MSAR 72-233
AC-860
Final Technical Report
to
Environmental Protection Agency
Office of Air Programs
Research Triangle Park, North Carolina 27711
on
HYDROCARBON POLLUTANT SYSTEMS STUDY
Volume I - Stationary Sources,
Effects, and Control
October 20, 1972
MSA RESEARCH CORPORATION
Evans City, Pennsylvania
-------
ACKNOWLEDGEMENTS
This study was performed for the Office of Air
Programs of the Environmental Protection Agency under the
technical administration of Mr. William R. King, EPA Pro-
ject Officer. Mr. King provided helpful assistance dur-
ing the literature collection and review phases of the
program, as well as constructive criticism of the interim
results used in compiling this report.
Th~ study was directed by Mr. William J. Cooper,
MSA Re~earch Corporation. MSAR personnel contributing sig-
nificantly to the study were Messrs. W.A. Everson, J.V.
Friel, J.S. Greer, and C.A. Palladino.
The MSAR study team was ably assisted by four
subcontractors: Industrial Health Foundation; Patent De-
velopment Associates; Singmaster and Br~yer; and University
Science Center. Space does not permit individual acknow-
ledgement of all contributing personnel, but the following
individuals are recognized for their subcontract direction
and contribution: Mr. Harry Bowman, IHF; Dr. B.J. Lerner,
PDA; Mr. Stanl~y Zukowsky, S&B; and Mr. Edwin Snow, USCg
This report wasfurniahed under
contract to the Environmental
Protection Agency (EPA). The
contents are reproduced herein as
received. Opinions, findings, and
conclusions expressed are those of
the author and not necessarily those
.0£ EPA. Mentioh of company or
. product names is not considered an
. endorsement by EPA.
i
-------
SUMMARY
This summary presents the results
of the Hydrocarbon Pollutant Systems Study.
the total study is followed by summaries of
port sectionso
Hydrocarbon air pollutants are emitted by the
major stationary sources in significantly greater quantities
than previously estimated. The total Quantity of hydro-
carbons emitted to the atmosphere is estimated in this study
to be in excess of 25 million tons per year, but this esti-
mate provides only a part of the hydrocarbon pollution pic-
ture. At concentrations measured in ambi~nt air, the non-
photochemically reactive hydrocarbon pollutants demonstrate
no apparent serious health ,threat. The major harmful effects
of hydrocarbon emissions result from their contribution to
photochemical smog, which is known to have adverse effects on
humans, vegetation and materials.'
and conclusions
An overview of
the major re-
Photochemically reactive hydrocarbon emissions are
estimated to constitute about 20 percent of total annual hy-
drocarbon emissions. These reactive hydrocarbons are derived
primarily from organic solvent evaporation, gasoline storage
and marketing, solid waste combustion, agricultural burning
of crop and field residues, and to a lesser extent, petroleum
production and refiningo
A state-of-the-art review of control technology
shows that hydrocarbon emissions could be significantly re-
duced by proper and widespread application of available con-
trol techniques. More specifically, emissions from uncon-
trolled sources could be reduced at least 50 percent by ap-
plication of available technology.
Certain available control techniques could and
should be improved; an R&D program to develop these improve-
ments and to fill other technology needs is recommended.
Specific R&D programs to improve characterization and mon-
itoring of emissions; documentation of adverse effects; and
interdisciplinary studies of air, water and solid w~ste pol-
lution are also recommended. For source categories that are
not directly amenable to applied controls (eg., the open
burning of solid and liquid wastes) or that ca~ be modified
to reduce the amount of control necessary (eg., solvent
evaporation sources), alternative practices and process mod-
ifications are suggested.
i i
-------
Insufficient quantitative data were available for
completion of a generalized systems model that would provide
the socio-economic benefit/cost relationships for application
of various control technologies to a given hydrocarbon pol-
lutant source. . Engineering cost studies do show, however,
that economic payout can be achieved in some applications of
control systems. Control techniques that do not offer payouts
can generally be applied without severe economic penalties.
Ranking of Pollutants and Sources
An attempt was made to combine emissions data with
data from a review of adverse effects to provide generalized
source-effect rankings and geographical distribution. Iden-
tifiable adverse health effects of hydrocarbon pollutants
were so limited, however, that relative health effects rank-
ings could not be justified. Certain pollutants can be iden-
tified as potentially hazardous, particularly to predisposed
segments of the population. Sensory irritation and aggravation
of existing respiratory ailments can result from chronic ex-
posure to low concentrations of formaldehyde and acrolein.
Laboratory studies of the polycyclic aromatic hydrocarbons,
particularly benzo-a-pyrene, have shown potential carcinogenic
activity, and although not specific, epidemiological studies
have revealed an urban factor that probably results from the
emission of greater amo~nts of polycyclic aromatic hydrocar-
bons in urban areas.
Because of the lack of data on adverse effects of
hydrocarbon pollutants specifically, the major factor util-
ized in ranking pollutants and sources was photochemical re-
activity, or relative smog contribution. Specific compounds
and some general chemical classifications have been demon-
strated to be photochemically reactive in smog chamber studies
and thus are principal contributors to smog. Application of
reactivity. parameters to the hydrocarbon sourc~s and their
total emissions resulted in the relative rankings shown in
Table 1.
Investigation of the .geographical distribution of
the major sources shows that certain categories, particularly
gasoline storage an~ marketing and solid waste combustion,
are highly population density dependent and thus are a wide-
spread urban problem. Other sources are more closely related
to i~dustrial processes; these sources may be generally wide-
spreaa, as are industrial solvent evaporation sources, or
they may be highly localized, as are petroleum refining and
petrochemical operations.
i i i
-------
Table 1 - "Best Estimate" Reactivity Ranking of Major Emission
Sources
Source Category
Solvent Evaporation
Total Hydrocarbon
.. 'Emissions
106 tons/yr
7 . 1
"High Reactivity"
Emissions
106 tons/yr
1.9
Solid Waste Combustion
(urban~ domestic,
commerci al, and
industrial)
Agricultural Waste
Combustion
4.5
1.4
4.2
1.1
Petroleum Products
Storage and
Marketing
Petroleum Production
and Refining
2.3
1.0
1.9
0.2
Chemical Process
Industry
Other Industrial
Processes
1.4
negligible
"'1
Fuel Combustion
0.4
0.2
negligible
negligible
Coal Refuse Burning
Forest Wildfires
not estimated
2.4-3.0
not estimated
Totals
25.4-26.0
5.6
iv
-------
Table 1 - "Best Estimate" Reactivity Ranking of Major Emission
Sources
Source Category
Solvent Evapo~ation
Total Hydrocarbon
EmissTons
106 tons/yr
7. 1
"High Reactivity"
Emissions
106 tons/yr
1.9
Solid Waste Combustion
(urban, domestic,
commerci a 1, and
industrial)
Agricultural Waste
Combustion
4.5
1.4
4.2
1.1
Petroleum Products
Storage and
t4arketi ng
Petroleum Production
and Refining
2.3
1.0
1.9
0.2
Chemical Process
Industry
Other Industrial
Processes
1.4
. negligible
"'1
Fuel Combustion
0.4
0.2
negligible
negligible
Coal Refuse Burning
Forest Wildfires
not estimated
2.4~3.0
not estimated
Totals
25.4-26.0
5.6
i v
-------
The necessary conditions for formation of photo-
chemical smog are highly dependent on local meteorological
conditions and geographical features. Reviews of meteoro-
logical variations and ambient air monitoring data show
that the highly urbanized regions of California are most, af-
fected by smog problems. Other major urban areas are much
less affected by severe smog conditions. However, almost
all urban areas are affected, at least a few times a year,
by excessive smog manifestations.
Vegetative damage from hydrocarbon air pollution
results primarily from photochemical reaction and, to a .
lesser extent, from ethylene emissions. Ozone, peroxyacety1
nitrate, and ethylene have been estimated to account for
nearly 80 percent of total crop damage from air pollution
in Ca1ffornia. Other areas of the country are much less af-
fected, but oxidants and ethylene do contribute significantly
to total u.s. crop damage.
Materials damage, although not widely investigated
or reported, does result from exposure of paint, rubber and
plastics to photochemical oxida~ts~
Major Sources and Their Emissions
Estimates of total hydrocarbon emissions were ob-
tained by application of selected emission factors to the
amounts of material processed by the major source categories.
Fuel combustion, solid waste combustion, solvent evaporation,
petroleum refining and marketing, chemical processing, and
some miscellaneous industrial sources were analyzed in this
fashion.
Emission data for many sources are limited to
estimates of total hydrocarbons, with little or no identi-
fication of specific chemical compounds or groups making
up the total emission. Where possible, estimates of com-
position were derived from published data, so that emission
and adverse effects data could be correlated.
The total emissions estimated for each of the
major source categories were included in Table 1, given in
the previous discussion of~r~'ative' ~~nking by smog con-
tribution.
v
-------
Adverse Effects
General reviews of the literature on human health
hazards, photochemical reactivity and vegetation damage were
made to provide input for the ranking of pollutants and
sources.
Reported data on human health effects of hydrocar-
bons are related primarily to industrial workroom exposure
and thus only indirectly relate to chronic, ambient air ex-
posures. Available data on chronic exposure to low concen-
trations of hydrocarbons do not, however, demonstrate a sig-
nificant health hazard from hydrocarbon emissions~ Low mol-
ecular weight oxygenated species, such as formaldehyde and
acrolein, have been noted to produce sensory and upper
respiratory irritation at low concentrations. Laboratory
studies, supported circumstantially by epidemiological
studies, show a potential human health hazard from exposure
to polycyclic aromatic hydrocarbons; some of which have been
shown to be carcinogenic in skin application studies of ex-
perimental animals.
Experimental data (i.e., hydrocarbon consumption,
oxidation rates, and eye irritation) from smog chamber
studies were correlated to give a Composite Reactivity Index.
Major organic compounds found or expected to be present in
ambient air are ranked on a scale of. relative reactivity of
from 0 to 10. Olefins, substituted aromatics and certain
oxygenated compounds constitute the most highly reactive
groupings.
A reveiw of vegetative damage by hydrocarbons
shows predominant contributions by ethylene and photochem-
ical oxidants. Exposure-injury relationships are reviewed
and typical injury patterns described. Economic cost data
on vegetative damage were found to be limited to estimates
for a few states and areas. Synergistic interactions and
problems of distinguishing injury patterns make the estimates
highly speculative.
Systems Model
A generalized benefit/cost model is developed
and illustrated. The lack of quantitative input data pre-
vented completion of the functional benefit/cost relation-
ship except in hypothetical terms. The model does provide
the basis for accurate cost/benefit estimates of pollution
in future. investigations.
vi
-------
The model examines damage caused by emissions in
five basic categories: human, material, animal, vegetable,
and indirect. Dam~ge functions are formulated and distri-
buted to allow the total net function to be computed for a
given stationary source. Application of respective control
costs then allows development of benefit/cost matrices.
Although incomplete, the model does point out the
need for future studies to overcome data deficiencies.
Control Technology
The state-of-the-art of current control technology
is reviewed. Deficiencies in existing techniques are iden-
tified, and specific studies of wet scrubbing design and
correlation are presented. A novel and highly useful pro-
cedure for correlation of particulate removal performance of
wet scrubbing devices is developed.
General case cost estimates are presented for gas-
oline storage tank control, waste gas incineration, solvent
adsorption, and wet scrubbing of particulate and soluble
organics emissions. The effects of process ranges and heat
recovery alternatives on capital and operating costs are
considered.
Floating roof controls for gasoline storage tanks
and solvent adsorption with recovery demonstrate significant
economic payouts. Waste gas incineration costs are shown
to be highly dependent on waste heat recovery and effluent
concentration. For effluent concentrations in the range
from about 13 to 18 percent of lower explosive limit (LEL),
the operating costs of catalytic and thermal incineration
are approximately equal. Analysis of cost data trends in-
dicates that catalytic incineration is more economical at
concentrations below this range and less economical at higher
concentrations. At concentrations equivalent to 25 percent
LEL and capacities in excess of 10,000 SCFM, heat recovery
from thermal incineration can provide economic payout in 10
years or less.
Process modifications and various control alter-
natives are considered in relation to their pqtential for
emission reduction. Existing and developable systems for
solid waste disposal are treated briefly. Generalized cost
considerations show that selection of an optimum disposal
system to replace open burning depends on several factors.
Transportation costs, land availability, waste types, and
collection procedures require individual studies for specific
regions.
vii
~
-------
Growth projections indicate hydrocarbon emissions,
if uncontrolled, will increase about 3 to 5 percent annually.
Widespread application of existing control technology can re-
sult in significant emission reduction, ba~ed on reported ex-
perience in California. San Francisco Bay Area studies show
that controls being used have reduced total emissions from
stationary sources nearly 70 percent. Controls in effect in
1969 reduced potential organic emissions from stationary
sources 52.5 percent. A complete ban on backyard waste burn-
ing in effect in 1970 will further increase control effective-
ness in the Bay Area.
R&D Recommendations
The identification and ranking of sources and types
of hydrocarbon emissions and their adverse effects delineated
areas for which application of control would result in sig-
nificant benefit. However, much further study is needed to
allow quantitative determination of the cost/benefit relation-
ships of specific strategies for control of hydrocarbon emis-
sions. To aid in future planning, a $5 million, 5-year
program is presented, emphasizing those research areas in
which current information is deficient.
The following major areas are recommended for re-
search and development:
Emissions and Sources - Detailed character-
ization of specific sources and emissions
is needed to provide a basis for quantitative
assessment of adverse effects and for proper
implementation of control planning. These
studies require improved monitoring tech-
niques and instrumentation to. provide con-
tinuous and chemically specific emission
analysis.
Adverse Effects - Chronic toxicity and epi-
demiological studies are needed because our
present knowledge of the health effects of
hydrocarbon pollutants is incomplete. These
programs should be complemented by smog
chamber studies over a broad range of sim-
ulated atmospheres with exposur.e of human
subjects to assess health manifestations.
The rather tenuous relationship of ambient
air hydrocarbon levels to photochemical oxi-
dant concentrations should be clarified and
viii
~
-------
refined by correlative investigation of
meteorological and air sampling data.
Control Technology - Studies are needed
to correct deficiencies in current con-
trol technology. These include investi-
gations of the performance, in terms of
control effectiveness or efficiency of
removal, of afterburners (both thermal
and catalytic) and wet scrubbers for par-
ticulate removal. Other recommended R&D
programs include studies to expand the
design base and range of application of
carbon adsorption, to improve the com-
bustion efficiency of waste gas after-
burners, and to update the data base for
estimation of storage losses.
The rapidly expanding and changing status of con-
trol application requires that a program be initiated to
revise and update, on a continuing basis, a comprehensive
'engineering manual for use by both plant engineers and con-
trol agency personnel. Finally, studies are recommended to
integrate the efforts of air pollution control implement-
ation with other efforts toward improvement of the total
environment. These studies would examine and recommend
steps to correct or avoid problems of water pollution or
solid waste disposal resulting from air pollution control.
ix
~
-------
TABLE OF CONTENTS - VOLUME I
ACKNOWLEDGEMENTS
SUMMARY
I.
II.
III.
I V.
INTRODUCTION
A.
B .
Purpose and Scope of Study
Background
RANKING OF POLLUTANTS AND SOURCES
A. Introduction
B. Health Effects
C. Photochemical Reactivity
D. Vegetative Damage
E. Materials Damage
References
MAJOR SOURCES AND THEIR EMISSIONS
A.
B.
C.
D.
Introduction
Fuel Combustion
Solid Waste Combustion
Petroleum Production, Refining and
Marketing
g. Solvent Evaporation
F. Chemical Process Industry
G. Other Industrial Processes
H. Geographical Distribution
References
ADVERSE EFFECTS OF HYDROCARBON POLLUTANTS
A. Introduction
B. Health Effects
C. Photochemical SmoQ
D. Vegetative Damage-
References
Page No.
i
i i
I .. 1
1-1
1-2
II ,..1
II -1
II.. 1
11-6
11-19
II - 20
II-21
111-1
III-1
111-5
111-15
I II - 29
111- 54
1II-67
111-71
II 1-84
111-93
IV-1
IV-1
IV-2
IV-27
IV-r39
I V". 50
-------
TABLE OF CONTENTS (continued)
V I I.
V.
SYSTEMS MODEL
A. Introduction
B. Model Development
C. Assumptions
:D. Mathematical Model
E. Examples
F. Discussion
References
VI.
CONTROL OF HYDROCARBON POLLUTANTS
A.
. B .
C.
Introduction
Control Technology
Economic Aspects of
Emission Control
D. Growth Projections
References
Hydrocarbon
RESEARCH AND DEVELOPMENT RECOMMENDATIONS
A.
B.
C..
D.
Introduction
R&D Priorities and Needs
Economic Impact of Control Planning
R&D Planning
. ,
Page No.
V-l
V-l
V- 3
V-6
V-9
V-17
V-28
V-3?
VI-l
VI-l
VI - 3
VI-23
V I - 85
VI-93
VII-l
VII-l
VII-l
VII-4
VII-6
-------
Table
1
11-1
II -2
II -3
II -4
I I 1-1
1II-2
III-3
111-4
.'
111-5
rr 1-6
III-7
1II-8
III-9
LIST OF TABLES. AND FIGURES
"Best Estimate" Reactivity Ranking
of Major Emission Sources
Total Estimated Hydrocarbon Emi~sions
From Stationary Source Categories
Reactivity Ranking of Stationary
Source Organic Emissions
Preliminary Reactivity Ranking of
Maj~r Emission Sources
"Best Estimate" Reactivity Ranking
of Major Emission Sources
Summary of Major Source Hydrocarbon
Emissions
Summary of Fuel Consumption - 1968
Stationary Sources
Uncontrolled Emission Factors for
Bituminous Coal (lb/ton of coal
burned)
Emission Factors for Natural Gas
Combustion
Emission Factors for Fuel Oil Com-
bustion
Fuel Wood Consumption
Estimated BaP Emissions from Fuel
Combustion
Regional and State Breakdown of Fuel
Combustion Emissions for 1968
Approximate Breakdown of Total Solid
Wastes Generation
Page No.
iv
II -2
q-ln
n."l
It-17
III-2
III-6
111-8
II 1..9
III-10
II 1...11
1II-13
1II-14
III-16
-------
LlsT OF TABLES AND FIGURES (continued)
Table
I1I:r10
III-ll
111-12
III..:13
III-14
111-15
III-16
II 1-17
III-18
III-19
I II - 20
III-21
III-22
111-23
III-24
III-25
Major Manufacturing Solid Waste
Generation
Average Solid Waste Collected
Household & Municipal Waste Gen-
eration and Disposal
Estimated Emission Factors for Solid
Waste CombustiQn Sources
Summary of Waste Generation and Com-
bustion
Summary of Estimated Organic Emissions
from Solid Waste Combustion Sources
Estimated Crude Oil Production Emis-
sions, Monterey County, 1967
Crude Losses in the Transportation
System
Hydrocarbon Emission Factors for the
Petroleum Industry
Refinery Emissions Reported by Com-
munity Surveys
Crude Storage Losses at Refineries
Gasoline Storage Loss at Refineries
Hydrocarbon Emissions from Refinerv
Operations
Gasoline and Non-Gasoline Type Pro-
ducts
Petroleum Product Storage Capacities
Petroleum Product Storage Losses
Page No.
III-21
III-25
III-25
III-27
I II - 30
III-31
II 1-36
I II - 38
II 1-40
1II-42
II 1-44
I II - 45
II1-47
lII-48
II I-50
I II-51
-------
LIST OF TABLES AND FIGURES (continued
Table
I II - 26
111-27
111-28
I II-29
I II - 30
III-31
I II - 32
II 1-33
II 1-34
I II - 35
II 1-36
111-37
II 1-38
111-39
1968 Summary of Petroleum Industrv
Hydrocarbon Emissions
Industrial Solvents
Recent Inventory Data on Solvent
Emissions
Reported Estimates of Solvent Con-
sumption by Coatings Industry
Estimated Solvent Emissions by End
Use Category
Estimated Hydrocarbon .Emissions From
Petrochemical Production
Chemical Industry Organic Emissions
Typical Oil, Air and Gas Rates Used
in Manufacture of Oil Furnace Blacks
Typical Composition of Oil Furnace
Carbon Black Exhaust Gas
Oven Coke Production and Distribution
in United States
Typical Material Balance for By-Product
Coke Oven Operation
Nature and Extent of Atmospheric Pol-
lution Cuased by Coke-Oven Batteries
Types and Quantities of Air Pollutants
in the Vicinity of Coking and By-
Product Plants
Estimated Non-Sulfur Organic Emissions
from Wood Pulping.
Page No.
III-55
III-57
III-58
III-61
III-66
II 1-68
II 1-69
III-72
I II - 72
1II-74
III-7-
III-77
III..78
III-81
-------
LIST OF TABLES AND FIGURES (continued)
Table
I II - 40
111-41
II 1-42
IV-l
IV-2
IV-3
IV-4
IV- 5
IV-6
I V-7
IV- 8
IV-9
IV-10
IV-ll
Special Naphtha Sales From Bulk
Terminals in Ten Highest Counties
Parameters for Regional Distribution
of Total Hydrocarbon Emissions
Distribution of Total Hydrocarbon
Emissions by Source Category and
Census Region
IndOstrially Important Solvents
Toxicity of Aldehydes to Animals via
Inhalation
Comparative Effects of Acute and
Chronic Exposures to Aroma~ic Hydro-
carbon Vapors in Air
Polycyclic Hydrocarbon Content of the
Air for Selected Cities
Aldehyde Yields From Photooxidation
of HydrocarDon-Nitrogen Oxide Mixtures
Atmospheric Reactions of Pollutants
With Ozone
Most Frequently Reported Odor Sources
Odor Qualities of Selected Odorants
Recognition Odor Thresholds and TLV's
of Some Odorants
Ranking of Reactivities of Hydrocarbons
When Photooxidized in Presence of Nitro-
gen Oxides Under Static Conditions
Reactivities of Hydrocarbons Based on
Ability to Participate in Photoox;dation
of Nitric Oxide to Nitrogen Dioxide
Page No.
II 1-87
I II - 89
111-90
I V- 3
IV-10
IV-14
IV-15
IV-18
IV-2D
IV-22
IV-24
IV-26
IV-3D
IV-31
-------
LIST OF TABLES AND FIGURES (continued)
Table
IV-12
IV-13
IV-14
IV-15
IV-16
IV-17
IV-18
\
IV-19
VI..1
VI-2
VI-3
VI-4
VI-5
VI-6
Hydrocarbon Reactivity in Nitric Oxide
Photooxidation
Eye Irritation R~activity
Comparison of Porduct Yields and Ef-
fects Caused by Various Organics
Composite Reactivity Index Va1ues for
Selected Hydrocarbons
Relative Concentrations of Some Unsat-
urated Hydrocarbons That Produce
Biological Response Similar to that
Produced by Ethylene
Typical Short-Term Effects on Photo-
chemicaJ Oxidahts on Vegetation
Page No.
IV-32
IV-34
I V - 35
IV-36
IV-41
IV-43
Typical Prolonged Effects on PhotoGhemica1
Oxidants on Vegetation IV..44
Relative Phytotoxicity of PAN Homo10gs
on Two Plant Species
Disposal Costs for Hypothetical Examp1e-
100,000 tons/yr
Typical Comparative Cost Date: Solvent
Vapor Degreasing Versus Aqueous Cleaning
Estimated Installed Costs of Gasoline
Storage Tanks
Summary of Estimated Gasoline Losses
Summary of Estimated Operating Cost~
for Gasoline Storage Tanks
Differential Savings vs Differential
Capital Investment and Payout
IV-45
VI-19
VI-24
VI-26
VI...29
VI-30
VI-32
-------
LIST OF TABLES AND FIGURES (continued)
Table
V 1-7
VI - 8
V~-9
VI-10
VI-"
VI-12
VI-13
VI-14
VI-15
VI-16
VI-17
VI-18
VI-19
Estimated Installed Costs Thermal
and Catalytic Incinerators - 15%
Lower Explosive Limit
Estimated Installed Costs Thermal
and Catalytic Incinerators -'25%
Lower ExPlosive Limit
Estimated Annual Operating Costs
Thermal and Catalytic Incinera~ors ~
15% Lower Explosive Limit
.Estimated Annual Operating Costs
Thermal and Cata1tyic Inci~erators r
25% Lower Explosive Limit
Differential Savings Versus Differential
Capital Investment and Payout
Differential Savings vs Differential
Capital Investment and Payout
Estimated Installed Costs of AdsorptiQn
Systems
Estimated Annual Operating Costs of
Adsorption Systems
Savings vs Capital Investment and Payout
for Adsorption/Solvent Recovery - 25%
"Lower Explosive Limit
Estimated Installed Costs of Inciner-
ator/Absorber Systems
Estimated Operating Costs of Inciner-
ator/Absorber Systems
Estimated Installed Costs of Sc~ubber/
Absorber Systems
Estimated Operating Costs of Scrubber/
Absorber Systems
Page No.
VI-42
VI-43
VI-48
VI-49
VI-56
VI-58
VI-67
V 1- 70
VI-73
VI-79
VI-81
VI-83
VI-8S
-------
LIST OF TABLES AND FIGURES (continued)
Table
VI-20
United States Total Gross Consumption
of Energy Resources by Major Sources
Page No.
VI..88
-------
LIST OF TABLES AND FIGURES (continued)
Figure
I 1-1
11-2
V-l
V-2
V- 3
V-4
V..5
V-6
V-7
V-8
V-9
V-10
VI-l
VI-2
VI-3
Petroleum Distribution Path and
Emission Points
Percentage Distribution of Emissions
and Population by Cen$us Region
Human Health Effects From Formaldehyde
Exposure
Annual Per Person Cost of Formaldehyde
Health Effects
Annual Per Person Cost as a Function
of Formaldehyde Exposure
Annual Human Benefit-Cost for Formalde-
hyde Emissions Control
Benefit-Cost Ratio for Formaldehyde
Control
Ethylene Damage to Vegetation
Per Acre Cost of Ethylene Damage tQ
Cotton
Per Acre Cotton Damage as a Fun.ction
of Ethylene Exposure
Annual Benefit-Cost for Ethylene
Emissions Control
Benefit-Cost Ratio for Ethylene Control
Estimated Installed Cost of Gasoline
Storage Tanks
Estimated Operating Cost of Gas91ine
Storage Tanks
Thermal Incinerator Without Heat Recovery
Flow Sheet
Page No.
111-34
III...92
V..19
V-20
V..2l
V:"23
V-24
V-26
V-27
V... 29
V-30
V- 31
VI-27
VI-31
VI:-36
-------
LIST OF TABLES AND FIGURES (continued)
Figure
VI-4
VI-5
VI-6
VI-7
VI-8
VI-9
VI-10
VI-ll
VI-12
VI-13
VI-14
VI-15
VI-16
VI-17
Thermal Incinerator With Primary Heat
Recovery - Flow Sheet
Thermal Incineration With Primary
and Secondary Heat Recovery - Flow
Sheet
Catalytic Incinerator Without Heat
Recovery ~ Flow Sheet
Catalytic Incinerator With Primary
Heat Recovery - Flow Sheet
Catalytic Incinerator With Primary and
Secondary Heat Recovery - Flow Sheet
Estimated Installed I~cinerator Cost -
15% L EL
Estimated Installed Incinerator Cost -
25% LEL
Estimated Catalytic Incinerator Operat-
ing Cost - 15% LEL
Estimated Thermal Incinerator Operating
Cost - 15% LEL
Estimated Catalytic Incinerator Operating
Cost
Estimated Thermal Incinerator Operating
Cost - 25% LEL
Adsorption With Solvent Recovery - Flow
Sheet
Adsorption With Thermal Incineration -
No Heat Recovery - Flow Sheet
Adsorption With Thermal Incineration/
Primary Heat Recovery - Flow Sheet
Page No.
VI-37
VI-38
V I - 39
VI-40
VI-41
VI-44
VI-45
VI-50
VI-51
VI-52
VI-53
VI-61
VI-63
VI-64
-------
LIST OF TABLES AND FIGURES (continued)
Figure
VI-18
VI-19
VI-20
VI-2l
VI-22
VI-23
VI-24
V I - 2'5
VI-26
VII-l
Adsorption With Thermal Incineration/
Primary and Secondary Heat Recovery'
Estimated Installed Adsorption System
Cost .
Adsorption Systems Estimat~d Operating
Cost
Incineration/Absorption - Process Flow
Sheet
Scrubbing/Absorption - Process F1Qw
Sheet
Estimated Installed Cost ~f Incinerator
and Absorber
Estimated Operating Cost of Incinerator
and Absorber System
Estimated Installed Cost of Scrubber and
Absorber
Estimated Operating Cost of Scrubber and
Absorber
Page No.
VI-65
VI-68
VI-71
VI-75
VI-77
V I - 80
VI-82
VI-84
V 1-86
. Milestone Schedule for 5 Year, $5 ,Million
R&D Program VII-15
-------
A.
1.
INTRODUCTION
PURPOSE AND SCOPE OF STUDY
This study of hydrocarbon air pollutants was con-
ducted for the Office of Air Programs ,of the Environmental
Protection Agency under Contract Number EHSD 71-12. The
study was performed by MSA Research Corporation, Evans City,
Pennsylvania with the assistance of four subcontractors:
Industrial Health Foundation, University Science Center,
and Patent Development Associates, all of Pittsburgh; and
Singmaster and Breyer of New York, New York. .
The study goal was the development of a problem-
solving R&D program for the control of hydrocarbon air pol-
lutants from major stationary sources. The primary goal
required the accomplishment of four major objectives:
(l)
(2)
(3)
(4)
Identification, ch~racterization and
ranking of all significant stationary
sources of hydrocarbon emissions.
Characterization of the effluent streams
from the major sources of hydrocarbon
e'missions.
Evaluation, both technical and economic,
of existing and developable technology
for control of hydrocarbon emissions.
Development of R&D priorities and recom-
mendations for a program that will ul-
timately lead to proven control hardware
and technology.
For. the purpose of this study, the term ~drocar-
bons included all organic compounds, such as alcoho s, acids,
aldehydes, esters and halides. Also included were such
borderline inorganic compounds as perchloroethylene. Spec-
ifically excluded from the study were the mercaptans and
other compounds whose predominant pollutant characteristic
is odor.
1-1
-------
As a result of preliminary consultation with the
EPA Project Officer, the major work emphasis was plac~d on
Objective (1), above; the identification and ranking of
sources and emissions. The approximate distributiQn of ef-
fort, expressed as percent of total manhours, wa~ a$ f01-
lows:
Objective (l) 75%
Objective (2) 6%
Objective (3) 12%
Objective (4) 7%
B. BACKGROUND
In earlier studies of hydrocarbon air pollutant$,
total emissions from major stationary sources were e$tima~ed
to be of the same order of magnitude as those from mobile
sources. The Nationwide Inventory of Air Pollutant Emis-
sions, 1968 (U.S. Department of Health, Educafion' a'ndWel-
fare, 1970) estimated stationary source hydrocarbon emissions
at 15.4 million tons per year and mobile source emis$;Qns at
16.6 million tons per year for 1968. This study WaS under-
taken to delineate the quantities, types and adverse effects
on human health and welfare of the major source hydro~arbQn
emissions in greater detail, and further, to review ~he ~on-
trol art and provide recommendations for future, control~
oriented R&D programs.
To achieve the study objectives, the project team
proposed the utilization of a systems study approach in
which th2 desired outputs, as defined by the Scope of Work
provided by EPA, were incorporated into a preliminary sys-
tems model. Identification of the input ~ata necess~ry to
complete the functional relationships of the mod~l thus pro-
vided a starting point for the survey of available inform-
ation. The major data input categories and their application
to the systems model are discussed briefly in the following
paragraphs.
1.
Major Sources and Emissions
The technical literature and reports of related
government contracts provided emission factors that relate
am?unts and types~'6f emissions to the quantities of mat-
erlals handled or processed by the major source categories.
I - 2
-------
These factors were applied to published statistics and esti-
mates of quantities processed to quantify source category
emissions.
For many of the major sources, published data in-
cluded only estimates of total hydrocarbon emissions.
Where no data were available on the individual species or
chemical groups making up the total, data from laboratory
studies, simulated source studies, and field tests of similar
processes or operations were used to estimate emissions of
specific chemical entities.
2.
Ranking of Adverse Effects
Estimates of total hydrocarbon emissions are sig-
nificant only if related to some adverse effect on human
health and welfare. A review of the ltterature on human
health effects of hydrocarbons at concentration levels ob-
served or estimated to be attainable in the ambient atmos-
phere revealed few documented instances of adverse effects
on human health. Vegetation and materials are damaged by
exposure to a few specific hydrocarbon compounds, but more
often by exposure to secondary pollutants resulting from
photochemical reaction of hydrocarbons with other pollutants
to form photochemical smog. Because.. photochemical smog
contributes to human discomfort and is known to damage
vegetation and materials, it is the principal parameter
used in this study to rank pollutants and sources.
Quantitative data on the socio-economic costs of
the adverse effects of hydrocarbon pollutants were insuf-
ficient to complete a functional systems model.
3.
Control of Hydrocarbon Emissions
A general review of the control art was made to
identify those existing control techniques for which eco-
nomic evaluations should be made. A second purpose of
the review was to identify deficiencies in the control tech-
nology. Deficiencies in the design bases and fundamental
mechanisms of wet scrubbing devices led to a more detailed
study of this technology.
Economic studies were made of the most generally
applicable control techniques. Industrial case histories
were utilized for the economic evaluation, and ranges of
pertinent operating perameters were investigated for their
effect on capital and operating costs.
1-3
-------
4.
R&D Program
The hypothetical systems model relationships were
used to identify specific data deficiencies that need to be
overcome in future studie$. Consideration of priorities
based on source rankings along with technological deficiencies
and control costs permitted development of recommendations
for basic and applied R&D, as well as alternative methods
that could be used to reduce hydrocarbon emissions.
1-4
-------
I I .
RANKING OF POLLUTANTS AND SOURCES
A.
INTRODUCTION
Identification and quantification of the major
hydrocarbon pollutants and their sources on a national
scale serve to define the total magnitude of the problem.
However, to place these data in perspective and to allow
priorities to be assigned for future R&D and control ap-
plication, relative rankings were developed in terms of
the adverse effects of these emissions and their geo-
graphical distribution.
In Chapter III, we present detailed estimates of
the total organic emissions from the major source categories
and, give breakdowns of their geographical distribution.
The total organic emissions from these sources are summarized
in Table 11-1, ranked in the order of total quantity of
emissions. The total estimated emissions of about 25 million
tons/year are significantly higher than earlier estimates
of about 15 million tons/year from stationary sources (U.S.
Department of Health, Education and Welfare, 1970).
In the following sections of this chapter, we
discuss pollutants from specific sources and their adverse
effects. Adverse effects considered were human health ef-
fects, photochemical smog contribution, vegetative damage
and some minor effects on materials.
B.
HEALTH EFFECTS
1 .
General
Human health effects are of paramount importance
in any consideration of the adverse effects of air pollu-
tants. This fact has been recognized and emphasized by
Congress in its enactment of legislation authorizing the
Environmental Protection Agency to promulgate standards
of ambient air quality and to expedite the control of air
pollutant emissions.
In the review of the human health effects
sented in the Air Quality Criteria for HYdrOC~rbo}Jl
Department of Health, Education and Welfare, 1970 ,
found that there is little documented evidence that
carbon air pollutants alone have adverse effects on
. . . .
p're-
(U.S.
it was
hydro-
the
I 1-1
-------
Table 11-1
Total Estimated Hydrocarbon Emissi9ns From
Stationary Source Categories
Source Category
Solvent Evaporation
Total Hydrocarbon Emi~s10ns
1 06' tons/~r
7, 1
Solid Waste Combustion
(urban, domestic, com~
mercial and industrial)
Agricultural Waste Combustion
4.~
Petroleum Products Storage
and Marketing
Petroleum Production and
Refining
4.2
2.3
1.9
Chemical Process Industry
Other Industrial Processes
1.4
"'1
Fuel Combusti on
0.4
0~2
Coal Refuse Burning
Forest Wildfires
2.4-3.0
Totals
25.4-26.0
II-2
-------
health of the general public. Our review of the literature
essentially confirmed this finding.
.> Documented health effects are limited to eye and
respiratory irritation from exposure to the low molecular
weight aldehydes and the organic peroxide and peroxy ni-
trate derivatives from photochemical reaction. Additionally
thereis.w,idespread concern over the potentially carcino-
genic effects of human exposure to the airborne polycyclic
aromatic hydrocarbons.
2.
Aldehydes and Smog Products
From our review of the health effects of the major
identified emissions and also of general organic chemical
group classifications, the only reported health effects at-
tributable to exposure to concentration levels approaching
those found in ambient air samples are eye and upper respir-
atory irritation observed on expsoure to the low molecular
weight aldehydes, particularly acrolein and formaldehyde.
Presence. in the atmosphere of these aldehydes is the result
of both direct emission from sources such as combustion of
fuels and wastes and the secondary or photochemical reactions
of reactive hydrocarbon emissions. Other photochemical re-
action products having similar irritant effects are the
organic peroxides and peroxy nitrates generally lumped with
ozone as atmospheric or "total" oxidants.
.' Smith (1~62) and Fassett (1963) reported findings
of sensory irritation on short term exposure to concentrations
in the range of from 2 to 4 ppm for formaldehyde and as low
as 0.25 ppm for acrolein. These concentrations are signifi-
cantly higher than those reported in ambient air but could
be readily achieved in localized areas near emission sources.
For example, Feldstein et al (1963) reported estimated con-
centrations of from 1 to 3 ppm of formaldehyde one to two
miles downwind of a large (2000 tons) open burn of land
clearing debris.
, Although some aldehyde emissions result from chem-
ical manufacture and processing, by far the largest emissions
of formaldehyde and other low molecular weight aldehydes re-
sult from incomplete oxidation in the combustion of fuels
and wastes. For example, total estimated emissions of al-
dehydes from fuel combustion are over 90 x 103 tons/year.
Inefficient combustion of solid wastes can result in alde-
hyde emissions which range from 10 to 30 percent of the
total organic emissions (Duprey, 1968; Feldstein, 1963).
11-3
-------
Both formaldehyde and acrolein have been i~enti~
fied as products from photochemical reaction (Altsh~ller,
1966). Other photochemical ~roducts identified in smog
chamber studies are the various organic peroxides ~nd per-
oxy nitrates. Peroxyacetyl nitrate (PAN) and perOXY~en~oyl
nitrate have been shown in smog chamber studies to be
potenti eye irritants. The presence of PAN has been dem9n-
strated in ambient air. However, according to Renletti and
Bryan (1961), it does not exist in sufficient concentration
to account for the degree of eye irritation experien~ed~
It has also been suggested (Smith, 1965) that expos~re to
PAN under conditions of exercise stress results in'increased
oxygen uptake.
The wide variety of both organic and inor9~nic com-
pounds present in smog effectively prevent singling out
specific compounds as contributors to specific noted adverse
effects. The severe sensory irritation and aggravation of
chronic respiratory ailments noted in smog areas' is undou,bt-
ed1y due to the combined and possibly synergistic effects
of all species present during these periods of pOor atmos-
pheric ventilation.
3.
Polycytlic Aromatic Hydrocarbons
Of considerable concern in relation to human
health are the polycyclic aromatic hydrocarbons (PAH). Cer-
tain of these PAH compounds, which are generally associated
with the particulate phase in various combustion an~ in~
dustrial process emissions, have been shown to proq~ce ~~r~
cinogenic activity on the skin of animals sUbjected to lab-
oratory studies. Epidemiological studies have also indicated
a correlation between the incidence of pulmonary cancer ~nd
exposure to atmospheres containing PAH. These studies are
reviewed in detail in Chapter IV, but several conclusions
and recommendations may be summarized here.
Even though most of the evidence relating to car-
cinogenicity of PAH compounds, particularly benzo-a-pyrene,
has been obtained from skin application of extracts to ex-
pertm!ntal animals, the potential danger is such that known
emiSSlon sources'of PAH should be controlled. The P9~entia1
so~rce$ of PAH emissions have been identified by H~ngebr~u~k
and coworkers (1964, 1965, 1967) as being primarily heat
generation and waste combustion processes, as well a~ in-
dustrial processes which involve heat processing of co~l or
petroleum based products such as asphalt, coke, and ~oa1 tar.
II-4
-------
Hangebrauck (1967), in summarizing experimental
data from tests of various sizes and types of heat gener-
ation processes, showed that, in spite of the wide range
of emission rates from anyone process, a correlation
exists between benzo-a-pyrene (SaP) emissions and the size
or heat output of the furnace. The highest SaP emission
rates occurred with the smaller units and decreased with
increasing furnace output, presumably in relation to the
efficiency of combustion. Although only limited data were
available on oil and gas fired units, the results tend to
emphasize that emission rates from coal fired units are
higher than those from oil or gas fired units of similar
size.
The influence of heat generation sources on the
total PAH emissions is emphasized in studies involving
sampling and analysis of atmospheric concentrations ofPAH,
such as reported by Sawicki (1960, 1962) and DeMaio and
Corn (1966). In general, significantly higher concentrations
of airborne PAH have been observed during winter months
than during summer or non-heating months.
. Studies of the formation of PAH compounds by
combustion sources have shown that airborne PAH are al-
most entirely adsorbed on the particulate or soot phase,
with negligible atmospheric existence of PAH vapors.
Commins (1962, 1969), Mukai (1964) and Thomas (1968) have
shown from laboratory studies that, at most, only trace
amounts of true vapor phase PAH are emitted from combu$tion
sources and then only in samples taken from very close'to
the combustion source. Thus, removal or control of partic-
ulate emissions should be effective in controlling atmos-
pheric PAH emissions.
This latter conclusion is partially .suppor.ted
by reported analyses of PAH in emissions from industrial
process and heat generation sources. Hangeb~~uck (1967)
reported analyses of emissions before and after wet
scrubber control of hot-road-mix asphalt processing.
Total benzene-soluble organic emissions were reduced by
about 50 percent by the combination dry cyclone and wet
spray scrubber. SaP emissions were reduced from 1,350
~g/m3 to <100 ~g/m3.
Hangebrauck (1967), Cuffe et al (1965), and
Diehl et al (1967) reported data on PAH emissions from
large coal-burning boilers. Cuffe and Diehl collected
samples only after fly-ash collection and thus do not
11-5
-------
show the effect of particulate removal. Hangebrauck reported
data from tests on two furnaces in which samples were taken
both before and after fly ash collection by electrostatic
precipitation. Three of these tests on a vertically-fired
furnace showed negligible reduction of total benzene-sQ1ub1e
organics and BaP (one test showed higher emissions after
the collector). The one test of a front-wa11-fired f~rnace
showed a 73 percent reduction of BaP emission rate by f1y-
ash collection.
Only limited conclusions may be drawn from these
scant data, but it appears that the various particulate re-
moval devices are not equally effective in removing PAH from
source emissions.
Hangebrauck (1967) estimated annual BaP emissions
from fuel combustion, waste combustion and selected stationary
industrial sources, using emission factors derived from source
testing data. In his estimate, residential and commerqia1
usage of coal accounted for 410 tons (nearly 90 percent) of
the total stationary source emissions of 461 tons~ However,
the base year used for his fuel usage distribution was not
specified. U.S. Bureau of Mines (1968) statistics on con-
sumer usage of coal show a marked decrease in retail deliveries
to residential and commercial coal users in recent year$
(Table 11-2). In 1968, coal usage accounted for only 4.2
percent of all fuel energy consumed by residential and ~pm-
mercial consumers.
From 1968 statistics (U.S. Bureau of Mines, 1968)
for the distribution of fuel consumption by consumer class,
we recalculated the BaP emissions from fuel combustion, us-
ing Hangebrauck's emission factors and percentage ~istribut;on
of hand-fired and stoker fired residential coal furnaces.
As discussed later in Chapter III, BaP emissions from the
residential and commercial coal usage category are thus esti-
mated at 203 tons/year. The decreasing trend in usage of
coal for residential and commerttal heating illustrated by
U.S. Bureau of Mines data for recent years (Figure 11-1)
will gradually reduce the BaP emissions from this source
category. However, efforts should be made to accelerate
this trend away from coal usage in residential furnaces
since this source segment could not sustain the economic
impact of app1jed controls. '
C.
PHOTOCHEMICAL REACTIVITY
1 .
General
Studies of photochemical smog formation have shown
that although hydrocarbons are essential smog precursors,
11-6
-------
5
5
4
4
4
4
4
Q
3
-3
ell
~ 3
+.J
UI 3
~ 3
-
ell 2
~ 2
~
(1) 2
>
;: 2
(1)
CI 20
o
o
o
o
o
o
o
1961 1962 1963 1964 1965 1966 1967 1968 1969
Year
Figure 11-1 - Annual Retail Deliveries of Bituminous Coal for
Recent Years (from U.S. Bur~au of Mines, 1969)
11-7
-------
not all hydrocarbons undergo photochemical reaction to pro~
duce the physical manifestations of smog. As in any chem~
ical reaction, differences in chemical structure and re~
activity affect the reaction rates and resulting reaction
products. Additionally, atmospheric reaGtions are comp1i~
cated by synergistic interaction of air contaminants other
than the usually considered smog precursors, hydrocarbons
and nitrogen oxides (NOx). Bufa1in1 (1971) and Wilson and
Levy (1970) have reviewed some of these syner~istic inter-
actions, particularly with respect to S02 oxidation and
interaction.
As detailed in Chapter IV, the literature data
on studies of the photochemical reactivity of the more pre-
valent organic air pollutants were correlated and cofupi1ed
to give a composite reactivity index ranking. Such factors
as nitric oxide oxidation rates, oxidant formation and eye
irritation were considered in developing this reactivity.
index. These and additional studies have been further re~
viewed to develop more exp1icit'ranking of the various pol-
lutants and sources.
Some of the more recent studies of photochemical
oxidation of hydrocarbons emphasize the highly arbitrary
and uncertain nature of attempts to obtain definitive rank-
ings of reactivity. Of particular interest are the studies
by Altshuller et a1 (1969) and Bufalini et al (1971) on the
oxidation of n-butane. Previous reviews and experimental
studies were generally concentrated on reaction systems in-
volving olefins and aromatics, since brief screening exper-
iments had shown the saturated hydrocarbons to have very low
reactivities. The work of A1tshu11er et a1 (1969) shows
that under conditions of high enough ratios of hydrocarbon
to NOx significant photochemical reaction can take place.
Bufalini et al (1971) have extended this work and shown that
although the amount of saturated hydrocarbon reacted is
10w,significant oxidation of nitric oxide occurs. Reaction
products common to other, more reactive systems are observe~,
such as ozone, peroxyacety1 nitrate, and aldehydes. Quot-
ing from A1tshu11er (1971), "therefore it appears that almost
every hydrocarbon except methane can produce some oxidant
when photooxidized in the presence of high enough ratios of
hydrocarbons to oxi des of ni trogen. II .
2.
Source Ranking
From studie~ made by the various air pollution
control agencies in California, the California Air Resources
, II-8
-------
Board (1969) reported total organic emissions from the major
source categories and also subdivided these total emissions
into two groups as either IIhighli or IIlowli reactivity.*
These ratings are based on the combined judgment of air pol-
lution control personnel knowledgeable in smog control ac-
tivities and thus should be given considerable weight. A
mitigating factor on the use of these percentage high re-
activity estimates is the probability that they reflect to
some extent the distributions forced by legislative controls
that are in effect in California, as compared with other
regions of the United States.
For preliminary ranking purposes, the California
data were reviewed and converted from the reported ton/day
basis to a percentage basis for each of the major emission
sources. The results are shown in Table 11-3, as averages
of data reported for the South Coast Air Basin (Los Angeles
and neiihboring counties) and the San Francisco Bay Area Air
Basin.*
The percentage factors given in Table 11-2 were
used to rank the major source categories according to their
relative IIhighli reactivity emissions, as shown in Table 11-3.
The two highest ranking source categories by this scheme
are solvent evaporation and gasoline storage and marketing.
Forest fires, which would also rank high, are not considered
in this or further discussions because of their highly vari-
able and uncertain nature and the lack of applicable control
procedures.
a. Solvent Emissions - The 20 percent IIhighli re-
activity classification for solvent evaporation emissions
given by the California Air Resources Board (1969) was un-
doubtedly based on the control regulations in effect, since
both the Los Angeles Rule 66 and the Bay Area Regulation 3
place an upper limit of 20 percent total reactive compounds
in any solvent formulation being emitted. Thus, the use of
20 percent IIhighli reactivity may be too low to represent
national solvent usage, which is generally not controlled by
regulation.
* IIHigh reactivityll emissions are so classified by comparison
of measured or estimated emission composition with, the de-
finitions of reactive hydrocarbons given by Los Angeles
Rule 66 and Bay Area Regulation 3.
** Reported values for the San Diego Air Basin were not based
on detailed emission inventory studies and thus were not
used in the averaging of reactivity percentages.
II-9
-------
Table 11-2 - Reactivity Ranking of Stationary Source Organic
Emissions (as percent of total emissions)*
Emission Source
% Reattivity
High Low
Petroleum
Production
Reffning
Marketing
o
10
45
100
90
55
Organic Solvent Users
Surface Coating
Dry Cleaning
Degreasing
Other
20
20
20
20
D
80
80
80
80
100
Chemical Industries
Metallurgical Industries
o
o
100
100
Mineral Industries
Incineration
Open burning,
Open burning,
Incinerators
dumps
backyard
11
11
11
89
89
89
Fuel Combustion
Steam Rower Plants
Other Industrial
Domestic and Commercial
o
o
o
100
100
100
Agriculture
Debris Burning
Orchard lieaters
Agriculture Product
Processing Plants
(Bay Area values only)
11
o
89
100
o
100
*Based on California Air Re$ources Board (1969)
11-10
-------
Table 11-3 - Preliminary Reactivity Ranking of Major
Emission Sources
Total Hydrocarbon
Emissions
106 tons r
Source Category
Solvent Evaporation
Petroleum Products
Storage and
Marketing
Solid Waste Combustion
(urban, domestic,
commercial and
industrial)
Agricultural Waste
Combustion
Petroleum Refining
Petroleum Production
Fuel Combustion
Chemical Process
Industry
Other Industrial
Processes
Forest Wildfires
Coal Refuse Burning
"High Reactivity"
Emissions
105 tons/yr
~ ,
7. 1
2.3
14.2
10.4
4.5
5.0
4.2
4.6
1.7
0.2
1.7
negligible
negligible
0.4
1.4
negligible
"'1
negligible
2.4-3.0
0.2
not estimated
not estimated
II-ll
-------
In order to arrive at a better or more represent-
ative estimate of solvent emissions reactivity, relative
photochemical reactivity factors were derived for the spec-
ific compounds making up the total solvent emissions cat-
egory. These reactivity factors are discussed in detail in
Chapter IV, (page IV- 27) bIt briefly, the methodology con-
sists of assigning a numerical index, on a scale from 0 to
10, to the individual solvent according to data on its rel-
ative reactivity and eye irritant production under simulated
smog conditions (smog chamber). These individual reactivity
assignments were arithmetically averaged to achieve a Com..
posite Reactivity Index.
Characterization of solvent emissions according
to photochemical reactivity through the use of individual
solvent reactivity indices is hampered by the uncertainties
associated with definition of the reactivity of the solvent
category of "special naphtha". "Special naphtha" is an a11-
inclusive term which includes all of the various petroleum
fractions that are used in solvent-type applications. These
petroleum fractions are generally classified by boiling
range, aromatic content, and solvency, and may vary from
narrow-boiling range, essentially single component fractions
to highly complex mixtures tailored for specific applications.
Kirk-Othmer (1969) reported a total of 12 billion
pounds of industrial solvents used in 1966 and gave the fol-
lowing breakdown by chemical groups: aliphatic hydrocarbons,
45%; aromatic hydrocarbons, 17%; chlorinated hydrocarbons,
13%; ~lcoho1s and esters, 10%; ketones, 8%; all others, 7%.
Lunche et a1 (1957) gave a breakdown of solvents used in
surface coatings in Los Angeles, aliphatic hydrocarbons ac-
counted for 56.7% and aromatic hydrocarbons for 22.6%. They
also gave sales volumes for the 11 most used solvents; ali-
phatic hydrocarbons accounted for 37.7% of the total volume.
, ,Aliphatic and aromatic solvents are often so-class-
ified on, the basis of solvency tests, such as the kauri-
butanol test (ASTM D 1133-54T), rather than on chemical com-
position. Aliphatic hydrocarbons thus classified have kauri-
butanol numbers in the range 25 to 55 (pure toluene = 100)
and aromatics have kauri-butanol numbers of greater than 55.
Thus aliphatic hydrocarbons may have aromatic contents as
high as about 30 percent. High naphthenic or cyclic paraffin
con tent can howev'er make th is' d'i s ti n ct i,on ,1 es s meah i ng fu 1',
sin'ce certain naphthenes can in'crease :so]v,ency wJthout in-
cre'as i'ngaroma t i c'i:ty 8 .
11-12
-------
. Because of the health hazard (in workroom atmos-
pheres) associated with benzene and the generally higher
solvency of the substituted aromatics, the benzene content
of aromatic solvents is generally kept quite low. Thus,
photochemical reactivity of special naphtha solvents may
be based primarily on substituted aromatic content.
Mellan (1957, 1970) gives boiling range
position data (as percent paraffin, naphthene and
for a number of commercial aliphatic and aromatic
formulations..
and com-
aromatic)
solvent
From a review of the data on solvent composition
and relative usage, we estimated the total "special naphtha"
category. as being 27 percent highly reactive.
By combining the above figure for the reactive
fraction of special naphtha with the total usage of other
solvents having a Composite Reactivity Index of 6 or higher,
we ar~ived at a total estimated emission of highly reactive
solvents of 38.4 x 108 lb/year, or approximately 27 percent
of total solvent emissions. Based on this figure as more
representing national usage, the total reactive solvent
emissions are increased from 14.2 x 105 ton/year to 19.2 x
105 ton/year.
Among the many factors complicating ranking of
industrial solvents is the widespread practice of baking
industrial coatings at sufficiently high temperatures to
produce photochemically reactive emissions from otherwise
unreactive or slightly reactive solvent formulations. This
problem has been recognized by both Los Angeles and Bay Area
Control Districts, in that bake oven emissions must be con-
trolled regardless of the ex :mpt or non-exempt status of
the solvent formulation used in the coatings application.
High temperature drying is a common production-
line operation in several industries. Major examples in-
.
The composition and reactivity of a number of commercial
solvents has been listed in Bay Area Information Bulletin
1-69 (1969). Based on aromatic and olefin content and
the definitions of Regulation 3, the solvents are classi-
fied as reactive, questionable or unreactive. Of the 143
solvents listed, 77 are reactive, 36 are questionable and
30 are unreactive. The listed solvents are almost entirely
petroleum based solvents used for surface coating and an-
alogous solvent usage, most of which would fall under the
category of "special naphtha",
II-13
-------
clude aircraft, furniture, metal can, appliance, ~nd auto-
motive painting. All types of surface coatings may be .
dried at elevated temperatures, with hydrocarbon emissions
ultimately dependent upon the specific coating formulation
and the elevated temperature drying conditiQns. Emissions
are maintained below the "25% lower explosive limit"
standard set by fire insurance underwriters, by diluting
with air. Exhaust gases from high-temperature drying ovens
contain partially oxidized and polymerized molecules in
addition to the paint solvent species. Often th~se reaction
products are present in the form of an aerosol 1n the ex-
haust. Solvents which contain long chain or unsaturated
groups may be cleaved, forming aldehydes, ketones, or acids
during th~ir exposure to these high temperatures and oxidiz-
ing conditions.
Thus, the ranking of a particular solvent as un-
reactive applies only for usage where little or nO heat is
applied during any subsequent drying or curing steps
b. Gasoline Evaporation Emissions - The California
rating of gasoline evaporative losses as being 45 percent
"high" reactive is probably based on the typical 40 to 50
percent total aromatic plus olefin content of most motor
gasolines.
Maynard and Sanders (1969) have reported the de-
termination of the composition and reactivity of some typical
premium grade gasolines. Their detailed analysis of a
p~emium grade fuel shows specific identification of 233
hydrocarbons. Their range of calculated reactivity, based
on the summation of component reactivities, is 1.32 to 2.10.
In order to be consistent with other approximations in
this study, the reported analysis of one premium grade gas-
oline has been combined with our reactivity factors and a
Composite Reactivity Index value developed. On the basis
of major identified components, accounting for about 80
moll of the total composition, a Composite Reactivity Index
value of 2.2 was obtained~ which is in reasonable agreement
with the range given by Maynard and Sanders.
Two distinct problem areas make it difficult to
predict the reactivity of gasoline emissions.
1.
Modern trends toward the elimination
of lead alkyls have resulted in in-
creased blending of aromatics to im-
prove octane ratings. The aromatic
11-14
-------
2.
compounds used. toluene and xylenes. have
greater photochemical reactivities than
most of the branched-chain alkanes. Ec-
cleston (1970) reported combined aromatic
and olefin contents of some prototype un-
leaded fuels of over 55 mol% (44.9 mol%
aromatic. 10.7 mol% olefin).
The highly volatile components in gasolines
have. in general. lower reactivities. but
are probably released during the first
storage and handling processes. Subsequent
emissions involve the less-volatile and
more reactive components. In addition.
not all emissions are initiated by the most
volatile components. An indefinite amount
of spillage occurs during handling; this
spillage may be treated on a 1:1 basis the
same as solvents (everything is assumed to
be emitted into the atmosphere).
Eccleston (1970) evaluated total evaporative losses
of leaded and prototype unleaded gasoline. Using U.5. Bureau
of Mines 'reactivity factors. he estimated a 12 percent in-
crease in photochemical reactivity for premium grade and about
28 percent increase for regular grade. in going from a typical
leaded to a prototype unleaded gasoline. .
c. Waste Combustion Emissions - In Table 11-3.
waste incineration and agricultural burning were ranked ac-
cording to the overall "high" reactivity factor calc~lated
from the data of the California Air Resources Board. This
factor indicated that 11 percent of all waste combustion
emissions are highly reactive. .
From an analysis of the emission factors reported
for the various waste combustion modes. the high reactivity
factor of 11 percent appears to be significantly low. When
the emission factors for unsaturated hydrocarbons and alde-
hydes are combined to estimate the reactive fraction of
total emissions. the percentage of reactive emissions ranges
from about 20 percent for industrial incineration to about
40 ~ercent for domestic incineration.
Open burning. the principal disposal mode for all
solid wastes. results in emissions that are about 27 percent
reactive. Agricultural waste disposal emissions. which are
almost entirely open burned. are thus estimated as being 27
11-15
-------
percent reactive. Municipal, domestic (both urban and rural),
and industrial wastes have been weighted accQrding to their
combustion modes to give an estimated average of ~O percent
reactive emissions.
Based on these factors, high reactivity emissions
from the major waste disposal categories of s01id waste burn-
ing and agricultural burning are thus incre~se~ tQ 13.5 mil-
lion, and 11.3 million tons/year, respectively.
d. Summary Source Ranking - In summary, the best
estimate of source ranking according to photpchemica1 re-
activity is given in Table 11-4. This ranking r~pr~sents the
relative source contributions to photochemical smog on a nat-
ional basis. Because of the very pronounced variations in
smog potential with regional geographic and me~eor010gical
conditions, such national estimates may not show a true
pitture. The regional parameters affecting smog formation
are discussed in the following subsection.
3.
Regional Factors Affecting Photoc~emica1 Smog
Significant to the discussion of source ranking ,by
photochemical reactivity is some consideration of the regional
differences that affect photochemical smQg formation and its
adverse effects.
Even though considerable study may yet be required
to elucidate completely the mechanism pf photochemic~l smOg
formation, experts tend to agree that certain atmospheric
conditions are necessary for significant photochemical smog
formation. These necessary conditions are gen~rally given
as:
1)
presence of reactive organics (above
some threshold value)
2)
presence of nitrogen oxides in suf-
ficient quantity to initiate photo-
chemical reaction
3)
a stagnant or poorly ventilated at~
mosphere to trap and retain reactants
intense sunlight
4)
The existence and revel of photochemical smog is
. generally monitored by total oxidant concentration which in-
II-16
-------
Table 11-4 - "Best Estimate" Reactivity Ranking of Major
Emission Sources
Source Category
Solvent Evaporation
Solid Waste Combustion
(urban, domestic, com-
mercial, and industrial)
Agricultural Waste Com-
bustion
Petroleum Products Stor-
age and Marketing
Petroleum Production and
Refining
Chemical Process Industry
Other Industrial Processes
Fuel Combustion
Coal Refuse Burning
Forest Wildfires
Totals
Total Hydrocarbon
Emissions
106 tons/yr
7. 1
"High Reactivity"
Emissions
106 tons/yr
1.9
4.5
1.4
4.2 1.1
2.3 1.0
1.9 0.2
1.4 negligible
"'1 negligible
0.4 negligible
0.2 not estimated
2.4-3.0 not estimated
25.4-26.0 5.6
11-17
-------
cludes not only ozone, but other photochemical oxidants such
as peroxyacetyl nitrate (PAN) and organic peroxides. Most
of the recent literature on episodic distribution of $mog is
based on the 1968 California standard for ambient ~ir quality
of 0.15 ppm oxidant for one hour by the neutral KI method.
The recent and more stringent national ambient air standard
for photochemical oxidant has not been considered in our
treatment of geographic distribution, but would greatly in-
crease the number of so-called smog days for the various af-
fected regions. .
Historically, California has been most affected by
photochemical smog, primarily because of the meteorological
conditions which predominate over much of the state during
the late:'summer and early fall months. These meteorological
conditions fulfill the third and fourth r~quirements listed
above for the formation of smog, namely stagnation and strong
sunlight. In recent years, however, the increased levels of
reactant precursors over other metropolitan areas of the U.S.
have resulted in almost all parts of the country being af-
fected at least occasionally by adverse smog levels.
Holzworth (1967, 1969) analyzed the meteorological
factors relating limited dispersion to smog pptential for
various regions of the United States. He showed that the
episodic nature of limited atmospheric dispersion ~an be
assessed in terms of morning and afternoon mixing heights
and averaged wind velocities through the mixing layers.
After defining an episode of limited dispersion
as no mixing height greater than 1500 meters, no average
wind speed greater than 4.0 m/sec, and no precipitation,
all for at least 2 days, Holzworth used this definition in
an analysis of the five year weather data for four stations.
In 5 years, the total of such episode-days were as follows:
Oakland, California, >200; Washington, D.C. and Denver,
Colorado, rv40; St. Cloud, Minnesota, 3. To be truly in-
dicative of photochemical smog potential, these weather
data on dispersion trends should be combined with solar
irradiation data, but they do, however, show the general
trend across the U.S.
Faith (1968) reported that analysis of ambient
ai~ monitoring data on oxidant concentration indicated
the following trend in annual frequency of smog days (oxi-
dant exceeding 0.15 ppm for 1 hr): Los Angeles Basin, 200
day~; ~an Francisco Bay Area, 50 days; Denver, 14 days;
St. Louis, 7 days; Cincinnati, 5 days; Washington, D.C.,
4 days; Chicago, none. The California Air Resources Board
11-18
-------
(1969), using the same oxidant level cri,terion, have re-
ported the following smog day frequencies (days/year) for
California air basins in 1968: Los Angeles, 206; .San
Francisco, 67; San Diego, 95.
Studies conducted in both Los Angeles and San
Francisco have shown that measured ambient oxidant l~vels
are quite dependent on geographic location of sampling
stations within an air basin. For example, Sandberg et al
(1971) reported oxidant concentration trends for the 8
monitoring stations in the San Francisco Bay Area basin.
The average high-hour oxidant concentrations follow the
general off-shore f}ow of air, with the highest concen-
trations observed at Livermore, which is in a sheltered
valley downwind of the central district. The 1969 average
high-hour concentrations for days with comparable weather
conditions show a four-fold range, from 0.04 ppm at San
Francisco to 0.18 ppm at Livermore. This local variation
within a particular air basin shows that the single station
data available in many metropolitan areas may not be a
true indic~tion of maximum levels within that area.
Reported data
California are affected
problems than the major
tailed studies, similar
needed to show the true
in any particular area.
indicate that regions outside of
by order-of-magnitude lower smog
California air basins. More de-
to those in San Francisco, are
magnitude of the smog potential
D.
VEGETATIVE DAMAGE
The secondary pollutants resulting from photo-
chemical reactions produce the greatest amount of veg-
etative damage by air pollutants in specific areas. Of
the primary hydrocarbon air pollutants, only ethylene
has been found' to produce significant vegetative damage.
Ethylene emitted by industrial processes and by
fuel and waste combustion acts as a growth modifier on a
wide variety of plants. Its greatest economic impact re-
sults from damage to citrus fruit and ornamental and cut
flowers. Exposure to ethylene concentrations as low as
0.1 ppm for a few hours have resulted in sepal damage to
sensitive flowering species.
Smog products, principally ozone and peroxy-
acetyl nitrate (PAN), have extensively damaged trees and
crops in the smog affected areas of California. Recent
11-19
-------
estimates of crop losses from air pollutant da~~ge in Cal-
ifornia, (Anon., 1971), show that smog produ~ts and e1:hy-
lene are responsible for most of the tot~l damage. The
individual pollutants and their estimated percentage con-
tribution to total crop damage are:
Ozone
PAN
Fluorides
Ethylene
S02
Particulates
- 50%
,.. 18%
- 15%
.. 14%
.. 2%
1%
Weidensaul and LaCasse (1970) pr~~en~ed results
of a statewide survey of air pollutant damage to vegetation
in Pennsylvania. Although their data are not representa-
tive of annual pollutant damage, the survey resul~$ showed
significant losses from oxidant exposure damage.
E.
MATERIALS DAMAGE
Materials damage by hydrocarbon pollution is not
well documented. The most often mentioned adverse effects
are those of photochemical oxidants on rubber and plastics
resulting in crac~ing and loss of elasticity. Other ef-
fects noted have been discoloration and deteriQration of
architectural coatings, and the formation of resistive
coatings on electrical contacts resulting from'polymeri~ation
of organic air pollutant species. .
Only crude estimates of total materials ~amage
costs have been reported. Wohlers and Feldstein (1965),
for example, estimated economic effects of smog products
in the Bay Area. Total yearly costs in materials damage
to fabrics, rubber, paint and electrical contact points
was estimated at $15 million.
lI-20
-------
CHAPTER II
REFERENCES
A1tshuller, A.P., D.L. Klosterman, P.W. Leach, I.J. Hindawi,
and J.E. Sigsby, Jr., "Products and Biological Effects
from Irradiation of Nitrogen Oxides with Hydrocarbons or
Aldehydes Under Dynamic Conditions", Int. J. Air Water
Pollution }Q, 81-91 (1966).
A 1 t s hull e r, A. P., S. L. K 0 P C zy n ski, D . Wi 1 s on. W. Lon n em an,
F.D. Sutterfield, "Photochemica1 Reactivities of N-Butane
and Other Paraffinic Hydrocarbons", J. Air Pollution Con-
trol Assoc. li, 787 (1969).
A1tshu11er, A.P. and J.J. Bufa1ini, "Photochemica1 Aspects
of Air Pollution: A Review", Environ. ScL Technol. 5
(1), 39-64 (1971).
Anonymous, "Contro1 District News, State of Ca1ifornia",
J. Air Pollution Control Assoc. fl (9). 580 (1971).
Bufalini. J.J.. B.W. Gay and S.L. Kopczynski, "Oxidation of
N-Butane by the Photolysis of N02"' Environ. Sci. Te'chno1.
~ (4), 333-6 (1971).
California Air Resources Board, "Emission Inventories",
November 1969.
Commins, B.T., liThe Spectrophotometric Determination of Poly-
cyclic Aromatic Hydrocarbons in Polluted Air", Ph.D.
Thesis, University of London (1962).
Commins, B.T., "Formation of Polycyclic Aromatic Hydrocarbons
During Pyrb1ysis and Combu~tion of Hydrocarbons", Atmos.
Environ. 1, 565-72 (1969).
Cuffe, S.T., R.W.
"Air Pollutant
Report No. 2",
353-62 (1964).
Gerstle, A.A. Orning~ and C.H. Schwartz,
Emissions from Coal-Fire~ Power Plants,
J. Air Pollution Control Assoc. 11(9),
DeMaio, L., and M. Corn, "Po1ynuclear Aromatic Hydrocarbons
Associated with Particulates in Pittsburgh Air", J. Air
Pollution Control Assoc. 1£", 67 (1966).
11-21
-------
REFERENCES, CHAPTER II (continued)
DiehL LK., F. duBreuil ~ and R.A. Glenn, "Polynuc1ear Hy-
drocarbon Emissions From Coal Fired Installations",
Trans. ASME 89, Series A, 276-82 (1967).
Duprey, R.L., Com i1ation of Air Pollutant Em1ssio~ F*ctors,
Public He a 1 t hS e r vie e, P B 190 245 1968. '"
Eccleston, B.H. and R.W. Hurn, "Comparativ~ Emissions from
Some Leaded and Prototype Lead-Free Automobile Fu~ls",
Bureau of Mines Report of Investi9ation 7390 (1970).
Faith, W.D., "The Photochemistry of Solvent Vapors", Air
Eng. lQ. (2), 16-7(1968).
Fassett, D.W., Industrial H iene
Ed., Vol. II, Interscience Pu
Feldstein, M., S. Duckworth,
"The Contribution of Open
to Air Po1.1ution", J. Air
542 - 5 (19-63).
