,a = 55,600 acfm
4.3.5 Catalyst Bed Requirement
The total volume of catalyst required for a given
destruction efficiency is determined from the de-
sign space velocity as follows:
Vbed = 60 Qcom/SV
where:
vbed = volume of catalyst bed required, ft3
(4.3-7)
Example Case
Using Equation 4.3-7:
Qcom = 20,170 scfm
SV = 30,000 hr1 (Table 4-3)
Vbed = 60x20,170/30,000
Vbed = 40 ft3
4.3.6 Heat Exchanger Size (for Systems with
Recuperative Heat Exchange Only)
To determine the size of the heat exchanger re-
quired for preheating the emission stream to The
use the following expression:
A = [60 Qe(1 + 0.002 Me) Cpair (The - Te)]/UATLM
(4.3-8)
where:
A = heat exchanger surface area, ft2
U = overall heat transfer coefficient, Btu/hr-ft2-°F
ATLM = logarithmic mean temperature difference, °F
and:
ATLM = [(Tc - The) - (The - Te)] In [(Tc - The)/(The - Te)]
For dilute emission streams that do not require
additional combustion air, then ATLM can be ap-
proximated by:
AT, M = Tr -
he
For a recuperative heat exchanger where the heat
transfer takes place between two gas streams, the
overall heat transfer coefficient (U) ranges from 2
to 8 Btu/hr-ft°F, generally depending on the heat
exchanger configuration and properties of the gas
streams.
-------
Example Case
Using Equation 4.3-8:
Qe = 20,000 scfm
Me = 2%
The = 550°F (based on heat recovery of 50%)
Te = 120T (input data)
Cpair = 0.0183 Btu/scf-°F for the interval
120°-550°F(reference 3)
U = 4 Btu/hr-ft2-°F
Since the emission stream is dilute, calculate
ATLM as follows:
AT, M =
T,
-Th
LM — ' co ' ' he
co = 1,000°F
ATLM= 1,000-550 = 450°F
The heat exchanger surface area from Equation
4.3-8 then becomes:
60 x 20,000 x 1.004 x 0.0183 x (550-100)1
A =
4x450
A = 5,500 ft2
Using Figure 4-9:
For all values of he
Thus, multiplying Qe by A/Qe, (20,000 x 0.275)
yields 5,500 ft2.
A/Qe is about 275 x 10'
Alternatively, Figure 4-9 can be used to determine
the heat exchanger size. In this figure, line (1) repre-
sents Equation 4.3-8 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 4-7. The overall heat
transfer coefficient is assumed as 4 Btu/hr-ft2-°F.
For emission streams that require additional com-
bustion air, use the solid line in Figure 4-9 to obtain
an estimate of the required heat exchanger size. As
in Figure 4-8, the solid line is based on maximum
combustion air requirements (i.e., no oxygen in the
emission stream). If Figure 4-9 is not directly appli-
cable in a particular situation, use the broken line in
the figure to obtain a conservative estimate for A.
4.3.7 Evaluation of Permit Application
Compare the results from the calculations and the
values supplied by the permit applicant using Table
4-4. The calculated values in the table are based on
the example case.
If the calculated values for Hf, Qc, Qcom, Vbed, and
A differ from the reported values for these varia-
bles, the differences may be due to the assump-
tions 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 handbook.
Figure 4-9. Heat exchanger size vs. emission stream heat
content.
S 340
§
f 330
1
0)
° 320
CD
CD
£C
310
LL
E
CD
CD
co 300
0
'co
CO
E 290
LU
t- 28°
CD
CD
% 270
t
D
CO
CD
g> 260
CD
.C
O
X
LU
*- 250
m our.
o> 0
I
p 280 1
(D
§ " <2) V
0 240 - \
X _ \
S? 200 - \
^. ^L
' ' ~ ^V
\
16A I I 1 I I 1 I I I I | I 1 I
0 2 46 8 10 12 14
he
(1) Dilute stream/no combustion air
(2) No oxygen in emission stream/
maximum combustion air
_
(1)
_
-
(2) _
^
f l I I I
12345
Table 4-4.
Emission Stream Heat Content, he (Btu/scf)
Comparison of Calculated Values and Values
Supplied by the Permit Applicant for Catalytic
Incineration
Calculated
Value
(Example Case)"
Reported
Value
Supplementary heat
requirement, Hf
Supplementary fuel flow
rate,Qf
Combustion air flow rate, Qc
Combined gas stream flow
rate, Qcom
Catalyst bed volume, Vbed
Heat exchanger surface area (if
recuperative heat recovery
is used), A
150,500 Btu
170 scfm
0
20,170 scfm
40ft3
5,500 ft2
a Based on Emission Stream 2.
59
-------
4.4 Flares
Flares use open flames for disposing of waste gas-
es during normal operations and emergencies.
They are typically applied when the heating value
of the waste gases cannot be recovered economi-
cally because of intermittent or uncertain flow, or
when process upsets occur. In some cases, flares
are operated in conjunction with baseload gas re-
covery 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 pres-
sure head flares. Typical flare operations can be
classified as "smokeless," "nonsmokeless," and
"fired" or "endothermic." For smokeless oper-
ation, flares use outside momentum sources (usu-
ally steam or air) to provide efficient gas/air mixing
and turbulence for complete combustion. Smoke-
less flaring is required for destruction of organics
heavier than methane. Nonsmokeless operation is
used for organic or other vapor streams which burn
readily and do not produce smoke. Fired, or en-
dothermic, flaring requires additional energy in or-
der 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 fac-
tors as flare gas exit velocity, emission stream
heating value, residence time in the combustion
zone, waste gas/oxygen mixing, and flame tem-
perature. 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 4-10. 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 prob-
lems. 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, how-
ever, 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 ignit-
ed by pilot burners. If conditions in the flame zone
are optimum (oxygen availability, adequate resi-
dence time, etc.), the VOC in the emission stream
may be completely burned (-100 percent efficien-
cy). 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.(9,10)
Typically, existing flare systems will be used to
control HAP emission streams. Therefore, the fol-
Figure 4-10. A typical steam-assisted flare system.
Steam
Nozzles
Gas Collection Header
and Transfer Line
Emission
Stream'
Knock-out Drum
Pilot Burners
Steam Line
Ignition Device
Air Line
Gas Line
Drain
60
-------
lowing sections describe how to evaluate the de-
struction efficiency of an existing flare system un-
der 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.(9) The cal-
culation 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 fol-
lowing calculation procedure can be applied.
4.4.1 Data Required
The data necessary to perform the calculations
consist of HAP emission stream characteristics pre-
viously compiled on the HAP Emission Stream
Data Form, flare dimensions, and the required HAP
control as determined by the applicable regula-
tions.
Example Case
Expected emission stream flowrate,
Qe = 30,000 scfm
Emission stream temperature, Te = 100°F
Heat content, he = 180 Btu/scf
Mean molecular weight of emission stream,
MWe = 33.5 Ib/lb-mole
Flare tip diameter, Dtlp = 48 in
Based on the control requirements for the emis-
sion stream:
Required destruction efficiency, DE = 98%
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. Worksheets are pro-
vided in Appendix C.5.
Flare system variables at standard conditions (70°F,
1 atm):
Flare tip diameter, Dtip, in
Expected emission stream flowrate, Qe, scfm
Emission stream heat content, he, Btu/scf
Temperature of emission stream, Te, °F
Mean molecular weight of emission stream,
MWe, Ib/lb-mole
Steam flowrate, Qs, Ib/min
Flare gas exit velocity, Uf|g, ft/sec
Supplementary fuel flow rate,3 Qf, scfm
Supplementary fuel heat content,3 hf, Btu/scf
Temperature of flare gas,b Tf|g, °F
Flare gas flow rate,b Qfig, scfm
Flare gas heat content, hf|g, Btu/scf
4.4.2 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 effi-
ciency when they are flared in steam- or air-assist-
ed flares.0 Therefore, the first step in the evaluation
procedure is to check the heat content of the emis-
sion stream and determine if additional fuel is
needed.
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 appli-
cation 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 4.4.2.3.
4.4.2.1 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 con-
tent to 300 Btu/scf. Calculate the required natural
gas requirements using the following equation:
Qf = [(300 - he)Qe]/582 (4.4-1)
where:
Qe = emission stream flow rate, scfm
Qf = natural gas flow rate, scfm
he = emission stream heat content, Btu/scf
(See Appendix B.7, reference 8, for details of all the
equations used in Section 4.4.) If the emission
stream heat content is greater than or equal to 300
Btu/scf, then Qf = 0.
Example Case
Using Equation 4.4-1:
Since he is less than 300 Btu/scf, supplementary
fuel is needed.
he = 180 Btu/scf
Qe = 30,000 scfm
Qf = (300-180)(30,000)/582
Qf = 6,200 scfm
"This information is needed if the emission stream heat content is less
than 300 Btu/scf.
b If no auxiliary fuel is added, the value for this variable will be the same
as that for the emission stream.
Tor unassisted flares, the lower limit is 200 Btu/scf.
4.4.2.2 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:
Qflfl = Qe + Qf (4.4-2)
where:
Qfig = flare gas flow rate, scfm
Note that if Qf = 0, then Qf,g = Qe.
The heat content of the flare gas (hf|g) is dependent
on whether supplementary fuel is added to the
61
-------
emission stream. When he is greater than or equal
to 300 Btu/scf, then hflg = he. If he is less than 300
Btu/scf, since supplementary fuel is added to in-
crease he to 300 Btu/scf, hfig = 300 Btu/scf.
Example Case
Using Equation 4.4-2:
Qe = 30,000 scf m
Qf = 6,200 scf m
Qfig = 36,200 scfm
Since he = 180 Btu/scf, hf,g = 300 Btu/scf.
4.4.2.3 Flare Gas Exit Velocity
The flare gas exit velocity values presented in Table
4-5 to achieve at least 98 percent destruction effi-
ciency in a steam-assisted flare system are based
on studies conducted by EPA.O) Flare gas exit ve-
locities are expressed as a function of flare gas heat
content. Determine the maximum allowable exit
velocity using the equation presented in Table 4-5.
Table 4-5. Flare Gas Exit Velocities (9)
Flare Gas Heat Content3
hflg (Btu/scf)
Maximum Exit Velocity
Umax (ft/sec)
h,ig < 300
300 s h,,g < 1,000
h,,g> 1,000
3 28
400
8 If no supplementary fuel is used, hfig = he.
b Based on studies conducted by EPA, waste gases having heat-
ing values less than 300 Btu/scf are not assured of achieving
98% destruction efficiency when they are flared in steam-
assisted flares.(4)
Example Case
Since hf|g = 300 Btu/scf, use the equation in
Table 4-5 to calculate Umax:
Umax = 3.28 [io(0-00118 h"= + °-908)
= 3 28 f-]o(000118 x 30°
Umax = 60 ft/sec
°-908)l
From the emission stream data (expected flow rate,
temperature) and information on flare diameter,
calculate the flare gas exit velocity (Ufig); compare
this value with Umax. Use the following equation to
calculate Uf|g:
Ufig = (Qflg,a/60)[4MDtlp/12)2]
= (3.06 Qfig,a)/(Dtlp)2 (4.4-3)
where:
Ufig = exit velocity of flare gas, ft/sec
Qfig.a = f'are 9as fl°w rate at actual conditions,
acfm
Dtip = flare tip diameter, in
Use the following expression to calculate Qfig,a:
Qfig,a =
460)1/530
62
(4.4-4)
If Ufig is less than Umax, then the 98 percent de-
struction level can be achieved. However, if Uflg
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 con-
sideration. Note that at very low flare gas exit ve-
locities, flame instability may occur. The minimum
flare gas exit velocity for a stable flame is assumed
as 0.03 ft/sec in this handbook.Cl) Thus, if Ufig is
below 0.03 ft/sec, the desired destruction efficiency
may not be achieved. In summary, Ufig should fall
in the range 0.03 ft/sec and Umax for a 98 percent
destruction efficiency level.
In a permit review case, if Ufig exceeds Umax, then
the application is not acceptable. If Ufig is below
Umax 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 4.4-3 and -4:
Qfig = 36,200 scfm
Tf,g = 95°F
Qfig.a = [36,200 (95 + 460)]/530
Qfig,a = 37,900 acfm
Ufig
Uf,g
= 48 in
= (3.06 x 37,900)/(48)2
= 50 ft/sec
Since 0.03 ft/sec < Uflg = 50 ft/sec < Umax =
60 ft/sec, the required level of 98% destruction
efficiency can be achieved under these condi-
tions.
4.4.2.4 Steam Requirements
Steam requirements for steam-assisted flare oper-
ation depend on the composition of the flare gas
and the flare-tip design. Typical values range from
0.15 to 0.50 Ib steam/lb flare gas. In this handbook,
the amount of steam required for 98 percent de-
struction efficiency is assumed as 0.4 Ib steam/lb
flare gas.(2) Use the following equation to deter-
mine steam requirements:
Qs = 1.03 x 10~3 x Qfig x MWf|g (4.4-5)
where:
Qs = steam requirement, Ib/min
= (Of x 16.7 + QeMWe)/Qf|g
Example Case
Using Equation 4.4-5:
Qfig = 36,200 scfm
MWfig = 30.6 Ib/lb-mole
Qs = 1.03 x10"3x 36,200x30.6
= 1,140 Ib/min
-------
4.4.3 Evaluation of Permit Application
Compare the results from the calculated and re-
ported values using Table 4-6. If the calculated val-
ues of Qf, Ufig, Qfig, and Qs 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 handbook.
Table 4-6. Comparison of Calculated Values and Values
Supplied by the Permit Applicant for Flares
Figure 4-11. Adsorption isotherms for toluene/activated
carbon system.(11)
Calculated
Value Reported
(Example Case)3 Value
Supplementary fuel flow rate, Qf
Flare gas exit velocity, Uf,g
Flare gas flow rate, Qfig
Steam flow rate, Qs
6,200 scfm
50 ft/sec
36,200 scfm
1,140lb/min
a Based on Emission Stream 3.
4.5 Boilers/Process Heaters
The application of boilers and/or process heaters as
emission control devices is very site-specific (see
Section 3.2.1.4). The level of detail required in the
calculations for sizing such devices is beyond the
scope of this handbook and thus is not presented.
4.6 Carbon Adsorption
Adsorption is a surface phenomenon whereby hy-
drocarbons and other compounds are selectively
adsorbed on the surface of such materials as acti-
vated carbon, silica gel, or alumina. Activated car-
bon is the most widely used adsorbent. The ad-
sorption 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 con-
stant temperature (see Figure 4-11). Typically, the
adsorption capacity increases with the molecular
weight of the VOC being adsorbed. In addition,
unsaturated compounds are generally more com-
pletely adsorbed than saturated compounds, and
cyclical compounds are more easily adsorbed than
linearly structured materials. Also, the adsorption
capacity is enhanced by lower operating tempera-
tures and higher concentrations. VOC's character-
ized by low vapor pressures are more easily ad-
sorbed 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 oper-
500 | 5,000 | 50,000
100 1,000 10,000 100,000
Toluene Concentration (ppmv)
ation 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 4-12) 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 break-
through 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 re-
generated 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 dif-
ficult to achieve and economically impractical.
During the last part of the steam regeneration cy-
cle, the hot bed saturated with water vapor 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 distil-
lation systems may be necessary, especially in
cases where the VOC's consist of mixtures of sol-
vents.
In other designs, continuous adsorption can be ac-
complished by fluidized bed adsorption. The fresh
adsorbent flows down the adsorption section that
consists of a series of fluidized trays. The emission
63
-------
Figure 4-12. A typical fixed-bed carbon adsorption system.
Emission Stream
Condenser
Low-Pressure
Steam
stream enters at the bottom of the adsorption sec-
tion and flows upward. The VOC's are progressive-
ly 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 re-
turned 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 ca-
pacity of the carbon for the specific VOC in ques-
tion (as determined from the adsorption isotherm),
(b) operating temperature, (c) adsorption and re-
generation cycle time, (d) amount and type of re-
generant, and (e) contaminants.
The discussion in the following sections will be
based on a fixed-bed carbon adsorption system
with two 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. Worksheets for
calculations are provided in Appendix C.6.
4.6.1 Data Required
The data necessary to perform the calculations
consist of HAP emission stream characteristics pre-
viously compiled on the HAP Emission Stream
Data Form and the required HAP control as deter-
mined by the applicable regulations.
64
Exhaust to
Atmosphere
Example Case
Maximum flowrate, Qe = 15,000 scfm
Temperature, Te = 90°F
Relative humidity, Rhum = 40%
HAP = toluene
Maximum HAP content, HAPe = 1,000 ppmv
Based on the control requirements for the emis-
sion 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 characteristics after
dilution.
In a permit review case for a carbon adsorber, the
following data 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):
Reported removal efficiency, RE^oned- °/°
HAP content, HAPe, ppmv
Emission stream flow rate, Qe, scfm
Adsorption capacity of carbon bed, AC, Ib HAP/100
Ib carbon
Number of beds, IN
Amount of carbon required, Creq, Ib
Cycle time for adsorption, 6ad, hr
Cycle time for regeneration, 0reg, hr
Emission stream velocity through the carbon bed,
Ue, ft/min
Bed depth, Zbed, ft
Bed diameter, Dbed, ft
Steam ratio, St, Ib steam/lb carbon
-------
4.6.2 Pretreatment of the Emission Stream
4.6.2.1 Cooling
Adsorption of VOC's is favored by lower tempera-
tures. 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. (See Appendix B.5, reference 8, for deter-
mining the size of a heat exchanger required for
such applications.
content will be limited to less than 25 percent of the
LEL See Table B.1-2 in Appendix B.1 for a list of
LEL values for several compounds.
Example Case
The HAP concentration of the emission stream is
1,000 ppmv (toluene). This is below 25% of the
LEL for toluene, which is 3,000 ppmv (see Table
B.1-2).
Example Case
The temperature of the emission stream is 90°F,
which is below 100°F. Therefore, cooling is not
necessary.
4.6.2.2 Dehumidification
Since water vapor competes with the VOC's in the
emission stream for adsorption sites on the carbon
surface, emission stream humidity levels exceed-
ing 50 percent (relative humidity) are not desirable.
In this handbook, 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 addi-
tional 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 4.8
where calculation procedures for sizing condensers
are described.
Another alternative for dehumidification is adding
dilution airto the emission stream if the dilution air
humidity is significantly less than that of the emis-
sion stream. However, since this will increase the
size of the adsorber system required, it may not be
cost effective.
Example Case
Since the relative humidity of the emission
stream is less than 50%, dehumidification is not
necessary.
4.6.2.3 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 to 50 percent of the LEL if proper
monitoring and controls are used. In addition, since
high bed temperatures may occur due to heat re-
leased during adsorption, high VOC concentrations
may need to be reduced. The maximum practical
inlet VOC concentration is usually about 10,000
ppmv. In this handbook, it is assumed that the VOC
4.6.3 Carbon Adsorption System Design Variables
Table 4-7 presents suggested values for the design
variables of a carbon adsorber system to achieve a
given outlet HAP concentration. If the emission lim-
it requirement is expressed as removal efficiency,
the outlet HAP concentration can be calculated
from the required removal efficiency and the inlet
HAP concentration.
Table 4-7. Carbon Adsorber System Design Variables (12,13)
Steam Requirement
Outlet Adsorption Degeneration for Regeneration
Concentration3 Cycle Time Cycle Timeb St(lb steam/
HAP0 (ppmv) M-ad(hr) (jLreg (hr) Ib carbon)
70
10-12
0.3
1.0
aEmission stream exiting the carbon adsorber.
bRegeneration cycle also includes the time necessary for drying
and cooling the bed.
For specific applications, other values may be ap-
propriate depending on the emission stream char-
acteristics 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 ca-
pacity is determined from the adsorption isotherm
of the compound under consideration. (See Appen-
dix B.8, reference 8, or references 14 and 15 for
adsorption isotherms for several compounds. Also
see reference 8 or 16 for adsorption capacities at
specific conditions for several compounds.)
Based on the required removal efficiency, deter-
mine the outlet HAP concentration using the fol-
lowing equation:
HAP0 = HAPe(1 -0.01 RE)
(4.6-1)
where:
HAP0 = HAP content of the emission stream exit-
ing the adsorber, ppmv
RE = removal efficiency, %
Next, specify the appropriate values for 0ad, 6req,
and St using Table 4-7.
65
-------
Example Case
Using Equation 4.6-1:
RE = 95 percent
HAPe = 1,000 ppmv
HAP0= 1,000(1 -0.95)
HAP0 = 50 ppmv
Assuming the conditions for HAP0 = 70 ppmv
are approximately the same as those for HAP0
= 50 ppmv, from Table 4-7,
6ad = 2 hrs
ereg = 2 hrs
St = 0.3 Ib steam/lb carbon
4.6.4 Determination of Carbon Adsorber System
Variables
4.6.4.1 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 adsorp-
tion 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 0adr the follow-
ing equation can be used to calculate the carbon
requirements:
Cr
-eq = 2 [1.55x 10"5N6adQe
(HAPe - HAP0)MWHAp/AC]
(4.6-2)
where:
Weq
N
ead
Qe
MWHAP
AC
= carbon requirement, Ib carbon
= number of carbon beds
= adsorption cycle time, hr
= emission stream flow rate, scfm
= molecular weight of HAP, Ib/lb-mole
(for a mixture of HAPs, MWHAp will be
defined as mean molecular weight)
= adsorption capacity of the carbon bed,
Ib HAP/100 Ib carbon
For design purposes, the carbon requirement is
generally multiplied by a factor of 2 as indicated in
Equation 4.6-2.(17) This safety factor is an
allowance for build-up of a heel during regenera-
tion (which results in a reduced capacity); fluctu-
ations 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.1-3 in
Appendix B.1 for molecular weight and boiling
point data for several compounds. For additional
data, see reference 18. If no data are available, use
a conservative value of 5 Ibs HAP/100 Ib carbon for
AC.
As an alternative, Figure 4-13 can be used to deter-
mine carbon requirements. The figure is based on
Equation 4.6-2 and evaluated at HAP0 = 10 and 70
ppmv for several inlet concentrations.
If the emission stream contains a mixture of HAP's,
Equation 4.6-2 should be evaluated using appropri-
ate adsorption isotherms for each component and
then summed to determine Creq.
Example Case
Using Equation 4.6-2:
Assume N = 2
= 2 hrs (from Table 4-7)
= 15,000 scfm
= 1,000 ppmv
= 50 ppmv
Qe
HAP
HAP
MWHAp = MWto,uene = 92 Ib/lb mole
AC = 20-25 Ib toluene/100 Ib carbon
This value is estimated from Figure 4-11 where
adsorption isotherms for toluene are plotted at
different temperatures. To obtain a conservative
estimate for Creq, assume AC = 20:
Creq = [2 x 1.55 x 10'5 x 2 x 2 x 15,000 (1,000 -
50) 92/20]
Creq = 8,100lb
Using Figure 4-13:
For HAPe = 1.000 ppmv, HAP0 = 50 ppmv, and
AC = 20 Ib, Creq/Qe is about 0.55, assuming the
curve for HAP0 = 70 ppmv is applicable. Thus,
Creq = 0.55 x 15,000 = 8,250 Ibs
4.6.4.2 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, calculate the
bed area. Typically, velocities used in industry
range from 50 to 100 ft/min depending on the sys-
tem 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 = Qe,a/Ue
(4.6-3)
where:
Abed = bed area, ft2
Qe-a = emission stream flow rate at actual condi-
tions, acfm
Ue = emission stream velocity, ft/min
In this expression, Qe a is determined as follows:
Qe,a = Qe [Te + 460)/530] (4.6-4)
66
-------
Figure 4-13. Carbon requirement vs. HAP inlet concentration.
MWHAp =100lb/lb-mole
N =2
ead = 2 hr
1,000 5,000 10,000
HAP Inlet Concentration, HAPe (ppmv)
where Qe and Te are the flow rate and temperature
of the emission stream. From the bed area, calcu-
late the bed diameter assuming a circular vessel;
use the following expression:
Dbed = 2 [Abed/Tr]°-5 =1.13 (Abed)°-5 (4.6-5)
where:
Dbed = bed diameter, ft
To calculate the bed depth, determine the volume
occupied by carbon in each bed. Assume a carbon
bed density of pbed, and use the following equa-
tion to calculate the volume of carbon (per bed):
V,
carbon ~~
q/N)/Pb(
ed
(4.6-6)
Having calculated Vcarbon- the bed depth can be
determined as follows:
-bed
= v
carbo
n/A:
'bed
(4.6-7)
where Zbed is the bed depth, ft. Hence, the re-
quired adsorber size for an adsorption cycle time
ead for obtaining RE percent removal efficiency for
an emission stream with flowrate Qe is: De ft (di-
ameter) by Zbed 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 4.6-3, -4, -5, -6, and -7:
= 90°F
= 15,000 scfm
= 15,000[(90 + 460)/530]
= 15,565 acfm
Te
Qe
Qe.a
Qe,a
Ue = 100ft/min
Abed - 15,565/100 = 155.7ft2
Dbed = 1.13x(155.7)°-5 = 14ft
Assume pbed = 30 Ib/ft3
Vcarbon = (8,100/2)/30 = 135ft3
Zbed = 135/155.7 = 0.88 = 1 ft
4.6.4.3 Steam Required for Regeneration
Carbon beds may be regenerated by various
means; the most common regenerant used is
steam. In this handbook, 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 to 70 percent, acts as a carrier
gas for the desorbed VOC's. It is not cost-effective
to achieve complete desorption; acceptable work-
ing capacities of adsorption can be obtained with-
out consuming large quantities of steam. For sol-
vent recovery 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 handbook, it is assumed that with a steam
ratio of 0.3 Ib steam/lb carbon, a HAP outlet con-
centration of 70 ppmv can be achieved after regen-
eration, and with a ratio of 1 Ib steam/lb carbon, a
HAP outlet concentration of 10-12 ppmv can be
achieved (Table 4-7).(11) The regeneration cycle
time, 0reg, 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:
Qs = [St(creq)/(ereg - edry_coo,)]/60
(4.6-8)
where:
Qs = steam flowrate, Ib/min
= cycle time for drying and cooling the
bed, hr
67
-------
Typically, cooling and drying the bed with air can
be carried out in about 15 minutes. Figure 4-14 can
also be used to estimate steam requirements. This
figure is based on 9reg = 2 hrs and edry.cooi =
0.25 hrs.
Steam flow rates based on cross-sectional area of
the bed (Qs/Abed) are generally limited to less than
4 Ib steam/min-ft2 to prevent the carbon from being
fluidized in the bed. If Qs/Abeci exceeds 4, the re-
generation cycle time or the steam ratio may need
to be modified.
Example Case
Using Equation 4.6-8:
RE = 95%
HAP0 = 50 ppmv
St = 0.3 Ib steam/lb carbon (Table 4-7)
reg
= 2 hrs (Table 4-7)
Assuming edry-cooi = 0.25 hrs:
Qs = [0.3 (8,100)7(2 -0.25)1/60
Qs = 23 Ib/min
d = 23/155.7 = 0.15 I b steam/min-ft2
Since Qs/Abed is less than 4 Ib steam/min ft2,
fluidization in the carbon bed is not expected.
Using Figure 4-14:
At Creq = 8,100 Ib and St = 0.3,
Qs = 20 Ib/min
Figure 4-14. Steam requirement vs. carbon requirement.
