EPA-600/5-74-007
MARCH 1974
Socioeconomic Environmental Studies Series
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and appli-
cation of environmental technology. Elimination of traditional grouping
was consciously planned to foster technology transfer and a maximum inter-
face in related fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the SOCIOECONOMIC ENVIRONMENTAL STUDIES
series. This series includes research that will assist EPA in implement-
ing its environmental protection responsibilities. This includes examining
alternative approaches to environmental protection; supporting social and
economic research; identifying new pollution control needs and alternate
control strategies; and estimating direct social, physical, and economic
cost impacts of environmental pollution.
EPA REVIEW NOTICE
This report has been reviewed by the Office of Research and Development,
EPA, and approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the Environmental
Protection Agency, nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.
-------
EPA-600/5-74-007
March 1974
FEASIBILITY OF EMISSION STANDARDS BASED ON PARTICLE SIZE
By
L. J. Shannon
P.G. Gorman
W. Park
Contract No. 68-01-0428
Program Element 1HA091
Project Officers
Mr. Howard Bergman
Mr. Paul Gerhardt •
Washington Environmental Research Center
Washington, B.C. 20460
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $2.50
-------
ABSTRACT
The technical and economic feasibility of participate emission standards
based on particle size was assessed in this program. Specific attention
was focused on standards to regulate the emission of fine particulates—
particulates below 2 u in size.
The program was divided into four major areas of effort:
1. Analysis of approaches for regulating fine particle emissions from
stationary sources.
2. Definition of technological and economic requirements necessary for
implementation of emission standards.
3. Identification of benefits that would accrue if control procedures for
fine particulates can be implemented.
4. Assessment of overall feasibility of implementation of fine particle
emission standards.
The analysis of the implications of emission standards based on particle
size identified some deficiencies in control technology that will limit
the type of standards that can be proposed and implemented in the near
future. The economic impact of fine particulate control was found to
vary substantially from industry to industry. Estimates of costs as-
sociated with the control of fine particulates varied from less than 1.0%
up to 207o of the value of the product.
The lack of data on the damages associated with fine particulate pollution
hampered efforts at meaningful cost/benefit analysis. The limited cost/
benefit analysis performed indicated that the optimum control efficiency
for fine particulates varied with population density.
In general, emission standards based on particle size were judged to be
both technically and economically feasible. The most realistic approach
would be to tailor the emission standard to specific sources of fine
ii
-------
particulate pollutants. Tailoring of standards permits a greater degree
of flexibility in an overall control plan for fine participates, and
acknowledges the differences in the importance and difficulty of control
of individual sources.
iii
-------
CONTENTS
Abstract ^
List of Figures vii
List of Tables ix
Acknowledgments x±±±
Sections
I Summary ]_
Goal(s) for Program to Control Fine Particulates 1
Approaches for Regulation of Fine Particulate Emissions 2
Technical Implications of Emission Standards 3
Economic Impact of Emission Standards 5
Benefit/Cost Relationships for Fine Particulate Control 5
Overall Feasibility of Emission Standards Based on
Particulate Size 9
II Recommendations for Future Work 11
Introduction 11
Source Sampling and Monitoring Methods 11
Fine Particle Emission Rates 11
Control Technology Evaluation 12
Control Technology Development 12
Analysis of Relationships Between Source Emissions and
Ambient Air Quality 12
Economic Impact Analysis 13
Cost/Benefit Analysis 13
' :.TV
III Introduction 14
IV Role of Fine Particles in Air Pollution 16
Introduction 16
Effects of Fine Particulates on Human Health 20
Deposition, Retention, and Clearance Processes in the
Respiratory System 2i
Clearance of Particulate Matter from the Respiratory
System 24
lexicological Studies of Atmospheric Particulate Matter 24
Epidemiological Studies of Atmospheric Particulate Matter 26
Modification of Properties of the Atmosphere 27
Visibility 29
Solar Radiation 33
Weather Modification 34
iv
-------
CONTENTS (Continued)
V Sources of Fine Particulate Emissions 36
Nature of the Particulate Pollution Problem 36
Principal Sources of Fine Particulates 40
Summary of Fine Particle Emissions 41
Priority List for Sources of Fine Particle Emissions 46
VI Approaches for Regulating Fine Particle Emissions 53
Introduction 53
Methods for Regulating Fine Particle Emissions 56
Basis for Emission Standards for Fine Particulates 58
Plume Opacity Standards 60
Minimum Collection Efficiency for Fine Particulates 62
Mass-Emission Regulations 63
Emission Standards Selected for Evaluation 66
VII Technological Implications of Fine Particle Emissions Regulations 68
Introduction 68
Requirements for Control Equipment Efficiency 68
Capability of Control Technology 71
Capability of Existing Control Equipment 71
Emerging and New Control Technology 74
Summary of Status of Control Technology 77
Requirements for Compliance Testing and Monitoring 78
Compliance Monitoring Methods for Plume Opacity 82
Compliance Monitoring Methods for Mass Emissions 85
Summary of Recommended Compliance Monitoring Methods 92
VIII Economic Impact of Fine Particle Emission Standards 93
Introduction 93
Determination of Cost vs Efficiency Relationships for Control
Devices 94
Methodology for Determining Costs Required for Compliance
with Fine Particle Emission Standards 98
Development of Model Plants for Important Industrial
Sources of Fine Particle Emissions 98
Determination of Control Device Performance Requirements 99
Costs to Model Plant and Industry for Compliance with
Specific Fine Particle Emission Standards 101
Estimated Reductions in Fine Particle Emissions 130
Economic Impact of Fine Particle Emission Standards on
Selected Sources 133
Example of Calculation of Economic Impact 135
Control Costs in Selected Industries 138
-------
CONTENTS (Continued)
Page
IX Benefit/Cost Relationships for Fine Particulate Control 147
Introduction 147
Determination of Economic Damage Attributable to a Specific
Source 147
Computing the Impact Area 149
Control Costs 151
Economic Damage 151
Total Economic Damages 153
Cost/Benefit Relationships 157
X Overall Feasibility of Emission Standards Based on Particulate
Size 166
Introduction 166
Feasibility of Specific Types of Standards 166
Opacity Regulations 167
Regulation Based on Best Installed Control System 168
Mass Emission Regulations 168
XI References 170
XII Glossary of Terms 178
XIII Appendices 180
Appendix A - Control Technology for Fine Particles 180
Appendix B - Example of Procedure for Determining Control 20,7
Costs for Compliance with Fine Particle 207
Emission Standards
Appendix C - Plume Opacity 213
vi
-------
FIGURES
No.
1 Relationship between particle mean residence time and
particle size 18
2 Effects of particulate air pollution in the community as
related to particle size
3 Fraction of particles deposited in the three respiratory
tract compartments as a function of particle diameter . . 23
4 Effect of irritants in major bronchi 28
5 Effect of irritants in terminal bronchioles 28
6 Effect of irritants in alveoli 28
7 Meteorological visibility vs particle size, ferric
sulfate aerosol 31
^
8 Meteorological visibility vs particle size, flyash
aerosol 32
9 plume opacity as a function of particle diameter and dust
loading 61
10 Representative emission standards based on potential emis-
sion rate 65
11 Extrapolated fractional efficiency of control devices ... 73
12 Forces operating on aerosol particles 75
13 Recommended manual method for monitoring compliance of
sources with fine particle emission regulations 87
vii
-------
FIGURES (Concluded)
No. Page
14 Semicontinuous method for monitoring fine particle
emissions 88
15 Schematic diagram of monitoring system utilizing quartz
crystal microbalance 90
16 Annualized cost for operation of high-voltage electro-
static precipitators 95
17 Installed costs for control equipment 96
18 Annualized costs for control equipment 97
19 Interrelationships among factors affecting the economics
of a firm 134
20 Structure for microeconomic impact analysis 136
21 Simplified hemispherical pollutant dispersion model . . . .150
22 Economic damage resulting from continuous exposure to
75 ug/m3 fine particulate concentration 156
23 Fine particulate control costs for 400-MW coal-fired
electric generating plant 159
24 Total annual economic costs of fine particulate control
at various population densities . . .., .164
25 Optimum fine particle control level at various population
densities 165
viii
-------
TABLES
No. Page
Estimates of Fine Particle Emissions as a Function of
Emission Standard
Effect of Control Criteria on Control Costs in
Selected Industries
Total Annual Cost of Achieving BICD-Level Control of
Fine Particulates in Selected Industries
4 Summary of Fine Particulate Control Costs at BICD-
Level Control ..................... 8
5 Classification of Particulate Pollutants ......... 17
6 Major Industrial Sources of Particulate Pollution .... 37
>- ->
7 ,,-, Nonindustrial Sources of Particulate Pollution ...... 38
8 All iMa^or Sources of Particulate Pollution ........ 39
' ;» i
9 Fine Particle Emissions from Industrial Sources ..... 42
• ,'*"' f v<"
10 Fine Particle Emissions from Mobile Sources ....... 45
11 Profile of the Characteristics of Particulate Pollutants
Emitted by Various Industrial Sources ......... 47
12 Priority List for Sources of Fine Particle Emissions. . . 52
13 . Estimated Concentrations of Los Angeles Aerosol
Particles by Source .................. 55
ix
-------
TABLES (Continued)
No. Page
14 Control Device Efficiency Required to Achieve Various
Plume Opacity Levels .................. 70
15 Measurement and Monitoring Methods for Fine Particu-
lates .......................... 79
16 Control Equipment Costs for Coal-Fired Electric Utility
Plants ......................... 102
17 Incremental Annualized Costs for Coal-Fired Electric
Utility Plants to Meet Fine Particle Emission Stand-
ards .......................... 103
18 Control Equipment Costs for Sinter Machine (Windbox) . . . 105
19 Incremental Annualized Costs for Sinter Machines (Iron
and Steel) to Meet Fine Particle Emission Standards. . . 106
20 Control Equipment Costs for Basic Oxygen Furnaces (Iron
and Steel Plants) .................... 107
21 Incremental Annualized Costs for Basic Oxygen Furnaces
to Meet Fine Particle Emission Standards ........
24
22 Control Equipment Costs for Electric Arc Furnaces (Iron
and Steel Plants) ................ .... 110
23 Incremental Annualized Costs for Electric Arc Furnaces
to Meet Fine Particle Emission Standards ........ Ill
Control Equipment Costs for Cement Plant Rotary Kilns. . . 112
25 Incremental Annualized Costs for Cement Kilns to Meet
Fine Particle Emission Standards 113
26 Control Equipment Costs for Hot-Mix Asphalt Plant Rotary
Dryers 115
27 Incremental Annualized Costs for Hot-Mix Asphalt Plant
Rotary Dryers to' Meet Fine Particle Emission Stand-
ards 116
-------
TABLES (Continued)
No. Page
28 Control Equipment Costs for Ferroalloy Furnaces
(Closed Electric Furnace) ................ 117
29 Control Equipment Costs for Ferroalloy Furnaces
(Hooded Open Electric Furnace) .............
30 Control Equipment Costs for Ferroalloy Furnaces
(Unhooded Open Furnace) ................. H9
31 Control Equipment Costs for Rotary Lime Kilns ....... 121
32 Incremental Annualized Costs for Rotary Lime Kilns to
Meet Fine Particle Emission Standards .......... 122
33 Control Equipment Costs for Municipal Incinerators .... 123
34 Incremental Annualized Costs for Municipal Incinerators
to Meet Fine Particle Emission Standards ........ 124
35 Control Equipment Costs for Iron Foundry Cupolas ..... 125
36 Incremental Annualized Costs for Iron Foundry Cupolas to
Meet Fine Particle Emission Standards .......... 126
37 Control Equipment Costs for Primary Aluminum-Electrolytic
Cells .......... ................ 128
38 Control Equipment Costs for Primary Copper Plants ..... 129
39 Estimates of Fine Particle Emissions as a Function of
Emission Standard .................... 131
40 Projections of Fine Particle Emissions from Industrial
Sources .........................
41 Example of Economic Impact of Air Pollution Control
Requirements on Electric Power Generating Costs ..... 137
42 Typical Financial Characteristics of 400-MW Electric
Generating Plant .................... 139
xi
-------
TABLES (Concluded)
No. Page
43 Estimated Control Costs for 400-MW Electric Generating
Plant 140
44 Effect of Control Costs on Costs of Electric Power. . . . 140
45 Effect of Control Criteria on Control Costs in Selected
Industries 142
46 Total Annual Cost of Achieving BICD-Level Control of
Fine Particulates in Selected Industries 143
47 Summary of Fine Particulate Control Costs at BICD-
Level Control 145
48 Unit Value of Output from Selected Industries 146
49 Estimated Damages Resulting from Fine Particle Pollu-
tion at an Ambient Concentration of 75 ug/m3 155
50 Comparison of Estimated Damages Resulting from Fine
Particle Pollution at Ambient Concentrations of 75
and 60 ug/m3- 158
51 Extent of Impact of Uncontrolled Fine Particle Emissions
for 400-MW Coal-Fired Electric Plant 160
52 Economic Costs of Control and Damage Caused by Fine
Particulate Emissions at Various Population Densi-
ties and Control Efficiencies
53 Economic Costs of Control and Damage Caused by Fine
Particulate Emissions 162
xii
-------
ACKNOWLEDGMENTS
This report was prepared by Midwest Research Institue, 425 Volker
Boulevard, Kansas City, Missouri 64110, under Contract No. 68-01-0428
with the Office of Research and Development of the Environmental
Protection Agency. Mr. Howard Bergman and Mr. Paul Gerhardt were
EPA's Project Monitors.
The program was centered in MRI's Physical Sciences Division, Dr. H.M.
Hubbard, Director. Dr. A.E. Vandegrift, Assistant Division Director,
served as Program Manager. Dr. L. J. Shannon, Head, Environmental
Systems Section, served as the Principal Investigator for MRI. Other
MRI staff members who contributed significantly to the program were
Mr. P.G. Gorman, Mr. W. Park, and Mr. T. Weast.
xiii
-------
SECTION I
SUMMARY
Particles smaller than 2 u (i.e., fine particulates) are a major factor
in air pollution, affecting both human health and the physical properties
of the atmosphere. While the overall impact of fine particulate pollu-
tants on man's environment is not precisely quantified, evidence continues
to point to many negative features of fine particulate pollution. Real-
izing the rapidly developing interest in all aspects of the fine par-
ticulate pollution problem, the Environmental Protection Agency through
its Standards Research Branch contracted with Midwest Research Institute
to perform a study to assess the technical and economic feasibility of
particulate emission standards based on particle size.
The effort in the current program was directed to an analysis of the tech-
nical and economic feasibility of various general bases for emission
standards for fine particulates. In practice, the diversity of both the
sources and particulate pollutants emitted from various sources will un-
doubtedly rule out the use of a general emission standard for all sources.
However, while the current study is broad in scope, the results have in-
dicated the general feasibility of fine particulate control and the areas
for future research and development activity have been pinpointed.
GOAL(S) FOR PROGRAM TO CONTROL FINE PARTICULATES
Definition of the size of particulates that must be controlled is the
first step in formulating a program to reduce the emission of fine par-
ticulates. If the aerosol burden in the urban atmosphere is to be re-
duced, a premium must be placed on the collection of particles smaller
than 5 u. If improved visibility is the goal of fine particulate control,
attention must be focused on particle collection in the 0.1-1.0 u range.
Concern for adverse health effects should direct attention to the control
of the 0.01-7 u size range and to the potentially hazardous constituents
(e.g., lead, cadmium, mercury) of the effluent stream. Because protection
of human health is the main justification for launching a program to con-
trol fine particulates, the particle size range below 7 u is the general
size range of interest. Currently available control equipment for par-
ticulate pollutants exhibit good collection efficiency down to a particle
-------
size of approximately 2 u. Below this particle size, collection efficiency
for most equipment decreases rapidly. Therefore, particles less than 2 u
in diameter are of most interest for future control activities for partic-
ulate pollutants.
APPROACHES FOR REGULATION OF FINE PARTICULATE EMISSIONS
Approaches that could be used to regulate fine particulate emissions in-
clude: (1) emission standards; (2) tax incentives; (3) process modifica-
tions; (4) substitution of ingredients or fuels; and (5) cessation of
processing operations that emit fine particles. The last alternative is
deemed untenable except in very special situations. The other alterna-
tives are viable, although methods (3) and (4) are limited in the extent
to which they can be used. A precedent exists for utilizing emission
standards as the main method of control of fine particulate emissions in
the near term. Current programs for the control of particulate pollutants
rely on emission standards, and minimum disruption in control agency
activities would occur if similar approaches were used for fine particu-
lates. However, the unavailability of high efficiency control equipment
and adequate source compliance monitoring methods and/or instrumentation
may limit the utility of emission standards—especially if fine particu-
lates must be controlled to a very high degree. Alternative regulatory
strategies based on tax incentives might be attractive in that event.
Effluent charges or fees for the discharge of fine particulates is one
definite possibility. The acquisition of the information necessary to
structure an effective and equitable effluent tax will require extensive
characterization of both sources and pollutants. While additional data
are being obtained to further the formulation of intermediate and long-
term strategies for the control of fine particulates, near-term activities
could be initiated using emission standards. lirlii W
Emission standards for the control of fine particulate emissions from in-
dustrial sources may be based on factors such as: (1) limitation of the
concentration of fine particulates in an effluent stream; (2) specifica-
tion of the required collection efficiency of control equipment in given
particle size ranges; (3) reduction in plume opacity; and (4) limitations
of the mass emission rate of fine particulates. Our analysis (see
Chapter 4) of the alternative routes for emission standards for fine par-
ticulates led to the following conclusions;
1. The use of emission standards based on plume opacity is a practical
means for reducing fine particulate emissions in the near term,
2. An emission standard based on the requirement of the installation of
the best installed technology on all sources in a specific source cate-
gory could be implemented in either the near- or intermediate-term, and
-------
3. A mass emission regulation based on the potential emission rate con-
cept is a very attractive approach for the long term.
Emission standards based on plume opacity would require the least amount
of additional data acquisition and they would readily interface with exist-
ing programs for control of particulate pollutants. An emission standard
requiring the installation of the best installed technology would represent
the attainment of the current practical limit of control equipment per-
formance. Regulations based on the potential emission rate concept would
establish emission limits which vary with the pollution potential of the
source, e.g., limitation of mass rate of fine particulate emissions in
pounds per hour as a function of potential emission rate, also pounds per
hour. A regulation of this type could be tailored to the control of
specific sources or specific constituents of a particulate effluent
stream. The existing data base on potential emission rates of fine par-
ticulates is clearly inadequate for the intelligent selection of allow-
able emission rates. An extensive data acquisition program will be re-
quired to obtain the needed emission rate data before viable standards
based on potential emission rate can be developed.
TECHNICAL IMPLICATIONS OF EMISSION STANDARDS
An analysis of the technical implications of emission standards requiring
107o and 5% plume opacity indicated that currently available control equip-
ment could achieve the required overall collection efficiency. Trans-
missometers are also available to permit instrumental evaluation of plume
opacity. An emission standard based on installation of the best installed
technology obviously stretches current technology to near its limit. Only
manual methods are currently available for monitoring compliance of sources
with a regulation of this type.
Substantial reductions in fine particulate emissions could be achieved by
either opacity standards or the installation of best installed technology.
Table 1 presents estimates of emission reductions that might be achieved.
The estimated reductions in emissions shown in Table 1 assume that the
control equipment is in operation 10070 of the time that a source is operat-
ing. This is seldom the case, but accurate data are generally not avail-
able on the reliability of control equipment. The reliability of control
equipment will have to be improved if efforts to control fine particulates
are to be successful.
Another area in which improvement will be necessary in order to achieve
control of fine particulates is the capture efficiency of hooding systems
that are associated with many installations of control devices. Capture
-------
Table 1. ESTIMATES OF FINE PARTICLE EMISSIONS AS A FUNCTION OF EMISSION STANDARD
Estimated Fine Particle Emissions
Uncontrolled
I.
II.
III.
IV.
V.
VI.
VII.
VIII.
Source
Coal combustion
Iron and steel
A. Sinter machines
B. Basic oxygen furnace
C. Electric arc furnace
Cement plants, rotary kilns
Asphalt plants, dryers
Ferroalloy plants
A. Closed electric furnace
B. Hooded open electric
furnace
C. Unhooded open electric
furnace
Lime plants, rotary kilns
Municipal incinerators
Iron foundry, cupola
Lb/Hr
796,000
7,100
400,000
27,400
243,000
3,260,000
92,000
129,500
72,000
75,800
15,000
22,900
Ton/Year
3,184,000
28,400
1,600,000
109,600
972,000
1,956,000
368,000
518,000
288,000
303,200
37,500
22,900
Present
Lb/Hr
243,000
1,400
43,600
3,600
44,300
257,000
11,300
84,100
60,800
22,000
13,100
13,100
Ton/Year
972,000
,
5,600
174,400
14,400
177,200
154,200
45,200
336,400
243,200
88,000
32,800
13,100
BICD Standard^./
Lb/Hr
15,000
64
2,560
645
2,000
25,600
11,300
15,800
28,800
1,200
490
356
Ton/Year
60,000
256
10,240
2,580
8,000
15,360
45,200
63,200
115,200
4,800
1,225
356
10% Opacity Standard 5% Opacity Standard
Lb/Hr
5,050
722
1,540
1,036
25,200
200,000
NC£/
NC
NC
10,500
3,400
5,900
Ton/Year Lb/Hr
20,200 2,510
2,888 354
6,160 800
4,144 683
100,800 12,200
120,000 167,000
NC NC
NC NC
NC NC
42,000 4,800
8,500 1,900
5,900 4,000
Ton/Year
10,040
1,416
3,200
2,732
48,800
100,200
NC
NC
NC
19,200
4,750
4,000
a/ Not calculated.
b_l BICD-Best installed control' device (not necessarily the highest efficiency device available, but rather the best that is generally being installed
at present time).
-------
efficiency of hood systems that are necessary for many sources is fre-
quently estimated to be less than 95%. If the control device efficiency
is in excess of 997o, the particulate matter which escapes the hooding
system becomes very significant.
ECONOMIC IMPACT OF EMISSION STANDARDS
Requirements for a given level of fine particulate control from stationary
sources may have significant economic impacts on an industry, with the ex-
tent of the impact depending on the characteristics of the industrial pro-
cesses involved and on the specific control equipment requirements.