H.C. Wohlers, and B. Linsky,
Burning of L~nd Clearing Debris
Pollution Control Assoc. 13 (11),
.. -
Hangebrauck, R.P., D.J. Von Lehmden, and J.E. Meeker, "Emis-
sions of Polynuclear Hydrocarbons and Other Pollutants
from Heat-Generation and Incineration Proc~$ses", J. Air
Poll uti. 0 n Con t r 0 1 Ass 0 c. li (7), 267 - 78 (1 964 ) .
Hangebrauck, R.P., D.J, Von Lehmden, and J.E, Meeker, "Poly-
nuclear Hydrocarbon Emissions from Selected Ind~~trial
Processes", J. Air Pollution Control Assoc. 15 (7), 306-12
(1965). -
Holzworth, G.C., "Mixing Depths, Wind Speeds and Air Pollution
Potential for Selected Locations in the United States", J.
Appl. Meteor, 6 (6), 1039-44 (1967).
II - 22
-------
REFERENCES, CHAPTER II (continued)
Holzwort~, G.C., "Large Scale Weather Influences on Community
Air Pollution Potential in the United States", J. Air Pol-
lution Control Assoc. 12 (4)~ 248-54 (1969).
Kirk-Othmer, Encyclopedia of Chemical Technology, 2nd Ed.,
Interscienc~ (1969).
Koyama, T., J. Geophys. Res. 68, 3971 (1963).
Lunche, R.G., A. Stein, A.J. Seymour, and R.L. Weimer, "Emis-
sions from Organic Solvent Usage in Los Angeles County",
J. Air Pollution Control Assoc. I (4), 275-83 (l957).
Maynard, J.B., and W.N. Sanders, "Determination of the De-
tailed Hydrocarbon Composition and Potential Atmospheric
Reactivity of Full-Range Motor Gasolines", J. Air pol-
lution Control Assoc. 12 (7), 505-10 (1969).
Mellan, I., Handbook of Solvents, Reinhold Publishing, N.Y.,
(l957).
Mellan, I., Industrial Solvents Handbook~ Noyes Data Cor-
poration, Park Ridge, N.J., (1970).
Mukai, M., B.D. Tebbens, and J.F. Thomas, "Multidimensional
Chromatography of Arenes Produced During Combustion", Anal.
Chern. l£, 1126-30 (1964).
Renzetti, N.A. and R.J. Bryan, "Atmospheric Sampling for Al-
dehydes and Eye Irritation in Los Angeles Smog - 1960", J.
Air Pollution Control Assoc. 11, 421-4, 427 (196l).
San Francisco Bay Area Air Pollution ContrQl District, Sol-
vent Compositions, Information Bulletin 1-69 (l969).---
S an db erg, J. S., R . T h u i 11 i e I" , an d M . Fe 1 d s t e in, "A S t u dy of
Oxidant Concentration Trends in the San Francisco Bay Area",
J. Air Pollution Control Assoc. £1 (3), 118-21 (1971).
Sawicki, E., ToRo Hauser, W.Co Elbert, FoTo Fox, and J.E.
Meeker, "Polynuclear Aromatic Hydrocarbon Composition of
the Atmosphere in Some large American Cities", Amo Indo
Hyg. Assoco Jo £1, 137-44 (l962)o .
II-23
-------
REFERENCES, 'CHAPTER II (continued)
Sawicki ~ E., W.C. E1bert~ 1.R. Huaser, F.L Fox, and T.W.
Stanley, "Benzo(a)pyrene Content of the Air of American
Communities", Am. Ind. Hyg. Assoc. J, 21,443-51 (1960).
......-
Smith, C.W., Acrolein, John Wiley, New York (1962).
Smith, L.E., "Peroxyacety1 Nitrate Inha1!itiol1 Cardiores.,.
piratory Effects", Arch. Environ. He~lth ],Q, 161.-4 (1965).
Thomas, J.F., M. Mukai~ ai1d B.D. Tebbens, "F~te of Airborne
Benzo(a)pyrene", Environ. Sci. Technol. £ (1), 33-9 (1968).
U.S. Bureau of Mines, Minerals Yearbook, (19p8).
U.S. Bureau of Mines, Minerals Yearbook, ,(1969).
U.S. Department of Health, Education and Welfare, Nationwide
Inventory of Air Pollutant Emissions - 1968, NAPCA Pub1.
No. AP-73 (T97O). ',. , '
U.S. Department of Health, Education and Welfare, Air Quality
Criteria for Hydrocarbons, NAPCA Publ. NO. AP-6~970).
Weidensaul, 1.C., and N.L. LaCasse, Statewide Survev of Air
Pollution Damage to Vegetation, 1969,Ce~ter for Air En-
vironment Studies, Pennsylvania State University, (1970)~
Wil son g W. E. and A. Levy ~ II A Study of S02 in Photoc;hemi ca 1
Smog 1. Effect of S02 ahd Water Vapor Concentration in
the l-butene/NOx/S02 System", J. Air Pollution Control
Assoc. £Q. (6), 385-90 (1970).
Wohlers~ H.C., and Mo Feldstein, "Investigation to Determine
the Possible Need for a Regulation on Organic Compound
Emissions from Stationary Sources in the Bay Areal!, J.
Air Pollution Control Assoc. Ii (5), 226-9 (1965).
II - 24
-------
III.
MAJOR SOURCES AND THEIR EMISSIONS
A.
INTRODUCTION
The development of ,the ranking of pollutants and
sources given in the preceding section required the identi-
fication, definition, and detailed analysis of hydrocarbon
emissions from the major stationary sources. In the initial
stages of the study, criteria and guidelines were developed
which limited the study to significant areas; these criteria
and guidelines were, however, broad enough to allow inclusion
of source categories not considered in the preliminary screen-
ing selection.
The major source categories treated in this chapter
are shown in Table 111-1. This table serves not only as a
summary of estimated total hydrocarbon emissions from the
various sources but also as an index to the chapter contents.
The major literature sources utilized for this
portion of the study were the trade magazines and:general
technical literature, as well as the reports of related gov-
ernment-sponsored contracts. Additional input was obtained
from questionnaire surveys of pollution control agencies and
trade associations. These latter data sources were of: limited
utility, but did provide some input in specific areas, Dis-
cussion and analysis of the questionnaire surveys are presented
in Appendix G.
The general approach to the review of each of the
major source categories was to define, where possible" the
total amounts of material processed and to apply published
emission factors for the calculation or estimation of hydro-
carbon emissions. Because of the qualitative and sometimes
contradictory nature of much of the literature information9
rather broad assumptions were made to arrive at emission
estimates. These assumptions are defined and, where pos-
sible, documented with actual examples. .
With some few exceptions, data from the foreign
literature were not utilized, primarily because of a lack
of information on the similarity of foreign practices with
those used in the United States. Since differences in raw
materials, processing steps, and types of control in prac-
111-1
-------
Table 111-1 - Summary of Major Source Hydrocarbon Emissions
r~ a j 0 r Sou r c e
Co a 1 Comb us ti on
Nai;ural Gas
Comb us t ion
rue 1 Oi 1 Com-
bustion
t~ood Comb us ti on
Coal Refuse
Combustion
Agri eul tura 1
Burning
Forest Hildfires
Industrial Solid
Waste Combustion
Secondary Source
Total Hydrocarbon
Emissions
105 tons/yr
Electric Utilities
Industrial
Domestic and Commercial
0.43
0.46
0.26
Electric Utilities
Industrial
Commercial and In-
stitutional
Domestic
O. 1 n
0.36
0.21
0.24
Electric Utilities
Industri al
Commerci al and In-
stitutional
Domestic
0.33
O. 15
0.15
0.62
Industrial
Domestic
0.17
0.83
2.0
Field and Crop Residue
Controlled Timber and
Land Clearance
25.0
17.0
24.0-30.0
III-2
Page
Re fe ren ce
111- 7
111- 9
III- 10
III-ll
111- 16
111- 17
III- 18
II 1- 19
-------
Table 111-1 (continued)
!1!Jor Source
f1 u n i c i pal and
Domes ti c Sol i d
Waste Combustion
Crude Oil Production
Petroleum Refining
Petroleum Product
Marketing
Organic Solvcnt
Usage
Total Hydrocarbon
Emissions
105 tons/yr
3.0
Secondary Source
Major Manufacturing
Miscellaneous Manu-
facturing and Com-
mcrcial
lumbering
Construction and
Oemol1tion. .
Food Processing and
Marketing
Urban
Urban
Rural
Rural
Collected
Uncollccted
Collected
Uncollected
Crude Storage at Rc-
fineries
Crude Storage in
Transportation System
Gasoline Storage at Re-
fineries
Refinery Operations
Gasoline Storage
Other Distillate Storage
Gasoline Transfers
Other Distillate Transfers
III-3
Reference
7.0
7.0
5.0
5.0
111-23
7.0
4.0
4.0
3.0
2.2 111-33
111-39
3.2
4.7
4.2
8.3
II 1 - 46
7.8
2.9
12.0
1.0
II I-54
-------
Table 111-1 (continued)
Hajor Source
Chemical Process
Industry
Specific Industrial
Processes
Secondary Source
Surface Coatings
Degreasing
Dry C 1 e ani n 9
Printing and Publishing
Rubber and Plastics
Othe~ Solvent Usage
(unidentified sources)
Carbon 131ack
Coke Ovens
Pulp and Paper
(Kraft, Softwood only)
Phthalic Anhydride
Rubber and Plastics
(vinyl chloride,
butadiene and
acrylonitrile only)
1II-4
To ta 1 By dro;ca rb on
Emts$ions
105 tons/yr
30.0-40.0
6.5-9.5
2.5-4.0
1 . 8- 2. 3
6.0-7.5
Page
Reference
8.1-24.6
14.0
II I -6 7
'"
3.0 I I 1_71
2.0 II 1_73
0.5 II 1-79
0.1-0.4 II I -82
0.4 I I I -83
-------
tice have a direct bearing on the resultant emissions, the
survey was generally limited to literature directly per-
taining to sources in the United States.
B.
FUEL COMBUSTION
Stationary fuel combustion sources are represented
by the wide variety of furnaces and boilers used to convert
fuel to heat energy. These include steam-electric utility
boilers, industrial and commercial process boilers and space
heaters, and domestic space heaters. The major fuels used
by such sources are coal, oil and natural gas. These fOssil
fuels currently supply nearly 95 percent of the United States
energy requirements. Other fuels cons~med are wood, lignite,
liquified petroleum gases, coke, refinery gas, and other by-
product gases, but the usage of these fuels is relatively
minor compared with coal, oil,and natural gas.
, The gross energy consumption by major fuel types
and consuming sectors are shown in Table 111-2.
In the review of the literature, it was found that
use of wood for fuel, although relatively minor from a total
energy consideration, does contribute significant amounts of
hydrocarbons to the atmosphere. Thus, estimates of wood
fuel consumption and the resulting emissions were made.
Reported studies of flue gas and combustion pro-
duct analysis from actual combustion sources which have
considered organic emissions have generally been limited to
the measurement of "total hydrocarbon", generally by flame
ionization methods. In a few reported studies, aldehydes
and organic acids are measured separately by other methods.
Flame ionization methods theoretically detect any
volatile C-H bonded species and thus can measure low mol-
ecular weight aldehydes and acids. In practice,differences
in sampling methods, e.g. heated or unheated sample probes,
res u 1 tin s i.g n i f i can t va ria t l' 0 n i nth earn 0 u n t s 0 f , 1 e s s vol-
atile species being detected. '
Where emission factors for "total hydrocarbons"
and "aldehydes" or "organic acids" have been reported,
these have been combined to arrive at total hydrocarbon
emissions. Thus, our combined values may result in some
III-5
-------
Table 111-2 - Summary of Fuel Consumption - 1968 Stationary Sources (a)
Coal (b) Petroleum (c) Natural Gas
Source 1012BTU 106 tons 1012BTU 106 bbl 1012 BTU 109 ft3
Electric Utilities 7 , 1 30 296.8 1 , 181 214.4 3,245 3,144
Industrial 5,616 191.6 4,474 812.0 9,258 8,971
Household and
...... Commercial 586 19.9 6,581 1,194.4 6,451 6,251
......
...... Miscellaneous and
I
m Unaccounted 1 O. 1 295 53.5
(a)
(b)
(c)
data from U.S. Bureau of Mines and Federal Power Commission.
includes bituminous, lignite and anthracite.
includes liquefied refinery gases and natural gas liquids,
values converted to equivalent bbl basis.
-------
indeterminate amount of double counting. This double count-
ing is felt to be minor in view of the overall uncertainty
of the reported emission factors.
1 .
Coal Combustion
Coal, more particularly bituminous coal, is our
most abundant fossil fuel. However, localization of natural
deposits coupled with transportation costs has limited its
significant usage to about two thirds of the nation, predom-
inantly in the central and eastern sections.
Emissions from coal combustion have been reviewed
and summarized by several authors. Smith (1966) covered the
available literature through about 1965. More recent sum-
maries have been presented by Duprey (1968) and McGraw and
Duprey (1971). These reviewers evaluated published data and
arrived at representative emission factors for typical coal
burning installations.
For large coal-burning power plants, the most com-
plete study of emissions is that reported by Cuffe et a1
(1964, 1965), in which several types of full-scale install-
ations were tested under both normal and partial load con-
ditions. Their measured emissions of total gaseous hydro-
carbons and total aldehydes are in reasonable agreement with
estimates obtained by using the emiss~on factors of McGraw
and Duprey (1971). However, in tests' on one boiler, Cuffe
et a1 measured total organic acid emissions, both before
and after fly ash collection (electrostatic). The average
value for five tests, after fly ash collection, was 224 ppm
by volume (dry basis), as acetic acid. Converting this value
to 1b/ton of coal gives 11.8 lb/ton. This value for organic
acids appears to be unusually high and, for this study, is
considered only insofar as it emphasizes the need for further
analytical studies.
For the estimate of emissions made in this present
study, the emission factors for coal combustion given by
McGraw and Duprey (1971) were felt to be a reasonable con-
census of the literature data. These emission factors are
given in Table 11-3 below.
111-7
-------
Table 111-3 - Uncontrolled Emission Factors for Bituminous
Coal (lb/ton of coal burned)
~106 BTU/hr heat input~
Greater than 100 (utility
and large industrial boil-
ers)
10 to 100 (large commer-
cial and general industrial
boilers)
Hydrocarbons
--(as CH41--
Aldehydes
Jas HCHOt
0.3
0.005
1
0.005
.
Less than 10 (commercial
and domestic furnaces)
Stoker
Hand fired
3
20
0.005
0.005
For domestic and commercial usage, the decreasing
usage of hand fired furnaces led us to choose the factor for
stoker fired units as being more representative of total
emissions from this source category.
Coal consumption data by various sectors of the
economy are given by the Bureau of Mines (1969) and for
steam-electric utilities by the Federal Power Commission
(1968). For the calculation of emissio~from coal burning
utilities, we chose to use the data from the Federal Power
Commission. These data cover some 528 steam-electric plants
which comprise 92 percent of total capacity and 96 percent
of the power generation in the contiguous United States.
The distribution of coal consumption by industrial
and domestic and commercial users is given for 1968 by the
United States Bureau of Mines (1969).
These data and the emission factors of McGraw and
Duprey (1971) were used to derive total organic emissions
by state and consuming sector. These detailed breakdowns
of coal consumption and emissions are given in Appendix H.
111-8
-------
2.. Natural Gas Combustion
Natural gas has long been considered a clean-burn-
ing, or relatively non-polluting, fuel. However, much of
this reasoning has been based on the absence of the particu-
late and SOx problems associated with coal and fuel oil and
has not taken into consideration the unburned or partially
oxidized organic emissions from gas combustion.
Only a few studies have been made of trace organic
emissions from natural gas combustion. McGraw and Duprey
(1971) reviewed available data and reported emission factors
for hydrocarbons (as CH4)' aldehydes, and organics. The
factors given by McGraw for hydrocarbons appear to be unusually
high when compared with emission factors reported" for other
fuels.
Chass (1960) reported the results of extensive source
testing of both gas and 011 burning sources in Los Angeles
County. Gaseous hydrocarbon emissions were not reported for
either gas or oil combustion because the concentrations
"were below the range of the analytical technique". However,
Chass does not describe the analytical technique used.
Magill and Benolie1 (1952) summarized in"one brief
table the emissions from all types of oil and gas burning
equipment. Converting their emission: factor for organic
emissions from gas combustion from 1b/ton to 1b/MMcf gives
an emission factor of about 75 1b/MMcf. Since this number
is a lumped factor for all types of sources and since Smith
(1962) rejected their similar number for fuel oil emissions,
for this study these particular data were discounted.
For the estimates of emissions from natural gas
combustion in this study, we chose to use the emission factors
for total hydrocarbons used by the Bay Area Air Pollution
Control District (1969) and the factors for aldehyde emis-
sions given by McGraw and Duprey (1971)0 These selected
emission factors are shown in Table 111-4.
Table 111-4 - Emission Factors for Natural Gas Combustion
(1b/MMcf burned)
Source
Total Hydrocarbons
Utility Boilers
Industrial Process Boilers
Domestic and Commercial
Heating Units
4
7
AldehYdes
3
3
1
10
III-9
-------
The consumption of natural gas by various sectors
of the economy was taken from U.S. Bureau of Mines data
(1969)0 Breakdowns of natural gas consumption and hydro-
carbon emissions by state and consuming sector are given
in Appendix Ho
3.
Fuel Oil Combustion
The various petroleum fractions used for fuel com-
bustion are generally described by the NBS Commercial Stand-
ards for Fuel Oils (Johnson and Auth, 1951). Grades 1 and 2
are distillate oils of sufficient volatility to allow usage
in general purpose burners not equipped with preheaters.
Grades 4 through 6 are residual oils used for industrial and
utility burners equipped with preheating facilities. Ad-
ditionally, some kerosene is used for heating purposes.
The emissions from fuel oil combustion have been
reviewed by Smith (1962), Duprey (1968), and McGraw and,
Duprey (1971)0 The emission factors selected for this
study were those of McGraw and Duprey and are shown in
Table III-50
Data on the consumption of fuel oils and kerosene
were obtained from the U.S. Bureau of Mines (1970)0 Emis-
sions from the usage of kerosene for space heating were not
broken down by state, but rather a total nationwide estimate
was made, using the emission factors for domestic distillate
fuel oil combustion.
Table 111-5 - Emission Factors for Fuel Oil Combustion
(lb/1000 gallons burned)
Source Total Hydrocarbons Aldehydes
Utility Boilers 5 1
Industrial and
Commercial Boilers
Residual - 3 1
Distillate - 3 2
Domes ti c 3 2
111-10
-------
4.
Wood Combustion
Wood is no longer a major fuel for industrial heat
or power generation, however, in those industries where con-
siderable quantities of wood wastes are generated, it is used
as fuel to' a limited extent. According to the Environmental
Engineering (1970) report on the wood pulping industry, the
consumption of wood wastes, primarily bark and chips, for
fuel amounted to about 15 million tons for 1968, at an esti-
mated fuel value of 4500 BTU/lb. Industrial usage of wood
for fuel by the furniture, plywood, veneer ~nd other lumber
industries was estimated by MSAR to be about 2 million tons.
The total industrial usage of wood and wood wastes
for fuel was estimated at 17 million tons. Based on the
emissions reported for wood and bark combustion in boilers of
2 lb/ton (McGraw, 1970), total hydrocarbon emissions from
this source category total 34 million pounds, or 17 x 10Y.
tons.
,
The usage of wood as domestic fuel is primarily
limited to rural areas where other fuels are not readily
available or are too expensive. The remaining domestic
usage may be considered as recreational or esthetic usage
of wood in fireplaces.
An estimated breakdown of wood fuel usage was given
by Schurr and Netschert (1960), citing data from Stanford
Research Institute. These data, shown in Table 111-6, give
actual consumption in 1950 and an extrapolated forecast
for 1975.
Table 111-6 - Fuel Wood Consumption (in units of 106 cords)
: heat,rura1 heatgrura1 heat;
Year farm non-farm urban Fireplace Other Total
1950 20.6 13.0 5.8 14.0 1.6 55.0
1975 2.3 3.7 2.4 17.0 0.2 25.6
More recent wood fuel consumption data are reported
in Statistical Abstract of the United States, 1969 (U.S.
Bureau of the Census, 1969). This latter source gives 1962
III-ll
-------
consumption of wood for fuel as 1.124 billion cubic feet.
At an average density of 40 1b/ft3; this is equivalent to
22.5 million tons, or about 18 million cords. Thus, it
can be seen that the Stanford forecast of 25.6 million
cords for 1975 is considerably too high. We estimate cur-
rent total domestic consumption of wood as fuel as being
about 15 million tons/yr, primarily in fireplaces.
Assuming the hydrocarbon emission factor for
fireplace and domestic fuel usage of wodd to be. similar
to that for incineration of wood in conical burners, the
average emission factor for such use is 11 1b/ton (McGraw
and Duprey [1971]). Based on this emission factor, total.
emissions amount to about 82.5 x 103 tons from domestic
usage.6f wood for fuel.
5.
Polycyclic Aromatic Hydrocarbon Emissions
Studies conducted by the U.S. Public Health Ser-
vice, as reported by Hangebrauck et a1 (1967), have shown
fuel combustion sources to be significant contributors of
polycyclic aromatic hydrocarbon. (PAH) emissions to the at-
mosphere. These PAH emissions are almost entirely assoc-
iated with the particulate emissions from combustion sources
and are directly related to the combustion efficiency of
the combustion source.
From measurements of total benzene soluble or-
ganic matter and chromatographic analyses for specific
compounds, particularly benzo-a-pyrene (BaP), Hangebrauck
derived estimates of emission factors for BaP from various
heat generation sources. Based on these emission factors
and Bureau of Mines data (1969) on fuel usage, current
estimates of BaP emissions were derived, as shown in Table
III-7.
For wood fuel usage, Hangebrauck's data on BaP
emissions from single-chamber incineration of cardboard
and wood packing crates were used to derive the average
emission factor for wood combustion of 106 mg/ton for use
with our estimated wood consumption.
6.
Summary of Emissions
The data sources and emission factors described
in the previous sections were used to derive a state-by-
state br~akdown of hydrocarbon and aldehyde emissions from
fuel combustion. This breakdown is given in Table 111-8.
III-12
-------
Table 111-7 - Estimated BaP Emissions from Fuel Combustion
1968 Emission Annual
Usage Factor Emissions
Fuel Burned (1012 BTU) (}.Ig/106 BTU) -( tons)
Coal
Utilities 7 ~074 90 0.7
Industrial 5,536 2,700 16.5
*Commercial 221 5,000 1.0
Residential
Stoker 98 44,000 5.0
Handfired 128 1 ,400,000 197
Oil 12,531 200 2.8
Gas 18,954 100 2. 1
Wood 32 x 106 tons 106 mg/ton 3.8
*U.S. Bureau of Mines total retail deliverIes for 1968
distributed among commercial and residential usage ac-
cording to Hangebrauck's (1967) relative distribution.
III-13
-------
Table 111-8 - Regional and State Breakdown of Fuel Combustion
Emissions for 1968 (tons/year)
Coal Fuel on Natural Gas Totals
Total Total Total Total
ReQion. and State Hydrocarbons Aldehydes Hydrocarbons Aldehydes Hydrocarbons Aldehydes Hydrocarbons Aldehydes
New England 1,357 585 8 29 3,868 1,776
Maine 242
New Hampshire 807 388
Vermont 379 229
Rhode Island 1,050 466 25 79
Connecticut 562 7 4,076 1,510 72 212 4,710 1,729
Mass achusetts 750 8 9,265 3,767 136 543 10,151 4,318
Middle Atlantic
New York 5,567 50 15,433 6,526 739 2,496 21,739 9,'172
New Jersey 1,205 15 8,884 3,544 371) 981 10,459 4,540
Pennsylvani a 9,447. 88 6,854 2,912 1,295 2,346 17,596 5,346
East North Central
Ohio 14,960 116 1,558 895 1 ,545 3,587 18,'163 4,598
Indiana 7,862 73 1,860 1,036 923 1,389 10,645 2,498
Il11nois 13,318 99 3,333 1,656 1,584 3,432 18,235 5,187
fl1 ch i gan 9,842 75 2,255 1,327 1,096 2,540 13,193 3,942
Wisconsin 6,347 35 1,468 919 512 849 8,327 1,803
West north Central
Minnesota 2,387 15 1,363 770 477 980 4,227 1,765
Iowa 1,709 13 517 324 466 870 2,692 1,207
Missouri 2,303 24 652 355 486 1,270 3,441 1,649
tlorth Dakota 1,049 9 271 166 17 75 1,534 467
South Dakota 170 103 27 114
Nebraska 316 3 171 97 276 594 1,760 1,843
Kans as 172 71 825 1,078
South Atlantic
Delal'lare 1,982 25 571 247 45 62 5,889 2,103
Maryland 1,801 816 186 604
Washington. D.C. 301 3 1,003 346
Viroinia 4,631 34 2,005 790 183 405 6,819 1,229
West Vi rginia 5,344 48 200 86 345 509 5,889 643
North Carolina 4,356 48 1,265 670 234 318 5,855 1,036
South Carolina 1,266 9 631 280 307 282 2,194 571
Georgia 2,335 31 776 315 580 831 8,756 2,773
Florida 4,483 1,106 582 490
East South Central
Kentucky 3,992 40 220 122 238 637 4,450 799
Tennessee 4,228 41 208 121 461 604 4,897 766
Alabama 3,298 42 189 66 667 709 4,878 1,414
fl1ssissippi 128 65 596 532
West South Central
Arkansas 78 59 20 678 790 13,786 10,075
louisi ana 225 123 3,677 2,406
Oklahoma 94 49 769 988
Texas 631 301 7,575 5,398
Mountain
Colorado 526 8 116 71 298 397 1,074 614
Idaho 1,136 9 221 102 411 795 1,768 906
Montana 635 1 261 164 92 101 1,258 553
Nevada + Arizona 67 41 67 97
New Mexico 14 35 21 348 463 397 484
Utah 430 2 550 197 220 330 1,200 529
Wyomi ng 476 6 271 148 134 174 881 328
Paci fi c
Alaska 374 2 279 150 12 39 665 191
California 24 3,563 1,025 3,608 5,365 7,195 6, 3CJO
Oregon 368 859 456 193 193 3,110 1,726
\~ashington 1,342 724 348 352
Sub total, United States 113,660 980 84,11 0 36,353 33,831 47,537 231,601 84,870
Wood Combustion, total United States 99,500
Undistributed Kerosene Us age, total United States 5,432 2.1..li
Total United States 336,533 88,145
II1-14
-------
More detailed treatment of the individual consuming sectors,
by region and state, is presented in Appendix H.
c.
Solid Waste Combustion
1 .
Introduction
The dimensions and nature of the problems associ-
ated with solid waste disposal have attained national recog-
nition during the past ten years, with many articles and
reports having been published in an effort to define these
problem areas. A dearth of quantitative data relegates many
of these pu~lications to general descriptions which contri-
bute little toward our understanding of the total situation.
Many of those studies conducted in sufficient de-
tail to arrive at reasonably quantitative estimates of waste
generation and disposal have been directed toward urban or
municipal areas, with little attention being paid to non-
urban problems. This limitation has been necessi~ated for
two principal reasons. One, the rapidly accelerating nature
of urban refuse generation and its resultant effect on the
major portion of the nation's population have combined to
direct the highest priority to urban needs. Two, it is only
in large urban areas that sufficient records are maintained
to allow even semi-quantitative estimates of generation and
disposal.
The widely dispersed and varied nature of the non-
urban waste sources, such as agriculture, mining, and rural
industrial operations, has prevented descriptions and measure-
ment in any terms except crude estimates or extrapolations
from limited data.
The most recent and well-documented study of waste
generation and disposal on a national scale is the National
Solid Wastes Survey. The survey results, as summarized by
Black et al (1968), when combined with conservative estimates
of waste generated by other sectors lead to a total national
waste generation estimate of over 3.5 billion tons per year.
Other, perhaps less conservative, estimates of
wastes from mineral and agricultural sources (for example,
Business Week, 1970) would place our total national waste
generation in the range of from 4 to 5 billion tons per
year. An approximate breakdown of total waste generation
is given in Table 111-9.
111-15
-------
For this study, development of estimates of hydro-
carbon emissions from disposal of solid wastes required (1)
esttmation of the fractions of waste burned by the various
combustion modes, i.e., municipal, industrial and commercial
and domestic incineration and open burning, and (2) estimation
of hydrocarbon emission factors most representative for these
combustion modes.
Table 111-9 - Approximate Breakdown of Total Solid Wastes
Generation (MSAR estimate)
Source Category
Municipal, commercial and domestic
was tes
Wastes Generartled
109 tons/yr
, -.
0.3
Agricultural refuse (crop & field
residues)
0,6
1.5-2.0
Animal wastes
Industrial wastes
0.6
1.0-1.5
Mineral wastes
Total
4,0-5.0
',: : 2,:,'-; ~Sol:td1'\.'laste Sources
An assessment of hydrocarbon emissions from solid
waste combustion req~ires the development of ~ system codify-
ing the amounts of disposal methods of the solid wastes pro-
duced by various source categories.
Four categories were chosen for the initial subdiv-
ision in this systematized profile. These included: a)
mining or mineral production; b) agricultural operations; c)
industrial and commercial operations; d) municipal and dom-
estic sources.
a.
Mining Wastes
Solid wastes generated during mining operations in-
clude overburden, rejected material from cleaning and refining,
111-16
-------
and construction wastes. Although a large amount of waste
is thus generated, much of it is innocuous in terms of air
pollution problems. Notable exceptions are the open burn-
ing of construction and land clearance wastes and the com-
bustion of coal refuse piles. Estimates of amounts of con-
struction and clearance wastes burned were assumed to be
lumped with the estimates of other similar wastes considered
in the industrial and commercial wastes category.
Coal refuse burnin~ has been reviewed by Sussman
and Mulhern (1964) and Hall (1962). The main source of
combustible refuse from coal mining derives from mechanical
cleaning operations. u.s. Bureau of Mines (1968) statistics
show that about 80 percent of total bituminous coal pro-
duction was cleaned, generating 97 million tons of refuse,
or about 18 percent of the total production.
The nationwide Inventorf of Air Pollutant Emissions,
1968 (U.S. Department of Health, ducation and Welfare, 1970
estimated that coal refuse banks totalling 19 billion cubic
feet were burning in 1964. GCA Technology (1970) cites an
unpublished Bureau of Mines study which reported the total
number of burning coal refuse banks in 1969 at 292, with a
total volume of 7.3 billion cubic feet and containing an
estimated equivalent of 18 million tons of coal. Sussman
(1964) cites a Bureau of Mines estimate of 488 burning re-
fuse banks. These figures would indicate significant pro-
gress in efforts to extinguish and prevent coal refuse com-
bustion.
The highly variable nature of this combustion source
makes any estimate of emissions highly questionable. Based
on the GCA technology reported value of the equivalent of
18 million tons of coal burning in refuse banks and the emis-
sion factor for hand fired domestic coal furnaces of 20
1b/ton (McGraw & Duprey, 1971), the total hydrocarbon emis-
sions are estimated at about 0.2 million tons/yr.
b .
Agricultural Wastes
Solid wastes generated by agricultural sources in-
clu~e animal wastes, field crop residues, wastes from pro-
cessing raw agricultural products, and land clearing debris.
Total estimates of agricultural wastes vary widely. Environ-
mental Science and Technology (1970) reported an estimated
2 billion tons/year of animal wastes alone. Black et al
(1968) reported an estimated 550 million tons of agricul-
tural waste and crop residue and approximately 1.5 billion
111-17
-------
tons of animal wastes. Business Week (1970) gave a total
of 2.3 billion tons/year of agricultural wastes, with no
breakdown given.
For our study, the amount of waste burned was of
primary interest. Thus, animal wastes were ignored, since
these are generally handled by field spreading and landfill
disposal methods rather than by burning.
The major emissions of organic air pollutants re-
sult from open burning of field residues and vegetation for
weed control and land clearance. Estimates of the total
extent of such burning were derived by extrapolation of
data reported for the San Francisco Bay Area by Johnson and
James (1970). Thus, field and crop residues burned would
total nearly 250 million tons/year on a national basis. Ex-
trapolated, estimates of wastes produced by land clearance
would contribute an additional 35 million tons/year.
The agricultural waste estimates derived above
are similar to totals reported in other studies. The
Nationwide Inventory of Air Pollutant Emissions, 1968,
gave an estlmated total of nearly 280 million tons of
crop residues, brush, weeds and other vegetation burned
annually. GCA Technology (1970) gave a breakdown of open
. burning in which the agricultural wastes burned annually
totalled 275 million tons.
Prescribed burning of timber and logging debris
is widely practiced,in several areas. For example, Bovee
et al (1970) report that 200,000 acres of logging slash
are burned annually in western Oregon and Washington. U.S.
Forest Service Statistics for 1966, reported in the Nation-
wide Inventory of Air Pollutant Emissions 1968, ind~cated
that controlled burning of timber lands cov~red some 3.5
million acres and represented about 76 million tons of com-
bustible material. More recent estimates given by Wadleigh
(1968) indicate a lower figure of 52 million tons burned by
prescribed burning of logging debris and vegetation.
Forest wildfires contribute significant amounts of
organic emissions annually. The combination of heterogen-
eous, wet fuel and insufficient oxygen combine to release
total organic emissions which can ex~eed those from other
open burning sources.
Forest Service statistics reported in the Statistical
Abstract of the United States, 1969 and in the Annual Fire Re-
port for the National Fore~ts, 1968 show that, for the period
111-18
-------
from 1966 through 1968, an average of 4.5 million acres of
forest land is burned annually by uncontrolled wildfires.
Combining this acreage with the factor of 32 tons of com-
bustible material per acre derived from the Nationwide In-
ventory (USDHEW, 1970) yields an estimate of 144 million
tons of vegetation consumed by wildfires. Consideration
of annual variation in acreage burned indicates that this
latter figure may range from about 120 to 150 million tons
per year.
, In summary, agricultural wastes burned annually
are estimated as: crop and field residues, 250 million
tons; land clearance, 35 million tons; prescribed forest
burning'~' 50 million tons. Forest wildfires, treated sep-
arately, result in an estimated annual burning of from 120
to 150 million tons of vegetation and logging residues.
c.
Industrial and Commercial Wastes
Estimates of industrial and commercial waste
generation vary widely and h~ve been found to be,:in many
instances, unrealistically low. Again, this is primarily
due to surveys and studies directed toward municipal or
urban generation and collection. Our estimates of waste
generation by the various sectors of industrial and com-
mercial operations are based on an analysis of the waste
generation factors for various industrial categories re-
ported by Combustion Engineering (1969), GCA Tech~ology
(1970), and the University of California (1970), supple-
mented by other literature data where applicable. '
. To arrive at meaningful estimates, we have
chosen to separate the industrial and commercial oper-
ations into five groups, primarily on the basis of their
differ'ent waste generation characteristics. These groups
are: ,major manufacturing; miscellaneous manufacturing and
commercial; forestry related industries; demolition and
construction; and food processing and marketing.
, ~r Manufacturing - Waste generation factors
in pounds or tons per employee year for the major manu-
facturing industries have been reported by Combustion
Engineering (1969) and the University of California (1970).
These waste generation factors are based on total solid
wastes generated, both combustible and non-combustible.
Excluding those industries treated separately, the
employment statistics for nineteen SIC groups were combined
III-19
-------
with the waste generation factors to estimate the total solid
wastes generated. Significant differences in the waste gen-
eration factors reported resulted in estimates of total wastes
generated ranging from 48 to 96 million tons/year. For our
selected value, we have taken the average or about 72 million
tons/year.
The breakdown of major manufacturing categories and
their estimated total solid waste generation is given in .
Table 111-10.
Estimates of the disposition or disposal mode for
industrial wastes vary widely. This variation is due not
only to the widely differing types of wastes be,ng generated
and disposed, but to the lack of clear definition of terms
used by the various investigators and reporting sources. We
have assumed that the fraction of wastes being disposed by
burning, either by incineration or open burning, is composed
totally of combustible waste. This assumption is an over-
simplification which could result in our estimates being
somewhat high, since wastes burned often include variable
amounts of non-combustible material.
. Based on their survey of community collection prac-
tices, B~ack et al estimated that from 30 to 40 percent of
generated industrial waste is self disposed. Combustion En-
gineering (1969), from their survey of industry groups,re-
ported that about 50 percent is handled on-site by the gen-
erator, 25 percent disposed off-site by the generator, and
about 21 percent by contract disposal, with only 3 percent
going to municipal collectors. Much of the off-site disposal,
however, ends up in municipally owned disposal sites.
As to ultimate disposal mode, Combustion Engineer-
ing (1969) estimated that incineration and open burning com-
bined account for from 30 to 40 percent of the total waste
disposed. For this study, 25 percent open burned and 10
percent incinerated were selected as being more realistic
in view of our assumption that wastes burned are totally
combustible.
Miscellaneous Manufacturing and Commercial - Using
an average waste generation factor for representative in-
dustries of 3.25 tons per employee year (Combustion Engineer-
ing, 1969), the total solid waste generated by miscellaneous,
commercial, and government sources was estimated to be about
150 million tons/year.
111-20
-------
Table 11-10 - Major Manfuacturing Solid Waste Generation
Was te Total Solid
1968 Generation Factors Wastes Generated
Employment (tons/employee/yr) (106tons/yr)
Industry Group .(1,000) ~L ~ ~ (b)
Ordnance 342 1. 45 0.66 0.50 0.23
Furniture 474 wood 5.75
meta 1 8.55 20. 16 2.37 9.56
Stone, clay &
glass 638 2.65 18.11 1.69. 11. 55
Primary metal 1 ,301 1. 50 6~73 1. 95 8.76
Fabricated
metal 1 ,389 1. 45 6.73 2.01 9.35
Machinery 1,958 1. 45 4.18 2.84 8.18
Electrical
Equipment 1,963 1. 05 2.98 2.06 5.85
Trahsportation
Equipment 2,026 1. 10 3.39 2.23 6.87
Instruments 451 2.65 2.52 1. 20 1. 14
Tobacco 86 5.40 2.49 0.46 0.21
Textile Mills 985 1. 25 0.53 1. 23 0.52
Apparel 1,417 0.30 0.53 0.43 0.75
Paper & Allied 698 8.75 12.54 6. 11 8.75
Pri nti ng. &
Publishing 1 ,063 8.25 13.20 8.77 14.03
Chemi ca 1 s (ex-
cept paint) 1 ,032 6.30 8.21 6.50 8.47
Rubber &
Plastics 558 5.95 1. 55 3.32 0.86
Leather 357 9.70 2.49 3.46 0.89
Petroleum Re-
fining 151 6. 30* 0.95
Paint 70 2.65 O. 19
Total 16,959 48.27 95.97
*Estimated as same as chemicals
(a) From Combustion Engineering (1969)
(b) From University of California (1970)
111-21
-------
Ultimate disposition was assumed to be similar to
that of the major manufacturing industries, i.e. 10 percent
incinerated and 25 percent open burned.
Forestry-Related Industries ~ Th~ lumbering and
forestry-related industries - saw mills, planing mills, etc.
generate exceptionally large amounts of solid wastes. Com-
bustion Engineering (1969) reported a waste generation factor
of 277 thousand lb per employee year. For the 1968 national
employment of 602,000 persons, this results fn estimated
generation of 83 million tons.
Sawmill wastes disposal was estimated by Combustion
Engineering to be over 75 percent burned in conical or tepee
burners. We have assumed ultimate disposal of all wastes
from forestry-related industries to be 65 percent conical
burning and 10 percent open burning.
Construction and Demolition Wastes - Contract con-
struction and demoli~on is another spetial category dis-
tinguished by high waste generation factors. Combustion En-
gineering gave a factor for construction waste generation
of 87 thousand lb/employee year. A similar generation factor
of 82.5 thousand lb/employee year was given by the University
of California. For demolition,'.the waste generation factor
given by Combustion Engineering was over 4.5 million lb/
employee year. Based on 1968 census data for total employ-
ment, the total waste generated by construction and demolition
was estimated to be about 140 million tons/year.
Ultimate disposition modes have not been reported,
but traditionally, open burning has been used for combustible
wastes. We assumed 25 percent disposal by open burning.
Food ~rocessing and M~rketing - Reported estimates
of food processing wastes have been limited to surveys of
canning and preserving, dairy products processing, and meat
packing. Waste generation factors for these industries
range from 3 to 5 tons per employee year. Combustion En-
gineering (1969) reported supermarkets as being especially
high in waste generation at nearly 18 tons/employee year.
A review of other literature sources indicated
that total waste generation should be considerably higher
than the 35 to 40 million tons/year derived from the above
generation factors. .
111-22
-------
Mercer (1963) cited typical waste production fig-
ures for fruit and vegetable processing in California. Solid
wastes generated ranged from 7 to over 15 percent of the
total tonnage of fruits and vegetables processed.
Shearon (1952), in discussing citrus fruit pro-
cessing, gave a figure of 60 percent wastage in juice prQ-
duction, Hickey (1966) gave amounts of "special wastes" in
the San Francisco Bay Area. About 10 percent of these
"special wastes" were derived from cannery operations.
If a 10 percent wastage factor is applied to all
fruit, vegetable, meat, fish, poultry and dairy products
processed ~er year, the solid wastes generated could range
as high as 200 million tons per year. For this study, .we
took a conservative estimate of 100 million tons per year
for all wastes from food processing and marketing.
Ultimate disposition was assumed similar to that
for the manufacturing industries and commercia1 operations,
or 10 percent incinerated and 25 percent open burned.
d.
Municipal and Domestic Wastes
Urbanization and industrializ~tion have served to
increase both the generation of solid wastes and the diffi-
culties of disposal for most communities. With nearly 75
percent of the population in urban environments, municipal
wastes are a dominant factor in urban governmental problems.
Although they represent only a small fraction of
the total solid wastes, municipal wastes have been given
paramount attention in technical surveys and investigations
because of the social and economic problems associated with
urban waste disposal. This public and governmental concern
has encouraged and supported studies to determine the various
aspects of the problem and to provide quantitative data.
Niessen et al (1970), Kaiser (1967), and Rogus
(1963) were used as general references for municipal wastes
generation and disposal practices. Combustion Engineering
(1969) and the National Emissions Standards Study (1970)
provided data for the location and capacity of municipal
incineration. The National Solid Wastes Survet provided re-
cent estimates of waste generation and disposition derived
from community collection practices.
1II-23
-------
Urb~n or municipal waste generation factors gen-
erally include not only household and municipal wastes, but
significant amounts of industrial and commercial wastes.
Thus, in our treatment, some unavoidable overlap or double
counting may have occurred, even though attempts were made
to minimize this. By comparison, however, uncertaint~es in
tfie tota 1 es trirpa tes ~f gener-a ti ng segmen ts obvi a tes the
significance of any double counting that may have occurred.
Waste generation factors given by Niessen (1970)
and urban population were used to derive state-by-state
estimates of municipal solid waste generation. These gen-
eration factors are heavily weighted by data on wastes col-
lected for municipal disposal and include unstated amounts
of commercial and industrial wastes as well as household or
domestic wastes. As given in Appendix I, this breakdown
also provides estimates of municipal incineration and muni-
cipal open burning. These latter estimates were derived
by assuming municipal incineration at 50 percent of reported
capacities and open burning of urban wastes at 20 to 25
percent, except in states that prohibit open burning. Where
specific data were made available as a result of our question-
naire survey of pollution control agencies, exceptions to
the above methods were made.
Totalling the state-by-state estimates indicates
that U.S. municipal waste generation amounts to about 170
million tons/year. .
The National Solid Wastes Survey (Black et a1,
1968) indicated that collected wastes ave~aged 5.32 lb/
person day on a national basis, as shown in Table 111-11.
These data reflect only the amount collected. Black et a1
estimated the total generation of both collected and uncol-
lected wastes at 10 1b/person day, with industrial wastes
accounting for 3 lb/person day and household, commercial
and municipal wastes (street & alley, etc.) accounting for
7 1b/person day. By excluding industrial and construction
demolition wastes from these per capita estimates, we ar-
rived at the estimates of urban and rural waste generation
and disposal shown in Table 111-12. Urban population was
assumed to be 73 percent of total population.
1II-24
-------
Table 111-11 - Average Solid Waste Collected (pounds per
person per day)
Waste Urban Rural National
Household 1. 26 0.72 1. 14
Commercial 0.46 O. 11 0.38
Combined* 2.63 2.60 2.63
Industrial 0.65 0.37 0.59
Demolition, Construction 0.23 0.02 0.18
Street & All ey O. 11 0.03 0.09
Miscellaneous 0.38 0.08 o. 31
Totals 5.72 3.93 5.32
*Combined household and commercial wastes from collections
that could not be characterized separately.
Table 111-12 - Household & Municipal Waste Generation and
Disposal (106 tons/year)
Source Generated In c i n era te d Open Burned
Urban collected 129 13 32
Urban uncollected 34 3 17
Rural collected 35 4 18
Rural uncollected 30 15
Totals 228 20 82
Municipal incineration of urban collected wastes
was assumed to amount to 50 percent of stated capacity for
all incineration capacity listed by Niessen (1970) and the
National Emission Standard Study. Ten percent of uncollected
urban wastes were assumed to be incinerated in domestic in-
cinerators. Ten percent of rural collected wastes were as-
sumed to be incinerated municipally. Open burning was assumed
III-25
-------
to be the disposal method for 25 percent of urb~n collected
wastes and 50 percent of all other urban and rural wastes,
both collected and uncollected.
3.
Wast~"Combustion'Emis~i6ns'
In order to estimate the organic air pollutant
emissions from the combustion of solid wastes as categorized
above, emission factors were selected as being representative
of the various combustion modes and types of wastes burned.
A general review of published studies of waste com-
bustion emissions showed an extremely wide range of experi-
mental results, depending on the analytical techniques em-
ployed and combustion parameters. Wide variations are not
unexpected since reproducible and representative combustion
conditions are difficult to achieve with such heterogeneous
mixtures as are represented by the various waste mixtures.
Even under the best conditions,such as those with a modern
multiple chamber incinerator, variations in waste composition,
moisture content, and combustion air can significantly affect
the type, amount, and composition of organic emissions.
For the estimates obtained in this study, the re-
ported emissions and emission factors were analyzed and com-
bined to give reasonably representative averages of total
organic gas emissions. Thus, our selected emission factors
may be somewhat higher than those reported on the basis of
only total hydrocarbon as determined by flame ionization.
Several investigators, namely Feldstein (1963), Darley (1966),
Gerstle and Kemnitz (1967), and Boubel (1968), reported
analyses for emission components such as organic acids, al-
dehydes or total oxygenates, and unsaturated hydrocarbons,
as well as total hydrocarbons. Special weight was given
these more detailed analyses in deriving our selected factors
for estimating total organic emissions.
Unfortunately, the failure of many investigators
to give sufficient detail about analytical techniques and
combustion conditions does not allow valid assessment of
their results. We attempted to give greater weight to
those studies which were obtained under field conditions.
Even so, an uncertainty of at least a factor of two exists
in the selected values.
The selected emission factors, in terms of total
organic emissions, are shown in Table 111-13. In Appendix
111-26
-------
I, the reported data from which the selected emission factors
were derived are detailed.*
Table 111-13 - Estimated Emission Factors for Solid Waste
Combustion Sources
Combustion Mode
Total Organic Emissions
(lb/ton burned) ,
Incineration
Municipal
Industrial and Commercial
Conical Burners (wood)
Domestic
4
14
20
40
Open Burning
Municipal
Industrial and Commercial
Lumbering
Field & Crop
Land Clearance
Demolition & Construction
Coa 1 Refuse
45
30
30
20
40
30
20
Forest Wildfires
40
The highest emissions, and also the widest vari-
ation in reported values, result from domestic single chamber
incinerators and open burning sources. Again, only a few
investigators have determined anything other than gaseous
hydrocarbons, generally by flame ionization analysis. Feld-
stein (1963) in a reworking of Yocom and Hein (1956) data
on single chamber domestic incinerators reported the high-
est values found. These data, also used by BAAPCD in their
*To illustrate the selection of representative emission
factors for waste combustion, the factor of 4 lb/ton of
waste burned by m~nicipal incineration was derived from
full-scale incinerator test data reported by Niessen
(1970). . These data show a range of total hydrocarbon
emissions of from 0.9 to 6.3 1b/ton (average = 1.84f2.16)
and aldehydes of from 0.001 to 0.84 lb/ton. The factor
of 4 lb/ton was selected as an upper mid-range value from
these data.
111-27
f~~~
-------
1969 inventory, show total organic emissions to be about
250 1b/ton, with 24 1b/ton of organic acids and 25 1b/ton
of aldehydes. These data were derived from burning a
50/50 mixture of grass clippings and paper, a perhaps
unrealistic mixture for any except domestic backyard burn-
ing.
Gerst1e & Kemnitz (1967) reported results for
simulated open burning of municipal refuse, landscape re-
fuse, and auto components. For both municipal and land-
scapes refuse they report about 50 1b/ton of gaseous hydro-
carbons (as CH4) and about 15 1b/ton of organic acids (as
CH2COOH). Repor.ted aldehyde concentrations were low, from
0.005 to 0.01 1b/ton. Gerst1e and Kemnitz also reported
that limited chromatographic analyses indicated that the
gaseous hydrocarbons consisted of 30 to 40 percent unsat-
urated hydrocarbons.
For domestic incineration, which may range from
fairly well controlled installations with a primary air
supply to poorly controlled burning cans and baskets with
insufficient combustion air, we chose to use an intermediate
factor of 40 1b/ton for the total organic emissions. Of
the 40 1b/ton total, we estimated that combined aldehydes
and organic acids account for 10 to 15 1b/ton.
The wide variety of wastes burned by open burn-
ing and the variation in emission factors reported do not
justify attempting to estimate separate emission factors
for each source category; however, some general distinctions
may be made.
For open burning of municipal and domestic refuse,
we selected the values given by Gerst1e and Kemnitz as being
reasonable mid-range values. Thus, for t~is source category,
we chose a factor of 45 1b/ton for total organic emissions,
of which about~15 1b/ton are from combined acids and alde-
hydes.
For industrial wastes not otherwise specified, an
emission factor of 30 1b/ton for total organic gases was
assumed for all open burning.
For agricultural burning and forest wild fires,
where the combustible materials may range from dry grass
and crop stubble to green brush and trees, we have chosen
to use two different values. For field crop' and residues,
an emission factor of 20 1b/ton was se1eited for total or-
111-28
-------
ganic emissions. This value is heavily weighted by the mid-
range values reported by Darley (1966) and Soubel (1969) for
total gaseous hydrocarbons. .
For land clearance and forest fires, a higher
factor of 40 lb/ton for total organic emissions was selected.
This value was weighted by the high values reported by Darley
and Soubel for green vegetation.
4.
Summa ry
The estimated solid waste gener&tion and disposal
by combustion for the various source categories are summarized
in Table 111-14. Application of the selected emission factors
from Table 111-13 to the amounts burned results in the esti-
mated total hydrocarbon emissions shown in Table 111-15.
D.
PETROLEUM PRODUCTION, REFINING AND MARKETING
1.
General
The processing of crude oil into salable products
and the marketing of these products constitute a significant
potential source of hydrocarbon emissions. Some hydrocarbon
losses occur during the production of crude oil; however,
these losses are small compared with those from refining and
marketing.
The magnitude of the petroleum industry may be il~
lustrated by some statistics for 1968 taken from the 1971
International Petroleum Encrclopedia and the 1968 Minerals
Yearbook. Domestic product on of crude oil in 1968 was 9
million bbl daily from almost 554,000 wells. Total U.S. re-
fining capacity was approximately 13 million bbl/day with an
average refinery throughput averaging nearly 11 million bbl/
day. Total U.S. consumption of petroleum products amount to
nearly 14 million bbl/day, with motor gasoline alone account-
ing for nearly 5.9 million bbl/day.
2.
Industry Description
Input crude from the field is treated to remove
sediment.which can be removed by sedimentation or settling.
In some cases, crude oil and water are in the form of an
emulsion that must be treated to achieve separation. Heat
and chemicals are used to break the emulsion and free the
oil. Gas-oil separators are employed to strip the liquid
crude of gaseous hydrocarbons.
111-29
-------
Table 111-14 - Summary of Waste Generation and Combustion
Total Was te Total Waste Burned
Gene ra te d Incineration Open Burning
Source Category -'106 tons/yrt 1106 tons /yr t -'106 tons/yrt
Mining Coal Refuse 97 18
Agriculture 7550
Crop and Field Residues 250
Controlled Timber and
...... Land Clearance 85
......
...... Fore s t Wi 1 d Fires 120-150
I
w
0 Industrial
Major Manufacturing 72 7.2 18
Miscellaneous Manufacturing
and Commercial 150 15 37.5
Forestry and Lumbering 83 62.3
Construction and Demolition 140 35
Food Processing and Marketing 100 10 25
Municipal & Domestic
Urban Collections 129 1 3 32
Urban Uncollected 34 3 1 7
Rural Collections 35 4 18
Rural Uncollected 30 15
-------
Table 111-15 - Summary of Estimated Organic Emissions From
Solid Waste Combustion Sources
Source Category
Mining (Coal Refuse Only)
'Agri cul ture
Field & Crop Residues
Controlled Timber &
Land Clearance
Total
Forest Wildfires
Industrial
Major Manufacturing
Miscellaneous Manufacturing
and Commercial
Lumbering
Construction & Demolition
Food Processing & Marketing
Total
Municipal and Domestic
Urban Collected
Urban Uncollected
Rural Collected
Rural Uncollected
Total
Grand Total
111-31
Total Organic ~miss1on
--'106 tons/year)
0.2
2.5
107
4.2
2.4-3.0
0.3
0.7
0.7
0.5
0.5
2.7
0.7
0.4
0.4
0.3
108
110 3-11 .9
-------
A transportation system consisting of pipelines,
tankers, barges, tank cars,and tank trucks move the crude
petroleum to the refineries for processing. Refineries
received 75.8 percent of their crude oil supply by pipe-
line, 23.0 percent by wate~ and 1.2 percent by tank cars
and tank trucks in 1968.
Operations at a modern refinery are highly com-
plex, but in general consist of four major steps: sep-
aration, conversion, b1ending,and treating. These steps
are made up of a number of unit operations, which include
distillation, catalytic conversion;, polymerization, and
isomerization. The products resulting from these oper-
ations are blended and treated to give the desired end
products. Because of the variable market demand for partic-
ular fractions and products, a high degree of flexibility
in processing is required.
Included at the refinery plant are innumerable
equipment-technique combinations which supply the service
necessary to maintain the continuity of the total refining
process. Equipment-technique combinations pertinent to
this study include: condensers, cooling towers, boilers
and heaters, compressors, oil-water separators, air blow-
ing,: pipeline valves and flanges, and pipeline blinds~
Although all of the above primary and secondary
processes and equipment contribute to total hydrocarbon
losses, the most significant refinery losses result f"om
the necessary use of vast storage facilities. As a rule,
refineries maintain about two barrels of storage capacity
for each barrel of inventory. The National Petroleum
Council (1963) suggests that this surplus storage factor
(2.0 in 1963) is the minimum amount of storage required
to maintain necessary flexibility in the refining process.
Evaporative loss from storage tanks is the greatest con-
tributor to hydrocarbon emissions.
The role of the refinery in the marketing of its
products is dependent upon the definition. 'Distinct sep-
aration of refining operations and marketing cannot be
readily made because of multiple usage of facilities.
The modern refinery serves as the first link in
the marketing and distribution chain. The bulk of all
refinery products are turned into pipelines, tankers,or
barges and transported to large holding facilities or
I II - 32
-------
bulk terminals. From these, gasoline and other products are
routed to the consumer either directly or through smaller
bulk stations and service stations. These bulk terminals
and stations may be refinery owned and operated or may be in-
dependent of the refinery. Storage and transfer operations
at each point along the distribution chain are potential
emission sources.
In order to simplify the estimation of hydrocarbon
emissions from the petroleum industry, generalizations and
assumptions 'were made of the flow of crude oil and refined
products through a hypothetical "typical" production, refin-
ing,and marketing operation. Figure II1-l shows schematic-
ally the distribution path and the major emission points
treated in subsequent portions of this section. By assuming
total industry crude and product flow through this hypothetical
path, total industry hydrocarbon emissions were derived.
Subsequent discussions treat each of the major emis-
sion points along the distribution path of Figure 111-1.
3.
Crude Petroleum Production
Approximately 70% of u.S. crude oil is produced in
the states of and offshore of Texas, Louisiana,and California.
Domestic production of crude in 1968 totalled 3,329,042,000
barrels from almost 554,000 wells, as reported by Bureau of
Mines (1968).
Data on crude petroleum production emissions are
sparse. Emission inventories of Los Angeles County (California
Air Resources Board, 1969) indicate hydrocarbon emissions of
60 tons/day from crude production, versus 50 tons/day from
refining and 110 tons/day from marketing. A breakdown of
crude production source emissions given in a pollution s'tudy
of Monterey and Santa Cruz Counties, California (Monterey-
Santa Cruz County Unified APCD, 1968) is shown in Table 111-
16. These losses resulted from an annual production of 18.5
million barrels of crude oil.
By assuming these crude production emissions to be
typical of total U.S. practice, extrapolated emissions for
the total U.S. crude production of 3.33 billion barrels
yields estimated annual crude production emissions of about
224,000 tons.
111-33
-------
Production
Losses
/to
I
Working and
Breathing Losses
l-
I
Working and
Breathing Losses
J...
I
Operating
Losses
..... .f'
-
-
-
I
W
.J:=o
All Distillate Products
Transfer
Losses
Trans fer
Losses
.....
I
Working and
Breathing Losses
.to
I
Working and
Breathing Losses
It..
I
Working and
Breathing Lbsses
A
I
4-
I
Servi-ce
Station
(motor gasoline,
a v i at ion gas 0;' - -
line & naphth
jet fuel
Consumer
Figure 11-1 - Petroleum Distribution Path
and Emission Points
Special naphtha, kerosene, kerosene-
type jet f.uel and distillate fuel 011
-------
4.
Crude Storage in Transportation System
Crude oil is supplied to refineries through a trans-
portation system which includes tank farms, bulk terminals,
and other storage points outside of the refinery. In order
to estimate storage emissions from this crude transportation
system, all of the crude delivered to the refinery is assumed
to have passed through available storage tankage facilities
enroute to the refinery.
Recommended methods for estimation of losses from
storage tanks have been given in a series of bulletins pre-
pared by the Evaporation Loss Committee of the American Pet-
roleum Institute. Most useful for estimating storage losses
are Bulletin 2513 (1959), Bulletin 2517 (1962), and Bulletin
2518 (1962). In order to use these methods of loss calculations,
values must be assigned for such factors as total storage
tankage, specific tank sizes and ranges, throughput rate, and
meteoro 1 ogi ca 1 condi-ti ons.
The National Petroleum Council (1963) reported
that storage capacity outside the r~finery in 1'960 was about
274 million parrels. Assuming storage capacity to be propor-
tional to refinery capacity, extrapolation of the 1960 stor-
age data to 1968 refinery capacity (U.S. Bureau of Mines,
1968) yields an estimate of crude storage capacity outside
of the refinery in 1968 of about 304 million barrels..
Estimated storage tank emissions are divided into
two categories; breathing losses and working losses. Breath-.
ing losses result from vapor expansion and contraction due
to atmospheric temperature changes and are influenced by the
vapor pressure of the stored liquid, tank coating reflectivity,
and ambient temperature changes. Working losses result from
escape of vapors during tank filling operations and are sig-
nificantly affected by the number offi11ing operations or
turnovers. The turnover rate is obtained from the ratio of
total annual throughput to tank capacity.
Storage tank emissions are also a function of tank
roof construction. Floating roof tanks are widely used to re-
duce emissions over those from cone or fixed roof tanks. For
this study, the percentage distribution of floating roof and
fixed roof tanks was that given by Duprey (1968) as being
typical of industry practice: 75 percent floating roof and
25 percent fixed roof.
II I - 35
-------
Table 111-16 - Estimated Crude Oil Production Emissions,
Monterey County, 1967 (Crude Production of
18.5 Million Barrels)
Point Source
Emissions~ tons/day
Hydrocarbons Ot er Organ1c Gases
Storage Tanks
Roi1ers and Heaters
O. 1
Gas
011
O. 1
0.5
Waste Water Separators
Pumps
0.2
1.9
Compressors
O. 1
0.2
Relief Valves
Pipeline Valves
0.3
II 1 - 36
-------
Rather than assuming all storage losses to occur
from a single, average tank size, we chose to distribute the
loss calculations over a range of tank sizes felt to be typ-
ical of industry practice. A Poisson distribution function
was thus used to generate the number of tanks of incremental
sizes over a range from 20..to 160 thousand barrel capacity.
For crude oil storage outside the refinery, the generating
function was induced to peak at an .assumed average tank size
of 80,000 barrel capacity.
The advantages of using a tank size distribution
are two-fold. First, it results in more realistic emis~ions
since emissions vary as a power function of the tank diameter
and hence the use of a single average tank size would not
necessarily yield average total emissions. Second, the range
of tank sizes allows for the application of a range in the
paint and roof factors used in the loss calculations. It
was assumed that the larger tanks were newer and more care-
fully maintained than the smaller tanks.
Meteorological conditions, temperature, and wind-
speed were taken from the API Bulletins and adjusted to re-
flect average crude storage conditions.
The turnover rate (annual throughput/tank capacity)
for estimation of the working losses was estimated at 13,
from the annual throughput of 3.8 billion barrels and the
estimated storage capacity outside the refinery of 304 mil-
lion barrels.
Tank outage, or the amount of vapor space above
the stored liquid is difficult to estimate. API Bulletin
2518 gives outage in feet for tank test data used in develop-
ing loss correlation methods but does not give total tank
height. We assumed a typical outage of 50 percent of tank
height.
The Reid vapor pressure, used to estimate the
true vapor pressure of the liquid under storage conditions,
was estimated from data presented in API Bulletin 2518.
Crude oil Reid vapor pressures range from about 3 to over
10 1b; a 'va1ue of 7.5 1b was adopted as being typical.
Estimated hydrocarbon emissions from crude oil
storage in the transportation system are tabulated in
Table 111-17. Hydrocarbon emissions from this s9urce of
crude storage totalled 470,000 tons.
111-37
-------
Table
III-17
Crude
Losses
; n
the
Transportation
Sys tern
Tant Olstrlbuted( t) Roof &(,) Control(') Control (0) StandIng
Tant SI ze . I!umber(') T~~~~~~~~t< ,) Palnt(') Wort I ng BreathIng Storage
She DistrIbutIon Capacity of Paint Factor loss lOH Factor ltJss
Storage ConditIons (10' bb1) (:) (10' bb1) Tants (10' bbl/tant) Factor Factor (:) (10' bb1) (10' bbl) (:) (10 I bbl)
Crude 011 20 8 24.] 1,215 260 1.3 3.2 2S 78.9 91.1 75 17S.0
Total storage capacIty,
304 x 10' bbl 40 15 45.5 1,142 520 1.2 2.B 25 112. B 171.3 75 2]9.8
liquId Ter.1p, 65°F 60 20 60.7 1,011 780 1.2 2.8 25 146.6 .~;.S 75 297.2
...... Temp Change, 17°F 80 20 60.7 758 1,040 1.2 2. S 25 144.0 Z31.? 75 a9.0
......
...... Reid V.P., 7.5 lb 100 15 4S.5 455 1,300 1.2 2.2 25 102.4 170.6 75 157.7
I True V.P., 5.2 ps fa 120 10 ]0.4 253 1,560 1.1 2.2 25 66.4 11]. g 75 96.0
W
00 ~lnd Speed. 9 mph 140 .. 21.2 151 1.820 1.1 1.3 25 4].4 77.8 75 ]8.3
Turnovers, 13 160 15.2 95 2.080 1.0 1.0 25 27.3 52.3 75 19.6
SUBTOTALS 721. 8 1,135.7 1,272.6
Total Crude loss I:! Percent dIstrIbutIon x total storage capacity
x 10' bbl Total capacIty' specIfIc tant yolume
Volume - 3.130 (:! SpecIfIc tant yolume . turnoyers
PaInt factors from API BulletIn 251B
HeIght - 470 x 10' tons (at ]00 lb/bb 1) !:) PaInt and roof factors from API BulletIn 2517
Percentage of tants wIth fIxed or floatIng roof
-------
5.
Refinery Emissions
Hydrocarbon emissions from the refineries were esti-
mated by combining the activity of all refineries into a com-
posite or single process. This was followed by application of
suggested emission factors for each source point~ A number of
assumptions and judgments were made to fill existing data voids.
In some cases the accepted basis for a judgment was a single
data "point". Major hydrocarbon sources are discussed sep-
arately and are preceded by a discussion which outlines the
mechanism of the tabulation and the guidelines used. The re-
finery storage area is treated in greater detail because of
its importance.
Reported emission factors for refinery operation
are based primarily on a detailed study of refinery operations
conducted in the Los Angeles Area and reported in Public Health
Service Publication 763 (1960). These have been summarized
by Duprey (1968) and are shown in Table 111-18. For many of
the point sources, ranges of emission factors are given, re-
flecting both uncontrolled and controlled operations.
A highly questionable aspect in any estimation of
total emissions is the extent of control in application. Com-
parison of community survey data from areas other than Los
Angeles indicates a range of total refinery emissi9ns of from
less than 0.1 percent of crude throughput t9 nearly 0.3 per-
cent. These data are listed in Table 111-19. The variations
shown for total refinery emissions in these survey ~stimates
may not represent any meaningful geographical or chronological
trend but may simply reflect differences in estimating pro-
cedures.
a.
Refinery Storage Emissions
In this study estimates of hydrocarbon emissions
from storage areas at refineries are limited to those from
crude oil and gasoline storage. Losses of other distillates
during storage are detailed in the later section on Petroleum
Product Marketing (Section III'D~6).
Storage tank losses at the refineries were treated
in a similar fashion to that described previously for crude
oil storage losses in the transportation system. To use the
APlr~commendations for storage loss calculation, values were
assigned for such factors as total storage tankage, specific
tank sizes and ranges, throughput and turnover rates, and
meteorological conditions.
II 1-39
-------
Table 111-18 - Hydrocarbon Emis~iQ9)Factors for the
Petroleum Industryl
CRUDE STORAGE
Fixed Roof
Breathing Loss -
Working Loss -
Floaii.ng Roof
Breathing Loss -
0.3 lb/day/1000 gal storage capacity
8.0 lb/1000 gal throughput
30-160 lb/day/tank
DISTILLATE STORAGE
Fixed Roof (vapor pressure >1.5 psia)
Breathing Loss -
W 0 r kin g L o's s -
0.4 lb/day/1000 gal storage capacity
11 lb/1000.gal throughput
Floating Roof (vapor pressure >1.5 psia)
Standing Storage -
Fi~ed Roof (vapor pressure ~1.5 psia)
4.8 lb/day/1000 bbl storage capacity
Standing Storage -
1.7' lb/day/1000 bbT sto~~ge capacity(2)
Floating Roof (vapor pressure <1.5 psia)
Standing Storage -
BOILERS AND PROCESS
HEATERS -
FLUID CATALYTIC
CRACK UNITS -
MOVING-BED CATALYTIC
CRACK UNIT -
COMPRESSOR ENGINES -
BLOWDOWN SYSTEM -
1.6 lb/day/1000 bbl storage capacity(2)
140 lb/1000 bbl oil burned 0.03 lb/
1000 ft3 gas burned
220 lb/1000 bb1 fresh feed
87 1.b/1000 bbl fresh feed
1.2 lb/1000 ft3 gas burned
5-300 1b/1000 bbl refinery capacity
111-40
-------
Table 111-18 (continued)
PROCESS DRAINS -
VACUUM JETS -
COOLING TOWER -
MISCELLANEOUS LOSSES -
Pipeline Valves &
Flanges -
Vessel Re~ief Valves -
Pump Seals -
Compressor Seals -
Others, (Direct
Blowing, Sampl-
ing, etc.) -
(1)
(2)
8-210 lb/1000 bbl waste water
0-130 lb/1000 bbl vacuum distillation
6 lb/106 gal cooling water
lb/1000 bbl refinery capacity
28 lb/1000 bbl refinery capacity
11 lb/1000 bbl refinery capacity
17 lb/1000 bbl refinery capacity
5 lb/bbl refinery capacity
10 lb/1000 bbl refinery capacity
Duprey (1968)
Public Health Service Publication 763
III-41
-------
Table 111-19 - Refinery Emissions Reported by Community Surveys
No. of
Region Date Refineries
Seattle-Tacoma 1963(a) 4
St. Louis, Mo. 1967-1968(b) 4
Erie County, N.Y. 1967(C) 1
1
S an Fran ci sco 1969(d) 5
......
...... 1969(e)
...... Louisville, Ky. 1
I
~
N
TOTAL
Estimated
total Hydrocarbon %
Refi nery Emissions Weight
Capacity tons/yr Loss
201,800 9,651 0.087
362,000 32,170 0. 16
37,000 3,200 O. 16
42,000 6 , 300 0.28
" _0. ~. '" 0;07(+)*
477,000 - 19,037*
22,000 2,300 O. 19
'1,141,000 72,658
bbl/cd tons/yr
*Does not include distillate storage.
(a) Seattle-Tacoma Air Pollutant Emission Inventory, 196~.
(b) St. touis Air Pollutant Emission Inventory, 1968.
(c) National Emission Standard Study, Report of the Secretary of HEW to the
U nit e d S tat e s Con g res s, March, 1 9 TO .
(d) Source Inventory of Air Pollutant Emissions, San Fran~isco Bay Area, 1969.