100
1,000 10,000
Carbon Requirement, Creq (Ib)
4.6.4.4 Condenser
The steam used for regenerating the carbon bed
containing the desorbed VOC's is typically con-
densed. 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 con-
denser size:
ACon - H|0ad/UATLM
(4.6-9)
where:
ACon = condenser surface area, ft2
= condenser heat load, Btu/hr
= overall heat transfer coefficient, Btu/hr-ft2-
°F
ATLM = logarithmic mean temperature difference,
°F
U
and:
ATLM
-[
(Tsti Two) (Tsto TW|)
In [(Tsti - Two)/(Tsto - Twi)]
where:
Tsti = steam inlet temperature, °F
Tsto = condensed steam outlet temperature,
I wo
1 wi
_ ___- p_
= cooling water outlet temperature, °F
= cooling water inlet temperature, °F
In this handbook, it is assumed that following re-
generation, steam will be condensed and sub-
cooled to 100°F with cooling water. Note that Equa-
tion 4.6-9 is an approximate expression;
condensation and subcooling processes are com-
bined and average values are used for U and ATLM-
To calculate H|oad, use the following equation:
H,oad = 1.1 x 60 x Qs [X + CPw (Tstl - T^o)] (4.6-10)
where:
X = latent heat of vaporization, Btu/lb
Cpw = average specific heat of water, Btu/lb-°F
In this expression, the condenser heat load is over-
sized 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.(19) 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.(12)
To determine cooling water requirements, use the
following equation:
Qcooi.w = H,oad/[Cpw (Two - Twi)] (4.6-11)
where:
,i,w = cooling water flow rate, Ib/hr
,i,w can be expressed in terms of gal/min as
follows:
68
-------
Qw = QCOOI.W x [(1/60) x (1/62.43) x 7.48]
= 0.002 X QCOol,w
(4.6-12)
where the factor 62.43 is the density of water and
the factor 7.48 is used for converting from "ft3" to
"gal" basis.
Example Case
Using Equation 4.6-10:
Qs = 23 Ib/min
X = 970 Btu/lb (reference 19)
= 1 Btu/lb-°F (reference 19)
= 212°F
= 100°F
= 1.1 x 60 x 23 [970 + 1 x (212 - 100)]
= 1,642,500 Btu/hr
Using Equation 4.6-9:
U
Ts,i
Tsto
Hload
ATLM
ATLM
= 150Btu/hr-ft2-°F
= [ (212-130)-(100-80) 1
Lln[(212)130)/(100)80)] J
= 44°F
= 1,642,500/( 150x44)
= 250 ft2
Using Equations 4.6-11 and -12:
AT = 50°F
Qw = 0.002 Qcool.w
Qw = 0.002 [1,642,500/11x50)1
Qw = 66 gal/min
4.6.4.5 Recovered Product
To calculate costs, the quantity of recovered prod-
uct that can be sold and/or recycled to the process
has to be calculated. Use the following equation:
Qrec = 60 x [Qe x (HAPe x 10~6)(1/387)
(0.01 RE) MWHAp] (4.6-13)
Qrec = 1.55 x 10~9Qex HAPe x RE xMWHAp
where Qrec 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 4.6-13:
Qe
HAPe
RE
MWHAp
Qrec
= 15,000 scf m
= 1,000ppmv
= 95 %
= 92 Ib/lb-mole
= 1.55 x10"9x 15,000x1,000x95x92
= 200 Ib/hr
4.6.5 Evaluation of Permit Application
Compare the results from the calculated values and
reported values using Table 4-8. If the calculated
Table 4-8. Comparison of Calculated Values and Values
Supplied by the Permit Applicant for Carbon
Adsorption
Calculated
Value Reported
(Example Case)3 Value
Carbon requirement, Creq
Bed diameter, Dbed
Bed depth, Zbed
Steam rate, Qs
Condenser surface area, Acon
Cooling water rate, Qw
Recovered product, Qrec
8,100 Ib
14ft
1 ft
23 Ib/min
250ft2
66gal/min
200 Ib/hr
'Based on Emission Stream 4.
values of Crpn, Dh
Qw, and Qrec are
'req» "-"bed' ^bed/ QS- ^
different from the reported values for these varia-
bles, the differences may be due to the assump-
tions 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 pro-
posed carbon adsorber system may be considered
appropriate based on the assumptions made in this
handbook.
4.7 Absorption
Absorption is an operation in which one or more
components of a gas mixture are selectively trans-
ferred into a relatively nonvolatile liquid. Absorp-
tion of a gaseous component by a liquid occurs
when the liquid contains less than the equilibrium
concentration of the gaseous component. The dif-
ference 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 contact-
ing 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 inor-
ganic compounds include water, mineral oils, non-
volatile 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
69
-------
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 liq-
uid on each plate. Bubble-cap plates have been
widely used; other types of plates include perforat-
ed 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 particu-
late matter (see Section 4.11).
Several different configurations of absorber sys-
tems 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 treat-
ment system or introduced as a process water
stream (see Figure 4-15). 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. An-
other configuration involves using the solvent
(usually water) on a once-through basis and strip-
ping 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 pollu-
tants from gaseous streams depends on several
factors, including (a) solubility of the pollutant in a
given solvent, (b) concentration, (c) temperature,
Figure 4-15. A typical countercurrent packed column absorber
system.
Emission
Stream Outlet
Solvent Inlet
Emission Stream
Inlet
tTTTt
(d) flow rates of gaseous and liquid streams (liquid
to gas ratio), (e) contact surface area, and (f) effi-
ciency of stripping (if solvent is recycled to the
absorber).
Determination of the absorber system variables
(absorber column diameter, height, etc.) is depen-
dent on the individual vapor/liquid equilibrium re-
lationship 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 absorb-
ers are not appropriate for this handbook; there-
fore, 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 ab-
sorbers. For illustration purposes, a simple con-
figuration is chosen for the absorber system: a
packed tower absorber using 2-inch ceramic Ras-
chig rings as the packing material with water used
as the absorbent on a once-through basis. The ef-
fluent from the absorber is assumed to be dis-
charged 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 18, 20, 21, 22, and 23).
As indicated in Chapter 3, absorption is the most
widely used control method for inorganic vapor
emissions; therefore. Emission Stream 5 contain-
ing inorganic vapors will be used in the example
case to illustrate the calculation procedures. Work-
sheets for calculations are provided in Appendix
C.7.
4.7.1 Data Required
The data necessary to perform the calculations
consist of HAP emission stream characteristics pre-
viously compiled on the HAP Emission Stream
Data Form and the required HAP control as deter-
mined by the applicable regulations.
Example Case
Maximum flow rate, Qe = 3,000 scfm
Temperature, Te = 85°F
HAP = ammonia
HAP concentration, HAPe = 20,000 ppmv
Pressure, Pe = 760 mm Hg
Based on the control requirements for the emis-
sion stream:
Required removal efficiency, RE = 98%
Solvent Outlet
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.
70
-------
Absorption system variables at standard condi-
tions (70°F, 1 atm):
Reported removal efficiency, RErep0rted, %
Emission stream flow rate, Qe, scfm
Temperature of emission stream, Te, °F
HAP
HAP concentration, HAPe, ppmv
Solvent used
Slope of the equilibrium curve, m
Solvent flow rate, Lgai, gal/min
Density of the emission stream, pG, Ib/ft3
Schmidt No. for the HAP/emission stream and
HAP/solvent systems, ScG, ScL (To calculate ScG
or ScL, see references 18 or 23 for viscosity, den-
sity, and diffusivity data.)
Properties of the solvent:
Density, pL, Ib/ft3
Viscosity, JJ,L, centipoise
Type of packing used
Packing constants a, b, c, d, e, Y, s, g, r
Column diameter, DCO|Umn, ft
Tower height (packed), HtCO|Umru ft
Pressure drop, APtotai, in H20
4.7.2. 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 de-
termined 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 force for absorption is small; there-
fore, long contact times, tall absorption towers,
and/or high liquid-gas ratios are required for ade-
quate performance (high removal efficiency and/or
low outlet concentrations). Hence, as a conserva-
tive guideline, assume that if m is greater than
about 50 for a given HAP/solvent system at atmo-
spheric pressure, then high removal efficiencies
(—99 percent) are not possible.
4.7.3 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 tem-
perature) are usually known. The composition of
the outlet gas is specified by the control require-
ments. 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 di-
ameter and height to accomplish the absorption
operation) for a selected solvent.
To keep the discussion simple, the following as-
sumptions 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 absorp-
tion column and the equilibrium curve can be ap-
proximated as a straight line). All of the data (e.g.,
packing factors, Schmidt numbers, etc.) required in
the calculation of the design variables can be found
in references 12, 18, 21, 23, 24, or Appendix B.9,
reference 8.
4.7.3.1 Solvent Flow Rate
The quantity of solvent to be used is typically esti-
mated from the minimum liquid-gas ratio as deter-
mined from material balances and equilibrium con-
siderations. As a rule of thumb for purposes of
rapid estimates, it has frequently been found that
the most economical value for the absorption fac-
tor (defined below) will be in the range from 1.25 to
2.0.{20)
AF = Lmo,/m G
(4.7-1]
where:
AF = absorption factor
Lmoi = liquid (solvent) flow rate, Ib-moles/hr
Gmoi = gas stream flow 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/sol-
vent system under consideration. (See references
12, 21 and 24 for equilibrium for specific systems.
For information on other systems, see references
18, 22, and 23). Assuming a value of 1.6 for AF, use
Equation 4.7-1 to calculate the solvent flow rate:
- 1.6 m Gm0|
(4.7-2)
The variable Gmoi can be expressed in terms of Qe
as follows:
Gmo, = 0.155 Qe
(4.7-3)
Note that Lmo| can be converted to gal/hr basis as
follows:
Lgai = [Lmo, x MWsolvent x (1/PL) x 7.48]/60 (4.7-4)
where:
Lgai = solvent flow rate, gal/min
MWSO|vent = molecular weight of solvent, Ib/lb-
mole
PL = density of solvent (liquid), Ib/ft3
The factor 7.48 is used to convert from ft3 to gal
basis. For water as the solvent, pL = 62.43 Ib/ft3
(reference 18) and MWSO|Vent = 18 Ib/lb mole; then:
= 0.036 Lmo,
(4.7-5)
77
-------
Example Case
Using Equations 4.7-2, -3, and -5:
m =1.3
(for the operating conditions in the system, con-
sult reference 21)
Qe = 3,000 scf m
3l —
mol
-mol
Lgal
Lgal
0.155x3,000x60
465 Ib-moles/hr
1.6x1.3x465
970 Ib-moles/hr
0.036 x 970
35 gal/min
4.7.3.2 Column Diameter
Once the gas and liquid streams entering and leav-
ing the absorber column and their concentrations
are identified, flow rates calculated, and operating
conditions (type of packing) determined, the phys-
ical dimensions of the column can be calculated.
The column must be of sufficient diameter to ac-
commodate 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 deter-
mining the column diameter is shown in Figure 4-
16. (12) The procedure to calculate the column di-
ameter is as follows: First, calculate the abscissa
(ABS):
ABS = (L/G) (pc/pj0-5 (4.7-6)
where:
L = solvent flow rate, Ib/hr
G = gas stream flow rate, Ib/hr
pG = density of emission stream, Ib/ft3
The values for the variables L and G can be calcu-
lated by multiplying Lmo, and Gmo| with their re-
spective molecular weights. Then proceed to the
flooding line in Figure 4-16 and read the ordinate
(ORD), and solve the ordinate expression for Garea;f
at flooding:
ORD = [(Garea,f)2 (a/e3)
Thus,
°-2)} /pGpLgc
= {[ORD PG pLgc]/[(a/e3)
(4.7-7)
(4.7-8)
where:
Garea.f
= gas stream flow rate based on column
cross sectional area (at flooding condi-
tions), Ib/ft2-sec
= packing factors (see reference 21)
= viscosity of solvent, centipoises
= gravitational constant, ft/sec2
a,e
IJLL
gc
Assuming f as the fraction of flooding velocity ap-
propriate for the proposed operation, the gas
Figure 4-16. Correlation for flooding rate in randomly packed
towers.(12)
i.o
0.1
^ 0.01
o
0.001
0.01 0.045 0.1 1.0
(L/G)pG/pL)05
10.0
Example Case
Using Equations 4.7-6, -7, -8, -9, -10, and-11:
L = MWso,vent x Lmol = 18x970
= 17,460 Ib/hr
G = MWe x Gmo, = 28.4 x 465 = 13,200 Ib/hr
(see Appendix B.1, reference 8, for
calculating MWe)
PG = 0.071 Ib/ft3 (from ideal gas law at 85°F)
(see reference 21 for calculating pG)
PL = 62.18 Ib/ft3 (reference 18; at 85°F)
ABS = (17,460/13,200) (0.071/62.18)° 5 = 0.045
From Figure 4-16, at ABS = 0.045, the value of
ORD at flooding conditions is about 0.15. For 2-
inch ceramic Raschig rings, from reference 21:
a = 28
e = 0.74
Also,
gc = 32.2 ft/sec2
IXL = 0.85 cp (reference 18, at 85°F)
Thus,
Garea,f = (0.15x0.071 x 62.18 x 32.2) °-5
L [28/(0.74)3] (0.85)0'2 J
,,f = 0.56 Ib/sec-ft2 (at flooding)
Assuming f = 0.60
Garea = 0.60 x 0.56 = 0.34 Ib/sec-ft2
Thus,
ACoiumn =13,200/(3,600x0.34)
AColumn = 10.8ft2
Dcolumn = 1.13(10.8)°'5 = 3.7~ 4 ft
72
-------
stream flow rate (based on cross-sectional area)
can be expressed as:
Garea = f Garea,f (4.7-9)
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:
Aco,umn = g/(3,600 Garea) (4.7-10)
The column diameter is then determined by:
Dcoiumn = (4M(Acolumn)°-5 = 1.13(Acolumn)°'5 (4.7-11)
where:
= column diameter, ft
Figure 4-17. NOG for absorption columns with constant
absorption factor AF.(12)
4.7.3.3 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 NOG or NOL depending on whether the gas film
or liquid film resistance controls the absorption
rate. In emission control applications, gas film re-
sistance will typically be controlling, therefore N0c
will be used in the following calculations.
The expression for the column height (packed) is:
Htcoiumn = NOG HOG (4.7-12)
where:
HtCoiumn = packed column height, ft
NOG = number of gas transfer units (based
on overall gas film coefficients)
HOG = height of an overall gas transfer unit
(based on overall gas film coeffi-
cients), ft
Although the determination of NOG is usually com-
plicated, when dilute solutions are involved, NOG
can be calculated using the following equation:
NOG = In {(HAPe/HAP0)[1 - (1/AF)] + (1/AF)}/
[1-11/AF)] (4.7-13)
This expression is simplified based on the assump-
tion that no HAP is present in the solvent as it
enters the column (see reference 8 for details). Al-
ternatively, use Figure 4-17 directly to determine
NOG-
The variable H0c is generally calculated from the
following equation:
HOG = HG + (1/AF) HL (4.7-14)
where:
HG = height of a gas transfer unit, ft
HL = height of a liquid transfer unit, ft
100
HAPe/HAPn
1,000
10,000
Generalized correlations are available to calculate
HG and HL; these are based on the type of packing
and the gas and solvent flow rates. The correlations
for HG and HL are as follows:(12)
(4.7-15)
(4.7-16)
05
HL = Y(L"VL")S (ScL>° 5
HG = [b (3,600 Garea)c/(L")dl (ScG)
=L
where:
b, c, d, Y, and s = empirical packing constants (see
reference 18)
L" = liquid flow rate, Ib/hr-ft2
JAL" = liquid viscosity, Ib/ft-hr
ScG = Schmidt number for the gas
stream
ScL = Schmidt number for the liquid
stream
The values for ScG and ScL are listed for several
compounds in references 12 and 23 (or see Appen-
dix B.9, reference 8). In the calculations, it is as-
sumed that the effect of temperature on Sc is negli-
gible. The value for the variable L" in this equation
is calculated as follows:
L" = L/Acolumn (4.7-17)
Use the following expression to calculate the total
column height (Ht,otai):(24)
Ht,otai = Htco,umn + 2 + 0.25 Dco,umn (4.7-18)
73
-------
Example Case
1. Calculation of NOG
Using Equation 4.7-13:
HAPe = 20,000 ppmv
HAP0 = 20,000 (1 - 0.98) = 400 ppmv
NOG = ln[(20,000/400) 0.375 + 0.625]/0.375
NOG = 7.9
Using Figure 4-17:
HAPe/HAP0 = 20,000/400 = 50
At AF = 1.6, 1/AF = 0.63, and NOG = 8
2. Calculation of HOG:
Using Equations 4.7-14, -15, -16, and -17:
L" = 17,460/10.8 = 1,617 Ib/hr-ft2
3,600 Garea = 1,224 Ib/hr-ft2
From reference 12, the packing factors are:
b = 3.82
c = 0.41
d = 0.45
Y= 0.0125
s = 0.22
Although 1,224 Ib/hr-ft2 is outside the range
shown in the table, assume that the packing
factors are applicable and the error intro-
duced into the calculations will be negligible.
From reference 12:
ScG = 0.66
ScL = 570
Also,
M.L" = 0.85 x 2.42 = 2.06 Ib/ft-hr
(The factor 2.42 is used to convert from centi-
poise to Ib/ft-hr.)
Hence,
HG = [3.82(1,224)a41/(1,617)°-45](0.66)a5
= 2.06
HL = 0.0125 (1,617/2.06)022(570)05
= 1.29
Using AF = 1.6,
HOG = 2.06 + (1/1.6) 1.29 = 2.87 -2.9
3. Calculation of Htcolumn:
Using Equation 4.7-12:
= 7.9 x 2.9 = 22.9 -23 ft
4. Calculation of Httotai:
Using Equation 4.7-18:
Httotai = 23 + 2 + (0.25 x 4) = 26 ft
For costing purposes, it is necessary to calculate
the column weight. Use the following equation:(24)
Wtcolumn = (48 Dco|umn X Httota|) + 39 (Dco|umn)
(4.7-19)
where:
WtCoiumn = column weight, Ib
Also, to determine packing costs, volume occupied
by the packing material (Vpacking) has to be calculated.
Use the following expression:
^packing = '/1T'4)(lJcolumn' X Htco|umn
"column ~~ "•'**•'>'•'column' ^ "^column
(4.7-20)
Example Case
Using Equation 4.7-19:
'-'column — 4 ft
Httotai = 26 ft
WtCO|umn = (48 x 4 x 26) + 39(4)2
Wtco,umn = 5,600 Ib
Using Equation 4.7-20:
HtColumn = 23 ft
Vpacking = 0.785 X (4)2 X 23
Vpacking = 290 ft3
4.7.3.4 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 consider-
ation in the design of such columns. For a particu-
lar packing, the most accurate data will be those
available from the manufacturer. For purposes of
estimation, use the following correlation:(12)
APa = (g x 10-8) [10(r^L)] (3,600 Garea)2/pG (4.7-21)
where:
APa = pressure drop, Ib/ft2-ft
g, r = packing constants (see reference 12)
The total pressure drop through the column is then
expressed as:
X Ht
column
(4.7-22)
Example Case
Using Equation 4.7-21 :
From reference 12:
g = 11.13
r = 0.00295
Also,
L"
3,600 Garea
pG
PL
Thus,
=1,617 Ib/hr-ft2
= 3,600 x 0.34 =
= 0.071 Ib/ft3
=62. 18 Ib/ft3
1,224 Ib/hr-ft2
1Q-8x 10(0-00295X1-617/62'18>(1,224)2]
APa = 2.8 Ib/ft2-ft
(0.071)
Using Equation 4.7-22:
Htcolumn = 23 ft
AP.otai = 2.8 x 23 = 64.4 Ib/ft2
AP,otai = 64.4/5.2 = 12inH2O
(The factor 5.2 is used to convert from Ib/ft2 to in
H20.)
74
-------
4.7.4 Evaluation of Permit Application
Compare the results from the calculations and the
values supplied by the permit applicant using Table
4-9. The calculated values in the table are based
on the Example Case. If the calculated values of
Lgaii Dcoiumn/ HtCO|Umn, Httota|, APtota| Wtco|umn, and
Vpacking are different from the reported values for
these variables, the differences may be due to the
assumptions involved in the calculations. There-
fore, 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 pro-
posed scrubber system may be considered appro-
priate based on the assumptions made in this
handbook.
Table 4-9. Comparison of Calculated Values and Values
Supplied by the Permit Applicant for Absorption
Calculated
Value Reported
(Example Case)3 Value
Solvent flow rate, Lgai
Column diameter, Dco,umn
Column height, HtCO|umn
Total column height, Httotai
Packing volume, Vpacking
Pressure drop, APtotai
Column weight, WtCO|umn
35gal/min
4ft
23ft
26ft
290 ft3
12inH20
5,600 Ib
aBased on Emission Stream 5.
4.8 Condensation
Condensation is a separation technique in which
one or more volatile components of a vapor mix-
ture are separated from the remaining vapors
through saturation followed by a phase change
(see Figure 4-18). The phase change from gas to
liquid can be accomplished in two ways: (a) the
system pressure may be increased at a given tem-
perature, or (b) the system temperature may be
reduced at constant pressure.
The design and operation of a condenser is signifi-
cantly 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), conden-
sation 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 nonconden-
sible and small quantities of condensible com-
pounds. To separate the condensible component
from the gas stream at a fixed pressure, the tem-
perature of the gas stream must be reduced. The
more volatile a compound (i.e., the lower the nor-
mal boiling point), the larger the amount that can
Figure 4-18. Flow diagram for a typical condensation system
with refrigeration.
Emission Stream Outlet
t
Emission
Stream
Inlet
Condenser
Condensed VOC
Coolant '
Refrigeration
Unit
remain as vapor at a given temperature; hence the
lower the temperature required for saturation (con-
densation).
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 re-
quired for condensation, a refrigeration unit may
be necessary to supply the coolant (see Section
4.8.3.2). The two most common types of condens-
ers used are surface and contact condensers. Sur-
face condensers are usually shell-and-tube heat ex-
changers. 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 condens-
ers 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.
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 25 and 26. In the following discussion,
Emission Stream 6, consisting of a single condens-
ible component and a single noncondensible com-
ponent, will be used to illustrate the calculation
procedure for surface condensers. It will be as-
sumed 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 proce-
dure will involve determining the condensation
temperature required, selection of coolant, and cal-
culation of condenser size and coolant require-
ments.
4.8.1. Data Required
The data necessary to perform the calculations
consist of HAP emission stream characteristics pre-
75
-------
viously compiled on the HAP Emission Stream
Data Form and the required HAP control as deter-
mined by the applicable regulations.
Example Case
Maximum flow rate, Qe = 2,000 scfm
Temperature, Te = 90°F
HAP = styrene
HAP concentration, HAPe = 13,000 ppmv
(corresponding to saturation conditions)
Moisture content, Me = negligible
Pressure, Pe = 760 mm Hg
Based on the control requirements for the emis-
sion stream:
Required removal efficiency, RE = 90%
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. Worksheets for calcu-
lations are provided in Appendix C.8.
Condenser system variables at standard conditions
(70°F, 1 atm):
Reported removal efficiency, REreponed, %
Emission stream flow rate, Qe, scfm
Temperature of emission stream, Te, °F
HAP
HAP concentration, HAPe, ppmv
Moisture content, Me, %
Temperature of condensation, Tcon. °F
Coolant used
Inlet temperature of coolant, TCOOi,i, °F
Coolant flow rate, QCOoiant, Ib/hr
Refrigeration capacity, Ref, tons
Condenser surface area, Acon, ft2
4.8.2 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 ex-
changer prior to the condenser.
Example Case
Since the moisture content of the emission
stream is negligible, no pretreatment is neces-
sary.
4.8.3 Condenser System Design Variables
The key design variable in condenser system de-
sign is the required condensation temperature for a
given removal efficiency or outlet concentration. A
condenser's removal efficiency depends on the na-
ture and concentration of emission stream compo-
nents. For example, compounds with high boiling
points (i.e., low volatility) condense more readily
compared to those with low boiling points. As-
sume, as a conservative starting point, that con-
densation 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 re-
moval efficiency (or outlet concentration) depends
on the vapor pressure of the HAP in question at the
vapor/liquid equilibrium. Once the removal effi-
ciency for a given HAP is specified, the required
temperature for condensation can be determined
from data on its vapor pressure-temperature rela-
tionship. Vapor pressure-temperature data can be
represented graphically (Cox charts) as shown in
Figure 4-19for typical VOC's. The coolant selection
is then based on the condensation temperature re-
quired. See Table 4-10 for a summary of practical
limits for coolant selection.
In a permit evaluation, use Table 4-10 to determine
if the values reported for the condensation tem-
perature (Tcon) and the type of coolant selected are
consistent. Also, check if the coolant inlet tempera-
ture is based on a reasonable approach tempera-
ture (a conservative value of 15°F is used in the
Figure 4-19. Vapor pressure-temperature relationship.
(1) Styrene
(2) Toluene
(3) Ethylene dichloride
(4) Hexane
0.1
0.00125 I 0.00175 I 0.00225 I 0.00275
0.00150 0.00200 0.00250 0.00300
[1/(Tcon + 460)] d/°R)
76
-------
Table 4-10. Coolant Selection
Required
Condensation
Temperature
Icon (°F)a
Coolant
Coolant
Temperature
Tcooi, <°F)b
Tcon:c 60-80 Water Tcon-15
60>Tcon>45 Chilled water Tcon-15
45>Tcona-30 Brine solutions (e.g., Tcon-15
calcium chloride,
ethylene glycol)
Chlorofluorocarbons Tcon-15
(e.g., Freon-12)
aAlso emission stream outlet temperature.
"Assume the approach as 15°F.
cSummer limit.
table). If the reported values are appropriate, pro-
ceed with the calculations. Otherwise, the appli-
cant's design is considered unacceptable. The re-
viewer may then follow the calculation procedure
outlined below.
4.8.4 Determining Condenser System Design
Variables
The condenser system evaluated in this handbook
consists of a shell-and-tube heat exchanger with
the hot fluid (emission stream) in the shell side and
the cold fluid (coolant) in the tube side. The emis-
sion stream is assumed to consist of a two-compo-
nent mixture: one condensible component (HAP)
and one noncondensible component (air). Typical-
ly, condensation for such a system occurs non-
isothermally. To simplify the calculations, it is as-
sumed that condensation occurs isothermally.
4.8.4.1 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. Calcula-
tions for cases involving mixtures of HAP's and
supersaturated streams are quite complex and will
not be treated here; for additional information,
consult references 25 and 26.
For a given removal efficiency, the first step in the
calculation procedure is to determine the concen-
tration at the outlet of the condenser. Use the fol-
lowing expression:
Ppartial = 760 {(1 - 0.01 RE)/[(1 - (RE X 10~8
x HAPe)]} HAPe x10~6 (4.8-1)
where Ppartjai 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 deter-
mining the temperature at which this condition oc-
curs, the condensation temperature (Tcon) can be
specified. To carry out this calculation, vapor pres-
sure-temperature data for the specific HAP are re-
quired (see Figure 4-19). Such data can be obtained
from references 18 and 27.
Example Case
Using Equation 4.8-1 and Figure 4-19:
HAPe
RE
p
'partial
= 13,000 ppmv (styrene)
= 90%
= 760{[1-(0.01 x90)]/[(1-(90x1Q-8
x 13,000)]} 13,000 x10'6
= 1.0 mm Hg
'partial
For styrene, the value of [1/(Tcon + 460)] corre-
sponding to 1.0 mm Hg in Figure 4-19 is about
0.00208.
Solving forTcon;
4.8.4.2 Selecting the Coolant
The next step is to select the coolant based on the
condensation temperature required. Use Table 4-
10 to specify the coolant type. For additional infor-
mation on coolants and their properties, see refer-
ences 18 and 27.
Example Case
Based on Tcon = 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 18).