The additional production costs incurred as a result of control practices
will include both operating costs and capital charges, again depending on
the type of control equipment installed. These costs have been estimated
for a variety of industries, with capital charges computed at four dif-
ferent rates to reflect differences in equipment life, depreciation policies,
and accounting procedures. In this study estimates of control costs and
the resultant impact on production costs were determined for emission
standards requiring the achievement of 107e or 57, plume opacity and the
installation of best demonstrated technology. Tables 2 to 4 summarize
the estimated costs for selected industrial sources. The estimates shown
in Table 4 indicate a widely varying impact among the industry studied.
In most cases, installation of the best installed technology will have
relatively minor effects on overall production costs, generally amounting
to less than 1.0% of the value of the product produced.
BENEFIT/COST RELATIONSHIPS FOR FINE PARTICULATE CONTROL
To assure the optimum utilization of economic resources, the costs of par-
ticulate control must be weighed against the benefits realized by having
reduced the fine particulate emissions. The economic damages resulting
from uncontrolled and controlled fine particulate emissions were investi-
gated, although insufficient data are available to draw any definite con-
clusions. When adequate data become available through additional research,
it will be possible to define the optimum control strategy for any given
set of conditions.
For illustrative purposes, the economic losses attributable to interactions
between fine particulates at ambient concentrations of 75 and 60 ug/m^,
and humans, animals, vegetation, materials and aesthetics, were estimated.
The estimates were based on the area over which exposure takes place, and
on the economic values exposed to damage.
-------
Table 2. EFFECT OF CONTROL CRITERIA ON CONTROL COSTS IN SELECTED INDUSTRIES
Source
Coal-fired electric plant
Municipal Incinerator
Cement plant (rotary kiln)
Asphalt plant (rotary dryers)
Iron and steel
(a) Basic oxygen furnace
(b) Electric arc furnace
(c) Sintering (windbox)
Lime plant (rotary kiln)
Iron foundry cupola
Control Fine Particle
Model Plant Total Annual Control Cost ($1,000)
Annual Production at Specified Capital Charge Rate
Criteria Control Efficiency (%) Rate 0.15
BICD
107. Opacity
57. Opacity
BICD
10% Opacity
57. Opacity
BICD
107. Opacity
57. Opacity
BICD
107. Opacity
57. Opacity
BICD
107. Opacity
57. Opacity
BICD
107. Opacity
57. Opacity
BICD
107. Opacity
57. Opacity
BICD
107. Opacity
57. Opacity
BICD
10% Opacity
5% Opacity
98.13
99.36
99.68
96.71
77.41
87.50
99.17
89.58
94.94
99.21
92.86
94.05
99.70
99.82
99.91
97.64
93.54
97.24
99.10
87.76
94.05
98.43
78.04
90.59
98.46
74.29
82.86
2.4 x 109 kwh 241.5
320.0
378.5
80,000 tons 35.5
25.3
29.8
3.0 x 106 bbls 125.1
89.2
98.1
90, 000. tons 13.3
20.9
31.2
1.0 x 106 tons 211.0
227.0
255.0
100,000 tons 50.1
36.1
42.2
1.46 x 106 tons ' 277.0
184.5
204.5
87,500 tons 17.5
41.6
63.4
10,000 tons 15.1
23.8
34.5
0.17
262.1
348.0
409.9
38.9
27.6
32.5
134.3
98.1
107.7
14.3
21.3
31.8
230.2 ,
247.0
274.2
53.7
39.3
46.3
297.0
199.9
221.5
18.9
42.4
64.6
16.3
24.3
35.2
0.20
293.0
390.0
457.0
44.0
31.1
36.6
148.2
111.4
122.1
15.9
22.0
32.7
259.0
277.0
303.0
59.2
44.1
52.4
327.0
223.0
247.0
21.0
43.7
66.3
18.1
25.2
36.3
0.30
396.0
530.0
614.0
61.0
42.7
50.2
194.4
155.8
170.2
21.2
24.1
35.6
355.0
377.0
399.0
77.4
60.2
72.8
427.0
300.0
332.0
28.0
47.9
72.1
24.2
27.9
39.7
-------
Table 3. TOTAL ANNUAL COST OF ACHIEVING BICDl/-LEVEL CONTROL OF FINE PARTICULATES IN SELECTED INDUSTRIES
Source
Coal-fired electric plant
Municipal incinerator
Cement plant (rotary kiln)
Model Plant
Production
Rate/Year
2.4 x 109 kwh
80,000 tons
3.0 x 106 bbls
Total
Annual Control Cost for
Model Plant at Specified
Capital Charge Rate ($1,000)
0.15
241.5
35.5
125.1
0.17
262.1
38.9
134.3
0.20
293.0
44.0
148.2
0.30
396.0
61.0
194.4
Total Annual
Industry
Production
516 x 109 kwh
18 x 106 tons
400 x 106 bbls
Total
Annual Control Cost for
Industry at Specified Capital
Charge Rate ($106)
0.15
51.9
8.0
16.7
0.17
56.4
8.8
17.9
0.20
63.0
9.9
19.8
0.30
85. L
13.7
25.9
Asphalt plant (rotary
dryers)
Iron and steel
(a) Basic oxygen furnace
(b) Electric arc furnace
(c) Sintering (windbox)
Ferroalloy
(a) Unhooded open electric
furnace
(b) Hooded open electric
furnace
(c) Closed electric
furnace
Lime plant (rotary kiln)
Iron foundry cupola
90,000 tons
1.0 x 10*> tons
100,000 tons
1.46 x 106 tons
8,000 tons
10,800 tons
13,600 tons
87,500 tons
10,000 tons
13.3
14.3
15.9 21.2 350 x 10$ tons
211.0 230.2 259.0 355.0 50 x 106 tons
50.1 53.7 59.2 77.4 21.5 x 10$ tons
277.0 297.0 327.0 427.0 54 x 10$ tons
205.5 220.5 243.0 318.0 600,000 tons
51.4 55.2 60.8 79.6 1.08 x 106 tons
3.8 4.1 4.5 6.0 820,000 tons
17.5 18.9 21.0 28.0 18.5 x 106 tons
15.1 16.3 18.1 24.2 13.1 x 10& tons
51.7 55.6 61.8 82.4
10.6 11.5 13.0 17.8
10.8 11.5 12.7 16.6
10.2 11.0 12.1 15.8
15.4 16.5 18.2 23.9
5.1 5.5 6.1 8.0
0.2 0.2 0.3 0.4
3.7 4.0 4.4 5.9
19.8 21.4 23.7 31.7
a/ See footnote (b), Table 1, for definition of BICD.
-------
Table 4. SUMMARY OF FINE PARTICULATE CONTROL COSTS AT BICD-LEVEL CONTROL
($/Unit of Production)
oo
Source
Coal-fired electric plant
Municipal incinerator
Cement plant (rotary kiln)
Asphalt plant (rotary dryers)
Iron and steel
(a) Basic oxygen furnace
(b) Electric arc furnace
(c) Sintering (windbox)
Ferroalloy
(a) Unhooded open electric furnace
(b) Hooded open electric furnace
(c) Closed electric furnace
Lime plant (rotary kiln)
Iron foundry cupola
Primary aluminum (electrolytic cells)
Primary copper
Annual Produc-
tion Rate for
Model Plant
2.4 x 109 kwh
80,000 tons
3.0 x 106 bbls
90,000 tons
1.0 x 106 tons
100,000 tons
1.46 x 10*> tons
8,000 tons
10,800 tons
13,600 tons
87,500 tons
10,000 tons
3,000 tons
76,000 tons
Fine
Particle
Control
Efficiency
98.13
96.71
99.17
99.21
99.70
97.64
99.10
60.00
87.80
97.52
98.43
98.46
94.85
98.85
Capital Charge Rate
0.15
0.00010
0.444
0.042
0.147
0.211
0.501
0.190
25.688
4.759
0.276
0.200
1.512
6.52
1.455
0.17
0.00011
0.486
0.045
0.159
0.230
0.537
0.203
27.563
5.107
0.298
0.216
1.613
6.99
1.595
0.20
0.00012
0.550
0.049
0.177
0.259
0.592
0.224
30.375
5.630
0.331
0.240
1.814
7.71
1.805
0.30
0.00017
0.763
0.065
0.236
0.355
0.774
0.292
39.750
7.370
0.441
0.320
2.418
10.08
2.505
-------
The most important variable is population density. Total economic losses
attributable to sustained exposure at the prescribed concentrations range
from about $5,000/mile2 in sparsely populated rural areas, to well over
$1 million/mile^ in densely populated urban locations, with per capita
costs ranging from over $500 to less than $30, depending on the char-
acteristics of the area in question.
The control efficiency at which the total of control costs and the costs
of remaining damages is a minimum identifies the optimum control level;
this is also the point at which incremental control costs equal the incre-
mental benefits of control. A low population density will have a relatively
low minimum net cost and will require a relatively low control efficiency.
Densely populated areas will experience much higher damage costs, and there-
fore justify much higher control efficiencies.
The effectiveness of any program for the control of fine particulate emis-
sions must ultimately be judged by improvements in ambient air quality.
Because both atmospheric conversion processes (chemical and physical) and
sedimentation contribute to the fine particulate burden (in the atmosphere,
it may be' difficult to relate atmospheric levels of fine particulates to
control activities on specific sources. For example, in Los Angeles, about
35% of the ambient particulate burden is formed in the atmosphere from
gaseous pollutants. In a city like Los Angeles, reductions in emissions
of fine particulates from stationary sources would not result in a propor-
tionate decrease in ambient fine particulate concentration. As a conse-
quence, development of meaningful benefit/cost relationships for fine
particulate control may be quite difficult in some urban locations. While
the assessment of the overall effectiveness of control efforts may be dif-
ficult, the total burden of fine particulates entering the atmosphere would
nonetheless be reduced by direct limitation of source emissions.
OVERALL FEASIBILITY OF EMISSION STANDARDS BASED ON PARTICLE SIZE
Our analysis of the implications of emission standards based on particle
size has identified some technical deficiencies that will limit the type
of standards that can be proposed and implemented in the near future.
However, because the technical deficiencies relate primarily to the lack
of data on particle size distributions of effluents and control equipment
fractional efficiency, there appear to be no insurmountable technical ob-
stacles to emission standards based on particle size.
Although our analysis of various formats for emission standards based on
particle size was performed in the context of general regulations that
.might be applied uniformly to all sources, a more realistic approach would
-------
be to tailor the emission standard to specific sources of fine particulate
pollutants. Tailoring of standards would permit a greater degree of flexi-
bility in an overall control plan for fine particulates, and would acknowl-
edge the differences in the importance and difficulty of control of in-
dividual sources.
The exact format of the emission standard(s) that could be proposed and
implemented for specific sources will be limited by: (1) collection ef-
ficiency ixx fine particle size range of available control equipment; and
(2) availability of source compliance monitoring techniques.
The economic impact of fine particulate control was found to vary sub-
stantially from industry to industry. Estimates of costs associated with
the control of fine particulates varied from less than 1.0% up to 20% of
,-r
the value of the product. The variation in economic impact suggests that
it may be necessary to consider less restrictive standards for industries
that experience a significant adverse economic impact.
Because our understanding of the damages resulting from fine particulate
pollutants is far from complete, definitive statements regarding the bene-
fits that would accrue from improved control of fine particulate emissions
cannot be made at this time. In view of our lack of knowledge of benefits
resulting from control activities, it seems prudent to consider a long-
range strategy based on the adoption and implementation of progressively
more stringent regulations for the control of fine particulates. Each
step toward more effective control of fine particulates would be based
upon improved knowledge of: (1) the effects of fine particulates on human
health and welfare; (2) the technical and economic impact of the more
stringent regulations; and (3) the benefits that would accrue to society
from a further decrease in fine particulate pollution.
10
-------
SECTION II
RECOMMENDATIONS FOR FUTURE WORK
INTRODUCTION
The existing information base in nearly every aspect of the fine particu-
late pollution problem is inadequate. Research or data acquisition programs
should be formulated and undertaken in several areas, especially: (1)
Source Sampling and Monitoring Methods; (2) Fine Particle Emission Rates;
(3) Control Technology Evaluation and Development; (4) Analysis of Relation-
ships Between Fine Particle Emissions and Ambient Air Quality; (5) Economic
Impact Analysis; and (6) Cost/Benefit Analysis. Specific programs in each
of the above categories are delineated in the following sections.
SOURCE SAMPLING AND MONITORING METHODS
Current capability to sample, size and monitor particulate emissions from
stationary sources is inadequate. Research should be initiated to improve
this capability. Specific areas of recommended research are:
1. Laboratory and field evaluations of promising methods for the measure-
ment of the concentration and particle size distribution of particulates
smaller than about 5 um. Methods that should be evaluated include im-
pactors, cyclones, beta-tape devices, and piezoelectric crystal devices.
2. Evaluation of optical techniques for monitoring fine particle emis-
sions. Attention should be directed to a comprehensive evaluation of the
performance of on-stack transmissometers on a variety of industrial sources.
•
FINE PARTICLE EMISSION RATES
Emission factors for fine particulates are nearly nonexistent at the pres-
ent time. Source testing programs shall be initiated to gather data on:
1. Rates of emission of particulates in the fine particle size range--
nominally less than 2 um.
11
-------
2. Chemical composition of particles and carrier gas emitted from sta-
tionary sources. Emphasis should be placed on trace metals and potentially
hazardous compounds.
A portion of this recommended research could be conducted in conjunction
with the programs outlined in the preceding section. However, because ex-
perience has generally shown that no single sizing device is suitable for
all the sampling circumstances encountered in a wide variety of stationary
emission sources, individual testing programs may have to be developed for
specific classes of sources.
CONTROL TECHNOLOGY EVALUATION
The ability of currently available control equipment to collect fine par-
ticulates is ill-defined. Programs should be developed to obtain basic
performance data on particulate control equipment as a function of particle
size. Emphasis should be placed on measuring collection efficiencies in
the fine particle size range. These programs should be interfaced with
those outlined in the preceding sections so that the best measuring tech-
niques are utilized.
CONTROL TECHNOLOGY DEVELOPMENT
Presently available information indicates that existing control equipment
may not be adequate for the collection of fine particulates from many in-
dustrial sources. Research and development programs should be formulated
to foster new control technology for fine particulates. Attention should
be focused in the following:areas:
1. Field evaluation of new or novel and emerging control technology.
2. Laboratory and field evaluation of particle conditioning processes
which increase the effective size of particles and thus makes them less
difficult to 'capture.
3. Research on methods to improve the performance of existing control
equipment.
ANALYSIS OF RELATIONSHIPS BETWEEN SOURCE EMISSIONS AND AMBIENT AIR QUALITY
The effectiveness of any program for the control of fine particulate emis-
sions will be judged primarily by improvements in ambient air quality.
Unfortunately, the fine particulate pollution problem is only partly a
12
-------
primary particle emission problem. Secondary conversion processes may con-
tribute significantly to the fine participate burden in some areas. Because
of the influence of atmospheric conversion processes and sedimentation on
the fine particulate burden, in the atmosphere, it may be difficult to relate
atmospheric fine particulate levels to control activities on specific sources.
Additional research on methods to relate air quality to emission sources is
highly recommended. Priority should be given to studies for urban and in-
dustrial regions with a high density of fine particulate sources with dif-
fering source characteristics.
ECONOMIC IMPACT ANALYSIS
More refined analysis of the economic impact of fine particulate emission
standards should be conducted. One phase of this research activity should
focus on the acquisition of more detailed data on the costs of high effi-
ciency control equipment. Costs of particulate control equipment vary
widely. The costs cover a range of values because of local conditions and
the nature of the particles, the gas stream, equipment size (gas volume),
and design collection efficiency. Published average cost figures frequently
do not reflect all cost components and fail to illustrate the great range
in costs.
Future economic impact analysis should focus on definite urban or industrial
regions. If possible the analysis should include consideration of factors
such as changes in industry structure, price elasticity in specific industries,
and number of potential plant closings.
COST/BENEFIT ANALYSIS
A more refined analysis of the trade-offs between costs and benefits as-
sociated with emission standards based on particle size should be per-
formed. A major part of this activity should be directed to the quantifi-
cation of the effects of fine particulates on various receptors. At the
present time, the lack of data on the damages associated with fine partic-
ulate pollution hampers efforts at meaningful cost/benefit analysis.
13
-------
SECTION III
INTRODUCTION
When air pollution became a major concern in the late 1950's and early
1960's, particulate pollutants from industrial processes were among the
first to receive attention and to be controlled. While we have succeeded
in removing the black cloud from the smoke stack and with it the major
fraction of mass emitted from industrial processes, we have not been nearly
as successful in eliminating the particulates which cause the major adverse
effects—the particulates below about 2 urn in size (i.e., fine particulates)
Fine particulates resulting frdfo man's activity contribute significantly
to all the major adverse aspects of air pollution. Fine particles ban
initiate or contribute to problems related to human health, changes in
atmospheric physical properties, and materials soiling and damage. The
effects of particulate matter on human health are, for the most part, re-
lated to injury to the surfaces of the respiratory system. Particulate
material in the respiratory system may produce injury itself, or it may
act in conjunction with gases, producing synergistic effects. Such injury
may be permanent or temporary. It may be confined to the surface, or it
may extend beyond, sometimes producing functional or other alterations.
Fine particulate pollutants also affect the physical properties of the
atmosphere. The chemical and physical properties of the atmosphere af-
fected include: its electrical properties; its ability to transmit radiant
energy; its ability to convert water vapor to fog, cloud, rain, and snow;
its ability to damage and to soil surfaces.. Concern with atmospheric
transmission of radiant energy encompasses the entire electromagnetic
spectrum, but of particular concern is the infrared region, as it affects
the terrestrial heat balance; the ultraviolet region, as it affects both
biological processes and photochemical reactions in the atmosphere; and
the visible spectrum, as it affects both our ability to see things, and
our need for artificial illumination.
14
-------
As man has become more knowledgeable about his environment, he has also
become increasingly aware of the importance to his well-being of the action
of fine particles. This awareness has resulted in increased attention
being focused on fine particulate pollutants by both governmental and
scientific groups. Regulations for the control of fine particle emissions
from industrial sources are being considered by both federal and state
pollution control agencies.
The Standards Research Branch, Implementation Research Division, Environ-
mental Protection Agency, realizing the developing interest in controlling
the emission of fine particulate pollutants from industrial operations,
initiated a program with Midwest Research Institute (MRI) to conduct a
study to define the technical and economic feasibility of particulate
emission standards based on particle size. The program was divided into
four major areas of effort:
1. Analysis of approaches for regulating fine particle emissions from
stationary sources.
2. Definition of technological and economic requirements necessary for
implementation of emission standards.
3. Identification of benefits that would accrue if control procedures
for fine particulates can be implemented.
4. Assessment of overall feasibility of implementation of fine particle
emission standards.
In the following chapters of this report we present a discussion of the
nature of fine particulate pollution, a description of the major stationary
sources of fine particle emissions, and the results of activities in the
four major areas of effort delineated in the previous paragraph. Chapter 2
presents a discussion of the role of fine particles in air pollution,
while Chapter 3 contains a description of the major stationary sources of
fine particle emissions. Chapters 2 and 3 are included in the report in
order to acquaint £he reader with the nature, magnitude, and sources of
fine particulate pollution. Chapter 4 presents our analysis of approaches
for regulating fine particle emissions. The technological and economic
implications of fine particle control are presented in Chapters 5 and 6.
A general discussion of potential benefits from the control of fine par-
ticle emissions is given in Chapter 7. Our assessment of the overall
feasibility of emission standards for fine particles and recommendations
for future work are given in Chapters 8 and 9.
15
-------
SECTION IV
ROLE OF FINE PARTICLES IN AIR POLLUTION
INTRODUCTION
Particulate pollutants resulting from man's activity are chemically and
physically a most diverse class of substances. They are variously de-
scribed as grit, dust, fume, smoke, aerosol or smog. The meanings gen-
erally given to these terms are listed in Table 5.V In this report the
term particulate pollutant is used to mean any dispersed matter, solid
or liquid, in which the individual aggregates are larger than single
small molecules (about 0.0002 um in diameter), but smaller than about
500 um. This is the same definition used in the document, Air Quality
Criteria for Particulate Matter.—/
Particulate pollutants may remain suspended in the atmosphere for times
ranging from a few minutes to many hours. Particulate removal from the
atmosphere occurs by diffusion, sedimentation, capture by cloud droplets
and scrubbing by precipitation. In the absence of precipitation, sedi-
mentation dominates and the removal of airborne particles from the
atmosphere depends upon the aerodynamic particle size of the particles.
Figure 1 presents a relationship between particle size (aerodynamic) and
the mean residence time for particles in the atmosphere in the absence
of precipitation.3/ As shown in Figure 1, particles in the range of 10 um
to 1 um have mean residence times of 10-100 hr, while the mean residence
time of submicron particles is on the order of 102-lcPhr.
A profile of the effects of particulate air pollution in the community is
presented in Figure 2. The effects are all related to particle size.
Comparison of Figures 1 and 2 indicates that with the exception of soiling
or damage of horizontal surfaces, the particles with residence times ex-
ceeding 100 hr (i.e., fine particulates) play the most important role.
The overall impact of fine particulate pollutants, broadly defined as
particles less than 2 um in size, on man's environment is not well known.
16
-------
Table 5. CLASSIFICATION OF PARTICULATE POLLUTANTS^
I/
Grit:
Dust:
Fume:
Mist:
Fogs:
Smoke:
Smog:
Soot:
Aerosols:
Coarse particles, greater than 76 um which is the size of the
opening in the 200 mesh sieve.
Particles smaller than 76 um (i.e., able to pass through a 200
mesh sieve) and larger than 1 um.
Solid particles smaller than 1 um.
Liquid particles, generally smaller than 10 um.
Mists are sometimes called fogs when they are sufficiently
dense to obscure vision.
This is the term used generally to describe the waste products
from combustion, and may either be fly ash or the products
of incomplete combustion, or both. The particles can be
liquid or solid.