(e) Louisville Metropolitan Area Afr Pollutant ,Emi~sion Inventory, 1969.
-------
The 1963 report of the National Petroleum Council
listed storage tankage at refineries for the 1962-1963 period.
Extrapolating these data to 1968 refining capacity data (U.S.
Bureau of Mines, 1968) results in an estimate of crude stor-
age capacity at refineries of 137 million barrels. Similarly,
estimated gasoline storage is 204 million barrels.
Establishment of typical tank sizes and ranges poses
a major problem. Data from the Los Angeles study (U.S. Public
Health Service Publication 763, 1960) lead to average tank
capacities of 55,000 barrels for crude oil and about 28,000
barrels for distillate. Private communications from know-
ledgeable petroleum industry personnel indicate that, for a
medium-sized refinery on the East Coast, crude tanks average
about 150,000 barrels and distillate tanks about 80,000 barrels.
For our study, a Poisson distribution function was
used to generate the number of tanks of incremental sizes over
a range felt to be representative of refinery practice. For
crude oil storage, the generating function covered the range
from 20 to 160 thousand bbl and was induced to peak at 80,000
bbl. For distillate storage the range was from 20 to 120
thousand bbl with the peak at 60,000 bbl.
Data on meteorological conditions, including tem-
perature and wind speed, were taken from API Bulletin 2513
and adjusted to reflect the predominance of warm coastal
regions as refinery locations.
API data (Bulletin 2518) suggest that the working
loss is essentially independent of the turnover rate (through-
pu'tlta'nk capa'city'}:~" for turnover rates of less than 36 per
year. The ratio of annual throughput to capacity gave turn-
over rates of about 30 for crude oil and about 10 for gasoline.
Tank outage was assumed to be 50 percent of tank
height. Reid vapor pressure was taken as 10.5 1b for gas-
oline (U.S. Bureau of Mines, 1971) and as 7.5 lb for crude
oi 1.
The distribution of floating and fixed roof tanks
was assumed to be 75 percent floating roof and 25 percent
fixed roof (Duprey, 1968).
Hydrocarbon emissions from refinery storage of
crude oil and gasoline are tabulated in Table 111-20 and
21. For 1968, the total estimated hydrocarbon emissions
are: refinery crude oil storage, 324,000 tons; refinery
gasoline storage, 424,000 tons.
III-43
-------
Table
III-20
Crude
Storage
losses
at
Refineries
Floatl ng Ro flosses
Tank , Fixed Roof losses Standi ng
Tank SI ze Dlstrlbuted(» Numbe r ( ,) T~~~~~~~~tC,)' Roof &(1) Control (I) Workl ng Breathl ng Control(') Storage
She Dlstrl but I on capaclt) of Palnt(') Pal nt Factor Loss loss Fa c to r loss
Storage Conditions (1D' bb1) (%) (1D' bbl Tanks (1~' bbl/tank)' Factor Factor (%) (ID' bbl) (1D' bb1) (%) (10' bbl)
Crude oil 20 B 11. 0 550 600 1.20 2.8 25 36.4 104.5 75 86.6
total storage capacity,
137 I 10' bbl 40 15 20.6 515 1,200 1. 20 2.5 25 54.7 199.6 75 106.2
liquid Temp, 70°F 60 20 27.4 457 1,800 1.16 2.2 25 70.3 260.5 75 116.9
Temp Change, 17°F 80 20 27.4 343 2.400 1. 16 2.2 25 6B.2 262.4 75 113.2
Reid V.P.. 7.5 Ib 100 15 20.6 206 3.000 1.044 1.3 25 45.8 196.2 75 45.3
True V.P.. 5.7 psla 120 10 13.7 114 3.600 I.D4 1.0 25 2B.4 131. 1 75 21. 4
Wind Speed, 9 mph 14D 7 9.6 69 4,200 1.04 O.g 25 20.4 93.2 75 13.1
Turnovers, 30 160 5 6.9 43 4.BOO 1.00 0.75 25 13.3 66.7 75 7.3
SUBTOTALS 337.5 1,314.2 510.0
......
......
......
I
~
~
Total Crude loss
Volume - 2.162 x 10' bbl
Wetght - 324 I 10' tons (at 3OQlb/bb1)
I) Percent distribution x total storage capacity
I) Total capacity; specific tank volume
'I Specific tank volume x turnovers
. Paint factors from API Bul1etln 2518
, Paint and roof factors from API Bul1etln.2917
, Percentage of tanks with filed or floating roof
-------
Table
111-21
Gasoline
S to rage
Loss
at
Refineries
.....
.....
.....
I
.j:oo
CJ'1
Floating R of losses
Tank Dlstrlbuted(1) Number(') Paint 6(') Control (.) F xed Roof loss s Control (.) StandIng
Tank She T~~~~~~~~d') Palnt(') Work I ng Breathing Storage
She Distribution Caeacl t~ of Roof Factor loss loss Factor loss
Storage Conditions 10' bbl) (S) (10 bbl Tanks (10' bbl/tank) Factor Factor (S) (10' bb1) (10' bbl) (S) (10' bb1)
GasoHne 20 19 39.3 1,965 200 1.20 2.B 25 199.0 267.7 75 350.B
Total storage capacity, 22 45.7 1,143 400 1.20 2.5 25 244.3 257.2 75 2B9.3
204 I 10' bbl 40
liquid Temp, 70°F 60 22 45.7 762 600 1.16 2.2 25 228.6 239.1 75 238.9
Temp Change, 17°F 80 17 34.3 429 800 1. 16 2.2 25 171.6 183.4 75 177.0
Reid Y.P.. 10.5 lb 100 10 20.6 206 1,000 1.04 1.3 25 103.0 97.9 75 56.3
True Y.P.. 6.8 pslo 120 9 -18.4 153 1,200 1.00 1.0 25 93.7 81.1 75 36.8
Wind Speed, g mph
Turnover., 10
SUBTOTALS 1,020.0 1,1'26.4 1,149.1
Total GasoHne loss
Volume - 3.296 I 10' bbl
WeIght - 424 I 10' tons (at 257 lb/bbl)
(I! Percent distribution I total storage capacity
(. Total capacity. specHlc tank volum.
! '/ SpecHlc tank volume I turnovers
. Paint factors from API Bu11etln 2518
. Paint and roof factors from API 8u11etln 2517
(, Percentage of tanks .lth fhed or f1 Gatt ng roof
-------
b.
Emissions from Refinery Operations
Data for 1968 operating capacities (charge capacities)
for the various refining operations (catalytic cracking, etc.)
were obtained from the survey of operating refineries given in
the Oil and Gas Journal (1968). These data were combined with
the emission factors for refinery operations listed in Table
111-18 to derive refinery emission estimates. Where ranges of
emission factors are given, as for b10wdown systems, process
. drains, and vacuum jets, midrange values were adopted as being
typical. Where emission factors were not based on refinery
'capacities, estimates were derived from data given in Public
Health Service Bulletin 763 and extrapolated to total U.S.
production figures.
Estimated hydrocarbon emissions from the various re-
fining operations are tabulated in Table 111-22. Annual emis-
sions from refining operation sources total 833,000 tQns.
6.
Petroleum Product Marketing Emission$
a.
Product Distribution
Many refineries, in addition to normal refinery op-
erations, serve as the first link in the marketing and dist-
ribution of products. Supplies for the local market and for
shipment to smaller storage facilities leave the refinery by
tank truck, car and barge. The bulk of all refinery products.
however, are turned into pipelines and transported to large
holding facilities (bulk terminals and bulk stations). Even-
tually gasoline is trucked from these sites to neighborhood
service stations and then transferred to the automobile tank.
In the previous discussion of refinery emissions
(Sectionlll D-5Y~ gasoline losses from refinery storage were
treated. For convenience of calculation, all other refinery
products are treated in this section by combining both re-
finery and non-refinery storage losses as produGt marketing
losses. Gasoline marketing losses are treated subsequent to
refinery storage.
b. . Hydrocarbon Emissions
Estimates of hydrocarbon emissions during the market-
ing of refinery products were obtained in the following manner:
first, the major refinery products were divided into gasoline-
like and non-gasoline type, as shown in Table 111-23.
111-46
-------
Table 111-22 - Hydrocarbon Emissions from Refinery Operations
Source Point
Sewers and Separators
Catalytic Regenerators
Fluid Bed
Moving Bed
Blowdowns-Turnarounds
Compressor Engines
Boilers and Heaters
Pipeline Valves and
Flanges
Vessel Relief Valves
Pump Seals
Compressor Seals
Others (Air Blowing,
Blind Changing, and
Sampling)
Cooling Towers
Totals
Emission Factor
Hydrocarbon
Emissions
~tons/yr)-
216,000
105 lb/l000 bbl
refinery capacity
22 lb/1000 bbl fresh 143,000
feed
57 lbl1000 bbl fresh 10,000
feed
120 lb/l000 bbl re- 243,000
finery capacity
1.2 lb/l000 ft3 fuel 33,000
gas
13.5 lb/l000 bbl re- 28,000
finery capacity
28 lb/1000 bbl re- 57,000
finery capacity
11 lb/l000 bbl re- 22,000
finery capacity
17 lb/l000 bbl re- 35,000
finery capacity
5 lb/l000 bbl refinery 10,000
capacity
10 lb/l000 bbl re- 20.000
finery capacity
16,000*
833.000
*Estimated from Los Angeles Data (Public Health Service.
1960)
111-47
-------
Table 111-23 - Gasoline and Non-Gasoline Type Products
Gasoline Type
Motor gasoline
Non-Gasoline Type
Jet fuel (kerosene)
Aviation gasoline
Kerosene
Special naphthas
j
Distillate oils
Jet fuel (naphthas)
Storage and transfer losses were tabulated for the
more volatile or gasoline-like products. In general, it was
assumed that products in this class were the same as motor
gasoline, both in the number of transfers and in the magni-
tude of evaporative loss. Special naphthas which are not
used as fuels were carried or routed only to the tank of the
user. The evaporative and transfer losses for naphtha
along this shorter route were assumed to be comparable to
losses for gasoline.
Only storage losses were tabulated for the less
volatile products. In general, these products consist of a
variety of blends or mixtures and especially in the case of
distilled oils undergo severe seasonal variations in stocks.
Products of this type were classed as having vapor pressure
less than" 1.5 psia; the emission factors suggested by Public
He.lth Service Publication No. 763 were employed.
Distillate Storage Losses - In the 1967 Census of
Business (Bureau of the Census, 1970), the number, storage
capacity and sales figures for bulk stations and terminals
were given for various petroleum products. Petroleum bulk
stations are defined as generally having capacities less
than 2 million gallons and receive their supply by truck or
rail transport. Bulk terminals are of larger capacity and
are supplied primarily by pipeline, tanker,or barge.
Bulk terminal and bulk station storage capacity
for gasoline is given as 171.7 million barrels. To obtain
the total storage capacity for other distillates it was
necessary to consider both refinery and non-refinery tank-
age. The 1967 Bureau of Census data detailed bulk stor-
age tanks for each specific product. A similar breakdown
111-48
-------
of refinery storage of products such as the jet fuels, avi-
ation gaso1ine,and special naphthas was not available. Total
refinery storage capacity for these distillates was estimated
by multiplying the available stock at refineries by the
factor 2.0. The factor represents a reasonable approximation
of the ratio of storage to inventory. The data obtained are
shown in Table 111-24.
To estimate evaporative and working losses of pet-
roleum products from storage and transfer during marketing,
we used the emission factors shown in Table 111-18. The wide
variety of tank sizes and storage conditions and the indeter-
minate turnover rates do not justify the more detailed loss
treatment given to refinery storage losses of crude oil and
gasoline. For this treatment, the distribution of floating
and fixed roof tanks was considered to be the same as for re-
finery storage. The estimated emissions from bulk storage of
the major petroleum products are summarized in Table 111-25.
Emissions for non-refinery bulk storage of gasoline
and total bulk storage of other petroleum products totals
about 644,000 tons/year.
Transfer Losses - Transfer losses in the gasoline
distributive chain occur during terminal loadings and dur-
ing the loading of service or retail station tanks and auto-
mobile tanks. Losses resulting from transferring gasoline
from storage to automobile tanks are as follows:
Terminal loadings of 83 x 109 gallons
with a loss factor of 6.4 1b/1000 ga1-
10n~ 'yie1d266~000.tons.
Service tank loadings of 83 x 109 gal-
lons with a loss factor of 9.0 1b/1000
gallons yield 373,000 tons.
Automobile tank loadings of 83 x 109
gallons with a loss factor of 12 1b/
1000 gallons yield 498,000 tons.
Losses for these three steps total 1,137,000 tons.
In all cases, loadings were assumed to be conducted under a
50-50 splash/submerged filling condition.
A reasonable approximation of transfer losses of
other distillates was obtained similarly. Total loadings of
each major distillate were based on the domestic demand of
each product as reported by the U.$, Bureau of Mines (1968).
111-49
-------
Table 111-24 - Petroleum Product Storage Capacities (exclud-
ing motor gasoline)
Bulk
Station& Total
Refiner1 Terminal Est.s~o~agYb)
Storage a) Storage Capacltles
Product (l06bb1) (1 0 6 b b 1 ). (106bb1 )-
Aviation Gasoline 14.0 5.8 19.8
Special Naphtha 11. 6 5.7 17. 3
Jet Fuel (Naphtha) 17.8 23.9
17.4
Jet Fuel (Kerosene) 30.8 42.1
Kerosene 46.2 29.6 75.8
Distillate Fuel Oil 346.2 169.5 515.7
(a)
(b)
Estimated as twice December 1968 stocks
Refinery plus bulk storage
1 II - 50
-------
Table 111-25 - Petroleum Product Storage Losses
Product
Motor Gasoline
Estimated storage capacity: gal.)(a)
171.7 x 106 bbl (7.214 x 109
Throughput: 83 x 109 gal
Loss Calculation
Fixed Roof (25%)
a) Breathing @ 0.4 lb/day/1000 gal
storage capacity(b}
b) Working @ 11)lb/1000 gal
throughput(b
Floating Roof
a) Standing
1000 bb1
(75%)
storage@ 4.8 lbL(d~YI
storage capacity a}
total
Aviation Gasoline
Estimated storage capacity:
19.8 x 106 bbl (831.6 x 106 9al)
Throughg~t: 1.28 x 109 gal (domestic
demand)\cJ
Loss Calculation
Fixed Roof (25%)
a) Breathing @ 0.4 lb/day/l000 gal
storage capacity(b)
b) Working @ ll)lb/1000 gal
throughput(',
Floating Roof
a) Standing
1000 bbl
(75%)
storage@ 4.8 lbL(d~Y/
storage capacity d}
total
Special Naphthas
Estimated storage capacity:
17.3 x 106 bbl (726.6 x 106 gal)
ThroughDut: 1.134 x 109 gal (domestic
demand)\C)
III-51
Emissions
131,700 tons/yr
114,100 tons/yr
112,800 tons/yr
358,600
15,200 tons/yr
1,800 tons/yr
13,000 tons/yr
30,000
-------
Table 111-25 (continued)
Loss CalcuJation
Fixed Roof (25%)
a) Breathing @ 0.4 lb)daY/l000 gal/
storage capacity(b
b) Working @ 11)lb/l000 gal
throughput(b
Floating Roof
a) Standing
1000 bbl
(75%)
storage@ 4.8 lbL(d~Y/
storage capacity a}
total
Jet Fuel (Naphtha)
Estimated storage capacit~:
23.9 x 106 bbl (1.00 x 10 gal)
Throughput: 5.31 x 109 (domestic
demand)\C)
Loss Calculation
Fixed Roof (25%)
a) Breathing @ 0~4 1b/day/l000 gal
storage capacity(b)
b) Working @ 11 lb/1000 gal
throughput(b)
Floating Roof
a) Standing
1000 bb1
(75%)
storage @ 4.8 lb/day/
storage capacity(a)
total
Jet Fuel (Kerosene)
Estimated storage capacity:
42.1 x 106 bbl
Loss Calculation
Fixed Roof (25%) @ 1.7 lb/day/
1000 bb1 storage capacity(a)
Floating Roof (75%) @ 1.6 1b/
day/1000 bb1 storage capacity(a)
total
III-52
13,300 tons/yr
1 ,600 tons/yr
11,400 tons/yr
26,300
18,300 tons/yr
7,300 tons/yr
15,700 tons/yr
41 ,300
3,300 tons/yr
9,200 tons/yr
12,500
-------
Table 111-25 (continued)
Kerosene
Estimated storage capacity:
75.8 x 106 bb1
Loss Calculation
Fixed Roof (25%) @ 1.7 1b/day/
1000 bb1 storage capacity(a)
Floating Roof (75%) @ 1.6 1b/
day/1000 bb1 storage capacity(a)
total
Distillate Oil
Estimated storage capacity:
515.7 x 106 bb1
Loss Calculation
Fixed Roof (25%) @ 1.7 1b/day/
1000 bb1 stora~e capacity(a)
Floating Roof (75%) @ 1.6 1b/
day/1000 bb1 storage capacity(a)
total
Total Product Storage Emissions,
excluding refinery stored gasoline
(a) Bulk Stations
(b) Duprey, 1968
(c) 1968 Minerals
(d) Public Health
and Terminals Only
Yearbook, Bureau of Mine~
Publications No. 763
III-53
5,900 tons/yr
16,600 tons/yr
22,500
40,000 tons/yr
112.,900 tons/yr
152,900
644,100 tons/yr
-------
. Estimated transfer losses for aviation gasoline
totalled 17,400 tons and for naphtha-type jet fuel 72,200
tons. Loss of special naphtha, from the storage tank to
the user's tank only, was estimated at 8,969 tons. Total
estimated loss of a'l these products was 98,000 tons.
These estimates were obtained by analogy to the transfer
loss calcuftation for gasoline.
7.
Summary of Petroleum Hydrocarbon Losses
Table 111-26 summarizes the total hydrocarbon emis-
sions within the petroleum industry. Total estimated emis-
sions were about 4.2 million tons for 1968. The breakdown
shows storage losses of"l.9 million tons and losses in
transfer amounting to about' 1.3 million tons. Process or
refinery' emissions totalled 0.8 million while about 0.2 x
106 tons were emitted during crude oil production.
Several potentially important emission sources
were considered but not estimated because of a lack of data
on which to base meaningful emission estimates. These in-
cJuded losses occurring during marine and pipeline transfers
of crude oil and finished products, as well as accidental
spills and pipeline leakage.
E.
SOLVENT EVAPORATION
1.
General
Since the pioneerin9 studies of smog formation by
Haagen-Smit et al (1950, 1952), organic solvent evaporation
has been recognized as a significant source of air pollution.
Lunche (1957) reported survey data which indicated that 50
percent of the stationary emissions of organics in the Los
Angeles area in 1955-1956 were due to solvent evaporation
sources. More recent data from S8n Francisco show that, even
with the stringent controls applied, solvent emissions are
the largest single category of stationary organic emissions,
amounting to about 36 percent of the total stationary source
organic emissions (Bay Area Air Pollution Control District,
1969).
The wide variety of solvents and end users makes
a detailed inventory of usage and resulting emissions highly
speculative. Additional complications arise from the fact
that for many of the organic chemicals used as solvents,
'significant amounts of the total production are used in non-
solvent applications. To arrive at total emissions estimates,
III-54
-------
Table 111-26 - 1968 Summary of Petroleum Industry Hydrocarbon
Emissions
Operation
1.
2 .
Crude Oil Production
Crude Storage in Transportation System
Crude Storage at Refinery
3.
4.
Refinery Operations
Separators
Catalytic Cracking Regenerators
B1owdowns and Turnarounds
Compressors
Boilers and Heaters
Miscellaneous (valves, flanges, seals,
blind changing, cooling towers, etc.)
Subtotal, Refinery Operations
Marketing
Gasoline Storage
Refinery
Bulk Stations and Terminals
Other Distillate Storage
Aviation Gasoline
Jet Fuel (naphtha)
Jet Fuel (kerosene)
Special Naphtha
Kerosene
Distillate Oils
Gasoline Transfers
Refinery
Terminals
Service Station Tanks
Automobile Tanks
Other Distillate Transfers
. . "' o. .
~ Subtotal, Ma~ketin~'
5.
6 !.'
total Industry Emissions
III-55
Emissions
Jtons/yrt
224,000 .
470,000
324,000
216,000
153,000
243,000
33,000
28,000
160,000
833,000
424,000
358,600
30,000
41,300
12,500
26,300
22,500
152,900
63,000
266,000
373,000
498,000
98,000
2,366,100
4,217,100
-------
we assumed that, for all solvents, 100 percent of the total
usage is eventually emitted to the atmosphere. This assump-
tion, coupled with estimated usage of the major industrial
chemicals as solvents, allows an estimate of total emissions.
The major end uses of solvents were reviewed, particularly
for data on their relative contribution to total emissions
and the geographical location or distribution of end use.
2.
Total Evaporative Emissions
Industrial solvent consumption has been reviewed
by Kirk-Othfuer (1969). For 1966, they report a total con-
sumption of approximately 12 billion pounds. Of this 12
billion, 4 billion were used in the coatings industry, 2
billion in dry cleaning, 1 billion in metal cleaning and
degreasing, and the remainder in the manufacture of inks,
adhesives, aerosols, etc.
To obtain a more recent estimate of total usage
and total emissions, data on total production of the major
chemicals used as solvents was obtained 'from the u.s. Tariff
Commission (1968) and combined with data on estimated sol-
vent usage obtained from the technical and trade literature.
The most useful source of estimated end usage was Chemical
Profiles [Oil, Paint and Drug Reporter (1966, 1968)]. Other
data were obtained from individual issues of the Oil, Paint
and Drug Reporter, Chemical and Engineering News, and Hydro-
carbon Processing. These data on national total production
and solvent usage were compiled to produce Table 111-27.
If eventual complete evaporation is assumed, annual organic
solvent emissions amount to about 7 x 106 tons/yr.
In the Nationwide ~urve~of Air Polluta~t Emissions,
1968 [U.S. Department of Health, Education and Welfare Cf97OT],
estimated total emissions from solvent evaporation sources of
3.1 million tons/yr were derived from extrapolation to a nat-
ional scale of data gathered in four metropolitan areas: Los
Angeles, San Francisco, St. Louis, and Washington, D.C.
Table 111-28 gives a comparison of total solvent
emissions and per capita emissions for some recent area in-
ventories obtained through our survey of pollution control
agencies. The per capita emissions range from a low of
7.5 lb/person year for Denver to a high of 60.4 lb/person
year for San Francisco. Since most of the inventories are
based on estimates rather than actual usage and sales sur-
veys, the range is probably more a reflection of differences
III-56
-------
Table 111-27 - Industrial Solvents
Solvent
Special Naphthas
Perchloroethylene
Ethanol
Trichloroethylene
Toluene
Acetone
Xylene
Fluorocarbons
Methyl Ethyl Ketone
l,l,l-trichloroethane
Methylene Chloride
Methanol
Ethylene Dichloride
Ethyl Acetate
Cyclohexane
Methyl Isobutyl
Hexanes
Benzene
n-Butanol
Nitrobenzene
Ketone
Turpentine
IS,bpropyl Acetate
Ethyl Ether
Monochlorobenzene
Isopropanol
Diethylene Glycol
Methyl Acetate
C re sol s
Phenol
Chloroethane (ethyl
chloride)
Carbon Tetrachloride
Pinene
Cyclohexanol
Cyclohexanone
Ethyl Benzene
Iso b u ty 1 A 1 co h 0 1
Chloromethane
n-Butylacetate
Methyl Chloride
Total
1968 Production
(l08 1b)
,
Estimated Solvent
Usage (lOB lb)-
86.5
5.7
5.3
4.9
4.8
86.5
6.4
21. 3
5.2
49.1
13.6
38.3
8.2 (1969)
4.5
3.0
4. 1
3.6
3.5
3.2
2.8
2.8
2.7
2.4
1 , 7
1.6
3.0
38.2
48.0
1.8
20.4
1.8
2.3
67.0
4.3
4.0
1.5
1.3
1.0
0.9
0.5
0.4
0,4
0.4
0.3
0.2
2.4
0.4
1.0
5.8
20.7
0.3
O. 1
1.2
15. 1
5.7
O. 1
0, 1
0.05
. 7.6
1.~
7.2
. . 4.8
40.3
1.1
1.4
0.6
3. 1
142.7
1t1-57
-------
Table 111-28 - Recent Inventory Data on Solvent Emissions
Total Solvent Per Capita
Population Emissions Emissions
Region Date ( thousands) Reported, ton/yr lb/person yr
Bay Area APCD 1969 4,100 123,880 60.4
Los Angeles County 1968 7,060 182,500 51. 7
Puget Sound APCD 1969 1,907 21 ,930 23.0.
San Diego Air Basin 1968 1,300 1 5,330 23.6
Orange County 1968 1,318 15,878 24. 1
.....
.....
..... Ventura County 1968 350 2,482 14.2
I
U'1
00 Monterey and Santa
Cruz Counties 1968 355 4,307 24.3
Den ve r 1969 515 1,919 7.5
Philadelphia AQCR 1968 5,500 43,909 16.0
Greater Cincinnati 1969 1 ,600 7,491 9.4
New York City 1966 7 ,980 99,320 24.8
( t1 S ARE s t i mat e 70)
-------
in assumptions made and sources inci~ded in the estimates
rather than actual differences in emissions. For those
areas which base their data on actual sales and usage sur-
veys (San Francisco and Los Angeles), per capita emissions
are at the high end of the range.
For comparison, the HEW estimate of 3.1 million
tons/yr is equivalent to about 31 1b/person year. Our esti-
mate, based on industrial usage estimates and total pro-
duction figures, is equivalent to about 70 1b/person year
and is more nearly in agreement with the values for Los
Angeles and San Francisco.
3.
End Uses and Emissions
The diversity of end uses for organic solvents re-
quires that any review include consideration of only broad,
general groupings or categories of end use. The principal
categories usually considered are surface coatings, metal
cleaning degreasing, dry cleaning, printing and publishing,
rubber and plastics processing, adhesives, solvent extraction
and miscellaneous usages such as specialty chemicals, drugs
and pharmaceuticals and aerosols. Distribution of solvent
usage among these end use categories varies widely, depend-
ing on the methodology and scope of estimation and the de-
finition of the individual categories. It is generally
agreed that surface coating is by far the largest end use
of solvents, ranging from 30 to 60 percent of total usage.
Other major users in the approximate order of total con-
sumption are degreasing, rubber and plastics, dry cleaning,
and printing. '
The most complete and detailed breakdown of re-
gional solvent usage has been made by the Los Angeles County
Air Pollution Control District in studies leading to leg-
islative action for the control of solvent emissions
(Rule 66). These studies and their results have been re-
ported by Lunche et al (1957) and Chass et al (1960, 1963),
and also in technical reports by Los Angeles County APCD
personnel. Similar studies by the San Francisco Bay Area
APCD have been reported by Wohlers (1965) and others.
a.
Surface Coatings
Surface coating operations employ the full spec-
trum of paints, varnishes, lacquers, stains~and shellacs
in a variety of application methods. Compositions of sur-
face coatings and particularly the amounts of solvents and
II I-59
-------
thinners used vary with the end use and type of application
. method. Volatile organic constituents range from less than
10 percent for high solids content coatings for dip and
roller application to over 70 percent for spray applicattons
of lacquers, stains and sealants.
Water-base paints, widely used as both interior
and exterior architectural coatings, have volatile organic
contents of the order of 2 to 5 percent.
By far the bulk of the emissions from surface
coating' operations result from the end use application and
evaporative drying or curing steps. Emissions from the man-
ufacture of coatings have been estimated by Lunche (1957) to
be only-l.2 percent of the total volume of coatings manu-
factured. These emissions result from open or uncontrolled
evaporation losses from milling and cooking of the coating
components.
Of the end use categories, surface coating operations
use the largest quantities of mixed petroleum solvents, both
aromatic and aliphatic, as well as significant amounts of
the oxygenated solvents - ketones and alcohols - and esters.
According to Lunche (1957), surface coating operations ac-
count for about 42 percent of the petroleum (n~phtha) type
solvents and about 38 percent of the ketones, alcohols, esters,
and miscellaneous solvents consumed in the LAAPCD.
Recent data from the Bay Area APCD (1969) show that
58 percent of estimated solvent evaporation emissions in the
BAAPCD derive from surface coati~g operations.
Estimates of solvent comsumption by the coatings
industry are given by':Kirk-Othmer (1969) for 1966 and, more
recently, by Yazujian (1971) for 1970. These sources pre-
sent somewhat conflicting estimates in their breakdown of
usage by solvent classes and in their total consumption
estimates, as shown in Table 111-29.
111-60
-------
Table 111-29 - Reported Estimates of Solvent Consumption by
Coatings Industry
Solvent
Hydrocarbon
Aliphatic
Aromatic
Other
Total
Alcohols
Esters
Ketones
Miscellaneous
Grand Total
1500
1000
85
2585
400
270
580
100
3935
600
690
1290
555
250
815
275
3185
Although not specifically stated in either of the
above references, it is believed that in both cases the esti-
mates are based on data for sales by the manufacturer. Thus,
we believe the totals are considerably low because they do
not account for solvents and thinners (primarily hydrocarbons)
added by the user before appliction.
For industrial applications, and some architectural
coatings, dilution or thinning before usage greatly increases
the solvent content in the applied coating and thus greatly
increases emissions. For example, a typical primer coat for
spray coating of appliances with an as-purchased solvent con-
tent of 42 percent by weight is reduced about 4:1 with an al-
iphatic thinner (Kirk-Othmer, 1969). Although this may be an
extreme case, estimates of solvent emissions based on as-pur-
chased formulations can be considerably low.
Our own estimate of total solvent usage in sur-
face coatings applications is about twice that given by Kirk-
Othmer and Yazujian, or in the range from about 6 to 8 billion
pounds per year. This amounts to from about 40 to 55 percent
of the total estimated solvent consumption or emissions by
all sources, putting our estimate in the range of surface
coating emissions estimated from Los Angeles and Bay Area sur-
vey data.
111-61
-------
b.
Degreasing
Degreasing and metal cleaning operations consume
large quantities of solvents in operations which range from
small parts cleaning in garages and small job shop operations
to large production line operations in automobile and air-
craft assembly plants. These degreasing operations employ
both liquid immersion and vapor phase treatment, or combin-
ations, but, predominantly, the large volume production op-
erations employ vapor phase degreasing.
Losses from degreasing result from vapor loss
through the loading ports and tank openings, and from drag-
out. Kearney: (1960) reports solvent makeup of 2.4 gal/hr
for a vapor degreasing operation cleaning 40,000 lb/hr of
automotive parts. Solvent evaporati"ori has been estimated
at 1/2 gallon per day per square foot of tank top opening
[Aerospace Industrtes Association (1968)] for typical vapor
degreasers.
Along with dry cleaning, degreasing consumes large
amounts of the halogenated solvents. Until the advent of
recent restrictions because of its photochemical reactivity,
trichloroethylene was the principal solvent used in vapor
degreasing.For those users in areas limiting the emissions
of trichloroethylene, inhibited l,l,l-trichloroethane has
been satisfactorily substituted with only slight modifications
of operating conditions. Other solvents used for specific
cleaning operations are perchloroethylene, trichlorotri-
f1uoroethane,and other mixed chlbro- and fluorocarbon sol-
vents.
" In addition to the halogenated solvents used in
vapor degreasing, the numerous smaller operations which re-
ly on immersion cleaning use a variety of available solvents,
both halogenated and petroleum solvents. Thus, degreasing
emission estimates based on consumption of the major hal-
ogenated solvents alone could lead to indeterminately low
estimates.
The best estimates of total solvent emissions from
degreasing operations are derived from survey data derived
from the detailed inventory studies made by the Los Angeles
and Bay Area Air Pollution Control Districts. These data
indicate that solvent emissions from degreasing range from
about 9 to 13 percent of total organic solvent emissions.
Thus, we estimated total emissions from this end use cate-
gory to be in the range from 1.3 to 1.9 billion pounds per
year.
111-62
-------
c.
Dry Cleaning
The dry cleaning of wearing apparel and other tex-
tiles consumes significant amounts of both petroleum naphtha
and chlorinated hydrocarbon solvents. Principal solvent
types are Stoddard solvent (aliphatic, boiling range 3100 to
390°F), 1400 safety solvent (aliphatic, boiling range 3500
to 400°F), and perchloroethylene.
In the 1963 Census of Business, dry cleaning plants
using synthetic solvents (principally perchloroethylene) out-
numbered those using petroleum solvent and accounted for about
51 percent of total sales receipts. Recent data from the
National Institute of Dry Cleaning, reported by McGraw and
Duprey (1971)0 indicate that 50 percent of the plants use
petroleum solvents and 50 percent use synthetic solvents. Of
the plants using synthetic solvents, 25 percent are reported
to be controlled, generally by adsorbent vapor recovery sys-
tems.
Emissions from dry cleaning operations occur prin-
cipally as a result of the drying operation, with lesser
amounts being evaporated from still and filter residues dis-
posed of as solid waste. McGraw and Duprey (1971) reported
estimated emission factors of 305 lb/ton of clothes cleaned
in plants using petroleum solvents and from 35 to 210 lb/ton
of clothes cleaned in plants using synthetic solvents, de-
pending on extent of control. These authors reported survey
data from California and Michigan w~ich indicated that about
18 lb/person year of clothes are cleaned in moderate climates
and about 25 lb/person year in colder climates. On this
basis and the 50/50 distribution of clothes cleaned by pet-
roleum and synthetic solvents, total dry cleaning emissions
range from about 300 million to over 500 million pounds per
year.
From data reported by the National Institute of
Dry Cleaning (1971), total solvent losses from plants using
petroleum solvents amount to nearly 500 lb/ton of clothes
cleaned in a plant employing a washer-extractor operation
with a regenerative filter. Of the nearly 500' lb/ton loss,
76 percent was evaporated to the atmosphere at the plant and
24 percent discarded as residue in the disposal of solids.
Based on this higher emission factor of 500 lb/ton of
clothes cleaned in plants using petroleum solvents and the
data of McGraw and Duprey for the pounds of clothes cleaned
per person year, total petroleum solvent emissions are esti-
mated at about 500 million 1b/yr.
111-63
-------
Regional survey estimates of dry cleaning emissions
vary widely, but recent Bay Area data (1969) indicate that
dry cleaning operations emit 6.5 percent of the total organic
solvent emissions. This is equivalent to 22.2 tons/day for
the Bay Area, or about 3.9 lb/person year. Since sources in
the Bay Area may be assumed to be better controlled than the
average, dry cleaning emissions in other urban areas may be
assumed to exceed 4 lb/person year. Extrapolation based on
this factor yields a national estimate of about 800 million
lb/yr.
For this study, we estimate dry cleaning solvent
emissions to be in the range from about 500 to 800 million
pounds per year.
d.
Printing and Publishing
The graphic arts industry consumes significant
quantities of solvents in inks and diluents and for solvent
cleaning operations. Emissions relating to this industry
are similar to those for surface coatings in that losses
occur predominantly at the point of application and end us-
age, rather than in the manufacture of component materials.
A general review of the air pollution control prob-
lems and practices of the graphic arts industry, as well as
industry survey data on ink and solvent consumption~ has
been presented by the Graphic Arts Technical Foundation,
Gadomski (1970). Recent reviews of inks and solvents have
been presented by Renson (1968), Salomon and Petrone (1969),
and Ianetti (1969).
The types and amounts of solvents emitted from
printing processes vary widely, depending on the printing
process being used. Probably most significant from a total
emission standpoint are the gravure processes, rotogravure
and flexogravure, since they utilize the so-called solvent
type inks, which are highly volatile and fast drying. News-
print, letterpress,and lithographic inks are the so-called
oil type inks which are generally low in volatile solvents
content. However, processes using oil type inks which re-
quire heat setting or drying at temperatures of 400°F or
greater can result in emissions of partially oxidized species.
These emissions from high temperature drying, although a
small part of total emissions, are often much more noxious
and photochemically reactive than emissions from lower tem-
perature evaporation processes.
111-64
-------
A current industry trend toward usage of water based
inks and solventless formulations may effect significant re-
ductions of solvent emission. However, these developments
have not progressed sufficiently to allow any estimate of their
total effect.
The solvents used for flexographic and gravure inks
listed by Salomon and Petrone (1969) are made up of eight
groups: aromatic hydrocarbons, aliphatic hydrocarbons, mixed
aromatic and aliphatic hydrocarbons, alcohols, glycol ethers,
esters, ketones,and miscellaneous. Boiling ranges vary from
below 150°F to about 400°F, but the majority have intermediate
raoges of 200° to 350°F.
Letterpress and lithographic ink solvents listed
by Ianetti (1969) are either aliphatic hydrocarbons or glycols,
with boiling ranges generally ranging 'from 400°F to over 600°F.
TOtal solvent usage in ink manufacture was estimated
by Renson (1968) at about 170 million pounds for a total ink
production of about 850 million pounds in 1967. Of this total
solvent in manufactured ink, about 65 percent was used in flex-
ographic and gravure inks.
In addition to the solvents used in the manufactured
inks, significant additional solvent' is used to dilute the
purchased ink before applicatlo~ and for various cleaning op-
erations. From data presented in the Graphic Arts Technical
Foundation Report (1970), the total solvent usage in gravure
processes is about twice the amount present in the purchased
ink. Lunche (1957) gave a breakdown of gravure solvent usage
that indicated total solvent usage to be about four times the
amount of solvent in inks purchased.
By applying these dilution factors to the estimate
of 110 million pounds of solvent in gravure inks (Renson, 1968),
total solvent emissions from gravure printing alone would
range from 220 to 440 million pounds per year.
Recent community inventories have not estimated
printing plant emissions as a separate category. Wohlers
(1957) presented a data summary which indicated that about
3.2 percent of total organic solvent emissions in the Bay
Area resulted from printing operations. Lunche (1957) gave
data for Los Angeles that indicated rotogravure plants emitted
about 2.5 percent of total solvent emissions.
111-65
-------
Based on these latter inventory estimates, total
solvent emissions from printing operations range from about
350 to 450 million pounds per year. We selected this range
to represent our best current estimate.
e.
Rubber and Plastics
The only estimates of total solvent usage in the
manufacture and processing of rubber and plastics are those
derived from the area surveys of Los Angeles and San Fran-
cisco. Wohlers (1,965) estimated total solvent emissions for
rubber and plastics processing at 24 tons/day or 8.6 percent
of total solvent emissions. Lunche (1957) attributes 10.9
percent of total solvent purchases to the rubber, plastics
and adhesives manufacturing industry.
Based on these
total nationwide solvent
processing of rubber and
billion pounds per year.
estimates by Lunche and Wohlers,
emissions from the manufacture and
plastics may range from 1.2 to 1.5
f.
End Use Summary
Estimated total consumption of solvents and, there-
fore, emissions are summarized on the basis of major end use
category in Table 111-30.
Table 111-30 - Estimated Solvent. Emissions by End Use Category
Identified Source
Solv.ent
"Emissions
(1011 lb/yr)
60-80
1 3- 1 9
5-8
3.5-4.5
12-15
93.5-126.5
Surface Coatings
Degreasing
Dry Cleaning
Printing and Publishing
Rubber and Plastics
Subtotal, Identified Sources
Unidentified Sources
Total
16.2-49.2
142.7
111-66
-------
F.
CHEMICAL PROCESS INDUSTRY
1.
General
The chemical process industry, as broadly defined
by the Standard Industrial Classification Manual, Bureau of
the Budget (1957), encompasses the spectrum of industries
lumped under the major product group designation of "chem-
ical and allied products," as SIC 28. The diversity of the
chemical industry is illustrated by the membership list of
the Manufacturing Chemists' Association, the major industry
association representing the chemical industry. Member com-
panies include major petroleum refineries, non-ferrous metal
producers, pharmaceutical manufacturers, and rubber and
plastics manufacturers, as well as those companies engaged in
the manufacture of the commonly considered industrial chem-
i ca 1 s .
In the development of criteria for defining and
limiting the scope of the present study, a candidate list-
ing of organic chemicals produced in excess of 100 million
pounds per year was prepared. A review of the literature
was made for information relating to organic emissions from
the manufacture or processing of the chemicals in the can-
didate listing. The trade and technical journals provided
essentially no quantitative data on organic emissions from
chemical processing, although a great many articles dis-
cussed in qualitative terms the successful application of
pollution control systems.
2.
Industry Emissions
Chemical process industry emissions may be esti-
mated by applying generalized loss factors developed by
analogy from the emission factors reported for the petroleum
industry. Mencher (1967) used this methodology in treating
a typical petrochemical plant producing ethylene from pet-
roleum feed stocks. For the various process equipment and
storage tanks in a plant producing 500 million 1b/yr ethylene,
Mencher estimated emissions by using emission factors de-
veloped in the Los Angeles studies of petroleum refining (PHS
Publication No. 763, 1960) adjusted for changes in volatility
and storage capacity. Using factors for a well-controlled
plant operation, Mencher arrived at a figure for total or-
ganic vapor loss of 10,430 lb/day or 0.21 percent loss by
weight based on feed throughput.
111-67
-------
Since statistical data are generally reported in
terms of amount of product, Mencher's total loss factor was
transformed from a feed throughput basis to a product basis.
Since the ethylene yield is about 30 percent of the feed-
stock, losses from ethylene manufacture are estimated at
about 0.7 percent of ethylene produced.
Mencher suggests that similar treatment of other
hydrocarbon processing plants would result in estimated
emissions ranging from 0.1 to 006 percent by weight of feed
throughput or about 0.3 to 2.0 percent based on product, as-
suming a feed-to-product ratio similar to t~at for the
~thylene plant. ",
Using these generalized loss factors, we "made
some estimates of total emissions for the entire petrochem-
ical industry. According to Bland (1970), the total ali-
phatic plus aromatic petrochemical output was 98~3 billoon
pounds in 1968 and about 113.5 billion pounds in 19690 As
shown in Table 111-31, the total estimated emissions for
this major segment of the chemical industry range from
about 2 x 105 to about 1 x 106 tons/yr.
Table 111-31 - Estimated Hydrocarbon Emissions From Petro-
chemical Production
Total Production
Estimated Emissions
lower Limita Upper LimitD
9803 X 109 lb (1968)
113.5 x 109 lb (1969)
a) Based on
b) Based on
1.5 X 105 ton/yr
1.7 x 105 ton/yr
9.8 X 105 ton/yr
1.1 x 106 ton/yr
0.3% of production
200% of production
The chemical industry8s own assessment of its pol-
lution problems, emissions,and extent of control was given
in the report of the 1967 survey of MCA member companies
(MCA, 1967)0 This report summarized data obtained from a
survey response from 992 chemical plants, representing
396,748 employees in 641 communities nationwide. As shown
in Table 111-32, the state-by-state breakdown of organic
vapor emissions totals 2088 million lb/day, under existing
controls. The estimated emissions would have been about 7
million lb/day without controls.
111-68
-------
Table 111-32 - Chemical Industry Organic Emissions
(pounds per day)
Alabama
Ari zona t Nevada t
New Mexicot
Oklahomat Utah
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Idahot Minnesotat
Montanat Nebraska
North Dakotat
South Dakotat
Wyomi ng
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Mainet New Hamp-
shiret Rhode
Islandt Vermont
Maryland
Massachusetts
Michigan
Mississippi
Missouri
New Jersey
New York
North Caro1inat
South Carolina
Ohio
Oregon
Pennsylvania
Tennessee
Organic
Particulates
(~olid 'or
1iquid)-
3tOOO
t
t
13,000
t
StOO~
1tOOO
1 t 000
1 t 000
4tOOO
4tOOO
7tOOO
t
t
3tOOO
13,000
t
t
2tOOO
11 t 000
t
t
6tOOO
StOO~
2tOOO
1StOOO
t
7tOOO
2tOOO
111-69
Other
Organics
(gas or
vapor)
46tOOO
t
3tOOO
84tOOO
2tOOO
18tOOO
4tOOO
7tOOO
1 tOOO
t
24tOOO
13tOOO
26tOOO
8tOOO
1S6tOOO
666tOOO
t
2tOOO
25tOOO
64tOOO
9tOOO
44tOOO
102tOOO
48tOOO
16tOOO
115tOOO
4tOOO
79tOOO
149tOOO
Estimated Emissions
if no Treatment or
Control Mea~ures
Were in Use
Organ; cs
118tOOO
t
3tOOO
549tOOO
35tOOO
33tOOO
17tOOO
14tOOO
4tOOO
27,000
68tOOO
271tOOO
36tOOO
11 t 000
525tOOO
lt072tOOO
4tOOO
10tOOO
54tOOO
104tOOO
14tOOO
115tOOO
441tOOO
87tOOO
42tOOO
297tOOO
12tOOO
231 tOOO
390tOOO
-------
Table 111-32 (continued)
Texas 6,000 925,000 1,814,000
V; rg i n i a 1,000 7,000 10,000
Washington t t 1 ,000
West Virginia 35,000 215,000 522,000
Wisconsin t 19,000 33,000
Total 150,000 2,883,000 7,014,000
tLess than 500 1b/day
Individual quantities rounded to nearest 1,000 1b
Sum of column entires may not exactly correspond to
total on account of rounding
*from MCA (1967)
II 1-70
-------
For our estimate of organic emissions from the
chemical industry, the MCA as-controlled value of 2.88
million lb/day was converted to an emission factor based
on employment of the plants surveyed, or 7.3 lb/employee
day. Application of this factor to the total chemical in-
dustry employment of about 1 million (Handbook of Labor
Statistics, 1970) resulted in total estimated emissions
from this source category of about 1.4 million ton/yr.
This extrapolation assumes no net reduction of
emissions due to increased control :activity since the
1967 survey. This may be rationalized by assuming that
the 1967 survey data were heavily weighted by data from
the larger, well-controlled plants and that extrapolation
to include all plants offsets any changes due to increased
control activity.
G.
OTHER INDUSTRIAL PROCESSES
Included in this section are several reviews of
industrial processes that do not logically fit into the
previous major source categories or for which there is suf-
ficient information to warrant detailed discussion.
1 .
Carbon Black Manufacture
The carbon black industry has been recently re-
viewed by the TI-2 Chemical Committee of the Air Pollution
Control Association (Drogin [1968]). This review described
current industry practices and discusses effluent emissions.
Bureau of Mines (1969) figures for carbon black
production in 1968 show that the furnace process, by far,
dominates the industry. Of the total 1968 production of
2.81 billion pounds, nearly 2.4 billion pounds were pro-
duced by the furnace process. The remaining production is
from the thermal black process and, to a lesser extent,
the channel process. Because particulate emission$ from
the channel black process are not controllable, this pro-
cess is slowly being phased out of existence.
Drogin (1968) presents typical process flowrates
and effluent gas compositions for various grades of oil
furnace blacks. These are given in Tables 111-33 and 111-
34. The only other effluent gas analyses found in the lit-
erature were for a typical furnace process in Great Britain
(Allan, 1955) and for a U.S. oil furnace plant of 25 x 106
111-71
-------
Table 111-33 - Typical Oil, Air and Gas Rates Us~d in
Manufacture of Oi1 Furnace B1acksla)
Air Oil Gas
Grade Black ft3/hr ga1/hr ft3/hr
Super abrasion 240,000 240 15,500
Intermediate super
abrasion 190,000 220 12,200
High abrasion 160,000 210 10,000
Fast extrusion 300,000 750 2,000
(a) From Drogin (1968)
Table 111-34 - Typical CompositiQn of Oil Furnace Carbon
Black Exhaust Gasla) .
Component
Carbon dioxide
Composition (vol%,. dry basis)-
Large Particle Small Particle
Size Black Size Black
Carbon monoxide
5 4.3
5 11.8
17-18 13.9
"'1 0.76
"'1 0.31
balance b a 1 an ce
300-400 300-400
200-400 200-400
Hydrogen
Methane
Acetylene
Nitrogen
Hydrogen Sulfide (in ppm)
Sulfur (in ppm)
(a) From Drogin (1968)
I II- 72
-------
1b/year capacity (Shearon, 1952). These latter data are, in
general agreement with those given in Table 111-34, in that
the major organic species are methane and acetylene, with some
smaller amounts of other low molecular wei~ht hydrocarbons.
Dependent on the grade of black being produced, the concen-
trations of total hydrocarbons emitted range from one to two
percent of the effluent gas (for the furnace process), with
the ratio of acetylene to methane ranging from 0.5:1 to about
1 : 1.
Calculated emission factors derived from these
data are in general agreement with the values given by McGraw
(1970). For oil furnace black production, total hydrocarbon
emissions average 400 lb/ton of product. Thus, the total hy-
drocarbon emissions from carbon black manufacture are esti-
mated to be 3 x 105 tons/yr.
The regional distribution of carbon black pro-
duction shows the emissions to be highly localizedg with
22 of the U.S. total of 29 plants being located in Texas and
Louisiana.
2.
Coke Manufacture
A major use of coal in the U.S. is for the pro-
duction of coke to be used primarily in blast furnace steel
production and, to a much lesser extent, in foundry oper-
ations and as fuel. Over 98 percent of the coke produced in
the U.S. is made in by-product ovens, with the remainder com-
ing from beehive oven operations. A breakdown of U.S. pro-
duction and consumption of oven coke by status is shown in
Table 111-35. Beehive coke production in 1968 was only
775,000 tons, almost entirely in Pennsylvania (355,000 tons)
and Virginia (419,000 tons).
A complete description of modern U.S. coke manu-
facture at an integrated iron and steel operation is given
by Varga (1969). Modern slot ovens may typically hold up
to 40 tons of coal per charge, with up to 100 ovens in a
battery. Charging, coking and discharging operations are
arranged sequentially within a battery, so that although
basically a batch operation overall operation may be treated
as being continuous. Typical coking periods range from 16
to 20 hours. A typical material balance for oven coke op-
eration is shown in Table 111-36.
III-73
-------
Table 111-35 - Oven Coke Production and Distribution in United States, 1968(a)
(in thousand short tons)
Commercial Sales
Used by Producer Res i de n t i a 1
State Produced Blast Furnace Other B 1 as t Furnace Foundries Industrial HeatinQ
A lab ama 5,462 3,425 472 501 557 369 9
C a 11 f 0 rn i a, Colorado,
Utah 3,174 2,941 14 b b
Connecticut, Maryland,
New Jersey, New York 7,599 6,421 49 985 415 100 b
..... Illinois 2,074 2,019 76 !> b
.....
.....
I Indiana 8,144 7,467 1 3 b b b b
'-J
~
Kentucky, fH s sou ri,
Tennessee, Texas 2,000 b b b b 96 b
Michigan 3,684 b b b b b b
Minnesota, IH s cons i n 844 b b b b b b
Ohio 8,428 7,224 137 440 b 184 b
Pennsylvani a 18,110 16,747 22 283 b 263 b .
West Virginia 3,360 2,999 b b
Undistributed 4,068 193 1 , 1 36 1,962 871 65
Totals 62,878 53,312 974 3,345 2, 9 ~4 1 ,883 "'3
fa~ Data from U.S. Bureau of Mines (1969)
b Included in undistributed
-------
Table 111-36 - Typical Material Balance for By-Product Coke
Oven Operation* (Basis - one net ton coke
produced)
Item
Amount (pounds)
Input
Coal
2950
Output
CQke + Breeze
By-product Chemicals
2121
290
Coke-oven gas
534 (15,000 ft3)
*Taken from Varga (1969)
Products of the coking operation are coke, coke oven
gas, normally used to fire the coke ovens, and various by-pro-
duct chemicals, such as phenol, benzene, toluene, naphthalene,
tars, and pitches.
The major emission sources and types of pollutants
emitted by U.S. coking operations have been described by
Varga (1969) and others (Dancy [1970], Fullerton [1967]), but
no quantitative data' on emissions have been reported. To ob-
tain quantitative estimates of pollutant emi$sions from cok-
ing operations, data from European and Russian sources have
been used (United Nations [1968], [1971], Kirchoff [1962]).
Except for minor differences in oven size and firing arrange-
ments, European coking practice is quite analogous to U.S.
practice and, thus, emissions should be quite similar.
Kirchoff (1962) reported the fo110win~ analyses of
of the gases released during charging of a 30 m oven (17
tons/charge):
II 1-75
-------
Constituent % by volume
C02 606-8~0
CmHn 0.4-0.8
CO 2.0-3.8
CH4 2.0-5.2
H 4.8-10.8
N 75.2-71.4
o 9-0
In actual practice, emissions during charging often ignite
and partially burn so that some reduction in organic emis-
sions to the atmosphere occurs.
The United Nations reports (1968, 1971) give esti-
mates of emissions from plants in Poland, Chechoslovakia,
France and the USSR. In Tables 111-37 and 111-38 the data
from Poland and the USSR are presented.
Combining the organic emission quantities given
in Table 111-37 for the Polish plant gives an organic emis-
sion total of 3.7 1b/ton coke, or approximately 505 lb/ton
coal charged.
Totalling emissions quantities for the USSR plant
given in Table 111-38 result in a considerably lower value
for total emissions of about 0.8 lb/ton coke, or 1.2 1b/ton
coal charged.
In the National Emission Standards Study, (U.S.
Department of Health, Education and Welfare, 1970) an esti-
mate of the total U.S. organic emission from the coking in-
dustry is given as 1.5 million tons/year (without support-
ing references). For the reported production of 66 million
tons of coke, this estimate amounts to about 30 1b/ton coal
charged or about 45 1b/ton coke produced.
A third estimate of emissions from coking is given
by McGraw. His figure of 4.2 1b/ton of coal charged is
based on data from the United Nations (1969). For beehive
ovens, McGraw recommends a factor of 8 lb/ton coal. Thus,
111-76
-------
Table 111-37 - Nature and Extent of Atmospheric Pollution Caused by Coke-Oven
Batteries*
Seria 1
Number
2
3
4
5
6
7
.....
.....
......
,
......,
......,
8
9
10
11
12
13
14
15
16
17
18
19
Pollutant
Coal (coke) dust
Tar
Methane and homo10gues
Ethylene and homo10gues
Acetylene
Ca-rbon monoxide
Benzene
Naphthalene
Phenol
Pyri di ne
Ammonia
Hydrogen sulphide
Hydrocyanic acid
Oxides of nitrogen
S u 1 p h u r d i 0 x i de
Carbon disulphide
Chlorine
Chlorides
Su1phates
Totals
Chemical
Formula
Discharge in kg/t of charge
Charging Discharging
of Coking and.
Chambers Cycle Quenching
0.621
0.104
CnH2n+2
CnH2n
C2H2
CO
0.103
0.015
0.001
0.037
C6H6
C10H8
CSH50H
C5HSN
tiff 3
SH2
HCN
rlo
0.015
0.005
0.006
0.061
0.008
0.012
0.001
0.013
S02
CS2
C12
C1-
S04
0.011
0.001
0.96
0.059
0.349
0.676
0.102
0.006
0.252
0.087
0.034
0.010
0.006
0.027
0.047
0.003
0.006
0.001
0.001
0.003
1. 45
O. 182
0.010
0.005
0.014
o . 079
0.057
0.038
0.005
0.028
0.072
0.49
Firing
0.005
0.001
2. 171
-
2.17
Total
0.867
0.463
0.779
0.117
0.007
0.295
0.162
0.053
0.095
0.007
0.092
0'.097
0.009
0.019
2.183
0.001
0.004
0.028
0.072
5.29
-------
Table
111-38 - Types and Quantities of Air Pollutants in the
Vicinity of Coking and By-Product Plants (USSR)*
Duration of
Source of pollution, or operation Quanti ty di s- emission for Number of
accompanied by discharge of Pollutant ch a rge d (i n each operation operations
noxious substances discharged tons per day) (in minutes) per day Remarks
1. Coke-oven charging Coal dust 0.530 4-5
Carbon
monoxide 0.220 4-5 106 The figures given are
Sulfur for a single battery
dioxide 0.060 4-5
Benzene
hydro-
carbons 0.057 4-5
Hydrogen
sulphide 0.030 4-5
2. Discharging Coke dust(a) 0.300 0.6 106 The figures given are
C a rb on for a single battery
monoxide 0.064 0.6
3. Quenching of coke with was te. Hydrogen
water steam-circulated to sulphide 0.580 2
remove the phenol Phenols 0.240 2 The figures given are
Ammonia 0.240 2 for a single quench-
Sulfur ing tower serving two
dioxide 0.145 2 212 batteri es
Cyanide
compounds 0.001 2
4. Coke-oven battery waste-gas
f1 ues:
(a) ovens heated with coke- Sulfur Per flue evacuating
oven gas having hydrogen dioxide 1. 830 one battery's com-
sulphide content reduced Continuous bustion products
to 3 g/flm'
(b) ovens heated with blast- C a rb on
furnace gas monoxide 21. 900
5. Wagon-tippler Coal dust 0.045 0.1 175 Unloading wagons of
60 T capacity
~. Chimneys of coal preparation Coal dust 15.600 Continuous
shed drier plant Sulfur
dioxide 6.720 For a twelve-drum drier
plant
7. Equipment and tanks of coke- Ammonia 0.0064
oven gas COnd?c,ation and Pheno 15 0.0025
cooling plant Pyridine
bases 0.0006
Hydrogen
sulphide 0.0006
Hydrogen
cyanide 0.0003
8. Condenser pump room Ammonia 0.0081
Pheno 15 0.0037 For a twelve-drum drier
Hydrogen plant
sulphide 0.0032
Pyri di ne
bas es 0.0005
Hydrogen
cyani de 0.0004
9. Engine room Ca rb on
monoxide 0.0173
Benzene
hydro-
carbons 0.0093
Hydrogen
sulphide 0.0053
Ammonia 0.0022
Naphthalene 0.0005
Organi c
sulfur
compounds 0.0005
Hydrogen
cyanide 0.0003
Pyri di ne
bases 0.0002
(a) Dust of relatively large particle size. settling rapidly and therefore not causing protracted pollution of
the atmosphere
(b) The discharge from these installations and from those listed subsequently is continuous
* Daily capacity of 10,000 (dry-charge). four battery operation
111-78
-------
estimates of total emissions from the various source data may
range from a high (USDHEW) of 1~5 x 106 ton/year to a low of
2.6 x 10~ tons/year (USSR).
For our estimate of organic emissions from the cok-
ing industry, we used McGraw's emission factors. Thus, our
estimate of total emissions from coking is about 2 x 105 tons/
year.
The specific compounds emitted are most probably
those types of organics listed by the UN report (see Tables
111-37 and 111-38).
The emissions of polynuclear aromatic hydrocarbons,
primarily associated with the particulate emissions from cok-
ing, are summarized in reports by Hangebrauck (1967) and GCA
Technology (1970). GCA Technology reviewed data from several
sources and derived an average emission factor for benzo-a-
pyrene (BaP) from coking operations of 1.8 grams per ton of
coal processed. Estimated total BaP emissions are thus about
180 tons/year.
A recent analytical study of polynuclear aromatic
hydrocarbbn emissions from coke production is reported by
Searl (1970). Searl summarizes the data from analysis of ef-
fluent samples taken from twenty different coke oven plants
and gives ranges of composition of eight specifically identi-
fied polynuclear aromatics.
3.
Wood Pulpin~ Industry
Emissions from the wood pulping industry have been
studied by Environmental Engineering, Incorporatedl (1970).
However, no mention is made of non-sulfur containing organic
emissions. In a further review of the literature, only two
publications relate any information on such emissions.
Walther and Amberg (1970) describe an advanced
control system in operation at a 500 ton per day Kraft mill
at Fairhaven, California and give the following values for
non-sulfur emissions from the major source points (after
control):
111-79
-------
Unidentified
Non-Sulfur
Acetone Methanol a-Pinene Organi cs
Source iJ b / d ay) ~ J 1 b / d ay t Jlb/dayt
W~sher seal vent 11 146 715
tank
Washer hood vent 25 92 116
Knotter hood
vent 14 95 20 500
Black liquor
oxidation 85 708 91
Digester* 50 27 118
Totals 185 1068 1060 500
*Without incineration of dige~ter non-condensib1es
Comparison of the flow sheet for the Fairhaven
pJant with the Kraft process flow sheets given in the EEl
report indicates that the Fairhaven plant may well be the
only in the U.S. using such advanced control technology.
However, the emissions tabulated by Walther and Amberg were
used in this study to obtain a crude estimate of the total
emissions from Kraft pulping, since a brief search of the
literature revealed no other tabulation of non-sulfur con-
taining- emissions.
Because the Fairhaven plant uses softwood only,
the estimate of emissions is based only on the softwood
Kraft tonnage, which for 1968 {was 18,440,000 tons or about
76 percent of the total Kraft tonnage. Assuming the
average emissions to be those of the Fairhaven plant with
no digester off-gas incineration, total U.S. emissions
(1968 levels) are estimated as shown in Table 111-39.
Thus, even though the estimates shown in Table 111-
39 are uncertain because of the assumptions made and data
are lacking on other pulping processes, it can be seen that
the pulping industry is a significant source of non-sulfur
containing organic emissions.
111-80
-------
-
Table 111-39 - Estimated Non-Sulfur Organic Emissions from Wood Pulping
Pulp
Production
Pulping Method -(1 0 6 ton t
Kraft, soft wood 18.4
Kraft, hard wood 5.9
Sulfite 2.5
NSSC 3.5
......
...... Dissolving 1.5
......
I
00 Soda 0.2
-'
Methanol
Organic Emissions (103 tons/yr)
Acetone a-Pinene Unidentified
20 3.4 20 9
no data
no data
no data
no data
no data
Total
52.4
-------
According to Kirk-Othm~r (1969), total 1968 U.S.
production of all grades of turpentine from wood pulp pro-
cessing was 32.9 million gallons. Approximately 15 per-
cent of this total was used for solvent purposes, equiva-
lent to about 17,600 tons. Thus, estimated emissions of
a-pinene from Kraft mills more than equals the total. used
for solvent purposes.
The only other article found relating to potential
non-sulfur organic emissions was by Miller (1968). In his
study of improved methods of turpentine recovery, Miller
estimated the total amount of turpentine lost to the atmos-
phere as a result of incomplete recovery as 15 million gal-
lons annually. This amounts to about 54 thousand tons, or
about 2-1/2 times the estimate of total industry emissions
of a-pinene based on Walther and Amberg data.
According to Miller, industry recovery of tur-
pentine currently averages 60 to 70 percent of the potentially
available turpentine. Thus, the 15 million gallons lost
would be in reasonable agreement with the 32.9 million gal-
lons total production cited by'Kirk-Othmer.
4.
Phthalic Anhydride Manufacture
, A specific study of air pollution from phthalic
anhydride manufacture was reported by Fawcett (1970). This
report reviewed the process chemistry, potential emissions,
and control technology.
The ranges of concentration for the major pollutants
in the effluent from a dry recovery (product condenser) pro-
cess prior to abatement are as follows:
Contaminant
Concentration (ppmt
phthalic anhydride
maleic anhydride
naphthoquinone
benzoic acid
aldehydes (as CH20)
40-200
100-600
10-30
5-40
10-100
Total organics emissions thus range from 300 to
1200 lb/hr for a 100 million pound per year plant, depend-
ing on type of process, condenser operation, etc.
111-82
-------
The two major control methods, wet scrubbing and
incineration, were described by Fawcett, with newer plants
tending to choose incineration as the preferred method.
However, no estimate of the extent of existing industry con-
trol is given.
Based on the above emissions and the 1968 pro-
duction of 744 million pounds, total organic emissions, with-
out controls, could range from 10,000 to 40,000 tons per
year. Since the total production is manufactured at only
sixteen locations in the U.S., the pollution load at a
specific location could cause a significant localized problem.
Of the sixteen plant locations, there are four each in New
Jersey and Illinois, three in Pennsylvania, two in California
and one each in Texas, Ohio, and West Virginia.
5.
Synthetic Rubber and Plastics
A limited amount of information on emissions from
synthetic rubber and plastics processing is reported from
plant studies made as part of the Louisville Air Pollution
Study (Public Health Service, 1961). For one plant engaged
in production of synthetic rubber (butadiene-styrene and
butadiene-acrylonitrile copolymers) and polyvinyl chloride
resins, air contaminant analysis and material balances result
in estimates of emissions from major processing areas. Prin-
cipal emissions are the monomers emitted during polymerization
and milling.
For PVC resin manufacture, conversion of plant
data to an emission factor basis results in an estimate of
17 1b of vinyl chloride per ton of plastic processed, from
all processing steps.
For synthetic rubber manufacture, similar treat-
ment results in emission factors for butadiene and acrylo-
nitrile of 40 and 17 1b/ton of rubber processed, respectively.
Applying these estimated emission factors to total
reported quantities manufactured gives crude estimates of
total emissions. For PVC resins and plastics, the 1968 pro-
duction of PVC and copolymers was 2.6 billion pounds, or
1.3 million tons. Total vinyl chloride-emissions thus could
approximate 22 million pounds, or 11 thousand tons.
The 1968 production of butadiene elastomers was
2.5 billion pounds for butadiene-styrene type and nearly
0.3 billion pbunds of butadiene-acrylonitrile type. Ap-
plication of the emission factors to these total amounts
111-83
.-
-------
results in estimated emissions of about 26 thousand tons of
butadiene and nearly 2 thousand tons of acrylonitrile.
H.
GEOGRAPHICAL DISTRIBUTION
1 .
Introduction
The estimation of organic air pollutant emissions
and ranking of sources on a nationwide basis presents only a
partial view of the total problem. Of significant importance
are the regional geographic and demographic distributions of
the sources, emissions and adverse effects. Since the primary
impetus of air pollution control is a concern for the welfare
of the human population, efforts have been made to analyze
those factors which illustrate the relative distribution of
pollutants and people.
2 .
Source Distribution
In previous discussions of major emission sources,
nationwide estimates of sources and emissions are broken out,
where possible, to show distribution at a state level, as
well as pointing out regional concentrations of certain in-
dustrial sources. In this section, further discussion of
the geographic distribution and ranking of sources and emis-
sion is considered. Several possible arrangements or group-
ings of the major emission source categories may be treated
in developing geographical ranking. For example, each of
the source categories could be handled independently and
then combined in various ways to show overall distribution
by regions, states, Air Quality Control Regions (AQCR's), or
Standard Metropolitan Statistical Areas (SMSA's).
A somewhat different, and perhaps more logical
method of arrangement is to initially treat the sources as
two separate groups. The first group is made up of those
sources for which actual geographic location can be speci-
fied or which show a strong correlation with general popu-
lation density. In this first group would fall the major
sources exemplified by petroleum refining, chemical process
industries, fuel combustion, gasoline storage and marketing,
and that portion of the solid waste combustion source that
derives from urban waste disposal.
The second group consists of those sources which
are so diffuse as to make specified source location impos-
sible or, at best, over1y detailed for a study such. as this
and which show little or no correlation with general population
111-84
-------
density distribution. This second group is exemplified pri-
marily by the largest single source category, solvent evap-
oration.
It is to this latter group that we devoted the
most attention, since any desired breakdown of those members
of the first group is obtained by a relatively straightfor-
ward compilation of available statistics.
a.
Solvent Emission Sources
Solvent evaporation emissions, except for those
attributable to dry cleaning and, perhaps, the trade sales
of paint and varnish products, are not directly population
dependent. Solvent emissions from dry cleaning, which ac-
count for about 5 to 8 X 108 Ib/year solvent emissions may
be distributed on a regional basis according to percentage
factors derived from dry cleaning receipts reported in the
1967 Census of Business (Selected Services, BC67-SA1, 1970).
Some information on regional distribution of sol-
vents used in surface coating applications may be obtained
from Commerce Department statistics on the regional ship-
ments of paints and allied products from manufacturing es-
tablishments. These data, however, do not necessarily re-
flect consumption patterns and, therefore, ultimate solvent
emissions.
Regional distribution of sales of paints and al-
lied products by merchant wholesalers is given in the 1967
Census of Business (BC67-W52, 1970). According to defin-
itions used to characterize paint sales, these data probably
represent the so-called IItrade sa1esll products. Trade sales
paints include primarily the interior and exterior archi-
tectural coatings, although sizable amounts of automotive
and machinery finishes and other non-architectural paints
are also handled by trade sales wholesalers.
Industrial coatings are sold directly to the man-
ufacturers of machinery and various finished goods and thus
distribution of these products may best be represented by
the regional distribution of the major manufacturing in-
dustries.
The 1967 Census of Business (BC67-WS6, 1970) re-
ports bulk storage capacity and state and county sales fig-
ures for the common petroleum products, including special
naphthas. Further examination of the sales of special
II I - 85
-------
naphtha shows, however, that the reported figures are, in
some cases., for very large bulk terminals serving state-
wide or even multi-state areas.
For example, total u.s. bulk terminal sales of
special naphtha for 1967 was 498.7 million gallons. Of
this total, 271 million gallons, or about 45 percent, are
represented in sales from ten counties, as shown in Table
111-40. Rather obviously, these sales cannot ref1ect ul-
timate consumption within that particular county. However,
if the reported sales are treated on a statistical region
basis, the distribution may better illustrate usage pat-
terns. Thus, the reported 5tate sales have been combined
to give regional totals.
Geographical distribution of solvent usage and,
therefore, emissions at levels more detailed than the major
statistical regions is thus seen to be highly speculative,
at least for industrial solvent usage. For this study,
distribution of solvent emissions was therefore limited to
estimates at a census region level.
The parameters finally used for estimating geo-"
graphical distribution of solvent emissions were regional
sales receipts data for dry cleaning and special naphtha
solvents and regional manufacturing employment for all
other solvents.
More detailed treatment of solvent emissions by
end use category was not believed to be justifiable. For
example, distribution of trade sales paints as a separate
category would require estimation of solvent content ~f
trade sales paints and also could result in some indeter-
minate amount of double counting of special naphtha sol-
ven ts .
b.