4.8.4.3 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, sensi-
ble heat change of the HAP, and the sensible heat
change in the emission stream. The calculation
steps are outlined below:
1. a. Calculate moles of HAP in the inlet emission
stream (Basis: 1 min):
HAPe,m = (Qe/387) HAPe x 10~6 (4.8-2)
The factor 387 is the volume (ft3) occupied by
1 Ib-mole of ideal gas at standard conditions
(70°Fand 1 atm).
77
-------
b. Calculate moles of HAP remaining in the out-
let emission stream (Basis: 1 min):
HAPom = (Qe/387)[1 - (HAPex
[Pvapor/(Pe - Pvapor)]
where Pvapor is equal to Ppartiai
(4.8-3)
c. Calculate moles of HAP condensed (Basis: 1
min):
HAPcon = HAPe.m - HAP
e,m
o,m
(4.8-4)
2. a. Determine the HAP's heat of vaporization
(AH): Typically the heat of vaporization will
vary with temperature. Using vapor pres-
sure-temperature data as shown in Figure 4-
19, AH can be estimated by linear regression
for the vapor pressure and temperature
range of interest. (See references 18 and 19
for details.)
b. Calculate the enthalpy change associated
with the condensed HAP (Basis: 1 min):
= HAPcon [A H +
(Te - Tcon)]
(4.8-5)
where CPHAP is the average specific heat of
the HAP for the temperature interval Tcon - Te
(Btu/lb-mole-°F).
c. Calculate the enthalpy change associated
with the uncondensed HAP (Basis: 1 min):
(Te - Tc
(4.8-6)
d. Calculate the enthalpy change associated
with the noncondensible vapors (i.e., air)
(Basis: 1 min):
Hnoncon = [JQe/387) - HAPe,J
Cpair (Te - Tcon)
(4.8-7)
where Cpair is the average specific heat of air
for the temperature interval Tcon - Te (Btu/lb-
mole-°F).
3. a. Calculate the condenser heat load (Btu/hr) by
combining Equations 4.8-5, -6, and -7:
H
,oad
= 1. 1x60
I |_1 i U \
' nuncon ~ nnoncon'
(4.8-8)
The factor 1.1 is included as a safety factor.
4.8.4.4. Condenser Size
Condenser systems are typically sized based on the
total heat load and the overall heat transfer coeffi-
cient estimated from individual heat transfer + co-
efficients 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 here, the value used for the overall
heat transfer coefficient is a conservative estimate.
Example Case
Using Equations 4.8-2 to -8:
1. a. Qe = 2,000 scfm
HAPe
HAPP
b. P,
vapor
Pe
HAP0
HAP
c. HAPr
o,m
= (2,000/387) 13,000 x 10 6
= 0.06718 Ib-moles/min
= 1.0 mm Hg
= 760 mm Hg
= 5.1008 [1.0/(760-1.0)]
= 0.00672 Ib-moles/min
= .06718-0.00672
= 0.0605 Ib-moles/min
2. a. AH = 17,445 Btu/lb-mole
(see Appendix B.10, reference 8, or refer-
ences 18 and 19)
b. MWHAP = 104.2 Ib/lb-mole
CPHAP = 24 Btu/lb-mole-°F
(extrapolated from data in reference 27)
Hcon = 0.0605(17,445 + 24(90-20)]
Hcon = 1,157Btu/min
c. Huncon = 0.00672x24x(90-20)
Huncon = 11.3Btu/min
d. Cpair = 6.96 Btu/lb-mole-°F
(see reference 3 for details)
Hn0ncon = ((2,000/387) - 0.06718] 6.96 x
(90-20)
Hnoncon = 2,485 Btu/min
= 1.1 x 60 (1,157 + 11.3 +
2,485)
H|0ad = 241,100 Btu/hr
3. a. H
|0ad
For additional information on how to calculate indi-
vidual heat transfer coefficients, consult reference
25.
To size condensers, use the following equation to
determine the required heat transfer area:
Aeon = H,oad/UATLM (4.8-9)
where:
ACOn = condenser (heat exchanger) surface area,
ft2
U = overall heat transfer coefficient, Btu/hr-ft2-
°F
ATLM = logarithmic mean temperature difference,
°F
and:
ATLM =
' ' e ' cool.o' ' I con ' i
con ' cool.i'
In [(Te TCOO| 0) (Tcon TCOO|])J
where:
Te = emission stream temperature, °F
TCOOI.O = coolant outlet temperature, °F
Tcon = condensation temperature, °F
TCooi,i = coolant inlet temperature, °F
78
-------
Assume that the approach temperature at the con-
denser exit is 15°F. In other words, TCOO|, = (Tcon -
15). Also, the temperature rise of the coolant fluid is
specified as 25°F, i.e., TCOO|,0 = (TCOO|,0 + 25) where
TCOOI.O is the coolant exit temperature. In estimating
Acon, the overall heat transfer coefficient can be
conservatively assumed as 20 Btu/hr-ft2-°F; the ac-
tual value will depend on the specific system under
consideration. This is based on reference 26 in
which guidelines on typical overall heat transfer
coefficients for condensing vapor-liquid media are
reported.
Example Case
Using Equation 4.8-9:
T6
' con
' cool, i
' cool.o
ATLM
ATLM
Hload
u
"con
"con
= SOT
= 20°F
= 20- 15 = 5°F
= TCOO|,i + 25 = SOT
f (90 - 30) - (20 -
[ In [(90 - 30)/(20 -
= 32°F
= 241,100Btu/hr
= 20 Btu/hr-ft2-°F
= 241,100/(20x32)
= 375 ft2
5)1
5)] J
4.8.4.5 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:
^coolant = H|oacj/[Cpcoo|ant (TCOO|/0 - Tcoo|j)] (4.8-10)
where:
= coolant flow rate, Ib/hr
= average specific heat of the coolant
over the temperature interval TCOO| ,
to TCOO|,0, Btu/lb-°F
Specific heat data for coolants are available in re-
ferences 18 and 27.
C_P
cooiant
Example Case
Using Equation 4.8-10:
Hload
Tcool.i
' cool.o
^Pcoolant
'-'coolant
'-'coolant
= 241,100 Btu/hr
= 5°F
= SOT
= 0.65 Btu/lb-T (reference 18)
= 241,100/[0.65(35-10)]
= 14,840 Ib/hr
4.8.4.6 Refrigeration Capacity
A refrigeration unit is assumed to supply the cool-
ant at the required temperature to the condenser.
For costing purposes, the required refrigeration ca-
pacity is expressed in terms of refrigeration tons as
follows:
Ref= Hioad/12,000 (4.8-11)
where Ref is the refrigeration capacity, tons.
Example Case
Using Equation 4.8-11:
H|0ad = 241,100 Btu/hr
Ref = 241,100/12,000
Ref = 20 tons
4.8.4.7 Recovered Product
To calculate costs, the quantity of recovered prod-
uct that can be sold and/or recycled to the process
must be determined. Use the following equation:
Qrec = 60 x HAPcon x MWHAp (4.8-12)
where Qrec is the quantity of product recovered,
Ib/hr.
Example Case
Using Equation 4.8-12:
HAPcon = 0.0605 Ib-moles/min
MWHAp = 104.2lb/lb-mole
Qrec =60x0.06065x104.2
Qrec = 378 Ib/hr
4.8.5 Evaluation of Permit Application
Compare the results from the calculations and the
values supplied by the permit applicant using Table
4-11. The calculated values in the table are based
on the Example Case. If the calculated values of
Tcon, coolant type, Acon, QCOO|ant, Ref, and Qrec are
different from the reported values for these varia-
bles, the differences may be due to the assump-
tions involved in the calculations. Therefore, the
reviewer may wish to discuss the details of the
proposed design with the permit applicant.
Table 4-11. Comparison of Calculated Values and Values
Supplied by the Permit Applicant for
Condensation
Condensation temperature, Tcon
Coolant type
Coolant flow rate, Qcoolam
Condenser surface area, Acon
Refrigeration capacity, Ref
Recovered product, Qrec
Calculated
Value
(Example Case)8
20°F
Brine solution
14,840 Ib/hr
375ft2
20 tons
378 Ib/hr
Reported
Value
3 Based on Emission Stream 6.
79
-------
If the calculated values agree with the reported
values, then the design and operation of the pro-
posed condenser system may be considered ap-
propriate based on the assumptions made in this
handbook.
4.9 Fabric Filters
Fabric filter collectors (also known as baghouses)
are one of the most efficient means of separating
particulate matter from a gas stream. Fabric filters
are capable of maintaining mass collection efficien-
cies of greater than 99 percent down to a particle
size approaching 0.3 jjim in most applica-
tions.(28,29,30) This efficiency is largely insensitive
to the physical characteristics of the gas and dust,
and, depending on fabric cleaning method, to the
inlet dust loading.(28,31) Physical limitations of the
fabric materials to the temperature, moisture con-
tent, 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 configu-
ration. The filter fabric, cleaning method, and air-
to-cloth ratio all should be selected concurrently;
choice of these parameters is mutually depen-
dent.(28) 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 meth-
ods.(28) 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 qualita-
tive guidance rather than predictive equations.
Generally, fabric filter design for HAP's is no differ-
ent than fabric filter design for control of any other
type of particulate matter. However, due to the haz-
ards 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 con-
sider selected fabric cleaning methods, and the de-
sign should specify an induced draft fan (i.e., a
negative pressure or suction baghouse) rather than
a forced draft fan (i.e., a positive pressure bagh-
ouse). Information presented in this section can be
used to provide guidance for or to evaluate 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.
4.9.1 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 deter-
mined 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).
Flow rate, Qe/a = 110,000 acfm
Moisture content, Me = 5% vol
Temperature, Te = 400°F
Particle mean diam. = 1.0 (Jtm
S03 content = 200 ppm (vol)
Particulate content = 3.2 grains/scf - flyash
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:
Filter fabric material
Cleaning method
Air-to-cloth ratio, ft/min
Baghouse configuration
The design criteria and considerations discussed in
this section will be used to evaluate the reasonable-
ness of the applicant's proposed design.
4.9.2 Pretreatment of the Emission Stream
As discussed in Section 3.3.1, the temperature of
the emission stream should be within 50° to 100°F
above the stream dew point. Procedures for deter-
mining the dew point of an emission stream are
provided in Appendix B.1. 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 reference 32. 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 tempera-
ture 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
80
-------
(20 to 30 |xm), pretreatment with mechanical dust
collectors is typically performed. (Appendix B.11,
reference 8, further describes the use of mechani-
cal dust collectors.)
4.9.3 Fabric Filter System Design Variables
Successful design of a fabric filter depends on the
proper selection 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 parti-
cles by fibers. Systems vary, however, in certain
key details of construction and in the operating
parameters. The design variables of particular in-
terest are filter bag material, fabric cleaning meth-
od, 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 secon-
dary, 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. Because
HAP control is similar to particulate control in gen-
eral, 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 Station-
ary Sources—Volume 2, or in the Mcllvaine Fabric
Filter Manual.(31,33) (Note: Because these design
variables are considered concurrently, the Example
Case is presented at the end of Section 4.9.3.)
4.9.3.1 Fabric Type
Several types of natural and synthetic fabric are
used in baghouse systems. Gas stream characteris-
tics such as temperature, acidity, alkalinity, and
particulate matter properties (e.g., abrasiveness
and hygroscoposity), determine the fabric type to
be used.(28,34) In many instances, several fabric
types will be appropriate, and a final selection 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 com-
mercial 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.(28,31,34) Synthetics can operate at tem-
peratures up to 550°F and generally have greater
chemical resistance.(28,31,34) Therefore, while the
initial cost of the synthetic filter fabric is greater
than the cost for natural fibers, the increased ser-
vice 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 fi-
berglass 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. 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 re-
main low.(28,33) Felted fabrics are thick enough
that a dust cake does not need to remain on the
fabric in order to maintain a good collection effi-
ciency.(28,31) This difference between woven and
felted fabrics has important implications for selec-
tion of fabric cleaning method, as described in Sec-
tion 4.9.3.2. Felted fabrics are more expensive than
woven fabrics.
Table 4-12 presents information on the maximum
continuous operating temperature and resistance
characteristics of commonly used filter fabrics.
Knowing the emission stream characteristics, Ta-
ble 4-12 can be used to select an appropriate fabric
filter type (or types). Although the information pre-
sented is qualitative, Table 4-12 provides a good
basis either for selecting a fabric or for evaluating
the appropriateness of a fabric in a permit applica-
tion.
When a number of fabrics are suitable for an appli-
cation, the relative cost of the fabrics may be the
key decision criterion. In general, fluorocarbon and
nylon aromatic bags are the most expensive, fol-
lowed by wool and fiberglass. The remaining com-
monly used synthetics are generally less expensive
than fiberglass (polypropylene, polyester, acrylic,
nylon polyamide, and modacrylic), while cotton is
generally the least expensive fabric.(29,30,31,35)
4.9.3.2 Cleaning Method
As dust accumulates on the filtering elements, the
pressure drop across the bag compartment in-
creases until cleaning of the bags occurs. A timer
can be used to control the cleaning cycle or pres-
sure 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 re-
81
-------
Table 4-12. Characteristics of Several Fibers Used in Fabric Filtration8
Resistance0
Fiber
Type"
Cottond
Wool8
Modacrylic8
(Dynel)
Polypropylene6
Nylon Polyamide6
(Nylon 6 & 66)
Acrylic6
(Orion)
Polyester6
(Dacron, Creslan)
Nylon Aromatic6
(Nomex)
Fluorocarbon6
(Teflon, TFE)
Fiberglassd
Max. Continuous
Operating
Temp. (°F)
180
200
175
200
220
260
275
450
500
550
Abrasion
G
F/G
F
E
e
G
E
E
F/G
P,Gh
Mineral
Acids
P
F
VG
E
P
G
G
F
E9
VG
Organic
Acids
G
F
VG
E
F
G
G
G
E9
E
Alkalies
G
P/F
G
E
VG
F
G
VG
E9
P
Solvent
E
G
G
G
E
E
E
E
E9
E
"References 8, 29, 31, 34, 35, and 37. 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.
"Woven or felted fabrics.
'Considered to surpass all other fibers in abrasion resistance.
9The most chemically resistant of all these fibers.
hAfter treatment with a lubricant coating.
versed air flow through the fabric to remove the
dust.(28) These may be used separately or in con-
junction with one another. The three principal
methods used to accomplish fabric cleaning are
mechanical shaking (manual or 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 man-
ufacturer'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.(28)
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.(28) The selection
of a cleaning method may be based on cost, espe-
cially where more than one method is applicable.
Table 4-13 contains a comparison of cleaning
methods. Cleaning methods are discussed individ-
ually 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.(28,30,34) The shaker mechanisms pro-
duce a violent action on the fabric filter bags and, in
general, produce more fabric wear than the other
types of cleaning mechanisms.(30) For this reason,
mechanical shaking is used in conjunction with
heavier and more durable fabric materials, such as
most woven fibers.(30,38) Bags with poor or fair
abrasion ratings in Table 4-12 (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 col-
lection efficiency.
Bags are usually taken off-line for cleaning by me-
chanical shaking so that no gas flows through the
bags being cleaned. Thus, reentrainment of parti-
cles 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 cy-
cles when using mechanical shaking.(28) Further
control efficiency is very high, and, in fact, properly
selected woven fabrics cleaned by mechanical
shaking can provide much greater particle collec-
tion than pulse-jet cleaned felted fabrics in many
applications.(31) For these reasons, mechanical
shaking is a good method to clean fabric filters
controlling emissions containing HAP's.(31)
Reverse air flow cleaning is used to flex or collapse
the filter bags by allowing a large volume of low
pressure air to pass countercurrent to the direction
82
-------
Table 4-13. Comparisons of Fabric Filter Bag Cleaning Methods (31)
Cleaning Method
Parameter
Cleaning On- or Off-line
Cleaning Time
Cleaning Uniformity
Bag Attrition
Equipment Ruggedness
Fabric Type
Filter Velocity
Power Cost
Dust Loading
Maximum Temperature3
Collection Efficiency
Mechanical
Shake
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
Individual
Bags
On-line
Low
Average
Average
Good
Felt
High
High
Very high
Medium
Lower
Pulse-jet
Compartmented
Bags
Off-line
Low
Good
Low
Good
Felt
High
Medium
High
Medium
Lower
aFabric limited.
of normal gas stream flow during filtration.(30,34)
Reverse air is provided either by a separate fan or
by a vent in the fan damper, which allows a back-
wash of air to clean the fabric filters.(30,34) 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.(30) 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 al-
most 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.(30,34) 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 ef-
fective collection on woven fabrics, felted fabrics
are generally used in pulse-jet cleaned fabric fil-
ters.(28) 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 pres-
sure (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 clean-
ing 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.(28,34) This
cost savings may be somewhat counterbalanced
by the greater expense and more frequent replace-
ment required of felted bags, the higher power use
that may occur, and the installation of the fabric
filter framework that pulse-jet cleaning re-
quires.(28,34)
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 emis-
sions can vary by as much as 100 times over a
filtration cycle.(31) Second, although collection effi-
ciencies 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 me-
chanically shaken or reverse air cleaned filters that
have a good cake buildup.(31) In one study, aver-
age outlet concentrations were two to three orders
of magnitude higher for pulse-jet cleaned filters
than for mechanically shaken filters.(31) Third,
emissions from pulse-jet systems are strongly de-
pendent on the inlet concentration; thus, the collec-
tion efficiency rather than the effluent concentra-
tion tends to be relatively constant for fabric filters
using pulse-jet cleaning.(28,31) 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.(29)
In cases of permit evaluation where pulse-jet clean-
ing is believed to be adequate to meet specific
regulations in specific applications, several options
are available to minimize the disadvantages of
83
-------
pulse-jet cleaning. First, pulse-jet filter bags can be
compartmentalized to permit off-line cleaning; ad-
ditional bags must be installed to allow this.(31)
Second, reduced filtration velocity, or pulse inten-
sity, will decrease average outlet concentration.(31)
Third, bags should be flexible, lightweight, and in-
elastic, with uniform pore structure, to obtain maxi-
mum particle collection.(31) These changes, in ef-
fect, 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.
4.9.3.3 Air-to-Cloth Ratio
The air-to-cloth (A/C) ratio, or filtration velocity, is a
traditional fabric filter design parameter defined as
the actual volumetric flow rate (acfm) divided by
the total active, or net, fabric area (ft2). The A/C
ratio is an important indicator of the amount of air
that can be filtered in a given time when consider-
ing the dust to be collected, cleaning method and
fabric to be used, and the characteristics 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 relation-
ship, but rather is based on industry and fabric filter
vendor experience from actual fabric filter installa-
tions. 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 4-14 summarizes the ranges of recommend-
ed 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 recom-
mended for HAP control situations; the A/C ratio
for control of streams containing HAP's will, there-
fore, be fairly low.) In addition to evaluating a par-
ticular fabric filter application, the A/C ratio and the
emission stream flow rate (Qe,a) are used to calcu-
late net cloth area (Anc):
Qe
A/C ratio
where:
Qe,a
= An
(4.9-1)
= emission stream flow rate at actual
conditions acfm
A/C ratio = air-to-cloth ratio, acfm/ft2 or ft/min
Anc = net cloth area, ft2
Net cloth area is the cloth area in active use at any
point in time. Gross cloth area (Atc), by compari-
son, 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
handbook, costing of the fabric filter structure uses
net cloth area, while costing of fabric filter bags
uses gross cloth area. Table 4-15 presents factors
to obtain gross cloth area from net cloth area:
Anc x Factor = A,,
(4.9-2)
where:
Factor = value from Table 4-15, dimensionless
Atc = gross cloth area, ft2
Fabric filters with a higher A/C ratio require fewer
bags to accomplish cleaning, and, therefore, re-
quire 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 require-
ments, etc., may counterbalance to some degree
the savings of high A/C ratio systems.
4.9.3.4 Baghouse Configuration
The basic configuration of a baghouse varies ac-
cording to whether the gases are pushed through
the system by a fan located on the upstream 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 posi-
tive-pressure baghouse; one using induced draft
fans is called a negative-pressure or suction bag-
house. Positive-pressure baghouses may be either
open to the atmosphere or closed (sealed and pres-
sure-isolated from the atmosphere). Negative-pres-
sure 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.(34) 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 oc-
cur, 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.
4.9.3.5 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 S03 or
where liquid-gas contact areas are involved, stain-
less steel may be required. Stainless steel will in-
crease the cost of the fabric filter significantly when
compared to carbon steel.(30) 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.
84
-------
Table 4-14. Recommended Air-to-Cloth (A/C) Ratios for Various Ousts and Fumes by Cleaning Method (28, 36)
A/C Ratios Recommended for Cleaning Method (ft/min)
Dust
or Fume
Abrasives
Alumina
Aluminum
Aluminum Oxide
Asbestos
Bauxite
Blast Cleaning
Carbon
Carbon Black
Chrome
Coal
Coke
Dyes
Fertilizer
Flint
Fly Ash
Foundry
Glass
Graphite
Gypsum
Iron Ore
Iron Oxide
Iron Sulfate
Lead Oxide
Leather
Lime
Limestone
Machining
Manganese
Metal Fumes
Metal Powders
Mica
Paint Pigments
Paper
Perchlorates
Plastics
Polyethylene
PVC
Resin
Silica
Silica Flour
Silicates
Silicon Carbide
Slate
Starch
Talc
Shaker
2.0-3.0
2.25-3.0
3.0
2.0
2.5-4.0
2.25-3.2
3.0 - 3.5
1.2-2.5
1.5-2.5
1.5-2.5
2.0 - 3.0
2.5
2.0
2.0-3.5
2.5
2.0
#
2.5
1.5-3.0
2.0-3.5
2.0-3.5
2.0-3.0
2.0-2.5
2.0-2.5
3.5-4.0
2.0-3.0
2.0 - 3.3
3.0
2.25
1.5
2.0
2.25 - 3.3
2.0
3.5 - 4.0
*
2.0-3.0
#
#
2.0
2.25-2.8
2.0-2.5
*
*
2.5 - 4.0
2.25
2.25
Reverse Air
*
*
*
*
*
#
*
#
1.1 -1.5
#
#
*
*
1.8-2.0
*
2.1 -2.3
#
*
1.5-2.0
1.8-2.0
#
1.5-2.0
1.5-2.0
1.5-1.8
*
1.5-2.0
#
X
*
1.5-1.8
#
1.8-2.0
*
#
*
#
#
#
*
1.2-1.5
*
*
*
*
#
*
Pulse-Jet
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
#
6-9
9-10
9-11
#
10-12
10
7-10
10
7
8-10
7-12
#
9-10
10
12- 14
*
#
*No information available.
4.9.4 Evaluation of Permit Application
Using Table 4-16, compare the results from this
section and the data supplied by the permit appli-
cant. The calculated values are based on the exam-
ple case. As pointed out in the discussion on fabric
filter design considerations, the basic design pa-
rameters are generally selected without the in-
volved, analytical approach that characterizes
many other control systems, such as an absorber
system (Section 4.7). 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 compati-
bility with the gas stream and paniculate condi-
tions and with the other selected parameters. The
following questions should be asked:
1. Is the temperature of the emission stream enter-
ing 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 4-12)?
85
-------
Table 4-15. Factors to Obtain Gross Cloth Area from Net
Cloth Area (30)
Net Cloth Area, Ano
(ft2)
1 - 4,000
4,001 - 12,000
12,001 - 24,000
24,001 - 36,000
36,001 - 48,000
48,001 - 60,000
60,001 - 72,000
72,001 - 84,000
84,001 - 96,000
96,001 - 108,000
108,001 - 132,000
132,001 - 180,000
180,001 +
Factorto Obtain
Gross Cloth Area, Atc
(ft2)
Multiply by 2
Multiply by 1.5
Multiply by 1.25
Multiply by 1.17
Multiply by 1.125
Multiply by 1.11
Multiply by 1.10
Multiply by 1.09
Multiply by 1.08
Multiply by 1.07
Multiply by 1.06
Multiply by 1.05
Multiply by 1.04
Example Case
Table 4-12 indicates that filter fabrics that can
withstand the 400°F emission stream tempera-
ture are nylon aromatic (Nomex), fluorocarbon
(Teflon), and fiberglass. Because there is a high
potential for acid damage (i.e., a high S03 con-
tent), however, Nomex bags should not be con-
sidered. Because HAP's are present, only me-
chanical shaking or reverse air flow cleaning
methods are advisable.
Using Table 4-14 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 possi-
bility. The documents that describe experience
in certain industry applications support the
choice of fiberglass bags with reverse air flow
cleaning.(31,33)
Table 4-16. Comparison of Calculated Values and Values
Supplied by the Permit Applicant for Fabric
Filters
Calculated Value Reported
(Example Case)3 Value
Emission Stream Temp.
Range"
Selected Fabric
Material
Baghouse Cleaning
Method
A/C ratio = Q"'a
Anc
Baghouse Configuration
365°-415°F
Fiberglass or
Teflon
Mechanical shaking
or reverse air flow
2-2.3 ft/min
Negative pressure
3. Is the baghouse cleaning method compatible
with the selected fabric material and its con-
struction; that is, material type and woven or
felted construction (see Section 4.9.3.2 and Ta-
ble 4-13)?
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 4-14)?
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 equa-
tion:
A/C ratio = §§•*-
^"*nc
where:
A/C ratio = air-to-cloth ratio, ft/min
(4.9-3)
"Based on the municipal incinerator emission stream.
bSee Section 3.3.1.
Qe,a = 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?
A particular manufacturer/customer combination
may employ somewhat different criteria in their
selection of design parameters (such as lower an-
nualized 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 efficien-
cies. Further discussions with the permit applicant
are recommended to evaluate the design assump-
tions and to reconcile any apparent discrepancies
with usual practice.
4.3.5 Determination of Baghouse Operating
Parameters
Many times, optimization of a fabric filter's collec-
tion efficiency occurs in the field after construction.
The following discussion does not pertain to the
preliminary design of a fabric filtration control sys-
tem; however, the information presented should
be helpful in achieving and maintaining the desired
collection efficiency for the installed control sys-
tem.
4.9.5.1 Collection Efficiency
A well designed fabric filter can achieve collection
efficiencies in excess of 99 percent, although opti-
mal 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 collec-
tion 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 per-
86
-------
formance include changing the A/C ratio, using a
different fabric, or replacing worn or leaking filter
bags. Collection efficiency can be improved by de-
creasing the frequency of cleaning or allowing the
system to operate over a greater pressure drop
before cleaning is initiated.
4.9.5.2 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 accu-
mulating 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.(29) In actual oper-
ation, variations in pressure drop outside of the
design range may be indicative of problems within
the fabric filter system. Higher than expected pres-
sure differentials 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 bag-
house.
As the dust cake builds up during filtration, both
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 operating costs, within reasonable
limits; (2) to clean bags as gently and/or infrequent-
ly as possible to minimize bag wear and to maxi-
mize 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 bal-
anced using engineering 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 re-
quirements of the system. This shift may require
more frequent manual adjustments to the automat-
ic control to achieve the minimum cleaning re-
quirements.