A term--a combination of smoke and fog—used to describe any
objectionable air pollution. There are two kinds, known as
the Los Angeles and the London type. The Los Angeles smog
is photo-chemical and comes from motor car exhausts. The
London type comes from the incomplete combustion of coal and
is characterized by its relatively high sulphur dioxide con-
centration and particle content.
Soot is the aggregated particles of unburned carbon produced
by incomplete combustion.
Initially this term was used for the fine relatively stable
aerial suspensions. In recent years the term has been
generally applied to all airborne suspensions.
17
-------
? 102 ~
0)
0)
u
c
01
c
o
u
10
10°
Limit of Residence Time
i i i i i i i I
. ..... .1
1 —
10-
Particle Size, de (u,m)
•Figure 1. Relationship between particle mean residence time and particle
size^/
-------
10
c
i
J
Atmospheric
Electricity
Atmospheric
Visibility
Condensation
Nuclei for Precipitation
Soiling Phenomena
(Horizontal Surfaces)
Upper Respiratory
Tract Deposition in Man
Peripheral Airways and
Alveolar Deposition-Man
Soiling Phenomena
(Vertical Surfaces)
Mmospneric
Chemistry (Gas-Solid)
Y/////////////A
w//,
Particles Comprising//
/Main Aerosol.Mass
10-
ID'3
io-'
10'1 10°
Particle Diameter - u.\
10'
102
Figure 2. Effects of particulate atr pollution in the community as related to particle sizeV
-------
However, evidence continues to point to the many negative aspects of fine
particle pollution. Research on health effects of air pollution indicate
links between fine particulate pollution and health effects of varying
severity. In addition to particle size, the chemical composition of par-
ticulates is an important factor in determining the effects of this type
of air pollution. For example, the health hazard associated with inhaled
airborne particles depends on: (1) the site of deposition in the respi-
ratory tract, which is determined by particle size; and (2) the effect on
biological tissues at the deposition site, which depends on chemical com-
position. The hazardous pollutant problem (e.g., mercury, lead, cadmium,
vanadium) is also directly linked to fine particle pollution because many
of the industrial processes that emit hazardous pollutants liberate them
in the form of micron or submicron particles. On both a regional and
global scale> fine particulate pollutants play a principal role in the
transport through the air of a variety of hazardous substances. It is
emphasized that the hazards posed to human health by trace amounts of
toxic materials in the form of fine particulates can be disproportionate
to the mass involved.
Suspended fine particles may have considerable influence on the behavior
of the atmosphere and thus on human activities. Fine particles in the
atmosphere absorb and scatter light, decrease visibility, influence solar
radiation, and interfere with astronomical observations.
The role of fine particles in air pollution is presented in more detail
in the following sections. Health effects will be discussed first, fol-
lowed by a review of the effects on the physical properties of the
atmosphere.
EFFECTS OF FINE PARTICULATES ON HUMAN HEALTH
Humans as biological organisms with varying degrees of resistance and
adaptive capacity continuously struggle with an essentially hostile en-
vironment. Anything lowering the resistance of man or increasing the
hostility of the environment, decreases his ability to adapt. A number
of studies have shown that children and marginal people with poor adap-
tive ability because of serious lung disease, heart disease, asthma,
or other serious chronic diseases, have increased respiratory illness
and cardiac and respiratory deaih rates parallel to the rise in pollution
levels. Thus, at every level of pollution and not at some defined thresh-
old, depending on the adaptive reserve of individuals, someone becomes
sick and someone's life is shortened.
20
-------
Any attempt to understand the relationship between air pollutants and human
health requires the integration of many diverse factors. The dynamics of
small particles, respiratory system and lung dynamics, mechanisms of respi-
ratory and lung disease, area of residence (i.e., urban vs rural), occupa-
tion, smoking habits, and age are among the more important elements that
dictate the influence of air pollutants on human health. An extensive
body of literature exists on the subject, and only a brief review of a
small segment of the literature will be presented, to orient the reader.
Deposition, Retention, and Clearance Processes in the Respiratory System
Laboratory studies of man and other animals show clearly that the deposi-
tion, clearance, and retention of inhaled particles is a very complex
process, which is only beginning to be understood. Understanding of these
phenomena requires knowledge of the following:
1. Mechanisms and efficiencies of particle deposition in the respiratory
system,
2. Retention mechanisms,
3. Clearance mechanisms, and
4. Secondary relocation to other sites in the body.
The physical forces which operate to bring about aerosol deposition within
the respiratory system vary in magnitude not only with particle size, but
also with the air velocities and times of transit of the air from place to
place within the system and from moment to moment throughout the breathing
cycle. Three mechanisms are of importance in the deposition of particulate
matter in the respiratory tract:
1. Inertial impaction - greatest importance in deposition of large parti-
cles of high density, and at points in the respiratory system where the
direction of flow changes at branching points in the airways,
2. Gravitational settling (sedimentation) - most important in the deposi-
tion of large particles or of high-density particles such as dusts of
heavy metals, and
3. Diffusion (Brownian motion) - major mechanism for the deposition of
small particles (below 0.1 urn) in the lower pulmonary tract.
The effectiveness with which the deposition forces remove particles from
the air at various sites depends upon the obstruction encountered, changes
21
-------
in direction of air flow, and the magnitude of particle displacement
necessary to remove them from the air stream. The anatomical arrangement
and physical dimensions of the respiratory system, transport mechanisms,
flow rates and gas mixing, and aerosol particle size are important factors
that must be considered in any physical analysis of the deposition of in-
haled aerosols.
The Task Group of Lung Dynamics has recently developed a model for the
deposition of particles in the respiratory tract,^J Findeisen's anatomi-
cal model"/ was chosen as the basis for the Task Group Model. Calcula-
tions, based on the model, of deposition in respiratory compartments are
shown in Figure 3. The curves indicate a relationship between the mass
median aerodynamic diameter and the gravimetric fraction of the inhaled
particles which would be deposited in each anatomical compartment.
Experimental studies of the deposition of inhaled particulate material may
be divided into two broad categories. The first group deals with the mea-
surement of total deposition in the respiratory tract, and the second group
is concerned with regional deposition within the various areas of the
respiratory tract. Details of these experiments are given in Ref. 7.
The following points can be made with respect to the overall retention
characteristics of the respiratory system:Z/
1. Percentage deposition increases with aerodynamic particle size from
a minimum value of about 25% at ** 0.5 urn and approaches 100% for particles
> 5 urn. Particles of different densities and shapes follow the same depo-
sition curve when size is expressed in terms of equivalent diameter of
unit-density spheres. Particles larger than 10 urn are essentially all
removed in the nasal chamber and therefore have little probability of
penetrating to the lungs. Upper respiratory efficiency drops off as size
decreases and becomes essentially zero at about 1 um.
2. The efficiency of particle removal is high being essentially 100%
down to around 5 um. Below this size it falls off to a minimum at about
0.5 iim. It then increases again as the force of precipitation by dif-
fusion increases with further reduction in size. For particles < 0.1 vim,
the percentage deposited out of the total respired air approaches, in
value, the fraction of tidal volume which reaches the pulmonary air spaces.
This fact suggests that the absolute efficiency of alveolar deposition of
these submicron particles approaches 100%.
3. Percentage deposition varies with breathing frequency, increasing, for
a heterogeneous aerosol, in both directions from a minimum level at fre-
quencies of 15-20 cycles/min. At slower rates, the probability of
22
-------
NASOPHARYNGEAL
3?&w PULMONARY
MASS MEDIAN DIAMETER,'MICRON
Figure 3. Fraction of particles deposited in the three
respiratory tract compartments as a func-
tion of particle diameter^/
23
-------
deposition by gravity settlement and diffusion goes up in proportion to
the increase in transit time of the dust-laden air into and out of the
lungs. With more rapid breathing percentage deposition of the coarser
particles increases because of the rise in force of inertial deposition
with increasing air velocity. r
4. Particles of hygroscopic materials are removed in higher percentages
than are nonhygroscopic particles of the same (dry) size because of the
growth of such particles by water adsorption from the moist air in the
respiratory system.
5. There is reasonably good agreement between the directly measured
values of overall deposition and the levels predicted from mathematical-
physical equations, with respect to both particle size and dynamics of
air flow into and out of the respiratory system.
Clearance of Particulate Matter from the Respiratory System
Different clearance mechanisms operate in the different portions of the
respiratory tract, so that the rate of clearance of a particle will depend
not only on its physical and chemical properties such as shape and size,
but also on the site of initial deposition. The fast phases of the lung
clearance mechanisms are different in ciliated and nonciliated regions.
In ciliated regions, a flow of mucus transports the particles to the
entrance of the gastrointestinal tract, while in the nonciliated pulmonary
region, phagocytosis by macrophages can transfer particles to the ciliated
region. The rate of clearance is an important factor in determining toxic
responses, especially for slow-acting toxicants such as carcinogens. The
presence of a nonparticulate irritant or the coexistence of a disease
state in the lungs may interfere with the efficiency of clearance mecha-
nisms and thus prolong the residence time of particulate material in a
given area of the respiratory tract. In addition, since the clearance of
particles from the respiratory system primarily leads to their entrance
into the gastrointestinal system, organs remote from the deposition site
may be affected.
The Task Group on Lung Dynamics developed a model for respiratory clearance
and also summarized experimental work on clearance. Details of the model
and the analysis of experimental data are presented in Refs. 2 and 5., and
the reader interested in a more complete discussion of respiratory clear-
ance is directed to these sources.
Toxicological Studies of Atmospheric Particulate Matter
Experimental toxicology develops information on the mode of action of
specific pollutants, on the relative potency of pollutants having a similar
24
-------
mode of action, and on the effect of one pollutant on the magnitude of
response to another. If man could be used as the experimental subject,
experimental toxicology would be the best means of deriving air quality
criteria. However, the impossibility of performing experiments using
human exposures to varying concentrations of a wide range of compounds
precludes this direct approach. A limited amount of intentional human
experimentation has been conducted, but most of the data for human toxi-
cology are derived from accidental or occupational exposures.
The use of laboratory animals in toxicological experiments is more straight-
forward, but the obvious anatomical and metabolic differences between ani-
mals and man require the exercise of caution in applying the results of
animal exposures to human health criteria. Furthermore, many of the
animal experiments have been conducted at exposure concentrations far in
excess of those likely to be found in the atmosphere.8~13/
In spite of these limitations, toxicological studies have shown that
atmospheric particles may elicit a pathological or physiological response.
Other conclusions from these studies are:
1. The presence of an inert particle in the respiratory tract may inter-
fere with the clearance of other airborne toxic materials,
2. Evaluation of irritant particulates on the basis of mass or concentra-
tion alone is not sufficient; data on particle size and number averages
per unit volume of carrier gas are needed for adequate interpretation,
3. Particles below 1 um have a greater irritant potency than larger
.particles, and
4. A small increase in concentration could produce a greater-than-linear
increase in irritant.response when the particles are < 1 um.
The possible influence of inert particulate matter on the toxicity of ir-
ritant gases has been the subject of considerable speculation and a limited
amount of experimental work. Work on the synergistic effects of aerosols
and irritant gases is reported in Refs. 14-27- Available information in-
dicates that gases and particulates acting together may cause greater damage
than either one acting separately.
i
Also, if the particulates and gases are delivered in an alternating cycle,
the damage is also, potentiated. This suggests that the defense evoked by
one of the challenges may impair the protection that the 'lung has against
the other. These observations, which admittedly have been made on occasion
25
-------
with greater concentrations of pollutants than normally encountered, should
at least alert us to the possibility that a certain ambient particle con-
centration might be without hazard in an otherwise clean environment, but
harmful if respired in association with gases commonly found in an urban
environment.
Epidemiological Studies of Atmospheric Particulate Matter
The ultimate assessment of the impact of air pollution on human health can
come only from epidemiology.—' Because particulate matter and gaseous
pollutants tend to occur together in a polluted atmosphere, few epidemic-
logic studies have been able adequately to differentiate the effects of
the two pollutants. Because we do not now have a good epidemiological
basis for stating the influence of particulates, only a brief synopsis
of epidemiological studies of atmosphere particulate matter will be
presented. References 2 and 28 present an extensive review of epidemiolog-
ical studies, and those seeking more detail are directed to these sources.
Epidemiologic studies of the relationship between pollutant concentrations
and their effects on health have used indices varying from disturbance of
lung function to death. British studies of acute episodes of increased
pollution show excess deaths occurring at smoke levels from 750 ug/nH to
2,000 ug/m3. High S02 levels are, of course, concurrently present. The
excesses of mortality are always accompanied by a very large increase in
illness, mainly exacerbations of chronic conditions. Similar, but less
spectacular, episodes have been reported in New York City..?./
Winkelstein found in Buffalo that increases in the mortality rate were
significantly linked to higher levels of suspended particulate pollution.29/
His studies showed that mortality from all causes, from chronic respiratory
diseases, and from gastric carcinoma increased from the lowest of his five
levels of pollution (less than 80 ug/m3) through the three higher ranges,
after the effects of socioeconomic status had been considered. Zeid*berg
found in Nashville significant increases in all respiratory deaths at soil-
ing levels over 1.1 cohs annual average.30,3I/ Neither of these studies
took smoking habits into account and the Nashville study only partially
allowed for socioeconomic factors.
Studies of illness in relation to residence in more- and less-polluted
areas contribute additional information. Fletcher, et al., noted a propor-
tional decline in the production of morning sputum in chronic bronchitics
in West London from 1961 to 1966 as smoke pollution in their residence
areas declined from 140 ug/m3 annual mean.r_£/ Douglas and Waller found
an increase in frequency and severity of lower respiratory illness at
26
-------
smoke and SC>2 levels over 130 ug/m3 annual average.33/ A study of Lunn,
et al. shows similar differences occurring with more morbidity measured
between about 100 ug/m3 and 200 ug/m3 of smoke, and for others between
200 ug/m3 aruj 300 ug/m3 annual average.2z/
Physiologic studies of lung function have also been made in both adults
and children. On the basis of present limited knowledge it appears that
the alterations found may be both temporary and permanent. The observa-
tions now available relate to long-term residence in a given area.34;35/
The study reported in Ref. 34 shows reduced pulmonary function in the
children living in areas of high dustfall as compared with those living
in low dustfall areas. In the Osaka study the dustfall levels were 6.5
gm/m2-month and 13.3 gm/m^-month.3JL/ Douglas and Waller33_/ have shown
that there is a three-fold increase in morbidity from lower chest infec-
tions in infants younger than 2 years in moderate and high pollution
regions, compared to very low pollution regions.
Bates36_/ has recently reviewed advanced concepts of lung function and has
made an attempt to differentiate in a preliminary way the effects that an
inhaled substance may have on the human lung. Figures 4 to 6 present in
a diagrammatic form Bates' analyses of the effects of different materials
on the major bronchi of the lung, on the terminal bronchioles, and on the
alveoli.
In summary, examination of the studies of the potential effects of air
pollutants on lung functions and available epidemiologic data indicates
that there is an association between air pollution, as measured by both
particulate matter and gaseous pollutants, and health effects of varying
severity. Increased respiratory disease morbidity is almost certainly
related to air pollution, and fine particulates play an important role
either along or in combination with gaseous pollutants.
MODIFICATION OF PROPERTIES OF THE ATMOSPHERE
The emission of fine particles into the atmosphere can increase the sus-
pended particulate burden, modify the properties of the atmosphere, and,
as a result, influence many facets of human activity. The suspended
atmospheric particles are of four general types. The first are the "light
ions" produced in the air by cosmic rays and radioactivity. They consist
of small aggregates of molecules having dimensions up to a few molecular
diameters. The second important type of particle consists of the so-called
"aitken nuclei." These particles range in radius from 2 x 10~7 cm up to
10~4 cm. They are particularly prevalent near cities and the earth's surface.
27
-------
PARTICLES
2.0-50.0/1
GAS EXPOSURE
SO2,NO2,O3etc.
1 PARALYSIS OF CILiA
2 HYPERSECRETION
3 MUCUS GLAND HYPERTROPHY
AND EXTENSION
4 SUSCEPTIBLITY TO INFECTION
CHRONIC PRODUCTIVE COUGH
Figure 4. Effect of irritants in major bronchi
PARTICLES
GAS EXPOSURE
SO,NO.O3 etc.
POTENTIATION IF BOTH
PRESENT
Loss of Normal Defences
Effect on Surfactant
Goblet Cell Metaplasia
Inflammation and Obliteration
Premature Closure
EFFECTS ON;
Collateral
Ventilation
Gas Exchange
Stress of Lung
Release of Proteo-
lytic Enzymes
Figure 5. Effect of irritants in terminal bronchioles
PARTICLES
0.01/1-0.5/1
GAS EXPOSURE
>k. Jf
POTENTIATION IF BOTH
PRESENT
1
INCREASE OF CELLS AND
MACROPHAGES IN LUNG
\
RELEASE OF PROTEOLYTIC
ENZYME
/
INFECTION
s^ Protection by
.x^ Anti-Proteolytic
S^ Enzymes
(a\~ Antitrypsin etc. )
«
EMPHYSEMA WITH
ALVEOLAR
DESTRUCTION
Figure 6. Effect of irritants in alveoli
28
-------
As a rule of thumb, one anticipates finding about 100,000 of these parti-
cles per cm3 in a large city, about 10,000/cm3 in the country, and about
I,000/cm3 at sea. The numbers decrease with increasing altitude and only
about 107o of the surface populations are found at an altitude of 7 km.
The fine particle pollution of the air is largely composed of these nuclei.
The third type of atmospheric particle is the cloud droplet having a radius
from 10~4 cm to 5 x 10~3 cm. Finally, the cloud droplets associate to
form raindrops or snowflakes that fall at velocities dependent upon their
size. Rain or snow represents the final stage for the removal of atmo-
spheric pollution by natural methods.
The interactions of fine particulate pollutants with atmospheric processes
are reviewed in the following sections.
Visibility
One of the most obvious effects of air pollution is the reduction in
visibility which results from the accumulation of particulate matter in
the atmosphere. Decreased visibility interferes with certain human
activities, such as safe operation of aircraft and automobiles and the
enjoyment of scenic vistas. Air pollution that reduces visibility, in
addition to endangering the safety of both air and land travel, results
in inconvenience and economic loss to the public, and to transportation
companies due to disruption of traffic schedules.
Deterioration of visibility caused by suspended particulate matter is the
result of adsorption and scattering of light. Both the brightness of the
viewed object and its visual contrast with the background are reduced by
attenuation of light due to scattering and absorption. In addition, a
further contrast reduction results from scattering of sunlight into the
observer's line of sight. Loss of brightness and contrast are responsible
for the subjective impression of impaired visibility.
Studies of the theory of visibility by numerous authors have led to a
very useful relationship between visual range and mass concentration of
atmospheric particulate matter. The visual range is defined as the dis-
tance, under daylight conditions, at which the apparent contrast between
object and background is just equal to the threshold contrast of the ob-
server. The usual assumption of threshold contrast of 0.02 for the human
eyes leads to expression"
L = 3.9/b , (1)
29
-------
where L is the visual range in meters and b is the extinction coefficient
per meter along the sight path for a black target.
Generally, the extinction coefficient can be depicted as the sum of several
components:
b = bscat + bRayleigh + babs-gas + babs-aerosol (2)
where bscat represents the component due to light scattering by aerosol,
bRayleigh the scattering due to gaseous air (the blue-sky scatter), and
babs-gas and babs-aerosol represent the absorption due to gases (like N02)
and particles (such as carbon black), respectively. Middleton and dthers
have suggested that bscat is dominant especially in situations where the
visual range is somewhat degraded due to haze.39/ Measurements of light
scatter in urban areas tend to support this assumption.
Mass concentration of atmospheric particulate matter has been found to be
approximately proportional to the scattering coefficient, and Char Is on,
et al.,^2/ and Noll, et al.,^1/ have developed a useful relationship be-
tween visual range and mass concentration of atmospheric particulate
matter.
Combining Eq. (1) with the observation that the mass concentration is ap-
proximately proportional to "b" results in Eq. (3) .
L = kM'1 (3)
Figures 7 and 8 illustrate visibilities calculated using the above approach
These and other calculations clearly show that, with respect to aerosol
particle size, visibility reduction related to air pollution is caused
'primarily by the 0.1-1.0 um radius particles.
Most industrial, combustion, and vehicular sources emit particles in a
wide variety of sizes. From the preceding discussion, it is clear that
the major interest insofar as visibility-reducing particles is concerned
centers on those emitted particles in the 0.1-1.0 um radius range.
30
-------
80
70
60
VI
0>
50
_i 40
u
o
g
o
2 so
20
10
Calculated Visibilities of
Ferric Sulfate Aerosol
Density = 3.09 gm/cm3
Refractive Index = 1.8
Incident Light = 5240 Angstroms
— Parameter: Concentration in Ambient Air
I
I
0.1
0.2 0.3
PARTICLE RADIUS, Microns
0.4
Figure 7. Meteorological visibility vs particle^./
size, ferric sulfate aerosol
31
-------
30 r-
Calculated Visibilities of
Flyash Particulates
Density = 1.7
Refractive Index = 1.55
Incident Light = 5240 Angstroms
in
0)
. 20
Parameter: Concentration in Ambient Air
CO
LO
NJ
O
g
o
9 10
o
0.1
0.2
Figure 8.
0.3
PARTICLE RADIUS, Microns
0.4
0.5
Meteorological visibility vs particle size,
flyash aerosol
-------
Solar Radiation
The intensity and spectral distribution of direct sunlight and scattered
daylight, and the variation of intensity with time of day, season, latitude,
altitude, and atmospheric conditions are important because they affect
photosynthesis in plants and the distribution of plants and animals on
earth, the weathering of natural and manmade materials, climate, and il-
lumination for human activity.—'
The attenuation of solar radiation through the atmosphere is caused by a
number of physical factors;43-45/
1. Rayleigh scattering by air molecules such as N£ and 02, and particles
in size ranges less than the wavelength of solar radiation,
2. Selective absorption by gaseous constituents of the atmosphere, and
3. Scattering and absorption by atmospheric dusts and particulate matter
of a size greater than the wavelength of solar radiation.