Total Emission Sources
As was mentioned in previous discussion, the
geographical distribution of the majority of emission
sources shows a correlation with population distribution
or can be estimated from industry statistics. Thus, the
development of a rationale for solvent distribution allows
estimation of total emissions from each of the major sources
on a regional basis.
II 1 - 86
-------
Table 111-40 - Special Naphtha Sales from Bulk Terminals in
Ten Highest Counties
County
Middlesex County. New Jersey
1967 Sales
103 gallons
97.975
Cook County. Illinois
Los Angeles County, California
41,616
36,453
Baltimore (City), Maryland
Mecklenburg County, North Carolina
22,368
19,645
St. Lou i s (C i ty ), Mis sou r i
Hamilton County, Ohio
12,014
11 ,868
Kings County, New York
10,273
10,109
Harris County, Texas
Multnomah County, Oregon
8,992
Total
271,313
111-87
-------
Table 111-41 tabulates the variables used to distri-
bute the emissions of each individual source category among
the census regions. The definition of geographical regions
used for this distribution is that employed by the Bureau of
the Census for census regions.
Application of the distribution parameters shown
in Table 111-41 to the total hydrocarbon emissions from the
major sources and subsources then allows regional distribution
of emissions, as shown in Table 111-42. The final columns of
this table give total hydrocarbon emissions for each region
and an emission density factor in tons per square mile/year.
To illustrate the calculations employed in develop-
ing Table 111-42, the calculation of the dry cleaning solvent
emissions for the New England region are described. From the
regional dry cleaning sales receipts data, sales receipts for
the New England region were $112 million or 5.8 percent of
total U.S. receipts. Applying this regional percentage to the
total national estimate of emissions from dry cleaning,
4 x 105 tons/yr, then yields annual emissions of 0.2 x 105
tons for the New England region.
Totalling similarly the regional emissions from
other source categories results in annual hydrocarbon emis-
sions from all sources in New England of 118. x 105 tons. By
dividing this total emissions estimate by the regional area,
66.6 thousand square miles, yields the estimated total emis-
sion density of 17.7 tons/square mile year.
Because of the simplifying assumptions used in
this estimation of geographical distribution of total hydro-
carbons, some minor inconsistencies occurred. For several
major source categories, subdivisions of the major sources
were lumped somewhat differently than in the national esti-
mate. The petroleum industry emissions were lumped into
four subdivisions: crude oil production; non-refinery crude
storage; petroleum refining and all refinery storage of
both crude and refined products; and all petroleum product
marketing. Also, the miscellaneous industrial process
sources were lumped together and distributed according to
manufacturing employment, even though such sources as coke
production and carbon black manufacture are known to be
highly localized.
111-88
-------
Table 111-41 - Parameters for Regional Distribution of Total
Hydrocarbon Emissions
Emission Source Category
Solvent Evaporation
Dry cleaning solvents
Special n~phtha solvents
All other solvents
Petroleum Industry
Crude oil production
Crude oil storage (non-
'.' reftnery)
Petroleum refining and
refi nery storage
Product marketing
Solid Waste Combustion
Municipal and domestic
Industrial
Agricultural Burning
Fue 1 CQmbus ti on
Chemical Industry
Other Industrial Processes
Forest Wildfires
I I I ~ 89
Distribution Parameters
Regional sales receipts
Wholesale trade sales
Manufacturing employment
Crude production statistics
Crude storage statistics
Refinery capacity statistics
Gasoline ~onsumption stat-
istics
Total population
Manufacturing employment
Agricultural sales statistics
Fuel usage statistics
MCA survey statistics
Manufacturing' employment
Forest acreage statistics
-------
Table 111-42 - Distribution of Total Hydrocarbon Emissions by Source Category and Census
Region (Emission quantities in 105 tons/year)
-
-
-
I
~
a
I I
CII I ... en'" I:
1:7' I: U CIIen CII - 0
Ia 00 - ::J S::J ... IO ,&)
~ 1-- "" ~ O,&) en - I-
... 0>, CII 0 os 00 I- 00
Source 1:7' ~ .... I- 0:: I- 0 :0: - ... u
Category c 0. tnCII CII 0.. ol: U 00 en 0
.... 00 c c ~ 01 -c I- C ::J I-
Cens u c Z I- -0 -.... 00 SI:7' -CII IOO ::J 0 ~en en ~en
00 en en ClIVI -- -.... >,1- ::J C 00... -- +> - ';~ CCII CII >,1: Total
Region CII'" -... ~... 0'" OCII 1-0 CII- ~II> I-+-> -1:7' ''' - en I- :::I:O
- I: IOC ...c U I- CII+-> -... - 00 ...en ::JC VI U... en '" - - Emission
UCII -CII OCII CII::J CIII Ctn OCII u:;: II>::J u- ::J - VI I-CII en.... - VI
> u> > ~~ ~c - I-~ - ::J,&) -c -,&) S::J CIIU CII~ 00 en Density
>,- CII- -- ::I 0 ::10 ....>, ...1- cuc ~s 1-1- cuS cu~ ~o 1-- ...-
1-0 0.0 - 0 1-1- I-c cu I- CIllO ::I..... 0 co 01::1 ::10 .s::.1: ...1- 0- os Tons/sq mi
Otn tn tn c(tn U 0.. u- o::cu 0..:;: :;:...- -u c(c:::J u.. u u- co.. u..:;:: I- UJ
New England 0.2 2.3 2.0 -- - --- -- - 0.9 1.0 2.2 0.8 0.3 0.2 0.8 ,.,1 11.8 17. 7
Middle Atlantic 0.8 11. 5. 5.5 --- O. 1 l,7 2.6 3.3 6.2 2.2 0.9 1.1 2.3 1.1 39.3 38.3
Eas t North Central 0.8 7.4 6.5 -- - 0.6 2.7 3.4 3.6 7.3 7. 1 0.9 1.1 2.7 1.7 45.8 18.4
West North Central 0.3 2.6 1.5 0.2 0.2 0.9 1.7 1.4 1.7 10.4 0.3 0.4 0.6 1.5 23.7 4.6
South Atlantic 0.7 7.3 3. 1 --- O. 1 0.3 2.7 2.7 3.5 4.9 0.6 1.2 1.3 3.9 32.3 11. 6
East South Central 0.3 2.3 1.3 O. 1 0.2 0.5 1.1 1. 2. 1.5 2.8 0.3 1.8 0.6 2.3 16.3 9.0
West South Central 0.4 2.9 1.3 2.2 2.5 6.6 2. 1 1.7 1.5 4.8 0.3 7.7 0.6 2.5 37.1 8.5
Mountain O. 1 0.9 0.4 0.4 0.4 0.6 0.8 0.7 0.4 3.0 O. 1 --- 0.2 5. 1 13. 1 1.5
Paci fi c 0.5 6. 1 2.5 0.5 0.7 2.4 2.3 2.3 2.8 5.7 0.3 0.5 1.0 7.7 35.3 3.9
Total U.S. 4.0 43.3 24. 1 3.4 4.7 15.7 17.6 18.0 27.0 42.0 4.0 14.0 10.0 27.0 254.7 7.0
-------
Of some interest for control planning purposes are
the regional distributions of total emissions relative to
total population. Figure 111-2 shows in bar graph form the
percentage of total u.s. hydrocarbon emissions and the per-
centage of total U.S~ population for each of the regional
areas.
111-91
-------
26
24
22
20
18
16
14
12
10
8
6
4
2
Figure 111-2 - Percentage Distribution of Emis-
sions and Population by Census
Region
lII-92
-------
:' CHAPTER I I I
REFERENCES
B.
FUEL COr1BUSTION
Bay Area Air Pollution Control District. Source Inventory of Air
Pollutant Emissions~ San Francisco nay A:~eaf} 1969.
i
ChassD R. L.. and R. E. ~eorge,,"Contaminant Emissions From the
Combustion of Fuels". J. Air Pollution Control Assoc. lQ. (1).
34-43 (1960). ~
C u f feD S. T." R . ~L G e :~ s t 1 e. A. A. 0 r n i n 9 Dan d C. H. S c h war t z "
"Air Pollutant Emissions From Coal-Fired Power Plants" Report
No.1". J. Air Pollution Control Assoc. 11 (9)" 353-62 (1964).
Cuffe. S. T.. R. W. Gerst1e. A. A. Orning" and C. H. Schwartz.
"Air Pollutant Emissions from Coal-Fired Power Plants. Report
No. 2". J. Air Pollution Control Assoc..12. (2).59-64 (1965).
Duprey~ R. L.. Compilation of Air ~ollutant Emission Factors.
Public Health Service. PB 190245 (1968).
Environmental Engineering. Inc.. and J. E. Sirrine Co.. Control
~~tmospheric Emis;;;ons in the Wood Pu1pinq Industry. PB
190~51. March (1970).
Federal Power Commission. Steam-Electric Plant Constr~ction Cost
and Annual Producti1n Expenses. Twenty-first Annual Supplement
(1968). .
Hangebrauck. R. p.. D. J. Von Lehmden. and J. E. Meeker. Sources
.2..f..1'olynuclear Hydrr)carbons in the AtmosPhere" U. S. Public
Hea1th Service. NAPCA Pub1. No. 999-AP-33. Second Printing.
(1970).
Johnson" A. J.. and G. H. Auth~Fu~l~ and "Combu~ti"on Handbook.
McGraw-Hi 1 1. N. Y. (1951).
Magi11p P. L.. and R. W. Beno1iel" "Air Pollution in Los Angeles
County. Contribution of Combustion Products". Ind. £ng. Chem.
ii. 1347-51 (1952).
McGra\'/o M. J. and R. L. Duprey." Air Polluta"n"t Emis"si'onFa"ctors"
Preliminary Document. Environmental Protection Agency (1971).
1II-93
-------
REFERENCES (continued)
Schurr, S., and B. C. Netschert~ Enerqy in the American Economy
1850-1975, Johns Hopkins Press, Baltimore (1960).
SmithQ W. S., Atmospheric Emissions From Fuel Oil Combustion,
An Inventory Guide, Public Health Service, PB 168874, November
(1962).
SmithD W. S., and C. W. Gruber, Atmospheric Emission from Coal
Combustionn An Inventory Guide, Public Health Service No.
999-AP-24 (1966).
u. So Bureau of Mines, Minerals Yearbook, (1968).
U. S. Bureau of Mines, Mineral Industry Surveys, Shipment~ 6f Fuel
Oil and Kerosene, (1970).
111-94
-------
C.
SOLID WASTE COMBUSTION
Black. R. J.. A. J. Muhich. A. J. Klee. H. L. Hickman. and R. D.
Vaughan~ The National Solid Wastes Survey - An Interim Report.
U. S. Department of Health. Education and Welfare (1968).
Boube1. R. W.. "Emissions from Burning Grass. Stubble and Straw".
Proceedings 61st Annual Meeting Air Pollution Control Assoc.
68-28. June (1968).
Business Week, "Special Report - Turning Junk into a Resource".
10 October (1970).
Combustion Engineering. Incorporated. Technical-Economic Study of
Solid Waste Disposal Needs and Practices, Public Health Service,
Publ. 1886 (1969).
Darley. E. F. et al. "Contribution of Burning of Agricultural
Wastes to Photochemical Air Pollution". J. Air Pollution
Control Assoc. li (12). 685 (1966).
Feldstein. M. et a1. "The Contribution of Open Burning of Land
Clearing Debris to Air Pollution". J. Air Pollution Control
Assoc. 11 (11). 542 (1963).
GCA Technology. Control Techniques for Polycyclic Orqanic Matter
Emissions, U. S. Department of Health. Education and Welfare.
Public Health Service. Draft Copy (1970).
Gerstle. R. W. and D. A. Kemmitz. "Atmospheric Emissions from
Open Burning". J. Air Pollution Control Assoc. 17 (5).324
(1967). -
Hall. E. P.o "Air Pollution From Coal Refuse Piles". Mining Congo
J. ~. 37-41 (1962).
Hickey. J. H.. "The Problem in Detail", in 'Solid Wa~tas Manaqe-
ment. Proceedings of the National Conference. University of
California. April (1966).
Johnson, H. C. and H. A. James. "Controlled Open Burning in the
San Francisco Bay Area". J. Air Pollution Control Assoc.' 20
(8).530-33 (1970).
Muhichg A. J.. A. J. K1ee. and P. W. Britton.' Preliminary Data
Analysis - 1968. Public Health Service Publ. 1867 (1968).
111-95
-------
National Emission Standards Study, Report to Congress by the
Secretary of Health, Education and Welfare (1970).
National Surve of Communit Solid Waste Practices, Public Health'
Service Publ. No. 1867 1968.
Niessen, W.R., S stems Stud of Air Pollution from Munici al
Incineration, Three Volumes PB 192 378, 379, 380, March 1970).
Rogus, C.H., IIRefuse Quantities and Characteristics", Proceedings
of the National Conference on Solid Wastes, Chicago, 17,
December (1963).
Shearon, W.H., "Citrus Fruit Processing", Modern Chemical Pro- .
cesses, Vol. 1,210-18, Reinhold Publishing Corp., N.Y. IT952).
Solid Wastes Management, Proceedings of the National Conference,:"
University of California, April (1966). '
Sussman, V.H. and J.J. Mulhern, IIAir Pollution from Coal Refuse
Disposal Areasll, J. Air Pollution Control Assoc. 14 (7), 279-84
(1964).
University of California, Comprehensive Studies of Solid Waste
Management, Public Health Service Publ. No. 2039 (1970).
U.S. Department of Agriculture, Forest Service, Annual Fire
Report for the National Forests. 1968.
UoS. Department of Commerce, Bureau of the Census, Statistical
Abstract of the United States, 1969, 90th Edition, Washington,
D . C., (1969).
U.S. Department of Health, Education and Welfare, Nationwide
Inventory of Air Pollutant Emissions, 1968, NAPCA Publ. 'AP-73
(1970).
Wadleigh, C.H., Wastes in Relation to Agriculture, Miscellaneous
Publication No. 1065, U.S. Department of Agriculture, 1968
Yo com, J. E ., G. M. He in, and H. W. N e 1 son, II A S t u dy 0 f E f flu e n t s
from Backyard Incineratorsll, J. Air Pollution Control Assoc.
i (2), 84 (1956).
111-96
-------
D.
PETROLEUM REFINING AND MARKETING
API Bulletin 2513. Evaporation Losses in the Petroleum Industry-
Causes and Control; American Petroleum Institute; New York (1959).
API Bulletin 2517. Evaporation loss from Floating Roof Tanks.
American Petroleum Institute. New York (1962).
API Bulletin 2518. Eva oration Loss from Fixed-Roof Tanks.
American Petroleum Institute. New York (1962 .
Baltimore Standard Metropolitan Statistical Area Emission Inventory.
Baltimore. Maryland (1968).
Bay Area Air Pollution Control District. Source Inventorv of Air
Pollutant Emissionso San Francisco Bay Area9 1969.
Danielson. J. A.. Air Pollution Enqineerin
Service Publ. 999-AP-40 1967).
Manual. Public Health
Dupreyo R. l.. Com ilation
Public Health Service, .
Louisville Metropolitan Area Air Pollutant Emission Inventory.
NAPCA Division of Air Quality and Emission Data. Durham. North
Carolina. January (1969).
Emission Factors.
Monterey-Santa Cruz County Unified Air Pollution Control District.
"Air Pollution in Monterey and Santa Cruz Counties". December
(1968).
National Petroleum Council. Report of the Committee on Petroleum
Storage CapacitYB 1962. National Petroleum Council. Washington.
D.C.o March 27 (1963).
National Petroleum Council. Current Key Issues Relating to Environ-
men tal Con s e r vat ion 0 The Gas and 0 il 'r n d u s t yo f e s'. ' . An' I n t e rim
Repo~t.. June 22 (1970).
National Emission Standards Study. Report to Congress by the
Secretary of Health. Education and Welfare (1970).
National Capital Interstate AQCR. Report for the Presentation of
Ambient Air Quality Standards for Carbon MonoxideD Hvdrocarbons
~Photochemica1 Oxidants. Maryland State Department of Health
and Mental Hygiene. October 20 (1970).
II I -9 7
-------
The Oil and Gas Journal, April (1968).
Public Health Service Publ. No. 763. Atmospheric Emissions from
Petroleum Refineries, U. S. Department of Health, Education,
and Welfare, Public Health Service (1960).
Seattle-Tacoma Air Pollutant
of Health. Education. and
Quality and Emission Data
December (1968).
Emission Inventory. U. S. Department
Welfare. Public Health Service. Air
Division. Durham. North Carolina.
Sto Louis Air Pollutant Emission Inventory. U. S. Department of
Health. Education. and Welfare. Public Health Service. Air
Quality and Emission Date Program. Durham. North Carolina.
August (1968).
U. S. Bureau of Mines. Mineral Industry Surveys, Crude Petroleum.'
Petroleum Productso and Natural Gas Liquids l1968).
Uo S. Bureau of Mines. Minerals Yearbook (1968).
U. S. Bureau of Mines. Motor Gasolines. Summer 1970 (197l).
U. S. Department of Commerce, Bureau of the Census, Statistical
Abstract of the United Stateso 1969, 90th Edition. Washington,
D. C. (1969).
U. S. Department of Commerce, 1967 Census of Business, Wholesale
Trade~ Petroleum Bulk Stations and Terminals, BC67-WS6 (1970).
111-98
-------
E.
SOLVENT EVAPORATION
Aerospace Industries Associationi Control of Orqanic Solvent
Em i s s ion sin to At m 0 s ph ere G T hi r d r n t e rim R e po r t G r~ C-:n (68 ) - 3
(1968).
Bay Area Air Pollution Control District. Source Inventory of Air
Pollution Emissionso 1969.
ChassD R. Lq P. S. Tow. R. G. LuncheD and N. R. Shaffero IITotal
Air Pollution Emissions in Los Angeles County II. J. Air Pollution
Control Assoc. ~ (5), (1960).
ChassG R. L., C. V. Kantero and J. H. E11iotto IIContribution of
Solvents to Air Pollution and Methods for Controlling Their
Emissionsll. J. Air Pollution Control Assoc. 11. (2),64-72 (1963).
DupreYD R. L.o Com ilation of Air Pollutant Emission 'Factors.
Public Health Service, PB 19 0245 1968).
Gadomski 0 R., Evaluation of Emissions and Control Technologies in
the Graphic Arts Industries, Graphic Arts Technical Foundation.
August (1970).
Haagen-Smit. A. J.. liThe Air Pollution Problem in Los Angelesll,
Eng. and Sci. Monthly (California Institute of Technology).
December (1950).
Haagen-SmitD A. J., IIChemistry and Physiology of Los Angeles
Smog II , Ind. Eng. Chern. 440 1342-6 (1952).
Ianetti 0 M.o IISolvents for Letterpress and Lithographic Inksll.
Am. Ink Maker. 51-2. 1450 October (1969).
Kearney 0 T. J. and C. E. Ki rcher 0 IIHow to Get the Mos t from
Solvent Vapor Degreasingllo Metal Progress, 87. April (1960).
Kirk-Othmer Encyclopedia of Chemical Technology. 2nd Edition.
Interscience. John Wiley and Sons (1969).
Lunche. R. G., A. Stein. C. J. Seymour. and R. L. Heimero IIEmissions
from Organic Solvent Usaqe in Los Angeles County II , Chern. Eng.
Progr. ~ (8). 375 (1957).
McGrawi M. J. and R. L. Duprey. IIAir Pollutant Emission Factors".
Public Health Service (1971).
National Institute of Dry Cleaning, Technical Bulletin. T-468,
Silver Spring. Marylando February (1971).
III-99
-------
Oil, Paint and DruQ Reporter~ Chemical Profiles, Schnell Pub-
lishing Company (1966) and (1968).
Renson, J. E., "Chemical Consumption Patterns in the Printing
Ink Industry". American Ink Maker. 58-60. 134. May (1968).
Salomon, G. and J. p. Petrone, "A Compilation of Solvents for
Flexographic and Gravure Inks", American Ink Maker. 28-38,
February (1969).
u. S. Bureau of Census. 1963 Census of Business. Wholesale Trade
Summary Statistics.
U. S. Department of Commerce, Bureau of the Census. Selected
Services. BC67-SAl (1970).
U. S. Department of Commerce, Census of Manufacturers. Paints
and Allied Products, August (1970).
U. S. Department of Health. Education and Welfare, "Nationwide
Inventory of Air Pollutant Emissions 1968". NAPCA Pub1. AP-73.
August (1970).
U. S. Tariff Commission, S~nthetic orvanic Chemicals~ U. S.
Production and Sales, 1 68, IC Pub. 327.
Wohlers. H. C. and M. Feldstein, "Investigation to Determine the
Possible Need for a Regulation on Organic Compound Emissions
from Stationary Sources in the San Francisco Bay Area", J. Air
Pollution Control Assoc. li (5). 226-9 (1965).
Yazujian. DOl "Chemicals in Coatings". Chemical Week. .l.Q2. (16).
35-51 (1971).
III-loa
-------
F.
CHEMICAL PROCESS INDUSTRY
Anonymous~ "Petrochemical Censusg Plant Listings"g Petro/Chem
Engineer 41 (3).13-41 (1969).
. -
Blandg W. F.. "Economic Climate Fails to Slow Petrochemicals"..
Petro/Chem Engineer 42 (4). 16-20 (1970).
. -
Bureau of the Budget. Office of Statistical Standards; 'Sta~dard
Industrial Classification Manual. U. S. Government Printing
Office (1957).
Fawcetto R. L.o "Air Pollution Potential of Phthalic Anhydride
Manufacture". J. Air Pollution Control Assoc. 20 (7). 461-5
(1970). -
Manufacturing Chemists' Associationo "Toward a Clean Environment.
a 1967 Survey of the Members of the Manufacturing Chemists'
Association" (1967).
Mencher. S. D.~ "Minimizing Waste in the Petrochemical Industry".
Chern. Eng. Prog. 63 (10). 80-8 (1967).
. -
Public Health Service Pub1. No. 763. Atmospheric Emissions from
Petroleum Refineries~ U. S. Department of Health. Education.
and Helfare. Public Health Service (1960).
U. S. Tariff Commission. "Synthetic Organic Chemicals. U. S.
Production and Sales. 1968". TC Publ. 327 (1970).
III-l01
-------
G.
OTHER INDUSTRIAL ~ROCESSES
Carbon Black
Allano D. l., "The Prevention of Atmospheric Pollution in the
Carbon Black Industry". Chern. and Ind., 1320-4, October 15
(1955).
Drogin, I., "Carbon Black", J. Air Pollution Control Assoc. 18
(4), 216-28 (1968). --
McGraw~ M. J., "Air Pollutant Emission Factors", Unpublished Draft
Copy, U. S. Department of Health, Education and Werf~re, August
(1970).
Shearon, H. H. Jr., R. A. Reinke, and T. A. Ruble, "Oil Black",
Ind. Eng. Chern. 44, 685-94 (1952).
. --
U. S. Bureau of Mines, Minerals Yearbook (1969).
Coke Industry
DancYD T. E., "Control of Coke Oven Emissions", Blast Furn. and"
Steel Plant ~ (ll), 811-19 (1970).
Full~rton. R. W.. "Impingement Baffles to Reduce Emissions From
Coke Quenching", J. Air Pollution Control Assoc. 17 (12),
801-9 (1967). --
GCA Technology, "Control Techniques for Polycyclic Organic Matter
Emissions", (1970).
Hangebrauck, R. P., D. J. Von Lehmden. and J. E. Meeker, "Emissions
of Polynuclear Hydrocarbons and Other Pollutants From Heat-
Generating and Incinerator Processes". J. Air Pollution Control
Assoc. li (7). 267 (1964).
Kirchoff. U., Aachener Bl. Aufbereiten Verkohen-Brikettereinen
11, 1-26 ~1962}.
Searlo T. D.. F. J. Cassidy. W. H. King, and R. A. Brm
-------
United Nations~ "Problems of Air and Water Pollution Arising From
the Iron and Steel Industry"~ ST/ECE/Steel/32 (1971).
U. S. Bureau of Mines. Minerals Yearbook (1969).
Varga. J. Jr.~ and H. W. Lownie. Jr.~ A Systems Analysis Study of
the Inteqrated Iron and Steel Industrv. Battelle Memorial Insti-
tute. PB 184577. May (1969).
Von Lehmden. D. V.. R. P. Hangebrauck. and J. E. Meeker. "Poly-
nuclear Hydrocarbon Emissions From Selected Industrial Pro-
cesses". J. Air Pollution Control Assoc. 15 (7). 306-12 (1965)0
Wood Pulpinq Industry
Kirk-Othmer, Encyclopedia of Chemical Technologyo Second Edition,
John Wiley and Sons (1969).
Miller. F. A., "Computer Approach Increases Yields", Chern. Eng.
Prog. ~ (12).62-7 (1968).
Walther. J. E.. and H. R. Amberg. "A Positive Air Quality Control
Program at a New Kraft Mill". J. Air Pollution Control Assoc.
~ (1).9-18 (1970).
Phthalic Anhydride
Fawcett. R. L.o "Air Pollution Potential of Phthalic Anhydride
Manufacture". J. Air Pollution Control Assoc. 20 (7). 461-5
(1970). --
Synthetic Rubber
Public Health Service. Division of Air Pollution. Louisville Air
Pollution Study. Cincinnati (1961).
111-103
-------
I V .
ADVERSE EFFECTS OF HYDROCARBON POLLUTANTS
Ao
INTRODUCTION
An analysis of the total picture of atmospheric
pollution by hydrocarbons requires consideration of the ad-
verse effects on human health and welfare. As a basis for
ranking of pollutants and sources establishing priorties
for future R&D,reviews of the literature were conducted to
assess the adverse effects of hydrocarbon emissions.
It ;s generally accepted that human health ;s the
predominant concern in efforts to control air pollution. As
an eKample, under the Clean Air Act as amende~, human health
was the only factor taken into account in setting primary
standards for ambient air quality. In this study, however,
attempts to define in quantitative terms the health effects
related to hydrocarbon air pollutants could not be supported
by data in the literature. Definition and quantification of
human health effects were precluded by lack of field or lab-
oratory studies under conditions of chronic exposure to low
concentrations of hydrocarbons and by complicating effects
of atmospheric reactions of hydrocarbons and their synergism
with other atmospheric contaminants.
The Air ualit Criteria for H drocarbons (U. S.
Department of Health, Education and Welfare, 1970 incorpo-
rated a review of the current state of knowledge relating
to adverse effects of hydrocarbons. The conclusions based
on that review were that hydrocarbons, per se, do not present
a significant health hazard in ambient air but should be con-
trolled or restricted on the basis of their contribution to
photochemical smog and the resultant adverse effects of the
smog products.
Literature data relating to the adverse effects
of hydrocarbon pollutants are reviewed in the following
sections, with particular emphasis on references that would
provide quantitative or semi-quantitative input for ranking
the pollutants and sources identified in Chapter III.
IV-1
-------
B.
HEALTH EFFECTS
1.
Introduction
In any review of air pollutants, a consideration
of human health effects is of prime importance. This section
presents an evaluation of the available information, pub-
lished and unpublished, on the contribution of hydrocarbon
emissions from stationary sources to adverse effects on human
health.
2.
Organic Solvents
A major contributor of hydrocarbon contamination
from stationary sourc~s is organic solvent evaporation.
Sources of these emissions include the manufacture of rubber
and plastic products, dry cleaning operations, chemical manu-
facturing, printing and the manufacture and application of
protective coatings. Sixteen of the industrially important
solvents are considered here (Table IV-l).
Solvents that exhibit essentially no buildup in
the body under conditions of occupational exposure, i.e.,
8 hours per day, 5 days per week, are listed below (Patty,
1962, Fairhall, 1969).
Methyl ethyl ketone
Methyl isobutyl ketone
Ethyl acetate
Acetone
Methylene chloride
Methyl chloroform (1,1,1-
trichloroethane)
Hexanes
Isopropyl alcohol
Ethyl alcohol
Of the remaining solvents in Table IV-l, toluene
and the xylenes are reported to undergo only slight buildup
in the human body at moderate levels of occupational exposure.
Chronic occupational exposure to 100 ppm toluene
produced no distinct symptoms or after effects. Although
some of the absorbed toluene is exhaled, the major portion
is oxidi2ed to benzoic acid, conjugated with glycin~, and
excreted as hippuric acid.
Although xylenes appear to have greater toxicity
than toluene, these substances tend to react similarly in the
body and are predominantly metabolized and/or excreted.
Methanol and n-butanol exhibit some buildup potential;
however, after occupational exposures of 100 ppm for 10 years,
no systemic effects were observed.
IV-2
-------
Table IV-l - Industrially Important Solvents
TLV(A) STL(B) LDSO(C) Smog Chamber(D) Vapor Chemical
Solvent (ppm) (ppm/min) (mg/kg-rats) Eye Response Irritant(E) Poison(F)
Petrpleum Nap~tha Va r. 500/30 119 (11) 0 1
*I~op~opanol . .; 400 400/30 5,840 318 (1) 1 2
Perch 1 or ethylene .100 200/30 1 2
*Ethanol . . 1000 5000/15 13,600 1 1
Trichlorethyle~e 100 .200/30 1 2
Toluene 200 600/30 106 (I I ) 1 2
*Acetone 1000 1000/30 10,700 303 (1) 1 0
Xylenes 100 300/30 111 (11) 1 2
*Methyl Ethyl Ketone 200 300/5 6,860 295 (1) 1 2
*l,l,l-Trichloroethane 350 1500/5 12,300 1 2
*Methylene Chloride 500 1000/30 2 2
Methanol 200 15,800 1 2
- *Ethyl Acetate 400 1000/15 5,000 1 2
<:
I *Methyl l-Butyl Ketone 100 400/5 2;080 213 (1) 1 ... 1
w *Hexanes 300-500 500/30 0 1
n-Butanol 100 150/30 4,360 1 2
A through F, see Notes on Table IV-l
*Compounds listed previously that e~hibit
no buildup in the body.
-------
NOTES.ON TABLE IV-l
A.
TLV (in parts per millio~) = Threshold Limits Values
published by American Conference of Governmental
Industrial Hygienists. They refer to airborne con-
concentrations of substances and represent conditions
under which it is believed that nearly all workers
may be repeatedly exposed day after day without ad-
verse effect.
B.
STL (Short Term Limits for Exposure to Airborne
Contaminants - A documentation - Pennsylvania De-
partment of Health, Division of Occupational Health).
500/30 = 500 parts per million for 30 minutes.
LD50 - The dosage which when applied to rats in a
single oral dose caused mortality in 50% of the ex-
posed animals.
C.
D.
Smog Chamber Eye Response - Expressed in arithmetic
mean seconds of exposure time to produce positive
feelings of irritation - a very subjective measure-
ment. (Bracketed numbers are relative ranking numbers,
with I expressing low reactivity, II intermediate re-
activity, and III high reactivity.)
Vapor Irritant - Based on the likelihood of developing
injury and on the severity and permanence of that injury.
E.
F.
o = chemicals that are non-volatile or the vapors
from which are non-irritating to-rhe eyes and
throat.
1 = chemicals that cause a slight smarting of the
eyes or respiratory systems (temporary) if
present in high concentrations.
2 = chemicals that cause moderate irritation
(temporary) such that personnel will find
high concentrations unpleasant.
Chemical Poison - Rates the hazard from chemicals that
enter the body through inhalation, oral ingestion, or
skin penetration to cause bodily harm.
o = no likelihood of ~roducing injury.
1 = minimum hazard. Includes most chemicals
having TLV above 500 ppm.
2 = some hazard, typically having TLV of 100
to 500 ppm.
IV-4
-------
Trichloroethylene is retained in the body for some
period after exposure, as shown by analysis of expired air
hours to days later. Experimentally, however, no demon-
strable intoxication or behavioral effect was evident in rats
exposed to 200 ppm for 7 hours per day, 5 days per week, for
a 6-month period.
Perch10rpethy1ene has been identified as an irri-
tating lachrymator with narcotic effects above 200 ppm but
moderate exposures are not expected to cause any permanent
injury. No pathological effects were observed in rats ex-
posed to 70 ppm for 8 hours per day, 5 days per week for 7
months.
Because petroleum naphtha is not a single compound
as are most of the other industrially important solvents, it
varies in composition and generalizations are therefore diffi-
cult. Although a significant portion of the mixture may be
aromatic in nature, these components are usually shown to be
substituted aromatics such as toluene, xy1enes, etc. The
classical studies of Gerarde (1962, 1963) show that changes
in chemical composition decrease the chronic toxicity potential
so markedly that no serious chronic health effects of the type
produced by benzene need be expected from inhalation exposure
to petroleum naphthas (Deichmann, 1969).
Table IV-1 summarizes available data regarding in-
dustrial solvents, including data on Threshold Limit Values
(TLV), short-term limits for exposure (STL), chemical poison
effects (National Academy of Sciences, 1970), smog chamber
eye irritation response (Levy and Miller, J970), and LD50*
(Fairha11, 1969; Deichmann, 1969; Patty, 1962; Sunshine, 1969).
Because of variations of conditions and considerations upon
which these data are based, this table can serve only as a
useful guide r~ther than as an authoritative ranking of the
health effects of the materials listed.
3.
Unsaturated Aliphatic Hydrocarbons
Frequent mention of the unsaturated aliphatic
hydrocarbons in ambient air warrants a review of their con-
tribution or potential contribution to adverse effects on
human health. Analyses of ambient air samples taken during
air pollution episodes in 1965 in Riverside, California,
identified the following unsaturated hydrocarbons (Stephens,
1967):
*LD50
=
Lethal dosage effecting 50 percent mortality in
exposed subjects.
IV-5
-------
Ethylene concentrations ranged from 1.1 to 15.4
parts per billion.
Propylene concentrations ranged from 0.3 to 3.0
parts per billion.
1-Butene concentrations ranged from 0.2 to 0.6
parts per billion.
Isobutene concentrations ranged from 0.3 to 1.0
parts per billion.
1-Pentene concentrations ranged from <0.1 to 0.4
parts per billion.
2-Methy1 butene-1 concentrations ranged from 0.2
to 0.8 parts per billion.
Acetylene concentrations ranged from 1.0 to 31.0
parts per billion.
Methyl acetylene concentrations ranged from <0.1
to 1.4 parts per billion.
The low concentrations reported were detected in
relatively clear weather, while the high concentrations were
detected during an unusually heavy haze. These and other
potentially important unsaturated aliphatic hydrocarbons
are discussed below.
a. 01efins - Industrially important olefins are
ethylene, propylene, and buty1enes formed as by-products of
the carcking of petroleum fractions and used mainly for polymer
synthesis. The lower members of the olefin series (2, 3 and
4 carbon chains) are weak anesthetics or simple asphysiants.
As the chain length increases, anesthetic potency increases
also.
No threshold limit values (TLV) have been established
for these olefins by the American Conference of Governmental
Industrial Hygienists (1970 listing). Ethylene, the only
olefin listed, is designated ~ simple asphyxiant. A TLV is
not recommended for each simple ashpyxiant because the limit-
ing factor is the available oxygen. Gerarde (1962) has
suggested 1000 ppm as the threshold limit value for these
olefins.
Effects on the health of workmen exposed to low con-
centrations of olefins for prolonged peri~ds or to higher con-
centrations for relatively short periods of time do not appear
to warrant serious consideration unless the olefins encountered
in sufficient concentrations to cause asphyxia.
IV-6
-------
While olefins in themselves do not appear to be
injurious to human health at concentrations usually found
in ambient air, they are instrumental as reactants in the
atmosphere in the formation of photochemical smog, via
peroxyacyl nitrates (PAN), which are reasonably well es-
tablished as lachrymators (Christman, 1969).
b. Diolefins - Butadiene and isoprene (2-methyl-l,
3-butadiene) we~e included as representative diolefins in
our review. Experiments on humans and on laboratory animals
have demonstrated that 1,3 butadiene has a low order of toxicity
and probably has no cumulative effects (API, 1959). The TLV
established by ACGIH (1970) is 1000 ppm. In high concen-
trations, 1,3-butadiene is an anesthetic which can cause
respiratory paralysis and even death.
Isoprene appears to be a more potent anesthetic
than butadiene, but pharmacologically is similar. No TLV
has been established (Gerarde, 1962).
c. Acetylene - No evidence of deleterious effects
on health was apparent in repeated exposure to tolerable
levels of acetylene. ACGIH (1970) lists acetylene as an in-
ert gas, a simple asphyxiant. On the basis of the lower ex-
plosive limit, however, acetylene conce~trations should not
be permitted to exceed 5000 ppm in t~ occupational environ-
ment.
Methyl acetylene is listed as a simple anesthetic
and in high concentrations as an asphyxiant. The recom-
mended TLV is 1000 ppm (Sax, 1963).
d. Gasoline Evaporation - Gasoline is a complex
mixture of hydrocarbons, princip~l'y paraffins, naphthenes,
olefins, and aromatics, as well as dyes, inhibitors and
antiknock agents (API Toxicological Review, 1967). Since
the composition of gasoline varies greatly depending on
blending, seasonal requirements, and intended use, a single
TLV for all types of gasoline no longer applies. To de-
termine which TLV is applicable for a particular gasoline,
its aromatic hydrocarbon content must be considered (Docu-
mentation of TLV, 1971).
Severe reactions, such as depression of the central
nervous system, anorexia, vomiting, narcosis and even death,
can be produced by inhalation of gasoline vapors at very high
concentrations. The most severe cases of gasoline vapor in-
halation reported most frequently involved persons entering
tanks containing nearly saturated vapor concentrations of
gasoline.
IV-7
-------
Reduced availability of oxygen may also play an
important role in effecting asphyxia in tank operations.
In repeated work day exposures to low concen-
trations of gaso1ine,effects have been varied and vague.
Considering periods of up to 10 and 12 years in some instances,
Mach1e (1941) concluded that no signs of gasoline intoxication
were evident in filling station attendants, garage mechanics
and tank wagon drivers. Barrel fillers, who are exposed in
their work day to concentrations that might be intolerable to
many people often complained of anorexia, nausea and nervous-
ness. Eye irritation produced from exposure to low concen-
trations of gasoline vapors (200 ppm and 500 ppm) was reported
by both Davis et a1 (1960) and Drinker (1943). Gerarde (1962)
suggests that no conclusive evidence can be found for harmful
effects due to repeated exposure to low levels of gasoline
vapors in work atmospheres.
4.
Aldehydes
Aldehydes have been reported extensively in the
literature as components of photochemical smog, the major
effects on humans being reported as irritation of the eyes,
respiratory tract and skin. Fassett (1963) indicates that
definitive cumulative organic damage to tissues is not com-
monly found, possibly because the aldehydes are readily meta-
bi1ized in the body. The irritant effect decreases with in-
creasing molecular weight within a given aldehyde series,
and the toxicity generally decreases as the chain length in-
creases. .
Most information on aldehyde toxicity is related
to the effects of actue exposures. There is a ~earth of
data pertaining to long-term exposures to low concentrations
to aldehydes which would correspond to the conditions of
community or ambient air pollution.
Fassett reports (1963) that no discomfort is noted
below 2 ppm of formaldehyde during an 8-hour exposure. At
concentrations above 2 but below 4 ppm for a similar period,
a mild tingling sensation may be noticed in the eyes, nose
and pharynx. Concentrations of 10 ppm cause profuse lacri-
mation and can be endured for only a few minutes. Above 10
ppm, breathing becomes difficult. However, when the subject
is removed from such exposure, the symptoms subside; lacri-
mation quite promptly and respiratory irritation more slowly.
According to Sax (1963), repeated exposures to formaldehyde
vapors may result in chronic irritation of the eyes, nose,
and upper respiratory tract.
IV-8
-------
A recent report gave the average concentration of
formaldehyde in the Los Angeles ambient air at less than 0.15
ppm, much lower than Fassett's reported level of discomfort.
However, in simulated atmospheric research, Schu~k et al (1960)
found that the human eye can detect and respond to as little
as 0.01 ppm of formaldehyde. They also observed the same
irritation intensity in most subjects exposed to concentrations
of 0.05 and of 0.5 ppm.
In a study of the photo-oxidation of hydrocarbons,
researchers at Stanford Research Institute found that for-
maldehyde and acrolein accounted for the majority of the eye
irritation produced by photochemical products, but the concen-
trations of these aldehydes required to produce eye irritation
were much in excess of measured atmospheric concentrations.
For example, again in the Los Angeles ambient air, only about
0.02 ppm of acrolein was found. Smith (1962) states that con-
centrations of acrolein of 0.25 ppm can cause moderate irri-
tation of the eyes and nose in exposures as short as 5 minutes.
Fassett (1963) reports no known cases of chronic toxicity, but
Sax (1963) says that inhalation may cause an asthmatic reaction
and that repeated skin contact may produce chronic irritation
and dermatitis.
In animal experimentation the effects of exposure
to low concentrations of both formaldehyde and acrolein were
found to be reversible when the animals were returned to clean
air.
Fassett (1963) provides a tabulation of the toxicity
of aldehydes to animals via inhalation (Table IV-2).
The information and data provided in th~ literature
indicate that aldehydes constitute little, if any, threat to
health at concentrations found in ambient air, but rather are
noticeable irritants. Before it can be stated with certainty,
however, such a conclusion must be supported by data obtained
both in emission inventories and in long-term low-level exposures.
A recent lit~rature review concerning aldehydes (Stahl, 1969)
covers the subject reasonably comprehensively and cites the
same gaps in available information as stated above.
5.
Aromatics
Lonneman, et al (1968) collected 126 samples in down-
town Los Angeles over a 26-day period from about 7:00 A.M. to
3:00 P.M. to determine aromatic hydrocarbon concentrations.
The average concentration of total aromatics was 0.106 ppm by
volume, the highest reading being 0.33 ppm.
IV-9
-------
TABLE IV-2 - TOXICITY OF ALDEHYDES TO ANIMALS VIA INHALATION
Compound Species .2Q!!! Time,hr Mortality
Saturated Aliphatic Aldehydes
Formaldehyde Rat 250 4 LC50a
Rat 815 0.5 LC50
Cat 650 8 LC approx
Cat 200 3.5 A1~Osurvived
Acetaldehyde Rat Sat vapb 2 min 6~JOOd
Rat 16,000 4
Rat 20,000 30 min LC10
Cat 13,600 0.25 1 /
Cat 4,100 3-5 0/1
Cat 256 5 0/1
Propiona1dehyde Rat 8,000 4 5/6
Rat 60,000 0.3 3/3
Rat 26,000 0.5 LC50
Ethoxypropiona1dehyde Rat 500 4 6/6
a8-Dich1oropropion- Concd vapc 2 min 6/6
aldehyde Rat 16 '4 4/6
n-Butyra1dehyde Rat 8,000 4 1/6
Ra t 60,000 0.5 LCSO
Isobutyraldehyde Rat 8,000 4 1/6
8-Hydroxybutyraldehyde Rat 4,000 4 2/6
(aldol, acetal dol) Rat Sat yap 0.5 No deaths
n-Valeraldehyde Rat 48,000 1.2 3/3
Rat 1 ,400 6 0/3
2-Methylbutyraldehyde Rat 67,000 0.3 3/3
Rat 3,800 6.0 0/3
Ra t 1 ,043 6.0 0/3
n-Hexa1dehyde Rat Concd yap 4 0/6
(hexanal ) Rat 2,000 4 1/6
2-Ethylbutyra1dehyde Rat Concd yap 5 min 0/6
Rat 8,000 1 5/6
2-Ethylhexy1adehyde Rat 25,000 13 min 3/3
(a-ethylcaproa1dehyde) Rat 4,000 4 l/6
Rat 2,000 23 min 3/3
Rat 145 6 0/3
IV-10
-------
Table IV-2 (Continued)
Compound
Species
1?.E.!!l
Unsaturated Aliphatic AldehYdes
Acrolein
Methacrylaldehyde
(Methacrolein)
2-Ethyl-3-propyl
acrolein
Crotonaldehyde
(a-methyl acrolein)
Methyl-a-ethyl acrolein
(2-methyl-2-penten-l-al)
Rat
Cat
Cat
Cat
Rat
Rat
Rat
Rat
Rat
Rat
8
690..1,lSO
18-92
11
130
2S0
Con cd vap
Concd vap
1 ,400
2,000
Aliphatic Dialdehydes
Succinaldehyde
(2S% in H20)
Hexa-2,4-(1ienal
3-Methyl glutaraldehyde
a-Hydroxyadipaldehyde
Rat
Rat
Rat
Rat
Rat
aLCSO ~ Lethal coneentrqtion at
bSat vap ~ Saturated vapor
cConcd vap = Concentrated vapor
d ~ fraction denotes deaths per
IV-11
Concd vap
(ca. lS,OOO
ug/liter)
2,000
Coned vap
Concd vap
Concd vap
Time,hr
4
2
3-4
3-10
30 min
4
8
1 min
30 min
4
6
4
8
6
8
Mortality
1/6
3/3
0/2
0/2
LCSO
5/6
0/6
0/6
LC50
3/6
0/3
1/6
0/6
0/3
0/6
which 50% mortality is observed
number of subjects
-------
Benzene, toluene, and the xylene isomers made up
0.082 ppm of the average of 0.106 ppm, or 77 percent of the
total aromatics measured. Toluene was the single most
abundant aromatic hydrocarbon, with concentrations about
twice as high as either benzene or m-xylene, the next highest.
The remaining 23 percent of the total aromatic fraction con-
sisted of ethylene; ethyl toluenes; trimethyl-propyl-, and
butyl-benzenes; and other unidentified higher homologs.
The American Conference of Governmental Industrial
Hygienists (ACGIH) has recommended 200 ppm as the threshold
limit value for toluene, an acceptable limit for exposures
not exceeding 8 hours daily. Von Oettingen (1940) reports
that at 200 ppm subjects experience a slight but definite
impairment of coordination and a slowing of reaction time,
suggesting that a lower limit might be desirable.
Chronic benzene poisoning results from repeated
or continuous exposure to relatively low concentrations of
benzene vapor. The level and degree of exposure necessary
to produce poisoning vary widely. There are at least two
well-authenticated cases of poisoning by repeated exposures
to 75 ppm. Inhalation of 50-150 ppm of benzene for 5 hours
caused slight headache, weariness, and lassitude (API, 1960).
Benzene is relatively insoluble in body fluids and
tissues. Practically complete elimination of benzene from
the blood occurs within a few minutes after exposure is
terminated. Benzene is eliminated from the body via the
lungs and the kidneys.
The ACGIH recommended threshold limit value for
benzene is 25 ppm for an 8-hour daily exposure.
Chronic' exposure to xylene may produce symptoms
similar to those of chronic toluene poisoning. There may
be systemic symptoms indicating nervous system involvement.
Serious blood changes are not usually encountered. Experi-
mental subjects exposed to various concentrations of xylene
considered 100 ppm the highest concentration tolerable for
an 8-hour exposure (API, 1960).
The ACGIH recommended threshold limit value for
xylene is 100 ppm for an 8-hour daily exposure.
Here again, with aromatics as with aldehydes, one
finds a paucity of data on low-level exposures. It appears
that community levels such as measured in the Los Angeles
Basin present no cause for concern regarding health hazards,
but additional information is required to substantiate a
definitive statement in this regard.
IV-12
-------
Table IV-3 provides comparative effects of acute
and chronic exposures to aromatic hydrocarbon v~pors in air
(API,1960).
6.
Polycyclic Aromatic Hydrocarbons
Polycyclic aromatic hydrocarbons are universally
present in the atmosphere (Coffin, 1971; Sawicki, et a1, 1960;
Heimann, 1967; Katz, 1962). A quantitative tabulation of the
polycyclic aromatic hydrocarbon (PAH) content of the air of
14 American cities is presented in Table IV-4. The stability
of PAH is consistent with their survival in the atmosphere
for sufficient periods of time to be respired by the general
population (Fa1k et a1, 1956).
The literature on PAH emissions shows that they are
products of incomplete combustion of carboniferous fuels and
evaporation of coal-tar pitch products (Coffin, 1971; Sawicki,
1962a; Faith, 1960). Qualitative identification of PAH was
made in stack gases from pulp mills; in the air surrounding
iron and steel works, heat generation, and incineration pro-
cesses; and in the fumes from coal-tire pitch operations
(Hendrickson et al, 1963; Tanimura, 1968; Hangebrauck~et a1,
1964; Sawicki et ~l, 1962b).
A nationwide inventory in 1968 of air pollutant
emissions showed that industrial process losses accounted
for 4.6 mn1ion tons of the approximately 32 million tons
of hydrocarbons emitted annually in the United States (USDHEW,
1970). Of this national total, 60 percent was emitted in
urban areas. Higher concentrations of PAH in urban areas
have been verified worldwide. In addition to variances in
PAH concentrations in urban and rural areas, differences in
PAH levels also stem from geographic location and climatic
conditions. PAH concentrations are generally greater in
winter, possibly because of seasonal differences in fuel use
practices and combustion efficiency (Sawicki et al, 1962a).
The benzo(d)pyrene coritent of particulate air pollutants also
differs markedly in eastern and western cities, the western
cities showing much lower concentrations (Katz, 1962).
Some polynuclear aromatic hydrocarbons exhibit
carcinogenic activity, i.e., chrysene, benz(~)pyrene, .
benzo(e)pyrene, and benzo(g,h,i)perylene. Of these, benzo
(~~pyrene is the strongest carcinogen as well as the most
stable. Numerous experimental studies on rats and mice
have demonstrated a possible carcinogenic hazard related
to some pollutants in the atmosphere. Numerous investigators
IV-13
-------
TABLE IV-3 - Comparative Effects of Acute and Chronic Exposures to
Aromatic Hydrocarbon Vapors in Air
Compound
Subject
Benzene
Man
Man
Mice
ToluenG
Man
Mice
Mice
Man
Mice
Rats
Man
Man
Man
Man
Man
Man
Man
Styrene
Mice
Mice
Mice
Rat
r~an
Guinea
pig
Rat
Rabbit
Guinea
pig
Guinea
pig &
rat
Guinea
pig &
rat
Guinea
pig &
rat
Mice
Xylene
Man
Rat &
Rabbit
Rabbit
Mice
tHce
Mice
Rat
3,000
4,700
7,400
7,500
14,100
17,800
20,000
50-100
200
300
400
600
600
2,700
6,700
9,500
13,500
100
650
1,300
1,300
1,300
2,500
5,000
10,090
174
200
690.
1,150
4,699
9,200
12,650
17,250
Concentration
ppm mg/m3
25
100
370
80
319
1,180
9,570
14,993
23,606
23,925
44,979
56,782
63,800
188-377
753
Da i 1.Y
Exposure,
hr
Expos u re,
days
1,130
1,506
2,259
2,259
10,166
25,226
35,768
50,828
418
2,714
5,428
5,428
5,428
10,438
20,875
41,750
755
868
2,995
4,991
20,394
39,928
54,901
74,865
Acute.
Acute
Acute
Acute
Acute
Acute
Acute
Acute
Acute
Acute
Acute
8
8
8
3
8
1
1
1
. 1
Acute
Acute
Acute
Acute
Acute
8
8
8
8
180
180
180
180
8
1/2-1
Acute
Acute
8
8
130
55
Acute
Acute
Acute
Acute
.Acute refers to a single short term exposure, not repeated
IV-14
Effect
Threshold limit value (TLV)
Mucous membrane irritation
Threshold for affecting the central
nervous system
Endurable for 1/2-1 hour
Prostration
LC50
Dangerous after 1/2-1 hour
LClOO
LC100
Fatal after 5-10 minutes
No effect
Mild fatigue, weakness, confusion,
skin parethesias
Symptoms more pronounced
Also: mental confusion
AlsD: nausea, headache, dizziness
Also: loss of coordination, staggering
gait,pupils dilated
Prostration'
LC50
LC100
LC100
Threshold limit value (TLV)
No effect
Eye and nasal irritation only
Eye and nasal irritation only
102: deaths
Some fatalities; varying degree of
weakness, stupor, incoordinaUon,
tremor, unconsciousness (in 10 hrs)
Unconsconsciouness in 1 hour
Unconsciousness in 10 minutes;
deaths in 30-60 minutes
Threshold for affecting central
ne rvous sys tem
Threshold limit value (TLV)
No hematological effects
Decreased leukocytes and red
cells; increased platelets
Prostration
LC50
LC100
LC100
blood
-------
Table. IV-4 - Polycyclic Hydrocarbon Content of the Air for Selected Cities
City
Winter, 1959
( ug!l 000 m3>. Month BghiP* BaP BeP BkF P Cor Per An th Total
At 1 an, t a Feb. 8.9 7.4 4.7 6.0 6.0 4.3 1.1 0.52 38.92
Birmingham . Feb. H~. 0 25.0 10.0 13.0 17.0 3.5 5.5 2.2 94.2
Detroit ~~b. .33.0 ~1..0 23.0 20.0 25.0 6.4 6.0 2.0 146.4
Los Angeles Feb. 18.0 5.3 8. 1 5.7 6.0 12.0 1.6 0.16 56.86
Nashville J an. 17.0 25.0 14.0 15.0 3.0 4.6 4.4 1.8 111.8
New Orleans Feb. 7.3 4. 1 6.4 3.9 2.3. 27.0 0.8 O. 10 27.6
.....
<: San Franci s'co Jan. 7.5 2.3 2.9 1.7 1.9 4.9 0.34 0.10 21.64
I
-
(J1 Seattle Jan.-Mar. 14.0 9..0 8. 1 6.7 15.0 2.4 1.0 56.2
Sioux Falls Jan.-Mar. 8.3 4.0 2.7 4.0 3.7 0.82 0.22 23.74
South Bend Jan.-Mar. 12.0 16.0 14.0 10.0 32.0. 4.7 3.0 1.5 93.2
Wheeling Jan.-Mar. 14.0 21.0 13.D 22.0 3.8 2.6 1.4 77.8
Y oun gs tOVIn Jan.-Mar. 22.0 28.0 18.0 34.0 3.4 7.4 3.4 116.2
*BghiP = Benzo(g.h,i)perylene; BaP = Benzo(a)pyrene; BeP - Benzo(e)Phrene: BkF = Benzo(k)
fluoranthene; P = Pyrene; C = Coronene; Per = Perylene; Ant = Anthanthrene
Sawicki, 1962a
-------
extracted tarry substances from air pollutants, and by em-
ploying a IIskin paintingll method on mice, produced tumors
which they attributed to the benzo(a)pyrene content (Shabad,
1960; Sawicki et a1, 1962a; Gurinov et a1, 1969). Thus far,
experimental production of long tumors by inhalation methods
in C57 black mice has been unsuccessful. A reason postulated
for this discrepancy is that carcinogenic hydrocarbons are
biologically ineffective when adsorbed on soot particles
within a given size range, becoming effective only upon elution
or displacement from the soot particles. The sebaceous
secretions on the skin may provide the liquid solvent to elute
the carcinogenic hydrocarbons from soot, while in the lungs no
such eluting agent is present.
It has been further postulated (Katz et a1, 1954)
that if gasoline vapors are simultaneously inhaled they may
act as elements which extract the carbinogenic hydrocarbons
from the soot so as to allow them to act on the respiratory
mucosa. Both Saffiotti et a1 (1965) and Gross et a1 (1969)
induced lung tumors in small laboratory animals with carcin-
ogens adsorbed on hematite (iron oxide, Fe203)' Inhalation
experiments conducted on mice (Gardner et a1, 1970) exposed
to both ambient and filtered Los Angeles air resulted in no
difference in histologic appearance and no increase in in-
cidence in lung tumors;however, prolonged exposure of mice
to ambient Los Angeles air suggested increased susceptibility
to pulmonary infection.
Concentrated efforts have been directed toward
studies of the acute effects of air pollution but apparently
not on the long-term effects of exposure to polluted atmos-
pheres. Fa1k and Kotin (1955) analyzed lungs of typical
urban dwellers and recovered approximately one gram of soot
(average per dweller). This failed to reveal the usual
aromatic hydrocarbons when subjected to spectrophotometric
analysis. They attribute the apparent elution of carcinogenic
polynuclear aromatic hydrocarbons from soot to intercellular
protein activity (Fa1k et a1, 1960; Kotin and Fa1k, 1959).
Even though carcinogenic hydrocarbons have been detected in
urban air the .1iterature provides no clear indication that
their presence is significant though their interaction with
biological and physiological substances might be the mec-
hanisms of carcinogenesis (Anderson, 1967). In reviewing
the .1itera.ture, we found no clear-cut correlations between
the effects of pollution by polynuclear aromatic hydrocarbons
and production of lung tumors in man.
IV-1€
-------
7.
Pollutants Created by Atmospheric
Hydrocarbon Reactions
The major concern with regard to health effects
of gaseous hydrocarbon emissions is due to their atmospheric
interactions to form photochemical smog. Photochemical re-
actions to form smog are complex but can be generally described
as reactions of reactive hydrocarbons with oxides of nitrogen
in the presence of ultraviolet radiation.
It is important at this point to differentiate be-
tween primary and secondary pollutants. Primary pollutants
are those emitted directly into the atmosphere such as the
hydrocarbons and oxides of nitrogen; secondary pollutants re-
sult from chemical reactions of primary pollutants emitted
into the atmosphere and exposed to sunlight. Examples of
secondary pollutants are the peroxyacyl nitrates (PAN), ozone,
aldehydes, and aerosols (Pitts, 1969). Some pollutants, such
as aldehydes, may be either primary or secondary pollutants.
a. Aldehydes - Photooxidation of hydrocarbon-
nitrogen oxide mixtures can result in the formation of low
molecular weight aldehydes, principally formaldehyde and
acrolein (Table IV-5). The effects of aldehydes on humans,
previously discussed (Sect. B.4), include the highly irri-
tating effects of formaldehyde and acrolein vapors on the
mucous membranes of the eyes, nose, and throat (Fassett,1963;
Renzetti and Bryan, 1961). Although concentrations of these
two aldehydes are generally higher than those of other alde-
hydes present in the air, neither substance exists in ambient
air in sufficient concentration to account for the degree of
eye irritation experienced.
b. Aerosols - Biological and biochemical effects
produced with organic aerosols are governed primarily by
particle size, which determines their site of deposition,
retention, and metabolic fate in human lungs. Ordinarily,
particles lO~ or larger are prevented from ent~ring the depths
of the lungs because of the filtering action of the upper
respiratory tract and its airways. Solubility of particles
in water is another factor affecting potential biological
effects. Soluble aerosols in adequate concentration may be
expected to produce an immediate effect; insoluble particles
must first be deposited and retained in the body, later to
be acted on by tissue fluids and chemically altered before
their effects occur (Hatch and Gross, 1964). Thus, the effects
of particulate pollutants may be immediate and acute, as with
sulfuric acid mist, or delayed and chronic, as with carcino-
gens such as the polynuclear aromatic hydrocarbons (Kotin,1964).
IV-17
-------
Table IV-5 - Aldehyde Yields From Photooxidation of
Hydrocarbon-Nitrogen Oxide Mixtures
Moles of Aldehyde per Mole of Hydrocarbon*
Aliphatic
Hydrocarbon Formaldehyde Acrolein Aldehydes
Ethylene 0.3-0.4 0.3-0.4
Propylene 0.4 0.6-0.8
l-Butene 0.4 0.65
Isobutene 0.5-0.7 0.5-0.7
Trans-2.butene 0.35 1.25-1.55
1,3-Butadiene 0.3-0.5 0.2-0.3 0.5-0.8
l-Pentane 0.5 1.0
2-Methyl-2-butene 0.3-0.5 0.8-1.2
Cis-2-hexane 0.9-1.0
Toluene 0.05 o. 1
Xylenes 0.15-0.2 0.2-0.3
1,3,5-Trimethylbenzene. 0.15-0.2 0.2-0.3
*Based on hydrocarbon initially present as reactant.
Altshuller, A.P., Air and Water Poll. Int. J. lQ.:713-733,1966
IV-18
-------
Both Saffiotti (1965) and Gross (1969) demonstrated the
ability of inert dusts to transport organic pollutants, in
these instances polynuclear aromatic hydrocarbons, into the
lungs of small laboratory animals by intratracheal ad-
ministration.
c. Photochemical Oxidants - Of the gaseous pollu-
tants resulting from photochemical reaction, ozone is perhaps
the best understood and most abundant. It is formed by a
cycle that involves the oxides of nitrogen, atmospheric oxygen
and hydrocarbons. The latter allow ozone to accumulate by
reacting to scavenge the nitric oxide, which would reduce the
ozone (Table IV-7). This process i~ illustrative of an antag-
onistic action between two gases (StepheITs, 1969; Stokinger,
1968).
Peroxyacy1 nitrates (PAN), another important group
of oxidants in ambient air, result from reactions between
oxides of nitrogen and organic pollutants. PAN has been
identified as an effective irritant in simple synthetic
photochemical systems; however," it is not present in ambient
air in concnetrations high enough to account for the degree
of eye irritation experienced" (Renzetti and Bryan, 1961).
Smith (1965) suggested that exposure to peroxyacy1 nitrate
results in increased oxygen uptake by human subjects when ex-
posed to the additional stress of exercise. Although sub-
stantiative evidence is necessary before any conclusive state-
ment can be made in this regard, exercise has been proscribed
in certain communities and schools during "smog a1ert" situ-
ations as a precautionary measure.
Investigation of the biological effects of photo-
chemical oxidants as pollutants is complicated by their
atmospheric instabi1ity,the necessity of employing unrealistic
concentrations to demonstrate biological effects, and finally
the extreme difficulty of combining gases quantitatively in
attempts to simulate atmospheric conditions (Kotin and Fa1k,
1964).
Gaseous oxidation reactions constitute one of the
most important types of reactions in community atmospheres.
Sudden increases in concentrations of oxidants are related
to eye irritation. To date, however, we find no good data
to indicate whether oxidants, nitrogen dioxide, and/or
hydrocarbons at existing community air levels play any role
in the causation of respiratory disease. It is difficult
to determine the importance of synergistic effects in actual
atmospheric conditions since in many cases the individual
IV-19
-------
Table IV-6 - Atmospheric Reactions of Pollutants
With Ozone*
Reaction
Rate (1 ppm of pOllutantst
NO + 03 --+
N02 + 03 --+
502 + 03 --+
H25 + 03 --+
CO + 03 ----..
N02+ 02
N03 + 02
503 + 02
502 + H20
C02 + 02
Very fast, 170 ppm/h
Fast, 0.3-0.6 ppm/h
Very slow, <0.001 ppm/h
Fast, 2 ppm/h
No measurable rate at ambient
temperature
Paraffins + 03 --. products
Alkylbenzenes + 03 ~ products
Acetylene + 03 --. products
Very slow
Very slow, <0.001 ppm/h
Very slow- <0.001 ppm/h
Ethylene + 03 --. products
Propylene + 03 --. products
Slow, 0.02 ppm/h
Slow, 0.07 ppm/h
Olefins with internal double
bonds + 03 ~ products
Fast, 0.2-2 ppm/h
*Altshuller, 1969
IV-20
-------
effects of pollutants may be either intensified or are can-
celled by interaction. However, no one pollutant can be
excluded from consideration because it alone has not been
found to produce adverse health effects at a given concen-
tration.
8.
Odors and Odorants
Odors may be one of the earliest manifestations of
air pollution since some odors are detectable even at ex-
tremely low concentrations. Any substance or mixture of
substances in the vapor or gaseous phase that stimulates
the sensation of smell is an odorant; an odor is defined
as the sensation perceived as a result of olfactory stimulus
(Duffee, 1968). Whether an odor is considered offensive may
depend on prior psychological conditions of the receptor as
well as on the nature and strength of the odor.
This brief section in no way represents an ex-
haustive reservoir of the available literature on odors, but
merely intends to include the subject as a part of the at-
mospheric health complex.
a. Odorant Sources - Offensive odors emanate
from a variety of industrial, domestic and natural sources.
These odors most frequently reported by city bureaus over
several years are listed in Table IV-7, as reported by Kerka
and Kaiser (1958). The most offensive odors originate from
plants that produce low molecular weight sulfur and nitrogen
compounds, such as mercaptans, hydrogen sulfide, ammonia
and amines. Major sources frequently reported by city bureaus
are animal rendering plants, petroleum refineries, chemical
and metallurgical plants, and sewers and sewage treatment
plants. Kraft mills are not frequently mentioned, probably
due to their location in non-urban areas. On occasion, odors
can be detected in a 10 to 20 mile radius from the source
(Sullivan, 1969).
b. Characteristics of Odors - The quality of an
odor is a subjective characteristic evaluated with descriptive
words that associate an odor with familiar materials, i.e.,
fruity, fishy, rank, etc. The odor qualities of some selected
odorants are listed in Table IV-B.
Odor intensity and pervasiveness are generally
measured by the response of a selected group of human sub-
jects or "odor panel ". Laboratory studies results are gen-
erally reported as threshold levels (in ppm) at which the
odor is barely perceptible by the human subjects.
IV-21
-------
Table IV-7 - Most Frequently Reported Odor Sources.
Source of Odor
Number Reported
Animal Odors
Meat packing and rendering plants
Fish oil odors from manufacturing plants
Poultry ranches and processing
Odors From Combustion Processes
12
5
4
Gasoline and diesel engine exhaust
Coke oven and coal gas odors (steel mills)
Poorly adjusted heating systems
Odors From Food Processing
10
8
3
Coffee roasting plants
Restaurants
Bakeries
8 .
4
3
Paint and Related Industries
Manufacturing of paint, lacquer, and varnish
Paint spraying
Commercial solvents
8
4
3
General Chemicals Odors
Hydrogen sulfide
Sulfur dioxide
Ammonia
7
4
3
General Industrial Odors
Burning rubber from smelting and debonding
Odors from dry cleaning shops
Fertilizer plants
Asphalt odors (roofing and street paving)
Asphalt odors (manufacturing)
Plastic manufacturing
Foundry Odors
5
5
4
4
3
3
Coke oven odors
Heat treating, oil quenching, and pickling
Smelting
4
3
2
IV-22
-------
Table IV-7 (Continued)
Source of Odor
Number Reported
Odors From Combustible Waste
Home incinerators and backyard trash fires
City incinerators burning garbage
Open-dump burning
Refinery Odors
4
3
2
Mercaptans
Crude oil and gasoline
Sulfur
3
3
1
Odors From Decomposition of Waste
Putrefaction and oxidation (organic acids**)
Organic nitrogen compounds (decomposition of protein**)
Decomposition of lignite (plant cells)
3
2
1
Sewage Odors
City sewers carrying industrial waste
Sewage treatment plants
3
2
*Kerka, 1958
**Probab1y related to meat processing plants
IV-23
-------
Table IV-8 - Odor Qualities of Selected Odorants*
Compound
Carbon disulfide
Odor Quality
,
Strong, objectionable
Nitrogen dioxide
Strong, irritating
Ammonia
Ammoniacal
Diethyl
amine
Fishy
Fishy
Triethyl amine
Hydrogen sulfide
Rotten eggs
Skunk
Methyl mercaptan
Dimethyl sulfide
Rotten cabbage
Rotten cabbage
Dimethyl disulfide
Phenol
Carbolic disinfectant
Acetic acid
Acrolein
Penetrating; vinegar when dilute
Irritating; snuffed candle
Van ill an
Vanilla
Methyl amine
Triethanolamine
Ammonia, boiled lobsters
Oily; slightly fishy
Thiophene
Faint; neutral
Pyridine
Rank, unpleasant
Indole
Alphanaphthyl amine when concen-
trated, but jasmine when dilute
Skatole
Fecal
*Moncrieff (1967)
IV-24
-------
Odor intensity can be defined as a numerical in-
dication of odor strength. The following set of values
associated with a numerical scale has been establi~hed to
allow convenient statistical treatment of odor levels (Turk,
1960; Kerka, 1958; Nader, 1958).
Odorless
Threshold (barely
perceptible)
De fi n i te
Strong
Overpoweri ng
o
1
2
3
4
Table IV-9 lists odor threshold values and TLV's
of some odorants. The upper odor threshold concentrations
reported for some odorants exceed the TLV, indicating that
the TLV may be exceeded without some subjects detecting the
presence of the odorant via the olfactory senses.
Thr acceptability or unacceptability of an odor
may range frum very pleasant to obnoxious, depending upon
quality and intensity of the odor and/or prior conditioning
of the subject exposed. Odors are best evaluated upon the
first and second inhalations because subsequent inhalations
tend to fatigue the receptor's sense of smell (Patty, 1948).
Pervasiveness, another significant characteristic
of odors, is defined as the 1I...quality of an odor to pervade
a large volume of air and still continue to possess a de-
tectable intensityll (Nader, 1958).
Odor source samples are normally evaluated by odor
panels using the technique of successive dilution with clean
air until the odor threshold is reached. Results are re-
ported in odor units per unit volume, equivalent to the number
of volume dilutions required to reach the odor threshold con-
centration.
c. Health Effects - The relationship of odors to
the health of man is difficult to prove, since odors in them-
selves do not appear to produce observable pathological
(anatomic) effects. No relationship of odors to toxicity
has been established; an odor can, however, serve a useful
function as a warning of the presence of a toxic substance,
as with hydrogen sulfide. Offensive odors, however, have
been reported to cause anusea, vomiting, headache, loss of
appetite, impaired respiration, insomnia, and me.ntal de-
pression.