4.10 Electrostatic Precipitators
Electrostatic precipitators (ESP's) use an electro-
static field to charge particulate matter contained in
the gas stream. The charged particles then migrate
to a grounded collecting surface. The collected par-
ticles are dislodged from the collector surface peri-
odically by vibrating or rapping the collector sur-
face, 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.(32,37) In the single stage precipita-
tor, which may be wet or dry, ionization and collec-
tion are combined, whereas in the two stage preci-
pitator, 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 de-
sign 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 func-
tion of the desired collection efficiency, gas stream
flow rate and particle drift velocity.(32,37,39,40)
Other design details to be estimated by the vendor
include (but are not limited to) expected secondary
voltage and current, electrical sections alignment,
and 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 phys-
ical and chemical properties of the particulate mat-
ter. The theoretical relationship of the drift velocity
to the variables is discussed extensively in the lit-
erature.(32,37,39,40) Unfortunately, there are no
empirical equations readily available to calculate
drift velocity directly from these variables. There-
fore, in determining drift velocity for a given emis-
sion stream, equipment vendors often rely upon
historical data for similar streams and data estab-
lished from pilot plant tests. Published information
on drift velocity (based on design data for actual
installations to represent typical gas characteris-
tics) are available for several industrial emission
streams.(37)
Appendix C.10 provides a worksheet to record the
information obtained during the performance of
the ESP design procedures.
4.10.1 Data Required
The data necessary to perform the design steps
consist of the data characteristics previously com-
piled on the HAP Emission Stream Data Forms and
the required HAP control as determined by the ap-
plicable regulations.
In the case of a permit review for an ESP, the fol-
lowing data should be supplied by the applicant:
Reported collection efficiency, %
87
-------
Reported drift velocity of particles, ft/sec
Reported collection plate area, ft2
The design criteria and considerations discussed in
this section will be used to evaluate the reasonable-
ness of the applicant's proposed design.
Example Case
Electrostatic precipitation was one of the select-
ed control techniques for the municipal inciner-
ator stream. The pertinent data for these proce-
dures are found on the HAP Emission Stream
Data Form (see Figure 3-8).
Flow rate, Qe/a = 110,000 acfm
Emission stream temperature, Te = 400°F
Particulate content = 3.2 grains/scf - flyash
Moisture content, Me = 5% (vol)
HAP content = 10% (mass) cadmium
Drift velocity of particles, Ud = 0.3 ft/s
Collection efficiency, CE = 99.9% mass
4.10.2 Pretreatment of the Emission Stream
As discussed in Section 3.3.1, the temperature of
the emission streams should be within 50° to 100°F
above the stream dew point. Procedures for deter-
mining the dew point of an emission stream are
provided in Appendix B.1. If the emission stream
temperature does not fall within the stated range,
pretreatment (i.e., emission stream preheat or cool-
ing) is necessary. (Methods of pretreatment are
briefly discussed in Appendix B.11, reference 8.)
The primary characteristics affecting ESP sizing are
drift velocity of the particles and flow rate. There-
fore, after selecting a temperature for the emission
stream, the new stream flow rate must be calculat-
ed. 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 be appropri-
ate. If the emission stream contains an appreciable
amount of large particles (20 to 30 (Jim), pretreat-
ment with mechanical dust collectors is typically
performed. (Appendix B.11, reference 8, further de-
scribes the use of mechanical dust collectors.)
4.70.3 ESP Design Variables
Estimating the collection plate area is the important
aspect of sizing an ESP. A secondary consideration
is the material of construction.
4.10.3.1 Collection Plate Area
Although precise specification of collection plate
area is best left to the vendor, an approximate col-
lection plate area can be calculated using the avail-
able drift velocity value for the gas stream.
As noted earlier, collection plate area is a function
of the emission stream flow rate, the paniculate
drift velocity, and desired control efficiency. The
Deutsch-Anderson equation relates these variables
asfollows:(32,37)
x ln <1 -CE)
(4.10-1)
where:
AP =
Qe,a =
Ud =
CE =
collection plate area, ft2
emission stream flow rate at actual condi-
tions as it enters the control device, acfm
drift velocity of particles, ft/s
required collection efficiency, decimal frac-
tion
Published data on drift velocities for a number of
industrial applications are presented in Table 4-17.
When unavailable, a drift velocity value for an in-
dustrial application can be obtained from an ESP
vendor or from literature sources.(30) If no value
for drift velocity is known, 0.30 ft/s for particles of
"average" resistivity (approximately 10 to 2 x 1010
ohm-cm) and 0.10 ft/s for particles having a "high"
resistivity (1011 to 1013 ohm-cm) can be used.(37)
Table 4-17.
Typical Values for Drift Velocity for Various
Particulate Matter Applications (37)
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 (H2SO4) 0.19 to 0.25
Acid Mist (TiO2) 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
Particles with low resistivities impose special de-
sign considerations on an ESP. Such particles (re-
sistivities from 104 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 coat-
ings may be used to reduce reentrainment.(32,37)
Particles with high resistivities also can cause ESP
operating difficulties. High resistivity particles ac-
cumulate on the collection plates and insulate the
collection plate, thus reducing the attraction be-
tween the particles and the collecting plate. In
these cases, oversizing an ESP and more frequent
cleaning or rapping of the collector plates are nec-
essary. 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.
88
-------
Example Case
Flow rate, Qe,a = 110,000 acfm
Drift velocity of particles, Ud = 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
Ap = 42,200ft2 of collection plate area
4.10.3.2 Materials of Construction
The most common material used in ESP construc-
tion is carbon steel. In cases where the gas stream
contains high concentrations of S03 or where liq-
uid-gas contact areas are involved, stainless steel
may be required.(30,32,37,39,40) However, by
keeping the emission stream temperature above
the dew point and by insulating the ESP (the tem-
perature drop across an insulated ESP should not
exceed 20°F) the use of stainless steel should not
be necessary.
4.10.4 Evaluation of Permit Application
Using Table 4-18, compare the results from this
section and the data supplied by the permit appli-
cant. The calculated values are based on the exam-
ple. In evaluating the reasonableness of ESP design
specifications 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 calculated 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.
Table 4-18. Comparison of Calculated Values and Values
Supplied by the Permit Applicant for ESP's
Drift velocity of particles, Ud
Collection efficiency, CE
Collection plate area, Ap
Calculated Value
(Example Case)a
0.30 ft/s
0.999
42,200 ft2
Reported
Value
"Based on the municipal incinerator emission stream.
4.10.5 Determination of ESP Operating
Parameters
Many times, optimization of an ESP's collection
efficiency occurs in the field after construction. The
following discussion does not pertain to the pre-
liminary design of an ESP control system, however,
the information presented should be helpful in
achieving and maintaining the desired collection
efficiency for the installed control system.
4.10.5.1 Electric Field Strength
Current in the form of ions from the charging elec-
trodes actually charge the particles. Once the parti-
cles are charged, the electric field strength deter-
mines the amount of charge on the particles.
Field strength is based on voltage and distance
between the collecting plates and elec-
trodes.(32,37,39) ESP's are usually operated at the
highest secondary voltage practicable with limited
sparking to maximize collection efficiency. Spark-
ing represents an instantaneous drop in voltage,
collapse of the electrostatic field, and momentary
cessation of paniculate collection. Sparking varies
with the density of the gas stream, material collect-
ed on the electrodes, and humidity and tempera-
ture of the gas stream. When automatic controls
are used, ESP's usually operate with a small
amount of sparking to ensure that the voltage is in
the correct range and the field strength is maxi-
mized. Automatic voltage controls can control
sparking to a specified sparking frequency (typical-
ly 50 to 150 sparks per minute per section of
ESP).(37) As the spark rate increases, a greater per-
centage of the input power is wasted in the spark
current. Consequently, less useful power is applied
to the collecting electrode.
4.10.5.2 Cleaning Frequency and intensity
Particles accumulating on the collecting plates
must be removed periodically. In wet ESP's the
liquid flowing down the collector surface removes
the particles.(40) In dry ESP's, the particles are re-
moved by vibrating or rapping the collector plates.
For dry ESP's this is a critical step in the overall
performance because improperly adjusted or oper-
ating rappers can cause reentrainment of collected
particles or sparking due to excessive particulate
buildup on the collection plates or discharge elec-
trodes. In normal operation, dust buildup of 6 to 25
mm is allowed before rapping of a given intensity
is initiated.(32) In this way, collected material falls
off in large clumps that would not be reentrained. If
rapping is initiated more frequently or if the inten-
sity of rapping is lowered, the resulting smaller
clumps of particulate 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.
4.10.5.3 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, bad-
ly adjusted rappers, full or nearly full dust hoppers,
and process upsets. Mechanical difficulties typical-
ly are the result of electrode misalignment or ex-
cessive dust buildup on the electrodes. Basic de-
89
-------
sign problems include undersized equipment,
reentrainment, or high resistivity particles. The per-
mit applicant should carefully examine each of
these items if the ESP is emitting paniculate emis-
sions from his facility that are in excess of permit-
ted levels.
4.11 Venturi Scrubbers
Venturi scrubbers are designed to serve as a con-
trol device for applications requiring very high col-
lection efficiencies of particles generally between
0.5 to 5.0 (Jim 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 scrub-
ber is called the throat. In general, the longer the
throat, the higher the collection efficiency at a giv-
en pressure drop, provided the throat is not so long
that frictional losses become significant.(42) Typi-
cally, a liquid (usually water) is introduced up-
stream of the throat and flows down the converg-
ing 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."(42) 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 im-
paction occurs causing the droplets to agglomer-
ate. 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.11 provides a worksheet to record the
information obtained during the performance of
the venturi scrubber design procedures.
4.11.1 Data Required
The data necessary to perform the design steps
consist of the HAP emission stream characteristics
previously compiled on the HAP Emission Stream
Data Forms, and the required HAP control as deter-
mined by the applicable regulations.
In the case of a permit review for a venturi scrub-
ber, the following data should be supplied by the
applicant:
Reported pressure drop across venturi, in H20
Performance curve applicable to the venturi
scrubber
Reported collection efficiency, %
Example Case
A venturi scrubber was one of the selected con-
trol techniques for the municipal incinerator
emission stream. The pertinent data for these
procedures are found on the HAP Emission
Stream Data Form.
Flow rate Qe/a = 110,000 acfm
Temperature, Te = 400°F
Moisture content, Me = 5% vol
Required collection efficiency, CE = 99.9%
Particle mean diameter, Dp = 1.0 JJUTI
Paniculate content = 3.2 grams/scf flyash
HAP content = 10% (mass) cadmium
4.11.2 Pretreatment of the Emission Stream
As discussed in Section 3.3.1, the temperature of
the emission stream should be within 50° to 100°F
above the stream dew point. Procedures for deter-
mining the dew point of an emission stream are
provided in Appendix B.1. 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 reference 32 and Appendix
B.11, references.) 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 func-
tion of the emission stream temperature (Te) and
flow rate at actual conditions (Qe,a).(42) 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
standard industrial equations. The use of pretreat-
ment mechanical dust collectors may also be ap-
propriate, particularly if a "nonwetted" venturi
scrubber is used.
4.11.3 Venturi Scrubber Design Variables
To design a venturi scrubber, any one of three
paths may be chosen: (1) rely on previous experi-
ence 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 parti-
cle 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 construc-
tion.
90
-------
4.11.3.1 Venturi Scrubber Pressure Drop
Performance curves are typically logarithmic plots
relating venturi collection efficiency, pressure drop,
and particle size.(23,32,43,44) Collection (control)
efficiency is usually plotted versus pressure drop
across the venturi (APv) for a particle mean diame-
ter (Dp). Figure 4-20 is a plot of venturi scrubber
pressure drops for a given collection efficiency and
particle mean diameter for venturi scrubbers man-
ufactured by a specific vendor. Thus, if the particle
mean diameter for an emission stream and re-
quired collection efficiency is known, the pressure
drop across the venturi can be estimated. Figure
4-20 is representative of plots likely to be used by
vendors, and does not necessarily represent char-
acteristics for all venturi scrubbers.
Estimating the pressure drop gives an indication of
whether a venturi scrubber is a feasible control
device for a given stream. Venturi scrubbers are
used in applications where pressure drops of be-
tween 10 and 80 inches water gauge occur across
the venturi. Venturi scrubbers can operate at pres-
sure drops higher than 80 inches; however, in gen-
eral, a pressure drop exceeding 80 inches H20 indi-
cates that a venturi scrubber will have difficulty
collecting the particles.(42) Therefore, if the pres-
sure drop indicated on the performance curve is
greater than 80 inches H20, assume that the ven-
turi scrubber cannot accomplish the desired con-
trol efficiency.
Table 4-19 lists typical pressure drops for venturi
scrubbers for a variety of applications. The pres-
sure drops are listed to provide general guidance
for typical values that occur in industry. The values
are not meant to supersede any specific informa-
tion known, and given application may have a pres-
sure drop outside those listed in Table 4-19.
Figure 4-20. Venturi scrubber collection efficiencies.
o
E
o>
DC
c
01
u
CD
Q-
99.9
99.8
99.5
99
98
97
96
95
90
80
70
60
50
40
30
20
10
rv .X" Venturi Pressure Drop
^' (in H20)
xxx
x x x^
I
I I
I
0.1
0.2
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Size of Particles (Aerodynamic Mean Diam.),
2.0
3.0
4.0 5.0
91
-------
Table 4-19. Pressure Drops for Typical Venturi Scrubber Applications (42)
Application
Boilers
Pulverized coal
Stoker coal
Bark
Combination
Recovery
Incinerators
Sewage sludge
Liquid waste
Solid waste
Municipal
Pathological
Hospital
Kilns
Lime
Soda Ash
Potassium chloride
Coal Processing
Dryers
Crushers
Dryers
General spray
Food spray
Fluid bed
Mining
Crushers
Screens
Transfer points
Pressure drop
(in H20)
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
25
6-20
20-60
20-30
20-30
6-20
6-20
6-20
Application
Iron and Steel
Cupolas
Arc furnaces
BOF's
Sand systems
Coke ovens
Blastfurnaces
Open hearths
Nonferrous metals
Zinc smelters
Copper and brass
smelters
Sinter operations
Aluminum reduction
Phosphorus
Phosphoric acid
Wet process
Furnace grade
Asphalt
Batch plants — dryer
Transfer points
Glass
Container
Plate
Borosilicate
Cement
Wet process kiln
Transfer points
Pressure drop
(in H20)
30-50
30-50
40-60
10
10
20-30
20-30
20-50
20-50
20
50
10-30
40-80
10-15
6-10
25-60
25-60
30-60
10-15
6-12
"Source: Reference 1.
4.11.3.2 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 de-
gree, the temperature of the gas stream. For any
given application, a vendor should be contacted to
ensure correct selection of materials. A venturi
scrubber will generally be constructed of either car-
bon or stainless steel or a nickel alloy; it may also
be lined with another material (e.g., ceramics). Ta-
ble 4-20 lists materials of construction for various
industries and is intended to serve as a general
guide rather than a definitive statement 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 n,m; therefore:
APV = 47 in H20 (Figure 4-20)
Since the estimated venturi pressure drop value
of APV is not greater than 80 in H20, this venturi
scrubber should be able to accomplish the de-
sired control efficiency. Table 4-20 indicates the
venturi scrubber should be constructed of 316L
stainless steel.
4. 1 1.4 Sizing of Venturi Scrubbers
If a venturi scrubber is found to be a feasible con-
trol 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 saturat-
ed gas flowrate (Qe,s).(45) Vendors may use either
parameter; the cost data presented in Chapter 5 are
based on Qea. However, more current cost curves
based on Qes may be available; therefore, Qes
should be calculated. A psychrometric chart (Figure
4-21) can be used to determine the saturated gas
temperature (Tes), and Qes can then be calculated
using the following formula:
Qe,s = Qe,a x (T
e,s
460)/(Te + 460)
(4.11-1)
where:
Qe/s = saturated emission stream flow rate, acfm
= temperature of the saturated emission
stream, °F
1 e,s
4.11.5 Evaluation of Permit Application
Using Table 4-21, compare the results of this sec-
tion and the data supplied by the permit applicant.
The calculated values in the table are based on the
example. Compare the estimated APV and the re-
ported pressure drop across the venturi, as sup-
plied by the permit applicant.
92
-------
Table 4-20. Construction Materials for Typical Venturi Scrubber Applications"
Application
Boilers
Pulverized coal
Stoker coal
Bark
Combination
Recovery
Incinerators
Sewage sludge
Liquid waste
Solid waste
Municipal
Pathological
Hospital
Kilns
Lime
Soda Ash
Potassium chloride
Coal Processing
Dryers
Crushers
Dryers
General spray dryer
Food spray dryer
Fluid bed dryer
Mining
Crushers
Screens
Transfer points
Construction
Material
31 6L stainless steel
31 6L stainless steel
Carbon steel
31 6L stainless steel
Carbon steel or 31 6L stainless steel
316L stainless steel
High nickel alloy
316L stainless steel
31 6L stainless steel
High nickel alloy
Carbon steel or stainless steel
Carbon steel or stainless steel
Carbon steel or stainless steel
304 stainless steel or 31 6L stainless steel
Carbon steel
Carbon steel or stainless steel
Food-grade stainless steel
Carbon steel or stainless steel
Carbon steel
Carbon steel
Carbon steel
Application
Iron and Steel
Cupolas
Arc furnaces
BOF's
Sand systems
Coke ovens
Blast furnaces
Open hearths
Nonferrous Metals
Zinc smelters
Copper and brass
smelters
Sinter operations
Aluminum reduction
Phosphorus
Phosphoric acid
Wet process
Furnace grade
Asphalt
Batch plants — dryer
Transfer points
Glass
Container
Plate
Borosilicate
Cement
Wet process kiln
Transfer points
Construction
Material
304-31 6L stainless steel
31 6L stainless steel
Carbon steel (ceramic lined)
Carbon steel
Carbon steel
Carbon steel (ceramic lined)
Carbon steel (ceramic lined)
Stainless steel or high nickel
Stainless steel or high nickel
Stainless steel or high nickel
High nickel
316L stainless steel
31 6L stainless steel
Stainless steel
Carbon steel
Stainless steel
Stainless steel
Stainless steel
Carbon steel or stainless steel
Carbon steel
Example Case
Determine 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 H20/lb dry air, decimal
fraction:
(Me/100) (18/29) = (5/100) (18/29) =
0.031 Ib H20/lb dry air
Tes = 127°F (Figure 4-21)
Qe,s = (110,000) x
75,000 acfm
(127 + 460)/(400 + 460) =
Table 4-21.
Comparison of Calculated Values and Values
Supplied by the Permit Applicant for Venturi
Scrubbers
Calculated Value
(Example Case)3
Reported
Value
Particle mean diameter, Dp 1.0 n,m
Collection efficiency, CE 0.999
Pressure drop across venturi, APV 47 in H20
"Based on the municipal incinerator emission stream.
If the estimated and reported values differ, the dif-
ferences may be due to the applicant's use of an-
other performance chart, or a discrepancy 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 differ-
ences between the estimated and reported values
for APV, the design and operation of the system can
be considered appropriate based on the assump-
tions employed in this handbook.
93
-------
Figure 4-21. Psychrometric chart, temperature range 0° - 500°F, 29.92 in Hg pressure.
0
50
100 150
200 250 300
Temperature, °F
350
400
4.12 References
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 Or-
ganic Chemical Manufacturing Industry —
Background Information for Proposed Stan-
dards. Draft EIS. October 1984.
3. Hougen, O.A., K. M. Watson, and R. A. Ragatz.
Chemical Process Principles Part I: Material
and Energy Balances. Asia Publishing House.
Bombay. 1962.
4. U.S. EPA. Parametric Evaluation of VOC/HAP
Destruction Via Catalytic Incineration. EPA-
600/2-85-041. April 1985.
5. U.S. EPA. Control of Volatile Organic Emis-
sions from Existing Stationary Sources — Vol-
ume I: Control Methods for Surface Coating
Operations. EPA-450/2-76-028. November
1976.
6. U.S. EPA. Guideline Series. Control of Volatile
Organic Compound Emissions From Manufac-
ture of High-Density Polyethylene, Polypropy-
lene, and Polystyrene Resins. EPA-450/3-83-
008. November 1983.
7. U.S. EPA. Afterburner Systems Study. EPA-R2-
72-062. August 1972.
8. U.S. EPA. Evaluation of Control Technologies
for Hazardous Air Pollutants — Appendices.
EPA-600/7-86-009b (NTIS PB 86-167/038/AS).
October 1985.
9. Federal Register. Volume 50. April 16,1985. pp.
14941-14945.
10. U.S. EPA. Evaluation of the Efficiency of Indus-
trial Flares: Test Results. EPA-600/2-84-095.
May 1984.
11. Chandrasekhar, R., and E. Poulin. Control of
Hydrocarbon Emissions From Cotton and Syn-
thetic Textile Finishing Plants. EPA Contract
No. 68-02-3134. May 1983.
94
-------
12. U.S. EPA. Organic Chemical Manufacturing
Volume 5: Adsorption, Condensation, and Ab-
sorption Devices. EPA-450/3-80-027. December
1980.
13. Parmele, C. S., W. L O'Connell, and H. S. Bas-
dekis. Vapor-phase Adsorption Cuts Pollution,
Recovers Solvents. Chemical Engineering. De-
cember 31, 1979. pp. 58-70.
14. Calgon Corporation, Pittsburgh, Pennsylvania.
In-house data.
15. Tomany, J. P. Air Pollution: The Emissions, the
Regulations, and the Controls. American Else-
vier Publishing Company, Inc. New York.
16. Manzone, R. R., and D. W. Oakes. Profitably
Recycling Solvents From Process Systems.
Pollution Engineering. October 1973. pp. 23-24.
17. Vatavuk, M. W., and R. B. Neveril. Part XIV:
Costs of Carbon Adsorbers. Chemical Engi-
neering. January 24, 1983. pp. 131-132.
18. Chemical Engineer's Handbook. Perry, R. H.,
and C. H. Chilton, eds. Fifth Edition. McGraw-
Hill Book Company. New York. 1973.
19. 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.
20. Treybal, R. E. Mass-Transfer Operations. Third
Edition. McGraw-Hill Book Company. New
York. 1980.
21. Buonicore, A. J., and L. Theodore. Industrial
Control Equipment for Gaseous Pollutants.
Volume I. CRC Press, Inc. Cleveland, Ohio.
1975.
22. Kohl, A., and F. Riesenfeld. Gas Purification.
Second Edition. Gulf Publishing Company.
Houston, Texas. 1974.
23. U.S. EPA. Wet Scrubber System Study, Volume
I: Scrubber Handbook. EPA-R2-72-118a. Au-
gust 1972.
24. Vatavuk, W. M., and R. B. Neveril. Part XIII.
Costs of Gas Absorbers. Chemical Engineering.
October 4, 1982. pp. 135-136.
25. Kern, D. Q. Process Heat Transfer. McGraw-Hill
Book Company, Inc. and Kogakusha Company,
Ltd. Tokyo. 1950.
26. Ludwig, E. E. Volume III. Applied Process De-
sign for Chemical and Petrochemical Plants.
Gulf Publishing Company. Houston, Texas.
1965.
27. Lange's Handbook of Chemistry. Dean, J. A.,
ed. Twelfth Edition. McGraw-Hill Book Com-
pany. New York. 1979.
28. Siebert, P. C. Handbook on Fabric Filtration. IIT
Research Institute. Chicago. April 1977.
29. U.S. EPA. Handbook of Fabric Filter Technol-
ogy, Volume I: Fabric Filter Systems Study.
APTD 0690. December 1970.
30. U.S. EPACapital and Operating Costs of Se-
lected Air Polution Control Systems. EPA-450/
5-80-002. December 1978.
31. The Fabric Filter Manual. The Mcllvaine Com-
pany. Northbrook, Illinois. 1975.
32. Environmental Engineers' Handbook, Volume
II: Air Pollution. Liptak, B. G., ed. Chilton Book
Company. Radnor, Pennsylvania. 1974.
33. U.S. EPA. Control Techniques for Particulate
Emissions from Stationary Sources — Volume
2. EPA-450/3-81-005b. September 1982.
34. U.S. EPA. Control Techniques for Particulate
Emissions From Stationary Sources — Volume
I. EPA-450/3-81-005a. September 1982.
35. Strauss, W. Industrial Gas Cleaning, 2nd Edi-
tion. Pergamon Press, Oxford, England. 1975.
36. U.S. EPA. Procedures Manual for Fabric Filter
Evaluation. EPA-600/7-78-113. June 1978.
37. U.S. EPA. Air Pollution Engineering Manual.
AP-40. May 1973.
38. U.S. EPA. Particulate Control Highlights: Re-
search on Fabric Filtration Technology. EPA-
600/8-78-005d. June 1978.
39. U.S. EPA. A Manual of Electrostatic Precipitator
Technology, Part 1 — Fundamentals. APTD
0610. 1970.
40. Perry's Chemical Engineers' Handbook. Perry,
R. H., and D. Green, eds. Sixth Edition.
McGraw-Hill Book Company. New York. 1984.
41. The Electrostatic Precipitator Manual. The Mcll-
vaine Company. Northbrook, Illinois. 1975.
42. Air Pollution Control and Design Handbook:
Part 2. Cheremisinoff, P. N., and R. A. Young,
eds. Marcel Dekker, Inc. New York. 1977.
43. U.S. EPA. Wet Scrubber Performance Model.
EPA 600/2-77-127. August 1977.
44. U.S. EPA. TI-59 Programmable Calculator Pro-
grams for Opacity, Venturi Scrubbers and Elec-
trostatic Precipitators. EPA-600/8-80-024. May
1980.
45. U.S. EPA. The Cost Digest. Cost Summaries of
Selected Environmental Control Technologies.
EPA-600/8-84-010. October 1984.
95
-------
Chapter 5
Cost Estimation Procedure
5.1 Objective
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 dispos-
al is outside the scope of this handbook; however,
this cost must be included in any rigorous control
cost estimation.) The procedures are presented in a
step-by-step format and illustrated at each step
with cost calculations pertaining to the thermal in-
cinerator system example discussed in Section 4.2.
Blank standard cost calculation worksheets are pro-
vided 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 are outside the scope of
this handbook; however, an EPA report entitled
Identification, Assessment, and Control of Fugitive
Paniculate Emissions (1) can be used for estimat-
ing costs of fugitive emission controls.
5.2 Total Capital Cost
In this handbook the total capital cost includes only
manufacturing area costs; therefore, it excludes
offsite costs. The total capital cost of a control sys-
tem is the sum of direct costs, indirect costs, and
contingency costs. Direct costs include the total
purchased equipment cost (i.e., the major equip-
ment purchased cost plus the auxiliary equipment
purchased cost), instrumentation and controls,
freight and taxes, and installation costs (i.e., foun-
dation and supports, erection and handling, electri-
cal, 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 costs, architect and engi-
neering contractor expenses, contractor fees, con-
struction expenses, and preliminary testing costs.
An example of contingency costs are penalties in-
curred for failure to meet completion dates or per-
formance specifications.
The capital cost estimation procedure presented in
this handbook is for a "factored" or "study" esti-
mate. 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 predeter-
mined 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 equipment cost by esti-
mating the purchased cost of major and auxiliary
equipment; (2) estimate the cost of instrumenta-
tion and controls plus freight and taxes as a per-
centage of the total purchased equipment cost; (3)
estimate 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.
5.2.1 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 presents a list of the de-
sign parameters needed for costing the HAP con-
trol equipment, and it identifies the figure that pre-
sents 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,
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
V0
A
Of.
creq
wtool
Dcol
A
"con
Ref
A
"nc
AP
Qa,a
Cost Curve
Figure No.
5-1
5-2
5-3
5-4b
5-5°
5-6
5-7
5-8
5-9
5-10
5-11
5-12
5-13
5-14
aSee Nomenclature for definitions of variables.
bPackaged carbon adsorbers.
cCustom carbon adsorbers.
97
-------
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 pre-
sented. These cost curves should not be extrapolat-
ed 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 esti-
mate by the ratio of the Chemical Engineering Fab-
ricated Equipment (FE) cost indices for June 1985
and the date of the cost data. For example, if a cost
is given in December 1977 dollars, it is converted to
June 1985 dollars using a factor of 1.49 (336.0
[June 19851/226.2 [Dec. 1977]). Table 5-2 presents
the monthly FE cost indices from December 1977
through June 1986.