Except in cases of heavy particulate pollution of the atmosphere, such as
may occur in large urban centers or heavy industrial areas, it appears
that the effect of turbidity is to scatter radiation out of the direct
solar beam and add an almost equal amount to the diffuse beam arriving
from the rest of the sky by forward-scattering. In cases of heavy parti-
cle concentrations, however, the loss from the direct solar beam greatly
exceeds the gain in the downward scattered beam, the difference being
lost to backscattering off the top of the pollution layer, and to ab-
sorption within the polluted layer or column.
Landsberg-t2/ and SteinhauserftZ./ report that cities in general receive
less solar radiation than do their rural environments. Seasonal, weekly
and daily variations in total solar radiation in urban communities have
also been noted.zZl^2./ These variations appear to be related, in part,
to cycles of industrial and commercial activity. Measurements at the
Moana Loa Observatory in Hawaii, which is remote from any local sources
of air pollution, indicate a long-term increase in turbidity or atmo-
spheric dustiness.lP./ McCormick and Ludwig have reported significant
increases in recent years of turbidity over Washington, D.C., and Davos,
Switzerland. Washington and Davos are probably the only localities with
reliable records over a 50-year span.^P./ Davos is an alpine station far
removed from major pollution sources. Cobb reports that man-generated
particulate pollution has not yet affected the atmosphere over 90% of the
33
-------
oceans.—*— Cobb made his conclusion after comparing electrical con-
ductivity readings taken in the late 1960's with those taken early in
the century. Exceptions to the general pattern include paths of anthro-
pogenic aerosol pollution extending from the U.S. across the North
Atlantic and from Japan and Asia across the North Pacific. A third path
was found across the Indian Ocean, probably caused by monsoon-carried dust.
Aerosol concentration is more varied now than it was 60 years ago, accord-
ing to Cobb. Concentration has doubled over the North Atlantic since 1911
but has remained fairly constant over the South Pacific. Cobb stressed
his results show only one aspect of man's effect on the atmosphere, but
he is confident that particulate concentration will decline and its in-
fluence on climatic conditions will decrease as regulations to reduce the
output of particulates are enforced.
Watt has suggested that one way to assess the influence of air pollution
on atmospheric properties is to find some natural events in the past that
operated in a manner analogous to modern pollution.-LI/ He analyzed the
changes in weather patterns following the eruptions of the Tambora and
Krakatau volcanoes, and suggests that these volcanoes had effects similar
to the global increase in atmospheric pollution.
The net influence of atmospheric turbidity on surface temperature is un-
certain, but the emission of long-lived particles may well be leading to
a decrease in world air temperature.54/ As more is learned aboub the
general circulation of the atmosphere and the delicate balance between
incoming and outgoing radiation, it seems increasingly probable that small
changes such as those occasioned by increasing particle loads in the atmo-
sphere may produce very long-term meteorological effects.
Weather Modification
Man's knowledge of his ability to inadvertently modify his climate is
still at best fragmentary. Insufficient information is known about the
long-term build-up of air pollutants, their effects and their inter-
relationships. Substantial evidence of precipitation modification is
meager but significant. An example of increased precipitation from air
pollution appears to exist at La Porte, Indiana, some 30 miles downwind
from the heavy industrial complex in Chicago. Precipitation and thunder-
storm activity have increased significantly since 1925, and precipitation
peaks have coincided with peaks in steel production in the Chicago area.—'
^i(\ 1
Hobbs, et al.,— have reported measurements of the concentrations of cloud
condensation nuclei in Washington State resulting from pulp and paper mills
and other industrial sources. Their study of the precipitation and stream
flow records in Washington State for the past 40 years revealed that in
34
-------
recent years there have been significant increases in precipitation in
several areas in the vicinity of large industrial sources of cloud-
condensation nuclei.
Excessive dustiness in the atmosphere can also reduce rainfall under cer-
tain conditions when overseeding occurs. This happens when many small
droplets, formed by condensation, do not fall to earth if there is insuf-
ficient moisture available to continue the droplet growth by condensation.
Consequently, what would have fallen as rainfall stays in the form of clouds.
A case in point of this reduction in rainfall has occurred in the sugar pro-
ducing areas in Queensland, Australia. During the cane harvesting season,
the common practice is to burn off the cane leaf before cutting and harvest-
ing. This results in fires over extensive areas and large palls of smoke.
The fine smoke particles have modified the cloud formation and hindered
the rainfall process. A reduction of up to 25% in the rainfall has occurred
downwind of these areas, but there is no such effect in neighboring areas
unaffected by the smoke plume.^Z/ Similar effects have been reported in
Puerto Rico, Africa, and Hawaii.58,597 Schaefer has also reported on ob-
servations of changes in micro-physics of clouds in the vicinity of large
cities during airplane flights through convective clouds .5_8/ His observa-
tions suggest that such clouds contain an increasing number of cloud drop-
lets. Schaefer's investigations of snow and rainstorm patterns in New York
State indicate that submicroscopic particulates from manmade pollution may
be initiating and controlling precipitation in a primary manner, rather
than being involved in the secondary process wherein precipitation elements
coming from natural mechanisms serve to remove the particles by diffusion,
collision, and similar scavenging processes.
In summary, available data indicate that man has detectably changed the
constitution of the atmosphere; in some respects globally, in another at
least on a subcontinental scale, in some cases locally. At the same time,
there have been variations of climate. Our present knowledge of atmo-
spheric processes suggests that these changes are what would be expected to
follow from man's interference with the atmosphere. But they are also
quite compatible with what we know of the statistics appropriate to an
atmosphere of undisturbed constitution and they are also compatible with
the possible behavior of a dynamical system as complex as the atmosphere.
35
-------
SECTION V
SOURCES OF FINE PARTICULATE EMISSIONS
NATURE OF THE PARTICULATE POLLUTION PROBLEM
Atmospheric particulatematter can be classified as primary--introduced
into the atmosphere in particulate form, or secondary--formed in the atmo-
sphere by chemical and physical processes. Atmospheric particulate matter
originates from natural causes such as the sea, volcanoes, and the soil
and from man-made sources such as industrial processes and internal com-
bustion engines. In order to cast the particulate pollution problem in
the proper perspective, it is helpful to identify the relative contribu-
tions of primary particulates from natural and man-made sources.
Tables 6, 7, and 8 summarize estimates of primary particulate emissions
from both natural and man-made sources. These estimates were prepared
by MRI as part of the work performed on a systems study of particulate
pollution.60-62/ Specific details of the procedures used to obtain the
estimates are presented in Ref. 60. Estimated emissions from various in-
dustrial operations are shown in Table 6. The emissions were computed in
most cases by means of the following equation:
(P) (ef) (1 - CcCt)
2,000
where,
E is the emissions rate, tons/year
ef is the emission factor (uncontrolled), pounds/ton
P is the production rate, tons/year
Cc is the average efficiency of control equipment
Ct is the percentage of the production capacity on which control
equipment has been installed
36
-------
Table 6. MAJOR INDUSTRIAL SOURCES OF PARTICULATE POLLUTION6.0./
Emissions
Source (tons/year)
1. Fuel combustion 5,953,000
2. Crushed stone, sand and gravel 4,600,000
3. Agricultural operations (grain elevators, feed mills
and cotton gins) 1,817,000
4. Iron and steel 1,442,000
5. Cement 934,000
6. Forest products 580,000
7. Lime 573,000
8. Clay 467,000
9. Primary nonferrous (copper, aluminum, zinc and lead) 476,000
10. Fertilizer and phosphate rock 328,000
11. Asphalt (batch plants and roofing) 218,000
12. Ferroalloy 160,000
13. Iron foundries 143,000
14. Secondary nonferrous (copper, aluminum, zinc and lead) 127,000
15. Coal cleaning 94,000
16. Carbon black 93,000
17- Petroleum 45,000
18. Acids (sulfuric and phosphoric) 16,000
Total 18,056,000
37
-------
Table 7. NONINDUSTRIAL SOURCES OF PARTICULATE POLLUTION^/
Source
A. Natural dusts
B. Forest fires
1. Wildfire
2. Controlled fire
(a) Slash burning
(b) Accumulated litter
3. Agricultural burning
C. Transpor tat ion
1. Motor vehicles
(a) Gasoline
(b) Diesel
2. Aircraft
3. Railroads
4. Water transport
5. Nonhighway use
(a) Agriculture
(b) Commercial
(c) Construction
(d) Other
D. Incineration
1. Municipal incineration
2. On-site incineration
3. Wigwam burners (excluding
forest products disposal)
4. Open dump
E. Other minor sources
1. Ruber from tires
2. Cigarette smoke
3. Cosmic dust
4. Aerosols from spray cans •
5. Ocean salt spray
Total
Emissions
(tons/year)
37,000,000
6,000,000
11,000,000
2,400,000
420,000
260,000
30,000
220,000
150,000
79,000
12,000
3,000
26,000
98,000
185,000
35,000
613,000
300,000
230,000
24,000
390,000
340,000
63,000,000
56,400,000
1,200,000
931,000
1,284,000
122,815,000
38
-------
Table 8. ALL MAJOR SOURCES OF PARTICULATE POLLUTION
Source
Natural dusts
Forest fires
Major stationary industrial sources
Transportation
Incineration
Other sources
Total
Emissions
(tons/year)
63,000,000
56,400,000
18,056,000
1,200,000
931,000
1,284,000
140,874,000
Percent by
Weight
44.7
40.1
12.9
0.8
0.7
0.8
100.0
39
-------
Production data for 1968 and application of control information* from 1969-
1970 were used in the calculations. Table 7 lists nonindustrial sources of
particulate pollutants with their corresponding estimated emissions, while
Table 8 presents a comparison of .particulate emissions from industrial and
nonindustrial sources.
As shown in Tables 7 and 8, the largest sources of particulates are natural
dusts and forest fires. These sources account for an estimated 85% of the
national atmospheric primary particulate loading on a mass basis, and are
a substantial portion of background levels. Emissions from these sources
are essentially beyond the scope of present air pollution control methodology.
Furthermore, their effect on the population is less than for man-made sources
because of the concentration of the latter in urban areas.
While the emission figures presented in Table 6 provide some indication of
the relative impact of individual industrial operations on the particulate
pollution burden, total mass emission data do not clearly portray the really
important and serious aspect of particulate air pollution. Not reflected
in Table 6 is the relative contribution that the emissions make to the long-
lived suspended particulate levels in the atmosphere. As was indicated in
Chapter 2, it is the long-lived particles (i.e., fine particulates) that
contribute to all the major adverse aspects of particulate air pollution.
PRINCIPAL SOURCES OF FINE PARTICULATES
As part of the systems study on particulate pollution, MRI estimated the
magnitude of the fine particle burden emanating from various particulate
pollutant sources .xL/ In this work, the fine particle size range was de-
fined as 0.01-2 um. In making the estimates of fine particle emissions,
the best data currently available on particle size distributions before
and after control devices, fractional efficiency curves for control devices,
and the degree of application of control equipment on specific sources was
used to estimate the present level of fine particle emissions from particu-
late pollution sources.
The method used to calculate the quantity of fine particle emissions is
similar to that used in calculating total mass emission in Table 6. The
same production figures and emission factors were used. However, to calcu-
late the quantity of fine particles (< 2 um) emitted it was necessary to
use the following additional factors:
* Application of control is defined as that fraction of the total produc-
tion which has particulate pollution controls.
40
-------
a. Particle-size distributions for particulates emitted by uncontrolled
sources.
b. Percent application of control on specific sources,with a breakdown
of this percent application of control into the percent application of
each type of control device.
c. Fractional efficiency characteristics of each type of control device.
Specific details of the calculations are given in Ref. 61.
Table 9 summarizes the estimates of fine particle emissions on the basis
of mass emitted in specific particle size ranges. Because of the lack of
adequate data on many specific process sources, the emission figures given
in Table 9 are not complete for some industrial categories. Also, fine
particle emissions for several sources (e.g., crushed stone operations)
could not be calculated for each size range shown in Table 9, and fine
particle emissions for potentially significant industrial categories, such
as primary nonferrous metallurgy and clay products, could not be estimated
because of incomplete data. Fine particle emissions from the industrial
sources listed in Table 9 are estimated to be at least 4 x 1C)6 tons/year.
This represents approximately 20-25% of the total mass emissions (Table 6)
from these sources. In view of the limitations of available data, the
figure of 4 x 10^ tons/year of fine particulate emissions from industrial
sources probably understates the fine particle burden emanating from those
sources.
To indicate the relative magnitude of the fine particle problem from
stationary sources, fine particle emissions were also estimated for mobile
sources. Particulate emissions from mobile sources were assumed to be un-
controlled and to consist of all < 2 jam particulates. Table 10 summarizes
the estimate of fine particle emissions from these sources. Fine particle
emissions from natural dusts, forest fires, and agricultural burning could
not be estimated because of lack of reliable particle size data..^JV
Summary of Fine Particle Emissions
Fine particle emissions (primary particulates) from man's activity are con-
servatively estimated to be over 5 x 10" tons/year of which about 75% re-
sults from stationary industrial operations. Furthermore, these emissions
compose about 257» of the total mass of particulates emitted from industrial
sources. Stationary combustion and metallurgical operations account for
about 50% of the quantity of fine particulates emitted from stationary
sources.
41
-------
Table 9. FINE PARTICLE EMISSIONS FROM INDUSTRIAL SOURCES^!/
(Mass Basis, 103 Tons/Year)
•IS
Source
1. Stationary combustion
A. Coal
1. Electric utility
a. Pulverized
b . Stoker
c . Cyclone
2. Industrial
a. Pulverized
b . Stoker
c. Cyclone
B. Fuel oil
1. Electric utility and
industrial
C. Natural gas and LPG
1. Electric utility and
industrial
Fine Particle Size Ranges (urn)
1-3 0.5-1.0 0.1-0.5 0.05-0.1 0.01-0.05
591.6 178.6 99.2 2.9
21.7 5.7 2.2
48.2 13.9 7.3 0.2
Total from electric utility
15.1 0.5
56.6 7.3 2.1
13.5 6.7 4.1 0.1
Total from industrial coal
126.9
97.2
Total from fuel oil and gas
Total from fuel combustion
Total
872.3
29.6
69.6
971.5
15.6
66.0
24.4
106.0
126.9
97.2
224.1
1,301.6
-------
Table 9. (Continued) FINE PARTICLE EMISSIONS FROM INDUSTRIAL SOURCES^!/
Fine Particle Size Ranges (urn)
2.
3.
4.
5.
6.
7.
Source
Crushed stone
Iron and steel
A. Sinter machines
B. Open hearth furnace
C. Basic oxygen furnace
D. Electric arc furnace
Kraft pulp mills
A. Bark boilers
B. Recovery furnace
C. Lime kiln
Cement plants, rotary kilns
Hot-mix asphalt plants
A. Rotary dryer
B. Vent line
Ferroalloys
A. Electric furnace
B. Blast furnace
1-3
868.0
3.8
60-80
0.9
3.8
49.1
95.3
1.6
130.8
96.4
14.4
18.4
0.8
0.5-1.0
1.2
20-56
18.9
2.5
Total from
11.9
78.0
0.2
Total from
32.7
36.3
1.7
Total from
27.8
0.1-0.5 0.05-0.1
0.6
8.5-234 0.1-22
153.7 1.0
5.2 1.3
iron and steel
6.5 0.3
74.7 1.4
kraft pulp mills
13.5
21.5
0.2
hot-mix asphalt plants
81.1 17.3
0.01-0.05 Total
868.0
5.6
0-3.6 108-376
174.5
1.5 14.3
302.4-570.4
67.8
249.4
1.8
319.0
177.0
154.2
16.3
170.5
7.7 152.3
0.8
Total from ferroalloys
153.1
-------
Table 9. (Concluded) FINE PARTICLE EMISSIONS FROM INDUSTRIAL SOURCES6.!/
8.
9.
10.
11.
12.
13.
14.
15.
16.
Source
Lime plants
A. Rotary kilns
B. Crushing and screening
Secondary nonferrous metallurgy
Carbon black
Coal preparation plants, thermal
dryer
•
Petroleum FC.C units
Municipal incinerators
Fertilizer, granulators and dryers
Iron foundries, cupolas
Acids
A. Sulfuric
B. Phosphoric (thermal)
1-3
40.6
25-99
127.0
93.0
63.5
45.0
10.4
7.1
6.8
2.7
1.0
Fine Particle Size Ranges (jim)
0.5-1.0 0.1-0.5 0.05-0.1 0.01-0.05 Total
18.8 23.6 3.0 1.8 87.8
25.0
Total from lime plants 113.0
127.0
93.0
63.5
45.0
6.7 11.5 3.5 4.3 36.4
3.5 3.1 13.7
2.4 3.1 0.4 0.4 13.1
2.7
1.0
Total from acids 3.7
Total from major industrial sources 3,800-4,142
Note: Potentially significant sources not evaluated because of lack of sufficient data: (1) operations re-
lated to agriculture, (2) primary nonferrous metallurgy, (3) clay products, (4) food processing
operations, and (5) fiberglass manufacture.
-------
Table 10. FINE PARTICLE EMISSIONS FROM MOBILE SOURCES
Emissions
Source (Tons/Year)
1. Motor vehicles
a. Gasoline 420,000
b. Diesel 260,000
2. Aircraft 30,000
3. Railroads 220,000
4. Water transport 150,000
5. Nonhighway use
a. Agriculture 79,000
b. Commercial . 12,000
c. Construction 3,000
d. Other 26.000
Total 1,200,000
45
-------
PRIORITY LIST FOR SOURCES OF FINE PARTICLE EMISSIONS
A complete description of the importance of a given source of fine partic-
ulate emissions requires not only the quantification of the amount of fine
particulate emitted, but also the identification of the chemical composi-
tion of the individual particle size fractions. Attention must be directed
to the chemical nature of the particles as it relates to toxicity or
hazardous aspects. For example, power plant flyash produced by the com-
bustion of pulverized coal consists of a diverse mixture of metal oxides
and silica while cement kilns produce little in the way of potentially
hazardous or toxic particulates. Other sources produce different mixtures
of substances. Incorporation of considerations of both quantity of fine
particulate emissions and chemical composition of particles can lead to a
priority ranking of sources of fine particulate emissions.
Data on the chemical composition of various particle size fractions emitted
from industrial sources are sketchy at best. However, the limited informa-
tion available indicates that many of the potentially hazardous materials
(e.g., trace metals) are generally in the micron to submicron particle
size range. While lack of detailed data hinders the clear-cut identifica-
tion of the most important sources of fine particulate pollution, available
information can be used to develop a relative priority list of emission
sources. By using available data on the mass of fine particles emitted
by sources, the amount and type of potentially hazardous pollutant emitted
by sources, and the general location of sources (i.e., urban or rural), a
general profile of the "adverse or negative" characteristics of the partic-
culate pollutants emitted by major industrial sources of fine particles can
be developed. Table 11 presents such a profile.
The data on the quantity of fine particles emitted from sources (column 1,
Table 11) were taken from Ref. 61, while the information on the amount and
type of potentially hazardous pollutant emitted was obtained from Ref. 63.
In analyzing the data from Refs. 61 and 63, there appeared to be some dif-
ference in the definition of particulate. Reference 61 considered only
primary particulates with no consideration of secondary particulates formed
by condensation or reaction in the atmosphere. Reference 63 apparently in-
cludes particulates formed by condensation.
Since some of the potentially hazardous pollutants may be emitted from a
source as a vapor which subsequently condenses in the atmosphere, some
sources deemed important in Ref. 63 were not considered in Ref. 61. This
accounts for some of the information gaps in Table 11. In addition, the
mass of fine particles emitted from some sources listed in Table 11 could
not be determined because of lack of sufficient data on the particle size
distributions of the emitted particulates.
46
-------
Table 11. PROFILE OF THE CHARACTERISTICS OF PARTICULATE POLLUTANTS
EMITTED BY VARIOUS INDUSTRIAL SOURCES
Industry and/or
Source
I. Stationary combustion
A. Coal
n.. Electric utility,
a. Pulverized
Mass of Fine
Particles Emitted
(103 tons/year)
b. Stoker
c. Cyclone
I. Industrial
a. Pulverized
b. Stoker
c. Cyclone
3. Commercial and
residential
872.3
29.6
69.6
15.6
66.0
24.4
Amount of Potentially
Hazardous Pollutant
Emitted (tons/year)
51,471
5,994
1,776
3,783
13,237
1,891
657
Type of Potentially
Hazardous Particulate
Pollutant Emitted
General Plant
Location
Inorganic/metal oxides, Urban, rural
fluorides, polyorganics,
As, Ba, Be, B, Cr, Cu,
Pb, Mn, Hg, Ni, Se, Sn,
V, Zn
As, Ba, Be, B, Cr, Cu, Urban, rural
fluorides, Pb, Mn, Hg,
Ni, POM, Se, Sn, V, Zn
As, Ba, Be, B, Cr, Cu, Urban, rural
fluorides, Pb, Mn, Hg,
Ni, POM, Se, Sn, V, Zn
As, Ba, Be, B, Cr, Cu, Urban
fluorides, Pb, Mn, Hg,
Ni, POM, Se, Sn, V, Zn
As, Ba, Be, B, Cr, Cu, Urban
fluorides, Pb, Mn, Hg,
Ni, POM, Se, Sn, V, Zn
As, Ba, Be, B, Cr, Cu, Urban
fluoride, Pb, Mn, Hg,
Ni, POM, Se, Sn, V, Zn
As, Ba, Be, B, Cr, Cu, Urban, rural
fluoride, Pb, Mn, Hg,
Ni, POM, Se, Sn, V, Zn
-------
Table 11 (Continued). PROFILE OF THE CHARACTERISTICS OF PARTICULATE POLLUTANTS
EMITTED BY VARIOUS INDUSTRIAL SOURCES
00
Mass of Fine Amount of Potentially
Industry and/or Particles Emitted Hazardous Pollutant
Source (103 tons/year) Emitted (tons/year)
B. Fuel oil
1. Electric utility
and industrial 126.9 28,326
2. Commercial and
residential - 44,063
C. Natural gas
1. Electric utility
and industrial 97.2 20,204
2. Commercial and
residential - 10,065
II. Crushed stone 868.0
III. Iron and steel
A. Sinter machines 5.6
B. Open hearth furnace 376.0 68,227
Type of Potentially
Hazardous Particulate
Pollutant Emitted
Inorganic/metal oxides,
polyorganics, As, Ba,
Be, Cr, Cu, Pb, Mn,
Hg, Ni, Se, V, Zn
Ba, Be, Cr, Cu, Pb, Mn,
Hg, Ni, POM, Se, Sn,
V, Zn
.