IV-25
-------
Table IV-9 - Recognition Odor Thresholds and
TLV's of Some Odorants
Odoran t
Odor Thresho1d(1)
(ppm)
.06, .07, .21
.21, .33,1.5,1.8
Acetaldehyde
Acrolein
Carbon disulfide
.21, .77
.003, .0059
Di ethyl .su1 fi de
Di me thy 1 ami ne
.047, 0.6
Dimethyl sulfide
Ethyl mercaptain
.001, .0025, .004, .02
.0004, .001, .002, .0033 '
.06-.09, 1.0, 0.4-6.6
Formaldehyde
Gasoline
10
.01, !0011, 0.13-1.0
Hydrogen sulfide
Methyl mercaptain
.00099, .0021, .0011, .04
4.0, 1-3
Nitrogen dioxide'
Ozone
.005, .02-.05, 0.1, 0.5, 2.0
.000000075, .019
Skatole
Sulfur dioxide
.03-1.0, 0.47, 3.0
.00077
Thiopene
(1) Taken from various reference sources by
R. Sullivan (1969).
(2) Documentation of Threshold Limit Values,
copyright 1966, by American Conference
of Governmental Industrial Higienists
IV-26
TLV(2)
(.ppm)-
200
0.5
20
10
2
5
500
20
2
5
O. 1
5
-------
Surveys have indicated that the general public al-
most invariably links air pollution with "bad" smells. Be-
cause of the subjectivity of odor complaints, as shown in
opinion surveys, the level of air pollution cannot be ob-
jectively appr~ised on the basis of complaints of bad smells.
Some allergic conditions, such as asthma attacks,
may be related specifically to odors (Horesh, 1966; Urbach,
1942) .
C.
PHOTOCHEMICAL SMOG
1.
Introduction
Photochemical smog is a term that describes the mani-
festations of the complex photochemical reactions of atmospheric
pollutants. The photochemical reactions that produce photo-
chemical smog can involve a variety of reactants, both organic
and inorganic, but it has been recognized for many years that
hydrocarbons in general are essential reactants for smog for-
mation. However, individual hydrocarbon species differ greatly
in their relative reactivities and in their production of the
manifestations of smog such as eye irritation, visibility re-
duction, and plant damage.
A considerable body of literature describes various
aspects of the reaction mechanisms, reaction kinetics, and
characteristics of photochemical smog. Several comprehensive
reviews have been written on the photochemical reactions in-
volved in smog formation (Leighton, 1961; A1tshuller and
Bufalini, 1965, 1971; System Development Corporation, 1970).
Much of the existing data on photochemical systems has been
reviewed and evaluated in the Air Quality Criteria for Photo-
chemical Oxidants, AP-63~ .and the Air Quality Criteria for
Hydrocarbons, AP~64.
The recognized variability of individual hydro-
carbons and chemical group classifications in their con-
tribution to photochemical smog formation requires that some
relative measure of this variability be developed in order
to determine which hydrocarbons should be controlled. The
following sections describe the development of relative
measures of smog contribution which were used to rank the
major source emissions.
2.
Photochemical Reactivity
Photochemical reactivity is the tendency of a
particular substance to participate in photochemical re-
actions and produce the adverse manifestations associated
IV-27
-------
with smog. As with most chemical reactions, photochemical re-
activity is a function of chemical structure and is additionally
inlfuenced by specific reaction conditions. The extremely com-
plex and uncontrollable nature of ambient atmospheres has in-
fluenced investigators to employ simulated and simplified re-
action conditions in studies of photochemical reactivity. The
laboratory conditions employed for photochemical reaction studies
vary widely, but they usually involve some form of simulated
atmosphere, generically termed a "smog chamber".
Hydrocarbon reactivity studies in the laboratory have
developed considerable information on the relative reactivities
of various homologous series and of specific compounds. The
usefulness of much of this information, however, is highly
questionable. Reaction conditions and techniques used to measure
reactivity vary widely, and thus the number of hydrocarbons in-
vestigated by anyone consistent approach has been limited.
Additionally, the degree of simulation of ambient atmospheric
conditions has often been quite poor and thus it has not been
apparent how much of the data can be extrapolated to actual at-
mospheric smog formation.
Recent investigative trends are being directed toward
reactivity studies of ambie~t atmospheric samples, so that in
the near future data more directly applicable to real con-
ditions may be available.
Laboratory or smog chamber data available for assess-
ment of relative hydrocarbon reactivities include hydrocarbon
consumption, nitric oxide oxidation, ozone or oxidant* for-
mation, and eye irritation. The various investigators and re-
viewers have generally attempted to rank the hydrocarbon in-
vestigated on reactivity scales which use one or more of the
above measurement parameters.
3.
Reactivity Scales
In their reviews of the literature data on photo-
chemical aspects of air pollution, Altshuller and Bufalini
(1965, 1971) compared reactivity data obtained by various in-
vestigators. Altshuller (1966a) summarized available reactivity
data and considered the problems involved in developing a single
index of reactivity for organic substances.
*The term "oxidant" is generally defined as that group of re-
action products that will oxidize potassium iodide.
IV-28
-------
Hydrocarbon consumption and nitric oxide formation
have been widely used as measures of relative hydrocarbon re-
activity in smog chamber studies. Despite differences in re-
action conditions and measurement techniques, the relative
ordering of hydrocarbon group classifications by these measures
is fairly consistent. The relative ordering by hydrocarbon
consumption from laboratory measurements has also been shown
to be consi'stent with that obtained by irradiation of ambient
atmospheric samples.
Relative reactivities based on hydrocarbon consumption
and nitric oxide oxidation do, however, show significant dif-
ferences when attempts were made to quantify reactivity re-
lationships. Investigators have generally developed reactivity
scales by choosing an arbitrary scale, such as 0 to 2 or 0 to 10,
and assigning the highest scale number to the most reactive com-
pound or group investigated. Other compounds are then assigned
a scale number in relation to their reactivity relative to the
highest ranking compound. Altshuller (1971) compared the re-
sults of several studies in this fashion, as shown in Tables
IV-10 and IV-ll. Table IV-ll also includes data from the more
recent studies by Glasson and Tuesday (1970) and Levy and
Miller (1970) which were not included in Altshuller's review.
In these tabulations, differences in scaling used by the original
investigators were adjusted to permit direct comparison. This
was done by a$signment of the same scale number to one compound
common to the studies being compared and adjusting other scale
numbers in a manner consistent with the original ordering.
Glasson and Tuesday (1970) investigated nitric oxide
oxidation rates of a large number of hydrocarbons under a stan-
dard set of conditions and demonstrated that reactivity ordering
according to hydrocarbon class requires rather detailed break-
down of hydrocarbon classes into specific structural groupings.
To avoid extreme overlapping of reactivity between groups or
structural classes, six structural classes were required.
Table IV-12 shows the structural or reactivity classes and
their relative reactivities on a 0 to 100 scale, as summarized
by Glasson and Tuesday.
Reported relative reactivity rankings based on oxidant
formation and eye irritation data show much less internal agree-
ment than those based on hydrocarbon consumption or nitric oxide
oxidation. Oxidant formation is highly dependent on irradiation
times and hydrocarbon to nitrogen oxides ratio and thus there
are significant variations in reported data.
Eye irritation data depend on human subject response
and may thus be subjective and unreliable. Additionally, the
chemical factors resulting in eye irritant production are
IV-29
-------
Table IV-10 - Ranking of Reactivities of Hydrocarbons When
Photooxidized in Presence of Nitrogen
Oxides Under Static Conditions
Schuck & Stephens &
Doy 1 e Scott Tuesday
Hydrocarbon (1959) ( 1969 )- _(1963t
Tetramethy1ethy1ene 10 10 10
trans-2-Butene 6 8
cis-3-Hexene 6
Isobutene 1.5 2 2
1 , 3-Butadi ene 1
Propylene 1 2 1
m-Xy1ene 1
p-Xy1ene 0.5 0.5
Ethylene O. 1 0.3
Hexanes, octanes <0.1
Pentanes <0.01
IV-30
-------
Table IV-ll - Reactivities of Hydrocarbons Based on Ability to
Participate in Photooxidationof Nitric Oxide to
Nitrogen Dioxide(~) .
Ranking
Altshuller Glasson Levy
and Cohen and Tuesday and Miller
Hydrocarbon ( 1 963 )- (1965) (1970'- (l 970 )-
2,3-Dimethylbutene-2 10
2-Methyl-2-butene 2 (a) 3
trans-2-Butene 2 2 2
Isobutene 1
Propylene 1 0.5 0.6
Ethylene 0.4 0.3 0.3 0.6
1,3,5-Trimethylbenzene 1.2 1.2 0.9
m-Xy1ene 1 0.9 0.7 0.6
1,2,3,5-Tetramethy1benzene 0.9 0.7
1,2,4-Trimethy1benzene 0.6 0.7 0.6
o-and p-Xylene 0.4 0.4 0.4
0- and p-Diethylbenzene 0.4 0.4
Propy1benzenes 0.3 0.2
Toluene 0.2 0.2 0.2 0.2
Benzene o. 15 0.04 0.05 0.03
n-Nonane O. 15
3-Methylheptane O. 15
n-Heptane 0.2
Methylpentanes <0.1 0.2 0.2
Pentanes 0.2
2,2,4-Trimethylpentane O. 1 5 o. 15
Butanes O. 1
Ethane 0.03 0.03
Methane <0.01
Acetlyene O. 1
(a)Arbitrarily adjusted to the same ranking as trans-2-
butene on Glasson and Tuesday's scale to permit com-
parison of other hydrocarbons.
IV-31
-------
Table IV-12 - Hydrocarbon Reactivity in Nitric
Oxide Photooxidation
Hydrocarbons in.
Reactivity Class
Reactivity
Class
Internal olefins with two
double bond substitutions
6
Cyclopentenes
Internal olefins with one
..double bond substitution
5
Internal olefins without
double bond substitution
4
Cyclohexenes
Tri- and tetraalkylbenzenes
Diolefins
Terminal olefins
Dialkylbenzenes
3
Monoa1ky1benzenes
C4 + paraffins
2,2-Dimethylpropane
Propane
Benzene
Ethane
Methane
2
1
IV-32
Relative
Reactivity
100
40
1 5
5
2
o
-------
apparently unrelated to the other parameters used for re-
activity measurement. Heuss and Glasson (1968)t in their
study of 25 'hydrocarbonst found no correlation between eye
irritation and chemical measurements of reactivity. A
correlation was observedt howevert between hydrocarbon
structure and eye irritation:
Altshuller (1971) presented a tabulation of rela-
tive reactivitYt on a 0 to 10. scalet based on eye irritation
measurements. This tabulation is shown in Table IV-13.
A program carried out for the National Paintt
Varnish and Lacquer Association (Levy and Millert 1970)
examined the reactivity of 45 organic solvents in terms of
eye iiritation measurementst chamical reactivitYt and pro-
duct yield. The study found poor coorelation between eye
irritation response times and any chemical parameters measured.
4.
Composite Reactivity Scales
episodest
effect on
tribution
effect.
The general lack of correlation of biological re-
sponset measured by eye irritationt with other laboratory re-
activity measurements has been used as the basis for the
agruement that r~ntings of hydrocarbons should not be based
solely on chemical parameters. Altshuller (1966b) suggested
that a ~omposite reactivity scale that included biological
response as well as chemical measurement of reactivity would
be more meaningful than a reactivity scale based on chemical
reactivity alone.
This is based on the fact thatt in actual smog
eye irritation is the most commonly noted adverse
humans. Thust estimates of relative smog con-
should include some measure of this adverse health
Table IV-14 shows Altshuller's tabulation of chemical
group classifications ranked on a 0 to 10 scale according to
various chemical and biological response parameters.
For our studYtPublished photochemical reactivity
scales were combined to give a Composite Reactivity Index
for a variety of individual hydrocarbon compounds. This Index
rankingt shown in Table IV-15 was based on a scale of 0 to lOt
since this scale allows a reasonable range for dispersion of
high to low reactivity without implying undue precision in the
placement of individual compounds relative to others with
similar structural characteristics.
IV-33
-------
Table IV-13 - Eye Irritation Reacti vi ty*
Eye Irritation Eye Irritation
Hydrocarbon Reactivity Hydrocarbon Reactivity
n-Butane 0 m,- XyJ en e.. 2.9
n-Hexane 0 1,3,5-Trimethylbenzene 3. 1
Isooctane 0.9
tert-Butylbenzene 0.9 l-Hexene 3.5
Benzene 1.0 Propylene 3.9
Ethylene 1.0 Ethylbenzene 4.3
l-Butene 1.3 Toluene 5.3
Tetramethylethylene 1.4 n-Propylbenzene 5.4
- cis-2-Butene 1.6 Isobutylbenzene 5.7
0:::::
I Isopropylbenzene 1.6 n-Butylbenzene 6.4
w sec-Butylbenzene 1.8 1,3-Butadiene 6.9
.;:..
2-Methyl-2-butene 1.9 c:x-Methylstyrene 7.4
trans-2-Butene 2.3 Alkylbenzene 8.4
o-Xylene 2.3 8-Methyl styrene 8.9
p-Xylene 2.5 Styrene 8.9
*Altshuller{197l)
-------
Table IV-14 - Comparison of Product Yields and Effects
Caused by Various Organics ..
Respone on 0 to 10 Scale
Peroxy- Eye
Substance or Ozone or acyl Formal- Irrita- Plant Averaged
Sub-Class Oxidant Nitrate dehyde Aerosol tion Damage Response
C1-C5 paraffins 0 0 0 0 0 0 0
acetylene 0 0 0 0 0 0 0
benzene 0 0 0 0 0 0 0
C6+ paraffins 0-4 0 0 0 0 0 1
- toluene (and other
< ND{a)
I monoalky1benzenes) 4 2 2 4 0-3 3
w
U1 +(b)
ethylene 6 0 6 1-2 5 4
l-alkenes 6-10 4-6 7-10 4-8 4-8 6-8 7
diolefins 6-8 0-2 8-10 10 10 0 6
dialkyl and trialkyl- +(b)
benzenes 6-10 5-10 2-4 4-8 5-10 6
internally double-
bonded olefins 5-10 8-10 4-6 6-10 4-8 10 8
aliphatic aldehydes 5-10 +(b) +(b) ND{a) +(b) +(b)
(a) No experimental data available
(b) Effect noted experimentally but data insufficient to quantitate.
-------
Table IV-15 - Composite Reactivity Index Values
For Selected Hydrocarbons
Compound
2-Butene,2,3-dimethyl
2-Pentene,2,3-dimethyl
2-Butene, trans-
2-Butene, cis-
l-Butene, 2-methyl-
l~Butene, 3-methyl-
2~Butene, 2-methyl-
2-Hexene, cis-
2-Hexene, trans-
3-Hexene, cis-
3-Hexene, trans-
2-Pentene
2-Pentene,
2-Pentene,
2-Pentene,
cis-
trans-
4-me thy 1-
l-Butene
l-Hexene
Formaldehyde
Isobutylene
l-Pentene
l-Pentene, 2-methyl-
l-Pentene, 4-methyl-
Propylene
Propylene Oxide
Styrene
Vinyl acetate, monomer
Vinyl chloride, monomer
a-Xylene
m-Xylene
p - Xy 1 en e
Benzene, 1,2,4-trimethy1-
Benzene, 1,3,5-trimethyl-
1,3-Butadiene
Butane, 2-methyl-
Cumene
Cyc 1 open tane
Isobutyl methyl ketone
Propadiene
Toluene
Toluene, diisocyanate
Toluene, m-ethyl-
Toluene, p-ethyl-
Trichloroethylene
IV-36
Composite Reactivity Index
(a to 10 scale)
10
9
8
8
8
8
8
8
8
8
8
8
8
8
8
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
6
6
6
6
6
6
6
6
6
6
6
6
6
-------
Table IV-15 (Continued)
Compound
Ethylene
Ethyl benzene
Cyclohexane
Cyclohexanone
Cyclopentane
Cyclomethyl
Ethyl methyl ketone
Hex a,n 0 1, 2 - e thy 1
Isopropyl alcohol
Methyl 'alcohol
n-Butyl alcohol
Acetone
Butyl acetates,
Ethyl acetate
Ethyl alcohol
mixed
Benzene
Butane, 2,2-dimethyl
Butane, 2,3-dimethy1
Hexane
Hexane, 2-methyl-
Hexane, 3-methyl-
Pentane, 2,3-dimethyl-
Pentane, 2,4-dimethyl-
Pentane, 2,-methyl-
Pentane, 3,-methyl-
Pentane, 2,2,4-trimethyl-
Acetylene
Butane
Cellulose acetate
Diethylene glycol
Ethane
Ethylene dichloride
Isobutane
Isopentane
Methane
Methane, trichlorofluoro
Methyl chloride
Methylerie chloride
Peptane
Propane
Perchloroethylene
Composite Reactivity Index
(0 to 10 scale)
5
4
3
3
3
3
3
3
3
3
3
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
IV-37
-------
In developing the Composite Reactivity Index, the
adverse effects of smog were heavily weighted by basing our
scale on that given by A1tshu11er (1966b) for relative re-
activity averaged from product yields and biological response
data (Table IV-14).
In order to incorporate reported reactivity data
which employed different scaling, e.g., the data of Levy and
Miller (1970) which were originally ranked on a sca1, of from
1 to 3, the reactivities were adjusted to ~he 0 to 10 scale
by assignment of a value of 8 (Altshu11erls aver~ged response)
to internally double bonded olefins. Other compounds were
then assigned values on the 0 to 10 scale relative to the in-
ternally double bonded olefins, consistent with the relative
chemical group arrangement of Table IV-14.
5 .
Chlorinated Hydrocarbon Solvents
Few specific studies have been made of the chlori-
nated hydrocarbons, with the exception of trichloroethylene.
Thus, we estimated the reactivities to the saturated chlori-
nated compounds to be similar to those of other saturated
hydrocarbons.
Altshuller (1971) cited the results of an unpublished
study by Kopczynski (1968) of trichloroethylene that included
determination of N02 formation rate, oxidant formation rate,
and eye irritation response. On the basis of chemical re-
activity, trichloroethylene was found to be between propylene
and ethylene; on the basis of eye irritation manifestations
it was somewhat higher than propylene.
Wilson et al, in their study of trichloroethylene,
reported both chemical reactivity and eye irritation mani-
festations to be lower than those observed by Kopczynski.
Chass (1970) took exception to certain of the results reported
by Wilson, on the basis of differences in experimental pro-
cedure and data interpretation, citing results of Los Angeles
APCD studies more nearly in agreement with Kopczynski. Chassl
interpretation places trichloroethylene in ~he reactivity
range of xylenes and olefins.
We used the data of Kopczynski to assign tri-
chloroethylene a reactivity of 6 on our 0 to 10, relative
index.
IV-38
-------
6.
Atmospheric Relationships
Schuck et al (lQ70) investigated the relationship
of ambient air hydrocarbon concentration to smog manifestations
by a compilation and correlation of air monitoring dat~. An
empirical relationship was developed by compari~ion of data
from the Continuous Air Monitoring Project (CAMP) for 6:00-
9:00 A.M. average non-methane hydrocarbon concentrations with
total oxidant concentration levels attained later in the same
day. Although only limited data were available for the
correlation, a relationship was observed which served to define
the upper limit of daily oxidant concentration as a function
of the 6:00-9:00 A.M. non-methane hydrocarbon levels. The
conclusion from the correlation was that if maximum l-hour
oxidant concentration is to be limited to about 0.1 ppm. the
6:00-9:00 A.M. non-methane hydrocarbon concentrations must be
less th~n about 0.3 ppm (as carbon). This approximately 3:1
relationship was used in the development of ambient air quality
standards for hydrocarbons set by the Federal Government
(Federal Register 36, 8186, 1971).
Do
VEGETATIVE DAMAGE
1.
Introduction
Vegetative damage has long been recognized to result
from exposure of growing plants to specific hydrocarbons,
notably ethyl~ne. More recently, vegetative damage was one of
the early manifestations observed in studie$ of photochemical
smog and its effects. As early as 1913, Knight and Crocker
(1913) suggested that ehtylene, and perhaps other carbon con-
taining gas.es, be considered phytotoxic. Haagen-Smit et al
(1952) observed damage to sensitive plants from exposure to
synthetic smog products.
A review of the literature indicated that plant damage
from exposure to ambient atmospheric concentrations of hydro-
carbons results predominantly from ethylene and the photo-
chemical smog products. Other inuurious hydrocarbons, such as
the aldehydes and unsaturated hydrocarbons other than ethylene,
produce plant damage only at concentrations much higher than
are present in ambient atmospheres.
2.
Relative Importance of Hydrocarbons Causing
Plant Damage
Various investigators have studied the relative im-
portance of hydrocarbons in terms of their damage effects on
sensitive plant species. General reviews of the effects of
I V - 39
-------
hydrocarbons on ve~etation have been given by Crocker (1948),
Slayton and Platt (1967), and the Air Quality Criteria for
Hydrocarbons (USDHEW, 1970). ...
Specific studies and results are discussed in the
following subsections: unsaturated hydrocarbons; aldehydes;
and photochemical smog products.
a. Unsaturated Hydrocarbons ~ Abeles and Gahagen
(1967) investigated the effects of exposure of $ensitive
plants to various concentrations of ethylene, propylene,
acetylene, l-butene, and 1,3-butadiene. They found that ex-
posure toO.l ppm ethylene resulted in about one-half maximum
. abscission. These authors then compared their results with
those of other studies (Burg and Burg, 1962; Crocker, 1934;
Zimmerman, 1935) as shown in Table IV-16. This tabulation
shows the relative concentrations of several unsaturated
hydrocarbons that produce the same degree of biological
response as ethylene.
Crocker (1935) found that exposure to concentrations
of 0.1 ppm ethylene produced epinasty in sensitive plant
species. Saturated hydrocarbons, 1,3-butadiene, and benzene
ring compounds were found to have no harmful effects under the
conditions studies.
Burg and Burg (1962) found that exposure to 0.1 ppm
ethylene inhibited elongation of peas. Methane, cis-2-butene,
trans-2-butene, and isobutene were inactive.
Dosage-response data from various inv~stigator$ on
the eff~cts of ethylene were summarized in the Air Quality
Criteria for Hydrocarbons. Damage to the more sensitive' plants,
such as orchids,has been observed from exposure to less than
O~ 1 ppm for as little as 1 hour.
The literature data thus show that, of the unsatu-
rated hydrocarbons, only ethylene produces significant plant
damage at low concentrations comparable to those attained in
ambient atmospheres. Additionally, the more limited investi-
. gation of other simple hydrocarbons indicates that paraffins
and benzene ring compounds have no harmful effects on vege-
tation. .
Stephens and Burlesen (1967) and Stephens et al
(1967) reported that ambient concentrations of ethylene in
congested urban areas were 25-150 ~g/m3 (0.02-0.13 ppm), well
within the harmful range of ethylene.
IV-40
-------
Table IV-16 - Relative Concentrations of Some Unsaturated Hydrocarbons That Produce
Biological Response Similar to that Produced by Ethylene
Inhibition of Growth
Compound Abscission Epinasty Pea Stem Tobacco
Ethylene 1 1 1 1
Propylene 60 500 100 100
Acetylene 1,250 500 2,800 100
..... 1-Butene 100,000+ 500,000 270,000+ 2,000
<
I
~ 1,3-Butadiene 100,000+ 5,000,000+
-
-------
b. Aldehydes - Haagen-Smit et al (1952) found that
exposure to relatively high concentrations of aldehydes re-
sulted in significant damage to sensitive plants. They con~
eluded, however, that no damage should be expected from e~-
posure to atmosphe~ic concentrations of aldehydes.
Investigation of the effects of smog products on
vegetation indicated that exposure to formaldehyde and
acetaldehyde did not result in plant injury (Stephens, 1961;
Altshuller and Hindawi, 1964; Altshuller, 1966).
Brennan (1964) reported that plant injury corre-
lated with total atmospheric aldehydes, but the work was
criticized by Hindawi and Altshuller because of poor analytical
procedures.
c. Photochemical Products - Harmful effects to vege-
tation from exposure to photochemical smog products are gen-
erally attributed to the oxidant products; ozone, peroxyacetyl
nitrate (PAN), and other homologs of PAN. As noted by Lacasse
and Moroz (1969), PAN and its higher homologs are important
phytotoxicants, even though they are present at concentrations
of only a few percent of the total oxidants.
Jaffe (1967) reviewed published data on vegetative
effects of PAN and other photochemical oxidants and summarized
studies of both short- and long-term effects on several plant
species. These summaries are shown in Tables IV-17 and IV-18.
Relative toxicity of PAN and its higher homolo~s in-
dicated by fumigation of two plant species (Taylor, 1969) is
shown in Table IV-19. Concentrations of the order of a few ppb
are seen to result in injury at exposure times of 0.5 to 4
hours. These time-concentration conditions are similar to
those observed in ambient atmospheres of smog-affected urban
are as ( T ay lor, 1 9 69 ) .
3.
General Symptoms of Pollution Injury to Vegetation
Assessment of the effects of pollution vegetation
is generally based on characterization of injury symptoms.
Brandt and Heck (1968) stated that symptoms of injury observed
on the leaves of plants attributable to air pollution are:
(1) leaf tissue collapse with necrotic patterns, (2) chlorosis
or other color changes, and (3) growth alterations and/or
suppression. In many cases, the injury pattern developed is
highly characteristic of the toxic agent; however, the patterns
may not always be specific. Disease, insects, nutrition and
other factors can produce the same leaf patterns.
IV-42
-------
Table IV-17 - Typical Short-Term Effects on Photochemical Oxidants on Vegetation
Concentration Len g th of
E ffec ts Spec; es Pollutant (ppm) Exposure
Lea f lesions Ifl~Cking) and tissue Tobacco, Bel-W3 Total oxidant 0.5-0.10 2 hr
destruction field studies) -( a vg)
Leaf tissue destruction Pinto beans Total oxidant 0.11-0.15 24 -hr
oxidant (fresh
maxima plants
exposed
da i 1 y)
Leaf lesions and tissue destruction Tobacco, Bel W3 Ozone 0.5-0.10 2 hr
- Ozone hr
<: Lea f lesions an d tissue destruction Tobacco, Bel-W3 0.02 8
I
~
w Leaf lesions and tissue destruction Petunia PBN 0.005 7-8 hr
Leaf lesions and tissue destruction Petunia PAN O. 10 5 hr
Pinto beans
Acute leaf tissue destruction Pinto beans Ozone 0.70 30 min
Acute leaf tissue destructi.on Pinto beans PAN LO 30 min
Inhibition of p~otosynthesis Pinto beans PAN 1.0 30 min
Changes in cellular chloroplast Pinto beans PAN ],0 30 min
structure
-------
Table IV-18 - Typical Prolonged Effects of Photochemical
Oxidants on Vegetation
Po11u... Concentration Length of
Effects tant (ppm) Exposure
Retardation of browth
and reduction in yield
Reduction in fres h wt PAN 0.005-0.015 5-6 hr/day,
(Tomato >22%) 5 days
Reduction in dry wt PAN 0.005-0.015 5-6 hr/day,
(Tomato >22%) 5 days
Reduction in fresh wt PAN 0.005-0.015 5-6 hr/day,
(Petunia >36%) 16 days
Reduction in dry wt PAN 0.005-0.015 5-6 hr/day,
(Petunia >28%) 16 days
Reduction in f1 owe ri ng Ozone 0.10-0.20 8 hr/day,
(Petunia >41%) 9 days
Reduction in stem Ozone 0.10-0.20 8 hr/day,
elongation 9 days
(Petunia >12%)
IV-44
-------
Table IV-19 - Relative Phytotoxicity of PAN Homo10gs on Two Plant Species (a)
PAN (1) PPN(2) PBN(3) Piso BN(4)
Fum. Con.c. . % Cone. % Cone. % Cone. %
hr. ~ In jury uL In jury ~ In jury ~ Injury
Bean (var. Pinto)
0.5 100 90 100 80 100 80
1.0 140 55 24 7 30 59
4.0 20 44 5 100
......
<:
I
.J::Io
C.11
Petunia (var. Rosy r~orn)
0.5 50 7 100 90 25 18
1.0 ] 40 33 24 35 12 45
(1) Peroxyacetyl
(2) Peroxypropionyl nit~ate
(3) Peroxybutyryl nitrate
(4) Peroxyisobuty1 nitrate
(a) Percent injury based on acute
symptoms of tissue collapse.
-------
General effects as described in The Air ~uality
Criteria for Hydrocarbons (1970) are: (1) acutes c arac-
terized by tissue collapse and necrosis of leaf partss
usually accompanied by a rapid change in leaf color; (2)
chronics identified by slow development of mild or severe
symptoms over a long periods such as chlorosis without death
of cells.
The terms "injury" and "damage" are often used
interchangeably. As recommended by Guderian (1960)s Heggestad
(1968)s Stern (1968) and in the handbook by Lacasse and Moroz
(1969)s plant damage indicates an economic loss and a change
in the usability of the plant. Injury includes all responses
of the plant to a pollutants but such injury does not nece-
ssartly constitute a basis for economic loss.
Brandt and Heck (1968) gave an excellent summary of
air pollutant injury patterns. Those pertinent to this study
are given as follows:
. a.
Ethylene
1) Leak markings
. ..Broad leaf - Epinasty and/or abscission
of leaves without markings. Sensitive plants may develop
general chlorosis of olde~' leavess necrosiss and leaf ab-
scission. Stimulation of lateral development and finally
death of plant may result. Abscission of young flower buds
and/or failure of floral blooms to open. More resistant
plants may show only retardation of growth and possible loss
of api~al dominance.
...Grasses - Retardation of growth with
high suckering rate. No visual damage even at high concen-
trations.
...Conifers - Abscission of needless retarded
elongation of new needless abscission of young cones or poor
cone development.
2) Similar markings
...Broad leaf - Water stress (welting)s
bacterial wiltss root upsetss nematode and aphid injurysearly
senescence.
...Grasses - Suppressed growth and lack of
markings are dissimilar to other disorders.
IV-46
-------
. ..Conifers - Water stress, other pollutants.
b .
PAN -
1) Leaf markings - These are the following
types:
. ..Broad leaf - On leafy-type veg~tab1es and
some other plants, a collapse of the tissue on the underside
of the leaf giving a glazed, si1ver~d, or bronzed appearance.
On tobacco, petunia and similar leaves, collapse may be through
the thickness of the leaf, usually in a banded pattern; on
some, a blotchy pattern slightly resembling sulfur dioxide in-
jury, but usually without the typicq1 bleach. An early maturity
or senescence is also associated with PAN.
...Grasses - Irregular collapsed banding
bleached yellow to tan~ sometimes appearing more as a chlorotic
or bleached band than necrotic.
...Conffers - Not highly specific needle blight
with some chlorosis or bleaching.
2) Similar markings -
...Broad leaf - "Sun scald", various virus
and fungus diseases which produce blotchy pattern. Typical
silvering is seldom duplicated by other agents.
...Grasses - Various fungal and bacterial
blights produce streaked and banded markings.
4.
Economic Crop Losses
It has been stated that plant damage is an important
economic loss to farmers and an important esthetic loss to
home owners. Although it is impossible to make a fire assess-
ment of the cost of air pollution damage to vegetation, a recent
estimate (Air/Water [1970]) indicates crop damage in California
alone exceeds $100 million per year and reaches $500 million
per year for the entire nation. It has been recently stated
that air pollution damage to agriculture in California's two
major smog areas totalled more than $44 million during 1969;
these were conservative estimates. This latter report showed
the citrus industry hardest hit, losing more than $33.5 million,
with extensive damage also done to the following crops: grapes -
$935,000; beans - $826,000; alfalfa - $500,000; celery -' $367,000;
tomatoes - $270,000; barley - $170,000; sweet corn - $163,000.
Significant losses also affected apples, avocados, figs and
pears; lighter damage was done to grass, hay, spinach, lettuce,
radishes, turnips, endive, and rhubarb.
IV-47
-------
Va 1 ue s ,glv e~~: \f,1).'r '~:.c,ononri'c::; r.eir
-------
5 .
Summary
Ethylene has been shown to be the only important
hydrocarbon causing injury to vegetation; aldehydes have
not been proved to produce injurious effects on crops. The
photochemical organic oxidant PAN appears to be a very im-
portant cause of plant injury and may be more widespread
than thought.
Ethylene produces symptoms of plant injury which
are not as recognizable as those produced by PAN; diagnosis
of plant injury by air pollutants is thus a difficult task.
Many environmental and pathological conditions produce plant
injury symptoms which are difficult to differentiate from
those caused by air pollutants. .
IV-49
-------
CHAPTER IV
REFERENCES
B.
HEALTH EFFECTS
ACGIH, Threshold Limit Values of Airborne Contaminants, adopted JY
the American Conference of Governmental Industrial Hygienists
for 1970.
Altshuller, A.P., "Reactivity of Organic Substances in Atmospher~c
Photooxidation Reactions", Intern. J. Air Water Pollution 10,
713-33 (1966).
Altshu11er, A.P., "Composition and Reactions of Pollutants in
Community Atmospheres", Bull. World Health Organ. (Geneva)
~ (4), 616-23 (1969).
Amdur, M.O., liThe Response of Guinea Pigs to Inhalation of For-
maldehyde and Formic Acid with a Sodium Chloride Aerosol",
Intern. J. Air Pollution 1, 201 (1960).
Anderson, D.O. ,liThe Effects of Air Contamination on Hea1th'~ .
Can. Med. Assoc. J. (Toronto), Part 11,585-93, September
9, 1967a.
Anderson, D.O., liThe Effects of Air Contamination on Hea1th".
Can. Med. Assoc. J. (Toronto), Part III, 802-6, September.
23, 1967b.
API Toxicological Review: Gasoline, 1st Edition, American
Petroleum Institute (1967). . .
API Toxicological Review of Benzene, American Petroleum In-
stitute (1960).
API Toxicological Review - Butadiene, 2nd Edition, American
Petroleum Institute (1959).
API Toxicological Review of Toluene, American Petroleum Institute
(1960).
API Toxicological Review of Xylene, American Petroleum Institute
(1960).
Bates, D.V., and R.V. Christie, Respiratory Function.i!!. Disease,
W.B. Sounders Company, Philadelphia and London, 1964.
Christman, R.F., Air Pollution Control Guidebook for Management,
A.T. Rossano, Jr., Ed., "Chemistry of Air Po1lution", En-
vironmental Sciences Service Div., ERA, Jnc., Stamford,
Connecticut, 1969. .
IV-50
-------
REFERENCES, CHAPTER IV (continued)
Coffin, D.L., IIHealth Aspects of Airborne POlycyclic Hydrocarbonsll,
t.1 e d i c a 1 As p e c t s 0 fA; r Poll uti On, 46 - 5 2, J an. 1 4, 1 9 7l.
Davis, A., L. Schafer, and Z. Bell, liThe Effects on Human Volunteers
of Exposure to Air Containing Gasoline Vaporsll, Arch. Environ.
Health 1, 548-54 (1960).
Deichmann, W.B., and H.W. Gerarde, Toxicology of Drugs and Chemicals,
, Academic Press, New York, 1969. --
Documentation of Threshold Limit Values, Copyright 1966 by American
Conference of Governmental Industrial Hygienists.
Documentation of Threshold Limit Values for Substances in Workroom
Air, American Conference of GovernmenfiT Industrial Hygienists,
3rd Ed., 1971.
Drinker, P., C. Yaglow, and M. Warren, liThe Threshold Toxicology of
of Gasoline Vapor II, 'J. Ind. Hyg. and Toxicology £2., 225-32 (1943).
Duffer~ 'R.A., "Appraisal of Odor Measurement Techniques", J. Air'
Pollution Control Assoc. ~ (7), 472 (1968).
Fairhall, L.T., Industrial Toxicology, 2nd ed., Hafner Publishing
Co., New York, 1969.
Faith, W.L., "Hydrocarbons as Air Pollutants II, Industrial Wastes
i, 86.-8 (1960).
Falk, H.L., and P. Kotin, liThe Fate of Polycyclic Aromatic Hydro-
carbons in Soot Following Their Emission into the Atmosphere
and Deposition in the Lungs", Proc. Am. Assoc. Cancer Research
£(1), 15 (1955).
Falk, H.L., P. Kotin, and A. Miller, "Aromatic Polycyclic Hydro-
carbons in Polluted Airll, Intern. J~ Air Pollution 2, 201 .(1960).
Fasset, D.W., "Aldehydes and Acetols", Industrial Hygiene and
Toxicology, F.A. Patty, Ed., Interscience Publications Inc.,
New York, 1963.
Gardner, M.D., C.G. Loosli, B. Hanes, W. Blackmore, and D. Tubkin,
IIPulmonary Changes in 7000 Mice Following Prolonged Exposure
to Ambient and Filtered Los Angeles Airll, Arch. Environ. Health
~ (3), 310-7 (1970).
IV-51
-------
REFERENCES, CHAPTER IV (continued)
Gerarde, H.W., "Aliphatic Hydrocarbons", Industrial Hygiene and
Toxicology, 2nd Ed., F.A. Patty, Ed., Interstience Publications,
Inc., New York, 1962.
Gerarde, H.W., "Toxicological Studies of Hydrocarbons. IX. The
Aspiration Hazard and Toxicity of Hydrocarbons and Hydrocarbon
Mixtures", Arch. Environ. Health £',329-41 (1963).
Gross, P., LA. Pfitzer, J. Watson, R.T.P. deTreville, M. Kaschak,
£. Tolker, and M.A. Babyak, "Experimental Carcinogenesis: Bronchial
Intramural Adenocarcinomas in Rats from X-ray Irradiation of the
Chest", Cancer fl, 1046-60 (1969).
Gurinov, G.P., F.D. Hashluts, and L.M. Shabad, "A Study of Blasto-
mogenic Properties of Some Tars Derived from Atmospheric Dust
and from Combustion Products", Air Pollution Translation I,
100 (1969).
Hangebrauck, R.P., D.J. vonLehmden, and J.E. Meeker, "Emission of
Polynuclear Hydrocarbons and Other Pollutants from Heat Generation
Sources and Incineration Processes", J. Air Pollution Control
Assoc. li, 267 (1964).
Hatch, T.F., and D. Gross, Pulmonary Deposition and Retention of
Inhaled Aerosols, Academic Press, New York, 1964.
Hendrickson, LR., C.G. Walker, and V.D. Chopnerkar, "Identification
of Non-Sulfur Organic Compounds in Stack Gases from Pulp Mills",
Am. Ind. Hyg. Assoc. J. £1., 121 (1963).
Horesh, A.J., "The Role of Odors and Vapors in Allergic Disease",
J. Asthma Res. 4 (2),125 (1966).
Katz, M., "Carcinogenic Hazards of Aromatic Polycyclic Hydrocarbons
in the Polluted Atmosphere", Occup. Health Rev. (Ottawa) li (1)
(1942).
Kerka, W.F., and LR. Kaiser, "An Evaluation of Environmental Odors",
J. Air Pollution Control Assoc. 2, 297-301 (1958).
Kotin, P., and H.L. Falk, "The Role and Action of Environmental
Agents in the Pahtogenesis of Lung Cancer. I. Air Pollutants",
Cancer, 1£, 147-63 ~1959).
Kotin, P., and H.L. Falk, "Atmospheric Pollutants", Annual Rev. of
Med. 15, 233-54 (1964).
Levy, A., and S.L Miller, Final Technical Report on the Role of
Solvents in Photochemical Smog Formation, Scientific Circular
No. 799, National Paint, Varnish, and Lacquer Association,
Washington, D.C. 19~O. .
IV-52
-------
REFERENCESt CHAPTER IV (continued)
Lonneman, ~J.A., LA. Bellart and A.P. Altshuller, "Aromatic Hydro-
carbons i/1 the Atmosphere of the Los Angeles Basis"t Environ.
Sci. Techno1. ~(11), 1017-20 (1968).
Mach1e, W., "Gasoline Intoxication", J. Am. Med. Assoc..JJ..7 (23)t
1965-71 (1941).
t1oncrieff, R.W., The Chemical Senses, 3rd Ed., Leonard Hill, London,
1967.
Nader, J.S., "Current Techniques for Odor Measurement", Arch. Ind.
Health 17 (5)~ 537~41 (1958).
National Academy of Sciences, National Research Council, Evaluation
of the Hazard of Bulk Water Transportation of Industrial Chemicals,
1970
Patty, F.A., Ed., Industrial Hygiene and Toxicology, Vol. I,
Interscience Publishers, Inc., New York, 1948.
Patty, F.A., Ed., Industrial Hygiene and Toxicology, Vol. II.
Interscience Publishers. Inc.. Npw York. lQh?
Pennsylvania Department of Health, Short-Term Limits for Exposure to
Airborne Contaminants - A Documentation.
Pitts, J.N., "Environmental Appraisal-Oxidants, Hydrocarbons, and
Ox i de s 0 f Nit r 0 9 en", J. Air Poll u tio n Con t r 0 1 As s 0 c. 12. (9),
658-67 (l969).
Renzetti, N.A., and R.J. Bryan, "Atmospheric Sampling for Aldehydes
and Eye Irritation in Los Angeles Smog-1960", J. Air Pollution
Control Assoc. 11, 421-4, 427 (1961).
R 0 s san 0, A. T ., Jr. Ed., Air Poll uti 0 n Can t r 0 1 G u ide boo k for f1 a nag e men t ,
Environmental Science Service Division, ERA, Inc., Stamford,
Connecticut, 1969.
Sablieski, J.J., Odor Control in Kraft Mills. A Summary of the
State of the Art, U.S. Public Health Service, National Center
for Air Pollution Control, Cincinnati, Ohio, 1967.
IV-53
-------
REFERENCES, CHAPTER IV (continued)
Saffioti, U., F. Cefis, L.H.
Studies of the Conditions
Cancer Inductionll, J. Air
23-5 (1965).
Kolb, and P. Shubik, IIExperimental
of Exposure to Carcinogens for Lung'
Pollution Control Assoc. .!i(1),
Saw i c k i, E., ~1. C. E 1 be r t, T. R. H au s e r ,F . T. Fox, and T. W. S tan 1 ey ,
IIBenzo{a)pyrene Content of the Air of American CommunitiesW,
Am. Ind. Hyg. Assoc. J..li, 443-51 (1960).
Sawicki, E., T.R. Hauser, W.C. Elbert, F.T. Fox, and J.E. i~e'eke'r,
IIPolynuclear Aromatic Hydrocarbon Composit'bn of the Atmosphere. .
in Some Large American Citiesll, Am. Ind. IJyg. Assoc. J. ll,
137-44 (1962a) .
Saw i c k i, E., F. T. Fox, W. E 1 be r t, T. R. H au s e r, and J. t1 e eke r ,
IIPolynuclear Aromatic Hydrocarbon Composition of Air Polluted
by Coal-Tar Pitch Fumesll, Am. Ind. Hyg. Assoc. J. ll, 482-6{l962b).
Sax, N.I., Dangerous Properties of Industrial Materials, Reinhold,
New York, 1963. .
Shabad, L.M., IIExperimental Research on the Ling Between Atmospheric
Pollumn and Lung Cancerll, Intern. J. Air Pollution 1,211 (l960).
Smith, C.W., Acrolein, Wiley, New York, 1962.
.Smith, L.E., IIperoxyacetyl Nitrate Inhalation Cardiorespiratory
Effectsll, Ar.ch. .Environ. Health lQ., 161-4 (l965).
Stahl, Q.R., Preliminary Air Pollution Survey of Aldehydes, Litton
Systems, 1969.
Stephens, E.R., and F.F. Burleson, IIAna1ysis of the Atmosphere for
Light Hydrocarbonsll, J. Air Pollution' Control Assoc. 17,174-53
(l967). -
Stephens, E.R., IIChemistry of Atmospheric Oxidants II, J. Air Pollution
Control Assoc. li(3), 181-5 (1969).
Stokinger, H.E., and D.L. Coffin, IIBiological Effects of Air Pollutionll,
Air Pollution, Vol. I, 2nd 'Ed., A.C. Stern, Ed., Academic Press,
New York, 1968.
Sunshine, I., Ed., Handbook of Analytical Toxicology, Chemical Rubber
Company, 1969.
IV-54
-------
REFERENCES, CIIAPTER IV (continued)
Sullivan, R., Air Pollution Aspects of Odorous Compounds, Public
Health Service, NAPCA Pub1. No. APTD 69-42~ 1969.
Tanimura, H., "Benzo{a)pyrene in an Iron and Steel ~Iorks", Arch.
Environ. Health II, 172-7 (1968).
Turk, A., "Appraisa1 of Odor Problems", Air Repair i (55), August i954.
Urbach, E., II Odors ( 0 s my 1 s ) as All erg en i c Agents ", J . All erg y 1 3 ,
387 (1942).
u.S. Department of Health, Education and Welfare, Air Quality Criteria
for Hydrocarbons, NAPCA Publ. No. AP-64, 1970a.
u.s. Department of Health, Education and Welfare, Nationwide Survey
of Air Pollutant Emissions-1968, NAPCA Publ. No. AP-73, 1970.
vonOettingen, W.F., P.A. Neal, and D.O. Donahue, The Toxicity and
Potential Dangers of Toluene with Soecial Reference to its
Maximal Permissible Concentration, U.S. Public Health Service
Bulletin No. 279, 1940.
Wagner, W.O., O.J. Dobrozorski, and H.E. Stokinqer, "Antagonistic
Action of Oil t.1ists on Air Pol1utants", Arch: Environ. Health
£,523-34 (1961).
Wagner,.W.D:, P. Wright, and H.E. Stokinger, "Inhalation Toxicity
o f 0 11M, s t s. I. C h r 0 n icE f f e c t s 0 f W hit e r'l i n era 1 0 i 1 ", Am. I n d .
Hyg. Assoc. J. 25, 158-68 (1964).
Wright, G.W., LC. Lloyd, P. Howell, and R. Prindle, "Epidemologic
Study of Obstructive Pulmonary Disease in Two Small Towns",
Am. Rev. Resp. Disease !!E., 713 (1962).
IV-55
-------
REFERENCES
CHAPTER IV
C.
PHOTOCHEMICAL SMOG
Altshuller, A.P., "Reactivity of Organic Substances in At-
mospheric Photooxidation Reactions", Intern. J. Air Water
Pollution lQ., 713-33. (1966a).
Altshuller, A.P.,
termination of
Emissions", J.
(1966b).
"An Evaluation of Techniques for the De-
the Photochemical Reactivity of Organic
Air Pollution Control Assoc. ~, 257-60
Altshuller, A.P., and J.J. Bufalini, "Photochemical Aspects
of Air Pollution: A Review", Photochem. Photobiol. i,
97-146 (1965).
Altshuller, A.P., and J.J. Bufalini, "Photo chemical Aspects
of Air Pollution: A Review", Environ. Sci. Technol. i,
39-64 (1971).
Altshuller, A.P., and J.R. Cohen, Intern. J. Air \tJater
Pollute I, 787 (1963).
Altshuller, A.P., D.L. Klosterman, P.W. leach, J.J. Hindawi,
and J.E. Sigsby, "Products and Biological Effects From
Irradiation of Nitrogen Oxides With Hydrocarbons or Al-
dehydes Under Dynami c Condi ti ons ", Intern. J. Ai r \tlater
Pollute lQ., 81-98 (1966).
Altshuller, A.P., S.L. Kopczyski, W.A. lonnemann, and F.D.
Sutterfield, "A Technique for ~1easurin9 Photochemical.
Rea c t ; on sin .A t Q1 0 s ph e r1 c Sam p 1 e s ", En v 1 r 0 n. S c i T e c h n 0 l.
. i, 50 3 - 6 (l 9 70 a ) .
Altshuller, A.P., S.l. Kopczyski, W.A. lonnemann, F.D. Sutter-
field, and D.l. Wilson, "Photochemical Reactivities of
Aromati c Hydrocarbon-Ni trogen Oxi de Rel ated Sys terns ",
Environ. Sct. Technol. i, 44-49 (1970b).
Anon., Inter-Association News, "Manufacturing Chemists As-
sociation", J. Air Pollution Control Assoc. 19(12),
968-9 (1969).
IV-56
-------
REFERENCES, CHAPTER IV (continued)
Chass, R.L., "Trichloroethylene", J. Air Pollution Control
Assoc. gQ(2), 104 (1970).
Glasson, W.A., C.S. Tuesday, Paper presented at 150th
National Meeting, ACS, Atlantic City, N.J., September
1965.
Glasson, W.A. and C.S. Tuesday, "Hydrocarbon Reactivities
in the Atmospheric Photooxidation of Nitric Oxide", En-
viron. Sci. Technol. i, 916-24 (1970).
Heuss, J. M., and W. A. Gl asson, "Hydrocarbon Reacti vi ty aVid
Eye Irritation", Environ. Sc1. Technol. l, 1109-16 (1968).
Leighton, P.A., Photochemistry of Air PollLltion, Academic
Press, New York, 1961.
Levy, A., and S.E. Miller, Final Technical Report on the
~ of Solvents in Photochemical Smog Formation, Sci-
entific Circular No. 799, National Paint, Varnish and.
Lacquer Association, Washington, D.C., 1970.
Maynard, J.B., and W.N. Sanders, "Determination of the De-
tailed Hydrocarbon Composition and Potential Atmospheric
Reactivity of Full-Range Motor Gasolines", J. Air Pol-
lution Control Assoc. }2, 505-10 (1969).
Schuck, E.A., A~P. Alts.huller, 0.5. Barth, and G.B. Morgan,
"Relationship of Hydrocarbons to Oxidants in Ambient At-
mospheres", J. Air Pollution Control Assoc, 20(5),297-
302 (1970).
Schuck, LA., and G.J. Doyle, "Photooxidation of Hydrocarbon
in Mixtures Containing Oxides of Nitrogen and Sulfur Di-
oxide", Rept. No. 29, Air Pollution Foundation, San Marino,
Calif. 1959.
Stephens, E.R., and W.E. Scott, Proc. Amer. Petrol. Inst.
E, 665 (1962).
Stern, A.C., Ed., Air Pollution, Vol. I, II, III, Academic
Press, New York, 1968.
IV-57
-------
REFERENCES, CHAPTER IV (continued)
System Development Corporation, "Comprehensive Technical Re.
port on all Atmospheric Contaminants Associated with
Photochemical Air Pollution", TM-(L)-4411/002/01, June
1970.
Tuesday, C.S., Arch. Environ. Health I, 188 (1963).
IV-58
-------
REFERENCES
CHAPTER IV
D.
VEGETATIVE DAMAGE
Abeles, F.B., and H.E. Gahagan, "Abscission: The Role of
Ethylene, Ethylene Analogues, Carbon Monoxide and Oxygen",
Plant Physiol. 43: 1253-1258, August, 1968.
"Air Quality Criteria for Hydrocarbon", U.S. Dept. HEW, PHS,
NAPCA PubL No. AP-64 (1970).
"Air/Water Pollution Report", Sept. 28, 1970.
Altshuller, A.P. et al, "Products and Biological Effects
From Irradiation of Nitrogen Oxides with Hydrocarbons
or Aldehydes Under Dynamic Conditions", Intern, Jour.
Air Water Pollution 10: 81-98, 1966.
-.....
I .
Brermen, E.G., LA. Leone, and R.H. Daines, "Atmospheric
Aldehydes Related to Petunia Leaf Damagell, Science 143:
818-819 (1964). -
Clayton, G.D., and 1.S. Platt, "Evaluation of Ethylene as
a n Air Poll uta n t A f f e c t in 9 P 1 ant L i fe", Am. I n d. H Y g .
Assoc. J. 28, 151.9 (1967).
Crocker, W., Growth of Plants, Reinhold Pub. Corp., New York,
1948.
Crocker, W., A.E. Hitchcock, and P.W. Zimmerman, IISimilar-
iti"es in the Effects of Ethylene and the Plant Auxins",
Contributions From'Boyce Thompson Institute'7(3): 231-
248, JUly-,Sept. 1935.
Guderian, R.H., vonhaut, and H. Stratman, "Problems of the
Recognition and Evaluation of the Effects of Gaseous Air
Impurities on Vegetation", Z. Pflanzerk iankh, Pflanzen-
schutz, 67: 257:264, U.S. Public Health Service Rept.
A61-37. .
Haagen-Smit, A.J. et al, "Investigation on Injury to Plants
From Air Pollution in the Los Angeles Areall, Plant Physiol.
27: 18-34, 1952.
I V-59
-------
REFERENCES, CHAPTER IV (continued)
Hall, W.C., G.B. Truchelut, C.L. Lernweber, and F.A. Herrero,
"Ethylene Production by the Cotton Plant and Its Effects
Under Experimental and Field Conditions", Physiol. Plant-
ar;um, lQ 306(1957).
He c k, W. W 0' and E.. G. P ire s, II G row t h - 0 f P 1 ant s F u m i gat e d Wit h
Saturated and Unsaturated Hydrocarbon Gases and Their
Derivatives", Texas Agric. Exp. Sta., Misc. Publ. No. 603,
1962. . .
Heggestad, HoWo, "Diseases of Crops and Ornamental Plants
Incited by Air Pollutants", Phytopathology 58:1089-1096,
19680
Hindawi, LJ., and A.Po Altshuller, "Plant Damage Caused by
Irradiation of Aldehydes", Science 146:540-542, 1964.
Jaffe, L.So, "Effects of Photochemical Air Pollution on Veg-
etation .With Relation to the Air Quality Requirements",
J. Air Pollution Control Assoc. 17:38-42, 1967.
Knight, L.I., and W. Crocker, "Toxicity of Smoke", Botan.
G a z. ~: 33 7 - 371, May 1 9 1 3 .
Lacasse, N.L., and WoJ. Moroz, Eds., Handbook of Effects
Assessment, Vegetation Damage, Center for Air Environment
Studies, Pennsylvania State University (1969)0
Litton Systems, "Preliminary Air Pollution Survey of Ethy1ene",
NAPCA Publo No. APTD 69-35 (1969).
Schonbeck, H., "A Method for Determining the Biological Ef-
fects of Air Pollution by Transplanted Lichens"; Staub-
Reinhold, Luft .li, No.1, Jano 1969.
Stephens, E.R., and FoR. Burleson, "Analysis of the Atmos-
phere for Light Hydrocarbons", J. Air Pollution Control
Assoc. ll: 147-153, March 19670.
Stephens, E.Ro, LFo Darley, and F.R. Burleson, "Sources and
Reactivity of Light Hydrocarbons in Ambient Air", Proc.
Div. Refining, Amero Petrol. Inst. 47: 466-483, 1967.
IV-60
-------
REFERENCES, CHAPTER IV (continued)
Sterno' A.C., Editor, Air Pollution, Vol. I., Academic Press,
New York, 1968 (Chapter 12, 13randt, C~S., and Heck, W.W.,
"Effects of Air Pollutants on Vegetation").
Taylor, D.C.. "Importance of Peroxyacetyl Nitrate (PAN)'as a
Phytotoxic Air Pol1utant""J. of Air Pollution Control
Assoc. 19: . 347-351 (1969L
Zimmerman, P.W., "Anaesthetic Properties of Carbon Monoxide
and Other Gases in Relation to Plants, Insects and Centi-
pedes", Contributions From the Boyce Thompson Institute
7(2): 147-155, April-June 1935.
IV-61
-------
REFERENCES
CHAPTER IV
B.
HEALTH EFFECTS - SUPPLEMENTAL BIBLIOGRAPHY
Altshuller, A.P., 1. R. Cohen, S.F. Sliva, and S-.L. Kopczynski,
IIAir Pollution: Photooxidation of Aromatic Hydrocarbonsll,
Science ~, 442 (1962).
A1tshuller, A.P., IIAtomospheric Reaction Studies Re1ated,to Air
Pq11utionll, Arch. Environ. Health 8,27-30 (1964).
A1tshuller, A.P., D.L. Klosterman, P.W. Leach, I.J. Hinda\'Ji and
J . E. S i g s by, Jr., II Pro d u c t san d, B i 0 log i c alE f f e c t s fro m
Irradiation of Nitrogen Oxides with Hydrocarbons or Aldehydes
under Dynamic Conditionsll, Intern. J. Air Water Pollution 10,
81-91 (1966). -
Appell, L., IIII. Odor Intensityll, Am. Perfume and Cosmetics 7..1,
43-8 (1964).
Ayer, H.E., IISamp1ing Methods for Oil Mist in Industryll, Am.
Ind. Hyg. Assoc. J. li, 151-7 (1964).
Baird, V.C., IIEffects of Atmospheric Contamination on Cancer
r~ortality' in Petroleum Refinery Employeesll, J. Occ. Med. 9
(8), 415-23 (1967).
Bartok, H., A.R. Crawford, A.R. Cunningham, H.J. Hall, E.H. Manny,
and A. Skopp, IIStationary Sources and Control d Nitrogen Oxide
Emissions~ Proceedings Digest, Second International Clean
Air Congress, December 6-11, 1970.
Benforado, D.M., \4.J. Rotella, and D.L. Horton, IIDevelopment of
an Odor Panel for Evaluation of Odor Control Equipmentll, J.
Air Pollution Control Assoc. .l2. (2), 101(1969).
Bingham, E., and H. Fa1k, IIEnvironmenta1 Carcinogens: The Modi-
fying Effect of Carcinogens on the Threshold Responsell,
Arch. Environ. Health .l2., 779-83 (1969).
Bretscher, H., IIE1imination of Malodorous Emissions in the
Chemical Industryll, The Chemical Engineer ~ (7),268-70 (1968).
Brooman, R.O. and E. Edgerly, Jr., IIConcentration and Recovery
of Atmospheric Odor Pollutants Using Activated Carbonll,
J. Air Pollution Control Assoc. ~, 25 (1966).
Bufa1ini, J.J., and A.P. A1tshu11er, liThe Effects of Temperature
on Photochemical Smog Reactionsll, Intern. J. Air'Water
Pollution I, 769-71 (1963).
IV-62
-------
REFERENCES, CHAPTER IV (continued)
Bufalini, J.J., and A.P. Altshuller, "Synergistic Effects in
the Photooxidation of Mixed Hydrocarbons", Environ. Sci.
Technol. 1(2), 133-8 (1967).
Burton, G., M. Corn, J. Bernard, L. Gee, C. Vasa10, and A.
Thomas, "Response of Healthy Men to Inhaled Low Concen-
trations of Gas-Aerosol Mixtures", Arch. Environ. Health
~(4), 681-2 (1969).
Calvert, W.R., "Catalysts at Work Eliminating Air Pollutants",
Proceedings Air Pollution Control Assoc., Paper No.9,
p. 49, 1956.
Caplan, J.D., "Causes and Control of Automotive Emissions",
Proc. Instr. Mech. Engrs. (A.P.) No.7, 1962-1963.
Carnow, B.W., R. Senior, R. Karsh, S. Hessler, and L. Avioli,
"The Role of Air Pollution in Chronic Obstructive Pulmonary
Disease", J. Am. Med. Assoc. 214(5), 894-9 (1970).
Clarke, G.S., L.O. Emik, and R.L. Plata, "The Atmospheric Con-
taminant Peroxyacety1 Nitrate. Acute Inhalation Toxicity
in Mice", Arch. Environ. Health 15(6), 739-44 (1967).
Commins, B.T., "Polycyclic Hydrocarbons in Rural and Urban Air",
Intern. J. Air Pollution 1, 14-7 (1958).
Cooper, R.L., and A.J. Lindsey, "Air Pollution by Polycyclic
Hydrocarbons", Chemistry and Industry, 1177-78, October
31, 1953.
Darley, E.F., J.P. Middleton, and M.J. Garber, "Plant Damage
and Eye Irritation from Ozone-Hydrocarbon Reactions",
Agricultural and Food Chemistry 8(6), 483-5 (1960).
Davis, W.H., "Po11ution: Whence and Whither", Arch. Environ.
Health £1, 3-4 (1970).
DeMaio, L., and M. Corn, "Polynuclear Aromatic Hydrocarbons
Associated with Particulates in Pittsburgh Air", J. Air
Pollution Control Assoc. 16, 67 (1966). .
, .
Douglas, J.S., P. Ridgway and M. Dennis, "Histamine Release
by Ai r Po 11 uta n t s II , Arch . En vir 0 n. He a 1 t h 1 8, 62 7 - 30 (1 96'9 ) .
Duprey, R.L., Compilation of Air Pollutant Emission Factors,
Public Health Service Publ. No. 999-AP-42, 1968.
IV-63
-------
REFERENCES, CHAPTER IV (continued)
Eckardt, R.E., "Air Pollution, Lung Cancer, and Chronic Lung
Disease", J. Occ. Hed. ~(4), 184-8 (1964).
Edwards, F.R., "Oxidation Processes for Controlling Fumes and
Odors", Plant Eng. 25(14),49-50 (1971).
Ehrlich, R., and M.C. Henry, "Chronic Toxicity of Nitrogen
Dioxide", Arch. Environ. Health li, 860 (1968).
Ellis, C.F., "Sunlight Photooxidation of Isobutylene and
Butene-l with N02 at Low- and High-Oxygen Concentrations",
Atmos. Environ. i, 83-8 (1971).
Endow, N., G.J. Doyle, and J.L. Jones, "The Nature of Some
~1 0 del Ph 0 t 0 c hem i c a 1 A e r 0 sol s .., J. Air Poll uti 0 n Con t r 01
Assoc. 1l(4) 141-7 (1963).
Epstein, S.S., "Photoactivation of Pol.ynuclear Hydrocarbons",
Arch. Environ. Health lQ, 233 (1965).
Epstein, S.S., "Carcinogenicity of Organic Extracts of Atmos-
p her i c Pollutants", J . Air P 0,11 uti 0 n Con t r 0 1 Assoc. 1 7 ,
728 (1967).
Falk. H.L., P.E. Steiner, S. Goldfein, A. Breslow and R. Hykes,
"Carcinogenic Hydrocarbons and Related Compounds in Pro-
cessed Rubber", Cancer Research li, 318 (1951).
Falk, H.L., 1. Morkul, and P. Kotin, "Aromatic Hydrocarbons.
IV. Their Fate Following Emission into the Atmosphere and
Experimental Exposure to Washed Air and Synthetic Smog",
Arch. Environ. Health ll, 13-7 (1956).
Falk, H.L., and P. Kotin, "The Chemical and Biological Con-
sideration of Atmospheric Carcinogenic Agents", J. Air
Pollution Control Assoc. I, 12(1957).
Falk, H.L., P. Kotin, and 1. Morkul, "The Disappearance of
Carcinogens from Soot in Human Lungs", Cancer li, 489 (1958).
Falk, H.L., P. Kotin, and A. Mehler, "Polycyclic Hydrocarbons
as Carcinogens for Man", Arch. Environ. Health !' 721-30' (1964).
Feldstein, M., and S. Balestrieri, "The Detection and Estimation
of Parts per Billion Concentrations of Hydrocarbons", J.
Air Pollution Control Assoc. li(4) 177-8 (1965).
Feldstein, M., "Toxicity and Analysis of Air Pollutants", J.
Forensic Sci. !i(3), 337-51 (1969).
IV-64
-------
REFERENCES, CHAPTER IV (continued)
Freeman, S.K., "Odor", Intern. Sci. and Technol. 70-80,
September 1967.
Fox, E.A., and V.E. Gex, "Procedure for Measuring Odor Con-
centration in Air and Gases", J. Air Pollution Control
Assoc. I, 60-1 (1957).
Glasson, W.A., and C.S. Tuesday, "Hydrocarbon Reactivity in
the Atmospheric Photooxidation of Nitric Oxide"~ Environ.
Sci. Technol. i, 916 (1970).
Glasson, W.A., and C.So Tuesday, "Reactivity Relationships of
Hydrocarbon Mixtures in Atmospheric Photooxidation"u Environ.
Sci. Techno1. ~, 151-4 (1971).
Goldman, A., liThe Odor Phase of Air Po11ution", Ind. Wastes 1,
217-20 (1956).
Goldsmith, J.R., "Air Pollution Epidemio10gy", Arch. Environ.
Health ~, 516-22 (1967).
Gross, P., liThe Relation of Chronic Lung Disease to Community
Type Air Pollution", J. Air Pollution Control Assoc. 15(4),
160-1 (1965).
Gross, P., "Breathing Air - Ho\'l Pure Need It Be?", presented at
the 34th Annual Meeting, Industrial Hygiene Foundation,
Pittsburgh, Pa., October 14-15, 1969.
Hammond, E.C., O. Averback, D. Kirmon, and L. Garfinkel, "Effects
of Cigarette Smoking on Dogs. I. Design of Experiment
Mortality and Findings in Lung Parenchyma", Arch. Environ.
Health £1, 740-53 (1970).
Hockheiser, S., F.J. Burmann, and G.B. Morgan, "Atmospheric
Surveillance: The Current State of Air Monitoring T~chno10gy",
Environ. Sci. Technol. 5(8),678-84 (1971).
Hodgson, T.A., Jr., "Short Term Effects of Air Pollution on
Mortality in New York City", Environ. Sci. Technol. 4(7),
589-97 (1970). -
Hore, 1., and D.E. Gibson, "Ozone Exposures and Intelligence
Tests", Arch. Environ. Health 17(1),77-9 (1968).
Horstman, S.W., R.F. Wromole, and A.N. Heller, "Identification
of Community Odor Problems by Use of an 0 ~erver Corps",
J. Air Pollution Control Assoc. 15(6), 261 (1965).
IV-65
-------
REFERENCES, CHAPTER IV (continued)
Horton, H., R. Tye, and K. Stemmer, "Experimental Carcinogenesis'
of the Lung. Inhalation of Gaseous Formaldehyde on an Aerosol
of Coal Tar by C3H Miceli, J. National Cancer Institute 30
(1), 31-42 (1963). --
Hueper, W.C., "Carcinogenesis in the Human Environment", Arch. .
Pathology Il, 237-67 (1961).
Huess, J.M., and W.A. Glasson, "Reactivity and Eye Irritation",'
Environ. Sci. Technol. 2, 1109 (1968).
Huey, N.A., L.C. Broering, G.A. Jutze, and C.W. Gruber,"Objective
Odor Pollution Control Investigations", J. Air Pollution
Control Assoc. lQ.(6) , 441 (1960).
Jaff~, L.S., liThe Biological Effects of Ozone on Man and Animals~
Am. Ind. Hyg. Assoc. J. ~(3), 267-77 (1967).
Jaffe, L.S., "The Biological Effects of Photochemical Air Pollution
on Man and Animals", Am. J. Public Health i?. (8),1269-77 (1967).
Jaffe, L.S., "Photochemical Air Pollutants and Their Effects.II.
Adverse Effects", Arch. Environ. Health li, 241 (1968).
Jones, K.H., J.F. Thomas, and D.L. Brink, "Control of Malodors
from Kraft Recovery Operations by Pyrolysis", J. Air Pollution
Control Assoc. 12.(7), 501-4 (1969).
Kaiser, E.R., "Odor and Its Measurement", Air Pollution, Vol. I,
A.C. Stern, Ed.. Academic Press, New York, 1962.
Katz, M., and J.L. ~10nkman, "The Organic Fraction of Particulate
Pollution", Occup. Health Rev. (Ottawa) li(l), 3-16 (1964).
Kneip, T.J.. M. Eisenbud, and C.D. Strehlow, "Airborne Particu-
lates in New York City", J. Air Pollution Cont~ol Assoc. 20
(3), 144-9 (1970).
Kotin, P., H. Falk, P. ~1ader, and M. Thomas, "Aromatic Hydro-
carbons. 1. Presence in Los Angeles Atmosphere and Car-
cinogenicity of Atmospheric Extracts". Arch. Ind. Hyg. Occup.
Med. i, 153 (1954).
Kotin, P., and H.L. Falk, "Local and Systemic Responses to Ex-
perimental Exposure to Atmosphere Pollutants", Proceedings
Third National Air Pollution Symposium, Pasadena, Cal. .
April 18-20, 1955.
IV-66
-------
REFERENCES, CHAPTER IV (continued)
Kotin, P., and H.L. Falk, "Host Factors in Relation to the
Action of Environmental Carcinogenic Agents", Arizona
Medicine 19(5), 95-101 (1962).
Kotin, P., and H.L. Falk, "Atmospheric Factors in the Patho-
genesis of Lung Cancer", Advan. in Cancer Reso I, 475 (1963).
Lave, L.B., and E.P. Seskin, "Air Pollution and Human Health",
Science 169, 723-32 (1970).
leonardos, G., D. Kendall and N. Barnard, "Odor Threshold De-
terminations of 53 Odorant Chemicals", J. Air Pollution
Control Assoc. 19 (2),91-5 (1969).
Los, L.I., A.G. Sadovnikova, RoM. Soboleva, and Do Va. Turets,
liThe Quality of Atmospheric Air in the Vicinity of Petroleum
Refinery Plants", Gigiena i Sanit., No.8, 8-13 (1950).
MacEwan, J.D., and R.P. Geckler, "Comparative Studies of gO-Day
Continuous Exposure of 03, N02 and CC14 at Reduced and
Ambient Pressures", Air Force Systems Command, Aerospace
Medical Research Laboratories, AMRL-TR-67-68, Feburary 1968.
Mac h 1 e, H., H. N. ~1 a c Fa r 1 and, and R. E. E c k a r d t, II M a j 0 r C hall en 9 e
in Setting S02 Standards in Avoiding the Simplistic Approach",
Environ. Sci. Technol. 3(6), 543-5 (1969).
r1acKenzie, V.G., liThe Power Industry and Air Pollution", 1962
Annual Meeting, Association of Edison Illuminating Companies,
Boca Raton, Florida, November 1962.
Manufacturing Chemists Association, "Odor Measurement and Control",
Air Pollution Abatement Manual, Washington, D.C., 1960.
Matthews, D.S., and J.J. Schveneman, "Management of Dade County's
Air Resources", U.S. Department of Health, Eduation and
Welfare, Robert A. Taft Sanitary Engineering Center,
Cincinnati, October 1962.
Matsushita, H., H. Arito, Y. Suzuki, and R. Sods, "Determination
of Threshold Values for Olfactory Perception of Primary
Odor Substances", Ind. Health ~, 221 (1967).
May, J., "Odor Thresholds of Solvents for Assessment of Solvent
Odors in the Air", Staub ~(9), 34 (1966).