Using the specific value for the design variable,
obtain purchased costs from the specific cost curve
for each major control system component. Present-
ed below are brief descriptions of the equipment
costs included in each HAP control cost curve.
The cost curve for thermal incinerators (Figure 5-1)
includes the fan plus instrumentation and control
costs, in addition to the major equipment pur-
chased cost. If the HAP control system includes a
heat exchanger, its cost (Figure 5-2) 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)
provides the cost of an incinerator less catalyst.
Catalyst costs (Table 5-3) and the cost of a heat
exchanger, if applicable, (Fig. 5-2) must be added to
obtain the major equipment purchased cost. All
auxiliary equipment (ductwork, fan, and stack) pur-
chased costs, the cost of instrumentation and con-
trols, and freight and taxes must be added to obtain
the total purchased cost.
Figure 5-1. Prices for thermal incinerators, including fan and
motor, and instrumentation and controls costs. (2)
110
l/> 100
CO
"o
•Q
£ qn
c^ yu
IN
°- 80
vt
- 7n
ft /U
O
(0
1 fin
C DU
'o
c
Rfl
^^,
^r
jr
/
/
/
/
i i i i i i
0 200 400 600 800 1,000 1,200
Combustion Chamber Volume (ft3)
Table 5-2. Chemical Engineering Fabricated Equipment (FE) Cost Indices'
Date FE Date FE
Dec. 1977 226.2 Jan. 1980 273.8
Feb. 1980 276.9
Jan. 1978 226.6 Mar. 1980 277.7
Feb. 1978 233.0 Apr. 1980 289.3
Mar. 1978 233.6 May 1980 290.9
Apr. 1978 237.1 June 1980 291.3
May 1978 237.3 July 1980 296.7
June 1978 237.4 Aug. 1980 297.3
July 1978 238.6 Sept. 1980 298.1
Aug. 1978 243.3 Oct. 1980 301.2
Sept. 1978 243.2 Nov. 1980 302.5
Oct. 1978 243.8 Dec. 1980 304.0
Nov. 1978 244.1
Dec. 1978 245.2 Jan. 1981 305.9
Feb. 1981 307.1
Jan. 1979 245.2 Mar. 1981 314.7
Feb. 1979 252.5 Apr. 1981 321.9
Mar. 1979 253.1 May 1981 321.6
Apr. 1979 253.7 June 1981 322.9
May 1979 258.3 July 1981 325.6
June 1979 259.9 Aug. 1981 325.7
July 1979 262.6 Sept. 1981 326.7
Aug. 1979 264.2 Oct. 1981 330.8
Sept. 1979 266.6 Nov. 1981 329.4
Oct. 1979 271.6 Dec. 1981 328.9
Nov. 1979 272.6
Dec. 1979 273.7 Jan. 1982 324.5
Feb. 1982 323.4
Date FE Date FE
Mar. 1982 324.1 May 1984 334.6
Apr. 1982 327.8 June 1984 333.8
May 1982 329.1 July 1984 335.4
June 1982 327.5 Aug. 1984 335.1
July 1982 327.1 Sept. 1984 335.9
Aug. 1982 326.2 Oct. 1984 335.0
Sept. 1982 326.7 Nov. 1984 335.4
Oct. 1982 325.8 Dec. 1984 336.5
Nov. 1982 324.8
Dec. 1982 325.1 Jan. 1985 336.9
Feb. 1985 336.5
Jan. 1983 324.4 Mar. 1985 336.6
Feb. 1983 327.6 Apr. 1985 338.0
Mar. 1983 326.8 May 1985 336.0
Apr. 1983 326.6 June 1985 336.2
May 1983 327.1 July 1985 336.4
June 1983 327.3 Aug. 1985 335.3
July 1983 327.0 Sept. 1985 335.9
Aug. 1983 327.1 Oct. 1985 335.4
Sept. 1983 328.0 Nov. 1985 335.7
Oct. 1983 327.8 Dec. 1985 336.8
Nov. 1983 328.9
Dec. 1983 330.1 Jan. 1986 332.5
Feb. 1986 319.2
Jan. 1984 331.5 Mar. 1986 317.0
Feb. 1984 333.0 Apr. 1986 310.6
Mar. 1984 332.9 May. 1986 310.9
Apr. 1984 333.8 June 1986b 310.8
aSource: Chemical Engineering, McGraw-Hill Publications.
bPreliminary.
98
-------
Figure 5-2. Prices for thermal oxidation recuperative heat
exchangers. (2)
J52.000-
"5
ro1
o
o
o
000-
800--
600-^
400--
200--
100-
0)
01
C
CO
.C
(J
X
LU
80-
60--
20-
200 500 1,000 10,000 100,000
Recuperative Heat Exchanger Surface Area (ft2)
Example Case
The example thermal incinerator system case
(see Section 4.2) consists of an incinerator with a
combustion chamber volume (Vc) of approxi-
mately 860 ft3 and a primary heat exchanger
with a surface area (A) of approximately 4,200
ft2. 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.)
Figure 5-3. Prices for catalytic incinerators, less catalyst.{4)
1,000
jo 500
"o
-o
O)
C:
CM
o
o
I 100
50
o
S
CD
C
'o
C
1051
1 1 1 1
1 1 1
1 1 1 1
1 1 1
MM
.00 1,000
5,000 10,000
100,000
50,000
Emission Stream Flow Rate, Qe (scfm)
Table 5-3.
Chemical
Unit Costs for Various Materials
(6/85 dollars)
Cost
Refrigerant
(ethylene glycol)
Activated Carbon
Catalyst (platinum-based)
$0.31/lb (8)
$1.92/lb(9)
$2,750/ft3(10)
Two cost curves are presented for carbon ad-
sorbers: Figure 5-4 for packaged carbon adsorbers
and Figure 5-5 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 pur-
chased cost. The cost curve for custom carbon ad-
sorbers does not include the cost of carbon (part of
the major equipment purchased cost), however, it
does include the cost of instrumentation and con-
trols; the cost of carbon is obtained from Table 5-3.
Figure 5-4. Prices for carbon adsorber packages.(5)
u>
jo 90
"o
E 80
CM
C. 70
60
50
40
CD
D)
3
o
30
20
10
j i i i i
01 2345678!
Carbon Weight, Creq (1,000 Ib)
Price includes carbon, beds, fan and motor, and
instrumentation and controls.
10
99
-------
Figure 5-5. Prices for custom carbon adsorbers, less
carbon.(5)
Figure 5-7. Prices for absorber platform and ladders.(6)
-S 800
o
73
1^
C;
CN
O
O
O
Q.
£
o
0)
d
o
.a
2
700
600
500
400
300
200
100
0 20 40 60 80 100 120 140 160 180 200
Carbon Weight, Creq (1,000 Ib)
Price includes beds, instrumentation and controls,
and a steam regenerator.
o
13
I
T3
TJ
JS
T3
C
CD
O
JO
Q.
Hc = 40 ft
Hc = 33 ft
Hc = 27 ft
Column Diameter, DCO|Umn (ft)
All auxiliary equipment (ductwork, fan, and stack)
purchased costs and freight and taxes must be add-
ed to obtain the total purchased cost.
The cost curve for absorbers (Figure 5-6) does not
include the cost of packing, platforms, and ladders.
The cost of platform and ladders (Figure 5-7) and
packing (Table 5-4) 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.
Figure 5-6. Prices for absorber columns.(6)
1,000 (=
_ 500
2?
o
T3
1
O
O
o
0)
.a
100
50
10
i I i M 11
i i i I i i 111
0 5 10 50 100 500 1,000
Column Weight, Wtcolumn (1,000 Ib)
Price includes manholes, skirts, and painting.
Table 5-4. Price of Packing for Absorber Systems (6)
Packing Type and
Material
Packing Diameter, inches
Pall Rings
Carbon steel
Stainless steel
Polypropylene
Berl Saddles
Stoneware
Porcelain
Intalox Saddles
Polypropylene
Porcelain
Stoneware
Packing Rings
Carbon steel
Porcelain
Stainless steel
Cost/Ft3
(6/81 Dollars)
1
24.3
92.1
21.9
28.1
34.5
21.9
19.4
18.2
30.3
13.2
109.0
1.5
16.5
70.3
14.8
21.7
25.6
14.8
13.3
19.8
10.6
82.6
2
15.1
0.8
13.8
—
—
13.6
13.3
12.2
17.0
9.7
22.9
3
_
—
—
—
—
7.0
12.2
11.0
13.9
8.1
—
The cost curve for condensers (Figure 5-8) yields
the total capital cost for cold water condenser sys-
tems. For systems needing refrigerant (ethylene
glycol), the applicable cost from Figure 5-9 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 Sec-
tion 5.3 to calculate annualized operating costs.
The cost curve for a negative pressure fabric filter
(Figure 5-10) does not include the the cost of bags
(Table 5-5), which depend upon type of fabric used.
This cost must be added to obtain the major equip-
ment purchased cost. All auxiliary equipment
(ductwork, fan, and stack) purchased costs, the cost
100
-------
Figure 5-8. Total capital costs for cold water condenser "
systems.(7)
5,000
"o
•a
(fl -
<38
— o
to -
•
« 1,000
CM
5 500
3 I
II
Ul
c
20,000 ft2
Pulse-jet"
Mechanical Shaker
Reverse Air
Dacron
0.40
0.35
0.60
0.25
0.25
Orion
0.65
0.50
0.95
0.35
0.35
Nylon
0.75
0.70
—
0.45
0.45
Nomex
1.15
1.05
1.30
0.65
0.65
Glass
0.50
0.45
—
0.30
0.30
Polypropylene
0.65
0.55
0.70
0.35
0.35
Cotton
0.45
0.40
—
0.40
0.40
'For heavy felt, multiply source by 1.5.
707
-------
Figure 5-11. Prices for insulated electrostatic precipitators.(2)
10,000
5,000
c;
CN
1,000
500
100
i i I i i i i I
i I i
10 50 100 500 1,000
Collection Plate Area, Ap (1,000 ft2)
Figure 5-13. Required steel thicknesses for venturi
scrubbers.(2)
2 5 10 50 100
Emission Stream Flowrate, Qe,a (1,000 acfm)
For use with Figure 5-12,
200
The cost curve for venturi scrubbers (Figure 5-12)
includes the cost of instrumentation and controls,
in addition to the major equipment purchased cost.
This cost curve is based on a venturi scrubber con-
structed from 1/8-inch carbon steel. Figure 5-13 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 yields a price adjust-
ment factor for various steel thicknesses; this fac-
tor is used to escalate the cost obtained from Fig-
ure 5-12. In addition, if stainless steel is required
(see Section 4.11.3.2) multiply the scrubber cost
estimate by 2.3 for 304L stainless steel or by 3.2 for
316L stainless steel. Costs of all auxiliary equip-
ment (ductwork, fan, and stack) and freight and
taxes must be added to obtain the total purchased
cost.
Figure 5-12. Prices for venturi scrubbers.(2)
60
50
40
30
0.
-------
5.2.2.1 Ductwork Purchase Cost
The ductwork purchase cost is typically proportion-
al to the ductwork weight, which is a function of: (1)
the material of construction, (2) length, (3) diame-
ter, and (4) thickness. Carbon steel ducts are nor-
mally used for noncorrosive flue gases at tempera-
tures below 1,150°F. Stainless steel ducts are
generally used with gas temperatures between
1,1 SOT to 1,500°F, or if the gas stream contains
corrosive materials. Figures 5-15 and 5-16 present
purchase costs for carbon steel and stainless steel
ducts, respectively. It is assumed that the major
portion 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
the ductwork.
Table 5-6. Identification of Design Parameters and
Cost Curves for Auxiliary Equipment
Auxiliary Equipment
Design Parameters
Cost Curve
Figure No.
Ductwork
Fan8
Stack
Diameter 5-15
Length 5-16
Material of construction
Actual air flow rate 5-17
Pressure drop
Gas stream velocity
Diameter 5-18
Length 5-19
Material of construction
a Assumed to be located downstream of the control system and
ductwork.
Figure 5-15. Carbon steel straight duct fabrication price at
various thicknesses.(2)
Without specific information, assume the following
items to simplify 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
duct
= 1 2 -
i9* = 0.3028 (Qe a) (5-1 )
uduct /
where:
Dduct - duct diameter, in
Qea = emission stream flow rate of actual
conditions, acfm
Uduct - velocity of gas stream in duct, ft/min
Example Case
In the example case, since no specific data on
the ductwork are available, use the above as-
sumptions to cost the ductwork. The duct diam-
eter is calculated according to item (3) above.
The emission stream flow rate at actual condi-
tions is approximately 16,500 acfm; therefore,
the duct diameter equals 39 inches.
0.3028 (16,500)* -
D
duct
39 in
The length of the ductwork is assumed to be 100
feet of 3/16-inch thick plate. Since the emission
stream treated contains no chlorine or sulfur
compounds (i.e., it is a noncorrosive emission
stream) and the gas temperature is 960°F, car-
bon steel ductwork is used. From Figure 5-15,
the cost of the ductwork is estimated as follows:
$52/ft x 100 ft x (336.2/226.2) = $7,700
(Note: 12/77 dollars escalated to 6/85 dollars.)
40
80 120 160
Duct Diameter, Dd (in)
200
Figure 5-16. Stainless steel straight duct fabrication price at
various thicknesses.(2)
1,600
1,400
1,200
1,000
CM
~ 800
c
g> 600
400
o
200
20 40 60 80 100 120
Duct Diameter,Dd (in)
140 160
103
-------
5.2.2.2 Fan Purchase Cost
The fan purchase cost (Figure 5-17) 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
(Qfg,a). Control system pressure drop (AP) is the
total of the pressure drops across the various con-
trol system equipment, including the stack and
ductwork. Table 5-7 presents conservative pressure
drops across specific control system components
which can be used if specific data are not available.
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 in 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.)
Figure 5-17. Fan prices.(2)
1,000,000 r
E 100,000 :
CD
DC
§
10,000 :
1,0000
Fan Wheel Diameter
98 in
80 in
60 in
q 100,000
- 10,000
o
T3
- 1,000 <5
100
Performance Range8
Single Width
1 2 4 6 8 10 20 30
Pressure Drop, AP (in H2O)
5 in @ 2,300 fpm to 2-1/2 in @ 3,200 fpm
S-Vi in @ 3,000 fpm to 4-1/4 in @ 4,175 fpm
13-V2 in @ 3,780 fpm to G-% in @ 5,260 fpm
Above Class III specifications
"Performance Range designations are indicated by static pres-
sure (inches of water) at fan outlet velocity (feet per minute).
Notes:
Fan price is lower at higher pressure drop because smaller fan
wheel is used at higher rpm.
For high temperature environment add 3% (250°F to 600°F).
For stainless steel construction multiply price by 2.5.
704
-------
Figure 5-18. Carbon steel stack fabrication price for 1/4 in
plate.(2)
o
•o
I*.
s
tf>
-------
(3) The stack diameter is calculated using a stack
exit velocity of 4000 ft/min. Therefore:
D
stack
= 12 -x
stack
= 0.2141(0^)* (5-2)
where:
Dstack - stack diameter, in
Qfg,a = flue gas flow rate at actual condi-
tions, acfm
= 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 assump-
tions to cost the stack. The stack diameter is
calculated according to item (3) above. The actu-
al gas flow rate exiting the heat exchanger (flue
gas flow rate) equals approximately 40,000
acfm. Therefore, the stack diameter is calculated
as follows:
Ds,ack = 0.2141 (40,000)* = 43 in
With the stack diameter known, use the appro-
priate 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 dol-
lars.)
5.2.3 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 ther-
mal incinerators, carbon adsorbers, and venturi
scrubbers also include the cost of instrumentation
and controls; therefore, this cost must be subtract-
ed to estimate the total purchased equipment cost
for these control devices.) Calculate the total pur-
chased equipment cost for thermal incinerators,
carbon adsorbers, and venturi scrubbers as fol-
lows: (1) multiply the summation of the major
equipment purchased cost and the auxiliary equip-
ment 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 equip-
ment cost is estimated as follows:
(a) Total purchased equipment cost plus cost of
instrumentation and controls (included in
cost curve)—
$145,700 + $104,400 + $7,700 + $6,700 =
$264,500
(b) Cost of instrumentation and controls—
$264,500x0.091 = $24,100
(c) Total purchased equipment cost—
$264,500-$24,100 = $240,400
5.2.4 Estimation of Instrumentation and Controls
Plus Freight and Taxes
For the majority of control equipment, instrumen-
tation costs are a small part of the total purchased
cost. Instrumentation requirements for a control
system depend upon control and safety require-
ments. 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 depends
upon the location of the control system and the
location of the supplier. Without specific data, esti-
mate the cost of freight and taxes at 8 percent of
the total purchased equipment cost.
Example Case
For the example case, the cost of instrumenta-
tion and controls and the cost of freight and
taxes are as follows:
(a) Instrumentation and controls—
$240,400x0.10 = $24,000
(b) Freight and taxes—
$240,400x0.08 = $19,200
5.2.5 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 pur-
chased cost.
Example Case
The total purchased cost for the example case is
as follows:
$240,400 + $24,000 + $19,200 = $283,600
706
-------
5.2.6 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 multiply-
ing the total purchased cost by the appropriate
factor listed in Table 5-8. 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 ex-
pensive (see footnote e, Table 5-8). Each compo-
nent 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
5.3 Annualized Operating Costs
The annualized cost of a control system can be
divided into direct operating costs, indirect operat-
ing costs, and credits. In this handbook, the infla-
Table 5-8. Capital Cost Elements and Factors" (2)
tion 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 ma-
terials 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 equipment
capacity data. Operating labor costs consist of op-
erator labor and supervision, while maintenance
costs consist of maintenance labor and materials.
The direct operating costs are established by esti-
mating annual quantities of utilities consumed and
operator and maintenance labor used and by ap-
plying unit costs to these quantities. The annual
quantities of utilities and labor requirements are
assumed to be proportional to the annual operat-
ing 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 estimat-
ed as applicable, and the cost of replacement labor
is assumed to equal the cost of 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
Control Technique
Cost Elements
DIRECT COSTS
Purchased Equipment Costb
Other Direct Costs:
Foundation and supports
Erection and 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
CONTINGENCY0
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
0.03
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 &
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
"As fractions of total purchased equipment cost. They must be applied to the total purchased equipment cost.
bTotal of purchased costs of major equipment and auxiliary equipment and others, which include instrumentation and controls at
10%, taxes and freight at 8% of the equipment purchase cost.
°Contingency costs are estimated to equal 3% of the total direct and indirect costs.
dFor retrofit applications, multiply the total by 1.25.
707
-------
and supervision costs plus maintenance labor
costs. Property tax, insurance, and administrative
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 appli-
cation, maintenance service, and operating duty.
For costing purposes, preestablished expected life
values are used.
Some control techniques recover the HAP's from a
given emission stream as a salable product. There-
fore, any cost credits associated with the recovered
material must be deducted from the total annua-
lized 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.
5.3.1 Direct Operating Costs
Table 5-9 presents June 1985 unit costs for utilities,
Table 5-9. Unit Costs to Calculate Annualized Cost
Cost Elements
Unit Costs/Factor
DIRECT OPERATING COSTS
1. Utilities:3
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. Materials6
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 (11)
$1.025 per gal (12)
$ 0.0003 per gal (2)
$0.00504 per Ib (2)
$ 0.059 per kWh (13)
As applicable
$11.53 per hour (14)
15% of Operator Labor
$ 11.53 per hour (14)
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
(CRFC) x Total Capital Cost
As applicable
a Refer to Tables 5-10 and 5-11 to estimate utility costs for each
HAP control technique.
bMaintenance materials include operating supplies (e.g., lubri-
cation, paper).
CCRF = capital recovery factor.
For an average interest rate of 10%, the CRF for specific control
devices are listed below.
ESP and fabric filter: CRF = 0.117 (based on 20-year life span).
Venturi scrubber, thermal and catalytic incinerators, adsorber,
absorber, and condenser: CRF = 0.163 (based on 10-year life-
span).
708
operator labor, and maintenance labor as well as
cost factors for other direct operating cost ele-
ments. The procedure used to estimate direct oper-
ating costs (including utilities, direct labor, mainte-
nance, 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 cap-
ital recovery cost for a multiple control device sys-
tem 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
estimate the annual costs for utilities consumed,
operator labor, and maintenance labor.
5.3.1.1 Determine Utility Requirements
The utility requirements for a control system are
obtained from each component's design calcula-
tions. Use the costing information in Table 5-9,
Table 5-10, and Table 5-11 to estimate the total
utility costs. A procedure to estimate fan electricity
costs is provided below, since these costs are appli-
cable to all control techniques. The fan horsepower
requirements are calculated as follows:
Fan horsepower, HP = °-000157 x Qfg,a x AP (5-3)
where:
HP = fan horsepower requirement, hp
AP = pressure drop across the control system,
in H2O
TI = fan efficiency (usually 60-70%)
Assuming a 65 percent fan efficiency and a 10 per-
cent additional capacity requirement for miscella-
neous purposes, and using the conversion factor of
0.746 kilowatt hour per horsepower-hour, estimate
the fan electricity requirement as follows:
(5-4)
FER = 2.0 x 10-4 (Qfg,a) (AP) (HRS)
where:
FER = fan electricity requirement, kWh
HRS = hours of operation per year
5.3.1.2 Determine Remaining Direct Operating
Costs
The remaining direct operating costs include re-
placement parts and labor, operating labor (i.e., the
summation of operator labor and supervision la-
bor), and maintenance (i.e., the summation of
-------
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 4.2, and Table 5-10, the only utility re-
quirement for the thermal incinerator, in addi-
tion 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 TO'4 (40,000 acfm) (7 in) (8,600 hr) =
481,600kWh
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 hr)($0.00425/ft3) =
$723,700
(b) Electricity cost—
481,600 kWh ($0.059/kWh) = $28,400
maintenance labor and materials). Tables 5-9 and
5-10 provide the necessary information to calculate
the cost of replacement parts and labor. Table 5-12
presents available data on estimated labor require-
ments for various control devices. The labor re-
quirements presented as "hours per shift" must be
converted to annual requirements. Total annual op-
erator and maintenance labor costs are obtained by
multiplying the estimated annual labor require-
ments with the applicable unit costs from Table 5-9.
These costs must be determined for each control
device in the control system. Operating labor su-
pervision and maintenance materials are estimated
as a percentage of operator labor and maintenance
labor, respectively. Again, these costs must be de-
termined for each control device if a multiple con-
trol device system is used.
Table 5-10. Utility/Replacement Operating Costs for HAP
Control Techniques"
HAP Control Device
Utilities/Replacement Parts
Thermal Incinerator
Catalytic Incinerator
Carbon Adsorber Systems
Absorber Systems
Condenser System
Fabric Filter Systems
Electrostatic Precipitators
Venturi Scrubbers
Natural gas or fuel oilb
Electricity (fan)
Catalyst costc (Vcat)
Natural gas or fuel oilb
Electricity (fan)
Carbond (Creq)
Steam b
Cooling waterb
Electricity (fan)
Absorbentb (water or solvent)
Electricity (fan)
Refrigerant6 (Ref)
Electricity (fan)
Bags' (Atc)
Electricityb (fan + control device)
Electricity15 (fan + control device)
Water"
Electricity (fan)
a Refer 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.
b See Table 5-11.
cAnnualized replacement catalyst costs are calculated as fol-
lows:
Annualized cost = Vcat
x $/ft3
(Current FE/Base FE)
o years
dAnnualized replacement carbon costs are calculated as fol-
lows:
Annualized cost =
Crea (Ib) x $/lb
5 years
(Current FE/Base FE)
e Refrigerant 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.
'Annualized replacement bag costs are calculated as follows:
Annualized cost = A'c
-------
Table 5-11. Additional Utility Requirements (2)
Fuel Requirement for Incinerators, ft3
(Note: The design sections for thermal and catalytic inciner-
ators 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 HRS
where:
Qf = supplementary fuel required, scfm
MRS = annual operating hours, hr
(Note: Use 8,600 hours unless otherwise specified.)
Steam Requirement for Carbon Adsorber, Ib
(Note: Assume 4 Ib of steam required for each Ib of recovered
product.)
Steam Requirement = 4 (Qrec) x MRS
where:
Qrec - quantity of HAP recovered, Ib/hr
MRS = annual operating hours, hr
(Note: Use 8,600 hours unless otherwise specified.)
Cooling Water Requirement for Carbon Adsorber, gal
(Note: Assume 12 gal of cooling water required per 100 Ibs
steam.)
Water Requirement = 0.48 (Qrec) x HRS
where:
Qrec = quantity of HAP recovered, Ib/hr
HRS = annual operating hours, hr
(Note: Use 8,600 hours unless otherwise specified.)
Absorbent Requirement for Absorbers, gal
(Note: Assume no recycle of absorbing fluid [water or solvent].)
Absorbent Requirement = 60 (Lgai) x HRS
where:
L9ai = absorbing fluid flow rate, gal/min
HRS = annual operating hours, hr
(Note: Use 8,600 hours unless otherwise specified.)
Water Requirement for Venturi Scrubbers, gal
(Note: Assume 0.01 gal H20 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.)
Baghouse Electricity Requirement, kWh
(Note: Assume 0.0002 kW are required per ft2 of gross cloth
area.)
Baghouse Electricity Requirement = 0.0002 (Atc) x HRS
where:
Atc = gross cloth area required, ft2
HRS = annual operating hours, hr
(Note: Use 8,600 unless otherwise specified.)
ESP Electricity Requirement, kWh
(Note: Assume 0.0015 kW are required per ft2 of collection
area.)
ESP Electricity Requirement = 0.0015 (Ap) x HRS
where:
Ap = collection plate area, ft2
HRS = annual operating hours, hr
(Note: Use 8,600 unless otherwise specified.)
Table 5-12. Estimated Labor Hours per Shift and Average
Equipment Life (2)
CRF-
+ i)n]/[(1 + i)n-1]
(5-5)
Labor Requirements
(hr/shift)
Average
Control Device
Operator Maintenance Equipment Life
Labor Labor (yr)
Electrostatic Precipitator 0.5-2
Fabric Filter 2-4
Venturi Scrubber 2-8
Incinerator 0.5
Adsorber 0.5
Absorber 0.5
Condenser 0.5
0.5-1
1 -2
1 -2
0.5
0.5
0.5
0.5
20
20
10
10
10
10
10
5.3.2 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 la-
bor 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 multi-
plying the total capital cost by a capital recovery
factor. The capital recovery factor (CRF) is calculat-
ed as follows:
where:
i = interest rate on borrowed capital, decimal
n = control device life, years
For the purpose of this handbook, an interest rate
of 10 percent is used. Table 5-12 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
lifetimes, respectively. If more than one control de-
vice is used by the control system, use a weighted
average capital recovery factor. A weighted aver-
age capital recovery factor (CRFw) is determined as
follows:
= CRF, [Pd/IPd + PC2)]
CRF2 [PC2/(Pd + PC2)1
(5-6)
where:
CRFi = the capital recovery factor for control
device #1
CRF2 = the capital recovery factor for control
device #2
Pd = the purchased equipment cost for control
device #1
PC2 = the purchased equipment cost for control
device #2
770
-------
Example Case
As estimated in Section 5.2.5, the total capital
cost for the thermal incinerator control system
example is $462,300. Section 5.3.1.2 estimated
that the direct and maintenance labor cost is
$13,300 (i.e., $6,200 + $900 + $6,200). There-
fore, the indirect operating costs are estimated
as follows:
(a) Overhead costs—
$13,300x0.80 = $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 us-
ing Equation 5-5: obtain the expected equip-
ment lifetime for an incinerator ffrom Table
5-12 (10 years), and assume an interest rate
of 10 percent.