-
-
Metal oxides, alkalis
Ba, Pb, Mn, Hg, Sn, V,
General Plant
Location
Urban, rural
Urban, rural
Urban, rural
Urban, rural
Predominantly
rural
Urban
Urban
C. Blast furnace
D. BOF
E. Electric arc furnace
F. Metallurgical coke
175.0
14.3
7,215
2,057
12,508
43,380
Zn oxides, fluorides,
POM
As, Cd, Mn, Hg, Ni, V,
Zn oxides, fluorides,
POM
Ba, fluorides, Mn, Hg,
POM, V, Zn
Ba, Mn, Hg, Zn
POM
Urban
Urban
Urban
Urban
-------
Table 11 (Continued). PROFILE OF THE CHARACTERISTICS OF PARTICULATE POLLUTANTS
EMITTED BY VARIOUS INDUSTRIAL SOURCES
IV.
V.
VI.
VII.
VIII.
IX.
X.
Industry and/or
Source
Kraft pulp mills
A. Bark boiler
B. Recovery furnace
C. Lime kiln
Cement plants,
rotary kilns
Hot-mix asphalt plants
A. Rotary dryer
Ferroalloys
A. Electric furnace
B. Blast furnace
Lime plants
A. Rotary kilns
Municipal incinerators
Iron foundry cupolas.
Mass of Fine
Particles Emitted
(103 tons/year)
67.8
249.4
1.8
177.0
154.2
152.3
0.8
87.8
36.4
13.1
Amount of Potentially Type of Potentially
Hazardous Pollutant Hazardous Particulate
Emitted (tons/year) Pollutant Emitted
- — f
15 Cr, Hg, POM
-
270 Fluorides
2,800 POM
4,382 Mn, Ni, POM, V, Zn
4,104 Mn, Ni, Zn, 0, POM
-
34,307 As, Cd, Cu, Pb, Hg,
POM, Se, Zn
6,151 As, Ba, Be, Pb, Mn,
General Plant
Location
Urban, rural
Urban, rural
Urban, rural
Primarily
rural
Urban, rural
Urban
Urban
_
Urban
Urban
XI. Primary copper
A. Roasting
B. Reverberatory
furnace
4,373
1,885
Hg, Ni, V, Zn oxides,
POM, fluorides
As, Cd, Cu, fluoride,
Pb, Se
As, Cd, Cu, fluoride,
Pb, Se
Rural
Rural
-------
Table 11 (Concluded). PROFILE OF THE CHARACTERISTICS OF PARTICULATE POLLUTANTS
EMITTED BY VARIOUS INDUSTRIAL SOURCES
Ui
O
Industry and/or
Source
C. Converters
D. Material handling
Mass of Fine
Particles Emitted
(103 tons/year)
XII. Primary zinc
A. Roasting
B. Sintering
C. Distillation
XIII. Primary lead
A. Sintering
B. Blast furnace
XIV. Primary aluminum
A. Reduction cells
XV. Iron ore pellet plant
XVI. Asphalt roofing materials
XVII. Secondary copper
XVIII. Secondary lead
XIX. Secondary zinc
XX. Structural clay products
Amount of Potentially
Hazardous Pollutant
Emitted (tons/year)
5,591
1,235
34,187
14,044
4,676
1,016
275
16,230
18,200
23,230
1,036
2,020
3,840
9,720
Type of Potentially
Hazardous Particulate
Pollutant Emitted
As, Cd, Cu, fluoride,
Pb, Se
As, Cd, Cu, fluoride,
Pb, Se
As, Cd, Cu, fluorides,
Pb, Se, Zn
Cd, fluorides, Pb, Zn
Cd, fluorides, Pb, Zn
As, Cd, fluorides, Pb,
Se
As, Cd, fluorides, Pb, Se
Fluorides (gas and solid)
Fluorides
POM
Cd, Cu, Pb, Zn, POM
Pb
Zn
Fluorides
General Plant
Location
Rural
Rural
Urban, rural
Urban, rural
Rural
Urban
Urban
Urban
Urban
Urban, rural
-------
In developing a priority ranking of the sources presented in Table 11,
equal weight was assigned to each of the adverse facets. Using this ap-
proach, the priority ranking in Table 12 was developed. Although quanti-
tative detail is less than desired in Tables 11 and 12, the tables do sug-
gest what should be the objective of initial efforts to control fine
particulate emissions. The main thrust of initial efforts should be
focused on the reduction of the mass of fine particulate emissions with
emphasis placed on control of those sources which emit potentially
hazardous materials and are located in or near urban areas. Potential
approaches to the regulation of fine particle emissions are discussed in
the next chapter.
51
-------
Table 12. PRIORITY LIST FOR SOURCES OF FINE PARTICLE EMISSIONS
Group I (High priority)
1. Stationary combustion (all fuel types)
a. Electric utility
b. Industrial
2. Iron and steel plants
a. Open hearth furnaces
b. EOF furnaces
c. Electric arc furnaces
d. Metallurgical coke ovens
3. Municipal incinerators
4. Ferroalloy plants
a. Electric furnace
b. Blast furnace
5. Primary nonferrous metallurgy
a. Zinc roasting, sintering and distillation
b. Copper roasting and converting
c. Aluminum reduction cells
Group II (Medium priority)
1. Hot-mix asphalt plant
2. Iron foundry cupolas
3. Asphalt roofing materials
a. Asphalt blowing
4. Secondary copper, lead and zinc
Group III (Low priority)
1. Iron ore pellet plants
2. Structural clay products
3. Cement and lime plants
4. Kraft pulp mills
5. Crushed stone
52
-------
SECTION VI
APPROACHES FOR REGULATING FINE PARTICLE EMISSIONS
INTRODUCTION
The goal(s) of the control program for fine particulates will be the main
factor in the selection of approaches for regulating the emission of fine
particulates. If the aerosol burden in the urban atmosphere is to be
reduced, a premium must be placed on the collection of particles smaller
than 5 um. If improved visibility is the goal of fine particulate control
programs, attention must be focused on particle collection in the 0.1-1.0 um
range. Concern for adverse health effects should direct attention to the
control of the0.01-7um size range, and to potentially hazardous consti-
tuents (e.g., lead, cadmium, vanadium, and particulate polycyclic organic
matter) of the effluent stream.
Two approaches, distinguished by their initial focal point, can be used
to regulate emissions of fine particulates from stationary sources:
1. Ambient air quality standards expressed in terms of fine particulates,
and
2. Direct reduction of emissions from specific sources.
The ambient air quality approach requires as an initial step a statement
specifying the desired objective in terms of atmospheric concentration,
e.g., so many micrograms per cubic meter of total particulate plus a
limitation on the contribution to total weight accounted for by particles
of given diameter(s). The second step requires determination by measure-
ment of existing ambient concentrations of particles of given diameter (s).
From the information developed in these two steps, a ratio can be derived
describing the fraction by which present emissions must be reduced to
achieve the ambient standard. The final step requires development of
data showing existing rates of emission from sources and the formulation
53
-------
of emission reduction plans derived by application of the ratio obtained
in Step 3.
The second approach focuses attention on the direct reduction of emissions
from specific sources, without initial recourse to ambient air quality.
This approach is based on the premise that, to the extent feasible, fine
particulates should be prevented from entering the atmosphere without
initial consideration of the ambient air quality. Measurements of changes
in ambient air quality before and after the initiation of control activi-
ties provide an indication of the overall effectiveness of the control
effort.
Before an approach based on ambient air quality could be utilized effec-
tively, data on (1) existing ambient air quality as a function of particle
diameter(s) and (2) existing rates of emission must be available from
numerous regions throughout the country. Acquisition of such a data base
would require the expenditure of a great deal of time and money. A signi-
ficant time lag would also occur before efforts to control fine particu-
late emissions could commence if this approach were selected.
If it were decided to initiate efforts to control fine particulate emis-
sions in the near future, direct reduction of emissions from specific
sources could be started immediately. Although the existing data base
on emissions of fine particulates from stationary sources is lacking in
detail and precision, it could be used to formulate initial emission
reduction programs. As more and improved information on emission rates
becomes available, the emission reduction programs could be refined.
The effectiveness of any program for the control of fine particulate emis-
sions must ultimately be judged by improvements in ambient air quality.
Because both atmospheric conversion processes (chemical and physical)
and sedimentation contribute to the fine particulate burden in the atmos-
phere, it may be quite difficult to relate atmospheric fine particulate
levels to control activities on specific sources. The City of Los Angeles
is a good example of this situation. In Los Angeles about 35% of the ambient
particulate burden is formed in the atmosphere from gaseous pollutants
(see Table 13) .^J In cities like Los Angeles, a program based on ambient
air quality might not be very effective. Atmospheric processes might
also make interpretation of the overall effectiveness of direct regula-
tion of sources difficult in some cases. For example, in Los Angeles
reductions in emissions of fine particulates from stationary sources
would not result in a proportionate decrease in ambient fine particulate
concentration. While the assessment of the overall effectiveness of direct
regulation of sources may be difficult in some cases, the total burden of
54
-------
Table 13. ESTIMATED CONCENTRATIONS OF LOS ANGELES
AEROSOL PARTICLES BY SOURCE**!/
(Annual Average)
Mass Concentration
Source
Natural background
Primary ............................. 14-26
Dust rise by wind 8-20
Na+ 3
Sea salt cl_ 3
Spores, pollen, etc. r Unknown
Secondary ............................ 4-7
Vegetation (organic vapors) 3- 6
Biological (soil bacterial action,
decay of organics) -NH3, NOX, S . . . 0.7
36
Man-made
Motor vehicles
Organic solvent usage
Petroleum
Aircraft
Combustion of fuels
Other
Reactive hydrocarbon vapors
N03 *
S04
15
6
1.3
4
5
5
11
13.5
14.4
39
Total Accounted For 93-108
Measured Total 119
55
-------
fine particulates entering the atmosphere would nonetheless be reduced by
direct limitation of source emissions. In the case of potentially hazard-
ous pollutants, which in many instances are really fine particulates,
direct regulation of source emissions is the only viable approach. The
Environmental Protection Agency has already set a precedent in this area
by adopting emission standards for sources of mercury, beryllium and as-
bestos. The preceding discussion indicates that direct reduction of fine
particulates from specific sources is the most effective method to insure
reduction of the fine particulate burden emanating from man's activity.
In the development of programs for the direct reduction of emissions from
specific sources of fine particulates, a decision must be made regarding
whether to limit regulation efforts to primary particulates or whether
both primary and secondary particulates emitted from sources will be
subjected to regulation. Primary particulates are defined as material
that is emitted from the source or its exhaust stack in a particulate
form. Secondary particulate is defined as particulate matter formed
(primarily by physical processes, e.g., condensation) in the atmosphere
from source effluents shortly after emission from the source or its
exhaust stack. These secondary particulates are not the particulates
formed in the atmosphere from gases and existing particles. As noted in
Chapter 3, some of the potentially hazardous pollutants may be emitted
from specific sources as a vapor which subsequently condenses into a fine
particulate in the atmosphere shortly after exit from the stack or ex-
haust point. Although cause-effect relationships for health effects as-
sociated with fine particulate pollutants are not quantitative, efforts
to regulate fine particle emissions from stationary sources should be
focused on minimizing the impact of fine particulate pollutants on human
health. In general, this stipulation requires control of both primary
and secondary particulates in the nominal size range of 0.01-5 um.
The preceding comments serve to highlight some of the more important
facts that must be considered in the development of effective control
strategies for fine particulate emissions. Various approaches that might
be used to regulate fine particulate emissions from stationary sources
are discussed in the following sections of this chapter.
METHODS- FOR REGULATING FINE PARTICLE EMISSIONS
i
Because our understanding of the importance of fine particulate pollution
to man's health and welfare is in its infancy, it seems prudent to con-
sider a long-range regulation strategy based on the adoption and imple-
mentation of progressively more stringent regulations for the control of
56
-------
fine particulates. Each step toward more effective control of fine
particulates would be based upon improved knowledge of (1) the effects
of fine particulates on human health and welfare; (2) the technical and
economic impact of the more stringent regulations; and (3) the benefits
that would accrue to society from a further decrease in fine particulate
pollution. Three time-frames that might be considered are:
1. Regulations that can be proposed and implemented in the near term
(i.e., by 1975).
2. Regulations that can be proposed and implemented in an intermediate
term (i.e., 1980-1985).
j
3. Regulations that can be proposed and implemented in the long term
(i.e., by 1990).
Any method(s) selected for regulation of fine particulate emissions
should be chosen to facilitate interfacing with existing strategies for
the control of particulate pollutants. Since the effluent or emission
standard is the backbone of current control programs, interfacing fine
particle control efforts with existing programs would be facilitated by
the use of emission standards as the main tool in the regulation of fine
particulates. Other methods that could be used to prevent or regulate
fine particle emissions include: (1) tax incentives; (2) process modi-
fications; (3) substitution of ingredients or fuels; and (4) cessation of
processing operations that emit fine particles. The last alternative is
deemed untenable except in very special situations. Each of the other
methods is a viable approach, although methods (2) and (3) are probably
limited in the extent to which they can be used.
, t
As noted in the preceding paragraphs, a precedent exists for utilizing
emission standards as the main method for control of fine particulate
emissions in the near term. Current programs for the control of particu-
late pollutants are built around emission standards, and minimum disrupt-
tion in control agency activities would occur if similar methods were
used for fine particulates. In the intermediate to long term, the un-
availability of sufficiently high efficiency control equipment and adequate
source compliance monitoring methods and/or'instrumentation may limit the
utility of emission standards, especially if fine particulate or specific
components of fine particulate emissions must be controlled to a very high
degree. Alternative regulatory strategies based on tax Incentives, might
57
-------
be attractive in that event. Both positive* and negative"1" tax incentives
could be considered, but the negative incentive in the form of an effluent
tax is deemed more beneficial. The negative tax incentive, in the form
of an emission or effluent tax, has attracted some support recently. The
proposal for charges on sulfur dioxide emissions is an example.
BASIS FOR EMISSION STANDARDS FOR FINE PARTICULATES
Emission standards for the control of fine particulates could be based
on concentration of fine particulates in an effluent stream, collection
efficiency of control equipment, plume opacity, mass-emission rate of
fine particulates, and hazard potential of the particulates.
Standards based only on concentration of fine particulates in an effluent
stream have a very significant deficiency; namely, they generally ignore
the volume of gas emitted from a source. As a consequence, such stand-
ards would not control the total fine particulate pollutant discharge
from a specific source unless they incorporated considerations of the
size of sources along with a specification of the allowable fine particu-
late grain loading. An extensive data base relating particle size and
grain loading to source size would need to be developed before this
approach could be implemented.
The requirement of a specific collection efficiency in given particle
size ranges, and the installation of the best installed control"*"
device"1""*" on all sources in a specific category, are two examples of stand-
ards based on the collection efficiency of control equipment. Combined
with a limitation on the total mass emission for a specific source, such
an approach has merit. An emission standard combining those facets
would limit both the total mass emitted as well as the mass in the fine
particle size range. The lack of reliable data on the efficiency of
control equipment as a function of particle size (i.e., fractional
efficiency) might make it difficult to utilize the approach of designat-
ing the collection efficiency in specific size ranges. The alternative
of requiring the installation of the best installed control on sources
is somewhat analogous to extending new source performance standards to
all sources in a specific categoryi A standard utilizing this approach
* Positive tax incentives—include not-only direct payments or subsidies
but also reductions in property taxes, accelerated depreciation and
tax credits.
+ Negative tax incentives—this category includes effluent charges or
fees for the discharge of specific amounts of pollutants or other
taxes on specific sources of pollution.
++ Best installed control device is not necessarily the highest efficiency
device available, but rather the best that is generally being in-
stalled at present time.
58
-------
might be expressed in terms of a maximum allowable total mass emission
and a minimum allowable control device collection efficiency. Since a
standard based on the specification of control device collection effi-
ciency requires the same degree of control for large and small sources,
a distinct possibility of undue regulation of small sources and insuf-
ficient regulation of large sources might occur with this type of regula-
tion. However, with some adjustment to account for plant size, a standard
based on the installation of the best installed control devices on all
sources in a specific category is a potential vehicle for the regulation
of fine particulate emissions. Determination of the effectiveness of such
an emission standard would require extensive source testing.
Because the opacity of a plume is a function of the size of entrained
particulate matter, and particles in the size range 0.1-1.0 um have the
dominant effect on plume opacity characteristics, plume opacity regula-
tions provide a vehicle to control fine particle emissions. Since opacity
regulations are currently utilized by many regulatory agencies, no prob-
lems would be encountered in interfacing with current particulate control
programs. The trend in opacity regulations is toward 20% opacity*!!/--a
regulation of 10% or 570 opacity could be used as a more stringent opacity
regulation.
Mass-emission regulations provide a direct approach to the control of
fine particle emissions. Regulations based on mass emissions could be
formulated to control the quantity of material emitted in specific size
ranges, and could be based on either process-weight rate or on potential-
emission rate. Process weight regulations specify the maximum allowable
discharge rates for particulates in pounds per hour in relation to the
process weight in pounds per hour. Process weight is defined as the
total weight of all materials introduced into a process, except liquid
or gaseous fuels or combustion air, divided by the time in hours to com-
plete the process. Process-weight regulations generally are formulated
so that the absolute allowable discharge increases with increasing process
weight. However, the percentage of process material which is permitted
to be discharged decreases as the process weight increases.
Emission standards can also be formulated to specify maximum allowable
emission rates (Ib/hr) of fine particulate as a function of the fine
particulate emission potential of the source (Ib/hr) . The potential
emission rate is defined as the total weight rate at which fine particu-
late matter is, or in the absence of an air-cleaning device would be,
emitted from an air contamination source when such source is operated at
its maximum rated capacity.
59
-------
The potential-emission rate concept appears to be a direct and useful
basis for formulating standards to control the emission of fine particu-
late pollutants. This approach should result in a better understanding
of the main parameters that influence fine particulate emissions from
specific sources. Once the potential rate of emission of fine particu-
lates is established for various sources, allowable emission rates can
be set to tailor the control of fine particulate emissions to any
desired level.
Our analysis has indicated that the most promising bases for emission
standards for fine particulates are: (1) plume opacity; (2) minimum
collection efficiency for fine particulates; and (3) potential emission
rate concept. Each of these is discussed in more detail in the follow-
ing sections.
Plume Opacity Standards
The opacity of a particulate emission depends on the mass concentration
of the particles comprising the plume, their size distribution and physi-
cal properties such as density and refractive index. Other parameters
such as plume width and wavelength of incident light are also important
in determining opacity. It is the dependence of plume opacity on mass
concentration and particle size, especially in the 0.1-1.0 urn size range,
which makes it possible to control the emission of fine particulates by
controlling the opacity of the exhaust plume.
i
As previously noted, Ref. 65 indicates that the trend in plume opacity
regulations is toward 20% opacity. Standards based on 10% or 5% (i.e.,
essentially no visible emission) could be used to reduce the emission of
fine particulates. The use of opacity regulations to control fine par-
ticulate emissions involves some subtleties, and care will have to be
taken in order to develop useful regulatory programs. A brief review of
the mathematical expressions used to define plume opacity is given in
Appendix C, and this review delineates some of the important implications
regarding the use of opacity regulations.
Because of the strong dependence of plume opacity on factors such as the
optical properties of the particulate-matter, each source under scrutiny
must be described in detail before plume opacity has practical meaning.
This fact is illustrated by Figure 9.£§/ Figure 9 presents plume
opacity as a function of particle diameter and dust loading for particles
with a density of 2 g/cm^ and for 5 ft dia.. stacks with exit temperatures
of 300°F.
60
-------
100 r-
80 -
U
LLJ
Q_
•v
>-
h-
u
60 -
40 -
20 -
0.01
QE = 2 for dp > 0.14
QE = 3.26(dp)l/4<0.14
Stack Diameter = 5 FT
Particle Density = 2g/cc
Exit Temperarure = 300° F
Dust Loading
.25 Grains/SCF)
0.05 0.10 0.50 1.0
AVERAGE PARTICLE DIAMETER (MICRONS)
Figure 9. Plume opacity as a function of particle—'
diameter and dust loading
10.0
-------
A fairly stringent standard for grain loading is 0.05 grain/scf. From
Figure 9 it is noted that if the particulates are large enough that
the average is greater than 5 um or so, an opacity requirement of 2070 is
not strict compared to the 0.05 grain loading requirement. If the average
particle diameter is 1/2 um or so, a 20% opacity requirement would be
more strict. In order to accomplish the desired control of fine particu-
lates via the regulation of plume opacity, it may be necessary to develop
regulations that require a different plume opacity for various sources.
More refined information on particulate properties will be required in
order to develop effective strategies for control of fine particulates
via opacity regulations. A more comprehensive data base will also be
needed to develop acceptable source compliance monitoring techniques.
Requirements for source monitoring will be discussed further in Chapter 5.
Minimum Collection Efficiency for Fine Particulates
An emission standard which designates a minimum collection efficiency
for fine particulates in combination with a limitation of total emission
as a function of source size could serve as a vehicle for the control of
fine particulates. A collection efficiency standard alone is not a
viable approach because it would not directly limit the total mass of
fine particle emissions. The minimum collection efficiency could be
achieved by (1) specifying minimum collection efficiency in designated
particle size ranges, or (2) requiring the installation of the best
installed control device on all sources in a specific category.
An emission standard incorporating minimum control efficiency in specific
particle size ranges could be designed to provide not only control of
fine particulates in general but also control of specific chemical constitu-
ents of the fine particulate stream. The Bay Area Air Pollution Control
District (San Francisco, California) has entertained discussions regarding
the feasibility of modification of their Regulation 2 regarding particulate
emissions to incorporate minimum collection efficiency in designated par-
ticle size ranges. The suggested modifications would supplement Regulation
2 with a new type of performance standard based on minimum control effi-
ciency in particle size ranges as follows: 227
Efficiency Particle
(weight 7,) ' Size Range
99 10 urn or larger
97 3-10 um
92 0-3
62
-------
The above performance standard is but one of an almost unlimited number
of standards that could be formulated using the basic concept of desig-
nated efficiencies as a function of particle size. Tailoring of a regu-
lation for a specific source category is an obvious possibility with this
type of standard. A considerable amount of data on particle size and
other effluent characteristics would have to be developed to show what
collection efficiencies are practical in various size ranges below about
2 um.