McCord, C.P., and W.N. Witheridge, Odor Physiology and Control,
1st Ed., McGraw-Hill Book Co., Inc., 1949.
IV-57
-------
REFERENCES, CHAPTER IV (continued)
McKee, H.C., and F.W. Church, IIParticulate Standards Keyed to
Visibility and Citizen Complaints are Interim t1easurell,
Environ. Sci. Technol. 1{6}, 542 (1969).
Middleton, J.T., IITrends in Air Pollution Damagell, Environ.
Health 8, 19-23 (1964).
Mills, J.L., R.T. Halsh, K.D. .Leudtke, and L.K. Smith, IIQuan-
titative Odor Measurementll, J. Air Pollution Control Assoc.
11 (10) 467-75 (1963).
Moncrieff, R.W., IIIndustrial Odors. Part III. Industrial Odor
Counteractionll, Ind. Water and Wastes 6,195-9 (1961)
t10 n k man, J. L., G. E. t'10 r e, and M. Kat z, II A n a 1 y s i s 0 f Pol y c y c 1 i c
Hydrocarbons in Particulate Pollutantsll, Am. Ind. Hyg. Assoc.
J. £1, 487-95 (1962).
Morrow, P.E., IIEvaluation of Inhalation Hazards Based on the
Respirable Dust Concept and the Philosophy and Application
of Selective Samplingll, Am. Ind. Hyg. Assoc. J. 25,213-36
(l964). .-
t1ountain, 1., E. Cassel, D. Wolter, J. Mountain, J. Diamond,
and J. ~1cCarroll, IIHealth and the Urban Environment. III.
Air Pollution and Disease Symptoms in the 'Normal Population'",
Arch. Environ. Health .J.l.(3), 343-52 (1968).
~1udd, J.B., "Effects of Environmental Agents at the Enzyme Level:
Air Pollutantsll ~sioloqical Characterization of Health
Hazards in Man's Environment, Section IV, National Academy
of Sciences, National Research Council, unofficial publication,
1966. .
~1ueller, P.K., and r.1. Hitchcock, "Air Quality Criteria - Toxico-
logical Appraisal for Oxidants, Nitrogen Dioxide and Hydro-
carbonsll, J. Air Pollution Control Assoc. 19 (9), 690-5 (1969).
Nau, C.A., J. Neal, and V. Stembrid~e, IIA Study of the Physio-
logical Effects of Carbon Blackll, Arch. Ind. Health .J.l.,
21-8 (1958).
Ottoson, D.G.R., "How We Recognize Odors", New Scientist li,
114-6 (1970).
Pennsylvania Department of Health, and U.S. Department of Health,
Education and Welfare, Public Health Service, "Pure Air For
Pennsylvaniall.
IV-68
-------
REFERENCESt CHAPTER IV (continued)
Pierrard, J.M., "Environmental Appraisal-Particulate Matter,
Oxides of Sulfur, and Sulfuric Acid", J. Air Pollution Control
Assoc. 19(9), 623-6 (1969).
Project Clean Air, Task Force No.2, "Human Health Effects,
Task Force Assessment", September 1, 1970.
Pybus, F.C., "Cancer and Atmospheric Pollution", Sanitation,
169-80, December 1963.
Rahn, J., J.J. Weslo\'lski, H. John, and H.R. Ralston, IIDiwinal
Variation of Aerosol Trace Element Concentrations in Livermore,
Californiall, J. Air Pollution Control Assoc. 21 (7), 406~9
(1971). -
Rap pap 0 r t, B. Z . and r~. M . H 0 f f man, II U r tic a ria Due to A 1 i p hat i c
Aldehydesll, J. Am. Med. Assoc. ill, 2656 (1941).
Ray, E.r-1., "Sampling and Determination by Ultraviolet Absorption
of Oil Mists and Solvent Vaporsll, Am. Ind. Hyg. Assoc. J. 31,
472-8 (1970). -
Rich, T.A., IIParticles and Air Pollutionll, J. Aerosol Sci. 2
(2), 183-92 (1971).
Roberson, J.E., liThe Effect of Odor Control on a Kraft Mill
Energy Balancell, J. Air Pollution Control Assoc. 20(6),
373-6 (1970). . -
Rogers, L.H., IIReport on Photochemical Smogll, J. Chern. Educ. 35,
310 (1958). -
Sawicki, E., "The Separation and Analysis of Polynuclear Aromatic
Hydrocarbons Present in the Human Environment", Chemist-Analyst
53, 24-30 (1964).
Schellinger, R.R., IIAir Pollution Health Effects", Plant Eng.. ~
(4), 88 (1970).
Schuck, E.A., E.R. Stevens and J.T. Middleton, "Eye Irritation
Responses at Low Concentrations of Irritantsll, Arch. Environ..
Hea 1 th li, 570 (1960). .
Sern, M. and R. B. Prattle, IIEffects of Possible Smog Irritants
on Human Subjects", J. Am. Med. Assoc. 165 (15), 1908 (1957).
IV-69
-------
REFERENCES, CHAPTER IV (continued)
Skinner, C.F., IIProperties of Airborne Po11utantsll, Plant Eng.
£1 (25), 56-8 (1969).
Skinner, C.F., 1I0dor Determination Evaluation and Contro111,
Plant Eng. 24, 66-8 (1970).
Speizer, F.E., IIAn Epidemiological Appraisal of the Effects of
Ambient Air on Health. Particulates and Oxides of Su1furll,
J. Air Pollution Control Assoc. ~ (9), 647-56 (1969).
Stalker, W.W., IIDefining the Odor Problem in a Community", Am.
Ind. Hyg. Assoc. J. £i, 600-5 (1963).
Stephens, LR. and W. E. Scott, "Relative Reactivity of Various
Hydrocarbons in Polluted Atmospheres", Am. Petroleum Inst.
Proceedings 1£, 665 (1962).
Stocks, P., B.T. Commins, and K.V. Aubrey, "A Study of Polycyclic
Hydrocarbons and Trace Elements in Smoke in Merseyside and
Other Northern Loca1itiesll, Intern. J. Air Water Pollution 4,
141-53 (1961). -
Stokinger, H.E" "Evaluation of the Hazards of Ozone and Oxides
of Nitrogenll, Arch. Environ. Health li, 181 (1957).
Stokinger, H.E" "Ozone Toxicology. A Review of Research and
Industrial Experience 1954-1965", Arch. Environ. Health 10,
719-31 (1965). -
Stone, R., "Sewage Treatment System Odors and Air Pollution",
J. Sanitary Eng. Div. Am. Soc. Civil Eng. 96 (SA 4), 905-9
(1970). -
S t r a u s s, l~., " The De vel 0 p men t 0 f a Con den s e r for 0 d 0 r Con t r 0 1
From Dry Rendering Plants", J. Air Pollution Control Assoc.
11 (10) 424 (1964).
Stresen-Reuter, J., "Catalytic Incinerator Controls Hydrocarbons
and Odors", Plant Eng. £1(8), 142 (1969).
S wan n, H. E., Jr., D . B run 0 1, and O. J. Sa 1 c hum, II P u 1 m 0 n a r y R e -
sistance Measurements of Guinea Pigs", Arch. Environ. Health
~, 24-32 (1965).
Swann, H.E., D. Bruno1, L.G. Wayne, and O.J. Ba1chum, "Biological
Effects of Urban Air Po11ution.II. Chronic Exposure of Guinea
Pigs", Arch. Environ. Health 11(6),765 (1965).
IV-70
-------
REFERENCES, CHAPTER IV (continued)
Tabershaw, I.R., F. Ottoboni and H.C. Cooper, 1I0xidants: Air
Quality Criteria Based on Health Effectsll, J. Occup. Med.
469-80, September 1968.
Tabor, E.C., T. Hauser, J.P. Lodge and R.H. Burttschell,
"Characteristics of the Organic Particulate Matter in the
Atmosphere of Certain American Cities", Am. Med. Assoc.
Arch. Ind. Health Jl., 58 (1958).
Tanimura, H., "Study on the Relation Between Benzo(a)pyrene of
the Suspended Matters and That of the Falling Matters in the
Atmosphere", Japan J. Hyg. £1, 269-78 (1966).
Tabbens, B.D., J.F. Thomas, and tL t.1ukai, "Hydrocarbon Synthesis
in Combustion", Am. ~1ed. Assoc. Arch. Ind. Health li, 567. (1956).
Teller, A.J., "Odor Abatement in the Rendering and Allied In- .
dustriesll, J. Air Pollution Control Assoc. li(4), 148 (1963).
Telling. H., IIAir Pollution Caused by Coking Plants and Gasworks",
Energietechnik Jl. (12), 556-9 (1967).
Thomas, M.D., O.B. Wimmer, and A.P. Altshuller, IIReactivities of
Smog Components Are Central Issue in Setting Control Standards",
Environ. Sci. Technol. ~(7), 629-33 (1969).
Tripler, A.B., Jr., and G.R. Smithson, Jr., IIA Review of Air
Pollution Problems and Control in the Ceramic Industries",
Preprint American Ceramic Society, Columbus, Ohio, May 1970.
Turk, A., "Odor Control Methods: A Critical Review", Symposium
on Odor,Special Technical Publication No. 164, Am. Soc.
Testing Materials, 1954.
T u r k, A., "C 0 n t r 0 1 - 0 d 0 r s ", Am. ~1 e d. Ass 0 c. Arc h . I n d. He a 1 t h Jl.,
542-4 (1958).
Tye, R., A.W. Horton, and I, Rapien, "Benzo(a)pyrene and Other
Aromatic Hydrocarbons Extractable from Bituminous Coal", Am.
Ind. Hyg. Assoc. J. ?2, 25-8 (1966).
Tye, R., t~.J. Burton, E. Bingham, Z. Bell, and W. Horton,
IICarcinogens in a Cracked Petroleum Residuumll, Arch. Environ.
Health li, 202 (1966).
IV-71
-------
REFERENCES, CHAPTER IV (continued)
U.S. Department of Health, Education, and Welfare, Air Quality
Criteria for Particulate Matter, NAPCA Pub1. No. AP-49, 1969.
U.S. Department of Health, Education and Welfare, Air Quality
Criteria for Photochemical Oxidants, NAPCA Pub1. No. AP-63,
1970. .
U.S. Department of Health, Education and Welfare, Air Quality
Data, 1964-1965, Public Health Service, 1966.
. .' .
U.S. Department of Health, Education and Welfare, Guide to
Research .i.!!. Air Pollution, 1969, Publ)c. Health Service; 1969.
Urbach, E., "Odors (osmyls) as Allergenic Agents", J. Allergy
li, 387 (1942). ..'
Vandegrift, A.E., l.J. Shannon, E.E. Sallee, P.G. Gorman, and
W.R. Park, "Particulate Air Pollution in the United State~",
J. Air Pollution Control Assoc. £1(6), 321-8 (197.1).
vonlehmden, D.J., R.P. Harigebrauc~, and J.E.Meeker, "Polynuclear
Hydrocarbon Emissions from Selected Industrial .Processes",
J. Air Pollution Control Assoc. li,306 (1965). . .
Waller~ R.E., and.B.T. Commins, "Siudies rif.Smoke ~nd Polycyclic
Aromatic Hydrocarbon Content of Air of large Urban Areas",.
Environ! Res. l, 295 (1967).
Walther, J.E. and H.R. Amberg, "A Positive Air Quality Control
Progra~ at a New Kraft Mill", J. Air Pollution Control Assoc.
£Q.(ll)~9:-18 (1970). .
Wang, C.C. ~nd G.V. Irons, "Acute Gasoline Intoxication", Arch.
Environ. Health £, 714-6(1961).
.. .
Wayne, L.G., "Eye Irritation as a Biological Indicator of Photo-
chemical Reactions in the Atmosphere", Atmos. Environ.l(2),
97-104 (1967). .
Wayne, l.G. and l. Chambers, "Biological Effects of Urban Pollution.
V. A Study of Effects of Los Angeles Atmosphere 'on laboratory
Rodents", Arch. Environ. Health 16, 871-85 (1968). ..
Welch, G.B., Air Pollution in the National Capitol Area, U.S.
Department of Health, Education and Welfa~e, Public Health
Service Pub1.No. 955, 1962. .
IV-72
-------
REFERENCES, CHAPTER IV (continued)
Wilby, F.V., "Variation in Recognition Odor Threshold of a Panel",
J. Air Poll~tion Control Assoc. }2.(2) , 96 (1969).
Winkelstein, H. and M. Gay, "Suspended Particulate
Arch. Environ. Health g(l), 174-7 (1971).
Wohlers, H.C., "Odor Intensity and Odor Travel From Industrial
Sources", Intern. J. Air Water Pollution I, 71-8 (1963)..
Air Pollutionll,
Wohlers, H.C., "Recommended Procedure for Measuring Odorous Con-
taminants in the Fieldll, J. Air Pollution Control Assoc. 17
(9), 609-13 (1967).
Wolkansky, P.M., IIPulmonary Effects of Air Pol1utionll, Arch.
Environ. Health }2.(4), 586-92 (1969).
Wright, G.~L, IIAn Appraisal of Epidemiologic Data Concerning the
Effect of Oxidants, Nitrogen Dioxide, and Hydrocarbons Upon
Human Populationsll, J. Air Pollution Control Assoc. .19(9),
679-89 (1969). . '-
Wynder, LL. and D. Hoffman, IISome Laboratory and Epidemiologic
Aspects of Air Po11utionll, J. Air Pollution Control Assoc.
li(4), 155-9 (1965).
Zeidberg, L.D., R.M.J. Horton, and E. Landau, liThe Nashville
Air Pollution Study. V. Mortality From Diseases of the
Respiratory System in Relation to Air Po11utionll, Arch.
Environ. Health li, 214 (1967).
Zimmer, C.E. and R.I. Larsen, IICalculating Air Quality and its
Controlll, J. Air Pollution Contr61 Assoc. 15, 562-72 (1965).
IV-73
-------
v.
SYSTEMS MODEL
A.
INTRODUCTION
This chapter describes the development of a sys-
tems model designed to quantitatively set forth the costs
and benefits to society of hydrocarbon pollutant control
systems. The model development was specifically oriented
toward stationary source hydrocarbon emissions; however
the model is general ~nd' can be adapted to other types of
air pollutant emissions.
The lack of sufficient data to quantify the
specific cost/benefit relationships resulted in a descriptive
model useful primarily for the identification of existing
data gaps and deficiencies to be filled by future R&D pro-
grams.
The specific goals of the modeling effort, ex-
pressed as desired outputs of the model, were:
. Definition of the socio-economic costs
of hydrocarbon pollution
. Quantification of measures of control
effectiveness
. Benefit/cost evaluation of control sys-
tems or control alternatives
. Enumeration of data deficiencies
The methodology of cost/benefit analysis of air
po1~ution and the associated problem areas have been dis-
cussed by a number of authors, ego Kneese and Wolozin
(1966); Kneese (1967); Lave (1970); and Ridker (1967a).
The anticipated problems, common to regulatory
benefit/cost analyses, included: the non-existence of a
market for determination of the benefits of pollution re-
duction; the measurement of economic externalities; and
the determination of realistic discount rates. Particular
attention was paid to factors such as double counting,
distributional effects, and discount rates, detailed later
under assumptions.
V~l
-------
Additionally, the review of adverse effects of
hydrocarbon pollutants showed that specific input data was
lacking to establish valid relationships between hydrocar-
bon pollutants and the adverse effects on society r~sult-
ing from their emission. As discussed in detail in Chapter
IV, documented human health effects of low ambient concen-
trations of hydrocarbon pollutants are limited to observ-
ations of eye and respiratory irritation from exposure to
the low molecular weight aldehydes. The major adverse
health effects derive from the secondary pollutants result-
ing from atmospheric photochemical reactions which involve
not only hydrocarbons but other reactive atmospheric pol-
lutants as well. Adverse effects on vegetation result al-
most solely from exposure to ethylene and the photochemical
oxidants in smog.
The model examines damage caused by hydrocarbon
emissions from stationary sources in five basic categories;
damage to humans, animals, materials, vegetation and also
indirect damage. "To the extent possible within the limit-
ations of the input data, eight functional relationships
were developed for each of the five categories. These re-
lationships are:
1.
Socio-economic costs as a function of
the adverse effects on the five af-
fected categories (humans, animals,
etc.).
2.
Adverse effects as a function of am-
bient pollutant concentration and syn-
ergism.
Ambient pollutant concentration as a
function of concentration in source
effluent and of meteorological con-
ditions.
3.
4.
Socio-economic costs as a function of
point-source effluent concentrations,
volume of emissions, and meteorological
conditions.
5.
Socio-economic costs as a function of
multiple-source synergism.
Socio-economic benefits as a function
of reduction of pollutant emissions.
6.
V-2
-------
7.
Reduction of pollutant concentrations as
a function of control systems costs.
Socio-economic benefit for reduction in
concentration of a specific pollutant as
a function of the control system cost.
8.
By formulating these functional relationships
for the affected categories and distributing them according
to the distributions of the categories within the given
demographic sector, one can compute the benefit-cost re-
lationships for a given stationary source. By looking at
the benefit-cost relationships as functional (distributional)
values rather than purely specific (single-point) values,
one can readily see the tradeoffs required to determine an
optimal allocation of resources.
Although the necessary input data were lacking
for the complete modeling of the cost/benefit relationships,
the methodology is illustrated by two examples; human health
effects of formaldehyde exposure and vegetation damage from
ethylene exposure. The assumptions required in developing
these examples limit their usefulness and they are presented
solely for illustration.
8 .
MODEL DEVELOPMENT
This section describes the mathematical model that
was developed to quantify the benefit-cost relationships of
hydrocarbon pollutant emissions from stationary sources and
corresponding control systems. The underlying concepts,
background thoughts, and the general process related to the
model's development are explained.
To decrease the chances of "double-counting" (i .e.,
including the same effect in more then one category) and to
allow inclusion of all possible effects, the benefit-cost
factors were assigned to one of five categories or adversely
affected groups. These categories were chosen to be as
orthogonal (disjoint and non-additive) as possible. The
five categories encompass costs of damage to (1) humans, (2)
materials, (3) animals, and (4) vegetation, and also include
(5) indirect costs. Descriptions of these categories follow.
1 ~
Human - Effects on humans are divided into
two sub-catego~tes; psychic and physical.
The psychic sub-category includes the ef-
fects of odors, loss of visibility and the
psychic aspects of physical damage to human
V- 3
-------
health as well as to the other adversely
affected categories (materials, animals,
etc.). Psychic effects and their assoc-
iated~costs cannot usually be express~d
in quantitative terms and thus were recog-
nized but not included in the model.
The physical sub-category includes: sensory ir-
ritations (eye, nose, throat, etc.); lost working time;
morbidity (illness or health damage more severe than that
causing lost working time); and mortality. There are two
possible ways to estimate the cost of physical effects on
humans: (a) to compute the actual medical costs and/or
the lost time costs and use these: as ,the cost to the per-
son; and (b) to determine the amount a person would be will-
ing to pay in order to avoid lost time and/or illness. The
latter method probably provides the best estimate of the
benefits that may be attained by reduction of human damages
due to air pollution (Lave, 1970), but the data in the form
of (a) above are much more readily obtainable (Rice, 1966;
Ridker, 1967a; Kneese, 1966a). For reaso"s of availability
only, the model development considered only actual costs of
illness and/or lost working time.
Even partial damage to the health of a susceptible
individual may affect work production (Rice, 1966) and thus
loss in production may be an additional cost (and a possible
benefit if the loss is eliminated). Again, of course, it
is very difficult to separate the portion of production loss
due to pollution from the portion that. ~i~ht have occurred
even if the pollution were not present.
Multivariate statistical an~lysis techniques, im-
plemented on some of today's large scale and very fast com-
puters, may be one possible approach to the determination of
loss of production due to pollution.
A relatively simple way of cost-eva~uating physical
damage to humans that results in mortality is to discount the
wages of those who work. This method constitutes a drastic
oversimplification, however, and of course provides no means
of evaluating costs 6f, non-workers. Much additional study is
needed in this area (Sirage1din, 1969). If mortality is
brought on by excessive pollution, the discounted cost of
burial must also be considered.
V-4
-------
2.
Materials - The assessment of economic costs
of materials damage by air pollution has been
examined by Kneese (1966a), Lave (1970), and
Ridker (1967a). Pertinent cost factors in-
clude the cost of increased maintenance (clean-
ing, painting, etc.), the cost of replacement
of deteriorated materials (or depreciation
costs if determinable), and reduced property
values. Data on these cost factors are avail-
able, but the major difficulty lies in sep-
arating the effects attributable to specific
pollutants from the total air pollution ef-
fects. Detailed study of isolated episodes
has been made (e.g. chapter 5 of Ridker, 1967a),
but further research is required to allow
separation of the effects of hydrocarbon pol-
lutants.
Wohlers and Feldstein (1965) estimated the
annual cost of damage to materials by photo-
chemical smog products in the San Francisco
Bay Area from limited data on the relative
service 1ive$ of fabrics, plastics, and rub-
ber products. Since the major damage resulted
from ozone or total oxidants attack on these
materials, it may be assumed that materials
damage from hydrocarbon pollutants in non-
smog areas may be slight.
3.
Animal - The animal category includes physical
damage (morbidity and mortality costs), pro-
duction losses, and psychic costs due to mor-
bidity and mortality of petso Physical damage
includes the increased cost of maintenance of
animals, eo go supplemental feeding, cleaning,
etc. Again, no data are available for separat-
ing those effects and related costs attribut-
able to hydrocarbon pollutantso
Veqetation - The effects of hydrocarbon pol-
lutants on vegetation may be assessed by treat-
ing the costs of replacement or increased
maintenance of ornamental plants and the loss
of production from commercial operations, such
as commercial farming, cut-flower production,
orchards, and forestryo Separation of the
effects of specific pOllutants on vegetation
is more readily attainable than in the others
4.
V-5
-------
(human, materials, etc.), however, much of
the vegetation damage results from exposure
to photochemical smog products derived from
not only stationary source hydrocarbo~ emis-
sions but other sources as well.
Again, psychic costs that are not readily
estimated may result from damage to orna-
mental plants and natural scenic areas.
Indirect Costs':- This category was included
in the model to handle economic effects not
direetly attributable to the other specific
categories. Examples of indirect costs could
include individual adjustments employed to
compensate for or alleviate the direct ef-
fects of pollution (Ridker, 1967a), such as
the cost of moving to a less polluted area
or the cost 6f installation of air condition-
ing to minimize respiratory irritation. Con-
siderable care must be taken in treating such
indirect costs or adjustments to avoid double
counting.
Given these five categories, the functional relation-
ships outlined earlier were developed mathematically to pro-
vide a model suitable for ~uantitative evaluation of the socio-
economic benefit for a reduction in quantity of emissions as
a function of the cost of control.
5.
Before developing the mathematical relationships,
certain important assumptions pertaining to the implement-
ation of the general relationships were required. These as-
sumptions are discussed in the following sub-section.
,
C.
ASSUMPTIONS
The original concept for treating the economics of
hydrocarbon pollution control was to develop a model that would
relate the cost of control to the social and economic benefits
resulting from control or reduction of emissions. Such a
model was developed but because of a lack of input data it
could not be adequately implemented. TQ demonstrate the use
of the model, various assumptions were made. This section
enumerates the important assumptions and discusses the reasons
for choosing one direction or specific value over other al-
ternatives. These assumptions relate to property values;
chronic vs. episodic exposures; derivative effects; distri-
butive effects; discount rates; meteorology; geographic and
V-6
-------
demographic distribution; control effectiveness; and recent
prices.
Property Values - An important assumption was that
property values cannot be used as a direct measure of the
effects of hydrocarbon air pollution, primarily because of
the difficulty of separating the effects of hydrocarbon pol-
lution from other factors affecting property values.
Another reason for not using property values is
that the most readily available data, that of assessed property
values, does not correlate well with real property values
(Davis, 1969).
Chronic Versus Episodic Exposures - Although much
more investigation of the chronic toxicity of hydrocarbons
remains to be done, the literature data indicate that no
serious chronic health effects result from expo9ure to average
ambient air hydrocarbon concentration. It has been assumed,
therefore, that health effects of hydrocarbons are primarily
related to acute or episodic exposure to higher than average
ambient concentrations. .
Even for localized areas near a continuous point
source emission there will generally be daily fluctuation in
wind directions and velocities that provide periods of suf-
ficiently low concentration to allow recovery (assuming no
physiological build-up). Thus, health effects may be esti-
mated, at least for illustrative purposes~ on the document-
ation used for developing Threshold Limit Values (TLV) (Am-
erican Conference of Governmental Industrial Hygienists,
1971).
Derivative Effects - The effects of hydrocarbon
pollutants may be minor in certain areas compared with the
effects of the hydrocarbon derivatives, such as photochem-
ical smog. The probability of smog formation in a given
area depends critically on the topography and meteorological
conditions present, in addition to the types and amounts of
hydrocarbon emitted from both stationary and mobile sources.
The difficulties in atmospheric diffusion modeling of smog
reactions and the lack of data for adequate separation of
stationary and mobile source smog contributions effectively
prevented consideration of derivative effects. For the il-
lustrative examples, we arbitrarily assumed that no photo-
chemical smog was involved.
V-7
-------
Distribution Effects - Distribution effects (i.e.,
the relooation and/or reallocation of people and/or resources)
were arbitrarily assumed to be negligible. This assumption
should be given careful consideration in future studies.
Changes in pollution levels brought about by applied controls
will probably result in significant reallocation of labor
resources and consumer dollars to other parts of the economy.
These distribution effects will be more pronounced for those
people in industries directly dependent on the effects of
pollution (e.g., service industries) than for the general
populace.
Discount Rate - Before making. any assumptions about
discount rates, several discount rates were examined to de-
termine the effects on long-range items such as mortality,
future burial costs, etc. (for burial costs see Ridker 1967a
and Kneese 1966). For this model it was arbitrarily assumed
that a long-term discount rate of 6 percent was representative
(Rice, 1966). A higher discount rate would decrease the pre-
sent value of future costs or benefits and a lower rate, which
discounts future expenditures by a smaller compounded amount,
would increase the present values. The discount rate does
not have a significant effect in the short term (2 to 3 years).
Meteorology - Although the model is designed to
allow inclusion of various meteorological distributions for
difficult regions of the country, simplifying assumptions
were made in the illustrative examples. It was assumed that
meteorological conditions for exposed people and materials
were linearly ordered. For the estimation of downwind pol-
lutant concentrations, the graphical approximations given
by Turner (1970, .revised) were used, assuming average low
wind velocity and atmospheric stability class (slightly un-
stable). Estimates of downwind concentrations for a number
of point source types showed that the principal effect of
different wind velocities and stability classes was to change
the downwind distance at which maximum concentration is at-
tained, with little effect on the maximum concentration (for
effective stack heights of 100 meters or less).
GeOgra~hY and Demography - Both geography and demo-
graphy are impor ant factors in evaluatton of air pollution
effects. Geography is important because of the different
meteorological conditions in differing types of terrain and
of land-use. Demography is important because factors such
as age, income, and susceptibility to disease have definite
effects on the human costs of air pollution. The model al-
lows geographical or regional effects to be adjusted for
V-8
-------
human regional differences as well as meteorological dif-
ferences.
For ease of calculation, the examples assume that
the human damage figures are for a person who lives in an
urban air basin, of average health, age 25 to 29, earns
average income for his age, makes no leisure time economic
contribution, and has linearly ordered pollutant damage.
Distributions for almost all of these factors are available
in published statistical compilations and can be put into
the model as des~red.
Control Effectiveness - The example calculations
are based on the assumption of essentially complete removal
of the pollutants under consideration.
Receht .Prices - In application of the cost-benefit
model, prices should be reviewed, and, where warranted, up-
dated. For example, the average income of 25 to 29 year old
males may change as may the recent cash price of 26t/lb for
cotton.
D.
MATHEMATICAL MODEL
This section describes the model in terms of the
mat.hematical relationships required for implementation.
Before describing the functional relationships, we define
the notation in which these relationships are expressed.
Definitions are given in four groups:
V-9
-------
Lower Case - Primarily sub or super scripts
1.
Letter
Range
a = 1 to 85
b = 1 to h
c = 1 to p
a
b
c
d
e
f
g
h
i
j
k
1
m
n
o
p
q
r
*
i = 1 to n
j = 1 to 5
k = 1 to m
*
*
*
*
*
s
t
u
v
s = 1 to q
t = 1 to 1
*
w
w = 1 to v
x
y
z
y = 1 to z
*
Symbol
1IJ
y
* depends on available data
-- range is not applicable
Description
a specific age
a human disease type
a type of crop or vegetation
a death
a burial
indicates a function
indicates a function'
the limit of human fisease types
a specific pollutant
a category that can be affected
by a pollutant
a specific region of the country
the.Jimit of the control systems
the. limit of the regions
the limit of the pollutants
normal replacement
the limit of the vegetation
the limit of the acreage
a counter used in discounting to
present value
area of affected crop (acres)
a specific control system
indicates a function
the limit of units of washing,
cleaning or maintenance
a unit of washing, cleaning or
ma~ntenance
earlier than normal replacement
a specific material
the limit of materials
indicates application of control
to indicate a function
the interest rate
V-10
-------
2.
Upper Case - Major notation
A
B
C
D
E
F
M
N
p
Q
R
S
T
V
Y
Description
ambient concentration
total benefit
component cost
adverse affect
an expected value
change in pollutant's concentration
meteorological condition
distribution of types (e.g.b's within vari-
ous categories
a probability or proportion
quantity of pollutant in effluent stream
benefit/cost.~atio
cost of cont~ol system
total cost
present value of future costs or benefits
yearly or annualized number occuring
Letter
V-l1
-------
3.
M!jor complete terms
Term
D. . k
1 J
Des cri pti on
variable specifying adverse affects of pollutant
i on category j (where the affected categories
are as originally specified - human, vegetation,
etc.) and k is the region being considered.
A.
1
ambient concentration of pollutant i
meteorological conditions for a given region, k
Mk
Qi
Sitk
quantity of pollutant i in effluent
Cik
cost of control system, t, for pollutant i in
region k
socio-economic cost of hydrocarbon pollution
for region k
Njk
Fitk
distribution of different types within category
j in the region k
reduction in concentration of pollutant i in
region k by control system t
Bit
total cost of controlling stationary source hy-
drocarbon pollutant i with control device t
total benefit of stationary source pollution
control of pollutant i with control device t
Tit
V-12
-------
4.
Benefit/Cost Functional Relationships
Working backwards, the mathematical functional re-
lationships are specified for determining the benefit/cost
ratio for reducing the concentration of pollutant i with con-
trol system t, summed over j categories and k regions.
5 m
Tit = r r Sitjk
j=l k=l
( 1 )
(2)
m
B.t = r (Ck - Ck~)
1 k=l
(3 )
Cik = fk (Dijk)'
m
where fk = r D.ok
j=l lJ
and Ck = fk (Dijk'
Sitk)
(4) Dijk = g(Aik' Njk)
(4~) Dijk = g~(Aik' Njk)
(5) Ai = 1P(Mk' Qi)
(6) Fitk = u(Sitk)
(7) Ai = 1P~(Mk' Qi' Fitk)
Combining these relationships yields R, the benefit/
cost ratio.
(8 )
BOt
- 1
R --
it Tit
V-13
-------
The benefit/cost ratio for all hydrocarbons is
then derived from a matrix analysis of the Rit values.
The calculation of Di'k across the affected
categories is straightforward and should present no sig-
nificant problems when the distribution of the types of
effects are available for each category. Examples of
several of the affected categories are described.
Human - Let j = 1 represent the human damage
category. Then, putting aside regional (k) and control
(t) distinctions for the moment,
n
Dl = E
i=l
costs) +
f (medical costs) +
n
E
i=l
f (psychic costs).
f (mortality
n
E
i=l .
As stated previously the examples used later assume
f(psychic costs) = O.
The medical cost of function is expressed as:
h
f(medical costs) = b~l Pib Cb E (Yb).
Pib is the proportion of disease b caused by pollutant i,
and is itself a function of ambient pollutant concentration.
the types of people that may be affected, the quantity of
pollutant in the effluent, meteorological condition$, and
changes in concentration due to control action (Lave, 1971;
Ridker, 1967a; Kneese, 1966a). CQ is the average annual
cost per case of disease b (see dlsease types below) and
E(Yb) is the total annual number of b diseases. The prin-
cipal disease types include but are not limited to the
respiratory diseases; cancer of the respiratory system,
pneumonfA;-chronfc bronchitis, acufe.bronchitis~ emphys~ma,
and asthma. .
The mortality cost function is expressed as:
85
f(mortality costs) = E Pida Cda E(Yda).
a=l
V-14
-------
Pida is the proportion of deaths (d) caused by pollutant
i at age a and is otherwise similar to Pib in the preced-
ing formula. Cda is the cost of death at age a and E(Yda)
is the expected number of annual deaths. Further, Dda is
the present value of future earnings (Vda) plus the pre-
sent value of the expected cost of burial (Vea).
Vda' the present value of future earnings, at
age a, is defined by the following expression (Rice, 1966):
85 XrWrPar
Vda = L ~a
r=a ,
where Xr is the annual mean earnings at age r; W~ is the
single year labor force participation rate (ie, proportion
of males, or females, of age r employed in any single year);
PaY is the incremental probability that a male of age a
survives to age r; and Y is the interest or discount rate.
When Xr is available only for age groups, the pre-
cedin~ expression can be readily modified, as was done by
Rice (1966). This formulation assumed the use of male earn-
ings only; it would be better to use full family incomes
(Sirageldin, 1969) in this expression, but some of the re-
quired parameters are difficult to measure and would be ex-
pensive to obtain.
V a is the cost of burial discounted to the pre-
sent by subtracting the net gain from delaying burial at
age a. It is calculated (Ridker, 1967a) as:
PaY
{ rr+Yfr-a}]
Vea = Ce[l
L
r=a
where Ce is the cost of burial and other terms are as pre-
viously defined.
Animal - Let j = 2 represent the animal damage
category. Domestic productive animals can be equated to
humans when their net productive contribution is considered
as a "wage" using the formula for evaluating pollution ef-
fects on humans, with perhaps some rendering benefit and
no burial cost.
V-15
-------
Vegetation - let j=3 represent the vegetation
damage category. Then, again putting aside regional and
control distinctions,
n
03 = L
i = 1
f(crop damage) +
n
L
r= 1
f(permanent soil
damage.
Assuming no permanent soil damage from hydrocar-
bon pollutants, the crop value, 0i3' is then the percent-
age of crop damaged and the area of the crop or:
f(crop damage) =
p q
L L PicsCcsE(Ycs)'
c= 1 a = 1
where Pics is the percent damage due to pollutant, i, on
a crop c, over s(acres) area, Ccs is the cost (or value of
crop per acre) and E(Y s) is the expected annual acres of
crop c. The discounte~ future cost of damaged crops could
be included if the pollutant causes any lasting soil dam-
age, but this discounting is not included in the present
treatment.
Material - let j=4 represent the material damage
category. The, 04=f(increased cleaning and maintenance)
+ f(earlier than normal replacement), or
04 = Cw + Vx-o.
Cw' the cost of extra washing, cleaning, or
maintennace, is expressed as:
n v z
L L L PiwyCwyE(Ywy), where Piwy is
i=l w=l y=l
the percentage of washing or cleaning of material due to
pollutant i; C~y is the average cost of cleaning material
y, and E(Ywy) 1S the expected annual units of cleaning or
washing material y.
C =
w
The present value of earlier than usual replace-
ment, Vx-o' is equal to
n z
Vx-o = L L
;=1 y=l
[E(CiY~)
( 1 +y ) r
E(Ciyo)]
(1 +y ) r '
V-16
-------
where E(Ciyx) is the expected cost of early replacement of
material y ~ue to pollutant i, E(C;yo) is the expected cost
of replacement of material y with t~e reduced exposure of
controlled pollutant i, and y is the years to replacement.
Indirect Costs - No indirect costs were included
in these examples. Procedures for calculating indirect
costs would be similar to those outlined above.
Some information was obtained on all of the pre-
ceding variables and functions, but little, if any, was
available on the ranges or distributions of the variable
values. Viable values for the benefit/cost ratios will re-
quire more information on all of the variables than is pre-
sently available. The preliminary examples shown in the
next section were completed by assuming specific values for:
Njk - The distribution of different types of
effects within each of the j categories.
Qi - The quantity of pollutant i in the ef-
fluent.
Mk - The meteorological conditions.
U(Sitk) - The function of the cost of control
system t for pOllutant i.
Values for all of these distributions affect the
distribution of 0i 'k and thus directly influence the cal-
culations given inJthe following examples for formaldehyde
and ethylene.
E.
EXAMPLES
In the examples that follow, additional assumptions
were required due to lack of explicit input data. These as-
sumptions are explained as they occur.
1 .
Formaldehyde
Formaldehyde was selected as an example pollutant
because it is one of the few hydrocarbon pollutants for
which there are some data on human health effects resulting
from exposure to low or moderate concentrations such as might
be attained locally near a pOint source emission. Since much
of the available data on formaldehyde effects are based on
V-l7
-------
short term exposure (see Chapter 4 and also Community Air
Quality Guides, IA1dehydes", 1968), the basis for estimat-
ing damage effects and costs was an 8 hour exposure period
rather than continuous exposure.
By assuming a linear relationship between human
health effects and formaldehyde concentration for an 8 hour
exposure, the limited data on health effects versus expos-
ure were used to develop the approximate relationship shown
in Figure V-1. The extremes of the human effects scale
used in Figure V-1 may be suspect, since Schu~k et al (1960)
reported mild eye irritation from exposure to concentrations
in the range from 0.01 to 0.5 ppm. Additionally, there are
no data on mortality resulting from chronic exposure.
Based on the assumptions described previously in
Section C, the relationship between annual cost per person
affected and the relative human health effects was estimated
as shown in Figure V-2. It should be obvious from these as-
sumptions that many of the important items (age, income, en-
vironment, physical condition, etc.) have wide and disparate
distributions over the total population. Without adequate
data on these distributions, computing the total effects,
even of formaldehyde on human health, is functionally impos-
sible and figures based on estimates or averages are prac-
tically meaningless. .
Figure V-2 shows that the cost to society of human
death is less than that of complete invalidism. As stated
previously, the choice of a discount rate higher than 6 per-
cent would lower the peak; a lower rate would make the peak
higher.* .
By combining the relationships shown in Figures
V~l and V-2, the annual cost per person affected as a
function of form~ldehyde exposure is obtained. This is
shown in Figure V=3.
*
As calculated by the use of the expression for the pre-
sent value of future earnings (p 15), the cost of mor-
tality for the individual male, age 25-29, at a discount
rate of 6 percent is $100,011 (Figure V-2), at a 4 per-
cent discount rate the cost would increase to $128,698.
The cost of complete invalidism is equivalent to the
mortality cost plus the cost of health care service over
the remaining life expectancy.
V-18
-------
VI
+-J
U
QJ
!f-
!f-
LU
.r.
+-J
-
Ie
QJ
....
-
Figure V-l
death
breathing
di ffi cul ty
persistent
irritation
mild irritation
odor
Human Health Effects From Formaldehyde
Exposure
,
14
o
I
10
.
1 2
2
4
6
8
Expos ure
(ppm/8 hr period)
V-19
-------
Figure V-2
Annual Per Person Cost of Formaldehyde Health
Effects
100
100,000 -
10,000
I/!
~ w~ Q. C
- Q. - -
V-20
-------
Figure V-3
100,000
10,000
III
-+-'
III
0
U
c 1 ,000
0-
111111
s..s..
QJta
a.. -
-
s..o 500
QJ"C
a..-
-
ta
~
C
C
ex::
100
Annual Per Person Cost as a Function of
Formaldehyde Exposure
40
20
2
4 6 8 10
Exposure.
(ppm/8 hr period)
12
14
V-21
-------
To complete this example, costs of formaldehyde
emission control must be estimatedo For simplification,
a point source emission affecting an urban region of mod-
erately level topography was choseno The point source
was a formaldehyde production plant producing 100 million
pounds per year of 37% formaldehyde by methanol oxidation.
Absorber vent gases (40,000 SCF/l,OOO lb product) were
assumed to contain 003 percent formaldehyde before con-
trol, or about 125 lb/hr formaldehyde emissions. For an
effective stack height of 20 meters or less, maximum down-
wind concentrations would range from about 0.3 to nearly
20 ppm, depending on wind velocities and atmospheric
stability (Turner, 1970 revised). At low wind velocities
and moderately unstable atmospheric conditions, downwind
concentrations would average about 4 ppm.
The cost of essentially complete control of
these emissions by direct flame incineration was estimated
to be $3.75/hour,* or $30,000/yro
Figure V-3 shows that the estimated annual cost
of human health damage from 4 ppm formaldehyde is about
$25 per person. Assuming this concentration is constant,
the costs and benefits as a function of the total affected
population is shown in Figure V-4.
The equation for the benefit (damage) line is
Yb = 0.025x and the control cost line is y = 300 The ben-
efit-cost ratio is therefore R = Yb/Yc or 8.025x/30. Putting
this into a single graphical expression yields Figure V-5,
in which the ratio line is described by Yr = 0.000833x or
in terms of thousands of people, Yr = 0.833x.
This latter figure (Figure V-5) shows that the
expected annual benefits exceed the control costs when the
population of the exposed area exceeds 1,200 people. The
assumption that all persons in the affected area receive
the same exposure is significant, since actual concentrations
in any air basis vary considerably; realistic estimates-of,
the concentration distribution would require the output of a
sophisticated atmospheric diffusion model.
, "
*
. I - I \. .. . .
Estimated from natural gas fuel costs ($0.60/106 BTU) of
$3.25/hr plus fixed costs of $0.50/hr for incineration
with no heat recovery.
V-22
-------
Figure V-4
+-J
111- 90
0111
us..
I I'd
+-J ,.....
.,...,..... 80
'+-0
(1)"'0
C
(1)"'0
co C 70
to
,..... 111
tO::S
::so
c.s::. 60
C+-J
c:(-
Annual Human Benefit-Cost for Formaldehyde
Emissions Control
130
120
110
100
Benefit from
Control (cost
of Damage)
50
40
30
Cost of Control
20
.10
1000
2000 3000 4000
Affected Population
V-23
5000
-------
Figure V-5
o
..-
~
It1
0:::
~
III 3
o
u
.......
~
or-
If-
QJ
~
QJ
IX) 2
Benefit-Cost Ratio
for Formaldehyde Control
5
4
1
1
2
3
4
5
Affected Population
(thousands)
V-24
-------
2.
Ethylene
The previous example treated a hydrocarbon pol-
lutant that has primarily human health effects (formalde-
hyde has not been shown to have demonstrable adverse effects
on vegetation or materials). For the second example, ethy-
lene was selected as a pollutant having primarily vegetation
damage effects. Human health effects of ethylene are limited
to primarily asphyxiant effects at fairly high concentrations
(TLV=lOOO ppm). Other human health effects are those related
to photochemical smog. Although ethylene is one of the known
smog precursors, this aspect of ethylene damage was not con-
sidered in the example because of the difficulties, cited
earlier, of evaluating health aspects of smog components.
Ethylene damage to vegetation, as detailed in
Chapter IV, results primarily from growth inhibition and
premature ripening of fruit.
Hall et al (1957) described the effects of ethylene
emissions on a cotton crop so severe that the plants died
within 2000 feet of the maximum downwind ground concentration.
The effects of ethylene on vegetation are exemplified in
Figure V-6, based on Hall's data for damage to cotton and
other data given in Chapter IV.
Treating Hall's example of cotton damage, the
vegetation damage from ethylene exposure can be expressed
as the loss of cotton production shown in Figure V-7. As-
sumptions employed were an average cotton yield per acre of
one net bale (~480 lb) and a recent cash price of $0.26 per
pound. Thus, the per acre cost of complete crop destruction
would be 480 lb/acre multiplied by $0.26/1b or $124.80 per
acre.
To complete the example, a point source of ethylene
emissions was assumed to be a petrochemical plant producing
200 million pounds per year of polyethylene. Ethylene emis-
sions were estimated at 450 lb/hr; this results in downwind
concentration estimates (Turner, 1970 revised) in general
agreement with those reported by Hall (1957), in the range
from 5 to 10 ppm.
The cost of essentially complete control of the
ethylene emissions by catalytic incineration was estimated
to be $1.75/hr (Estimated from data in Appendix 8 for cat-
alytic incineration with no heat recovery).
V-25
-------
Figure V-6
Complete
Destruction
~ Partial
u Destruction
Q)
~
«+-
I.I.J
Q)
>
....
....,
ItS
....,
Q)
en
Q)
:>
Growth
Inhibition
Loss of
Apical
Dominance
Abscission
Ethylene Damage to Vegetation
o.
o. 1
1
2
3
Ethylene Exposure, ppm/8 hr period
V-26
4
-------
Figure V-7
125
QJ
~
U
1'0 100
........
v->
II)
~
II)
o
u
75
QJ
en
1'0
E
1'0
C
c:
o
~
~
o
u
50
25
Per Acre Cost of Ethylene Damage to Cotton
o
c: .r; c: - c: QJ c:
O. ~o 1'00 ~o
.... 3:'''' -,... -,.. QJ....
II) o~ ~~ -~
II) ~.... ~u o.u
.... UJ..Q 1'0::1 E ::I
U .... o..~ 0 ~
II) .r; ~ U~
..Q c: II) II)
c:t: - QJ QJ
C c
Exposure Effects
V-27
-------
Combining the relationships given in Figure V-6
and V-7 yields the economic costs of cotton damage as a
function of ethylene exposuret shown in Figure V-8.
Assuming an average uncontrolled downwind con-
centration of 5 ppm ethylene and the cost of control given
abovet the annual benefits (damage costs) and control costs
as a function of exposed cotton acreage are shown in Figure
V-g. Calculation of the benefit-cost ratio for ethylene
control results in the benefit-cost ratio function shown in
Figure V-10. This shows that control costs are exceeded by
benefits when the exposed acreage exceeds 115 acres of cotton.
The input data and assumptions do not warrant this degree of
precision and do not take into account the concentration
distribution. .
F.
DISCUSSION
The preceding examples required considerable over-
simplification and extensive assumptions in order to illustrate
the application of the benefit-cost model. It must be empha-
sized that these examples were purely illustrative and that
considerable effort must be expended to provide the input
data that would allow more meaningful modeling. For further
testingt developmentt and eventual validation of the systems
modelt additional data are required. The more important
missing input values are: Dijkt fk; U(Sitk); Mk; Qit
~(Mkt Qi); and Njk.
V-28
-------
Fig u re V - 8
125
Q)
s...
u
~ 100
~
ell
~
ell
o
U
Q) 75
C')
I'CI
E
I'CI
C
c:
o
~ 50
~
o
u
25
o
0.01
Per Acre Cotton Damage as a Function of Ethylene
Exposure
o. 1
1
2
3
4
5
Ethylene Exposure, ppm/8 hr period
V-29
-------
Fig u re V - 9
+J
~-
o~
U~
Ita
+J ,....
.,.... ,....
'1-0
~ "0 30
(1.1"0
c:::3C
ta
,.... ~
ta:;,
:;'0
c.c
C+J
cx:- 20
Annual Benefit-Cost for Ethylene Emissions
Control
50
40
Benefit from
Control (Cost
of Damage)
Cost of Control
10
100
200 300 4nO
Acres of Cotton Affected
500
V-30
-------
Figure V-10
Benefit-Cost Ratio for Ethylene Control
o
.,...
-+oJ
ItS
~
-+oJ
1/1
c3 3
"'-
-+oJ
.,...
'to-
OJ
C
OJ
co
5
4
2
1
o
1
2
3
4
Acres of Cotton Affected
(hundreds)
V- 31
5
-------
CHAPTER V
REFERENCES
Davis, O.A. and K. L. Wertz, "Consistency of the Assessment
of Property: Some Empirical Results and Managerial Sug-
gestions", Applied Economics 1, 151-7, May 1969.
Hall, W.C., G.B. Truchelut, C.L. Leinweber, and F.A. Herrers,
"Ethy1ene Production by the Cotton Plant and Its Effects
Under Experimental and Field Conditions", Physiologia
Planetarium, ~, 306 {1957}.
Kneese, A.V. and H. Wo10zin, The Economics of Air Pollution,
W.W. Norton, New York, 1966.
Kneese, A.V., "How Much is Air Pollution Costing Us in the
Uni ted States? ", Presented at Economi c and Soc; al Aspects
of Air Pollution Control, 3rd National Conference on Air
Pollution, December 1966.
Kneese, A.V., "Economics and the Quality of the Environment
Some Empirical Experiences", Costs of Air Pollution, M.
Garnsey and J. Hibbs, Eds., Praeger, New York, 1967.
Lave, L.B., "Air Pollution Damage: Some Difficulties in-
Estimating the Value of Abatement", Presented at RFF Con-
ference, Research on Environmental Quality: Theoretical
and Methodology Studies in the Social Sciences, Washington,
D. C., 1970.
Lave, L.B., liThe Economics of Urban Air Pollution", Graduate
School of Industrial Administration, Carnegie-Mellon Uni-
ve rs i ty, 1 9 71 .
Lave, L.B. and E.P. Seskin, "Air Pollution and Human Health",
Science, 169 {3947} 723-33 {1970}.
Rice, D.P., "Estimating the Cost of Illness, U.S. Department
of Health, Education, and Welfare, Public Health Service
Publication No. 947-6, 1966.
Ridker, R.G., Economic Costs of Air Pollution; Studies and
Measurements, Praeger, New York, 1967a.
V-32
-------
REFERENCES, CHAPTER V (continued)
S c h u c k, E. A., E. R. S t e ph ens and J. T. Mid d 1 e ton, .. Eye I r r i -
tation Response at Low Concentrations of Irritants",
Arch. Environ. Health li, 57 (l966).
Turner, 0.8., "Workbook of Atmospheric Dispersion Estimates",
U.S. Department of Health, Edl,lcation and Welfare, Public
Health Service Publication No. 999-AP-26 (1970, revised).
Wohlers, A.C. and M. Feldstein, "Investigation to Determine
the Possible Need for a Regulation on Organic Compound
Emissions from Stationary Sources in the San Francisco
Bay Area", J. Air Pollution Control Assoc. 15 (5), 226-9
(1965). - .
V-33
-------
CHAPTER V
SUPPLEMENTAL BIBLIOGRAPHY
Adeliman, H.M., "Exploration of the Relationship Between Air
Pollution and City Systems", Public Systems Research In-
stitute, Los Angeles, California.
Alexander, R., "Social and Economic Effects of Changes in Air
Quality", Oregon State University, Contract CPA-70-l09, May
1969.
Anderson, R.J~, Jro and ToO. Crocker, "Air Pollution and
Residential Property Values", Paper Presented at Meeting
of Econometric Society, New York, December 1969.
Barrett, L.B. and I.E. Waddell, "The Cost of Air Pollution
Damages, A Status Report", National Air Pollution Control
Administration, Public Health Service, July 1970.
Battelle Memorial Institute, A Survey and Economic Assess-
ment of the Effects of Air Pollution on Elastomersll, Con-
tract CPA-22-69-146, June 19700
Bautnol, W. and D. Bradford, "Optimal Departures from Marginal
Cost Pricing", American Economic Review !:I, 265-83 (1970).
Becker, G.S., "A Theory of the Allocation of Time", Economic
Journal LXXV, 493-517 (1965)0
Booz-Allen and Hamilton, Incq "Survey to Determine Residential
Soiling Costs of Particulate Air Pollution" Contract CPA-
22-69-103 (1970)0
"Community Air Quality Guides: Aldehydes" American Ind.
Health Assoc. Jo, Sept.-Octo 19680
Crocker, ToO., "Urban Air Pollution Damage Functions: Theory
and Measurement", Department of Economics, University of
Wisconsin.
Hodgson, T.A., "Short Term Effects of Air Polluti-onon Mor-
tal i ty inN ew Yo r k City", En v i ro n. S c i. and Tech n 0 1. 4
(7) 589-97 (1970)0
.V-34
-------
SUPPLEMENTAL BIBLIOGRAPHY, CHAPTER V (continued)
Ipsen, J., F. Ingenito, and M. Deane, "Episodic Morbidity
and Mortality in Relation to Air Pollution", Archives
Environ. Health ~, 458-61 (1969).
Jackson, W.E., Wohlers, H.C., and W. DeCoursey, "Determin-
ing the Costs of Air Pollution", J. Air Pollution Control
Assoc. !i, 917-23 (1969).
Kohn, R.E., "Linear Programming Model for Air Pollution
Control: A Pilot Study of the St. Louis Airshed", J.
Air Pollution Control Assoc. ~ (2) 78-82 (1970).
Lancaster, K.Jo, "A New Approach to Consumer Theory", The
Journal of Political Economy, LXXIV, April 1966.
LeSourd, D.Ao, "Comprehensive Study of Specified Air Pol-
lution Sources to Assess the Economic Effects of Air
Quality Standards", Final Report, Research Triangle In-
stitute, December 1970.
Michelson, I. and S. Tourin, "Comparative Method for Study-
in9 Costs of Air Pollution", Public Health Reports 81
(6) 505-11 (1966). -
Michelson, 10 and B. Tourin, "Report on Validity of Ex-
tension of Economic Effects of Air Pollution Data",
Environ. Health and Safety Assoc., Contract PH 27-68-22,
August 1967.
McCaldin, R.L., "How Much Does Air Pollution Cost", Air
Engineering li, 18 (1963).
Prest, A. and R. Turvey, "Cost-Benefit Analysis:
Economic Journal Zi, 683-735 (1965).
A Survey II ,
Uhlig, H.H., "The Cost of Corrosion in the U.S.", Corrosion
n (1) 29-33 (1950).
U.S. Senate, The Cost of Clean Air, Report of the Secretary
of Health, Education and Welfare to the Congress of the
United States, U.S. Government Printing Office, Washing-
ton, D.C. First Reporto June 1969; Second Report, March
1970.
v- 35
-------
SUPPLEMENTAL BIBLIOGRAPHY, CHAPTER V (continued)
Wilkenstein, W., Jr., S. Kanter, E.W. Davis ~ C.S. Maneri,
and W.E. Mosher, "The Relationships of Air Pollution
and Economic Status to Total and Selected Respiratory
System Mortality in Men", Presented at AMA Air Pollution
Medical Research Conference, Los Angeles, March 1966.
V-36
-------
VI~
CONTROL OF HYDROCARBON POLLUTANTS
A.
INTRODUCTION
The purpose of this phase of the study 'was to
identify and evaluate engineering technqiues for the control
of' hydrocarbon emissions from stationary sources, with the
ultimate objective of presenting a coordinated R&D planning
program. To aid in establishing priorities and properly
channelling future efforts, particular emphasis was placed
on the identification of gaps in existing technology and de-
sign bases and the formulation of remedial experimental pro-
grams.
A review of the literature relating to hydrocarbon
emission control was conducted to (1) assess the disadvantages
and deficiencies in existing technology and (2) identify those
control techniques for which general case economic evaluations
would be of value.
This review of the open literature showed that,
aside from the compilation of design procedures and case
histories presented in the Air Pollution Engineering Manual
(Danielson, 1967), there is very little in the way of pub-
lished quantitative information on industrial application of
hydrocarbon control processes and equipment. A review of
hydrocarbon control technology was presented in Control Tech-
ni ues for H drocarbons and Or anic Solvent Emissions from
~~~;~~~~r19~g}:c~~t ~h;. t~~~~~~~~n~a~fl~~~eiy'qua~f~~~~~e a~~d
descriptive. .
Much of the detailed information on the engineering
design and application of control techniques resides in the
files of engineering and equipment firms specializing in par-
ticular types of pollution control. Discussions with several
such firms resulted in the release of hitherto proprietary in-
formation on design and application procedures. Some of this
information is presented in the detailed engineering and eco-
nomic studies in Appendices A through E.
VI-l
-------
Technology for the control of hydrocarbon emissions
has developed rather unevenly. Primarily through the impetus
of smog control programs in Southern California, the control
and abatement of petroleum products and organic solvents by
incineration or afterburning and carbon adsorption has been
developed into rather straightforward procedures of engineer-
ing design and application. Other techniques, particularly
wet scrubbing for absorption, direct-contact condensation,
and wet particulate removal, lack a:suitable design base.
Our review and ranking of organic pollutants and
their effects showed that identifiable and proven adverse
effects of organic vapors or gaseous emissions were limited
to their contribution to photochemical smog and certain rel-
atively minor health effects. This lack of documented adverse
effects undoubtedly accounts for the unevenly developed state-
of-the-art with smog control and nuisance abatement being
strongly emphasized. In spite of the concern in some quarters
over the status of the polycyclic aromatic hydrocarbons as
potential carcinogenic agents, very little development of
control technology for sources of these PAH's has been re-
ported.
The control technologies that are the most advanced
both theoretically and practically, are those used in ap-
plications offering a clear-cut economic advantage, either
as prevented losses or as recovered values. Predominant among
these are the application of storage tank controls for volatile
petroleum products and the utilization of activated carbon for
solvent recovery. The incentive for development has been
heavily weighted for an economic return on investment rather
than concern for environmental pollution control, although a
number of instances have been reported recently in which pol-
lution control applications have resulted in unexpected eco-
nomic returns.
The most conspicuous lack of published data is in
quantitative information relating to the effectiveness or
efficiency of the various control techniques and equipment.
The crux of the control evaluation problem was identified in
a recently completed study of polycyclic aromatic hydrocarbons.
Although specifically cited with respect to PAH emissions,
the statement applies much more broadly. In their draft re-
port on Control Techni(ues for Polycyclic Or9anic Matter Emis-
sions, GCA Technology 1970), after considerlng the available
design data, stated, "in the absence therefore of any practical
information on the effects of existing types of gas cleaning
equipment on polycyclic organic emissions, we must rely on
VI-2
-------
theoretical considerations to determine which cleaning tech-
niques will be effective in removing polycyclic organic emis-
sions". Although the outlook for control technology for
certain other hydrocarbon classes is not quite so bleak, the
degree of effectiveness offered by a specific design for a
specific emission is still uncertain in the best of cases.
Apparent control effectiveness may result~ in many
cases, from variable definitions of pollution or from failure
to measure the process output under operative controls. An
example is provided by Fawcett (1970) in a review of emissions
common to the phthalic anhydride manufacturing industry and
the type of controls used. Emissions are generally the acids
or acid anhydrides, aldehydes, o-xylene or naphthalene, and
occasionally, particulates. In one set of locations, the air
pollution problem is that of nuisance odors and lachrymatory
emissions, which are characteristic of the organic acid com-
ponents. In West Coast locations or other areas of poor air
shed ventilation, however, the pollution problem is caused by
the photochemically reactive aldehyde and aromatic constituents.
Fawcett reports that wet scrubbing affords 99 percent removal
of the acids arid anhydrides, but gives relatively poor control
of aldehydes. Thus, in smog-prone areas where the aldehyde
is of primary concern, "control" would involve incineration
of the process off-gas, either by direct flame or catalytic
methods; in other areas "control" would call for acid removal
by wet scrubbing. The impact of control is therefore measur-
able only in the context of local pollution problems.
With a few exceptions, the hazards to health and
welfare posed by organic compounds have not been fully deter-
mined. Until these basic parameters are defined, control ef-
fectiveness cannot be absolutely fixed. Thus, quantitative
information on design and "efficiency" for existing control
techniques can best be assembled on a "case history" basis.
In the following sections, the generally used con-
trol techniques are briefly reviewed and the economic costs
of general case control techniques are presented. Detailed
treatment of two areas of wet scrubbing technology is given
in Appendices E and F, along with the calculations and other
documentation supporting the general case economic eValuation
summaries, presented in Appendices A through D.
B .
CONTROL TECHNOLOGY
1.
Emission Elimination Via Add-On Devices
VI-3
-------
a.
Direct Flame Incineration
Control of hydrocarbon or organic emissions by in-
cineration involves direct (non-catalytic) oxidation of the
combustible portion of the effluent, the desired ultimate
products of combustion being water and carbon dioxide. Ef-
fluent streams from incineration of organic emissions may
be characterized as either fuel-rich or fuel-poor. Relatively
few effluent streams are fuel-rich and they are more readily
amenable to control. The cli1ute or fuel-poor effluent
streams are of more concern, because of their ubiquity and
because of the diversity of types and concentrations of or-
ganic species involved.
Fuel-rich effluents are most common in the petroleum
and petrochemical industries and are generally handled by the
so-called smokeless flares. Smokeless flares depend on three
principal factors: (1) sufficient fuel values in the effluent
stream to attain theoretical combustion temperatures, (2) suf-
ficient combustion air, and (3) turbulent mixing of fuel and
combustion air. Given a fuel-rich effluent, such as a highly
volatile h~drocarbon waste stream from a refinery operation,
factors (2) and (3) are generally accounted for by the use of
steam injectiong which provides a low-cost energy source for
the inspiration of combustion air and for production of tur-
bulent mixing. Steam also aids in combustion by reaction in
the fJime to form molecular species and radicals, which burn
or decompose completely at relatively low flame temperatures.
The design and application of smokeless flares have
been reviewed by Danielson (1967), Murray (1967), and Miller
(1968), and rather genera11y,in Control Techni9ues for Hydro-
~anic Solvent Emissions from Statl0nary Sources
USDHEW, 1970 .
The same controlling factors, listed for flare com-
bustion of fuel-rich effluents influence the combustion of
fuel-lean effluents. In these effluents, however, the ef-
fluent stream usually lacks sufficient fuel values to attain
stable or self-sustaining combustion temperatures, and a
secondary fuel source must be supplied. The most commonly
used supplemental fuel for direct flame incineration is nat-
ural gas, although in some applications, petroleum-based fuels
are used. In some industrial operations, waste organic gas
streams are burned in existing boiler furnaces, but, because
of increased problems in maintaining proper boiler performance,
this technique has limited application.
VI-4
-------
For most hydrocarbon compounds, the lower limit of
flammability corresponds to about 52 BTU/SCf, and waste gas
streams are generally limited, for reasons of safety, to 25%
of the flammability limit, or 13 BTU/SCF (Stern, 1968). Be-
cause of this necessary dilution to a level well below the
flammability limit, additional energy is required to maintain
combustion of the hydrocarbons; this energy is added by pre-
heating and by supplementary fuel burners. The minimum tem-
perature level required in afterburning is the autogenous com-
bustion temperature. The other conditions necessary for com-
plete oxidation include good mixing of the waste gas with the
hot combustion gases and residence times of 0.3 to 0.5 second.
The Air Pollution Engineering Manual (HEW, 1967) presents
basic design procedures for direct-flame afterburners and pro-
vides data on stack analyses for diverse applications.
Although design methods present no particular en-
gineering difficulty, the safety problems attendant on handl-
ing a gas whose components may lie in the explosive. range
often require the services of an engineering group specializ-
ing in combustion or furnace operations. With any incineration
process, including afterburning, the principal factor, from
an emission standpoint, is the degree of completion of the
oxidation reactions. Even when incompletion of combustion
is limited to 1 or 2% the result may be intermediate oxidation
products whose intrinsic hazard or nuisance value is greater
than that of the original hydrocarbons. The problem is then
one of quality rather than of quantity. A case in point is
the study published by Sandomirsky et al (1966) of the incin-
erative disposal of an odorous mineral oil aerosol fume created
in rubber processing. Although direct-flame incineration was
stated to yield 84-92% oxidative removal, "samples taken from
the stack gas indicated a completely changed material, compared
with samples from the inlet process gas ... Products indicated
as being present in trace amounts in the stack gas were acids,
anhydrides, lactones and unsaturates. These were due to thermal
cracking and oxidation in the combustor". Whereas the mineral
oil caused odors and visible fumes, the oxidation products
were certainly more reactive smog-formers than was the pre-
cursor oil. Thus, although complete afterburning is an ef-
fective method for disposal of hydrocarbon vapors, incomplete
and inefficient afterburning may create a pollution problem
that is more serious qualitatively than that faced initially.
The open literature is replete with articles describ-
ing, in general qualitative terms, the application of direct
flame incineration to a wide variety of sources. Very few of
these reported studies, however, provide sufficient data to
VI-5
-------
assess the performance of a particular system. Much of the
art resides in the proprietary files of companies specializ-
ing in incineration equipment. Danielson (1967) reviewed
and summarized much of the available information and provided
some very helpful case studies. General engineering data on
the combustion of organic gases may be obtained from such
sources as the North American Combustion Handbook (1952).
Waid (1969) presented some design considerations and limited
laboratory test data on a specific burner configuration.
The design of afterburners for application to
varnish cookers was considered by Mills (1959). Benforado
(1968a) described five methods (including incineration) for
controlling air pollutants from painting and baking operations
in finishing plants. He also (1968b) presented data on the
applicability of direct-flame incineration to wire enameling
operations and to the paint industry (1967). Elliott (1962)
performed an experimental program on control of organic emis-
sions from protective coating operations. Wallach (1962)
provided data on the combustion of gaseous effluents from
baked lithograph coatings.
The main point brought out by these and other authors
is that proper design and performance of direct-flame incin-
eration equipment depends greatly upon knowledge of the type
of organic species in the effluent stream and on adequate test-
ing and monitoring to assure that the desired performance is
achieved., With simple and easily combusted waste streams,
minimum flame temperatures and residence times can achieve
complete combustion. Complex and refractory waste emissions,
such as those from the baking and heat treatment of vinyl
and other organic-based plastic and polymeric coatings, may
require long residence times at temperatures above 1600°F to
assure satisfactort1y complete combustion.
One of the primary advantages of combustion processes
is that they generally convert the pollutant directly to non-
noxious substances. Sometimes, however, the organic contami-
nant contains constituents other than carbon, oxygen,'and
hydrogen which, when combusted, result in a product more nox-
ious than the hydrocarbon itself. Organic substances contain-
ing sulfur and halogens are primary examples. With sulfur
the product is S02' which is generally considerably less nox-
ious than the organo-sulfur precursor. With halogens, the
products are generally acid gases such as HC1, which can
severely corrode the control equipment or can produce nuisance
or corrosion problems outside of and adjacent to the industrial
complex. In such situations, secondary purification, generally
VI-6
-------
by liquid scrubbing, is used to remove inorganic compounds
from the effluent.
In an improperly designed combustion system it is
also possible that hydrocarbons are only partially oxidized
and aldehydes are formed. Aldehydes are associated with
smog formation and are more objectionable than the original
hydrocarbon. The possibility of formation of benzo(a)pyrene,
a known carcin0gen, by controlled combustion is discussed
by Boubel (l963).
, ..
Another disadvantage of combustion processes is
the possible formation of nitrogen oxides. In attempts to
increase the efficiency of combustion of hydrocarbons, the
efficiency of oxidation of nitrogen increases, with the re-
sult that NOx is added to the already existing NOx burden
produced by automobiles and by other stationary combustion
sources.
A significant factor in direct-flame incineration
is the cost of the secondary fuel. Because of insurance under-
writers' restrictions, most waste ~as streams are diluted to
well below the flammable or explos1ve limits, usually below
25 percent of the lower explosive limit. In these cases,
fuel costs can be high enough to require heat-recovery equip-
ment. The wide variety of heat recovery systems now in use
generally involve using the waste heat to preheat the effluent
gas stream or to provide process heat to some other part
of the operation. Careful evaluation of the total system is
required to assess the cost effectiveness of direct-flame
incineration in comparison with alternative control methods.
b.
Catalytic Oxidation
As a means of atmospheric pollution control cat-
alytic oxidation is applied most often to reducing or elim-
inating organic emissions where recovery of these emissions
by regenerable processes is not warranted. Catalytic oxi-
dation is thus applicable to dilute emission streams and to
emissions that are difficult or impracticable to recover by
other methods. Conventional flame incineration and catalytic
oxidation are often quite competitive methods of emission
control; careful engineering and economic evaluations are
required for a proper choice of combustion methods. Several
pertinent reviews of flame incineration versus catalytic
oxidation have been made in recent years. Hein (1964) and
Brewer (1968) compared the costs of installing and operat-
ing thes~ two processes and showed that in many instances
catalytic oxidation processes are more economical. Thomaides
VI-7
-------
(1971) also compared the costs and presented advantages and
disadvantages of catalytic oxidation and direct flame incin-
eration.
The literature dealing with comparative economics
of incineration must be viewed with extreme caution, however,
because presentation of the evaluation methods is often
sketchy and incomplete, and sometimes a deliberate choice of
examples tends to show a clear advantage for a particular
control system.
The practical applications of catalytic oxidation
in control of industrial solvent emissions remain highly pro-
prietary. Literature is available, however, on application of
catalyst combustion in paint manufacture (Stenburg, 1958),
paint baking ovens (Danielson, 1967), insulation board manu-
facure (Duerden, 1965), and coffee roasting (Sullivan, 1965).
Betz (1963) showed that catalytic combustion was the most
efficient method of removing effluents from a phthalic an-
hydride plant. lli Air.Pollut~on Engi.f1eering Manual (Danielson,
1967) gives case histories with design data and operating con-
ditions. Review articles such as those by Brewer (1968) give
a limited amount of practical information on the choice of
incineration techniques.
A significant advance in the art of catalytic in-
cineration has been the development in recent years of im-
proved catalyst support systems, namely honeycomb ceramics.
An excellent article by Miller (1967) presents the results of
a pilot scale study of several catalyst support systems.
Measurements of ojfdation efficiency over a range of temper-
atures and space velocities show clear advantages of honey-
comb ceramics over other catalyst support systems.