[0.1 (1 + 0.1)10]/[(1 + 0.1)10-1] = 0.163
The capital recovery cost is then estimated as
follows:
$462,300x0.163 = $75,400
(f) Total indirect operating costs—
$10,600 + $4,600 + $4,600 + $9,200 +
$75,400 = $104,400
5.3.3 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.
Example Case
For the example case, there are no recovered
products since a thermal incinerator destroys
the organic vapors contained in the emission
stream.
Example Case
Total direct and indirect operating costs for the
example case are $771,600 and $104,400, re-
spectively. There are no recovery credits. Thus,
the net annualized cost of the example HAP con-
trol system is as follows:
$771,600 + $104,400 - $0 = $876,000
5.3.4 Net Annualized Costs
The direct and indirect operating costs less credits
received equal the net annualized cost of the HAP
control system.
5.4 References
1. U.S. EPA. Identification, Assessment, and Con-
trol of Fugitive Paniculate Emissions. Draft fi-
nal report. EPA Contract No. 68-02-3922. April
10, 1985.
2. U.S. EPA. Capital and Operating Costs of Se-
lected 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. Vatavuk, W.M. and Neveril, R.B. "Estimate the
Size and Cost of Incinerators." Chemical Engi-
neering. July 12, 1982.
5. Vatavuk, W.M. and Neveril, R.B. "Costs of Car-
bon Adsorbers." Chemical Engineering. Janu-
ary 23, 1983.
6. Vatavuk, W.M. and Neveril, R.B. "Costs of Gas
Adsorbers." Chemical Engineering. October
12, 1982.
7. U.S. EPA. Organic Chemical Manufacturing.
Volume 5: Adsorption, Condensation, and Ab-
sorption Devices. EPA-450/3-80-077. December
1980.
8. Chemical Marketing Reporter. Schnell Publish-
ing Company. New York, NY. Volume 228. Sep-
tember 16, 1985.
9. Telecon. Charlotte Clark (PES) to David
Chalmers (Calgon Corp.). June 12,1985.
10. Telecon. Gunseli Sagun Shareef (Radian Corp.)
to Roy Uhlman (Engelhart Specialty Chemi-
cals). September 30, 1985.
11. Phillip Gennarelli, Manager, American Gas As-
sociation (AGA). Monthly Gas Utility Statistical
Report. AGA, 1515 Wilson Blvd., Arlington, Va.
22209. July 1985.
12. Bureau of Labor Statistics. August 1985. U.S.
consumer price for fuel oil.
13. Monthly Energy Review. October 1982 electric-
ity costs. DOE/EIA-003583/01. January 1983.
14. Survey of Current Business. November 1982.
Primary Metal Industry Costs. December 1982.
777
-------
Appendix A. 1
Potential HAP's for Solvent Usage Operations
Aliphatic
Hydrocarbons
Aromatic
Hydrocarbons
Halogenated12
Hydrocarbons
Alcohols
Glycols
Ethers
Epoxides
Phenols
1 f
OI 13 0) m
,E £ | 1 8 1 § = |
QS<0?2~-t
sS.y^llaSS
l^att^i^ll
COQO$tOtO(/)l0(/3
Specific Compounds
Cyclohexane •
Generic Compounds
Stoddard solvent •
Alicyclics • •
Specific Compounds
Benzene « •
Napthalene"
Generic Compounds
Specific Compounds
Chloromethane11
Methylene chloride • •
Chloroform"
Carbon tetrachloride • • •
1,1-dichloroethane
Trichloroethylene • •
1,1,1-trichloroethane
Tetrachloroethylene •
Trichlorotrifluoroethane • •
Chlorobenzene
o,p-Dichlorobenzene
Generic Compounds
Halogenated solvents • •
Spec/fie Compounds
Methanol • • •
Ethylene glycol • • • •
Propylene oxide
Cresols
Phenol •
Generic Compounds
Glycols •
Ethers • • • •
Phenols •
Epoxides
01 .±
CO U) »-
"5 ? 5 ^ 3
-S ~ .9 £ o
H 1 1 1 1 1 1 1
* 1= 1 ? 5 i * 3 .?
Or^ fc-O^-Crf"Oo) ^
'S ? 8 v « -c j; o c
-------
I §
V IB !
S;p3
ccu-
o
o
« •=
.5
•5
O)
Ketones Specific Compounds
Aldehydes Formaldehyde
Acetaldehyde11
Furfural
Acetone
Acrolein (propenal)11
Methyl ethyl ketone
Methyl isobutyl ketone
Cyclohexanone
Generic Compounds
Aldehydes
Ketones
Esters
Amides
Specific Compounds
Ethyl acetate
Generic Compounds
Esters
Amides
Nitrosamines
Particulates
Specific Compounds
Cadmium
Chromium
Lead
Zinc
Acids
Nitriles
Specific Compounds
Nitrobenzene11
Generic Compounds
Organic acids
Nitriles
Nitrocompounds
Heterocyclic
Compounds
Specific Compounds
Tetrahydrofuran
Furfural
Generic Compounds
Pyrrolidones
Miscellaneous
Trade Solvents
So Cal I + II
Solvesso 100 + 150
Panasolve
Hi Sol 100
TennecoT-125
774
-------
Appendix B. 1
Gas Stream Parameters Calculations
At many plants, it is common that one pollution
control system serves several emission sources. In
such situations, the combined emission stream pa-
rameters must be calculated from mass and heat
balances. Procedures for calculating the combined
emission stream and single emission stream pa-
rameters listed below are provided in this appen-
dix.
B.1.1 Flow Rate and Temperature
B.1.2 Moisture Content, S03 Content, and Dew
Point
B.1.3 Particulate Matter Loading
B.1.4 Heat Content
B.1.1. Emission Stream Flow Rate and
Temperature Calculations
Only gas volumes at standard conditions (70°F,
1 atm) can be added together. Thus, volumes of all
gas streams must first be converted to volumes at
standard conditions. This calculation is shown be-
low. (Note: It is assumed that the emission streams
are approximately at atmospheric conditions;
therefore, pressure corrections are not necessary.)
Qe1,a X
530
460 + Tel
= 0,
•el
where:
Qei,a = flow rate of gas stream #1 at actual condi-
tions (acfm)
Tei = temperature of gas stream #1 (°F)
Qe1 = flow rate of gas stream #1 at standard condi-
tions (scfm)
This calculation is repeated for each emission
stream which, when combined, will be served by
the control system. The total gas stream volumetric
flow rate at standard conditions (Qe) is calculated
by adding all gas streams, as follows:
Qe1 + Qe2 + ... = Qe
where:
Qe = flow rate of combined gas stream (scfm)
The temperature of the combined gas stream (Te)
must be calculated to convert this combined volu-
metric flow rate at standard conditions (Qe) to actu-
al conditions (Qe,a)-
The temperature of the combined gas stream (Te)
is determined by first calculating the enthalpy (sen-
sible heat content) of each individual stream. The
calculation procedures are shown below.
n 0.018 Btu ,T 7n. _ H
e1 ft3_°F X ( I e1 - /U) - Hs1
where:
Tei = temperature of gas stream #1 (°F)
Hs1 = sensible heat content of gas stream #1 (Btu/
min)
This calculation is repeated for each emission
stream. The total sensible heat is calculated as fol-
lows:
H.i +
... = He
where:
Hs = sensible heat of combined gas stream (Btu/
min)
The combined gas stream temperature (Te) is cal-
culated as follows:
u v
ft3 - °F
ll r
1
_ _ _
s 0.018 Btu Qe e
where:
Te = temperature of combined gas stream (°F)
The actual combined gas stream volumetric flow
rate at actual conditions (Qe,a) is then determined
as follows:
Q
e
460
530
*e,a
where:
Qe,a = flow rate of combined gas stream at actual
conditions (acfm)
B.1.2 Moisture Content, SO3 Content,
and Dew Point Calculations
Moisture content is typically reported as a volume
percent. The calculation procedures require that
the volume percent moisture content of each
stream be converted to a Ib-mole basis, added to-
gether, and then divided by the total combined gas
stream volumetric flow rate (Qe) to obtain the mois-
ture content of the combined gas stream. The
moisture content is calculated below both on a
volume percent and mass percent basis. The mass
basis is to allow for the dew point calculation.
775
-------
The moisture content is converted from a vol %
basis to a Ib-mole basis as follows:
M
1
e1
100%
Qe1 X
Ib-mole
414scf
~ M
el.lm
where:
Me1 = moisture content of gas stream #1 (% vol)
Mei,im = moisture content of gas stream #1 (Ib-
mole/min)
This calculation is repeated for each emission
stream to be combined. The moisture content of
the combined gas stream on a volume percent ba-
sis (Me) is calculated by adding, as follows:
Me1,lm + Me2,|m + .- = Me,|m
where
Me,im = moisture content of combined gas stream
(Ib-mole/min)
Me = moisture content of combined gas stream (%
vol)
The moisture content of the combined stream must
be reported on a mass basis (Me,m) to determine
the dew point. This is calculated as follows:
M
e,lm
18 Ib ..
Ib-mole ~ e-m
where:
Mem = moisture content of combined gas stream
(Ib/min)
The amount of dry air in the combined gas stream
(DAe) is calculated as follows:
Q Ib-mole 29 Ib _ DA
UeX 414scf Ib-mole e
where:
DAe = dry air content of combined gas stream
(Ib/min)
Calculate the psychrometric ratio as follows:
Me,m/(DAe - Me/m) = psychrometric ratio (Ib of wa-
ter/lb dry air)
Knowing the psychrometric ratio and the gas
stream temperature, the dew point temperature is
selected from Table B.1-1.
The presence of sulfur trioxide (S03) in the gas
stream increases the dew point of the stream. If the
S03 component is ignored during the dew point
determination, condensation may occur when not
expected. In addition to the problems associated
with the entrainment of liquid droplets in the gas
stream, the S03 will combine with the water drop-
lets to form su If uric acid, which causes severe cor-
rosion on metal surfaces and deterioration of many
fabrics used in baghouses. Therefore, the determi-
nation of the stream dew point must consider the
presence of S03. With information on the S03 con-
tent (ppm vol) and the moisture content (% vol) of
the gas stream, the "acid" dew point temperature
can be determined from Figure B.1-1. Figure B.1-1
provides dew points for two moisture levels, how-
ever, dew points can be estimated for other mois-
ture values.
The SO3 content of a combined gas stream is cal-
culated by first converting the S03 concentration of
Table B.1-1 Dew Point Temperatures
Psychrometric
Ratio
Gas Stream Temperatures (°F)
70
80
90
100
120
140
160
180
200
220
240
0.000
0.005
0.010
0.015
0.020
0.025
0.030
0.035
0.040
0.045
0.050
0.055
0.060
0.065
0.070
0.075
0.080
0.085
0.090
0.095
0
54
62
68
0
58
65
72
77
0
61
68
75
80
85
89
0
65
71
77
82
87
91
95
98
Dew Point Temperatures i
0
70
77
82
87
91
95
98
101
104
107
109
111
114
116
118
119
0
76
82
86
91
94
98
101
104
107
109
112
114
116
118
120
122
123
124
128
0
81
86
90
94
98
100
104
107
109
112
114
116
118
120
122
123
125
130
140
0
86
90
94
97
101
104
107
109
112
114
116
118
120
122
124
125
130
140
165
0
89
94
97
100
103
107
109
111
114
116
118
120
122
123
125
130
143
162
180
0
93
97
100
103
106
109
110
114
116
118
120
122
124
125
130
140
168
180
205
0
96
100
103
106
109
111
114
116
118
120
122
124
125
130
150
170
182
205
225
776
-------
Figure B.1-1. "Acid" dew points in stack gases.
10
160
Temperature, °C
200
each individual stream to a Ib-mole (Im) basis. The
S03 content is calculated as follows:
S 1 Q x Ib-mole _ g
where:
Sei = S03 content of gas stream #1 (ppm vol)
Sei,im = S03 content of gas stream #1 (Ib-mole/
min)
This is repeated for each separate gas stream.
These are then added to obtain the total S03 con-
tent of the combined gas stream to the control
device as follows:
5e1,lm
Se,lm X
414scf
Ib-mole * QP
= ... = S,
106
e,lm
where:
Se,im = S03 content of combined gas stream (Ib-
mole/min)
Se = S03 content of combined gas stream (ppm
vol)
With information for the S03 content of the com-
bined gas stream (Se) and the moisture content of
the combined gas stream (Me), the acid dew point
is determined from Figure B.1-1.
B.1.3 Participate Matter Loading
Particulate matter concentrations usually are re-
ported in grains per acf. The procedures below may
be used to determine the particulate loading to a
control device (in Ibs/hr) when gas streams are
combined.
Wei,B X Qel,a
v 60 min
A f~ A
hr
Ib
7,000 gr
= WP
where:
Wel/g = particulate loading for gas stream #1
(gr/acf)
We1/i = particulate loading for gas stream #1 (Ib/hr)
This is repeated for each gas stream and the results
are added to obtain the particulate loading for the
combined gas stream.
... = We,,
W
ei.|
W
e2.,
where:
We| = particulate loading for combined gas stream
(Ib/hr)
The particulate loading of the combined gas stream
can be converted to a concentration as follows:
We,,X
7,000 gr „ 1 hr
1
Ib
60 min A Q,
•= W,
•e,a
e,g
where:
W
e,g
= particulate loading for combined gas
stream (gr/acf)
B.I. 4 Heat Content Calculation
The heat content of gas stream #1 (hei) can be
determined from the heat of combustion of its
components using the following equation:
n
he1 = (0.01) 2 ye1il x he1,i
i = 1
where:
he1 = heat content in gas stream #1 (Btu/scf)
yeu = volume percent of component "i" in gas
stream #1 (% vol)
he1 , = heat of combustion of component "i" in gas
stream #1 : see Table B.1-2 (Btu/scf)
n = number of components in gas stream #1
The heat content of a combined emission stream
can be determined from the heat content of the
individual emission streams as follows:
He = (0.01)
m
yej x he
777
-------
where:
hfi -
Yej =
hei =
m =
n-lt
in combined
combined emission stream heat content
(Btu/scf)
volume percent of stream
gas stream (% vol)
heat content of stream "j" in combined gas
stream: see previous discussion (Btu/scf)
number of individual gas streams in com-
bined gas stream
Table B. 1 -2. Heats of Combustion and Lower Exposive Limit
(LEL) Data for Selected Compounds*
Compound
Methane
Ethane
Propane
n-Butane
Isobutane
n-Pentane
Isopentane
Neopentane
n-Hexane
Ethylene
Propylene
n-Butene
1-Pentene
Benzene
Toluene
Xylene
Acetylene
Naphthalene
Methyl alcohol
Ethyl alcohol
Ammonia
Hydrogen sulfide
LEL
(ppmv)
50,000
30,000
21,000
16,000
18,000
15,000
14,000
14,000
11,000
27,000
20,000
16,000
15,000
13,000
12,000
11,000
25,000
9,000
60,000
33,000
160,000
40,000
Net Heat of
Combustion"'0
(Btu/scf)
892
1,588
2,274
2,956
2,947
3,640
3,631
3,616
4,324
1,472
2,114
2,825
3,511
3,527
4,196
1,877
1,397
5,537
751
1,419
356
583
"Sources: Steam/Its Generation and Use, The Babcock & Wilcox
Company. New York, NY. 1975.
Fire Hazard Properties of Flammable Liquids, Gases,
Volatile Solids -1977. National Fire Protection Associ-
ation. Boston, MA. 1977.
lower heat of combustion.
'Based on 70°F and 1 atm.
Example Case
Calculate the heat content of an emission stream
from a paper coating operation (gas stream #1)
with the following composition data: methane
(44 ppmv), toluene (73 ppmv), and others (4
ppmv). Let subscripts "1" and "2" denote meth-
ane and toluene, respectively.
he1 = (0.01) (ye1(1 x he1<1 + yei,2 x hei,2)
Convert the concentrations to volume percent
basis:
Methane: ye1i1 = 0.0048 (assume "others" is
equivalent to meth-
ane)
Toluene: ye1,2 = 0.0073
From Table B.1-2:
Methane: he1/1 = 892 Btu/scf
Toluene: he12 = 4,196 Btu/scf
Substituting these values in the above equation
yields:
he1 = 0.35 Btu/scf
Table B.1-3 Properties of Selected Organic Compounds*
Compound
Acetone
Benzene
n-Butyl acetate
n-Butyl alcohol
Carbon tetrachloride
Chloroform
Cyclohexane
Ethyl acetate
Ethyl alcohol
Heptane
Hexane
Isobutyl alcohol
Isopropyl acetate
Isopropyl alcohol
Methyl acetate
Methyl alcohol
Methylene chloride
Methyl ethyl ketone
Methyl isobutyl ketone
Perchloroethylene
Toluene
Trichlorethylene
Trichlorotrifluoroethane
Xylene
Molecular
Weight
(Ib/lb-mole)
58
78
116
74
154
119
54
88
46
100
86
74
103
60
74
32
85
72
100
166
92
131
187
106
Boiling
Point
(°F)
133
176
257
243
170
142
176
171
173
209
156
225
191
181
135
148
104
175
244
250
231
189
118
281-292
"Source: Chemical Engineer's Handbook. Perry, R.H. and Chil-
ton, C.H. (eds). Fifth Edition. McGraw-Hill Book Com-
pany. New York, NY. 1973.
778
-------
Appendix B.2
Dilution Air Requirements Calculations
B.2.1 Dilution Air Calculations
The quantity of dilution air (Qd) needed to decrease
the heat content of the emission stream to hd is
given by the following equation:
= [(IVhd)-1]Qe
(1)
where:
Qd = dilution air flow rate, scfm
he = emission stream heat content before dilution,
Btu/scf
hd = emission stream heat content after dilution,
Btu/scf
Qe = emission stream flow rate before dilution,
scfm
The concentrations of the various components and
flow rate of the emission stream have to be adjust-
ed after dilution as follows:
02,d = 02 (hd/he) + 21 [1 - (hd/he)] (2)
Me,d = Me (hd/he) + 2 [1 - (hd/he)] (3)
Qe,d = Qe (he/ hd) (4)
where:
O2d = oxygen content of diluted emission stream,
vol %
Me_d = moisture content of diluted emission
stream, vol %
Qe/d = flow rate of the diluted emission stream,
scfm
The factor 21 in Equation 2 denotes the volumetric
percentage of oxygen in air and the factor 2 in
Equation 3 is the volumetric percentage of mois-
ture in air at 70°F and 80 percent humidity.
After dilution, the HAP emission stream character-
istics are redesignated as follows:
02
Me
he
Qe
02,d =
Me,d =
hd =
Qe,d =
. Btu/scf
.scfm
Appendix C.2 is a worksheet for calculating dilution
air requirements.
779
-------
Appendix C. 1
HAP Emission Stream Data Form
121
-------
/I DATA FORM*
«=
UJ
CC
t,
o
t/5
c/>
^
UJ
CL
X
mt Contact
IU
>
C
re
c
f
L.
^
C
0
•z.
01
c
o
x:
c
a
a
r~
^*<
\
a
(7
•-._
K
: c
1 C
i- '£
: n
^
) f
t
CO
:^j
c
O
CJ
>
CJ
L C
03
C
i
> ">
; ±
) C.
•• -*~
5
j
3
>
\
^
J
_a
'>
a:
nr
^^
k.
03
"a
c
D
w
a
t/
1
a
E
. LU
M-
o
^ o
^
^m
-<— •
C
K
. a
< 0
. 4-
) cy
•» ^^
L
i
h.
)
i
1
!
c
c
'4^
re
c
J4^
^"
C
a
"C
4-
C
re
E
—
k.
a
.c
£
z
E
re
a
4-
cr
c
c
'a
a
£
LL
<
C
X
re
!
: a
: ^
- c
It
: C
! S
II
J a
i§
: a
) C
I J2
03
•—
I to
) u
. i£
) c/
: w
1 -2
: U
i S
: 1
: co
J C
c
.c
CC
(/•
Q.
2
£
re
o
k.
t/-
c
c
V
u
E
LL
C
) C
X
re
E
k.
c
LL
!^
o
: o
> J!
i C
'§
I Z
I LL
) 0
2 .a
I CO
*
: *.
: co
> CM
• •
] c
) 1
J 0
) 0
. Q.
J LL
U
-Q
03
j.— i.
c\
*^_-
^
a
' «
CL
0
a
CO
^
a_
I
G
c
^:
CO
CS
r—
**—
>
1
L —
C
CO
CL
<
I
Z
u
_a
(0
cs
-•^^
a
- 2
- Q.
' 03
- >
C
0
w
•a
<
CL
<
I
0
JQ
CO
^— •
- CN
" *
i- a
a
?
CO
Lc3
a:
i
a.
<
X
^
*
#
*
-w
C
o;
c
0
O
0
'c
03
O
CD
oj
'
«!*•
- o-
cs
^ ^.
"—
4-
c
a
•£
c
t-'
2:
mm
$
c
^
^
. -^
*• •
4-<
a
o
U
r
O
4-^
) CC
a
I
a
"
i
- cr
CS
) r-
—
£
-
i «
k.
i a
:f
: h
• _•
<.• .
CO
•t
• «
C
IL.
1
*— «
CS
^
^«
v_
I a
,|
> ~£
1
! «
z
: c
'
0
K
.^~
C/
•>•»
1 1
"E
' 0
CC
_a
re
D .S:
) -£
) C
: <
i z
h
> c!
If
: c
> ^-
U
L a
J >
C/)
T3
C
> .c
4-
'1
> c
i £
1 c
!l
u
i 0
: t/
>
!
)
t
)
)
j
)
»
»
O
4->
0
E
"o
'c
o :
0 LU
T3 =
0 TO
O
0 CD
0 =
CO (J
j- • CO
4^ T3 Q.
o 0 0
^ 0 4>2
c c w
~ co c
>• "c
J-: +3 CD
4^ 4^
C 0 CD i,'
05 0 "°
k. "O "4^ CD
1- § ?
J^ 05 OJ
c a. •- •-,
E < T3 -1
CD T= C c
0 -L 0 o
i; 0 Q- ._
co 0 0 E
T3 £ ^ §
C C^ 0 "^
15 CD ~
E £ a. ^
O i- CO .—
k. 0 0 t?
o E -Q 3
0 T3 -°
OJ Dl 3 ^
.E_0 o g
•*•* >. w Q-
E»- CD <
CD 4™* ^r
CD CO CD
0 CO T3 CO
| § ^ c |
.2 w ^ §'<« ^
co •- 0 ~ .<2 J2
•|S o 8 E § x
E CD c: F 0 ~ ^
0 Q. 42 S CO « ^3
c W <-! ^ CO •= C
^ _ co co 0 £ c
10 CD 05 CO O Ji^ O
The data presented are for
Use extra forms if addition
The numbers in parenthes<
1 = organic vapor proce
2 = inorganic vapor pro
3 = particulate process f
Organic emission stream c
* * *
* *
*
-------
Appendix C.2
Calculation Sheet for Dilution Air Requirements
Dilution airflow rate:
Qd = [(he/hd)-1]Qe
Qd = _ scf m
Diluted emission stream characteristics:
02,d = 02 ( hd/he) + 21 [1 - ( hd/he)]
02,d = _ %
Me,d = Me ( hd/he) + 2 [1 - ( hd/he)]
Me/d = _ %
Qe,d = Qe (he/ hd)
Qe d = _ scf m
Redesignate emission stream characteristics:
02 = 02,d = - %
Me = Me,d = _ %
he = hd = _ Btu/scf
Qe = Qe,d = - scf m
723
-------
Appendix C.3
Calculation Sheet for Thermal Incineration
4.2.1 Data Required
HAP emission stream characteristics:3
1. Maximum flow rate, Qe = scfm
2. Temperature, Te = °F
3. Heat content, he = Btu/scf
4. Oxygen content,6 02 = %
5. Moisture content, Me = %
6. Halogenated organics: Yes No
Required destruction efficiency, DE = %
In the case of a permit review, the following data should be supplied by the applicant:
Thermal incinerator system variables at standard conditions (70°F, 1 atm):
1. Reported destruction efficiency, DEreported = %
2. Temperature of emission stream entering the incinerator,
Te = °F (if no heat recovery);
The = °F (if a heat exchanger is employed)
3. Combustion temperature, Tc = °F
4. Residence time, tr = sec
5. Maximum emission stream flow rate, Qe = scfm
6. Excess air, ex = %
alf dilution air is added to the emission stream upon exit from the process, the data
required are the resulting characteristics after dilution.
bThe oxygen content depends on the oxygen content of the organic compounds (fixed
oxygen) and the free oxygen in the emission stream. Since emission streams treated by
thermal incineration are generally dilute VOC and air mixtures, the fixed oxygen in the
organic compounds can be neglected.
724
-------
7. Fuel heating value , hf = _ Btu/scf (assume natural gas)
8. Supplementary heat requirement, Hf = _ Btu/min
9. Combustion chamber volume, Vc = _ ft3
10. Flue gas flow rate, Qfg = _ scfm
11 . Heat exchanger surface area (if a heat exchanger is employed),
A = _ ft2
4.2.2 Pretreatment of the Emissions Stream: Dilution Air Requirements
Typically, dilution will not be required. However, if the emission stream heat content (he)
is greater than 13 Btu/scf with oxygen concentration greater than 16 percent, see
Appendix C.2 where a blank calculation sheet for determining dilution air requirements
is provided.
4.2.3 Thermal Incinerator System Design Variables
Based on the required destruction efficiency (DE), select appropriate values for Tc and tr
from Table 4-1.
T = °F
i c - r
tr = _ sec
For a permit evaluation, if the applicant's values for Tc and tr are sufficient to achieve the
required DE (compare the reported values with the values presented in Table 4-1),
proceed with the calculations. If the applicant's values for Tc and tr are not sufficient, the
applicant's design is unacceptable. The reviewer may then use the values for Tc and tr
from Table 4-1.
T = °F
ic - r
tr = sec
(Note: If DE is less than 98 percent, obtain information from literature and incinerator
vendors to determine appropriate values for Tc and tr.)
4.2.4 Determination of Incinerator Operating Variables
4.2.4.1 Supplementary Heat Requirements
1. For dilute emission streams that require no additional combustion air:
a. Use Figure 4-2:
Hf = (Hf/Qe)fjgUre Qe
125
-------
Hf = Btu/min
or
b. Use Equation 4.2-1:
+ 0.002 Me) [Cpair (Tc-Tr) - Cpair (The-Tr) -^
f~-f hf-1.4Cpair(Tc-Tr)
The values for the parameters in this equation an be determined as
follows:
Qe, he, Me Input data.
hf Assume a value of 882 Btu/scf if no other information
is available.
Cpair Assume a value of 0.0190 Btu/scf-°F if no other infor-
mation is available.
Tc Obtain value from Table 4-1 or from permit applicant.
The Use the following equation if the value for The is not
specified:
The = (HR/100) Tc + [1 - (HR/100)] Te
where HR = heat recovery in the heat exchanger (per-
cent). Assume a value of 50 percent for HR if no other
information is available.
Tr 70°F
Hf = Btu/min
If Hf is less than 5 Btu/min, redefine Hf = 5 Btu/min.