An emission standard utilizing the requirement of installation of the
best installed control technology on all sources in a specific category
would in effect specify the minimum allowable collection efficiency on
a total mass basis. Since the collection efficiency on a total mass
basis is the summation of efficiencies for all particle sizes, control
of particles in specific size ranges would be achieved by this approach.
In terms of existing technology for control systems, this approach would
also represent the current practical limit of an emission standard based
on designation of minimum control efficiency in specific particle size
ranges. Determination of the effectiveness of a standard requiring the
installation of the best installed control technology would require
extensive source-testing in order to develop accurate data on control
device fractional efficiencies for a spectrum of sources of fine particu-
late pollutants.
Mass-Emission Regulations
The use of emission standards which limit the quantity of fine partic-
ulate emitted from specific sources is the most direct approach to
regulating fine particulate emissions. Process-weight rate, potential
emission rate, or parameters that reflect source size (e.g., heat input
in 106 Btu/hr) could form the basis for this type of regulation. In
the latter category, the New Mexico Environmental Improvement Board has
taken a pioneering role and has adopted a regulation for coal-burning
equipment that is aimed at regulating fine particulate emissions from
that type of source. The New Mexico regulation reads as follows:
"After December 31, 1974, no person owning or operating coal-
burning equipment shall permit, cause, suffer, or allow:
"1. Particulate matter emissions to the atmosphere in excess
of 0.05 Ib/million British thermal units of heat input; or
"2. Fine particulate matter emissions of less than 2 um equi-
valent aerodynamic diameter and unit density to the atmosphere
in excess of 0.02 Ib/million British thermal units of heat input.
63
-------
"Fine parttculate matter emissions for this regulation shall be
collected and measured at stack conditions and in such a manner
that no condensation of gaseous material is included with the
sample."
Pro cess-weight rate regulations are currently used as an integral part
of many air pollution control programs. Historically, process-weight
rate regulations have been developed from data obtained from well-con-
trolled and well-operated plants. Source testing data acquired from such
plants form the basis for determining the degree of control that is tech-
nically and economically feasible. To utilize this type of regulation
to reduce the emission of fine particulates, it will be necessary to have
accurate data regarding:
1. Quantity of particulate emitted from various well-controlled sources;
2. Particle size distribution of the particulate emitted; and
3. Fractional efficiency characteristics of various control devices.
Currently available data in the above areas are inadequate for the formu-
lation of effective emission standards.
With regard to the regulation of fine particulate emissions, process-
weight rate may not be a good basis for an emissions standard because
the quantity of fine particulate emitted from a source may not be directly
related to process weight. Available data suggest that factors such as
nature of ingredients, processing conditions, and equipment and hooding
and ducting configurations have a much stronger influence on fine particu-
late emissions than does the process-weight.
The concept of potential-emission rate offers an alternate basis for
emission standards to regulate fine particulate pollution. Regulations
based on this concept establish emission limits which vary with the pol-
lution potential of the source, e.g., limitation of mass rate of emis-
sions in Ib/hr as a function of potential-emission rate, also Ib/hr.
This type of regulation appears to be adaptable to sources of fine par-
ticulate pollutants. Three variations of the potential emission rate
concept are depicted in Figure 10.
1. Specification of the minimum efficiency (on a mass basis) of collec-
tion of the potential emission of less than 2 \ua particulates. Curves
A, B, C, and D, in Figure 10, represent four of the many collection effi-
ciency regulations that could be utilized. Curve C, 99 wt % collection
-------
ON
Ln
90wt%(A)
95 wt % (B)
1,000 —
99«t%(C)
99.9 «t % (D)
100 1.000
POTENTIAL EMISSION RATE OF LESS THAN 2 MICRON PARTICULATE (Ib/hi)
10.000
100.000
Figure 10. Representative emission standards based on potential emission rate
-------
efficiency for less than 2 urn particulate, represents the probable up-
per limit of efficiency for currently available control equipment. This
variation actually corresponds to the minimum collection efficiency con-
cept previously discussed with potential-emission rate as the basis.
2. Direct limitation of the quantity (i.e., pounds per hour) of less
than 2 jam particulate that can be emitted from any source without con-
sideration of the emission potential of any specific source. Curves E
(100 Ib/hr allowable emission rate) and F (10 Ib/hr allowable emission
rate) are representative of this type of regulation.
3. Regulations based on a sliding scale of allowable emission as a
function of potential emission. The sliding scale is adjusted to force
sources with large emission potential to use higher efficiency collec-
tion methods than sources with smaller emission potentials. Curves G,
H, and I are representative of this form of regulation.
Potential-emission rates of fine particulates for specific sources could
be determined by sampling the uncontrolled effluent gases, and source
compliance could be determined by measuring control equipment efficiency.
Alternatively potential-emission rates could be assigned to sources
through use of pre-established emission factors for fine particulates.
Because an emission factor ideally represents the average measured emis-
sion rate from a number of similar installations, the use of such factors
is a logical and equitable substitute for determining potential-emission
rates for each individual source. Since currently available data are
inadequate, an extensive data base would need to be developed before this
concept could be implemented.
EMISSION STANDARDS SELECTED FOR EVALUATION
Our analysis of the alternative routes for emission standards for fine
particulates 'led to the following observations:
1. The use of standards involving plume opacity is the most practical
means for controlling fine particulate emissions in the near term.
Standards based on plume opacity would require the least amount of addi-
tional data acquisition and they would readily interface with existing
air pollution control programs.
2. An emission standard based on the requirement of the installation of
the best installed technology on all sources in a specific category
66
-------
could be implemented in the intermediate term. A standard of this type
would represent the attainment of the current practical limit of control
equipment performance.
3. A mass emission regulation based on the potential emission rate
concept is a very attractive approach for the long term. A regulation
of this type could be tailored to the control of specific sources or
specific pollutants.
The technical and economic implications of these types of emission
standards are analyzed in Chapters 5, 6, and 7-
67
-------
SECTION VII
TECHNOLOGICAL IMPLICATIONS OF FINE PARTICLE EMISSIONS REGULATIONS
INTRODUCTION
Implementation of any regulation(s) for the control of fine particulate
pollutants from stationary sources requires the availability of technology
in two distinct areas:
1. Control equipment, and
2. Compliance testing and monitoring.
In selecting a regulation for the control of fine particulates, it is
necessary that the regulation be realistic in the sense that control tech-
nology must be available to permit the requisite collection of fine par-
ticulates. In addition, methods must be available to ascertain whether
or not specific sources are in compliance with the regulation(s).
REQUIREMENTS FOR CONTROL EQUIPMENT EFFICIENCY
Regulations to control the emission of fine particulate pollutants will
require the use of high-efficiency control equipment. Depending upon the
type of regulation adopted, the performance capability of control systems
may -be stretched to the limit of existing technology. An analysis of the
collection efficiency required to meet stringent plume opacity regulations
provides an indication of the level of performance that will probably be
required. ,
The translation of an opacity regulation into terms ofL,the overall mass
efficiency required of control equipment necessitates the use of a rela-
tionship between plume opacity and various particulate and source char-
acteristics. Ensor and Pilat have 'developed a procedure to calculate
plume opacity from plume diameter, particle size distribution, particle
mass concentration, average particle density, and particle refractive
68
-------
index.(Also see Appendix C.) Equation(4) presents the relationship
developed by these authors to calculate the expected mass concentration
for various values of plume transmittance (or opacity), average particle
density, and plume diameter.
W = -K p/L In (I/I0) (4)
In Eq. (4),
W = Total particulate mass concentration (i.e., grain loading) at
stack conditions
p = Average particle density
L = Diameter of plume
I/I0 = Light transmittance
_ Specific particulate volume
Extinction coefficient
The assumptions and simplifications involved in the derivation of Eq. (4)
are discussed in Ref. 68. Although Eq. (4) is not exact, it can be
utilized to determine, with sufficient accuracy for the present purpose,
the total particulate mass concentration corresponding to various combina-
tions of p, L, K, and I/IO values.
The parameter K is a function of the particle size distribution, the re-
fractive index of the particulate, and the wavelength of incident light.
Reference 67 presents graphs of K vs the geometric mass mean particle radius
with the geometric standard deviation of the particle size distribution
and the refractive index of the particles as parameters. A wavelength of
light of 0.55 u was used as an average for visible light. This is approxi-
mately the wavelength of maximum sensitivity for the human eye. Equation
(4), combined with the graphs for the parameter K, and information on ef-
fluent properties from specific sources of particulate pollution presented
in Ref. 62, was used to develop the control efficiency vs plume opacity
data presented' in Table 14.
The selected values of plume opacity (20%, 107o and 570) correspond to
Ringlemann No. 1, No. 1/2, and "no visible emissions." The total mass ef-
ficiency required for a control device to reduce plume opacity to these
69
-------
Table 14. CONTROL DEVICE EFFICIENCY REQUIRED TO ACHIEVE VARIOUS PLUME OPACITY LEVELS
K f L
Source (cm3/m2) (g/cm3) (meters)
Coal-fired 0.64 2.0 2
power plant
5
10
Basic-oxygen 0.20 3.5 2
furnace
5
10
Cement plant — 1.1 3.0 2
rotary kiln
5
10
Asphalt plant— 10 2.6 2
rotary dryer
5
10
Grain Loading at Inlet to
Plume Opacity Control Devices (gi/Bcl)
(I/Ioi 70 Low Average High
20 1.0 3.0 6.0
10
5
20
10
5
20
10
5
20 2.0 5.0 10.0
10
5
20
10
5
20 1.0 6.4 17.0
10
5
20
10
5
20
10
5
20 10.0 30.0 70.0
10
5
20
10
5
20
10
5
W
Grain Loading at Exit
of Control Device
(Rr/scf)
0.09
0.037
0.02
0.036
0.017
0.008
0.018
0.0085
0.004
0.07
0.033
0.016
0.028
0.013
0.006
0.014
0.007
0.003
0.30
0.15
0.069
0.12
0.06
0.0275
0.06
0.03
0.0138
1.70
0.81
0.39
0.68
0.32
0.16
0.34
0.162
0.078
Total Mass
Efficiency Required
for Control Device (%)
Low
Inlet
91.0
96.3
98.0
96.4
98.3
99.2
98.2
99.1
99.6
96.5
98.3
99.2
98.6
99.3
99.7
99.3
99.7
99.9
70.0
85.0
93.0
88.0
94.0
97.0
94.0
97.0
98.7
83.0
91.9
96.1
93.2
96.8
98.4
96.6
98.4
99.2
Average
Inlet
97.0
98.8
99.5
98.8
99.4
99.7
99.5
99.7
99.9
98.6
99.3
99.7
99.4
99.7
99.9
99.7
99.9
99.94
95.5
97.8
99.0
98.1
99.0
99.6
99.1
99.5
99.8
94.3
97.3
98.7
97.7
98.9 •
99.5
98.9
99.5
99.7
High
Inlet
98.5
99.3
99.7
99.4
99.7
99.9
99.7
99.8
99.9
99.3
99.7
99.8
99.7
99.9
99.94
99.9
99.9
99.97
98.2
99.2
99.8
99.0
99.7
99.9
99.7
99.9
99.95
97.6
98.8
99.4
99.0
99.5
99.8
99.5
99.8
99.9
-------
values is shown in Table 14 for various values of the inlet grain loading
to the control device and plume diameter. The range of inlet grain load-
ings for specific uncontrolled sources was taken from data in Ref. 68.
The plume diameters were arbitrarily chosen to represent a spectrum of
diameters.
Inspection of the data in Table 14 indicates that a plume opacity limita-
tion of 20% (Ringlemann No. 1), which is the current trend in opacity
regulations, would require an efficiency ranging from 94-99.5% on a total
mass basis for sources with an average inlet grain loading. That degree
of collection efficiency represents control equipment of medium to high
efficiency. A plume opacity limitation of 10% would require a collection
efficiency of 97-99.9% for sources with an average inlet grain loading.
If a 5% plume opacity limitation were imposed, sources with an average
inlet grain loading would be required to control emissions to a level of
99-99.9% efficiency.
Compliance with a 10% or 57» plume opacity standard would require the use
of high-efficiency control systems which approach the limits of current
technology. The capability of current control equipment obviously places
a constraint on the degree of reduction of fine particulate emissions that
can be achieved by any type of emission standard in the near term. The
status of existing and emerging control technology for fine particulates
is discussed in more detail in the following sections.
CAPABILITY OF CONTROL TECHNOLOGY
The current capability of control equipment to collect fine particulates
is of immediate concern, while the potential of new control methods or
emerging technology is of importance with regard to regulations that might
be devised for the intermediate to long term. Appendix A presents a de-
tailed review of existing and emerging control technology, and only the
major points of the review are presented in the following sections.
Capability of Existing Control Equipment
Four basic types of control devices (electrostatic precipitators, wet
scrubbers, fabric filters, and afterburners) can currently be used to col-
lect fine particulates. Afterburners have limited utility since the par-
ticulate must be combustible in order to employ this type of control device.
The fabric filter is usually considered to offer the highest collection
efficiency; for the greatest number of sources. However, it is not ap-
plicable to some sources of particulate pollutants because of the nature
of the particulate, or the high cost involved in conditioning the gas
stream.
71
-------
The ability of control devices to collect participate matter, and especially
the fine particles, is commonly expressed in terms of their fractional ef-
ficiency. MRI has accumulated much of the available data on the fractional
efficiency of control devices. Analysis of these data (as explained in
Appendix A) has resulted in the generalized fractional efficiency graph
shown in Figure 11. The curves shown in Figure 11 indicate the efficiencies
that may be expected for the basic types of control devices as a function
of particle size. There are, however, several variations in design of
those control devices which is one reason for the broad ranges in ef-
ficiency and may be the reason why reported efficiencies for specific de-
vices are not necessarily in agreement with these curves.
A comparison of the efficiency data presented in Figure 11 and the total
mass efficiency required for control devices to meet stringent opacity
regulations (10% or 5% opacity, Table 14) shows that only high-efficiency
electrostatic precipitators, Venturi scrubbers or fabric filters are capable
of meeting the opacity standards.
Achievement of a significant reduction in fine particulate emissions will
require not only the use of high-efficiency control equipment, but also
improvements in the reliability of control equipment. It is often assumed
that control equipment is in operation 10070 of the time that a source is
operating. This is seldom the case, but data are generally not available
on control equipment operational availability. An indication of the po-
tential error in this type of assumption is provided by the results of a
survey reported in Ref. 69 . In Ref. 69 , Greco and Wynot report the results
of a study dealing with the performance and availability of electrostatic
precipitators of 16 different design types serving 51 power generating
units of TVA. The overall weighted average availability was reported to be
92.6% for a 1-year period of operation. For most industries the records of
availability have not been reported.
The availability factor assumes considerable importance in the control of
fine particulate pollutants. For example, an electrostatic precipitator of
99.5% overall mass efficiency which operates only 92.67o of the time a source
operates is really equivalent to a net efficiency of 92.1%--a reduction of
~- 7.5% in actual collection efficiency. Reliability will obviously have
to be improved to insure maximum equipment availability if efforts to con-
trol fine particulates are to be successful. More comprehensive maintenance
programs for control equipment will undoubtedly be required.
Another area in which improvements will be necessary in order to achieve
control of fine particulates is the capture efficiency of hooding systems
that are associated with many installations of control devices. Capture
72
-------
99.99
i 0.01
co
0.01
0.01
FARTICU DIAMETER - MICRONS
Figure 11. Extrapolated fractional efficiency of control devices
-------
efficiency of hooding systems that are necessary for many sources such as
metallurgical furnaces is frequently estimated to be less than 95%. If the
control device efficiency is in excess of 997o, the particulate matter which
escapes the hooding system becomes very significant.
The preceding discussion serves to emphasize that the control of fine par-
ticulates will require improvements in all phases of the technology. The
major obstacle that will impede control efforts is the inherent limitations
in ability of existing control equipment to collect submicron particles.
Control technology, involving either improved control devices or particle
conditioning processes, will need to be developed if we are to achieve a
significant reduction in the emission of fine particulates from station-
ary sources. Recent work on new control technology is reviewed in the
next section.
Emerging and New Control Technology
There are several avenues that might lead to improved control of fine
particulates: (1) development of new or novel particulate control devices;
(2) augumentation of commonly used collection mechanisms by additional
forces which do not approach zero in the ^ 2 urn size range; and (3) utili-
zation of particle conditioning or agglomeration techniques. The current
status of technology in each of these categories is reviewed briefly in
the following sections.
New or Novel Control Devices - A variety of devices which are claimed to
have high collection efficiencies in the fine particle range have been
reported in the technical and patent literature.ZP_i_Zl/ However, for most
of these devices the supporting data for the claims are often meager,
unavailable, or inconclusive. Additional testing programs will be required
to determine the capability of these devices.
ADTEC system - One of the more promising new control devices is the ADTEC
system. The ADTEC system is a wet scrubbing system that operates on the
conventional Venturi collection mechanism of inertial impaction, but es-
tablishes the requisite particle-droplet differential velocity by utilizing
waste process heat rather than external energy. On the basis of currently
available information, this system appears to offer significant improve-
ment in the collection of fine particles, at modest energy consumption
rates, where a waste gas is available which contains a sufficient amount
of thermal energy. Z?_iZ^/
Condensation scrubbers - Control devices utilizing steam condensation also
appear to offer promise for improved collection of fine particulates. The
main areas for utilization of steam condensation appear to be (1) in a
particle conditioning device, and (2) in conjunction with various types of
74
-------
wet scrubbers. In the former category, a wide variety of condensation
chambers might be designed to achieve growth of particles prior to intro-
duction into a collection device.
Some existing types of wet scrubbers utilize, to some extent, condensa-
tion scrubbing mechanisms and particle growth by steam condensation. In
a Venturi scrubber, for instance, if the aerosol is saturated in the pres-
sure zone in the Venturi throat, the subsequent expansion (which is sudden
and therefore approximately adiabatic), will cool the aerosol and cause
condensation. The water tends to condense on the particles and the grown
particles are more readily collected by impaction with the water drop
spray. If the water drops are colder than the gas, the sweep mechanisms
(i.e., diffusiophoresis and thermophoresis) may also cause the particles
to move toward the water drops.
The main emission sources where the use of condensation mechanisms for
fine particulate removal appear economically attractive are sources that
have wet or saturated gas streams at temperatures on the order of 125°-
175°F or sources with a waste heat stream in which the gas can be saturated
down to 125°-175°F. Sieve-plate scrubbers may be an attractive configura-
tion that would lend itself to maximization of condensation effects by
utilizing combinations of cold and hot water plates.
Direct steam injection appears to be too expensive if the steam must be
purchased. If a waste heat stream is available, steam generation from
the waste heat might permit use of a control device employing direct steam
injection.
Charged droplet scrubbers - The use of charged drops to agglomerate fine
particulates has some potential for transition to commercial equipment.
However, it appears that there is little reason to expect that systems
based on charged drops will represent a breakthrough for collecting sub-
micron particles. Devices utilizing charged drops may match electro-
static precipitators in performance and may offer advantages in volume
savings over a conventional electrostatic precipitator. Also, charged
drop scrubbers may be quite attractive for use on sources that emit cor-
rosive and high-moisture content pollutant streams.
Augmentation of Collection Mechanisms - A potentially fruitful avenue to
pursue in order to improve the collection of fine particulates is to aug-
ment commonly used collection mechanisms (e.g., inertia, impaction, inter-
ception) by forces which do not approach zero in the fine particle size
range. Figure 12 presents the ratio of several forces acting on an aerosol
75
-------
10
8
c
o
D
6
O
0)
&
o
o
102
10°
•Electrical Force E = 11 Kv/cm
and Maximal Surface Change
10
r2
10
r4
Force 140 dB
Diffusiophoretic
Force Assumes
107 dynes/cm3
Thermal Force
£ = l,000°K/cm
I I
I _J
10
,-2
,0
10'
Particle Diameter
Figure 12. Forces operating on aerosol particles
76
-------
particle to the weight of the particle (gravitational force) for particles
from 0.01-104- um in size. For particles < 1 ]im the forces shown in Figure
12 are many times greater than gravitational forces. Thermal, diffusio-
phoretic, electrical and other forces shown in Figure 12 might be used ad-
vantageously for cleaning industrial gas streams containing fine particulates
if devices can be designed and operated with reasonable energy requirements.
Agglomeration of Particles - The agglomeration or coagulation of partic-
ulates could be used as a step in a sequence of operations aimed at con-
trolling the emission of fine particles. If sufficiently large particles
can be produced, it may be possible to use conventional low-cost techniques
as the final collection step. Only sonic agglomeration and the agglomera-
tion of charged particles appear attractive.
Coagulation by a sonic field has as its principal advantages its applica-
bility to any aerosol, including those comprised of submicron particles.
The principal disadvantage of sonic coagulation is its relatively high
energy requirements. A second major disadvantage is the low efficiency of
acoustic coagulators and their inability to handle highly dispersed suspen-
sions. Even with long residence times, sonic precipitators which incorporate
inertial separators, cannot treat suspensions having particle loadings of
< 0.5-1.0 grains/ft^. It is therefore necessary to augment highly dis-
persed suspensions with a water mist or other particles to increase the
particle loading and obtain satisfactory separation.
One method of increasing the rate of agglomeration of fine particulates
is to add a bipolar charge, either with or without an externally imposed
field. To a limited degree, this occurs in a standard electrostatic pre-
cipitator but not sufficiently to permit efficient collection of fine par-
ticulates. With proper conditions the large electrostatic forces between
particulates can produce a large increase in the rate of agglomeration of
submicron particulates.
The limited theoretical and experimental evidence available indicates that
fine particulates can be coagulated with water mists in a bipolarly charged
mixture on a practical scale. The resulting agglomerates can then be pre-
cipitated in either a conventional or space-charge electrostatic precipitator.