Basic research on the susceptibility of homologous
series of hydrocarbons to catalytic oxidation was carried
out by Accomazza (1965) and Carretto (1966); the industrial
aspects of design, economics,-and operation were reviewed by
Brewer (1968) and Werner (1968). Werner classified the ~om-
mercial catalytic afterburning catalysts as (a) active metal/
met~llic carrier (b) active oxide/oxide carrier, and (c)
a~t1ve metal/oxide carr~er. P~isoning is a major problem
w1th catalyst beds, as 1S part1culate plugging; both Brewer
and Werner provided information on the nature of poisons for
the various types of catalysts.
Effluents containiog sulfur, halogens, and silicon
are known to be poisons for platinum and palladium-type cat-
VI-8
-------
alysts. Although the mechanisms are not fully known, silicon
oxidizes to silica, which blocks the pores of the catalyst.
Metal oxide catalysts developed for specific applications
have been shown to be less susceptible to halogen poisoning
than the precious metal catalysts. Other problems such as
thermal degradation, however, limit the application of these
systems. .
The primary advantage of catalytic incineration is
that extremely dilute concentrations of noxious organics can
be oxidized with only small amounts of supplemental fuel be-
ing required, often only that needed to preheat the catalyst
bed. Thus, the lower fuel costs present a significant ad-
vantage over direct-flame incineration for dilute waste gas
streams. .
Again, we emphasize that knowledge of the type and
concentration of the waste gases being incinerated is required
to select and design a properly performing catalytic system.
In the past, improper application of catalyst systems resulted
in installations that woul~ not meet local code requirements
for emission control. Recognition of these problem areas and
more careful design studies are currently resulting in success-
ful application.
c.
Adsorption
Although a. vArtety:of,b1gh~surface-area materials
have been used for vapor-phase adsorption, only activated
carbon has been used widely in practical applications involv-
ing organic vapors. The great advantage of carbon over other
adsorbents is its ability to strongly and regenerably adsorb
a wide variety of organic molecules, even in the presence of
water vapor.
Because the literature on the theory and application
of activated carbon for adsorption of organics is voluminous,
no attempt was made to review the sUbject completely. Thorough
reviews have been presented in a nu~ber of standard texts,
such as Mantell (1961) and Hassler (1961). In addition, com-
mercial suppliers of activated carbons and carbon adsorption
systems maintain extensive data files and cooperate in making
available this specific information on design and operation.
Theoretical and experimental studies have been made
to describe the various param~ters influencing ~he operative
cycle of adsorption-regeneration on carbon. Mathematical re-
VI-9
-------
lationships correlating adsorptive capacity, bed life, and
pressure drop with specific and generalized vapor properties,
vapor concentrations, temperature, flow rate, and carbon
properties have been reported. Perhaps the most useful re-
lationship for preliminary screening of the applicability
of carbon adsorption to a particular system is that based on
the Polanyi potential theory, described and developed by
Grant et al (1962) and Robel1 et al (1965).
The POlanyi potential, A, is related to temperature,
concen~ration, and molar volume of adsorbate by the expression,
T Co
A = Vm log Ci
where T = adsorption temperatures in oK
Vm = molar volume of liquid adsorbate
at the normal boiling point in
cm3jmol
Co = saturation vapor concentration
Ci = influent vapor concentration.
By experimentally determining the adsorptive capacity, q, of
a.patttt~lir carbon for several organic vapors over a range
of A, a generalized semilog plot of q versus A may then be
used to predict q values for other organic vapors.
Generalized curves of q vs A (such as given by
Grant et al, 1962) for several typical carbons may thus be
utilized to screen the app~icability of carbon adsorption
for the recovery of any known organic vapor in the concen-
tration range of interest.
The app~ication of carbon adsorption for solvent
recovery, particularly where the vapor concentration is fairly
high (>500 ppm) and the solvent has recovery value, is a
standard operation. Integral, "packaged" systems are avai 1-
able from several commercial sources. A number of papers
have been published on these applications as well as those
involving removal of low concentrations of noxious vapors and,
more recently, the concentration of vapors for subsequent
incineration. Reviews on the elimination of organic vapors
by carbon were given by Barnebey (1959) and Morgani (1951).
Parameters associated with the operation of adsorption equip-
ment were discussed by Chetfie1d (1967) and LeDuc (1967).
Adsorption is used for the control of emissions from surface
coating operations (Kanter, 1959; Elliott, 1961), pharmaceutical
VI-10
-------
m~nufacture (Johnson, 1959), and solvent extraction oper-
ations (Barnebey, 1965).
Barry (1960), after reviewing the published in-
formation on carbon systems design, concluded that the
design and scale-up procedures at that time were not satis-
factory. Confirming this finding, the Los Angeles County
Air Pollution Control District and the U.S. Public Health
Service undertook the evaluation of adsorption design methods;
they obtained and published updated design information
(Elliott, 1961); much of which was incorporated by Danielson
(1967).
The economics of operation of carbon adsorption
systems are strongly influenced by the specific vapors ad-
sorbed, the influent concentration, and the ease of regen-
eration. Steam is the usual medium used to strip the loaded
carbon bed, but for some conditions the temperature range of
saturated steam is not high enough to strip the sorption bed
in a r~asonable regeneration time cycle. Mattia (1970) stated
that ~;5 lb of steam/lb of stripped organic is required when
the influent vapor concentration is 0.2 percent by volume,
whereas over 30 lb of steam/lb of organic may be required when
the pollutant concentration is reduced to 200 ppm.
Because steam regeneration requires corrosion-re-
sistant construction, capital costs are generally high. Cap-
ital costs were reviewed in Control Equipment, AIHA (1968),
and data on operating costs were reviewed by Mattia (1970).
Lee (1970) reviewed design and operating conditions of success-
ful carbon adsorption systems.
Mattia (1970) presented some preliminary studies
and costs of a process that utilizes carbon adsorption to
concentrate hydrocarbon effluents for subsequent incineration.
This technique should be useful for effluent streams having
little or no economic recovery value, in' that it allows in-
cineration of a more concentrated stream from the carbon de-
sorption, with concommitant lower incineration costs.
d.
Adsorption
Absorp.tion is distinguished from adsorption in that
with absorption the sorbed species is physically transferred
across the sorbent interface, while in adsorption the sorbed
species is held, physically or chemically, at the interface.
In effluent gas purification, the most commonly used absorbents
are liquids, primarily water.
VI-ll
-------
Even though wet scrubbing methods are widely used
for removal of particulates and hydrocarbons, usually con-
currently, assessment of current design methods and the state
of technical development of wet scrubbing showed a very ser-
ious deficiency. This defitiency is examined in detail in
Appendices E & F of this report. A technical literature re-
view showed an unbridged gap between theory and field practice.
There was no rationale for scrubber selection or design in
the following problem areas:
(a)
Removal of liquid hydrocarbon aerosols or
of particulates in association with heavy
hydrocarbon co-particles, particularly
polynuclear hydrocarbons.
Combustion gas scrubbing with secondary
absorption of water-soluble oxidized
hydrocarbons.
(b)
Condensation (direct or indirect)9 followed
by mist-filtration or scrubbing.
Thus, the need was obvious for an intensive critical
examination of scrubber technology, re-evaluation of design
approaches, and correlation of data.
(c)
At the outset of this-syftems study, it became ap-
parent that the major stationary sources of hydrocarbon emis-
sions were combustion processes. Invariably, such processes
emit particulates in association with hydrocarbons. Power
generation, incineration of solid wastes, and disposal of
industrial organic liquid sludges are examples of such op-
erations. Attention was focused on.~ydrocarbon-particulate
emission because of the health problems attendant on the
polycyclic ar~matic hydrocarbons normally found in combustion
off-gases, particularly fn inefficient combustor units. Sev-
eral of the polycyclic aromatics, notably benzo{a)pyrene (SaP),
are known to::be :carcinogenic, and for this reason, control in-
formation for these hydrocarbons merits priority.
Hangebrauck (1967) measured rates of emission of
polynuclear hydrocarbons in refuse burning and various in-
dustrial operations, both before and after the exhaust gases
passed through control equipment. Data were presented on
the effect of CO waste heat boilers and plume burners on cat-
alytic cracking regenerators in petroleum refining, and of
water-spray scrubbers and steam "sprays" in batch asphalt
VI-12
-------
hot-road-mix plants. In the latter, water-spraying was found
to be 92% effective in BaP removal; the data were extremely
limited, however, and no information was given on the design
of the control equipment.
Cuffe (1967) studied emissions of seven key poly-
nuclear hydrocarbons from coal-fired power plants using var-
ious boiler designs. Preliminary tests showed that consider-
able recovery of the polynuclear aromatics was effected by
the fly-ash collectors, although no quantitative data were
provided. Full-scale tests showed that even some formaldehyde
was removed with the particulates, an indication of probable
adsorption. In terms of control relevance, this means that
control of hydrocarbon emissions from a combustion process
can and should be accomplished via control of particulates.
Appendix F presents results of a review. of the
performance characteristics of wet scrubbing devices in
particulate removal. This presentation describes a novel
correlating technique that shows unique promise as a tool
for valid absolute performance comparison and establishes
the test conditions necessary to validate such comparisons.
Briefly summarized, this study showed that perform-
ance of dispersed liquid wet scrubbers is mechanistically
explained by turbulent agglomeration contacting and secondary
inertial removal, rather than by inertial impaction theory.
It was found that three primary independent variables control
scrubber performance: inlet dust loading, water/gas ratio,
and gas velocity (or pressure drop equivalent). Further,
when scrubber performance was expressed in terms of the per-
cent mass penetration*, the performance was inversely pro-
portional to both inlet dust loading and the liquid/gas ratio.
When the latter variables were combined into a single pro-
duct group, termed the "agglomeration index", the literature
data on dust penetration showed that penetration was inversely
proportional to this agglomeration index. Application of this
new concept to a wide variety of experimental and field data
on wet scrubber performance showed that the agglomeration
index can be a sensitive and powerful correlating parameter.
Deviations of reported data from the correlation were found
to be indicative of dust feed pre-agglomeration or design
deficiencies in the removal mechanism (as distinguished from
the contacting mechanism) of the scrubber.
The third independent variable, pressure drop, was
readily correlated when performance data were plotted at con-
stant agglomeration index conditions.
* Percent mass penetration is defined as the weight percent-
age of inlet dust that penetrates the scrubbing device;
the inverse of percent removal efficiency.
VI-13
-------
The final generalized equation obtained was:
% Penetration = (Cd)(~/G)(~p)n
where
K = performance constant
WCdG --= inlet dust loading, grains/CF
I water/gas ratio, gallons/MCF
P = pressure drop, inches of water
n = 1 or 2, depending on pressure
drop regime.
In Appendix E, new and hitherto proprietary design
techniques are presented covering applications for the re-
moval of soluble pollutants, such as partially oxidized hydro-
carbons from waste incineration, in a combination gas quench-
cooling operation.
2 .
Modification
Reviews of pollutant sources and existing control
technology have pointed out certain categories of hydrocarbon
emissions and sources for which existing control techniques,
per se, are not directly applicable (e.g. open burning of
solid wastes). For these sources, reduction of emissions
may best be accomplished, at least for the present and near
future, by alternatives or processrmodifications.
Emission Elimination via Process or Equipment
a.
Storage and Transfer Controls
Atmospheric emissions from storage and transfer of
volatile organics, particularly petroleum products, have been
identified as a significant source of organic pollutants.
These emissions may be controlled by a variety of mechanical
devices aimed at preventing the losses from occurring, rather
than by systems for destroying or recovering emissions from
deliberately vented waste streams.
Control methods for"preventing losses of volatile
hydrocarbons during storage and transfer were reviewed by
Danielson (1967) and by USDHEW (1970). The American Petroleum
Institute published an excellent series of technical bulletins
covering methods of calculating losses and recommended practices
for reducing or controlling emissions from storage and trans-
fer o~;petroleum products.
VI-14
-------
A recent review by the Petroleum Committee of the
Air Pollution Control Association (TI-3 Petroleum Committee,
1971) ~ummarized the state of the art of storage tank con-
trols.
The most generally applied control device for re-
duction of storage tank emissions is the floating roof stor-
age tank. Although a variety of floating roof designs have
been developed, all consist of an impermeable cover, which
floats on the stored liquid and provides a sealing device be-
tween the floating roof and the tank wall to prevent evaporation.
Other devices, such as sealed or pressure tanks and
vapor recovery systems, are used much less than floating roof
controls. Vapor recovery systems are more commonly used in
transfer operations, in conjunction with sealed and submerged
filling devices, to prevent and recover both evaporative and
spillage losses.
b.
Waste Burning Alternatives
Sections II and III identified the disposal of
solid wastes by open burning as a highly significant con-
tributor to the total hydrocarbon emissions problem. The
large number and widely dispersed distribution of individual
point sources of waste disposal by open burning make the
application of conventional control techniques prohibitive,
both technologically and economically. The review of exist-
ing technology has identified three principal alternatives
to the open burning of solid wastes: multiple chamber in-
cineration, sanitary landfilling, and composting. Other
potential alternatives, which are in various early stages of
development, are touched upon briefly after discussion of
the three principal methods.
Sanitary Landfilling - Of the three alternatives,
landfilling is the most widely practiced in the United States.
The general consensus is that at least 50 percent of the col-
lected refuse in the United States is disposed of in land-
fill operations. Unfortunately, most of the current landfill
operations would not meet the definition of "sanitary" as
given by the American Society of Civil Engineers (1959). The
National Solid Wastes Survey (1968) found that of 6,000 sites
studied in detail, only 6 percent met minimum requirements
for designiation as "sanitary landfills".
VI-l5
-------
Provided proper design and control, landfill oper-
ations can be handled in a manner that meets sanitary re-
quirements with essentially no air pollutant emissions. The
cost and availability of adequate land areas must be con-
sidered in any comparison of landfills with other disposal
methods; in some municipal areas, these factors present a
distinct disadvantage. A recently reported study of remain-
ing'landfill life at 34 cities (Solid Waste Report, 1970)
shows a range of from 0 to 25 years remaining life, with land-
fills in approximately half of the cities having less than
5 years remaining.
Although the costs of collection and transportation
are significant factors in the total cost of any disposal sys-
tem, landfill operations are generally most heavily affected
by transportation costs. With the rapidly increasing 'sprawl'
of most citi~s;l~ndr~it~s-within reasonable hauling distances
are fast diminishing. In a survey of 439 cities (Combustion
Engineering, 1969) 90 cities, or about 20 percent, reported
hauling distances of over 10 miles to the disposal site.
Thus, the feasibility of sanitary landfilling as an
alternative method is highly dependent on site availability
and hauling distance. Sinc~ these factors cannot be accurately
generalized, specific studie~ are required for each area con-
sidered.
Multiple Chamber Incineration - Recent studies of
disposal of municipal and industrial wastes indicate that
technology is ~vailable for the satisfactory (from an air
poll uti on s tan dp 0 i n t ) d i s po s a 1 of m 0 s t w as t e s by i n c i n era t i on
(A.D. Little, 1970; Engdahl, 1969; Combustion Engineering,
1969). Although development of municipal incinerators in the
U.S. has lagged considerably behind European practice, re-
cent developments have shown that with, proper combustion
chamber design and application of a,ir 'pollution controls, the
operation is technically feasible, with acceptable emission
levels. Depending on the type of operation, a solid residue
of 5 to 15 percent of the original wast~ volume must be dis-
posed of by landfill or other ultimate disposal or reuse.
Application of incineration to the disposal of in-
dustrial solid wastes has been limited historically to those
industries generating significant amounts of wastes having a
reasonable fuel value. For example, the lumbering and paper
pulp industries utilize large amounts of bark and wood wastes
as fuel, generally admixed with conventional fuels such as
VI-16
-------
fuel oilo Much of the so-called incineration of industrial
wastes is accomplished in single chamber incinerators and
conical or "tepee" burners, which are little more than con-
fined open burning.
Application of incineration as a feasible disposal
alternative will require the installation of modern, properly
designed equipment and the establishment of meaningful, rigid
standards of performance to accomplish the desired goal of
reducing or eliminating organic and other air pollutant emis-
sionso
Compostinq - Of the three principal alternatives,
composting has undergone the least study in the u.s. and
thus entails the greatest uncertainty as to practical ap-
plication and realistic costo Most, if not all, of the de-
monstration and "model" composting operations started in re-
cent years have failed, primarily for two reasons. First,
no steady market for the end product could be developed and,
second, odo~s from the aerobic decomposition were not con-
trolledo
The low organic content of typical American wastes
requires significant segregation, before composting and the
addition of large amounts of nutrients to achieve an accept-
able soil supplement or fertilizero Some agricultural ex-
perts believe that compost can never be marketed competitively
with the inorganic fertilizerso
Other Waste Disposal Alternatives - Other alterna-
tive disposal methods under study by various investigators in-
volve the segregation of wastes into various fractions such
as metal, paper, and plastic, and conversion or recycling to
provide usable materials. The Bureau of Mines (1970) in-
vestigated conversion of garbage and organic wastes into us-
able petroleum substituteso By heating the organic wastes
to 400°C under pressure with CO and H20, crude petroleum pro-
ducts were obtained. No cost estimates were reported, but
it was estimated that each ton of urban refuse could produce
one barrel of low-sulfur oil by this process.
Much publicity has been given to numerous studies
related to the recycling of wastes; with few exceptions, how-
ever, technological and economic feasibility is yet to be
demonstrated.
VI-17
-------
Advanced methods of incineration and mechanical
waste reduction are being investigated. These methods in-
clude pyrolysis, high temperature incineration, and mech-
anical shredding and baling for volume reduction.
The Solid Wastes Report (1970) described pilot
plants using pyrolysis and high temperature incineration.
At a 35 ton/day pyrolysis pilot plant, volume reduction of
94 percent was reported. Capital costs of a 1000 ton/day
unit were estimated at $10 million and operating costs at
$7.50/ton. Based on operation of a 100 ton/day high-tem-
perature incineration pilot plant, volume reduction was
reported as 97 percent, with capital costs estimated at
$15,000/ton and operating costs at $6-$8/ton.
Cost Comparison of A1ternattve Solid Waste Dis-
posal Methods - Accurate economic cost comparisons of the
alternatives to open burning depend on a number of para-
meters that vary widely with locale. For most municipal
regions, no one simple solution is generally applicable;
detailed study of the many factors involved in a satis-
factory systems mix of alternatives is required for real-
istic cost comparisons.
For our cursory analysis, a simple, hypothetical
situation illustrates some economic variables. An urban
region is postulated which can be serviced by a centralized
collection and transfer depot. This depot serves as the
starting point in our comparison, and thus we need not com-
pute the costs of various collection modes.
The centralized collection depot provides 100,000
tons/year of mixed total wastes for ultimate disposal.
Average hauling distances to the various disposal sites are
postulated as 2 miles to incinerator site, 5 miles to com-
posting site, and 10 miles to landfill site. Transportation
costs are estimated at $O.lS/ton mile plus $l/ton for dump-
ing or transfer. Thus, comparative costs of.transportation
and transfer are:
Round Trip Transportation and
Disposal Mode Distance (miles) Transfer Cost ($/ton)
Incineration 4 1.60
Composting 10 2.50
Landfilling 20 4.00
VI-18
-------
Incinerator cost studies were reported by A.D.
Little (1970). for a number of existing types of install-
ations in the capacity range near 400 ton/day. Battery
limits for evaluation were standardized; water quenching
and 99 percent efficient electrostatic precipitation were
assumed. Operation was established at three shifts per
day, 250 days per year. No waste heat recovery, byproduct
credits, or residue disposal debits were assumed. Disposal
costs range from $4.47/ton for a continuous, horizontal
cylinder incinerator to $9.50/ton for a residue fusion
(Dravo/FLK) type incinerator. A weighted average cost of
$5.50/ton is assumed for our analysis as being represent-
ative of incineration practice in the United States.
Both Combustion Engineering
General (1969) have analyzed landfill
this analysis we used the generalized
by Combustion Engineering:
Cost in $/ton = 0.50 + 6,000
tons/yr
(1969) and Aerojet-
operations. For
cost expression given
Thus, for 100,000 ton/year, the landfill cost is $0.56/ton.
Not enough examples of composting disposal in the
U.S. are available to allow estimation of average costs.
Such variables as market acceptance of the product cannot
be assessed. Aerojet-General reported unit costs of $5/ton
for a windrow operation and $lO/ton for a mechanical com-
poster. Kupchick (1966) reported cost data from 14 foreign
plants averaging $4.55/ton. Combustion Engineering (1969)
reported costs of foreign practices as ranging from 30 to
60 percent of the cost of municipal incineration. For our
example, we selected a value of $5/ton.
A cost of comparison of the three alternative dis-
posal modes for our hypothetical example is summarized in
Table VI-l.
Table VI-l - Disposal Costs for Hypotnetical Example -
100,000 tons/yr
Transportation Disposal Total
Cost Cost Cos t
Disposal Mode $/ton lli..r. j/ ton $/yr ...JiB:.
Landfi11i ng 4.00 400,000 0.56 56,000 456,000
Incineration 1. 60 160,000 5.50 550,000 710,000
Composting 2.50 250,000 5.00 500,000 750,000
VI-19
-------
A detailed investigation of a specific geographical
area, with recommendations tailored to the community needs and
practices was made by Aerojet-General (1969) in their study
of Fresno, California. Specific costs developed for the var-
ious disposal and processing methods were generally higher per
ton than those used in our example. For waste disposal in
the Fresno area, the recommended systems mix consisted of es-
sentially no incineration, composting of all suitable wastes,
and landfilling of the non-compostable portion. The projected
total cost for the year 2000 was about $14/ton (in 1967 dol-
lars) for the total waste management program.
c.
Solvent Reformulation or Substitution
Since the advent of Rule 66 in the Los Angeles area
and, more recently, similar legislative codes in other areas,
solvent reformulation has been an accepted and widely practiced
alternative for controlling solvent emissions.
Organic solvent emissions can be significantly miti-
gated by using solvents with low vapor pressures or photochem-
ical reactivities. These reformulations or substitutions are
governed by three basic factors: 1) adaptability to an exist-
ing process or formulation, 2) solvent performance, and 3)
economics.
Equipment and process modifications can be minimized
by substituting solvents that are in the same chemical class
as those originally used. The decreased vapor pressure of
higher homologs in any organic series benefits air pollution
control and may not require serious modifications to an exist-
ing process or formulation. Such a substitution does not
eliminate emissions, however, since the physical and chem~cal
characteristics of the original and the alternative solvents
are often similar.
Standards for measuring solvent performance depend
on the application. Solvency can, be classified according to
the aniline point, dilution ratio, or Kauri-Butanol number.
Solvency is a basic measure of solvent performance, but paint
formulations and solvent extractions are notable examples of
applications in which other parameters are needed to measure
solvent performance. Basically, solvent performance is
measured by the amount of solvent necessary to dissolve, or
otherwise carry, a unit amount of solute. This criterion is
applied to all solvents, but must be adjusted in accordance
with other standards for an overall evaluation of solvent per-
formance.
VI-20
-------
Both adaptability ,and performance influence the
economics of solvent reformUlation. A complete economic
evaluation would reflect influences from the initial pro-
duction to final consumptiono In commercial processes,
production and maintenance costs are more concrete, and
more easily predicted, than such variables as public opin-
ions on air pollution control criteriao
Solvent producers now have solvent blends that
can be used for comparison and compromise to suit most
processes. Values for solvency and other chemical and
physical characteristics are available for evaluation in
any application.
Surface Coatinqs - Larson and Sipple (1967) pub-
lished some experimental data and guidelines for reformulat-
ing paints to conf6rm to Rule 660 This report can be sum-
marized by the statement that oxygenated hydrocarbons ap-
pear to be the most promising alternatives for paint solvent
reformulations at the present timeo C4 through C9 ketones
have been mixed in various proportions to develop desirable
paint solvent characteristics, often in mixtures with other
oxygen-containing solventso Oxygenated hydrocarbon solvents
can be substituted into existing paint production equipment
without excessive modifications, and these mixtures will
usually meet solvency requirementso
Solvent reformulation has been used successfully
to reduce the photochemical reactivity index of oil-base
paint solvents. As yet, however, no solvent mixtures have
been prepared which can combine low photochemical reactivity
with solvent properties equal to those of the reactive aro-
matic hydrocarbons.
Oil-base paints enjoyed a monopoly of the industrial
surface coatings market, until recently. Industrial coatings
such as machinery paints, paper and wire coatings, wood pre-
servatives, marine paints, and traffic paints were dominated
by hydrocarbon solvents. The variety of problems involved
in developing solvent reformulations for such diverse ao-
plications has required a considerable amount of research.
A more recent advance in reformulation has been
the development of water-base paints. The advantages and
variations available in this class of paints were described
by Parker (1967). Applications for water-base paints are
being continually developed, and they have become an influ-
ential substttute for oil-base paints in architectural (trade
sales) application.
VI-2l
-------
Substitution of water as the main solvent does not
eliminate the problems of hydrocarbon emissions, however.
Most film-forming polymers are based on carbon chains, which
are insoluble in water. Water-solutions or emulsions of
these materials require the suspension of unstable, partially
polymerized species composed of molecules such as vinyl
chloride, butadiene, acrylonitrile, methyl, ethyl and butyl
acry1ates, and glycerol derivatives. Solution or suspension
of these resins in water also requires emulsifying agents such
as dibuty1- and diocty1-phthalate, hexylene glycol, esters of
fatty acids~ and paraffin waxes. In addition, fungicides and
other preservatives must usually be included in water-based
formulations.
We found in the literature no data on experimental
measurements of emissions from water-base formulations. Em-
pirical evaluations indicate that the excess of water in these
formulations should preclude hydrocarbon pollution problems,
but these evaluations may overlook the reactivity of hydro-
carbons present in these mixtures. In general, the volatile
hydrocarbon content of water based formulations is in the
range of from 3 to 5 percent.
in the c1e~~~a~~~gp;0~~~~~~~ ~~g~:~~~n~u~~a~~~e}~rP~~~~i~~~
erations as salvage and repair, plating, and surface coating.
Many of the smaller, non-automated operations utilize 1iquid-
phase solvent degreasing, either cold or hot; most larger pro-
duction and processing facilities utilize vapor-phase degreas-
ing.
A wide variety of solvents used in small shops for
liquid-phase degreasing ranges from gasoline and kerosene to
the chlorinated hydrocarbons. These small operations are al-
most always uncontrolled and types and amounts of emissions
are impossible to assess.
Until the application of Rule 66 in Los Angeles and
similar regulations in other areas, the predominant solvent
for vapor degreasing was trichloroethylene. In unregulated
areas, it still accounts for the majority of applications.
A review of studies made by one of the major industries us-
ing tric~10roethy1ene.was reported by the Aerospace Industries
Association (1968). In order to comply with the Rule 66
classification of trichloroethylene as a non-exempt solvent,
studies were made of the feasibility and economics of alter-
native procedures. These studies included testing of alterna-
tive organic solvents, testing of aqueous cleaning solutions
and economic evaluation of possible control techniques and
equipment modifications.
VI-22
-------
The investigation of alternative organic solvents
resulted in selection of inhibited 1 ,1,1-trichloroethane as
a satisfactory replacement for trichloroethylene. The
physical and chemical (solvent and corrosive) properties of
these two solvents are quite similar, 1,1,1-trichloroethane
having a slight heat requirement advantage due to its lower
vaporization temperature (166°F versus 188°F for trichloro-
ethylene). This may be offset by higher vaporization losses
for the more volatile 1,1, 1-tri chloroethane.
Detailed economics of controlling emissions by
equipment modification and control devices were not reported.
Estimates from four suppliers were given, however, ranging
from $70,000 to $170,000 per degreaser, with no guarantee of
meeting Rule 66 compliance.
The reported investigation of the use of aqueous
cleaning solutions as alternatives to vapor degreasing in-
cluded an economic comparison of the two methods. The com-
parison summary, shown in Table VI-2, is for comparably
sized facilities. The aqueous system requires, in addition
to the cleaning solution tank, a rinse tank and drying oven,
which are not required for the vapor degreaser. Aqueous
solution and rinse tanks were heated to 165°F, while the
drying oven operated at 150°F. Solvent costs were based on
a price of $1:20 per gallon, and an industry average usage
loss of 1/2 gallon per day per square foot of tank top open-
ing. Usage of the aqueous cleaning compound was based on
complete renewal each month, with 15 percent makeup additions
during the month. Labor costs are not included in the com-
parison, since these probably would be nearly identical.
c.
ECONOMIC ASPECTS OF HYDROCARBON EMISSION CONTROL
L
General
As a result of the review of emission control tech-
nology, several systems were identified as being specifically
pertinent to the control of hydrocarbon emissions from station-
ary sources. Additionally~ the study of sources and their
emissions identified and ranked those categories of sources
which, if controlled, would result in the greatest benefit by
reduction of emissions. In order to assess the economic im-
pact of emission control of these various sources and to pro-
vide a frame of reference for evaluation of other control
methods or alternatives, engineering cost studies were made
for four types of control schemes: gasoline storage control,
waste gas incineration, adsorption, and wet scrubbing.
VI-23
-------
Table ~I-2 - Typical Comparative Cost Data:
Solvent Vapor Degreasing Versus Aqueous Cleaning
Aqueous
Degreasing
Solvent
Degreasing
Basic Equipment
Aqueous - 2 tanks + drying oven,
including heating and exhaust
systems
- Degreaser
$4050
$3050
3600
1500
$7650 $4550
Degreaser + still + exhaust
Floor space $15/sq ft - Aqueous
TOTAL FACILITIES COST
Operating Costs (21-day month,
24-hour day)
Solution cost - Aqueous
Degreasing solvent cost
600
504
Heating - steam 15 psi at
$1.50/10001b
Aqueous - 2 tanks - heat loss
Drying oven
173
45
Degreaser - Vaporizing solvent
+ heat loss + still
117
Water cost - $0.14/1000 gallons
~.
Aqueous - Changing water in rinse
tank each 2 hours
53
Degreaser and still - Cooling water
TOTAL OPERATING COST
$871
55
$676
Note: Based on facilities 10 feet long by 4 feet wide
by 7 feet working depth.
VI-24
-------
2.
Gasoline Storage Control
Evaporative losses from the storage of gasoline
and other volatile petroleum and petrochemical liquids con-
stitute a significant portion of the total hydrocarbon emis-
sions from industrial sources. The general case of control
of gasoline storage losses was chosen to illustrate the
economic costs of such emissions.
a.
General Case Description
Control costs for pontoon floating roof tanks and
fixed roof tanks with retro-fitted internal floating covers
were compared with costs for conventional cone roof or fixed
roof tanks. Three tank sizes were selected to cover the
range of general operating practice: 50,000; 100,000; and
150,000 barrel capacity.
b.
Insta]]ed Capital Cost
Total installed costs for the three tank types
were obtained by adding estimated values for excavation,
foundation, electrical grounding, piping, and painting to
the average of installed-cost quotes provided by tank vendors.
An allowance for engineering was also included in the total.
These estimated costs are tabulated in Table VI-3 and plotted
in Figure VI-l. A cost breakdown consisting of vendor
quotes and construction cost estimate sheets is presented in
Appendix A.
The total instal1~d prices of the fixed or cone
roof tank and the converted cone roof tank with internal
floating cover are the lowest and highest costs, respectively;
the total installed cost of the pontoon floating'roof tank
lies between these two. These prices indicate that for a
new installation, it is less expensive to install a pontoon
roof tank than a fixed roof tank with an internal floating
cover for a gasoline vapor emission control. Where a means
of emission control is required for existing fixed roof tanks,
however, the addition of an internal floating cover is the
most economical conversion. According to tank manufacturers
contacted, conversion cost of removing the cone roof and
installing a pontoon roof in its place is roughly twice the
cost of adding an internal floating cover.
c.
Opera ti ng Cos ts
Annual operating ~osts for each type of tank were
taken to consist of the following:
VI-25
-------
Table VI-3 - Estimated Installed Costs of Gasoline Storage Tanks
Nominal Tank Capacity, 50,000 100,000 150,000
Barrels
Size 90' dia x 481 120. dia x 481 1501 d i a x 48.1
Installed Costs, $
Fixed Roof Tank 161,000 257,000 379,000
Pontoon Floating
Roof Tank 176,000 279,000 403,000
Internal Floating
Cover in Existing
Roof Tank 34,000 54,000 68,000
NOTE:
Installed cost estimates are developed
in Appendix A.
VI-26
-------
500
400
tOt
,
o
-
x 300
v.-
..
+J
VI
o
U
"'0
C1J
::: 200
10
+J
VI
c::
-
100
o
Total Cost Cone Roof Tank
Converted with Internal
Floating Roof
.",
./
.".,.'"
/
/
/
/ //
/ /
/ /.
/
/ /
/
Internal Float Cover on
Existing Cone Roof Tank
(Incremental Cost-Conversion)
VI-27
--
--
-
200
Figure VI-l - Estimated Installed Cost of Gasoline
Storage Tanks
~
Pontoon Floating
Roof Tan k
./
./.",
./ ""
,,/ '"
./
,
--
--
-
--
o
50 100 150
Capacity, Barrels x 10~
-------
Maintenance
@ 2% Capital Cost
@ 10% Capital Cost
Depreciation
Property Taxes
Insurance
@ 1% Capital Cost
@ .5% Capital Cost
Corporate Overhead
Gasoline Loss
@ 3% Plant Level Cost
@ $5.50/Barrel or
$.131/gal.
Loss Calculations
Gasoline losses for fixed roof and pontoon float-
ing roof tanks were calculated in accord with loss equations
presented in American Petroleum Institute publications.
Losses for fixed roof tanks with internal floating covers
were calculated as a percentage of the losses determined for
equivalent fixed roof tanks.
d.
Evaluation of Storage Costs
Evaporation losses calculated for gasoline storage
tanks are summarized in Table VI-4. Loss values were ob-
tained by assuming Reid vapor pressure at 9.0 PSIA, tank
outage (height of vapor space above liquid) at 50 percent or
241, and tank location at Philadelphia, Pennsylvania. Meteor-
ological data presented at API Bulletin 2513 were used as
required. The pontoon floating roof, internal floating cover
conversion, and fixed roof storage systems, in that order,
are decreasingly effective in reducing vapor emissions.
Operating costs are summarized in Table VI-5 and
Figure VI-2 on an annual basis as total cost and cost per
1,000 barrels of capacity. Manufacturing estimate sheets
for development of operating costs are included in Appendix A.
In general, ,the highest operating costs occur in the type of
storage system which exhibits the highest vapor loss, and
vice versa.
An incremental capital investment vs operating
cost analysis, which uses the fixed roof tank system as a
basis for comparison, is illustrated in Table VI-6. A payout*
of 2 to 4 years is realized for the additional capital re-
quired to construct a new pontoon floating roof tank instead
of a fixed roof tank. Where internal floating covers are in-
* Payout (in years) = capital cost/after tax savings plus
depreciation, thus estimated payout is critically de-
pendent on choice of depreciation rate.
VI-28
-------
Table VI-4 - Summary of Estimated Gasoline Losses
Tank Capacity (Bb1s) 50,000 100,000 150,000
Fixed Roof: Bb 1 /yr 1,365 2,345 3,560
Ga1/yr 57,300 99,500 149,500
Pontoon Floating Roof: Bb1/yr 98 150 205
Ga1/yr 4,120 6,300 8,820
Fixed Roof/Internal
Floating Roof: Bbl/yr 273 469 712
Gal/yr 11,450 19,600 29,800
NOTE: Gasoline losses are calculated in Appendix A.
VI-29
-------
Table VI-5 - Summary of Estimated Operating Costs
For Gasoline Storage Tanks
Tank Capacity, Barrels
50,000
Annual Operating Costs, $
Fixed Roof Tank
29,900
25,000
Pontoon Floating Roof Tank
Internal Floating Cover
(In Fixed Roof Tank)
28,600
Annual Operating Cost/Capacity,
$/1 ,000 Ba,rrels
Fixed Roof Tank
600
500
Pontoon Floating Roof Tank
Internal Floating Cover
(In fixed Roof Tank)
570
100,000
48,600
39,500
45,600
490
400
460
NOTE: Operating costs are developed in Appendix A
V 1- 30
150,000
72,200
57,100
66,000
480
380
440
-------
100
...
.
o
, -
)( 70
4A-
..
'+, 60
en
o
,u
0) 50
c
'r-
+'
ItS
'- 40
C1I
c.
o
- 30
ItS
:::::I
C
C
C1: 20
90
80
10
o
(a) Cone Roof Tank
(b) Pontoon Floating Roof Tank
(c) Internal Float Cover/Cone Roof Tank
/
'/
'/
'/ /
'/
/ ,/
/ '/
/
c),. ,/
./
,/
/'
- - --
- - --
_...---
1000
:>
900 ;
c
s:»
-
800 0
'0
(1)
"'1
700 s:»
r+
.....
:::::I
~
600 n
o
en
r+
500 ;;-
s:»
~
s:»
n
400 -
r+
'<
..
300 ~
.......
-
..
o
200 0
o
c:r
c:r
100 -
o
200
Figure VI-2 - Estimated Operating Cost of Gasoline Storage
Tanks
"
",
././
." ................ ./
- - "...
- -
-
o
50 100 150
Capacity. Barrels x 10.3
VI-31
-------
Table VI-6 - Differential Savinqs Vs Differential Capital
Investment and Payout (Base Case -
Fixed Roof Tank)
Tank Capacity, Barrels
50,000
Pontoon Roof
Differntial Installed
Capital Cost, $
Differential Net Savings
(Before Tax), $
15,000
4,900
Differntial Tax, @ 50%, $
2,450
2,450
Differntial Savings
(After Tax), $
Payout After Taxes (yr)
3.80
Internal Floating Cover in
Existing Fixed Roof Tank
Differential Installed
Capital Cost, $
34,000
Differential Net Savings
(Before Tax), $
Differential Tax @ 50%, $
1 ,300
650
650
Differential Net Savings
(After Tax), $
Payout After Taxes (yr)
8.40
VI-32
100,000
22,000
9,100
4,550
4,550
3.26
54,000
3,000
1,500
1 ,500
7.83
150.000
24,000
15,100
7,550
7,550
2.41
68,000
6,200
3,100
3,100
6.87
-------
stalled in existing fixed roof tanks to control vapor emis-
sions, payouts based on gasoline savings are much longer.
This method, however, is more economical than converting
the fixed roof tank to a pontoon floating roof tank, in
which case no payout exists at all. This statement is sup-
ported by the cost data in Appendix A, assuming a tank con-
version cost double that for the internal floating cover.
3.
Waste Gas Incineration
Because of the wide variety of sources and emis-
sions controlled by both thermal and catalytic incineration,
the choice of a representative general case system was some-
what arbitrary. For our evaluation of incineration, both
thermal and catalytic, we chose an example, given by Danielson
(1967), of a paint bake oven operating on a surface coati~g
production line.
a.
General Case Description
The emission source being controlled by incineration
is a paint bake oven exhausting an air stream at 375°F, con-
taining a 50/50 weight percent mixture of benzene and hexane.
The oven was assumed to operate two shifts per day, 365 days
per year. The range of operating variables included oven ex-
haust volumes of 1,000 SCFM, 10,000 SCFM, and 20,000 SCFM and
total organic vapor concentrations equivalent to 15 and 25
percent of the lower explosive limit (LEl).
For th~ 50/50 benzene/hexane mixture the calculat~d
LEL is 1.2 percent by v9lume. Thus, the 15 and 25 percent
LEL levels are equivalent to 0.18 and 0.30 percent by vQlume.
Cost comparisons are made for cases of no heat re-
covery, primary heat recovery, and primary plus secondary
heat recovery.
The design parameters for the thermal incinerator
were a residence time of 0.5 second and an exhaust tempera-
ture reported by Danielson (1967) to give 90 percent con-
version of hydrocarbons and carbon monoxide to carbon diox-
ide. For the catalytic incinerator, the oven exhaust gas
was preheated to 650°F and the incinerator exhaust design
temperature range was 1000° to 1200°F.
Reaction kinetics computations performed by one
incinerator vendor (shown in Appendix B) indicate that
catalytic conversion of the hexane/benzene system at 25 per-
VI-33
-------
cent lEL is highly difficult because of th0 nresence of hex-
ane. This is generally true of most higher carbon aliphatic
compounds. Under these conditions satisfactory conversion
usually requires that the catalyst surface area be increased
or the space velocity be reduced as much as possible. However,
sufficient turbulence must be maintained to assure uniform
combustion. The aliphatic aromatic system of hexane/benzene
is chosen because it is typical of paint system diluents (e.g.,
naphthas may have 30% or higher aromaticity level), and be-
cause it entails considerations that are significant in select-
ing an incineration system.
Schematic diagrams of the systems without heat re-
covery are shown in Figures VI-3 and VI-6.
b .
Heat Recovery
Primary, and primary with secondary, heat recovery
systems were incorporated in the catalytic and thermal incin-
eration systems and evaluated over the ranges of exhaust capacity
and concentration specified for the basic incinerators.
The primary heat recovery process consists of utiliz-
ing a heat exchanger to transfer heat from incinerator exhaust
gases to oven exhaust gases, which are being fed to the incin-
erator.
Secondary heat recovery is defined as the process in
which incinerator exhaust gases cooled by primary heat exchange
are used to heat make-up air being fed to the baking oven.
The heat recovery processes substantially reduce,
and in some cases eliminate, the fuel required to heat the
oven exhaust and makeup streams to required temperature levels.
Thermal efficiencies between 45 percent and 47 per-
cent were specified with equipment vendor's order-of-magnitude
quotes for primary heat exchangers. These values were also
applied to secondary heat recovery systems. Thermal efficiency
was defined by the equation:
t2 - tl x 100
Percent Efficiency = Tl - tl
tl = Entering cold stream temperature, OF
t2 = Leaving cold stream temperature, of
Tl = Entering hot stream temperature, of
VI-34
-------
Process schemes for thermal and catalytic incinerators
with primary heat recovery are shown on the flow sheets
in Figures VI-4 and VI-7, respectively.
Thermal and catalytic incinerators with primary
and secondary heat recovery are shown in the flow sheets of
Figures VI-5 and VI-8. The additional fan used to handle
make-up air is included to account for pressure drop in
the exchanger and make-up air line.
'c.
Installed Capital Cost - Incineration
The total installed capital costs for thermal
and catalytic tncineration systems were based on written
and verbal quotes received from two vendors and given in
Appendix B. Each vendor quoted for both types of incin-
erators at one of the concentration levels, 15% LEL or
25% LEL, investigated. Within the scope of this study,
these quotes were interpreted as being consistent with
each other, or as provided by a single vendor. This as-
sumed consistency of quotations was based on one vandor's
additional quote which was in reasonable .agreement with
his competitor's prices for three systems. The quotations
cover the majority of incineration systems with and without
primary heat recovery. Estimates for secondary heat re-
covery systems were computed by utilizing plots of vendors'
quotes, and assuming the following for secondary heat ex-
changers: a) thermal efficiency of 45 percent, b) overall
heat transfer coefficient of 5.0 Btu/hr ft2 of, c) cost
curve of the form, f(x) = Ax.s6where
f(x) = Cost, $
A = Constant
x = Heat Exchanger Area, ft2
Quotations also included vendors best estimate for install-
ation cost ~s a percentage of capital equipment cost.
Fifty percent was used for the integral erection of a pro-
cess and a control system. One hundred percent was used for
a retrofit. Installation included costs for excavation,
foundations, piping, ductwork, insulation, electrical,
painting, and engineering. Total installed costs of incin-
erator systems are presented in detail in Appendix B on
construction cost estimate sheets. Values for 15 percent
and 25 percent LEL systems over a capacity range of 1,000
SCFM to 20,000 SCFM are summarized in Table VI-7 and VI-8
and plotted in Figures VI-9 and VI-10, respectively.
VI-35
-------
Exhaust
Wet Pain
Materia
1400° F -1450° F
Natural Gas
Fuel
...
Thermal
Incinerator -
375°F (" I
~
Booster
Fan
ted Finished Material
1 in Baking Oven Out
Make-Up
Air
Figure VI-3 - Thermal Incinerator Without Heat Recovery -
Flow Sheet
V I - 36
-------
Exhaust
935°F...9500F
Natural Gas
Fuel
Primary
Heat
Exchanger
840°F
875°F
Thermal
Incinerator
Boos ter
Fan
14000F...14500F
Wet Painted
Materi al n
Finished Material
Baking Oven
Out
Make...Up
Air
Figure VI...4 ... Thermal Incinerator With Primary Heat
Recovery... Flow Sheet
VI-37
-------
Wet Painted
Material in
Exhaust
Secondary
Heat
Exchanger
LG< 65°F
/ "-
Make-Up
Air Fan
625-640°
375°F
935°F-
950° F
Primary
Heat
Exchanger
375°F
Booster
Fan
Baking Oven
Finished Material
). Out
Natural Gas
Fuel
840°F
875°F
Figure VI-5 - Thermal Incineration With Primary
and Secondary Heat Recovery - Flow
Sheet
VI-38
Thermal
Inci nerator .
l400°F-1450°F
-------
Wet Painted
Material in
Baking Oven
Make-Up
Air
Exhaust
I
lOOO°F-12000F
Natural
Gas
Fuel
375°F
655°F
Catalytic
Incinerator
Preheat
Burner
Booster
Fan
Finished Material
Out
Figure VI-6 - Catalytic Incinerator Without Heat Recovery -
Flow Sheet
VI-39
-------
E xh a us t
f
720°F-920°F
Pri mary
Heat
Exchanger
655°F
Catalytic
I nc1 nerator
375°F
Boos ter
Fan
1000°F-1200°F
Wet Painted
Materi al n
Baking Oven
Finished Material
Out
Make-Up
Air
Figure VI-7 - Catalytic Incinerator With Primary Heat
Recovery - Flow Sheet
VI-40
-------
Wet Painted
Materi al n
Exhaust
Secondary
Heat
Exchanger
410°F.fl
610°F,
375°F
Booster
Fan
Baking Oven
Primary
Heat
Exchanger
375° F
LG<
Make-Up
Air Fan
720°F-
920°F
Finished Material
') Out
65°F
655°F
Catalyti c
Incinerator.
lOOOOF-1200°F
Figure VI-8 - Catalytic Incinerator With Primary and
Secondary Heat Recovery - Flow Sheet
VI-41
-------
Table VI-7 - Estimated Installed Costs Thermal and Catalytic
Incinerators - 15% Lower Explosive Limit
Incinerator Capacity, scfm
Installed Costs, $
Catalytic without Heat Recovery
Catalytic with Primary Heat
Recovery
Catalytic with Primary and
Secondary Heat Recovery
Thermal without Heat Recovery
Thermal with Primary Heat
Recovery
Thermal with Primary and
Secondary Heat Recovery
VI-42
1,000
22,300
39,600
60,700
20,300
37,600
53,600
10,000
60,900
121,800
196,900
40,600
101,500
158,300
20,000
91,400
182,700
295,400
50,800
142,100
228,400
-------
Table VI-8 - Estimated Installed Costs Thermal and Catalytic
Incinerators - 25% Lower Explosive Limit
Incinerator Capacity, scfm
1,000
Installed Costs, $
Catalytic without Heat Recovery
32,500
40,400
Catalytic with Primary Heat
Recovery
Catalytic with Primary and
Secondary Heat Recovery
51 ,000
Thermal without Heat Recovery
Thermal with Primary Heat
Recovery
20,300
30,100
Thermal with Primary and
Secondary Heat Recovery
40,600
VI-43
10,000
2@3,000
228,400
270,000
69,000
107,600
149,200
20,000
376,000
428,300
497,400
102,500
173,600
240,600
-------
300 .
'
Legend /
Thermal Incineration
Catalytic Incineration /
ill Without Heat Recovery.
b With Primary Heat Recovery /
250 c) With Primary and Secondary
Heat Recovery /
(c) /
... /
I
0 /
-
200
)(
~ /
.. ,/
+.)
1/1 /
0 /.
u
~ ./'
.3 150 / (b) /'
ItS
~
Q) /
c /
.r- (b)
u /
c
....
"'0
Q)
- 100
-
ItS
+.) -
1/1 ---
C (~ --
....
----
~
.~ ( a )
50 ~
,/
~
o L-r-.---'--T--'-,
o 5 10 15
Incinerator Capacity, SCFM x 10-3
Figure IV-9 - Estimated Installed Incinerator Cost-15% LEL
(90% Combustion of 50/50 - Hexane/Benzene
Mixture @ 15% Lower Explosive Limit)
20
VI-44
-------
500
400
'"
I
0
-
)(
-4A-
..
-+-J 300
VI
o
u
~
0
-+-J
"'
~
Q)
c
.,...
u 200
.c
-
"'0
Q)
-
-
"'
-+-J
VI
C
-
100
Legend
/
/
Thermal Incineration
Catalytic Incineration
raT Without Heat Recovery
(b) With Primary Heat Recovery
(c) With Primary and Secondary
Hea t. Recovery .
/
/
/ /
/ /
.. (cV / . /
. /(b)/ /
/ / /
/ / (a) /
1//
/ / /
/ /
////
// //
/ / //
//~
//
~
o
o
1
Incinerator Capacity, SCFM x 10-9
20
Figure IV-10 - Estimated Installed Incinerator Cost-25% LEL
(90% Combustion of 50/50 - Hexane/Benzene
Mi.xture @ 25% Lower Explosive Limit)
VI-45
-------
Analysis of these data indicated that total in-
stalled costs of catalytic incinerator systems are higher
than those of thermal incinerator systems over the concen~
tration and capacity ranges studied. Moreover, as concen~
tration and capacity increase, the cQst difference of cat-
a1yti~ ~ver thermal incineration increases.
It is generally accepted that installed cost for
a given incineration method increases with concentr~tion
level. However, exceptions to this appear in cases of
thermal incineration with heat recovery systems at low
volume capacities. This discrepancy may be the result of
reduced burner and heat exchanger sizes or invalidity in
extr~po1ating budget quote prices to low capacities~
A reduction in burner size occurs when the stream's
thermal value or concentration increases at a given capa~ity.
Heat exch~nger surface area becomes smaller when temperature
driving force is increased, as is the case with the higher
exhaust temperature of the thermal vs catalytic incinerators.
Extrapolation of the data presented indicates that
installed costs of both types of incineration, when compared
on an equal flow rate basis, are the same at a point b~tween
10 percent and 15 percent LEL concentration. Above this con-
centration range, thermal systems are economically favoreQ;
catalytic systems are favored at lower concentrations.
d.
Operating Cost - Incineration
Operating costs for incineration control processes
were calculated on the manufacturing estim~te sheets included
in Appendix B. The values were determined by incorporating
the following percentages and rates in the calculations:
Operating - 2 shifts/day, 365 days/year,
0.5 man hour per sjoft direct
labor
Maintenance @ 5% Capital Cost
Depreciation @ 10% Capital Cost
Property Taxes @ 1% Capital Cost
Insurance @ 0.5% Capital Cost
Corporate Overhe~d @ 3% Plant Level Cost
VI-46
-------
Supervision @ 10% Direct Labor
Payroll Overhead @ 30% Direct Labor
Plant Overhead @ 50% Direct Labor
Labor @ $4.00/hr
Natural Gas @ $.60/106 Btu
Electricity @ $.015/KWH
Catalyst:
(Estimated Catalyst Life of 15,000 hr)
@ $1.00/103 SCFM of Capacity
@ $6.00 to $7.00/103 SCFM of
Capacity
For convenience of calculation, fuel consumption was cal-
culated on the basis of total adiabatic combustion of fuel,
adiabatic system operation, and combustion within the con-
fines of the incinerator only. In practice, fuel con-
sumption can vary up to 30 percent higher than the calcu-
lated values, depending upon system design and operating
conditions. Partial combustion is also known to occur in
the primary heat exchanger if autoignition conditions
are approached.
15% LEL -
25% LEL -
The high value and wide range of catalyst cost
is caused by the presence of hexane, which is difficult
to burn catalytically at the higher solvent concentration
levels investigated. Computer printouts that confirm the
larger volume of catalyst needed to convert the hexane are
given in Appendix B.
Costs of operation are presented on an annual
basis as total cost and cost per 1,000 SCFM processed.
Values for 15 percent LEL catalytic and thermal systems
are tabulated in Table VI-9. Plots of this information
are illustrated in Figures VI-ll and VI-12. In the same
manner the values for 25 percent LEL systems are pre-
sented in Table VI-10 and Figures VI-13 and VI-14.
Inspection of Tables VI-9 and VI-10 and the
graphical presentations in Figures VI-ll through VI-14
leads to the conclusion that operating costs of catalytic
VI-47
-------
Table VI-9 - Estimated Annual Operating Costs Therm~l and
Catalytic Incinerators - 15% Lower Explosive Limit
Incinerator Capacity, scfm 1 ,000 10,000 20 ,000
Catalytic without Heat Recovery 8,600 31,500 55,000
$/yr
$/1000 scfm processed .0246 .0090 .0079
Catalytic with Primary Heat
Recovery, $/yr 10,400 29,800 46,400
$/1000 scfm processed .0296 .0085 .0066
Catalytic with Primary and
Secondary Heat Recovery, $/yr 13,000 31,20Q 43,000
$/1000 scfm processed .0372 .0090 .0062
Thermal without Heat Recovery, $/yr 9,800 43,100 78,300
$/1000 scfm processed .0280 .0123 .0112
Thermal with Primary Heat
Recovery, $/yr 10,500 31 ,600 50,000
$/1000 scfm processed .0300 .0090 .0070
Thermal with Primary and
Secondary Heat'Recovery, $/yr 12,300 29,900 42,100
$/1000 scfm processed .0350 .0085 .0060
VI-48
-------
Ta~le VI-l0 - Estimated Annual Operating Costs Thermal and
Catalytic Incinerators- 25% Lower Explosive Limit
Incinerator Capacity, scfm
Gatalytic without Heat Recovery
$/yr
$/1000 scfm processed
C~talytic with Primary Heat
Recovery, $/yr
$/1000 scfm processed
Catalytic with Primary and
Secondary Heat Recovery, $/yr
$/1000 scfm processed
. Thermal without Heat Recovery
$/yr
$/1000 scfm processed
Thermal with Primary Heat
Recovery, $/yr
$/1000 scfm processed
Thermal with Primary and
Secondary Heat Recovery, $/yr
$/1000 scfm processed
VI-49
1 ,000
11 ,800
.0337
12,10Q
.0344
12,900
.0369
8,800
.0292
8,500
.0282
9,300
.0330
1 O,~,OOO,
74,900
,0211
68,900
.0197
66,400
.0190
39,000
.0110
26,300
.0075
23,700
.0068
20,000
,
142,000
.0203
130,000
.0186
122,700
.0175
69,000
.0099
42,700
.0061
34,800
.0049
-------
100
90 ( c)
... 80
I
0
-
)( 70
VI-
.. 60
~
II)
0
u
C) 50
s:::
.r-
~
ItS
~ 40
.C1J
c..
0
- 30
ItS
:;,
s:::
~ 20
10
0
0
(a) Without Heat Recovery
(b) With Primary Heat Recovery
(c) With Primary and Secondary
Heat Recovery
.04
o
"'0
(!)
-s
QI
.03 rfo
.....
:;,
to
n
o
II)
rfo
.(,A
"-
.02 -'
(b) 0
o
( c) 0
(,I)
n
-r,
-c
-s
o
.01 (")
(!)
(a) II)
II)
(b) (!)
0.
i c)
5 10 15
Incinerator Capacity, SCFM x 10-3
20
Figure VI-ll -
Estimated Catalytic Incinerator Operating
Cost-15% LEL (90% Combustion of 50/50 -
Hexane/Benzene Mixture @ 15% Lower Ex-
plosive Limit)
VI-50
o
-------
100
90
...
I 80
0
-
)(
~ 70
..
+J
II) 60
0
U
01
c:::: 50
'r-
+J
to
~
C1J 40
0.
0
-
to 30
~
c::::
c::::
c:C
20
10
0
0 5
(a) Wi~hout Heat RecQvery
(b) With Primary Heat RecQvery
(c) With Primary and Second~ry
Heat Recovery
(a)
Ab)
---
10 15
Incinerator Capacity, SCFM x 10-3
20
Figure VI-12 - Estimated Thermal Incinerator Operating Cost-
15% LEL (90% Combustion of 50/50 - Hexane/
Benzene Mixture @ 15% Low~r Explosive Limit)
VI-51
.04
o
"0
t1)
"OS
CI
("to
-
. 03 ~
("')
o
II)
("to
..
-CA
.....
-
o
.02 0
o
U')
("')
- ."
~ ""C
"OS
o
( a) n
t1)
1- II)
.01 II)
t1)
(b) 0.
(c)
o
-------
150
140
(a) Without Heat Recovery
(b) With Primary Heat Recovery
(c) With Primary and Secondary
Heat Recovery
130
120
.06
.05
110
100 .04
(c)
90 -
(b
80 (a
.03
70
60
50
40
30
20 l'
1: L--
o .
Incinerator Capacity, SCFM x 10-3
Figure VI-13 - Estimated Catalytic Incinerator Operating
Cost (90% CDmbustion of SO/50 - Hexane/
Benzene Mixture @ 25% Lower Explosive Limit)
VI-52
.02
.01
- 0
-------
100
90
'"
I 80
o .
-
x
~ 70
..
+'
~ 60
Co.)
en
c:
.,.. 50
+'
co
s..
~ 40
o
-
~ 3Q
c:
c:
c(.
20
10
(a) Without Heat Recovery
(b) With Primary Heat Recovery
(c) With Primary and S~condary
Heat Recovery
\ \
\\-- (~
\~""'''''''---- ----/~~-( c) ---
t ~.... >----<--~:
-'. /~
~-
-;~. . ~=--_..__._-_.Jb)
~ -- -\c1
o
I -r--r-r'--r--..,'--,--- I'-r--r
5 10 15
Incinerator Capacity. SCFM x 10.3
20
o
Figure VI.14 . Estimated Thermal Incinerator Operating Cost-
25% LEL (90% Combustion of SO/50 . Hexane/
Benzene Mixture @ 25% Lower Explosive Limit)
VI-53
.04
o
"0
(1)
"'1
QI
. 03 ~
::s
to
n
o
VI
r1'
~
......
.02 .....
o
o
o
V)
n
"'T1
""0
"'1
o
n
.01 ~
VI
(1)
Q.
o
-------
systems are considerably higher than those for thermal sys-
tems at 25 percent LEl but tend to approach or become lower
than costs of thermal systems near 15 percent lEl.
Extrapolation of the data to concentrations below
15 percent lEl justifies this conclusion; costs for all
thermal and catalytic systems evaluated appear to reach
equivalent operating costs within the range of 13 percent
to 18 percent lEl. On this basis, thermal and catalytic
units above and below this concentration cost range are pre-
ferred as the control method provided incineration is es-
tablished as the most economical fume abatement process.
The addition of heat recovery equipment in the
15 percent to 25 percent lEl range has a greater effect
in reducing the operating costs of thermal than of cat-
alytic incinerators. This is confirmed by data in Figures
VI-ll and VI-12 and Figures VI-13 and VI-14. The displace-
ment between the (a) and (b) curves is greater for the
thermal units.
Figures VI-ll through VI-14 show further that
equivalent operating costs for incinerators with and with-
out heat recovery occur at lower capacities for thermal
than for catalytic incinerators. Also, as solvent concen-
tration increases, the incinerator capacity at which the
equivalent cost point occurs decreases.
e.
Control Method Evaluation - Incineration
Within the scope of this report, the use of incin-
eration as an emission control process is economically just-
ified when a combustible solvent cannot be recovered in a
sufficiently uncontaminated condition for reuse in production.
The choice of catalytic or thermal incineration is
based on process feasibility and economics.
The presence of catalyst poisons or hydrocarbons
that are difficult to convert catalytically usually indicates
the use of thermal incineration.
Solvent concentration level in the exhaust air
stream is the greatest factor influencing control process
economics. Thermal incineration is economically preferred
at concentrations above the 13 percent to 18 percent lEl
range; catalytic incineration is preferred at concentrations
VI-54
-------
below this range. For systems in which stream concentrations
fall within this range, an incineration control method should
be selected only after thorough consideration of alternatives.
Addition of heat recovery systems to an incineration
process is usually analyzed on the basis of the plant's need
for more heat and control process payout. '
An incremental analysis of savings versus capital
investment and payout is presented for primary and primary
with secondary heat recovery systems as applied to the incin-
erator processes evaluated.
Tables VI-ll and VI-12 present data pertinent to
operation at 15 percent and 25 percent LEL exhaust concen-
trations, respectively. The cost analyses are made by de-
termining differences between total costs (installed and
operating) for a given incineration system with and without.
heat recovery and then calculating a payout after taxes.
The analyses show that payout time for heat re-
covery is affected by system capacity and hydrocarbon con-
centration of these factors increase. Payout time for heat
recovery is also reduced as incinerator op~rating temperature
increases. because the higher thermal driving force lessens
heat exchanger costs. The shorter payout time of thermal
versus catalytic systems illustrates this point.
4.. Adsorption
Control of hydrocarbon emissions by adsorption on
activated carbon is generally applied when recovery of ad-
sorbed material is economically desirable. Other applications
are for control of very low concentrations of noxious organics
not readily handled by catalytic incineration and for collect-
ing and concentrating low-concentration emissions for sub-
sequent disposal by incineration.
a.
General Case Description
For convenience of calculation and to allow compar-
ison with thermal and catalytic incineration systems, the
adsorption systems studies were based on the same emission
source: a paint bake oven eXhausting benzene/he~ane in air
at 375°F.
VI-55
-------
Table VI-ll - Differential Savings Versus
Differential Capital Investment and Payout
15% Lower Explosive Limit - (B~se Case - In9inerators
Without Heat Recovery)
Incinerator Capacity, scfm
Catalytic with Primary Heat Recovery
vs Catalytic without Heat Recovery
1 ,000
10,000
, .
20,000
,
Differential Installed Cap. Cost, $
Differential Gross Savings (Loss), $
Differential Tax @ 50%, $
Differential Net Savings (After Tax),$
...
60,~00 91 ,300
1 ,700 8,600
850 4,300
8~0 4,300
8.8 6~8
17,300
(1,800)
Payout Cap. Inve~tment, Jr.
Catalytic with Primar{ and Secondary
Heat Recove~y vs Cata ytic without
Heat Recover~ .
Di fferenti a 1 Installed Cap. Cost, $ 38,400 136,000 204,000
Differential Gross Savings (Loss), $ (4,400) 300 12,000
Differential Tax @ 50%, $ 150 6,000
Differential Net Savings (After Tax), $ 150 6,000
Payout Cap. Investment, Yr. 9.9 7,7
Thermal with Primary Heat Recovery
vs Thermal without Heat Recovery
.
Differential Installed Cap. Cost, $ 17,300 60,900 91 ,300
Differential G ros s Savings (Loss), $ (700) 11 ,500 28,300)
Differential Tax @ 50% 5,750 14,150
Differential Net Savings (After Tax), $ 5,750 14,150
Payout Cap. Investment, Yr. 5. 1 3,9
VI-56
-------
Table VI-11 (Continued)
Incinerator Capacity, s cfm 1 ,000 10,000 20,000
The rma 1 with Primary and Se con da ry
Heat Recovery vs Thermal without
Heat Recovery
Differential Installed Cap. Cost, $ 33,300 117,700 117,600
Differential Gross Savings (Loss), $ (2,500) 13,200 36,200
Differential Tax @ 50%, $ 6,600 18,100
Di fferenti a1 Net Savings (After Tax),$ 6,600 18,100
Payout Cap. Investment, Yr. 6.4 3.9
VI-oS7
-------
Table IV-12 - Differential Savings Vs Differential
Capital Investment and Payout
25% Lower Explosive Limit - (Base Ca$~ - Incinera~ors
Without Heat Recovery)
Incinerator Capacity, scfm
Catalytic with Primary Heat Recovery
vs Catalytic without Heat Recovery
Differential Installed Cap. Cost, $
Differential Gross Savings (Loss), $
Differential Tax @ 50%, $
Differential Net Savings (After Tax),$
Payout Cap. Investment, Yr.
Catalytic with Primary and Secondary
Heat Recovery vs Catalytic Without
Heat .Recovery
Differential Installed Cap. Cost, $
Differential Gross Savings (Loss), $
Differential Tax @ 50%, $
Differential Net Savings (After Tax),$
Payout Cap. Investment, Yr.
T~erma1 with Primary Heat Recovery
vs Thermal without Heat Recovery
Differential Installed Cap. Cost, $
Differential Gross Savings (Loss), $
Differential Tax @ 50%, $
Differential Net Savings (After Tax), $
Payout Cap. Investment, Yr.
VI-58
1 ,000
7,900
(300)
,.
18,500
(1,100)
9,800
300
150
150
8.7
10,000
20,000
25,400 51 .700
6,000 12,000
3,000 6,000
3,000 6,000
4.6 4.6
67,000
8,500
120,800
19,300
4,250
9,650
9,650
4,250
6. 1
5.6
38,600
12,700
71 ,100
26,300
6,350
6,350
13,150
13,150
3,8
3.5
-------
Table VI-12 (Continued)
Incinerator Capacity, scfm
Thermal with Primary and Secondary
Heat Recovery vs Thermal without
Heat Recovery
Differential Installed Cap. Cost, $
Differential Gross Savings (Loss), $
Differential Tax @ 50%, $
Differential Net Savings (After Tax),$
Payout Cap. Investment, Yr.
VI-59
1,000
20,300
(500)
10,000
80,200
15,300
7,650
7,650
5. 1
20,000
138,100
34,200
17,100
17,100
4.5
-------
Three base cases were evaluated: adsorption with
solvent recovery; adsorption with incineration and no heat
recovery; and adsorption with incineration plus heat recovery~
b .
Adsorption With Solvent Recovery
Adsorption is achieved by passing cool, particulate-
free and hydrocarbon-laden exhaust gases through a bed of
activated carbon. As required, the gases are filtered or
scrubbed to avoid fouling the adsorbent with dust and then
cooled to approximately 100°F to increase the driving force
in the transfer processo
This section of the study evaluated dual horizontal
fixed-bed adsorption systems with capacities of 1,000 SCFM and
20;000 SCFM processing 25 percent LEL concentrations of 50/50
benzene-hexane vapor mixtures (equivalent to about 0.3 percent
by volume). A working capacity equal to 50 percent of sat-
uration capacity or 001 pound adsorbate per pound adsorbent
was assumed in determining cycle times and utility require-
ments. This value is based on dynamic column tests performed
by a vendor and on the adsorption isotherms for benzene and
hexane. The dynamic column test indicates preferential qd-
sorption of hexaneo
Bed dimensions and carbon charge were .determined by
the adsorption system vendoro An emission reduction value of
90 percent across the carbon bed was used in the analysis.
Steam regeneration was held constant at 3 pounds of steam
per pound of solvento Condensation, subcooling, and decant-
ation recovery were employed in operating cost computations.
Distillation was not considered, since the purity of de-
canted solvent was assumed to be adequate for recycle to the
coating processo The process flow sequence is shown in Fig-
ure VI-150 The process consists of filtering and then cool-
ing oven gases from 375°F to 100°F, passing the stream through
one of the beds to adsorb solvent vapors, and venting the
clean gases to the atmosphereo The remaining bed is simul-
taneously regenerated with low pressure steam. The resultant
vapors are condensed at 212°F and subcooled to 100°F. Sep-
aration of the solvent-water condensate is accomplished by
decantationo
Adsorption With Incineration - No Heat
Recovery - No Solvent Recovery
The process consisted of a dual-bed adsorption
system, a booster fan, and a thermal incinerator.
c.
VI-60
-------
t
Solvent
Stream
Exhaust
100°F
I"
I
,~'1
! . '
I
! 212°F.
_l J.-
\ \
-or -
Adsorber #1
/'
Adsorber.#2
\
~
\
S te am ,
,1/.
100°F
CO~lingl
Wa te r :
----- --.--J
lOQoF
L[~nt~r
Has te
Wa te r
Wet Painted
---_.
Material in
Pre-Cooler
I
T
Condenser
c
Cool; n g'
'>-
Water
375°F
I
I
i
o
Boo$t~r
Fan
Pre-Filter
lJaking Oven
Finished
--- >--
Material Out
Figure VI-15 - Adsorption With Solvent Recovery-Flow Sheet
VI..61
-------
The process flow diagram for this scheme is pre-
sented in Figure VI-16. The stream temperatures shown are
common to both oven exhaust concentrations considered.
The stripped oven exhaust gases are vented to the atmosphere.
The solvent-laden stream from the regeneration cycle is fed
to the incinerator.
Oven exhaust rates of 1,000 SCFM, 10,000 SCFM,
and 20,000 SCFM with hexane-benzene solvent concentration
levels at 200 ppm and 25 percent LEL were incorporated
in the system analysis. At the 200 ppm concentration, th~
adsorber carbon charge was reduced to the minimum amount for
each SCFM capacity level, while incinerator capacity was re-
duced to and fixed at 500 SCFM for all oven volume capacity
levels. This was done because the 500 SCFM incinerator
appeared to be the smallest size unit readily available from
vendors and the closest to meeting process requirements. At'
the 25 percent LEL concentration, the maximum carbon charge
for each adsorber was utilized. Incinerator capacity was
based on adsorber regeneration rate plus the amount of com-
bustion air needed to maintain a 1450°F stack temperat4re.
The extremes in concentration were chosen to illustrate areas
suited to application of t~is process.
d.
Adsorption With Incineration and Heat "
Recovery
The schemes covered under this section are a con-
tinuation of the process shown in Figure VI-16 and described
under "Adsorption with Incineration - No Heat Recovery - No
Solvent Recovery". The additions of primary, and primary
with secondary, heat recovery to the thermal incinerators
are shown in Figures VI-17 and VI-1B, respectively. The
same thermal efficiencies used to evaluate heat recovery
for the incineration schemes were incorporated in thi$ phase
of the study (s ee p. 34).