2. For emission streams that are not dilute and require additional combustion
air:
Use Figure 4-3 to obtain a conservative estimate:
Hf = (Hf/Qe)figure Qe
Hf = Btu/min
726
-------
4.2.4.2 Flue Gas Flow Rate
1. For dilute emission streams, use Equation 4.2-2:
Qfg = Q6 + Qf + QC
where:
and
Qf = Hf/hf
Qfg = scf m
2. For emission streams that require additional combustion air, use the follow-
ing equation to calculate Qc:
Qc = [(0.01 He + 9.4 Qf) (1 + 0.01 ex) - 0.0476 02 Qe]
He = Qehe
Assume ex = 18 percent if no other information is available.
Qc = scf m
Then use Equation 4.2-2 to calculate Qfg:
Qfg = scf m
4.2.5 Combustion Chamber Volume
a. Use Equation 4.2-4 to Convert Qfg (standard conditions) to Qfg/a
(actual conditions):
Qfg,a = Qfg [(Tc + 460)/530]
(Note: Pressure effects are negligible.)
Qfg/a = acfm
b. Use Equation 4.2-5 to calculate combustion chamber volume:
Vc= [(Qfg,a/60)tr]1.05
Obtain value for tr from Table 4-1 or from permit applicant.
Vc = ft3
127
-------
If Vc is less than 36 ft3 (minimum commercially available size),
Vr = 36 ft3
' c
4.2.6 Heat Exchanger Size
1. For dilute emission streams that do not require additional combustion air:
a. Use Figure 4-4:
A = ft2
b. Use Equation 4.2-6:
A = [60 Qe (1 + 0.002 Me) Cpair (The - Te)]/U ATLM
The values for the parameters in this equation can be determined as
follows:
Qe, Cpair, The, Me As specified for Equation 4.2-1.
Te Input data.
U Use a value of 4 Btu/hr-ft2-°F unless the inquirer/
applicant has provided a value.
Tc As specified for Equation 4.2-1.
ATLM Calculate ATLM using the following expression:
ATLM = Tc - The
ATLM = °F
Heat exchanger surface area:
A = ft2
2. For emission streams that are not dilute and require additional combustion
air:
Use Figure 4-5 :
A = (A/Qe)figure Qe
A = ft2
728
-------
4.2.7 Evaluation of Permit Application
Compare the calculated values and reported values using Table 4-2. The combustion
volume (Vc) is calculated from flue gas flow rate (Qfg) and Qfg is determined by emission
stream flow rate (Qe), supplementary fuel flow rate (Qf), and combustion air requirement
(Qc). Therefore, if there are differences between the calculated and reported values for Vc
and Qfg, these are dependent on the differences between the calculated and reported
values for Qc and Qf.
If the calculated and reported values are different, the differences may be due to the
assumptions involved in the calculations. Discuss the details of the design and operation
of the system with the applicant.
If the calculated and reported values are not different, then the design and operation of
the system can be considered appropriate based on the assumptions employed in this
handbook.
Table 4-2. Comparison of Calculated Values and Values Supplied by the Permit
Applicant for Thermal Incineration
Calculated Reported
Value Value
Supplementary heat requirement, Hf ... ...
Supplementary fuel flow rate, Qf ... ...
Flue gas flow rate, Qfg ... ...
Combustion chamber size, Vc ... ...
Heat exchanger surface area, A ... ...
723
-------
Appendix C.4
Calculation Sheet for Catalytic Incineration
4.3.1 Data Required
HAP emission stream characteristics:3
1. Maximum flow rate, Qe = scfm
2. Temperature, Te = °F
3. Heat content, he = Btu/scf
4. Oxygen content13, 02 = %
5. Moisture content, Me = %
Required destruction efficiency, DE %
In the case of a permit review, the following data should be supplied by the applicant:
Catalytic incinerator system variables at standard conditions (70°F, 1 atm):
1. Reported destruction efficiency, DEreported = %
2. Temperature of emission stream entering the incinerator,
Te = °F (if no heat recovery),
The = °F (if emission stream is preheated)
3. Temperature of flue gas leaving the catalyst bed,
T = °F
i co r
alf dilution air is added to the emission stream upon exit from the process, the data
required are the resulting characteristics after dilution.
bThe oxygen content depends on the oxygen content of the organic compounds (fixed
oxygen) and the free oxygen in the emission stream. Since emission streams treated by
catalytic incineration are generally dilute VOC and air mixtures, the fixed oxygen in the
organic compounds can be neglected.
730
-------
4. Temperature of combined gas stream (emission stream + supplementary
fuel combustion products) entering the catalyst bed,3
5. Space velocity, SV = hr"1
6. Supplementary heat requirement, Hf = Btu/min
7. Flow rate of combined gas stream entering the catalyst bed,
Qcom = scf m
8. Combustion air flow rate, Qc = scfm
9. Excess air, ex = %
10. Catalyst bed requirement, Vbed = ft3
11. Fuel heating value, hf = Btu/scf
12. Heat exchanger surface area (if a heat exchanger is employed),
A = ft2
4.3.2 Pretreatment of the Emission Stream: Dilution Air Requirements
For emission streams treated by catalytic incineration, dilution air typically will not be
required. However, if the emission stream heat content is greater than 10 Btu/scf for air
+ VOC mixtures or if the emission stream heat content is greater than 15 Btu/scf for inert
+ VOC mixures, dilution air is necessary. For emission streams that cannot be character-
ized as air + VOC or inert + VOC mixtures, assume that dilution air will be required if the
heat content is greater than 12 Btu/scf. In such cases, refer to Appendix C.2 where a blank
calculation sheet for determining dilution air requirements is provided.
4.3.3 Catalytic Incinerator System Design Variables
Based on the required destruction efficiency (DE), specify the appropriate ranges for Tci
and Tco and select the value for SV from Table 4-3.
Tci (minimum) = 600°F
Tco (minimum) = 1,000°F
Tco (maximum) = 1,200°F
SV = hr1
alf no supplementary fuel is used, the value for this variable will be the same as that for
the emission stream.
737
-------
In a permit review, determine if the reported values for Tci, Tco, and SV are appropriate to
achieve the required destruction efficiency. Compare the applicant's values with the
values in Table 4-3 and check if:
Tci (applicant) > 600°F and 1,200°F > Tco (applicant) > 1,000°F
and
SV (applicant) < SV (Table 4-3)
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
values in Table 4-3.
4.3.4 Determination of Incinerator Operating Parameters
4.3.4.1 Supplementary Heat Requirements
1. For dilute emission streams that require no additional combustion air:
a. Use Equation 4.3-1 to determine if Tci = 600°F from Table 4-3 is
sufficient to ensure an adequate overall reaction rate without dam-
aging the catalyst, i.e., check if Tco falls in the interval 1,000° -
1,200°F:
Tco = 600 + 50 he
T = °F
i co r
If Tco fa||s in the interval 1,000° - 1,200°F, proceed with the calcula-
tions. If Tco is less than 1,000°F, assume Tco is equal to 1,000°F and
use Equation 4.3-2 to determine an appropriate value for Tci; and
then proceed with the calculations:
Tci = 1,000-50he
T . - _ °p
i c, - r
(Note: If Tco is greater than 1,200°F, decline in catalyst activity may
occur due to exposure to high temperatures.)
b. Use Figure 4-7 to determine supplementary heat requirements:
Hf = (Hf/Qe)figure Qe
Hf = _ Btu/min
c. Use Equation 4.3-3 to determine supplementary heat requirements:
732
-------
Hf - 1.1 hfQe(1 + 0.002 Me)
r[Cpair(Tci-Tr)-Cpair(The-Tr)]l
L h, - 1.4 Cpair (Tci - Tr) J
The values for the variables in this equation can be determined as
follows:
Qe, Me, Te Input data.
hf Assume a value of 882 Btu/scf (for natural gas) if no
other information is available.
Cpair Assume a value of 0.0190 Btu/scf-°F if no other
information is available.
Tci Obtain value from part a above or from permit
applicant.
The For no heat recovery case, The = Te. For heat recov-
ery case, use the following equation if the value for
The is not specified:
The = (HR/100)TCO + [1 - (HR/100)] Te
where HR = heat recovery in the heat exchanger
(percent). Assume a value of 50 percent for HR if no
other information is available.
Tr 70°F
Hf = Btu/min
2. For emission streams that are not dilute and require additional combustion
air:
Use Figure 4-8 to obtain a conservative estimate:
Hf = (Hf/Qe)figure Qe
Hf = Btu/min
4.3.4.2. Flow Rate of Combined Gas Stream Entering the Catalyst Bed
1. For dilute emission streams that require no additional combustion air, use
Equations 4.3-4 and -5:
Qcom = Qe + Qf + Qc
Qf = Hf/hf
733
-------
Qf = scf m
Qcom = scf m
2. For emission streams that require additional combustion air, use the
following equation to calculate Qc:
Qc = [(0.01 he Qe + 9.4 Qf) (1 + 0.01 ex) - 0.0476 02 Qe]
Qc = scf m
Then use Equation 4.3-4 to calculate QCOm:
Qcom = scf m
4.3.4.3 Flow Rate of Flue Gas Leaving the Catalyst Bed
a. Use the result from the previous calculation:
Qfg = Qcom
Qfg = scf m
If Qfg is less than 500 scfm, define Qfg as 500 scfm.
b. Use Equation 4.3-6 to calculate Qfg;a:
Qfg,a = Qfg KTCO + 460)7530]
Qfg,a = acfm
4.3.5 Catalyst Bed Requirement
Use Equation 4.3-7:
Vbed = 60 Qcom/SV
Vbed = ft3
4.3.6 Heat Exchanger Size (for Systems with Recuperative Heat Exchange Only)
1. For dilute emission streams that do not require additional combustion air:
a. Use Figure 4-9 (line 1):
A = (A/Qe)figure Qe
A = ft2
b. Use Equation 4.3-8:
734
-------
A = [60 Qe (1 + 0.002 Me) Cpair (The-Te)/UATLM]
The values for the parameters in this equation can be determined as fol-
lows:
r' The' Me, he As specified for Equations 4.3-1 and -3.
Te Input data.
U Use a value of 4 Btu/hr-ft2-°F unless the inquirer/
applicant has provided a value.
Tco As calculated in Step 1 of 4.3.4.1 :
T - _ °P
1 CO - r
ATLM Calculate ATLM using the following expression:
ATLM = Tco - The
ATLM = °F
Heat exchanger surface area:
A = ft2
2. For emission streams that are not dilute and require additional combustion
air:
Use Figure 4-9 (line 2):
A = (A/Qe)figure Qe
A = ft2
4.3.7 Evaluation of Permit Application
Compare the calculated values and the values supplied by the applicant using Table 4-4.
If the calculated values for Hf, Qc, QCOm' Vbed, ancl A differ from the applicant's values, the
differences may be due to the assumptions involved in the calculations. Discuss the
details of the design and operation of the system with the applicant.
If the calculated and reported values are not different, then the design and operation of
the system can be considered appropriate based on the assumptions employed in this
handbook.
735
-------
Table 4-4. Comparison of Calculated Values and Values Supplied by the Permit
Applicant for Catalytic Incineration
Calculated Reported
Value Value
Supplementary heat requirement, Hf
Supplementary fuel flow rate, Qf
Combustion air flow rate, QfC
Flue gas stream flow rate, CCOm
Catalyst bed volume, Vbed
Heat exchanger surface area,
(if recuperative heat recovery
is used), A
136
-------
Appendix C.5
Calculation Sheet for Flares
4.4.1 Data Required
HAP emission stream characteristics:
1. Expected emission stream flowrate, Qe = scfm
2. Emission stream temperature, Te = °F
3. Heat content, he = Btu/scf
4. Mean molecular weight of emission stream MWe = Ib/lb-mole
Flare tip diameter, Dtip = in
Required destruction efficiency, DE = %
In the case of a permit review, the following data should be supplied by the applicant:
Flare system design parameters at standard conditions (70°F, 1 atm):
1. Flare tip diameter, Dtjp = in
2. Expected emission stream flowrate, Qe = scfm
3. Emission stream heat content, he = Btu/scf
4. Temperature of emission stream, Te = °F
5. Mean molecular weight of emission stream, MWe = Ib/lb-mole
6. Steam flowrate, Qs = Ib/min
7. Flare gas exit velocity, Ufig = ft/sec
8. Supplementary fuel flow rate,3 Qf = scfm
9. Supplementary fuel heat content,3 hf = Btu/scf
aThis information is needed if the emission stream heat content is less than 300 Btu/scf.
737
-------
10. Temperature of flare gas,bTfig = °F
11. Flare gas flowrate,b Qfig = scfm
12. Flare gas heat content,6 hfig = Btu/scf
4.4.2 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
In a permit review case, if he is below 300 Btu/scf and no supplementary fuel is used,
then the application is rejected. The reviewer may then wish to proceed with the
calculations below. If he is equal to or above 300 Btu/scf, then the reviewer should skip to
Section 4.4.2.3.
4.4.2.1 Supplementary Fuel Requirements
For emission streams with heat contents less than 300 Btu/scf, additional fuel is required.
Use Equation 4.4-1 to calculate natural gas requirements:
Qf = [(300 - he) Qel/582
Qf = scfm
4.4.2.2 Flare Gas Flow Rate and Heat Content
a. Use Equation 4.4-2 to calculate the flare gas flow rate:
Qf|g = Q6 + Qf
Qfig = scfm
b. Determine the flare gas heat content as follows:
hfig = 300 Btu/scf if Qf > 0
hfig = he if Qf = 0
hfig = Btu/scf
4.4.2.3 Flare Gas Exit Velocity
a. Use Table 4-5 to calculate Umax:
blf no auxiliary fuel is added, the value for this variable will be the same as that for the
emission stream.
c For unassisted flares, the lower limit is 200 Btu/scf.
738
-------
If 300 s hfig < 1,000, use the following equation:
Umax = 3.28 [io 1,000 Btu/scf, Umax = 400 ft/sec
b. Use Equation 4.4-3 to calculate Ufig:
Uf,g = 3.06 Qfig,a/(Dtip)2
where Qfig,a is given by Equation 4.4-4:
Qflg,a = [Qflg (Tf|g + 460)]/530
See Appendix B.7, reference 8, for calculating Tfig.
Qfig,a = acfm
Ufig = ft/sec
c. Compare Ufig and Umax:
If Ufig < Umax/ the desired destruction efficiency level of 98 percent
can be achieved. (Note: Ufig should exceed 0.03 ft/sec for flame
stability.) If Ufig > Umax, 98 percent destruction efficiency cannot be
achieved. When evaluating a permit, reject the application in such a
case.
4.4.2.4 Steam Requirements
a. Assume that the amount of steam required is 0.4 Ib steam/lb flare
gas. Use Equation 4.4-5 to calculate Qs:
QS = 1.03XlO"3XQf|gXMWf|g
Qs = Ib/min
4.4.3 Evaluation of Permit Application
Compare the calculated and reported values using Table 4-6. If the calculated values of
Qf, Ufig, Qfig, and Qs are different from the reported values for these variables, the
differences may be due to the assumptions (e.g., heating value of fuel, ratio of steam to
flare gas, etc.) involved in the calculations. Discuss the details of the design and oper-
ation of the system with the applicant. If the calculated and reported values are not
different, then the operation of the system can be considered appropriate based on the
assumptions employed in the handbook.
735
-------
Table 4-6. Comparison of Calculated Values and Values Supplied by the Permit
Applicant for Flares
Calculated Reported
Value Value
Supplementary fuel flow rate, Qf ... ...
Flare gas exit velocity, Uf|Q
Flare gas flow rate, Qfig
Steam flow rate, Qs ... ...
140
-------
Appendix C.6
Calculation Sheet for Carbon Adsorption
4.6.1 Data Required
HAP Emission stream characteristics:
1. Maximum flow rate, Qe = scfm
2. Temperature, Te = °F
3. Relative humidity, Rhum = %
4. HAP =
5. Maximum HAP content, HAPe = ppmv
Required removal efficiency, RE = %
In the case of a permit review, the following data should be supplied by the applicant:
Carbon adsorber (fixed-bed) system variables at standard conditions (70°F, 1 atm):
1. Reported removal efficiency, REreported = %
2. HAP content, HAPe = ppmv
3. Emission stream flow rate, Qe = scfm
4. Adsorption capacity of carbon bed,
AC = Ib HAP/100 Ib carbon
5. Number of beds =
6. Amount of carbon required, Creq = Ib
7. Cycle time for adsorption, 6ad = hr
8. Cycle time for regeneration, 0reg = hr
9. Emission stream velocity through the bed, Ue = ft/min
10. Bed depth, Zbed = ft
11. Bed diameter, Dbed = ft
12. Steam ratio, St = Ib steam/lb carbon
747
-------
4.6.2 Pretreatment of the Emission Stream
4.6.2.1 Cooling
If the temperature of the emission stream is significantly higher than 100°F, a heat
exchanger is needed to cool it to 100°F. Refer to Appendix B.5, reference 8, for the
calculation procedure.
4.6.2.2 Dehumidification
If the relative humidity level is above 50 percent, a condenser is required to cool and
condense the water vapor in the emission stream. Refer to Section 4.8 for more details.
4.6.2.3 High VOC Concentrations
HAPe = _ ppmv
If flammable vapors are present in the emission stream, VOC content will be limited to
below 25 percent of the LEL.
LEL = _ ppmv (from Table B.1-1)
25% of LEL = 0.25 x LEL (ppmv) = _ ppmv
The maximum practical inlet concentration for carbon beds is about 10,000 ppmv. If
HAPe is greater than 10,000 ppmv, carbon adsorption may not be applicable.
4.6.3 Carbon Adsorption System Design Variables
a. Use Equation 4.6-1 to calculate the required outlet HAP concentra-
tion:
HAP0 = HAPe (1 - 0.01 RE)
HAP0 = _ ppmv
b. Specify the appropriate values of 0ad, 0reg, and St from Table 4-7.
0ad = - hr
0reg = - hr
St = _ Ib steam/lb carbon
742
-------
4.6.4 Determination of Carbon Adsorber System Variables
4.6.4.1 Carbon Requirements
a. Use Equation 4.6-2:
Creq = 2 x 1.55 1(T5 N6ad Qe (HAPe - HAP0) MWHAP/AC
Assume N = 2
Obtain MWHAP from Table B.1-2 or reference 18. If no data are
available, use a conservative value of 5 Ib HAP/100 Ib carbon.
Creq = - Ib
b. Use Figure 4-13 to obtain (Creq/Qe):
Weq = ^req'^e'figure ^e
Creq = - Ib
4.6.4.2 Carbon Adsorber Size
a. Use Equation 4.6-3 to calculate Abed:
Abed = Qe,a/Ue
Calculate Qe/a using Equation 4.6-4:
Qe,a = Qe KTe + 460)/530]
Qe a = _ acfm
Assume Ue = 100 ft/sec
Abed = - ft2
b. Use Equation 4.6-5 to calculate Dbed:
Dbed = 1.13 (Abed)a5
c. Use Equation 4.6-6 to calculate volume of carbon per bed:
Vcarbon = (Creq/N)/Pbed
Assume pbed = 30 Ib/ft3
Vcarbon ~ - ft
743
-------
d. Use Equation 4.6-7 to calculate Zbed:
^•bed ~ vcarbon/Abed
Note: If Qe is greater than about 20,000 scfm, three or more carbon beds
may need to be used.
4.6.4.3 Steam Required for Regeneration
a. Use Equation 4.6-8 to calculate steam requirements:
Qs - [St X Creq/(ereg - 6dry-cool)]/60
Assume 0dry-cooi = °-25 nrs-
Qs = Ib/min
b. Use Figure 4-14:
Qs = Ib/min
Calculate Qs/Abed:
Qs/Abed = Ib steam/min-ft2
If Qs/Abed is greater than 4 Ib steam/min-ft2, fluidization of the carbon
bed may occur.
4.6.4.4 Condenser
a. Use Equation 4.6-10 to calculate H|oad:
Hload = 1.1 X 60 X Qs [X + Cpw(Tsti - Tgto)]
Obtain X and Cpw from reference 19 based on the values assumed
forTstiandTsto.
Hload = Btu/hr
b. Use Equation 4.6-9 to calculate Acon:
ACon = H|oad/UATLM
Assume U = 150 Btu/hr-ft2-°F if no other data are available.
AT _ I ' ' Sti " ' WO' ' ' StO " ' Wi' |
lnl(Tsti-Two)-
144
-------
where Twi = 80°FandTwo = 130°F.
ATLM = - °F
A - ft2
AACOn — - I I
c. Use Equations 4.6-11 and -12 to calculate Qw:
QCOO!,W ~ H|oacj/[Cpw(Two - Twi)]
QCOOI.W =
,,w
Qw = 0.002 Qcoo,,
Qw = _ gal/min
4.6.4.5 Recovered Product
a. Use Equation 4.6-13 to calculate Qrec:
Qrec = 1.55x10~9xQexHAPexRExMWHAp
Q = _ Ib/hr
rec
4.6.5 Evaluation of Permit Application
Compare the results from the calculations and the reported values using Table 4-8.
If the calculated values of Creq, Dbed, Zbed, Qs, Acon, Qw, and Qrec, are different from the
reported values, the differences may be due to the assumptions involved in the calcula-
tions. Discuss the details of the design and operation of the system with the 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 handbook.
745
-------
Table 4-8. Comparison of Calculated Values and Values Supplied by the Permit
Applicant for Carbon Adsorption
Calculated Reported
Value Value
Carbon requirement, Creq
Bed diameter, Dbed
Bed depth, Zbed
Steam rate, Qs
Condenser surface area, Acon
Cooling water rate, Qw
Recovered product, Qrec
146
-------
Appendix C. 7
Calculation Sheet for Absorption
4.7.1 Data Required
HAP emission stream characteristics:
1. Maximum flow rate, Qe = scfm
2. Temperature, Te = °F
3. HAP =
4. HAP concentration, HAPe = ppmv
5. Pressure, Pe = mm Hg
Required removal efficiency, RE = %
In the case of a permit review, the following data should be supplied by the applicant:
Absorption system variables at standard conditions (70°F, 1 atm):
1. Reported removal efficiency, REreported = %
2. Emission stream flow rate, Qe = scfm
3. Temperature of emission stream, Te = °F
4. HAP -
5. HAP concentration, HAPe = ppmv
6. Solvent used =
7. Slope of the equilibrium curve, m =
8. Solvent flow rate, Lga| = gal/min
9. Density of the emission stream, pG = Ib/ft3
10. Schmidt No. for the HAP/emission stream and HAP/solvent systems:
ScG =
ScL =
(Refer to Appendix B.9, reference 8, or reference 12 for definition and
calculation of ScG and ScL)
147
-------
11. Properties of the solvent:
Density, pL = Ib/ft3
Viscosity, (XL = centipoise
12. Type of packing used =
13. Packing constants:
a = b = c = d =
e = Y = s = g =
r =
14. Column diameter, DCO|Umn = ft
15. Tower height, (packed) HtCO|Umn = ft
16. Pressure drop, APtotai = in H20
4.7.3 Determination of Absorber System Design and Operating Variables
4.7.3.1 Solvent Flow Rate
a. Assume a value of 1.6 for AF.
Determine "m" from the equilibrium data for the HAP/solvent sys-
tem under consideration (see references 18, 22, and 23 for equilibri-
um data).
m =
Use Equation 4.7-3:
Qe = scf m
Gmol = 0.155Qe
Gm0| = Ib-moles/hr
b. Use Equation 4.7-2:
l_moi = 1.6m Gmo,
Lm0| = Ib-moles/hr
748
-------
c. Use Equation 4.7-4:
Lga, = 0.036 Lmo,
Lgai = gal/min
4.7.3.2 Column Diameter
a. Use Figure 4-17:
Calculate the abscissa (ABS):
MWSO|Vent = Ib/lb-mole
L - Lmo, x MWSO|vent
L = Ib/hr
MWe = Ib/lb-mole
G = Gmol x MWe
G = Ib/hr
PG = Ib/ft3 (refer to Appendix B.9, reference 8, for calcu-
lating this variable)
pL = Ib/ft3 (from reference 18)
ABS = (L/G)(PG/PL)°-5
ABS =
b. From Figure 4-17, determine the value of the ordinate (ORD) at
flooding conditions.
ORD =
c. For the type of packing used, determine the packing constants from
reference 21:
a =
e =
Determine |iL (from reference 18):
M-L = cp
149
-------
d. Use Equation 4.7-8 to calculate Garea/f:
Garea,f = ([ORD pGpL gc]/[(a/e3)W)0-2]}0-5
Garea/f = Ib/sec-ft2
e. Assume a value for the fraction of flooding velocity for the proposed
design:
f =
Use Equation 4.7-9 to calculate Garea:
~ '
G = _ Ib/hr-ft2
f. Use Equation 4.7-10 to calculate the column cross-sectional area:
Acolumn = G/(3,600 Garea)
A , ft2
"column - ' l
g. Use Equation 4.7-11 to calculate the column diameter:
'-'column = ' • '3 \Aco|umn)
'-'column ~~ - '*
4.7.3.3 Column Height
a. Use Equation 4.7-13 or Figure 4-18 to calculate NOG:
Using Equation 4.7-13:
HAPe = _ ppmv
HAP0 = HAPe (1 - 0.01 RE)
HAP0 = _ ppmv
NOG = ln{(HAPe/HAP0)[1 - (1/AF)] + (1/AF)}/[1 - (1/AF)]
NOG = -
Using Figure 4-18:
HAPP/HAPn = _
Ql I I/-VI Q
150
-------
At HAPe/HAP0 and 1/AF = 1/1.6 = 0.63, determine NOG:
NOG =
b. Use Equations 4.7-14,-15, and -16 to calculate HG, HL, and HOG. Determine the
packing constants in Equation 4.7-15 using Tables B.9-2 and -3, reference 8,
or reference 12:
b = c - d =
Y = s -
Determine ScG and ScL using Tables B.9-4 and -5, reference 8, or reference
12:
ScG =
ScL =
*- ~~ L/ACO|urnn
L" = Ib/hr-ft2
(JLL" = Ib/hr-ft (from reference 18)
Calculate HG and HL:
HG = [b (3,600 Garea)c/(L")d] (ScG)°-5
HG = ft
HL = Y(L"/jiL")8 (ScL)0'5
HL = ft
Calculate HOG using AF - 1.6:
HOG = HG + (1/AF) HL
HOG = ft
c. Use Equation 4.7-12 to calculate HtCO|umn:
Htcolumn = NOG HQG
H*column = '*
757
-------
d. Use Equation 4.7-18 to calculate Httota|-.
= HtCO|umn + 2 + (0.25 DCO|umn)
e. Use Equation 4.7-19 to calculate WtCO|umn:
WtCO|umn = (48 DCO|Umn x Httota|) + 39(DCO|umn)
WtCO|umn - - Ib
f. Use Equation 4.7-20 to calculate Vpacking:
"packing ~ 0./ob(IJco|urnn) X Mtco|umn
\/ — ft3
•packing - ' l
4.7.3.4 Pressure Drop Through the Column
a. Use Equation 4.7-21 to calculate APa:
Determine the constants using Table B.9-6, reference 8, or reference 12:
9 = -
r — r
APa = g x TO'8 [10(rL"/pL)] (3,600 Garea)2/PG
APa = _ Ib/ft2-ft
b. Use Equation 4.7-22 to calculate APtota|:
APtotal = AP X Htco|umn
APtotal = - Ib/ft2
APtotal (1/5.2) = - in H20
4.7.4 Evaluation of Permit Application
Compare the results from the calculations and the values supplied by the permit appli-
cant using Table 4-9. If the calculated values are different from the reported values, the
differences may be due to the assumptions involved in the calculations. Therefore,
discuss the details of the proposed design with the applicant.
If the calculated values agree with the reported values, then the design of the proposed
absorber system may be considered appropriate based on the assumptions made in this
handbook.
752
-------
Table 4-9. Comparison of Calculated Values and Values Supplied by the Permit
Applicant for Absorption
Calculated Reported
Value Value
Solvent flow rate, Lgai ... ...