SUMMARY OF STATUS OF CONTROL TECHNOLOGY
The only conventional control devices capable of significant collection of
fine particulates are high-efficiency electrostatic precipitators, Venturi
scrubbers and fabric filters. Existing data, as well as theory, on fine
particulate collection efficiency (both on a mass and number basis) of
77
-------
conventional equipment indicate that inefficiency in the fine particulate
range is an inherent characteristic of the equipment. Therefore, definite
constraints will be imposed on the types of emission standards for fine
particulates that can be implemented in the near term. The inherent limita-
tions of existing control equipment may require that process modifications
or other alternatives be used to meet near-term regulations.
Several physical phenomena exist that suggest possible avenues to improved
fine particulate control. These include condensation scrubbing, sonic and
charged droplet agglomeration and electrostatics. In general, their po-
tential for enhancing the fine particulate collection efficiency for con-
ventional control equipment as well as their potential for forming the
basis for a unique collector cannot yet be fully assessed.
REQUIREMENTS FOR COMPLIANCE TESTING AND MONITORING
Emission standards for fine particulates, if they are to be implemented
and enforced, must specify some method(s) whereby the emission of fine
particles can be measured or monitored. The emission standards selected
for detailed study in this program involve regulation of two characteris-
tics of the effluent stream from a source: (1) the opacity of the plume;
and (2) the mass of fine particulates emitted. The method(s) used to
determine compliance with these standards must,' therefore, be able to
accurately measure these characteristics. Since the ultimate success of
efforts to control fine particulate emissions will be judged, at least in
part, in terms of improvements in ambient air quality, the methods used
to determine rates of emissions from sources should also be compatible
with or relate to methods for determination of particulate concentrations
in the ambient atmosphere.
i
A literature review was conducted as part of this program to identify mea-
surement and: monitoring methods that might be useful to determine com-
pliance with emission regulations for fine particles. The search and
subsequent evaluation activities were directed primarily to methods for
determining plume opacity and manual or continuous methods for determin-
ing the mass concentration of particulates < 3 urn.
Several methods were identified and evaluated. A listing of the more
promising methods is given in Table 15. General* comments regarding some
of the major advantages or disadvantages of each method are included in
Table 15. In evaluating the methods, cost of the equipment was not an
important criterion, but complexity of the methods was considered to be
important. The primary consideration, however, was whether the method
78
-------
Table 15. MEASUREMENT AND MONITORING METHODS FOR FINE PARTICIPATES*/
A. .Opacity methods
Visual
Transmissometer
B. Mass emission methods
Transmissometer
Comment
Precision not adequate for 5% and 1070
opacity regulations.
Recommended method, but not necessarily
applicable for plumes containing con-
densed water vapor or for detached
plumes. Source and detector must be
purged with clean air. Calibration J
may be a problem.
Comment
Correlation of transmissometer reading
with concentration is possible, but
dependent on properties of partic-
ulate matter.
Cutoff impactor and filter
Cutoff cyclone and filter
Cutoff method and
piezoelectric crystal
Cutoff cyclone and beta-tape
Impactors subject to error due to
pluggage, blowoff, moisture and over-
loading .
Recommended manual method, but relies
on single point anisokinetic sample.
May require use of impingers if ef-
fluent contains condensible vapors.
Crystal must capture particles and will
probably require frequent cleaning or
replacement. Not applicable for high
particulate loadings. Might serve as
a manual method to eliminate weighing
of filters.
Offers best possibility of semicon-
tinuous measurement *
79
-------
Table 15. (Concluded) MEASUREMENT AND MONITORING METHODS
FOR FINE PARTICULATES
B. Mass emission methods (Cone1udedj
Sonic impingers
Microscopy
Light scattering
Comment
A simplified manual method, but
collecting solution must be
evaporated and weighed. Might
possibly be fed continuously to
Coulter counter.
An accurate method of counting and
sizing particles, but method is
tedious and expensive and not
amenable to continuous monitoring.
Most methods are rather complex.
They indicate number concentra-
tion rather than mass concentra-
tion.
Electric mobility analyzer
(Whitby)
Condensation nucleic counter
Diffusion batteries ^
Lidar
Holography
Intended for measurement of size
distribution rather than mass
concentration. Development might
enable use of this method for
mass concentration.
Similar to light-scattering methods.
Limited to lower concentrations.
Limited potential.
a/ Compliance monitoring applications only. Comments not meant to assess
~ suitability of methods for research applications.
80
-------
was presently developed to the point that it could be used for determining
compliance without serious limitations or disadvantages. Most of the
methods were judged to have unacceptable limitations.
A number.of other more advanced techniques and methods were also included
in our review. These techniques include acoustic attenuation, pressure
drop across a nozzle, pressure drop across a filter, unbalance of a
centrifuge, acoustic particle counting, hot-wire anemometry, decrease in
natural frequency of a vibrating band or wire, automatic weighing, tape-
spot photometry, light scattering photometry or nephelometry, lidar,
single- particle light scattering, holography, electrostatic particle
bounce, electrostatic probe-in-nozzle, electrostatic ion capture, and
electrostatic contact charging. Although some of these techniques may
be useful for other effluent measurements, each has at least one serious
limitation for the monitoring of particulate mass emissions.
A more complete review of most of these methods was conducted by Thermo-
Systems, Inc., under contract from the Environmental Protection Agency.
Their report contains a more complete description of each method and the
advantages and disadvantages of each.l-t' A summary of this work was also
published in Ref. 75.
The methods selected and recommended for determining compliance with each
type of emission regulation for fine particles selected in Chapter 4 are
presented in the following sections.
Compliance Monitoring Methods for Plume Opacity
Stringent plume opacity regulations will present problems in regard to the
monitoring of sources for compliance with these standards. The main tech-
nique currently used to determine compliance with opacity standards is to
utilize trained observers. The "calibrated eyeball" method of judging
densities and opacities of plumes is adequate for the large majority of
situations confronting air pollution control agencies at the present time.
However, visual evaluation of plume opacity will probably not be adequate
to judge compliance with stringent plume opacity regulations (5% or 10%
opacity).
Reproducibility of readings for opacity of plumes is already a problem in
enforcing existing opacity regulations. Currently, an observer is required
to reproduce his reading of opacity usually within 10% of actual plume
transmlttance. The ability of observers to accurately read Ringelmann No. 1
or 20% equivalent opacity has been questioned.Z§/ In response to this ques-
tion, the Bay Area Air Pollution Control District (BAAPCD) has stated that
BAAPCD inspectors can now read Ringelmann No. 1 or equivalent opacity with
an error of ± 1 Ringelmann number .12J With improved calibration it is believed
81
-------
that the error can be reduced to ± 1/2 Ringelmann number. Thus, aRingelmann
No. 1 (20% opacity) would be read between 1/2 (10% opacity) and 1-1/2
(30% opacity). Based on the above comments, an opacity regulation of
20% opacity seems to be the most stringent regulation that can be en-
forced by use of trained observers.
A trained observer also has difficulty in adequately evaluating a wet
plume or a detached plume. For an emitted aerosol that does not change
in the atmosphere except to become diluted with air, relative opacity at
the emission point is a reasonable measure of the potential effect it will
have on visibility downwind of the source. But in the case of the wet
plume and the detached plume, the apparent opacity at the emission point
has no relationship whatsoever with the visibility reduction downwind of
the source.
Many industrial effluents are saturated with moisture at temperatures
above ambient, and in many situations the plume from such operations will
be highly opaque. The effluent from most wet scrubbers fits in this cate-
gory, and the apparent opacity of the plume bears more relationship to
ambient temperature and humidity than the amount of pollutant being
emitted to the atmosphere. Since high-efficiency wet scrubbers are can-
didates for the control of fine particle emissions from many sources, a
problem would exist for a trained observer attempting to judge the com-
pliance of these control systems with stringent opacity regulations.
The stack effluent from the burning of high sulfur fuel oil or high sul-
fur waste gas can very often produce what is known as a "detached plume"
when the atmosphere is cool and humid. In this situation the gases im-
mediately above the emission point may be essentially transparent, but at
some distance downwind of the emission point a plume develops, presumably
from the formation of sulfuric acid droplets produced from mixing cool,
moist air with the 803 in the exhaust gases.
In view of the limitations of the trained observer, instrumental evalua-
tion of plume opacity will undoubtedly be required if stringent opacity
regulations are imposed. The best instrumental method currently available
to monitor the opacity of particulate emissions is the on-stack transmis-
someter. Commercial transmissometers with a variety of designs are avail-
*
able for measuring the in-stack opacity of particulate emissions. For
these instruments to properly measure the in-stack opacity and for the
measurement to reflect the opacity of the plume emitted by the source,
standardization of performance and installation specifications will be
necessary. EPA has prepared a set of proposed specifications for trans-
missometers and, in addition, has initiated a program to evaluate the
82
-------
performance of a commercially available transmissometer as a continuous
monitor of the in-stack opacity, plume opacity and in-stack mass concen-
tration of particulates for several different types of sources.
The Texas Air Control Board has investigated the use of a transmissometer
in over 100 installations in various industrial plants in the State of
Texas.7o?79/ -jhe types of processes investigated include: (1) alumina
kilns; (b) carbon black furnaces; (c) cement kilns; (d) copper smelters;
(e) fluid catalytic cracking units; (f) glass furnaces; (g) incinerators;
(h) lignite-fired boilers; and (i) instant coffee driers. Because of the
mostly favorable experience, the State of Texas has amended their regula-
tions to make use of the transmissometer instrument mandatory. Effective
in December 1973, all industrial processes producing visible emissions
must install a transmissometer instrument on the stack if the total flow
rate is in excess of 100,000 cfm. Consideration is being given to reduc-
ing this minimum flow rate to 50,000 cfm in order to cover more of the
sources of visible emissions in the State of Texas.
Previous experience has pinpointed some disadvantages in the use of trans-
missometer instruments. One of the major problems in some applications
occurs in rechecking the zero and span adjustments of the instrument. How-
ever, recent advances in instrumentation have resulted in the availability
of an instrument with an automatic calibration sequence.Z-L-§2/ Some
problems have also occurred because of dirt in the flue gases depositing
on the optical surfaces of the instrument, giving unrealistically low read-
ings. If the instrument is readily accessible, cleaning the optical sur-
faces every few hours to insure accurate readings is not an excessive burden
although obviously some labor cost is involved. However, if the instrument
is mounted several hundred feet up on a stack and is accessible only by an
exposed ladder, frequent cleaning becomes an even greater burden. This
problem can be reduced by proper design of the installation although it may
never be completely eliminated. The common practice with a negative pres-
sure stack is to leave a small space between the mating flanges where the
light source is bolted to the pipe which extends through the stack so that
outside air is aspirated into the slot, passes across the optical surface,
and then through the pipe, into the stack, and out the top of the stack.
A similar arrangement is provided on the other side where the bolometer or
other detector is mounted. With a positive pressure stack, a sparging point
is placed in the flanges and clean air is forced in under pressure to sweep
across the optical surfaces. Sorae installations of this type have been
quite successful and lens cleaning is only required at very infrequent
intervals, sometimes as long as a month or more. In other cases, especially
where the particulate matter in the stack is sticky or gummy, these measures
have been only partially successful. The transmissometer being marketed by
83
-------
Lear-Siegler, Inc., is equipped with high volume air purging attachments
to keep the optical windows free of dirt deposits. The purge air is sup-
plied by one or two blowers with filters, depending on pressure conditions
within the stack or duct. If the pressure is very negative, filters alone
can be used to clean the air drawn into the instrument.
As was the case with direct visual observation, effluent streams that con-
tain condensed water vapor present operational difficulties for transmis-
someters. Experience has shown that these instruments cannot operate
properly in stacks with an appreciable amount of condensed water vapor.
Such problems can be avoided by withdrawing a portion of the flue gas,
passing this stream through a heated chamber to raise the temperature
above the dew point, and making a measurement in this chamber, after which
the sample stream is then returned to the stack.ZijJLL/ However, very few
installations of this type have been constructed and operated successfully,
and moisture condensation remains a problem in certain types of industrial
operations.Zz/
Even utilizing variations in transmissometer techniques, like those described
above, it is still not possible to representatively measure plume opacity
in those cases where condensation of hydrocarbons or other gaseous pollu-
tants occurs beyond the stack (i.e., detached plume). Even though the
definition of "fine particulate11 may include these condensing vapors, the
phenomenon occurs in so few effluent streams and usually only under such
specific atmospheric conditions that this situation does not impair the
otherwise wide applicability of the transmissometer method.
In summary, it is our opinion that the transmissometer has been developed
and tested sufficiently to justify recommending it as a method that can be
used for monitoring the opacity of effluent streams in order to determine
compliance with 5% or 10% opacity standards.
Compliance Monitoring Methods for Mass Emissions
Determination of compliance with an emission standard based on the utiliza-
tion of best installed technology or the potential-emission rate concept
will require the measurement of the concentration of mass of fine par-
ticulates emitted for a source. Because a measurement of the mass of
fine particulate is involved in both cases, one monitoring procedure can
probably be developed for use with either standard. The specific require-
ments for compliance monitoring associated with each type of standard are
briefly summarized in the following paragraphs.
Emission standards for fine particles based on best installed technology
will require a monitoring method which measures the concentration (grains/
ft3) of fine particulate in the inlet and outlet of the control device.
84
-------
If the quantity of gas into and out of the control device is significantly
different because of use of dilution air or vaporization of water, etc.,
then it would be necessary to determine efficiency on the basis of pounds
per hour of fine particles at the inlet and the outlet. This means that
in addition to determining the concentration of fine particles, the gas
flow rate may also have to be determined by a velocity traverse in the in-
let and outlet ducting. Standards which utilize the concept of potential-
emission rate will require a monitoring method which measures the concen-
tration of fine particulates (grains/ft3) and the total gas flow (cfm).
Continuous monitoring on both variables would be difficult.
Based on our review of potential methods that might be used in this applica-
tion, we have selected a manual method involving the use of a sampling probe
with a 2 urn cut-off cyclone preceding a filter, similar to the EPA sampling
train, or an in-stack system as depicted in Figure 13. This method of de-
termining the concentration of fine particles has the disadvantage of being
a manual method.
This recommended manual method, consisting of a cutoff cyclone followed by
a filter, would need to be modified if "particulates" are defined to in-
clude material that exists as a particle at conditions of standard tempera-
ture and pressure. The effluent stream may contain condensible vapors
which would not be collected by the filter in the recommended sampling
method if the sampling is conducted at temperatures above standard. In-
deed, it is desirable to maintain the sampling temperature above the dew
point of the effluent gas stream. Therefore, when the effluent stream
may contain significant quantities of condensibles, it will be necessary
to follow the filter with impingers immersed in an ice bath.
A continuous monitoring method would be preferable to a manual method,
but none are available at present that, in our assessment, would be satis-
factory for monitoring the concentration of fine particles. The best po-
tential method that might be further developed for this purpose is the
beta-tape device. Although the beta-tape is a semicontinuous monitoring
method, it does appear to have possibilities for monitoring the fine par-
ticles in a sample stream from a cutoff cyclone as depicted in Figure 14.2^L/
Several particle collection techniques other than filters can be used with
beta—radiation attenuation. Among these are impactors and electrostatic
precipitators and possibly thermal precipitators. Further investigation
is necessary to fully evaluate each collection method for use with the beta
monitor. Reference 82 reports the results of field and laboratory tests
conducted to evaluate inertial sizing techniques, cascade impactors and
cyclones, for field measurements of control device efficiency as a function
85
-------
Probe
00
ON
I
S
Filter
Valve Flowmeter Pump
Cutoff
Cyclone
Flue Gas
Figure 13.
Recommended manual method for monitoring compliance of sources
with fine particle emission regulations
-------
00
Probt
f
Flue Gas
jr
Y
Cutoff
Cyclone
Output
Indicator Recorder
Geiger-
Muller
Counter
Filter Tape
Approximately
Beta Source 60 Liters/Minute
Valve Flowmeter Pump
Figure 14. Semicontinuous method for monitoring fine particle emissions
-------
of particle size. One of the conclusions of the study was that no single
inertial sizing device was suitable for all the sampling circumstances en-
countered on a wide variety of stationary emission sources. The results
of the work reported in Ref. 82 suggest that it will be necessary to modify
monitoring techniques to suit specific source types.
As noted in Chapter 4, the State of New Mexico has adopted a fine particle
regulation for coal-burning equipment. This regulation limits emission
of < 2 urn particles to 0.02 lb/10^ Btu. The following test methods are
recommended by the State of New Mexico.
Method 1 (Manual) - The sample is first passed through a cascade impactor
(Andersen type) to remove < 2 ]im particles. Particulate which escapes the
impactor is collected on the integral filter.
Method 2 (Continuous) - The sample is passed through the impactor as above,
and particulate which escapes the impactor is directed to an adhesive-
coated quartz crystal microbalance.
Method 3 - Any method as accurate as Method 1 or 2.
The above Method 1 is similar to our recommended method except that the
use of a cutoff cyclone is, we feel, subject to fewer inaccuracies and is
much more amenable to continuous sampling. Method 2 of the New Mexico
Regulation involves use of the quartz crystal microbalance (Figure 15).
One possible problem with the piezoelectric microbalance is that particles
must adhere to the surface strongly enough to overcome inertial forces of
the vibrating crystal. Most testing of this device has been done on ambient
particles and it appears that most particles smaller than 20 um adhere
well enough to be weighted.!^/ This device is relatively new and untried
in effluent gas streams but it does have the desirable feature of high
sensitivity for direct sensing of particulate mass. The primary problem
with its use may be frequent need to replace or clean the crystal, and
this must be evaluated along with testing the applicability on stack ef-
fluents before it could be recommended for compliance testing or monitor-
ing. Even if this device is not applicable for continuous monitoring,
further development may enable its use as. a substitute for the cutoff
eyelone-filter method, thereby eliminating the tedious task of condition-
ing and weighing of the filters.
t
Whether one uses the cutoff cyclone or a cascade impactor for the frac-
tionation of the large and small particles, it is a requirement of both
that a constant sample flow rate be used. Because the flow rate into the
cutoff cyclone (or impactor) is fixed, it is difficult to sample the stack
88
-------
Probe
(r
t
Flue Gas
Frequency Oscillator
Recorder Monitor Circuit
Particle
Collection
Region
Cutoff
Cyclone
Figure 15.
Schematic diagram of monitoring
1 system utilizing quartz
crystal microbalance
89
-------
gas isokinetically. However, this may not be necessary because the analyses
are only concerned with the concentration of fine particles in the stack
gas.
Studies of errors in sampling at anisokinetic conditions indicate that it
is rarely necessary to bother about isokinetic sampling for particles
of unit density below about 5 urn diameter.83^86/ Therefore, single
point sampling would be representative. This conclusion assumes that
problems of air-in-leakage, etc., in stacks which can cause localized
concentration gradients in both the gases and particulates are negligible.
Unless further experimental study develops evidence to the contrary,
sampling can be conducted anisokinetically at a single point without sig-
nificant error in order to measure fine particle concentrations by the
methods previously discussed.
The recommended method will determine the concentration of fine particles
and thereby yield the control device fine particle efficiency only if the
effluent flow rate (SCFM) is the same at the inlet and outlet. If the
flow rates (SCFM) are significantly different, a velocity traverse may
also be necessary. A velocity traverse is also a manual method that would
be very difficult to adapt to continuous measurement unless single-point
measurements at the inlet and outlet were determined to be representative,
and a calibration curve prepared to enable calculation of total flow based
on velocity at the single point.
If currently available data on particle-size dsitribution of effluents were
used to formulate the fine particle emission standard, some problems of
equality in enforcement might result during the initial stages of the im-
plementation of a mass emission standard using the recommended testing
method. Most of the particle size distribution data currently available,
which would be used to define the potential fine particle emission rate
of sources, has been obtained by a variety of methods, e.g., ASME train-
Bacho sizing, EPA train-Bacho sizing, and cascade impactors. Measurements
of the mass fraction of fine particles by these methods probably differ
considerably from those which would be obtained by the cutoff cyclone-
filter method. Depending upon the extent of discrepancy, it may be neces-
sary to measure fine particles emitted from a source by one of the more
common previously used methods as well as by the cutoff cyclone-filter
method. When an extensive data base has been developed using the recom-
mended measuring method, it may be appropriate to modify the emission
standard.
90
-------
SUMMARY OF RECOMMENDED COMPLIANCE MONITORING METHODS
Presently available methods for determining source compliance are amenable
to use with the three types of emission standards considered for the regu-
lation of fine particulate emissions. In the case of the opacity stand-
ard, the on-stack transmissometer can be used for continuous monitoring.
For the standard based on best installed technology and the potential
emission-rate, the manual methods of determining fine particle concentra-
tion by cutoff cyclone-filter and total gas flow by velocity traverse can
be used. It also appears that both of these standards might make use of
beta-tape instruments for continuous monitoring of the concentration of
fine particles, which may be the only measurement necessary for compliance
testing. Further development of the cutoff cyclone-piezoelectric crystal
method may permit its use as a substitute for the cutoff cyclone-filter
method to eliminate filter weighing and provide more immediate results.
91
-------
ECONOMIC IMPACT OF FINE PARTICLE EMISSION STANDARDS
INTRODUCTION
Solutions to environmental problems require more than just improved or new
technology. Economic problems must also be overcome. The supply of
capital—whether private, corporate or government--is always limited. It
is, therefore, imperative that the funds that are available are allocated
in a way that will return the greatest benefits per dollar invested.
Adoption of emission standards based on particle size will increase the
capital investment requirements of various industries and raise manufactur-
ing costs by whatever amount is required to operate and maintain the air
pollution control facilities. In this chapter, estimates of control costs
are presented for selected industrial sources of fine particulates, and a
limited analysis of the economic impact of the control costs on the se-
lected industrial operations is also presented.
The capital and annual costs for control equipment required to meet emis-
sion standards based on particle size were determined using model plants
for the selected industrial operations. Accurate description of the model
plant for a specific operation requires knowledge of production rates,
emission rates, carrier gas flow rates, effluent particle size distribu-
tion, stack diameter, etc. Data on the performance capability of control
equipment and the capital and operating costs of control equipment as a
function of collection efficiency and capacity (cfm) are also required.
The following sections of this chapter discuss (1) determination of cost
versus efficiency relationships for control devices, (2) methodology for
determining costs required for compliance with fine particle emission
standards, (3) potential reduction in fine particle emissions resulting
from implementation of fine particle emission standards, and (4) economic
impact resulting from costs of control equipment required for compliance
with emission standards for fine particles.