Systems were again evaluated at capacities of
1,000 SCFM, 10,000 SCFM, and 20,000 SCFM and a 25 percent
LEL hexane-benzene concentration. In each case, incin-
erator size was determined by the quantity of air req~ired
to mix with the regenerated adsorber stream and maintain
the incinerator outlet temperature at 1450°F.
e.
Installed Capital Cost - Adsorption
Installed capital costs of the adsorption-solvent
recovery systems were based on one vendor's quotation, which
VI-62
-------
Exhaust
100°F
Adsorber #1
j"
/~
Adsorber #2
,Steam
I
\ / f-1
100°F
Pre-Coole
Exhaust
l450°F
~,,~Air
Thermal
Incinerator
Booster
Fan
375°F
Booster
Fan
Pre-Filter
..-_.n~-" . ---'__.'_'n--__~.-- --.
I
I
i
riniShed
Material Out
Wet Painted
---:-+-
Material in
Baking Oven
Figure VI-16 - Adsorption With Thermal Incineration -
No Heat Recovery - Flow Sheet
VI-53
Cooling
~
W ate r
+--
-------
212° F I
r I! IJ 'I A. ~
i:., IPr~ma~.Y ea ~---l"";~'-- lr 1"
: .- r x.~~anger I
Therma 1 !
i I nci nera to~? F - !._, .
Exhaust
100°F
,""J..--
,
L-'~k:-~
I ,
<
......
I
0"1
.j:::.
1450°F
Adsorber #1
J
Adsorber #2
~.
,
i
!
j~1
!.-<
I
.---J
/
/'
S te am
100°F
Dcoo1~ng
~1 a te r
~
Pre-Filter
Booster
Fan
1
J
-Baking Oven
Figure VI-17 - Adsorption With Thermal Incineration/Primary Heat Recovery -
Flow Sheet
>- Exhaust
1450°F
/
T
i
~
l'
I/'~
-------
Exhaust
1100. F Adsorber #1
1/1
I
t----1/ 1
I /
\
J Steam
Adsorber #2
1000F
1/ /1 A
212° F
Pre-Cooler
-<
.....
I
en
U1
o
1450°F
I
7
/1<
Air
Thermal
Incinerator 715
375°f
Booster
Fan
375
Primary
Heat
Exchanger
0, :~ Exhaust
455°F
Secondary
Heat Exchanger
815
Cooling
)
Water
~
Pre-F; 1 ter
~ Booster
'8 J Fan
Baking Oven
Make-Up Air Fan
Figure VI-18 - Adsorption With Thermal Incineration/Primary and
Secondary Heat Recovery
-------
itemized the equipment and installation costs for the pack-
age units. The quotation is shown in Appendix C. Additional
prices for the precooler and prefilter, not included in
vendor's package, were obtained by verbal quotations and
through use of correlations.
Installed capital costs of adsorption-thermal in-
cineration systems without and with heat recovery were based
on previously discussed quotations for the adsorbers and in-
cinerators.
Installed costs are tabulated in Table VI-13 and
shown graphically in Figure VI-19; a detailed presentation
is given on construction cost estimate sheets in Appendix C.
Adsorption with solvent recovery shows the lowest installed
price of the adsorption systems investigated at 25 percent
LEL concentration. Installed costs increase, in order, for
the systems of adsorption-incineration without heat recovery,
adsorption-incineration with primary heat recovery, and ad~
sorption-incineration with primary and secondary heat recovery.
For the adsorption-incineration system used to pro-
cess exhaust streams at the 200 ppm level, the installed ad-
sorption-solvent..recovery system and the adsorption-inciner-
ation no heat recovery system at 25 percent LEL.
When compared on the basis of equivalent heat re-
covery schemes, the cost curves of Figure VI-19 lie between
those for thermal and catalytic incineration, shown in Fig-
ure VI-10.
f.
Operating Cost - Adsorption
Operating costs for adsorption-solvent recovery
and adsorption-thermal incineration systems are shown on
the manufacturing estimate sheets in Appendix C. Values
were calculated for all systems at the 25 percent LEL level
and in addition for the adsorption-incineration, no heat
recovery system at the 200 ppm level. Costs were determined
by employing the following rates and percentages:
Operating
- 2 shifts/day, 365 days,
0.5 man hours direct
labor per shift
@ 5% Capital Cost
Maintenance
VI-66
-------
Table VI-13 - Estimated Installed Costs
of Adsorption Systems
Adsorber Capacity scfm 1 ,000 10,000 20,000
25% Lower Explosive Limit
With Solvent Recovery, $ 54,700 120,300 207,300
With Thermal Incineration/No
Heat Recovery, $ 66,800 150,700 256,600
With Thermal Incineration/
Primary Heat Recovery, % 75,800 190,300 321 ,300
With Thermal Incineration/
Primary & Secondary Heat
Recovery, $ 88,800 2~9,900 398,500
200 ppm
With Thermal Incineration/No
Heat Recovery, % 66,800 130,200 241,900
VI-67
-------
...
I
o
- 300
)(
-tI)-
....,
'"
o
u
....,
c
~ 200
c..
.....
:::I
C'"
LLJ
"'C
CIJ
-
-
Z 100
'"
c
-
400 -
o
Legend
25% Lower
(a) With
(b) With
(c) With
(d) With
Heat
Explosive Limit Concentration
Solvent Recovery
Incineration/No Heat
Incineration/Primary
Incineration/Primary
Recovery
Recovery
Heat Recover
and Second y
200 PPM Concentration
(e) With Incineratiion/
No Heat Recovery
o
-T----,
5 10 15
Adsorber Capacity, SCFM x 10-3
20
Figure VI-19 -
Estimated Installed Adsorption
(90% Combustion or Recovery of
Benzene Mixture)
VI-68
System Cost
SO/50-Hexane/
-------
Depreciation
Property Taxes
@ 10% Capital Cost
@ 1% Capital Cost
Insurance
@ 0.5% Capital Cost
@ 3% Plant level Cost
Corporate Overhead
Supervision
Payroll Overhead
@ 10% Direct Labor
@ 30% Direct labor
Plant Overhead
@ 50% Direct Labor
labor
Cooling Water
@ $4.00/hr
.. $.03/1.000 gal.
Steam
.. $1.00/1.000 gal.
.. $.015/KWH
Electricity
Activated Carbon
$.50/1b. 5 yr ser~ice
1 i fe
This list is the same as that used for the incineration eval~
uation. except that prices for additional utilities and differ~
ent operating materials are included. The values of recovered
solvents were taken at $.03 per pound for benzene and $.0375
per pound for hexane. Adiabatic system operation was assumed
in computing utility consumption; these utility calculations
are included in Appendix C.
A summary of annual operating costs is presented in
Table VI-14 and Figure VI-20 as total dollars and dollars per
1.000 SCF processed. .
These data show that adsorption with solvent recovery
is the most economical process considered at the 25 percent
LEl level. A savings with this process is apparent for the
hexane..benzene system at capacities in excess of 5.000 SCFM.
The savings rate appears to level off at $.005/1.000 SCF
processed between 15.000 and 20.000 SCFM capacity values.
Operating costs become higher for the adsorption..
incineration and adsorption..incineration-heat recovery sys..
terns in that order. This trend is generally in line with
that shown for the installed costs of these systems.
VI-69
-------
Table VI-14 - Estimated Annual Operating
Costs of Adsorption Systems
Adsorber CapacitYt scfm
2~% Lower Explosive limit
With Solvent RecoverYt $/yr
$/1000 scf Processed
With Thermal Inci~eration/No
Heat RecoverYt $/yr
$/1000 scf Processed
With Thermal Incineration/
Prim~ry Heat RecoverYt $/yr
$/1000 scf Processed
With Thermal Incineration/Primary
& Secondary Heat RecoverYt $/yr
$/1000 sef Processed
200 ppm
With Thermal Incineration/No Heat
Recovery t $/yr
$/1000 scf Processed
1 t 000
9t500
.0270
17t100
.0490
18t700
.0540
19t900
.0662
15t900
.0453
10tOOO
9t500*
.0027*
51 t 500
.0147
59t400
.0170
58t200
.0166
36t600
.0100
*Indicates a savings as opposed to operating cost
VI-70
20tOOO
30t700*
.0044*
89t300
.0128
140t700
.0150
96t600
.0138
65tOOO
.0093
-------
80 ,
'" 70
I
0
po-
)( . 60
~
11\
+a 50
1/1
0
U
en 40
'c
-
+a
/G
r.. 30
~
c.
,~
po- 20
,/G
~
.C
C
< 10
o
10
20
30
20
o
110
100
90
Legend
25%, Lower Explosive Limit concentratio~
(a) With Solvent Recovery
(b) With Incineration/No Heat Recover~c
(~) With Incin~ration/Primary Heat )
Recovery , d
(d) With Incineration/Primary
and Secondary Heat Recov-
ery
200 PPM Concentration
(~) wiih Inci~eration/
No Heat Recovery
(c)
---(d)
'"CD}
e)
Cost
Savings
Figure VI-20 - Adsorption Sys~ems Estimated Operating Cost
(90% Combustion or Recovery of SO/50 -
Hexane/Benlene Mixture)
VI-71
5 10 15
Adsorber Capacity. SCFM x 10-3
20
.05
.04
o
-a
CI).
f'$
.03 ~
....
~
co
n
o
1/1
C'+
.02 ..
~
......
-
o
, 0
o
V)
n
.01 ."
."
.,
o
n
(I)
III
1/1
(I)
-- 0 Q.
.01
.02
-------
The operating cost for the adsorption and intin-
eration with no heat recovery decreases as concentration is
reduced from 25 percent LEL to 200 ppm. This cost decrease
is caused by decreases in the size of the carbon bed, number
of regeneration cycles, and consumption of utilities. The
cost decrease is further accented as capacity increases.
g .
Control Method Evaluation-Adsorption
The process of activated carbon adsorption with
solvent recovery should be used for emission control in
plants where the recovered solvent is free enough, or can
feasibly be made free enough, of contaminants to be reused
in the process. This is the most economica1 of the control
systems evaluated.
Results of computing savings ver$US capital in-
vestment and payout for the 25 percent LEL benzene-hexane
system are summarized in Table VI-15. Review of Table VI-15
in conjunction with curve (a) Of Figure VI~20 demonstrates
that this proces~ begins to show a ~ayout ~t c~pacities
greater than 5,000 SCFM. At capacities in excess of 20,000
SCFM, this process is economically attractive regardless of
emission control considerations.
Processes employing adsorption with inciner~tion
cannot be justified on economic grounds under any conditions
that allow normal incinerator operations. Adsorption-incin-
eration without heat recovery does appear to be attractive
for disposal of very low concentration effluents «500 ppm)
that cannot 'be controlled by catalytic incineration (eg.
effluents t~at contain catalytic poisons).
Evaluation of the adsorption-incineration process
at 200 ppm was included to illustrate costs associated with
this application. For this example, the low concentration
effluent was assumed to be contaminated by a cat~lyst poison,
thus eliminating catalytic incineration. Also, at this low
concentration, direct flame incineration is not economically
feasible because of the high supplemental fuel costs. There-
fore, solvent must be collected or in effect concentrated
for destruction by thermal methods. The basis for choosing
this type of system is one of control process feasibility and
economics.
VI-72
-------
Table VI-15 - Savings vs Capital Investment and
Payout for Adsorption/Solvent Recovery -
25% Lower ~xplosive Limit
Adsorber Capacity, scfm 1 ,000 10,000 20,000
Installed Capital Cost, $ 54,700 120,300 207,300
Savings (Loss) Before Taxes, $ (9,500) 9,500 30,700
Tax @ 50%, $ 4,750 15,350
Savings (Loss) After Taxes, $ (9,460) 4,750 15,350
Payout on Investment, yr 7.2 5. 7
VI-73
-------
5.
Wet Scrubbing
a.
General
In Section B of this chapter, wet scrubbing methods
were identified as suitable for control of emissions contain-
ing organic particulates, aerosols, and sol~ble gases. To il-
lustrate the economic variables involved in application of
scrubbing techniques, two industrial applications were eva1u-
ated. The actual design capacity ranges were extrapolated to
show the effects of variation of exhaust gas flow rates on
the capital and operating costs.
Process Description - Incineration/
Scrubbing .
The control process shown on the flow she-et of
Figure VI-21 is designed to dispose of a 5 percent by weight
solution of DDT in kerosene.
b .
The operational sequence is comprised of collecting
waste liquid in:.the holding tank, pumping the liquid to the
incinerator, where plant air is provided for atomization -of
the mixture, and burning the mixture at 1600°F after blending
it with 20 percent excess combustion air supplied by the
blower. The hot gases are sent to the packed absorption
tower, where the gases are adiabatically cooled and hydrogen
chloride is atisorbed with water. The cleaned and cooled
gases then pass out through the stack to the atmosphere.
This evaluation is based on the assumption that the
waste stream will be an intermittent flow of varying mag-
nitude, which is contingent on upstream production require-
ments. A two-shift operating schedule was selected for the
emission control process; this schedule does not necessarily
coincide with that of the main process and was selected to
keep manpower requirements at a minimum. The additional
control process capacity realized by 5asing the evaluation
on this premise provides for future plant expansion and op-
erating flexibility.
A more detailed description of the installation is
provided in Appendix D. Although the calculations and economic
evaluation are based solely on the absorption of the HCl
component of the incinerator exhaust stream, trace organics,
either as particulates or soluble gases would also be col-
lected.
VI-74
-------
Proces s1.---
Water
Process Haste
DDT - 7-
Kerosen~
<:
.....
I
......
<.11
Plan t
Air
Ai r't
-1
~
H old i n gl "-7 \.J
Tank
~~
'.""-"-t-''''. ...
ta--(PI)
Feed r-~
Pump . i>\
a
Blower
~
.I
1/
Incinerato,
1
Stack
'(
Figure VI-2l - Incineration/Absorption - Process Flow Sheet (Kerosene & DDT)
Abs orbe r
-------
The major incinerator combustion products calcu-
lated for the feed and operating conditions are as follows:
Components Mol %
C02 16.5948
N2 78.2904
~2 3.4679
NO 0.0131
OH 0.0001
C12 0.0012
HC1 0.1448
H20 1.4874
99.9997
c.
Process Description Scrubbing/Absorption
This scheme is shown diagramatica11y in Figure
VI-22. The plant operates 11 reactors for the manufacture
of specialty organic chemicals. The formulations are made
batchwise. The steps consist of feeding the raw materials
to the reactor and mixing them under a heated and/or cooled
reactor jacket.
Each reactor exhaust line is provided with a valve
ahead of its connection to the scrubber system feed duct.
Since only some reactors are on stream at a given time, the
individual va1ving provides for more efficient operation.
The exhaust gases pass through a venturi scrubber,
where entrained solids are knocked out by the recirculating
water. The venturi scrubber is physically close-coupled to
the packed tower. The water-soluble vapors are removed in
the absorber. Scrubbed gases then pass out of the packed
tower through the fan, and thence to a stack.
Water is recirculated from the bottom of the
column, and split between the venturi scrubber and the
spray nozzles in the column. A portion of the water is
VI-76
-------
t
,
I
<:
......
I
'-I
'-I
1
Bleed
t lTota1 Number of
t: Reactors = 11
J.
Venturi
Scrubber L
Absorber
Pump
I
~ --------...- --------..---.
X
Reactor
(Typical)
Reactor
(Typ i ca 1)
~/
Figure VI=22 - Scrubbing/Absorption - Process Flow Sheet
~
Fan
t
I
I
>-
Stack
-------
bled off. Fresh makeup water is brought in automatically
to maintain the level in the bottom of the absorber.
Production incorporates a large number of raw
materials, including organic and inorganic compounds in
both liquid and in solid forms. Solids are charged manually
to the reactors via a chute through the manhole. Liquids
are charged to the reactors from drums and bulk $torage
tanks. Transfer from drums is accomplished with portable
pumps and hoses. Because of the significant quantities in-
volved, bulk liquids are charged by pumping directly from
storage.
The venturi scrubber is designed to reduce solids
loading of the stack discharge to less than .01 grain/cubic
foot.
A variety of gaseous compounds are exhausted to
the scrubbing system. Most of the liquid chemicals are used
in small quantities and have relatively low vapor pres~ures.
Because isopropanol exhibits the highest vapor pressure of
the raw materials used in production, it is the contaminant
upon which the absorber design is based. The design data
for the absorber are as follows:
Gas Rate - 642 lb-mol/hr (4000 SCFM)
Liquid Rate - 1400 lb-mol/hr
Mol Fraction Isopropanol in Gas Inlet - 0.1
Mol Fraction Isopropanol in Gas Outlet - 0.0073
Tower Diameter - 4 ft.
Packed Height of Tower - 6 ft.
Packing - 111 Ceramic Intalox Saddles
Height of a Transfer Unit - 1.5 ft.
Number of Transfer Units - 4
d.
Installed Cost - Incineration/Absorption
Process equipment costs were based on the 100 GPH
DOT - kerosene waste stream case illustrated in Appendix E.
Equipment costs for the 500 GPH systems were determined through
VI-78
-------
use of the six-tenths factor rule; an exception was the cap-
ital costs of the direct flame incinerator which were ob-
tained from a correlation developed from one vendor's budget
quote.
The incinerator cost correlation is shown in Appendix
D, where incinerator exhaust rates of 2000 SCFM, 10,000 SCFM,
and 20,000 SCFM are shown to be equal to waste stream flow
rates of 100 GPH, 500 GPH, and 1000 GPH.
Cost for installation of equipment was estimated
at 100 percent of purchased equipment cost.
These costs are summarized in Table VI-16 and shown
graphically in Figure VII~23; they are detailed on construction
estimate sheets in Appendix D.
Table VI-16 - Estimated Installed Costs of Incinerator/
Absorber Systems (Kerosene and DDT Waste)
System Capacity, SCFM
(Based on Absorber Feed Rate)
Installed Capital Cost, $
2000
60,000
10,000
136,000
20,000
202,000
e .
Operating Costs - Incineration/Absorption
Operating costs for the incineration/absorption
control processes are developed on manufacturing estimate
sheets in Appendix D. The values were calculated through
use of the following percentages and rates:
Operating - 2 shifts/day, 330 days/year
Maintenance @ 5% Capital .Cost
Depreciation @ 10% Capital Cost
Property Taxes @ 1% Capital Cost
Insurance @ 0.5% Capital Cost
Corporate Overhead @ 3% Plant Level Cost
Supervision @ 10% Direct Labor
VI-79
-------
250
200
...
I
o
-
)( 150
~
+oJ
VI
o
U
"'0
cv
:::100 -
°ca
+oJ
VI
c:
-
50
o
o
5
--.--,- I
10 15
System Capacity, SCFM x 10-3
(Based on Absorber Feed Rate)
I
20
25
Figure VI-23 - Estimated Installed Cost of Incinerator and
Absorber (Disposal of DOT/Kerosene Waste)
VI-80
-------
Payroll Overhead @ 30% Direct ~abor
Plant Overhead @ 50% Direct Labor
Labor @ $4.00/hr.
Process Water @ 30t/thousand gallons
Electricity @ $.015/KWH
Estimated operating costs for absorber feed rates
of 2,000 SCFM, 10,000 SCFM, and 20,000 SCFM are $ummarized
in Table VI-17 and plotted in Figure VI-24 as annual manu-
facturing cost per thousand SCF processed. These gas flows
are equivalent to liquid flow rates of 100 GPH, 500 GPH,
and 1000 GPH. The high operating costs of this system are
the result of using the absorber water on a once~through
basis. Operating cost could potentially be reduced by using
a closed-loop water system for the absorber. This would re-
quire installation of a chemical treatment system, a water
chilling system, and supporting process equipment.
Table VI-17 - Estimated Operating Costs of Incinerator/
Absorber Systems (Kerosene and DDT Waste)
System CapaGity, SCFM
(Based on Absorber Feed Rate) 2,000 10,000 20,000
Annual Operating Cost, $ 32,400 123,200 232,900
Operating Cost, $/1000 SCF .051 .033 .0314
f.
Installed Cost - Scrubbing/Absorption
Process equipment costs were based on the 4000 SCFM
unit actually designed. Equipment costs for other system
capacities were estimated using the six-tenths factor rule
defined by equation (1).
Cost of installation was estimated at 100 percent
of equipment cost. Total installed costs are summarized in
Table VI-18 and shown graphically in Figure VI-25; they are
developed on construction cost estimate sheets in Appendix D.
VI-81
-------
250
200
...
I
o
-
x
~
.. 150
+J
III
o
U
~
c:::
....
+J
10
s..
~ 100
o
-
10
~
c:::
c:::
<
50
o
.~ ..
/
/
,
I ' r--l---r-,-r--r--T - -,---,
5 10 15
System Capacity, SCFM x 10-3
o
Estimated Operating
and Absorber System
Kerosene Waste)
VI-82
Cost of Incinerator
(Disposal of DDTI
Figure IV-24 -
.05
.040
"'0
(1)
,
QI
a-t-
r ....
~
co
n
o
.03111
c-t
~
......
r
[
! VI
}.. . 02 ~
....
6
o
a
"
"'1
o
o
(1)
III
III
(1)
Q.
.01
o
20
-------
Table VI-18 - Estimated Installed Costs of Scrubber/Absorber
Systems (Organic Dust and Isopropanol)
System Capacity, SCFM
Installed Capital
Cos t, $
1 ,000
4,000
8,000
20,000
10,900
24,800
37,400
65,500
g.
Operating Costs - Scrubbing/Absorption'
Operating costs for the control system are developed
on the manufacturing estimate sheets in Appendix D. Values
were calculated through use of the following percentages and
rates.
Operating - 2 shift/day, 330 day/year
.Maintenance @ 5% capital cost
Depreciation @ 10% capital cost
Property Taxes @ 1% capital cost
Insurance @ 0.5% capital cost
Corporate Overhead @ 3% plant level cost
Supervision @ 10% direct labor
Payroll overhead @ 30% direct labor
Plant overhead @ 50% dire.ct labor
Labor @ $4.00/hr
Process Water @ 30~/thousand gallons
Electricity @ $.015/KWH
The operating costs are summarized in Table VI-19
and plotted on Figure VI-26.
VI-83
-------
100 J
~
...
I
o 80
-
x
v,
..
~
II)
o
u
'- 60
C11
.&:J
'-
o
II)
.&:J
c(
......
.'-
C11
~ 40
:3
'-
U
V)
"C
C11
-
-
10
~
~ 20
-
o
/
o
//
/
/
//
//
//
/
10
15
20
Scrubber/Absorber Capacity, SCFM x 10-3
Figure IV-25 - Estimated Installed Cost of Scrubber and
Absorber (Organic Dust and Isopropanol)
VI-84
-------
Table VI-19 - Estimated Operating Costs of Scrubber/Absorber
Systems
Operating Cost.
$/1000 SCF
1 .000 4.000 8.000 20,000
5,900 12,100 19,500 402200
.019 .0095 .0077 .0063
System Capacity, SCFM
Annu~l Operating
Cos t, $
D.
GROWTH PROJECTIONS
1.
General
Any consideration of the regulation and control of
air pollution must take into account the projected growth
(or diminution) of the various segments of the national
economy that contribute to or are affected by the pollution
problem, Such projections or forecasts are necessitated by
several factors: (1) The sources, quantities, and types of
pollutant emissions are a function of the patterns of social,
economic, and technological change, and are thus time-dependent
parameters; (2) the often undesirable. but neverthe1ess real,
time lag between recognition of a problem and application of
a solution requires that some estimate of the f~ture be made;
(3) The economic and'technological implications of the ap-
plication of controls often require incremental reduction of
emissions over a period of years, in order to achieve a sat-
isfactory level of air quality. Additionally, p9pulation and
industrial growth may require increasing levels of control to
maintain air quality.
The hazards of forecasting technological and economic
changes are well d~c~mented. The acceleration of such changes,
as seen in recent years will undoubtedly continue; thus even
short-range forecasts are highly uncertain. If we are to at-
tain future air quality standards, however, general growth
trends must be recognized and allowed for in planning recom-
mendations. The growth aspects and short-range projections
of several areas of the U.S. economy are presented briefly
in this section.
VI-85
-------
...
I 80
o
-
; 70i
.. I
+.I
VI 60
o
u
0')
c: 50
'r-
+.I
~
L-
~ 40
o
-
~ 30
::I
c:
c:
<
100
90
20
10
o
o
. .
.-*
----
. ~~.
~
~~y
--_._-~----
.-._._~.- ..-----
,- I
5 10 15
Scrubber/Absorber Capacity, SCFM x 10-3
20
Figure VI-26 - Estimated Operating Cost of Scrubber and
Absorber (Organic Dust and Isopropanol)
VI-86
.04
o
"'C
(1)
~
QI
. 03 ~
::I
I.Q
n
o
VI
c-t'
~
......
.02 -
o
o
o
VI
n
."
-a
~
o
n
.01 (1)
VI
VI
(1)
Q.
o
-------
2.
Energy Resources
A review of energy resources with projections for
the future was made by Morrison (1968). This review presents
a simplified energy model for the United States and, for var-
ious assumptions relating to technological changes and in-
novations, projects energy consumption to the years 1980 and
2000. The medium-term forecasts, along with historical data~
are shown in Table VI-20 for the major sources of energy.
In spite of significant changes in nuclear energy
sources, the increase in total use of fossil fuels from 1970
to 1980 ;s projected to be nearly 30 percent. Assuming a
proportionate increase in organic emissions from fossil fuel
use, if uncontrolled, the emissions from combustion of coal,
oil, and gas by 1980 would reach about 0.3 x 106 tons/year.
In addition to the effect of increased nuclear en~
ergy sources, such factors as the increased efforts in coal
gasification and oil shale usage could significantly affect
these projections. A recent article (Washington Energy Memo,
March 29, 1971) cited a Department of Interior projection of
a possible 36 plants producing fuel gas from coal, producing
3 trillion cubic feet of gas a year. This is estimated at
10 percent of the predicted gas consumption.
3.
Petroleum Industry
Martin (1970) reviewed the near future of the re-
fining industry and presented some medium-range growth pro-
jections. Petroleum industry experts predict a total U.S.
crude petroleum throughput of about double the present
capacity by the mid 1980's. Construction engineering firms
specializing in refinery processes project somewhat lower
estimates of 14 to 17 x 106 bbl/day refinery throughput by
1985, from the current level of about 11 x 106 bbl/day.
Martin predicts that the increased throughput will
be handled with no net increase, or perhaps even a decrease,
in the total number of plants. Thus, new plants will be of
much higher capacity and many present small units will be
phased out. Although some coastal plants might attain as
much as 500,000 bbl/day capacity (currently only five re-
fineries yield more than 300,000 bbl/day), most of the in-
dustry believes that capacity of new plants will be in the
range of 200,000 to 300,000 bbl/day .
VI-87
-------
Table VI-20 - United States Total Gross Consumption of Energy Resoufces by Major Sources*
1947-65 and Projections to 1970, 1975, and 1980
(Trillion Btu)
Bituminous Natural Total Total gross
and Gas Fossil Nuclear Energy
Year Anthracite Li gnite Dry2 Petroleum3 Fuels Hydropower4 Powe r Inputs
Historical year:
1947 1,224.2 14,599.7 4,518.4 11,367.0 31,709.3 1,459.0 33,168.3
1.948 1,275.1 13,621.6 5,032.6 12,558.0 32,487.3 1,507.0 33,994.3
1949 957.6 11,673.1 5,288.5 12,120.0 30,039.2 1,565.0 31,604.2
1950 1,013.5 11,900.1 6,150.0 13,489.0 32,552.6 1,601.0 34,153.6
1951 939.8 12,285.3 7,247.6 14,848.0 35,320.7 1,592.0 36,912.7
1952 896.6 10,971.4 7,760.4 15,334.0 34,962.4 1,614.0 36,576.4
1953 711.2 11 , 1 82 . 1 8,156.0 16,098.0 36 , 147. 3 1,550.0 37,697.3
1954 683.2 9,512.2 8,547.6 16,138.0 34,881.0 1,479.0 36,360.0
1955 599.4 11,104.0 9,232.0 17,524.0 38,459.4 1,497.0 39,956.4
1956 609.6 11,340.8 9,834.4 18,624.0 40,408.8 1,598.0 42,006.8
1957 528.3 10,838.1 10,416.2 18,570.0 40,352.6 1,568.0 1.2 41,921.8
1958 482.6 9,607.6 10,995.2 19,214.0 40,299.4 1,740.0 1.5 42,040.9
1959 477.5 9,595.9 11,990.3 19,747.0 41,810.7 1,695.0 2.2 43,507.9
1960 447.0 9,967.2 12,698.7 20,067.0 43,179.9 1,775.0 5.5 44.960.4
1961 403.8 9,809.4 13,228.0 20,487.0 43,928.2 1,628.0 17.0 45,573.2
< 1962 381.0 10,159.7 14,120.8 21,267.0 45,928.5 1 ,780.0 23.0 47,731.5
......
I 1963 361.0 10,722.0 14,843.0 21,950.0 47,876.0 1,740.0 33.0 49,640.0
OJ 1964 365.8 11,295.0 15,647.5 22,385.8 49,694.1 1,873.0 34.0 51,601.0
OJ 1965 preliminary 327.7 12,030.0 16,136.1 23,209.3 51,703.1 2,050.0 38.0 53,791.1
Projected years:
1970 309.0 14,251.0 19,374.0 27,27"5.0 61,209.0 2,193.0 874.0 64,276.0
1975 280.0 16,865.0 22,360.0 31,875.0 71,380.0 2,422.0 1,803.0 75,605.0
1980 250.0 19,290.0 25,455.0 35,978.0 80,973.0 3,026.0 4,076.0 88,075.0
lGross energy is that contained in all types of commercial energy at the time it is incorporated
in the economy, whether the energy is produced domestically or imported. Gross energy comprises
inputs of.primary fuels (or. their derivatives), and outputs of hydropower and nuclear power con-
verted to theoretical energy inputs. Gross energy includes the energy used for the production,
processing, and transportation of energy proper.
2Excludes natural gas liquids.
3petroleum products including still gas, liquefied refinery gas, and natural gas liquids.
4Represents .projections of outputs of hydropower and nuclear power converted to theoretical
energy inputs at projected rates of pounds of coal per kilowatt hour at central electric
stations. Excludes inputs for power generated by nonutility plants, which are i.nc1uded
within the other consuming sectors.
Sources:
Compiled by Division of Mineral Economics, Bureau of Mines, U.S. Department of the
Interior, supplemented by estimates for conventional power plants and for nuclear
power from the Federal Power Commission and the Atomic Energy Commission.
*From Morrison and Readling (1968)
-------
A recent news article (Washington Energy Memo,
April 5, 1971) reports an API estimate that proved recoverable
crude reserves in North America totaled 39 billion barrels as
of December 31, 1970. This represents an increase over the
previous year of 9L4 billion. This increase, however, is due
entirely to Alaskan dri1ting 'activity; the lower 48 states
show a net decrease. .
4.
Solvents
Growth trends in solvent production and usage are
upward but erratic. The impact of pollution control by such
legislation as Rule 66 in Los Angeles, Regulation 3 in San
Francisco, and other similar restrictions has already been
noticeable in certain segments of the solvent industry. Re-
formulation of mixed solvents for specific applications has
been one of the approaches favored by the industry for com-
pliance with control legislation. Reformulation has not.
significantly changed the total solvent consumption, the 'net
result being a decrease in the non-exempt or reactive solvents
and a concomitant increase in other non-reactive components.
The use of water-base or completely solventless coat-
ings and inks has been accelerated by pollution control reg-
ulations. For those applications, such as architectural coat-
ing$, which are not amenable to recovery or c6ntrol, reform-
ulation and increased usage of water-base paints has been
the only recourse.
Recent news releases by .the printing industry in-
dicate significant progress in development of completely sol-
ventless inks for some applications.
Production of paint, varnish, and lacquer has in-
creased from .398 million gallons in 1966 to about 445 million
gallons in 1970, or about 12 percent. Similar, or even
slightly increased growth may be projected for the near future.
However, pressure for reduction of emissions may be. expected
to generally reduce the total usage of organic solvents in
coatings.
~
The net result of industrial and population growth
coupled with increased emission control efforts will be a
continued overall growth rate of perhaps.4 or 5 percent per
year; considerable rearrangement of the consumpt10n of 1n-
d1v1dual solvents will take place, however, reflecting the
trend toward the usage of less photochemically reactive sol-
vent blends. .
VI-89
-------
5 .
Solid Wastes
As discussed in Chapter III, Section C, the current
status of the generation and disposal modes of solid wa~tes
is only partially defined. Thus, general forecasts, even
over short time spans, are highly speculative and subjective.
For municipal and domestic wastes, where there is some gen-
eral agreement as to current status, the growth is e~pected
to follow patterns of population change and demographic dis-
tribution. Added to this population-dependent growth is a
variable factor reflecting an apparently accelerating rate
of per capita waste generation.
Fogel et al (1970), in projections of pollution
control costs, assumed this latter portio~ 9f the waste gen-
eration growth to be represented by a 3 percent per year in-
crease. Black et al (1968), in summarizing the results of
the National Solid Wastes Survey, extrapolated the per capita
increase in waste production according to recent increases.
in consumption of durable and non-durable goods, or about 4
to 6 percent per year. .
A.D. Little (1969) reported projections of solid
waste generation through the year 2000. By combinatibn of
their population growth rate and refuse quantity grow~h rate,
the total generated refuse for 1980 is projected as about
1.2 times the 1970 level.
The shifting patterns of land usage and the pres~
sures of all types of pollution control programs will greatly
affect the growth or diminution of the various disposal modes.
Open burning of domestic and municipal wastes is now banned
in many areas, and the number of communities applying such'
bans is expected to increase rapidly during ~he next 2 to
5 years. Open burning of industrial, commercial, and agri-
cultural wastes is being increasingly controlled by local
variance rulings. Meteorologically scheduled open burning
of bulk wastes not suitably handled by other disposal methods
is increasing, with a concomitant reduction of the local ad.
verse effects, if not of the total pollution load.
6 .
Total Hydrocarbon Emission
Estimates of the growth rates of the total organic
emissions may be obtained from comparison of emission in-
ventories and estimates made by pollution control agencies.
As an example, data are presented for the San Francisco Bay
Area (BAAPCD, 1969) in which potential or uncontrolled emis-
VI-gO
-------
sions are compared with actual or partly controlled
These data show an estimated reduction of potential
trolled stationary source organic emissions of 1041
a reduction of nearly 53 percent.
The Bay Area estimates also show that potential or-
ganic emissions from stationary sources would have grown
from 1922 tons/day in 1955 to 1982 tons/day in 1969, for an
average uncontrolled compounded growth rate of about 3 percent
per year. This would forecast a doubling of organic emissions
from stationary sources in about 25 years, if uncontrolled.
These data show a net reduction from 1955 to 1969 of ~1500 ton/
day of organic emissions from all sources, both mobile and
stationary. Allowing for estimated growth of uncontrolled
emissions, this shows the achievement of a reduction of po-
tential 1969 emissions of ~53no ton/day, or a 67 percent re-
duction.
emissions.
or uncon-
tons/day;
The impact of such an uncontrolled growth rate may
be visualized by the application of a rollback analysis to
ambient air quality standards for hydrocarbons. The usual
expression for a rollback calculation assumes a 1:1 relation-
ship between the reduction in emissions and the improvement
in ambient air quality and was expressed in mathematical
form by Jensen (1971) as
~B
A -
- G U-B , where
A = allowable emission fraction
Q = ambient air quality standard
U = present air quality
B = background
G = growth factor
For the purposes of this calculation, a current
air quality level (U) may be obtained from recent CAMP data
(Larsen, 1970) for total hydrocarbon concentration and the
assumption that the non-methane fraction is about 40 percent
of the total concentration. Thus, a typical urban non- .
~ethane hydrocarboh concentration (3 hr average) would be
about 2 ppm (as C). The recommended national standard for
ambient non-methane hydrocarbon (Federal Register, 1971) is
0.24 ppm (as C).
The background level (B) for non-methane hydro-
carbon is difficult to define readily, but the value of
0.5 ppm (USDHEW, 1970) is not unreasonable.
VI-91
-------
Substituting these values into
expression, and applying a growth factor
pected doubling of emissions in about 25
allowable emission fraction is
the above rollback
of 2, for the ex-
years, the estimated
A - 0.24 - 0.05 = 0.049
- 2(2 - 0.05)
This is equivalent to a reduction in all emissions over the
25 year time span of 95.1 percent.
Even if a zero growth rate (G = 1) in emissions is
projected', the necessary reduction of emissions over the
time span of interest would be about 90 percent. Thus, the
achievement of current air quality standards for hydrocarbons
over any reasonable time span is seen to be a formidable task.
Lacking any basis for estimation of the amount and
effectiveness of applied control that will be put into prac-
tice over the next few years, the only forecast that can be
made is one based on uncontrolled emission growth rates.
From the estimates made by San Francisco Bay Areas, APCD,
the overall uncontrolled growth rate for stationary source
organic emissions may be assumed to be about 3 percent per
year. At this rate, the currently estimated total hydro-
carbon emissions from stationary sources of 25 to 26 million
tons per year would reach nearly 35 million tons per year
by 1980.
In view of the assumptions, and resultant uncer-
tainty, involved in making this overall extrapolation, more
detailed estimates of future hydrocarbon emissions is unwar-
ranted.
VI-92
-------
CHAPTER VI
REFERENCES
B.
CONTROL TECHNOLOGY
Accomazzo, M.A., and K. Nobe, Ind. Eng. Chern., Prod. Res.
Dev. i, 425-30 (1965).
Am. Ind. Hyg. Assoc., l\ir Pollution Manual Part II, Control
Equipment, Detroit (1968). .
Anon., "Plant Licks Solvent Emission Problems", Environ.
Science and Technol. ! (2), 107 (1970).
Anon., "Catalytic Oxidation Controls Emissions", Environ.
Science and Technol. 1 (11), 1159 (1969).
Barnebey, H.L., "Elimination of Organic Vapors", Proc. 52nd
Annual Meeting APCA, Los Angeles, Calif., June 21-26,
1959.
Barnebey, H. L., II Remova 1 of Exhaus t Odors from Sol vent Ex-
traction Operation by Activated Charcoal Adsorption", J.
Air Pollution Control Assoc. li (9), 422 (Sept. 1965).
Barry, H.M., Chern. Eng. &.I, 105-7 (1960).
Benforado, D.M., "Control of Air Pollutants in the Finishing
Industries", Ind. Finishing ii (7),24-7 (june 1968)a.
Benforado, D.M., and J.J. Waitkus, "Fume Control in Wire
Enameling by Direct-Flame Incineration", J. Air Pollution
Control Assoc. 18 (1), 24-6 (1968)b.
Benforado, D.M., "Air Pollution Control by Dfrect Flame In-
. ," c i n era t ion i nth e P a i n tIn d u s try ", J. P a i n t T e c h n 01. 39
(508), 265-6 (May 1967).
Benz, E.C., "Elimination of Atmospheric Pollution Caused b,y
Phthalic and Maleic Anhydride", Fumi and Polveri (Milan)
1. 275-8 (Sept. 1963).
Boubel, R.W., "Benzo(a)pyrene Production During Controlled
Combustion", J. Air Pollution Control Assoc. l1., 553-7
(Nov. 1963).
VI-93
-------
REFERENCES, CHAPTER VI (continued)
Brewer, G~L., "Fume Incineration", Chern. Eng. Ii (22),160-5
(1968).
Chatfield, H.E., in Danielson, Air Pollution Engineering Man-
ual,' pp. 681-8, 999-AP-40 (1967).
Manual, U.S. Public
Duerden, C., Public Health Inspector (London), 74, 21-9
(Oct. 1965). -
Elliott, J.H., N. Kayne, and M.F. Leduc, "Experimental Program
for the Control of Organic Emissions From Protective Coat-
ing Operations", Report No.7, Los Angeles APCD (1961).
Fawcett, R.L., "Air Pollution Potential of Phthalic Anhydride
Manufacture", J. Air Poll. Control Assoc. £Q., 461-5 (1970).
GCA Technology Division, Control Techniques for Polycyclic
Organic Matter Emissions, U.S. Department of Health, Ed-
ucation & Welfare, Draft Copy, August 1970.
Grant, R.J., M. Manes, and S.B. Smith, "Adsorption of Normal
Paraffins and Sulfur Compounds on Activated Carbon", A.loCh.E.
J. 8 (3), 403 (l96 2) .
Hein, G.M., "Odor Control by Catalytic and High Temperature
Oxidation", Annals of the New York Academy of Sciences,
ill, 656 (l964).
Johnson, W.C., and L.L. Kempe, "Experiences in Controlling
Atmospheric Pollution Encountered in Pharamceutical Manu-
facturing Processes", J. Air Pollution Control Assoc. 9,
32 - 5, 3 7 - 41 ( May 1 9 5 9 ) . -
Kanter, C.V., et al, "Control of Organic Emissions From Sur-
face Coating Operations", Proc. 52nd Annual Meeting, APCA,
Los Angeles, Calif. (June 1959).
Leduc, M.F., in Danielson, Air Pollution Engineering Manual,
pp. 192-201, 999-AP-40 Ci967).
Vt..94
-------
REFERENCES, CHAPTER VI (continued)
Lee, D.R., "How to Design Charcoal Adsorption Systems for
Solvent Vapor Recovery", Heating, Piping and Air Con-
ditioning, 1£ (5), 80-3, (4), 76-9 (1970).
Matti a, M. M., II Pro c e s s for So 1 vent Po 11 uti on Con tr 0 1" ,
Chem. Eng. Prog. M (12),74-79 (1970).
Miller, M.R., and Wi11hoyte, "Catalyst Support Systems for
Incineration of Hydrocarbon Solvents Emissions", J. Air
Pollution Control Assoc. II (12), 791-5 (1967).
Miller, P.D. et al, liThe Design of Smokeless Non-Luminous
Flares", Proce. Amer. Petrol. Inst. 11! (3), 276-81 (1958).
Mills, J.L. et al, "Design of Afterburners for Varnish
Cookers", Proc. 52nd Annual Meeting APCA, Los Angeles,
June 21-26, 1959.
M 0 r g an, D. F., II Use 0 fAd s 0 r p t ion to C.o n t r 0 1 Sol ve n t Va p 0 r s
Released from Industrial Processes", Proc. 1st Tech.
Meeting, West Coast Section APCA, Los Angeles, 1957.
North American Combustion Handbook, North American Manu-
facturing Company, Cleveland (1952).
Robell, A.J., E.V. Ballow, and F.G. Borgardt, "Basic Studies
of Gas-Solid Interactions", Lockheed Missiles and Space
Co., Report No. 6-75-65-22 (1965).
Sandomirsky, A.G. et al, "Fume Control in Rubber Processing
by Direct Flame Incineration", J. Air Pollution Control
Assoc. li, 673-6 (1966).
Stenburg, R.L., "Control of Atmospheric Emissions from
Paint and Varnish Manufacturing Operations ", Robert A.
Taft Sanitary Engineering Center, 1958.
Stern, A.C., Air Pollution, Vol. III, "Sources of Air Pol-
lution and Their Control II , Academic Press, New Vor~, N.V.
(1968).
Sullivan, J.L. et al, "An Evaluation of Cataltyic and Direct
Fired Afterburners for Coffee and Chicory Roasting Odors ",
J. Air Pollution Control Assoc. !i, 583-6 (Dec. 1965).
VI-95
-------
REFERENCES, CHAPTER VI (continued)
Tho m aid e s, L., II Why Cat a 1 y tic I n c; n era t ion? II , Pollution E n -
gineering 1 (3), .32-3 (1971).
TI-3 Petroleum Committee, IIContr01 of Atmospheric Emissions
from Petroleum Storage Tanksll, J. Air Pollution Control
Assoc. n (5),260-8 (1971).
u.S. Department of Health, Education and Welfare, IIContr01
Techniques for Hydrocarbon and Organic Solvent Emissions
from Stationary Sources II, NAPCA Publ. AP-68 (1970).
Waid, D.E., IIIncineration of Organic Materials by Direct
Gas Flame for Air Pollution Control II, Am. Ind. Hygiene
Ass 0 c. J., May / J un e 1 9 69 .
Wallach, A., liS orne Date and Observations on Combustion of
Gaseous Effluents from Baked Lithograph Coatings II, J. Air
Pollution Control Assoc. 1£, 109-10 (March 1962).
Werner, K.D., Chern. Eng. 75 (24), 179-84 (1968).
VI-96
-------
CHAPTER VI
REFERENCES
C.
CONTROL ALTERNATIVES
A.D. Little, Inc., "Systems Study of Air Pollution from
Municipal Incineration", PB 192378, March 1970.
Aerojet-General Corp., "A Systems Study of Solid Waste Man-
agement in the Fresno Area", U.S. Public Health Service
Publication No. 1959 (1969).
American Society of Civil Engineers, Civil Engineer~Man-
ual of Practice, No. 39, Sanitary Landfill (1959 .
Combustion Engineering, Inc., "Technical-Economic Study of
Solid Waste Disposal Needs and Practices", U.S. Public
Health Service Publication No. 1886 (1969).
Engdahl, R.B., Solid Waste Processing, Public Health Service
Publication No. 1856 (1969).
Kupchick, G., liThe Economics of Composting Municipal Refusell,
Public Works 21., 127"(l966) (Abstract only).
Larson, E.C., and -H.E.' Sipple, "Los Angeles Rule 66 and Ex-
empt Solvents", J. Paint Technol. l2. (508), 258 (l967).
Technolo
Solid Waste Report, Vol. 1, No.4, November 30, 1970.
U.S. Bureau of Mines, "Bureau of Mines Research and Accom-
plishments in Utilization of Solid Wastes", Information
Circular 8460, March 1970.
U.S. Department of Health, Education and Welfare, liThe Nat-
ional Solid Waste Survey, An Interim Report", Environ-
mental Health Service (1968).
VI-gj]
-------
CHAPTER VI
REFERENCES
D.
ECONOMIC ASPECTS
API Bulletin 2513, Evaporation Loss in the Petroleum Industry-
Causes & Control, American Petroleum Institute, 1959:
API Bulletin 2517, Evapor~ion Loss from Floating-Roof Tanks,
American Petroleum Institute, New York, 1962.
API Bulletin 2518, Evaporation Loss From Fixed-Roof Tanks,
American Petroleum Institute, New York, 1962.
API Bulletin 2519, Use of Internal Floating Covers for Fixed-
Roof Tanks to Reduce Evaporation Loss, American Petroleum
Institute, New York, 1962.
Bay Area Air Pollution Control District, Source Inventory of
MrPol1ution Emissions, San Francisco Bay Area, 1969.
Brewer, G.L., "Fume Incineration", Chern. Eng., n (22), 160-5
(1968).
Danielson, J.A., Air Pollution Co~neering Manual,
U.S. Public Health Service Pub ication No. 999-AP-4crT1967).
Duprey, R.L., Com i1ation of Air Pollution Emission Factors,
U.S. Public Health Service Publication, PB 190245 1968.
Dunmyer, J.C. et a1, "Tommorow's Gasoline-Best Route", Hydro-
carbon Processing, May 1971. .
Hein, G.M., Odor Control by Catalytic and High-Temperature
Oxidation, Annals New York Academy of Sciences, 116, 656
(1964).
Lee, D.R., "How to Design Charcoal Adsorption SY$tems For
Solvent Vapor Recovery", Heating, Piping & Air Conditioning,
42 (5), 80-3; (4), 76-9 (1970).
VI;:;g8
-------
REFERENCES, CHAPTER VI (continued)
Lunche, R.G., A. Stein, C.J. Seymour, and R.L. Weimer, "Emis-
sions From Organic Solvent Usage in Los Angeles County",
Chern. Eng. Prog., g (8), 375 (1957).
Mattia, M.M., "Process for Solvent Pollution Control", Chern.
Eng; Prog., M (12),74-9 (1970).
Thomaides, L., "Why Catalytic Incineration?", Pollution En-
gineering, 1 (3), 32-3 (1971).
Waid, D.E., Incineration of Organic Materials by Direct Gas
Flame for Air Pollution Control, May/June, 1969.
VI-99
-------
CHAPTER VI
REFERENCES
E.
GROWTH PROJECTIONS
A.D. Little, Systems , Study of Air Pollution From Municipal
Inci nerati on, PB 192 378 ['f'9'75l.
Bay Area Air Pollution Control District, "Source Inventory
of Air Pollutant Emissions, 1969".
Federal Register ~, 8186 (1971).
Jensen, D., "From Air Quality Criteria to Control Regu1ations",
~~g~~~~~:l~~S of Air Pollution, Society of Automotive
Larsen, R.r., "Re1ating Air Pollutant Effects to Concentration
and Control II , J. Air Poll. Control Assoc. £Q. (4),214-25
(1970).
Martin, R., liThe Refining Industry of the Future:
Petro/Chern Engineer 11 (3), 16-31 (1970).
1985",
Morrison, W.E., and C.L. Readling, "An Energy Model for the
United States", U.S. Bureau of Mines Information Circular
8384 (1968).
U,Scr~~~~~~m~~; ~:d~~~~;~~n~~~c~~~~~ ~~~1~e~~~r~p_~~r(~~~~1ty
Black, R.S., A.S. Muhich, A.J. K1ee, H.L. Hickman and R.D.
Vaughan, liThe National Solid Wastes Survey, An Interim
Report", Paper Presented at the 1968 Annual Meeting of
the Institute for Solid Wastes of the American Public
Works Association, October 1968.
VI-10D
-------
VIr.
A.
INTRODUCTION
RESEARCH AND DEVELOPMENT RECOMMENDATIONS
Current awareness of the problems associated with
pollution of our environment has resulted in the development
of federal, state, and local programs to identify, assess,
and control emissions that have adverse effects on our.
society and economy. This present study was one of many
such programs sponsored by the Environmental Protection
Agency having the ultimate goal of providing problem-solving
research and development programs to aid in future planning
for governmental action.
The R&D program developed from this study was based
on results of the review of the sources, emissions, adverse
effects, and control of hydrocarbon pollutants. Evaluation
of these results in terms of priorities, economic impact, and
data needs resulted in the formulation of detailed R&D recom-
mendations.
These recommendations include objectives, priorities,
technical effort, and estimated cost for needed $tudies in the
following areas:
1)
Definition and quantification of the ad-
verse effects of hydrocarbon pollutants,
particularly those related to human health.
2)
Characterization of hydrocarbon emission
streams, relative to chemical composition
and engineering control design parameters.
Development of cost/benefit data for total
systems analysis.
3)
4)
Correction of deficiencies in control
technology, emission characterization and
monitoring, and feasible alternatives
for reduction of total environmental
pollution.
B.
R&D PRIORITIES AND NEEDS
1.
Application of Existing Technology
Significant reduction in the emissions of atmos-
pherically reactive hydrocarbons and, concomitantly,
diminution of the impact of the most widely recognized
VII-l
-------
adverse effect, photochemical smog, can be achieved by appli-
cation of existing control techniques and alternatives. This
fact has been amply proven by the experience of local pollution
control agencies in the major smog areas of California.
For the most immediate short-range benefit, imple-
mentation of available technology must be given high priority.
Control or restriction of the use of photochemically reactive
solvents, limitations or bans on open burning of wastes, and
control of evaporative losses during gasoline storage and
marketing, when coupled with the current implementation of
mobile emission source controls, would result in significant
benefits to that majority of the populace now centered in
major metropolitan areas.
The impact of existing controls on the industrial
sector of the economy has not proved to be the severe hard-
ship expected by many. Experiences with such controls on
the West Coast have not significantly affected overall econo-
mic growth, although in certain instances adjustments and
modifications did pose some difficulty. Our limited investi-
gation of general-case control systems showed that econo~ic
payouts or benefits can result,in some cases, from the
application of controls. These findings are supported by
numerous industrial case histories reported in the literature.
Thus, although it does not directly relate to our
R&D planning objectives, the highest priority in'terms OT
direct benefit to society must be the rapid implementation of
existing control technology.
2 .
Updating Existing Literature on Control Technology
Relevant to the short-range benefits achievable by
application of existing technology is the need for updating
much of the published information on design and application
of control systems. Although much helpful information can be
obtained through data and services provided by individual ex-
perts, and from engineering and equi~ment firms specializing
in pollution control~ the pUblished literature is woefully
inadequate for the i~gineer faced with a specific problem.
High priority thus should be placed on updating
and revising the existing texts and manuals to provide mean-
ingful basic design and engineering data for practicing in-
dustrial pollution control engineers and for control agency
personnel responsible for the specification and approval of
control instaliations. The Air Pollution Engineering Manual
(Danielson, 1967) made an excellent approach toward providing
such information and, thus, could serve as a model for revision
VIr~2
-------
and updating. The revised manual should be in a format suit-
able for periodic addenda provided by a continuing program of
data compilation and evaluation.
3.
Pollutants and Effects
Although evidence relating hydrocarbons to photo-
chemical smog is sufficient to require corrective action,
studies of the health-effects of chronic exposure to specific
organic compounds and mixtures thereof should be con9ucted.
The results of such studies, combined with the fragmented
data available in the literature, would provide the infor-
mation necessary to adequately protect society from unknown
or suspected but unproven hazards arising from long-term
exposure to organic species not eliminated by smog control
action.
To support such health studies, more detailed
characterization of specific source emissions of organic com-
pounds is required. For even the most thoroughly studied
emission sources, published data are barely adequate to allow
estimates of total organic emissions; detailed characteri-
zation of organic emissions as to chemical composition and
concentration ranges is-almost entirely lacking. To accomplish
these studies, improvements in analytical techniques and equip-
ment would be highly desirable. Particular emphasis should
be placed on the development of dfscriminatory continuous
monitoring techniques, since grab sample and spot check tests
of emission sources can yield highly erroneous and conflicting
results.
Recent investigations of the photochemical reactivity
of hydrocarbons that were previously considered unreactive have
shown that smog manifestations can be produced in laboratory
experiment with even completely saturated paraffinic hydro-
carbons. Investigations of this type, over wide ranges of
experimental conditions, should be continued. These studies,
as well as investigations of synergistic interactions, would
further our knowledge of the critical control steps required
for eliminatiun of photochemical smog.
4.
Control Technology Improvements
Recognizable deficiencies in current control tech-
nologies must be corrected if long-range plants to eliminate
hydrocarbon pollution are to be realized. Significant areas
for-study are:
VII...3
-------
1 )
Afterburners - studies of both direct-fired
and catalytic systems to better define control
effectiveness and to improve combustion effi-
ciencies.
2)
Wet Scrubbers - data correlation qnd appropriate
experimental studies to correlate performance
data and improve design bases.
Adsorption - selected experimental studies to
better system design bases and to widen the
range of applicability.
3)
4)
Storage Tank Controls - studies to update data
base for estimation of losses and to investi-
gate usage of novel materials as sealants.
Dry Particulate Collectors - collection and
analyses of performance data on the removal of
polycyclic aromatic hydrocarbons.
5)
5 .
Water and Solid Waste Pollution
Existing technology for the control of hydrocarbon
air pollutant emissions often simply transfers pollutants from
one facet of the environment to another; thus control of air
pollutjon can create problems of water pollution or of solid
waste disposal. Programs directed toward control of hydrocarbon
air pollutants should be integrated with the overall environ-
mental pollution control objectives. Studies of air pollutant
control devices that create potential water or solid waste
problems should be implemented. Particular areas for study
are the clarification, neutralization, and recycle or disposal
of wet scrubber liquors and the disposal of organic particulates
which may result in potential carcinogenic hazard.
The need for interdisciplinary correlation of con-
trol programs is pointed up by the significant amounts of
hydrocarbon air pollutants emitted by the indjscriminate dis-
posal of solid or liquid wastes. Control planning for such
waste disposal must consider the potential side effects of
even trace concentrations of organic emissions to the at-
mosphere.
C.
ECONOMIC IMPACT OF CONTROL PLANNING
To achieve a realistic balance between our presently
undesirable lack of emission control and the extreme of cQm-
plete prohibition at any cost, the economic impact of control
VII~4
-------
planning on the various sectors of society must be considered.
Certain aspects of the socio-economic relationships were pre-
sented in the discussion of applications of existing controls
(VII. B.1) and in Chapter V on systems modelling.
Determination of the cost of control for specific
cases presents only a portion of the total picture. To com-
plete the picture, the benefits to "society resu1ti~g from
control application should be assessed. Unfortunately, the
lack of data quantifying the costs of adverse effects (or,
conversely, the benefits of control) effectively prevents
this assessment.
Attempts to estimate total costs of control for all
emissions or even for any significant segment or source
category are 'felt to be fruitless.. The wide range of sizes
and types of individual sources and control devices within
any broad category, coupled with the lack of data on extant
controls, prohibits meaningful estimates.
To even qualitatively assess, in the light of pre-
sent knowledge, the impact of cuntro1s on the economy, re-
course must be made to the experiences of those few regional
areas having a history of app1ied"controls. The best in-
dication of economic cost and benefit related to control of
hydrocarbon emission may be derived from studies of the Los
Angeles and Bay Area control districts. Significant re-
ductions in potential or uncontrolled emissions of hydrocarbons
from stationary sources have been achieved with little or no
effect on economic growth of the areas.
For example, data from the Bay Area APCD indicate
that potential organic emissions from stationary sources in
1969 were reduced by about 53 percent by the app1icatinn of
controls. The estimated dollar cost of applied control of
stationary source organic emissions in the Bay Area was about
$200 million, equivalent to nearly $50 per person living in
the area. The cost per annual ton of organic emission pre-
vented by control was estimated from these data to be about
$525. The cost of complete reduction of emissions was esti-
mated by BAAPCD to be double the current total cost, or about
$400 mi 11 i on .
Economic indicators and projections for the San
Francisco AQCR (Office of Business Economics, 1970) show
that total personal income and earnings from industrial em-
ployment for the region both increased about 16 percent
during the. p~riod 1967 to 1970. Similar figures for the
total U.S. are 15 and 14 percent, respectively.
VII..5
-------
For the Los Angeles AQCR, which is under similarly
stringent control, increases in personal income and indust-
~ia1 earnings for the 1967-1970 period were both nearly 17
percent.
D.
R&D PROGRAMS
Specific recommended programs of research and de-
ve10pmentare based on the needs and priorities identified
in this study.
1 .
Emissions and Sources
A series of programs is needed to further define
and characterize selected emissions in terms of their chem-
ical composition, source-dependent engineering control para-
meters,and atmospheric distribution.
a. Program D-1a - Characterization of Selected
Emissions and Sources
Objective - To provide detailed chemical and en-
gineering data characterizing emissions from selected ~ources.
Discussion - Emissions and sources requiring more
detailed study include all types of waste combustion, fuel
combustion,and industrial processes such as coke manufacture,
rubber and plastics processing, synth~tic organic chemical
processing,and surface coatings application and processing.
Results of these studies would provide input for adverse
effects studies and control systems design.
Priority - High
Duration - 3 years
Effort - 12 man-years, $600,000.
b.
Program D-lb- Atmospheric Modeling
Objective - To improve procedures for relating
emissions to ambient air quality.
Discussion - Existing models and mathematical tech-
niques for estimation of ambient air distribution of emissions
are limited in their ability to treat real atmospheric con-
ditions and interactions. Because of the apparent key role
Vfl~6.
-------
of hydrocarbons in smog formationp realistic modeling to
account for atmospheric reactions must be developed. Theo-
retical studies should be implemented by additioITal at-
mospheric monitoring studies to better define the relation-
ships of hydrocarbons and photochemical oxidants.
Priority - Medium to high
Duration - 5 years
Effort - 10 man-yearsp $500pOOO.
c.
Program D-lc - Analytical Development
Objective - To develop improved instruments and
techniques for continuous monitoring of both emission sources
and ambient air.
Discussion - Current monitoring techniques are
limited in their ability to provide continuous, real time
determination of specific organic species or chemical
groups. Laboratory development and field testing studies
are needed to provide improved methods of monitoring.
Priority - High to medium
Duration - 3 years
Effort - 6 man-years; $300,000.
2.
Adverse Effects
A series of programs is needed to supplement exist-
ing limited data on the adverse effects of hydrocarbon pollu-
tants. Particular emphasis should be placed on human health
effects and on assessment of economic costs to all sectors of
society.
a.
Program D-2a - Animal Toxicity Studies
Objective - To provide data on both short-term and
chronic exposures for assessment of health effects of hydro-
carbon pollutants.
Discussion - Limited data on health effects of low
concentrations of hydrocarbons should be supplemented by
laboratory toxicity studies of animals. Of particular im-
portance are investigation of the effects of exposures to
VII-7
.,~
-------
polycyclic aromatic hydrocarbons, oxygenated hydroca~bons ~
and photochemical smog products.
Priority - Hig~
Duration - 3 years
Effort - 6 man-years; $350,000.
b.
Program D-2b - Epidemiological Studies
Objective - To provide statistically validated
correlations of hydrocarbon pollutants and human health.
Discussion - Field data on urban and non-urban
population samples should be compiled and analyzed. Areas
and population samples selected should provide ranges and
extremes of exposure. Particular emphasis should be placed
on the inter-relationships of exposures with predisposing
health conditions and synergistic factors.
Priority - Medium to low
Duration - 5 years
Effort - 3 man-years; $175,000.
c.
Program D-2c - Smog Studies
Objective - To compile and analyze data on human
response to both actual and synthetic smog atmospheres.
Discussion - Smog chamber studies, coupled with
field data compilation, are needed to establish human re-
sponses to exposures over wider ranges of time and con-
centration. Synthetic atmospheres should be developed to
correspond with the range of actual exposures in polluted
atmospheres.
Priority - Medium
Duration - 3 years
Effort - 6 man-years; $300,000.
VII.8
-------
d.
Program D-2d - Economic Costs
Objective - To compile and analyze field data on
economic costs of hydrocarbon pollutants.
Discussion - To completely analyze cause and ef-
fect relationships, better data' are needed on the economic
costs of the effects of hydrocarbon pollutants. Costs of
health effects and of damage to vegetation and materi.als
should be developed and analyzed. Field survey teams would
compile data for input to a cost - benefit model analyses.
Priority - Medium
Duration - 3 years
Effort - 6 man-years; $300,000.
3.
Control Systems
A series of programs is recommended to improve the
design bases, (assessment 'of ton'tro1 effectiveness and imple-
mentation of control systems.
a. ~ Program D-3a - Control Effectiveness of
Direct Flame Afterb~rners
Objective - To compile and analyze data on the
control effectiveness or control efficiency of currently
u t 1'1i zed s y s te m s .
Discussion - To implement more effective use ~f
direct flame afterburners, studies should be made, under
typical field co~ditions, of the control effectiveness or
efficiency of tn~se systems. Studies would include analysis
of influent and effluent contaminant concentrations as a
function of the combustion parameters (flow rate, residence
time, temperature and fuel and air rates). Detailed cost
data should be compiled and analyzed.
Pri ori ty - Hi gh
Duration - 1.5 years
Effort - 4 man-years; $200,000.
VII;;.9
-------
b. Program D-3b - Control Effectiveness of
Catalytic Afterburners
Objective - Same as D-3a, but for catalytic sys-
terns
Discussion - Same as D-3a, but for catalytic sys-
terns
Priority - High
Duration - 1.5 years
Effort - 4 man-years; $200,000.
c.
Program D-3c - Improved Afterburner Designs
Objective - To improve the combustion efficiency of
both flame and catalytic afterburners.
Discussion - This program would consist of several
phases of study to improve current designs of afterburners
systems: a theoretical and laboratory study of combustion
chamber configuration, utilizing flow modeling techniques
developed for combustion furnace design; laboratory and pilot
scale studies of novel designs under simulated operating con-
ditions; field testing of best design configurations.
Priortty - High:
Duration - 2 years
Effort - 5 man-years; $350,000.
d.
Program D-3d - Adsorption Systems
Objective - To improve existing design base fQr
carbon adsorption systems.
Discussion - Existing design bases for activated
carbon adsorption system should be updated, particularly by
studies of adsorption and desorption of over a wider range
of operating conditions. Studies to collect and compile
such operating data should also be complemented by engineer-
ing cost data.
Priority - Medium
VII~tO
-------
Duration - 1.5 years
Effort - 3 man-years; $150,000.
e.
Program D-3e - Wet Scrubber Systems
Objective - to develop improved design base for wet
scrubbers systems.
Discussion - A three-phase program is recommended
to provide, ultimately, a sound engineering design base for
selection and operation of wet scrubbers for particulate re-
moval. The study phases are 1) further compilation and cor-
relation of existing test data in context of correlation with
agglomeration index; 2) theoretical model development; 3) ex-
perimental performance studies to confirm correlation and
theory. Results should be complemented by detailed cost
studies of selected devices.
Priority - High
Duration
- Phase 1)
Phase 2)
Phase 3)
Total
0.75 year
1 year
~ars
3.75 years
Effort - Phase 1)
Phase 2)
Phase 3)
Total
1 man-year; 50,000
0.5 man-year; $25,OQO
6 man-years; j400,000
$475,000
f.
Program D-3f - Dry Collectors
Objective - To analyze performance of dry particu-
late collectors with respect to removal of condensed-phase
polycyclic a~omatic hydrocarbons (PAH).
Discussion - PAH and other condensible hydrocarbons
are known to be associated with particulate emissions. The
performance of dry particulate collectors should be analyzed
for effectiveness in PAH removal, particularly from combustion
sources.
Priority - Medium
Duration - 1.5 years
VII;" 11
-------
Effort - 3 man-years; $150s000.
g .
Program D-3g - Storage Tank Controls
Objective - To update data base on control effective-
ness and to investigate new or novel materials for roof seals.
Discussion - The existing data base for control ef-
fectiveness of floating roof controls on storage tanks is con-
siderably outdated. A cooperative industry-government program
would provide input to update loss calculations. A second phase
of the study would experimentally investigate the potential
of novel materials as improved roof seals.
Priority - Medium to low
Duration - Phase 1) 3 years
Rhase 2) 1 year
Effort - Phase 1) 1.5 man-years; $75,000
Phase 2) 1 man-year; $75s000
h .
Program D-3h - Vapor Recovery Systems
Objective - to provide engineering and cost data
on vapor recovery systems
Discussion - Existing data on vapor recovery sys-
tems are quite limited. Compilation and analysis of en-
gineering performance and cost data are required.
Priority - Medium to low
Duration - 0.75 year
Effort - 1 man-year; $50,000.
i .
Program D-31 - Engineering Manual
Objective - To revise and update control engineer-
ing manual
Discussion - A program is recommended to revise and
update the Air Pollution Engineering Manuals with provision
for a continuing program.of revision to maintain currency of
data.
VII~12
-------
Priority - High
Duration -5 years
Effort - 2.5 man-years; $250,000
4.
Water and Solid Waste Studies
The inter-relationship of air pollution with water
and solid 'waste pollution requires careful analysis of air
pollution control devices to avoid creating other pollution
problems.
ao
Program D-4a - Wet Scrubber Waste Disposal
. Objective - To prevent problems of water pollution
~~r solid waste disposal from operation of wet scrubber de-
vices for air pollution control.
Discussion - The suggested program should review
current problems of waste disposal from wet scrubbing oper-
ations and investigate new or improved techniques for neutral-
ization, recycling or disposal of scrubber liquors. Cost
analyses of disposal or treatment should be applied to con-
trol system cost evaluation.
Priority - Medium
Duration - 2 years
Effort - 5 man-years; $250,000.
b 0
Program D-4b - Dry Collection Waste Disposal
Objective - Similar to D-4a
Discussion - Specific emphasis should be placed on
determination of potential hazards from disposal of wastes
having high PAH content.
Priority - Medium
Duration - 2 years
Effort - 5 man-years; $250,000.
VII~13
-------
e.
Milestone Schedule
Figure VII-l provides a suggested milestone ~hedule
for the 5-year, $5 million R&D program described in th~)revious
section (VII-D).
'.
VII:14
-------
r
c:;
.....
.....
I.'
--'
(J'1
Figure VII-l - Milestone Schedule for 5 Year, $5 Million R&D Program
Cost Prllnr;:lm V~;:Irf:.
Program Des cri pti on ($1,000) 1 2 3 4 5
0-1 Emission and Sources
ao Characterization of Selected
Emissions and Sources 600 ....
-
b 0 Atmospheric Modeling 500 ~ -:.--
co Analytical Development 300 I..
- ...,.
0-2 Adverse E ffe cts
a. Animal Toxi ci ty Studies 350 L-
./
b. Epidemiological Studies 175 --
....
c. Smog Studies 300 '- ......
... ....
d. Economic Cost Studies 300 -
...
0-3 Control Sys terns
a. Control Effectiveness of [; -
Direct flame Afterburners 200 ...
-
b. Control Effectiveness of
Cata1ytic Afterburner$ 2{)0 v ~
.' ....
c. Improved Afterburner Design
Studies 350 -
-.,
-------
Figure VII-l (continued)
c:::
-
....
..oJ
.....
Cost Proqram Yea rs
Program Description ($1,000) 1 2 3 4 5
do Adsorption System Design
Studies 150 ....
,
eo Wet Scrubber Systems
Studies
Phase 1 50
Phase 2 25 '
I' ,
Phase 3 400 I" ,.
f. Dry Collector Studies 150 /0 ....
" ,
go Storage Tank Controls
Phase 1 75 -
,.
Phase 2 75 ...
,
h 0 Vapor Recovery Sys terns 50 ~ ~
...... ,.
i. Engineering Manual 250 ...
" ,.
0-4 Water & Solid Waste Studies
a 0 Wet Scrubber Waste Disposa] 250 ' ...
.... -
b 0 Dry Collector Waste Disposal 250 ~ ,
"
TOT At S 5,000 910 1,310 1,360 860 560
m
------- |