Column diameter, DCO|Umn ... ...
Column height, Htco|umn
Total column height, Httota|
Packing volume, Vpacking
Pressure drop, APtotai
Column weight, WtCO|Umn
753
-------
Appendix C.8
Calculation Sheet for Condensation
4.8.1 Data Required
HAP emission stream characteristics:
1. Maximum flow rate, Qe = scfm
2. Temperature, Te = °F
3. HAP =
4. HAP concentration, HAPe = ppmv
5. Moisture content, Me = %
6. Pressure, Pe = mm Hg
Required removal efficiency, RE = %
In the case of a permit review for a condenser, the following data should be supplied by
the applicant:
Condenser system variables at standard conditions (70°F, 1 atm):
1. Reported removal efficiency, REreported = %
2. Emission stream flow rate, Qe = scfm
3. Temperature of emission stream, Te = °F
4. HAP =
5. HAP concentration, HAPe = ppmv
6. Moisture content, Me = %
7. Temperature of condensation, Tcon = °F
8. Coolant used =
9. Temperature of inlet coolant, T(
coolj
10. Coolant flow rate, QCOO|ant = 'D/hr
11. Refrigeration capacity, Ref = tons
12. Condenser surface area, Acon - ft2
754
-------
4.8.2 Pretreatment of the Emission Stream
Check to see if moisture content of the emission stream is high. If it is high, dehumidifica-
tion is necessary. This can be carried out in a heat exchanger prior to the condenser.
4.8.3 Condenser System Design Variables
The key design variable is the condensation temperature. Coolant selection will be based
on this temperature.
In evaluating a permit application, use Table 4-10 to determine if the applicant's values
for Tcon, coolant type, and Tcoo(ri are appropriate:
T = °F
1 con '
Coolant type =
' cool,i ~
If they are appropriate, proceed with the calculations. Otherwise, reject the proposed
design. The reviewer may then wish to follow the calculation procedure outlined below.
4.8.4 Determination of Condenser System Design Variables
4.8.4.1 Estimation of Condensation Temperature
a. Use Equation 4.8-1 to calculate Ppartiai:
Ppartiai = 760{(1- 0.01 RE)/[1 - (RE x 1Q-8 HAPe)]}HAPex 10'6
Ppartiai = mm Hg
b. Use Figure 4-20 to determine Tcon:
T = °F
i con i
4.8.4.2 Selection of Coolant
Use Table 4-10 to specify the coolant (also see references 18 and 27):
Coolant =
4.8.4.3 Condenser Heat Load
1. a. Use Equation 4.8-2 to calculate HAPe,m:
HAPe,m = (Qe/387) HAPex 10'6
HAPe/m = Ib-moles/min
b. Use Equation 4.8-3 to calculate HAP0,m:
HAP0,m = (Qe/387)[1-(HAPex10-6)][Pvapor/(Pe-Pvapor)]
755
-------
where Pvapor = Ppartia|
HAP0 m = _ Ib-moles/min
c. Use Equation 4.8-4 to calculate HAPcon:
HAPcon = HAPe,m - HAP0,m
HAPcon = _ Ib-moles/min
2. a. Calculate heat of vaporization (AH) of the HAP from the slope of the graph
IWPvapor)] vs [1/(Tcon + 460)] for the Pvapor and Tcon ranges of interest.
See Appendix B.10, reference 8, for details.
AH = _ Btu/lb-mole
b. Use Equation 4.8-5 to calculate Hcon:
Hcon = HAPcon[AH + CPHAP (Te - Tcon)]
where CpHAP can be obtained from references 18 and 27.
Hcon = _ Btu/min
c. Use Equation 4.8-6 to calculate Huncon:
Huncon - HAP0(m CpHAP(Te - Tcon)
Huncon = - Btu/min
d. Use Equation 4.8-7 to calculate Hnoncon:
Hnoncon = [(Qe/387)-HAPe,m] Cpair (Te - Tcon)
where Cpair can be obtained from Table B.4-1, reference 8.
Hnoncon = - Btu/min
3. a. Use Equation 4.8-8 to calculate H|0ad:
Hioad = 1-1 x 60 (Hcon + Huncon + Hnoncon)
H,oad = - Btu/hr
4.8.4.4 Condenser Size
Use Equation 4.8-9 to calculate Acon:
756
-------
ACOn - H|oad/U ATLM
where ATLM is calculated as follows:
AT~LM = t(Te - TCOO|/0) - (Tcon - Tcoo(;j)]/ln[(Te - TCOO|;0)/(Tcon - TCOO| f)]
Assume: TcooU = Tcon-15, and Tcoo,,0 - TcooU = 25°F
T _ or
1 cool,i ~ - r
T , = °F
1 cool,o - r
ATLM = - °F
Assume: U = 20 Btu/hr-ft2 -°F (if no other estimate is available).
A = _ ft2
"con - ' L
4.8.4.5 Coolant Flow Rate
Use Equation 4.8-10 to calculate QCOOiant:
'-'coolant ~ ''load' l^Pcoolant '' cool, o " 'cool,i'J
The value for CpCOO|ant for different coolants can be obtained from references 18 or 27. If
water is used as the coolant, Cpwater can be taken as 1 Btu/lb-°F.
Cpcoo,ant = - Btu/lb-°F
Qcooiant = - Ib/hr
4.8.4.6 Refrigeration Capacity
Use Equation 4.8-11 to calculate Ref:
Ref = H,oad/1 2,000
Ref = _ tons
4.8.4.7 Recovered Product
Use Equation 4.8-12 to calculate Qrec:
Q = 60 x HAP x MW
4.8.5 Evaluation of Permit Application
Compare the results from the calculations and the values supplied by the permit appli-
cant using Table 4-11. If the calculated values Tcon, coolant type, Qcooiant' Acon, Ref, and
757
-------
Qrec are different from the reported values for these variables, the differences may be
due to the assumptions involved in the calculations. 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 assump-
tions made in this handbook.
Table 4-11. Comparison of Calculated Values and Values Supplied by the Permit
Applicant for Condensation
Calculated Reported
Value Value
Condensation temperature, Tcon ... ...
Coolant type ... ...
Coolant flow rate, QCOOiant
Condenser surface area, Acon ... ...
Refrigeration capacity, Ref
Recovered product, Qrec
158
-------
Appendix C.9
Calculation Sheet for Fabric Filters
4.9.1 Data Required
HAP emission stream characteristics:
1. Flow rate, Qea = acfm
2. Moisture content, Me = % (vol)
3. Temperature, Te = °F
4. Particle mean diameter, Dp = |xm
5. SO3 content = ppm (vol)
6. Particulate content = grains/scf
7. HAP content = % (mass)
In the case of a permit review, the following data should be supplied by the applicant:
1. Filter fabric material
2. Cleaning method (mechanical shaking, reverse air, pulse-jet)
3. Air-to-cloth ratio ft/min
4. Baghouse construction configuration (open pressure, closed pressure,
closed suction)
4.9.2 Pretreatment Considerations
If emission stream temperature is not from 50° to 100°F above the dew point, pretreat-
ment is necessary (see Section 3.3.1 and Appendix B.1). Pretreatment will cause two of
the pertinent emission stream characteristics to change; list the new values below.
1. Maximum flow rate at actual conditions, Qe,a = acfm
2. Temperature, Te = °F
4.9.3 Fabric Filter System Design Variables
1. Fabric Type(s) (use Table 4-12):
a.
b.
c.
759
-------
2. Cleaning Method(s) (Section 4.8.3.2):
a. _
b. _
3. Air-to-cloth ratio, point or range (Table 4-14) _ ft/min
4. Net cloth area, Anc:
Anc = Qe,a / (A/C ratio)
where:
Anc = net cloth area, ft2
Qe/a = maximum flow rate at actual conditions, acfm
A/C ratio = air-to-cloth ratio, ft/min
A = - ' -
DC
Anc = _ ft2
5. Gross cloth area, Atc:
Ate = Ancx Factor
where:
Atc = gross cloth area, ft2
Factor = value from Table 4-15, dimensionless
Atc = _ ft2
6. Baghouse configuration
4.9.4 Evaluation of Permit Application
Using Table 4-16, compare the results from this section and the data supplied by the
permit applicant. 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 4.7). Therefore, in evaluating the reasonableness of any system specifications
760
-------
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 paniculate 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 4-12)?
3. Is the baghouse cleaning method compatible with the selected fabric materi-
al and its construction; that is, material type and woven or felted construc-
tion (see Section 4.9.3.2 and Table 4-13)?
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 4-14)?
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:
A/C ratio = -^La-
^nc
where:
A/C ratio = air-to-cloth ratio, ft/min
Qe,a = 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?
767
-------
Appendix C. 10
Calculation Sheet for Electrostatic Precipitators
4.10.1 Data Required
HAP emission stream characteristics:
1. Flow rate, Qe,a = acfm
2. Emission stream temperature, Te =
3. Particulate content = grams/scf
4. Moisture content, Me = % (vol)
5. HAP content = % (mass)
6. Drift velocity of particles, Ud = ft/s
7. Collection efficiency, CE = % mass
In case of a permit review, the following data should be supplied by the applicant. The
design considerations in this section will then be used to check the applicant's design.
1. Reported collection efficiency = %
2. Reported drift velocity of particles = ft/sec
3. Reported collection plate area = ft2
4.10.2 Pretreatment of Emission Stream
If the emission stream temperature is not from 50° to 100°F above the dew point,
pretreatment is necessary (see Section 3.3.1 and Appendix B.1). Pretreatment will cause
two of the pertinent emission stream characteristics to change; list the new values
below.
1. Maximum flow rate at actual conditions, Qe/a = acfm
2. Temperature, Te = °F
4.10.3 ESP Design Variables
Collection plate area is a function of the emission stream flow rate, drift velocity of the
particles (Table 4-17), and desired control efficiency. The variables are related by the
Deutsch-Anderson equation:
AP = 60%TJ7 x ln |1 - CE>
762
-------
where:
Ap = collection plate area, ft2
Qe>a = emission stream flow rate at actual conditions as it enters the
control device, acfm
Ud = drift velocity of particles, ft/sec
CE = required collection efficiency, decimal fraction
60 x
Ap = _ ft2
4,10.4 Evaluation of Permit Application
Using Table 4-18, compare the results from this section and the data supplied by the
permit applicant. In evaluating the reasonableness of ESP design specifications in a
permit application, the main task will be to examine each parameter in terms of its
capability with the gas stream conditions.
If the applicant's collection plate area is less than the calculated area, the discrepancy
will most likely be the selected drift velocity. Further discussions with the permit appli-
cant are recommended to evaluate the design assumptions and to reconcile any appar-
ent discrepancies.
763
-------
Appendix C.11
Calculation Sheet for Venturi Scrubbers
4.11.1 Data Required
HAP emission stream characteristics:
1. Flow rate Qe/a = acfm
2. Temperature, Te = °F
3. Moisture content, Me = %
4. Required collection efficiency, CE = %
5. Particle mean diameter, Dp = pm
6. Particulate content = grains/scf
7. HAP content = % (mass)
In the case of a permit review, the following data should be supplied by the applicant:
1. Reported pressure drop across venturi = in H20
2. An applicable performance curve for the venturi scrubber
3. Reported collection efficiency = %
4.11.2 Pretreatment of Emission Stream
If the emission stream temperature is not from 50° to 100°F above the dew point,
pretreatment is necessary (see Section 3.3.1 and Appendix B.1). Pretreatment will cause
two of the pertinent emission stream characteristics to change; list the new values
below:
1. Maximum flow rate at actual conditions, Qe a = acfm
2. Temperature, Te = °F
4.11.3 Venturi Scrubber Design Variables
4.11.3.1 Venturi Scrubber Pressure Drop
The pressure drop across the venturi (APV) can be estimated through the use of a venturi
scrubber performance curve (Figure 4-22) and known values for the required collection
efficiency (CE) and the particle mean diameter (Dp).
APV - in H2O
If the estimated APV is greater than 80 in H20, assume that the venturi scrubber cannot
achieve the desired control efficiency.
164
-------
4.11.3.2 Materials of Construction
Select the proper material of construction by contacting a vendor, or as a lesser alterna-
tive, by using Table 4-20.
Material of construction
4.11.4 Sizing of Venturi Scrubbers
Some performance curves and cost curves are based on the saturated gas flow rate
(Qe s). If Qe s is needed, it can be calculated as follows:
Qe,s = Qe,a x (Te,s + 460)/(Te + 460)
where:
Qes = saturated emission stream flow rate, acfm
Te s = temperature of the saturated emission stream, °F
Use Figure 4-22 to determine Tes; the moisture content of the emission stream (Me) must
be in units of Ibs H20/lbs dry air.
Convert Me (% vol) to units of Ibs H20/lbs dry air, decimal fraction:
(Me/100) x (18/29) = Ib H20/lb dry air
From Figure 4-22:
Te,s = °F
Qe/s = ( ) x ( + 460) / ( + 460)
Qe s = acfm
4.11.5 Evaluation of Permit Application
Using Table 4-21, compare the results of this section and the data supplied by the permit
applicant. Compare the estimated APv and the reported pressure drop across the ven-
turi, 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 discrepancy 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 handbook.
765
-------
Table 4-21. Comparison of Calculated Values and Values Supplied by the Permit
Applicant for Venturi Scrubbers
Calculated Reported
Value Value
Particle Mean Diameter, Dp ... ...
Collection efficiency, CE ... ...
Pressure drop across venturi, APV ... ...
166
-------
Appendix C. 12
Capital and Annualized Cost Calculation Worksheet
167
-------
Table C.12-1. Preliminary Calculations for Capital Cost Algorithm
(1) Calculation of Duct Diameter, Dduct (in)
DH,,,t =12 — x
4 Qe.
a
2
where:
Qe,a = emission stream flow rate at actual conditions, acfm
= velocity of gas stream in duct, ft/min
= 12
If velocity of gas stream in duct is unknown, use 2,000 ft/min; the equation then
becomes:
Dduct = 0.3028 (Qe.a)*
Dduct = 0.3028 ( _ )* = _ in
(2) Calculation of Stack Diameter, Dstack (in)
= 12 (*-x Qfg'a V
X
where:
Qfg,a - actual flue gas flow rate, acfm
Ustack = velocity of gas in stack, ft/min
4
D- 12
The gas stream velocity in the stack should be at least 4,000 ft/min. If velocity is
unknown, use 4,000 ft/min; the equation then becomes:
Dstack = 0.2141
Dstack = 0.2141 ( - )* - - in
(3) Calculation of Total System Pressure Drop, APt (in H20)
APt = APduct + APstack + APdevice#1 + APdevice#2 + APdevice#3
(Note: See Table 5-7 for AP values.)
APt = _ + _ + _ + _ +
_ in H20
768
-------
Table C.12-2. Estimate of Capital Costs in Current Dollars
Cost Elements
Figure or
Table Cost
Escalation Factor
(Current FE/Base FE)
see Table 5-2
Current Cost
1. Major Equipment Purchase Cost
Thermal Incinerator
Heat Exchanger
Catalytic Incinerator
Catalyst0, Vcat =
Carbon Adsorbed
Carbond, Creq =
Absorber6
Packing6, Vpack =
Condenser'
Refrigerant'
Fabric Filter9
Bags9, Atc =
ESPh
Venturi Scrubber'
Design Factors'
ratnr3 $
rb $
ratnrc $
ft3*
fird $
Ibx
$
H I aHHfirs6 $
k - ft3 Y
$
$
$
ft2*
$
fir' $
tors'
x ( / )
X ( / )
X ( / )
$/ft3 X ( / )
X ( / )
$/lb x ( / )
x ( / )
x ( / )
$/ft3 x ( / )
x ( / )
x ( / )
x ( / )
$/ft2 X ( / )
X ( / )
X ( / )
X
(Thickness Factor)
= $.
$-
$-
(Composition Factor)
$-
$-
$-
$-
$-
$-
$-
SUBTOTAL
2. Auxiliary Equipment Purchase Cost
Ductwork1 $.
Fank
Motor1
Stack"1
(Fan
x
(Length)
Current Cost)
x ( /
X ( /
. x 0.15
x ( /
) = $
) = $
= $
) = $
SUBTOTAL
3. Pre-Total Purchase Equipment Cost
Adjustments"
4. TOTAL Purchase Equipment Cost
Item 1 Subtotal + Item 2 Subtotal
(Item 3) x -0.091
Item 3 + Adjustments
5. Instrumentation and Controls
6. Freight and Taxes
7. TOTAL Purchased Cost
10% of Item 4
8% of Item 4
Item 4 + Item 5 + Item 6
8. TOTAL CAPITAL COSTS
F°x (Item?); where F =
169
-------
Footnotes to Table C. 12-2
a Thermal Incinerator: Figure 5-1, includes fan plus instrumenta-
tion and control costs for thermal incinerators, in addition to
the major equipment purchased cost. Additional auxiliary
equipment (ductwork and stack) purchased costs and costs of
freight and taxes must be added to obtain the total purchased
cost.
bHeat Exchangers: If the HAP control system requires a heat
exchanger, obtain the cost from Figure 5-2, escalate this cost
using the appropriate factor, and add to the major equipment
purchased cost.
0 Catalytic Incinerator: Figure 5-3 provides the cost of a catalytic
incinerator, less catalyst costs. The "Table" catalyst cost is
estimated by multiplying the volume of catalyst required (Vcat)
by the catalyst cost factor ($/ft3) found on Table 5-3. Catalyst
costs, all auxiliary equipment (ductwork, fan, and stack) pur-
chased costs and the cost of instrumentation and controls, and
freight and taxes must be added to obtain the total purchased
cost.
dCarbon adsorber: Figure 5-4 (packaged carbon adsorber sys-
tems) includes the cost of carbon, beds, fan and motor, instru-
mentation and controls, and a steam regenerator. Additional
auxiliary equipment (ductwork and stack) purchased costs and
costs of freight and taxes must be added to obtain the total
purchased cost. Figure 5-5 (custom carbon adsorber systems)
includes beds, instrumentation and controls, and a steam re-
generator, less carbon. The "Table" carbon cost for custom
carbon adsorbers is estimated by multiplying the weight of
carbon required (Creq) by the carbon cost factor ($/lb) found on
Table 5-3. Costs of carbon, all auxiliary equipment (duct, fan,
stack) purchased costs, and freight and taxes must be added to
obtain the total purchased cost.
e Absorber: Figure 5-6 does not include the cost of packing,
platforms, and ladders. The cost of platforms and ladders (Fig-
ure 5-7) and packing must be added to obtain the major pur-
chased equipment cost. The "Table" packing cost is estimated
by multiplying the volume of packing required (Vpack) by the
appropriate packing cost factor found on Table 5-4. All auxil-
iary equipment (ductwork, fan, and stack) purchased costs,
and costs of freight and taxes must be added to obtain the total
purchased cost.
1 Condenser Systems: Figure 5-8 yields total capital costs for
cold water condenser systems. For systems needing refriger-
ant, the applicable cost from Figure 5-9 must be added to
obtain the total capital costs. In either case, the escalated cost
estimate is then placed on Line 8, "TOTAL CAPITAL COSTS."
9 Fabric Filter Systems: Figure 5-10 gives the cost of a negative
pressure, insulated baghouse. The curve does not include bag
costs. The "Table" bag cost is estimated by multiplying the
gross cloth area required (Atc) by the appropriate bag cost
factor found on Table 5-5. Bag costs, all auxiliary equipment
(duct, fan, and stack) purchased costs, the cost of instrumenta-
tion and controls, and freight and taxes must be added to
obtain the total purchased cost.
h Electrostatic Precipitators: Figure 5-11 provides the cost for an
insulated ESP. 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.
' Venturi Scrubber: Figure 5-12 includes the cost of instrumen-
tation and controls in addition to the major equipment pur-
chased cost. This cost curve is based on a venturi scrubber
constructed from 1/8-inch carbon steel. Figure 5-13 is used to
determine if 1/8-inch steel is appropriate for a given applica-
tion (use the higher curve). If thicker steel is required, Figure
5-14 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 required (see Section
4.11.3.2) 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.
1 Ductwork: Figure 5-15 gives the cost of straight ductwork
made of carbon steel for various thicknesses, based on the
required duct diameter. Figure 5-16 gives the cost of straight
ductwork made of stainless steel for various thicknesses,
based on the required duct diameter. Preliminary calculations
(duct diameter, see Table C.12-1) are necessary to estimate
ductwork costs.
k Fan: Figure 5-17 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 H20).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.
' The cost of a motor is estimated as 15% of the fan cost.
"Stack: Figure 5-18 gives the cost of a carbon steel stack at
various stack heights and diameters. Figure 5-19 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.
n For thermal incinerators, carbon adsorbers, and venturi scrub-
bers, the purchase cost curve includes the cost for instrumen-
tation and controls. This cost (i.e., the "Adjustment") must be
subtracted out to estimate the total purchased equipment cost.
This is done by adding the Item 1 subtotal and the Item 2
subtotal and multiplying the result 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" equals zero.
0 Obtain factor "F" from "TOTAL" line in Table 5-8.
770
-------
Table C.12-3. Preliminary Calculations for Annualized Cost Algorithm
(1) Calculation of Annual Electricity Requirement, AER (Line 5, Table C.12-6)
a. Fan Electricity Requirement, PER
FER = 0.0002 (Qfg/a) x AP x MRS
where:
Qfg,a = actual flue gas flow rate, acfm
AP = total HAP control system pressure drop, in H20 (see Table 5-7)
MRS = annual operating hours, hr
(Note: Use 8,600 unless otherwise specified.)
FER = 0.0002 ( ) x x = kWh
b. Baghouse Electricity Requirement, BER
(Note: Assume 0.0002 kW are required per ft2 of gross cloth area.)
BER = 0.0002 (Atc) x MRS
where:
Ate = gross cloth area required, ft2
BER = 0.0002 ( ) x = kWh
c. ESP Electricity Requirement, EER
(Note: Assume 0.0015 kWare required per ft2 of collection area.)
EER = 0.0015(Ap)xHRS
where:
Ap = collection plate area, ft2
EER = 0.0015 ( ) x = kWh
d. Annual Electricity Requirement, AER
AER = FER + BER + EER
AER = + + = kWh
777
-------
(2) Calculation of Capital Recovery Factor, CRF (Line 18, Table C.12-6)
CRF = [i(l + i)n]/[(l + i)n-1]
where:
i = interest rate on borrowed capital, decimal fraction
(Note: Unless otherwise specified use 10%.)
n = control device lifetime, years (see Table 5-12)
CRF = [ _ x(1+ _ )( )]/[(1+ _ )( )-1]
(3) Calculation of Annual Operator Labor, OL (Line 9, Table C.1 2-6)
OL = (MRS) (operator hours per shift) / (operating hours per shift)
(Note: Obtain operator hr/shift value from Table 5-12)
OL = ( _ ) x ( _ ) / ( _ ) = _ hr
(4) Calculation of Annual Maintenance Labor, ML (Line 11, Table C.12-6)
ML = (MRS) (maintenance hours per shift) / (operating hours per shift)
(Note: Obtain maintenance hr/shift value from Table 5-12)
ML= ( _ )x( _ )/( _ ) = hr
772
-------
Table C.12-4. Additional Utility Requirements
(1) Fuel Requirement for Incinerators (Line 1 or Line 2, Table C.12-6)
(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 resultant product of the equation [gallons of fuel oil
required] is then used on Line 2 of Table C.12-6.)
Fuel Requirement = 60 (Qf) x MRS
where:
Qf = supplementary fuel required, scfm
MRS = annual operating hours, hr
(Note: Use 8,600 hours unless otherwise specified.)
Fuel Requirement = 60 ( ) x = ft3
(2) Steam Requirement for Carbon Adsorber (Line 4, Table C.12-6)
(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
MRS = 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, Table C.12-6)
(Note: Assume 12 gal of cooling water required per 100 Ibs steam.)
Water Requirement = 0.48 (Qrec) x MRS
where:
Qrec = quantity of HAP recovered, Ib/hr
HRS = annual operating hours, hr
(Note: Use 8,600 hours unless otherwise specified.)
Water Requirement = 0.48 ( ) x = gal
773
-------
(4) Absorbent Requirement for Absorbers (Line 3 or 6, Table C.12-6)
(Note: Assume no recycle of absorbing fluid [water or solvent].)
Absorbent Requirement = 60 (Lga|) x MRS
where:
Lgai = absorbing fluid flow rate, gal/min
MRS = annual operating hours, hr
(Note: Use 8,600 hours unless otherwise specified.)
Absorbent Requirement = 60 ( ) x = gal
(5) Water Requirement for Venturi Scrubbers (Line 3, Table C.12-6)
(Note: Assume 0.01 gal of water required per acf of emission stream.)
Water Requirement = 0.6 (Qe/a) x MRS
where:
Qe a = emission stream flow rate into scrubber, acfm
MRS = annual operating hours, hr
(Note: Use 8,600 hours unless otherwise specified.)
Water Requirement = 0.6 ( ) x = gal
774
-------
Table C.I2-5. Estimation of Replacement Parts Annualized 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 be
replaced (assume catalyst lifetime is 3 years):
Annual Catalyst Cost = (Catalyst Current Cost8) / 3
Annual Catalyst Cost = ( ) / 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 Cost8) / 5
Annual Carbon Cost = ( ) / 5 = $
(3) Annualized Refrigerant Replacement Costs
Refrigerant in a condenser needs to be replaced periodically due to system leaks;
however, the loss rate is typically very low. Therefore, assume the cost of refrigerant
replacement is negligible.
(4) Annualized Bag Replacement Costs (Line 7, Table C.12-6)
Over the lifetime of a fabric filter system the bags become worn and must be
replaced (assume bag lifetime is 2 years):
Annual Bag Cost = (Bag Current Cost3) / 2
Annual Bag Cost = ( ) / 2 = $
3See Table C. 12-2.
175
-------
Table C.12-6. Estimate of Annualized Costs in Current Dollars
Cost Elements
Units Costs/Factor
Annual Expenditure
Current Dollars
Direct Operating Costs
1. Natural Gasa
2. Fuel Oil8
3. Water3
4. Steam3
5. Electricity"
6. Solvent8
7. Replacement Parts
8. Replacement Labor
9. Operator Labor*3
10. Supervision Labor
11. Maintenance Laborb
12. Maintenance Materials
13. SUBTOTAL
Indirect Operating Costs
14. Overhead
15. Property Tax
16. Insurance
17. Administration
18. Capital Recovery"
19. SUBTOTAL
20. CREDITS
NET ANNUALIZED COSTS
$0.00425 per ft3 x
$1.025 per gal x
$0.0003 per gal x
$0.00504 per Ib x
$0.059 per kWh x
$ pergal0 x
As applicable (see Table C.12-5)
100% of Line 7
$11.53 per hr x
15% of Line 9
$11.53 per hr x
100% of Line 11
Add Items 1 through 12
80% of Sum of Lines 8, 9,10, and 11
1% of Total Capital Costd
1% of Total Capita! Costd
2% of Total Capital Costd
(CRF) x Total Capital Costd; where CRF =
Add Items 14 through 18
As applicable (see Section 5.2.3)
Item 13 + Item 19- Item 20
_ft3
-gal
-gal
Jb
_kWh
-gal
$-
$-
$-
$-
$-
. hr
hr
$-
$-
$-
$-
$-
$-
$-
$-
$-
$-
$-
$-
$-
8 See Table C. 12-4.
b See Tabled 2-3.
0 As applicable.
dTotal Capital Cost from Line 8 of Table C. 12-2.
6U.8. GOVERNMENT PRINTING OFFICE: 1986-646-116x1*061+7
776
------- |