93
-------
DETERMINATION OF COST VS EFFICIENCY RELATIONSHIPS FOR CONTROL DEVICES
During this program, a review of the literature and other available sources
of data on control equipment costs was conducted. Numerous installations
of control devices are described in the literature and prior EPA studies,
but it is difficult to obtain sufficient information to correlate the costs
of individual installations on a common basis.
Our analysis of available information indicated that the most comprehen-
sive analysis of the cost of control equipment as a function of capacity
has been presented in the EPA publication "Control Techniques for Partic-
ulate Air Pollutants."87/ Figure 16 illustrates typical data from Ref.
87. The format used in Figure 16 is not a convenient method of presenta-
tion of data. Other methods of data presentation were reviewed, and a
format suggested by Ref. 88 was selected as being the most convenient
for the purposes of the current study. Reference 88 indicates that for
a given equipment capacity (cfm), equipment costs, either capital or an-
nual ized, plotted as a function of an efficiency factor (E/100-E) yield
useful working relationships for estimating costs. The data from Ref. 87
are plotted in Figures 17 and 18 in the format suggested by Ref. 88 with
equipment capacities (cfm) as a parameter. The data in Ref. 87 (as shown
in Figure 16) present generalized efficiency ranges and only three values
of the nominal efficiency of each control device can be assigned to the
cost data given in Ref. 87- It was assumed that a linear relationship
could be used to depict the data as shown in Figures 17 and 18.
The lines shown in Figures 17 and 18 represent our judgment of the best
fit of available data for each equipment capacity (cfm). In the case of
fabric filters, only one efficiency value is given in Ref. 87 and the re-
lationship of cost to efficiency is not known. It seems probable that
the cost of fabric filters would increase as the required efficiency is
increased, because the higher efficiencies would generally require more
cloth area per cfm (i.e., a lower air-to-cloth ratio). Therefore, we
again assumed that the straight-line relationship is valid for fabric
filters, and the lines were drawn with an arbitrary slope between that
for electrostatic precipitators (ESP) and wet scrubbers.
It is evident from the above discussion that the cost vs efficiency re-
lationships (Figures 17 and 18) are based on limited data and that it has
been necessary to make several assumptions which could lead to errors in
subsequent calculations. However, Figures 17 and 18 represent the best
estimations that are possible based on available information.
94
-------
10 50 100 500 1000
GAS VOLUME THROUGH COLLECTOR, 103 ocfm
Figure 16. Annualized cost for operation of high-voltagt
electrostatic precipitators
87/
95
-------
10.0
50%
75%
90%
95%
E
99%
99 5%
I
99.9%
I
99.99%
PARAMETER ON CURVES: EQUIPMENT CAPACITY IN CFM
VO
I i.c
A FA1MC FILTER
O H.ECTROSTATIC PKCIPITATOR
D WETSCWMER
0.1
I I I
I III
I I I I I
I III
I I I I I
1 I I I
10
100
E
100-E
1.000
10.000
Figure 17. Installed costs for control equipment
-------
10.0
50%
75%
90%
95%
E
99%
99.5%
99.9%
I
99.99%
PARAMETER ON CURVES: EQUIPMENT CAPACITY IN CFM
A FABRIC FILTER
O ELECTROSTATIC PKCIHTATC*
Q WETSCRUHER
VO
0.1
10
100
E
J I I I I I I I I I 1 I I I I I I
1,000
10.000
Figure 18. Annualized costs for control equipment
-------
METHODOLOGY FOR DETERMINING COSTS REQUIRED FOR COMPLIANCE WITH FINE PARTI-
CLE EMISSION STANDARDS
The determination of the control costs required for compliance with various
emission standards for fine particles involves the following major steps:
1. Development of model plants for each industry category.-
2. Determination of control-device performance required to meet a specific
emission standard for fine particles.
3. Determination of costs for model plant and industry for compliance with
specific emission standard for fine particles.
General details involved in each of these steps are discussed in the fol-
lowing subsections.
Development of Model Plants for Important Industrial Sources of Fine
Particle Emissions
A review of the literature and previous EPA industry studies was conducted
in order to formulate "model plants" for each important source of fine
particle emissions. The objective of this activity was to select an average
or representative size plant in terms of production rate. Other important
parameters for these model plants were carrier gas flow rate and tempera-
ture, emission rate of particulates, particle size distribution of emitted
particulates, and "best installed control device" (type and efficiency).
In several industries the model plant consisted of only one source such
as a coal-fired power plant or a cement kiln. Other categories were
divided into two or three model plants because of the different types of
sources (ferroalloy furnaces). In still other cases only one model plant
was selected but it consisted of several different sources and more than
one control device (primary nonferrous).
It was not possible to develop model plants and carry out the cost calcula-
tions for all sources because of lack of information—especially particle
size distribution data. However, an attempt was made to carry out these
calculations for the most important industrial sources of fine particles
even though it was necessary in some cases to make generalizations and
assumptions.
98
-------
Model plant sizes were selected with the realization that such a generali-
zation does not truly reflect the situation for those plants that are much
smaller or much larger. However, the primary concern of this study was
the total industry impact so it was felt that use of the model plant was
justified.
Determination of Control Device Performance Requirements
Control equipment costs presented in Figures 17 and 18 are based on the
overall mass efficiency of the control equipment. In order to utilize these
data to determine costs, the control device performance requirements neces-
sary to achieve a fine particle emission standard must be translated into
overall mass efficiency. The specific standards that will be considered in
the calculation of costs are: (1) plume opacities of 10% and 5%, and
(2) utilization of the best installed technology. Although the potential
emission rate concept offers one of the most direct methods for regulat-
ing fine particle emissions, calculations of control equipment costs were
not performed for this type of regulation because of the need for more
detailed and extensive source testing to determine both potential emis-
sion rates and source compliance requirements.
The techniques used to translate the plume opacity and best installed
technology standards into overall mass efficiency requirements are pre-
sented in the following subsections.
Control Device Performance Requirements for Plume Opacity Standards - An
opacity regulation is perhaps the most difficult to translate into terms
of overall mass efficiency required of control equipment. A suitable
relationship between plume opacity and particulate properties and stack
diameter must be used to determine the degree of collection efficiency
required to meet a plume opacity standard. Ensor and Pilat have recently
developed a procedure for calculating plume opacity from particulate air
pollutant properties.—' (Also see Chapter 5.) These authors developed
the following equation to calculate the expected mass concentration for
various values of plume transmittance (or opacity), average particle
density, and plume diameter.
W = -K p/L In (I/Io) (4)
In Eq. (4),
W = Total particulate mass concentration of plume (i.e., grain
loading),
p = Average particle density,
99
-------
L = Diameter of plume,
I/Io = Light transmittance,
K _ Specific particulate volume
Extinction coefficient ratio.
The parameter K is a function of the particle size distribution, the re-
fractive index of the particulate, and the wavelength of incident light.
Reference 68 presents graphs of K versus the geometric mass mean particle
radius with the geometric standard deviation of the particle size distribu-
tion and the refractive index of the particles as parameters.
The grain loading from an uncontrolled source was determined by using the
effluent flow rate from the model plant, the production rate of the model
plant, and the appropriate emission factor. Equation (4) was then used
to compute the grain loading of a controlled source corresponding to the
opacity limitation for each model plant. In some cases the characteris-
tics of the particles (parameters p and K) were not available, and it was
necessary to assume that they were approximately equivalent to other
similar sources for. which the data were available.
When data were available to permit calculation of the outlet grain loading
corresponding to a selected opacity, the efficiency of the control device
required was calculated from Eq. (5) assuming no dilution of the stack
gases.
_,,.,... Inlet Grain Loading - Outlet Grain Loading /CN
Etriciency = ^ "• (_j)
Inlet Grain Loading
In order to determine, from Figures 17 and 18, the cost of the control
device with the efficiency determined from Eq. (5), it is necessary to
specify the type of control device. If more than one type of control de-
vice could provide the required efficiency and function on the given
source, the selection of the type of control device was based on knowl-
edge of usual industry practice and lowest cost.
t
Control Device Performance Requirements'for Best Installed Technology
Standard - The emission standard based on the installation of best in-
stalled control technology on all plants would require that all sources
in each industrial category be equipped with control equipment that is
at least as efficient as that of the best control device that has been
installed on each type of plant in recent years. Our data base was used
to identify this "best installed control device" (BICD) for each industry
100
-------
category and its overall efficiency. This device was not necessarily the
highest efficiency device available, but rather the best that is generally
being installed in a given industry at the present time.
Costs to Model Plant and Industry for Compliance with Specific Fine Particle
Emission Standards
The procedure for determining the costs for compliance with specific fine
particle emission standards is illustrated in detail in Appendix B using
coal-fired power plants as an example. Only the results of the calcula-
tions will be presented for the various sources in the following subsections,
The annualized costs associated with the control devices were based on
8,000 hr of operation per year. Not all plants operate that many hours,
and indications of the likely operating hours are given for those industries
which differ significantly from the assumed 8,000 hr/year.
Coal-Fired Power Plants - Table 16 presents the estimated control equip-
ment costs for coal-fired electric utility plants. An electrostatic
precipitator with an overall mass efficiency of 99% was selected as the
best installed control device (BICD). The overall collection efficiencies,
computed from Eq. (4) using effluent property data from Refs. 68,60,61
required to achieve the 10 and 5% opacity regulations are 99.66 and 99.83%.
Electrostatic precipitators were chosen as the control devices that would
be used to achieve compliance with the opacity standards.
•
A comparison of the incremental annualized costs for compliance with the
fine particle regulations is presented in Table 17. Since only a small
percentage of electric utility plants are currently equipped with the BICD,
it was assumed that all existing plants would have to install new control
•
devices to meet all the regulations. The incremental annualized costs
shown in Table 17 for the case where all the electric utility plants are
required to install the BICD are based on the preceding assumption. . The
incremental costs for the opacity regulations represent the difference
between the costs associated with the opacity regulations and the BICD
regulation.
Estimates of the reduction in fine particle emissions from coal-fired
power plants that would result from installation of control equipment re-
quired for compliance with the emission standards for fine particles are
also presented in Table 16. Currently, fine particle emissions from the
total industry are estimated at 243,000 Ib/hr. Installation of the best
installed control system on all plants would reduce emissions to 15,000
Ib/hr—a 94% reduction. Compliance with a 10% opacity regulation would
reduce the emissions to 5,050 Ib/hr (a 97% reduction), while compliance
101.
-------
Table 16. CONTROL EQUIPMENT COSTS FOR COAL-FIRED ELECTRIC UTILITY PLANTS
I. Source description
Capacity
Coal burned
Model Plant
400 raw (100% load)
1.2 x 10 tons/year
at 68.5% load
factor
Industry Total
150,000 raw
258 x 10 tons/year
Fine particle emissions
(a) Uncontrolled
(b) Present
II. Best installed control
systemSL'
Installed cost
Annualized cost
Fine particle emissions
3,700 lb/hr
c/
III. 10% Opacity regulation-
Installed cost
Annualized cost
Fine particle emissions
IV. 5% Opacity regulation!/
Installed cost
Annualized cost
Fine particle emissions
$1.03 x 106
$0.19 x 106/year
69.1 lb/hr
$1.4 x 106
$0.25 x 106/year
23.5 lb/hr
$1.57 x 106
$0.30 x 106/year
11.7 lb/hr
796,000 lb/hr
243,000 lb/hr
$386 x 10°
$71 x 106/year
15,000 lb/hr
$525 x 106
$94 x 106/year
5,050 lb/hr
$589 x 106
$112 x 106/year
2,510 lb/hr
a/ Quantity emitted depends on control device installed.
b/ Control device required to meet standard is 99% efficient electro-
static precipitator.
£/ Control device required to meet standard is 99.66% efficient electro-
static precipitator.
d/ Control device required to meet standard is 99.83% efficient electro-
static precipitator.
102
-------
Table 17. INCREMENTAL ANNUALIZED COSTS FOR COAL-FIRED ELECTRIC UTILITY PLANTS
TO MEET FINE PARTICLE EMISSION STANDARDS
Emission Standard
All plants install
best installed con-
trol device (BICD)
10% Opacity
5% Opacity
Control System
99% Efficient electrostatic
precipitator (BICD)
99.66% Efficient electro-
static precipitator
99.83% Efficient electro-
static precipitator
Incremental Annualized
Annualized Cost ($/Year) Cost ($/Year)
Model Plant All Plants Model Plant All Plants
190,000 71 x 106
250,000 94 x 106
300,000 112 x 106
190,000
60,000
50,000
71 x 106
23 x 106
18 x 10(
-------
with a 5% opacity regulation would further reduce the emissions to 2,510
Ib/hr (a 99% reduction) . In making the above estimates of emission reduc-
tions, the control device was assumed to be available and operating at
peak efficiency at all times.
Iron and Steel Plants. - The costs for compliance with fine particle emis-
sion regulations were estimated for three of the major sources of partic-
ulate pollutants in iron and steel plants—sinter machines, basic oxygen
furnaces, and electric arc furnaces. Open hearth furnaces were not in-
cluded because these furnaces are gradually being phased out of production.
The cost estimates for the three sources are presented separately in the
following sections.
Sinter machine - Estimated control equipment costs for sinter machines are
presented in Table 18. A model plant with a production capacity of 4,000
tons/day (1.46 x 10" tons/year) was used as a basis for the calculations.
Sinter plants in current use in the industry range in capacity from 800
to 14,000 tons/day. For this source, 10% and 570 opacity regulations would
not require the installation of control equipment that is as efficient as
the best installed control device (i.e., a fabric filter system). Since
only about 19% of the existing sinter machines are equipped with fabric
filters and electrostatic precipitators are the prevalent control systems
in use, electrostatic precipitators were selected for compliance with the
opacity standards.
Table 19 gives a comparison of the incremental annualized costs for com-
pliance with the fine particle regulations. The costs given in Table 19
reflect the fact that 1970 of the existing sinter machines are already
equipped with fabric filters.
The potential reduction in fine particle emissions resulting from imple-
mentation of an emission standard for fine particles for all sinter
machines ranges from 48.5 to 96%.
Basic oxygen furance - Control equipment costs for basic oxygen furnaces
in iron and steel plants are shown in Tables 20 and 21. The model furnace
selected as the basis for the estimates of control costs was assumed to
have a capacity of 1 x 10 tons/year or 200 tons/heat. The costs presented
in Tables 20 and 21 recognize the fact that 20% of the existing basic
oxygen furnaces are already equipped with the best installed control device.
Implementation of an emission regulation for fine particles could result
in a significant reduction in fine particulate emissions from basic oxygen
furnaces.
104
-------
Table 18. CONTROL EQUIPMENT COSTS FOR SINTER MACHINE (WINDBOX)
(IRON AND STEEL PLANTS)
Model Plant
I. Source description
Capacity
Fine particle emissions
(a) Uncontrolled
(b) Present
II. Best installed control
system^/
Installed cost
Annualized cost
Fine particle emissions
III. 10% Opacity regulation!/
t
Installed cost
Annualized cost
Fine particle emissions
IV. 5% Opacity regulation6-/
Installed cost
Annualized cost
Fine particle emissions
4,000 tons/day
116 Ib/hr
a/
$1 x 106
$227,000/year
1.04 Ib/hr
$770,000
$146,000/year
14.2 Ib/hr
$850,000
$162,000/year
6.9 Ib/hr
Industry Total
< /•
90 x 10 tons/year
7?150 Ib/hr
1,400 Ib/hr
$49.9 x 10£
$11.3 x 106/year£/
64 Ib/hr
$38.4 x l
7.3 x 106/year£/
722 Ib/hr£/
$42.4 x 10£
$8*1 x 106/year£-/
354 lb/hr£/
aj Quantity emitted depends on control device installed.
b/ Control device required to meet standard is.a fabric filter.
c_/ Based on fact that approximately .19% of sinter machines are already
equipped with fabric filter.
d/ Control device selected to meet standard is 98.6% efficient electro-
static precipitator.
e/ Control device selected to meet standard is 99.4% efficient electro-
static precipitatqr. .
105
-------
Table 19. INCREMENTAL ANNUALIZED COSTS FOR SINTER MACHINES (IRON AND STEEL)
TO MEET FINE PARTICLE EMISSION STANDARDS
H
O
Emission Standard
10% Opacity
5% Opacity
All plants install
BICD
Control System
98.6% Efficient electro-
static precipitator
99.4% Efficient electro-
static precipitator
Fabric filter
Incremental Annualized
Annualized Cost ($/Year) Cost ($/Year)
Model Plant All Plants Model Plant All Plants
146,000 7.3 x 106 146,000 7.3 x 106
162,000 8.1 x 106 16,000 0.8 x 106
227,000 11.3 x 106 65,000 3.2 x 106
-------
Table 20. CONTROL EQUIPMENT COSTS FOR BASIC OXYGEN FURNACES
(IRON AND STEEL PLANTS)
Model Plant
Industry Total
I. Source description
Capacity
Fine particle emissions
(a) Uncontrolled
(b) Present
II. Best installed control
systemk'
Installed cost
Annualized cost
Fine particle emissions
III. 10% Opacity regulation^/
Installed cost
Annualized cost
Fine particle emissions
IV. 5% Opacity regulation6-/
Installed cost
Annualized cost
Fine particle emissions
1 x 10 tons/year 80 x 10^ tons/year
(200 tons/heat)
10,700 Ib/hr
a/
$960,000
$163,000/year
32 Ib/hr
$1 x 106
$177,000/year
19 Ib/hr
$960,000!/
$207,000/year
10 Ib/hr
400,000 Ib/hr
43,600 Ib/hr
$61.5 x
$10.4 x 106/year£/
2,560 Ib/hr
$80 x 106
$14.2 x 106/year
1,540 Ib/hr
$77 x 106
$16.5 x 106/year
800 Ib/hr
a/ Quantity emitted depends on control device installed.
b/ Control device required to meet standard is 99.7% efficiency electro-
static precipitator. This is a very high efficiency for an electro-
static precipitator operating on a source emitting essentially 100%
< 3 urn particulates, and, in fact, this efficiency is inconsistent
with fractional efficiency data for electrostatic precipitators.
However, the particle size distribution at the exit of the furnace
may not be what the electrostatic precipitator "sees" (i.e., ag-
glomeration may occur before the particulate reaches the control
device).
£/ Based on fact that approximately 20% of basic oxygen furnaces are
already equipped with 99.77. efficiency electrostatic precipitator.
d/ Control device required to meet standard is 99.8% efficiency electro-
static precipitator.
e/ Control device required to meet standard is 99.91% efficiency fabric
filter.
f/ Installed cost of 99.91% efficiency fabric filter is the same as for
99,7% electrostatic precipitator, but annualized cost is higher.
107
-------
Table 21. INCREMENTAL ANNUALIZED COSTS FOR BASIC OXYGEN FURNACES TO MEET FINE
PARTICLE EMISSION STANDARDS
Emission Standard
All plants install
BICD
10% Opacity
5% Opacity
Control Systems
99.7% Efficient electro-
static precipitator
99.8% Efficient electro-
static precipitator
99.91% Efficient fabric
filter
Incremental Annualized
Annualized Cost ($/Year) Cost ($/Year)
Model Plant All Plants Model' Plant All Plants
163,000 10.4 x 106 163,000 10.4 x 106
177,000 14.2 x 106 143,000 4.2 x 106
207,000 16.5 x 106 30,000 2.3 x 106
O
00
-------
Installation of the BICD on all sources could reduce emissions from
43,600 Ib/hr to 2,560 lb/hr--a reduction of 94.5%. The more stringent
opacity regulations could result in a reduction of 96.5 to 98%.
Electric arc furnaces - A model furnace with a capacity of 80 tons/heat
was used as the basis for the estimates of control equipment costs pre-
sented in Tables 22 and 23. Electric arc furnaces represent a relatively
well controlled source with 65% of existing furnaces already equipped
with fabric filters. Costs presented in Tables 22 and 23 reflect this
fact.
The costs given in Table 22 show that the annualized cost associated with
the opacity standards are lower than that for the best installed control
device and reflect the fact that these standards require a control ef-
ficiency that is lower than that of the best installed control system
(i.e., the fabric filter). However, the installed cost of the control
device selected for the 5% opacity standard is higher than that for the
best installed control system. It would, therefore, be reasonable to
assume that the higher efficiency best installed control system might be
chosen to meet the 5% opacity standard, but this depends on whether the
choice is made on the basis of installed cost or annualized cost.
The estimated reduction in fine particle emissions that might be achieved
by implementation of fine particle emission standards ranges from 71% for
10% opacity to 82% for the best installed control system.
i
Cement Plants (Rotary Kilns) - In this source category, control equipment
costs for compliance with fine particle emission standards were estimated
only for the rotary kiln—the major emission source in a cement plant. A
cement plant with a production capacity of 3 x 10° barrels/year was chosen
as the basis for the model plant used to estimate the control equipment
costs. Cement plants range in size from 1-14 x 10*> barrels/year produc-
tion with approximately 50% of the plants in the capacity range of 2-4 x
10^ barrels/year.
Tables 24 and 25 summarize the estimated costs for compliance with the
fine particle emission standards. For cement kilns, the opacity regula-
tions are less stringent than the standard requiring the use of the best
installed control device on all sources. The cost estimates shown in
Tables 24 and 25 reflect the fact that approximately 17% of the cement
kilns are already equipped with the best installed control device—a
fabric filter system.
109
-------
Table 22. CONTROL EQUIPMENT COSTS FOR ELECTRIC ARC
FURNACES (IRON AND STEEL PLANTS)
Model Plant
Industry Total
I. Source description
Capacity
Production
Fine particle emissions
(a) Uncontrolled
(b) Present
80 tons/heat
100,000 tons/year
127 Ib/hr
a/
10 to 250 tons/heat
21.5 x 106 tons/year
27,400 Ib/hr
3,600 Ib/hr
II. Best installed control
system^'
Installed cost
Annualized cost
Fine particle emissions
III. 10% Opacity^/
Installed cost
Annualized cost
Fine particle emissions
IV. 5% Opacity^/
Installed cost
Annualized cost
Fine particle emissions
$182,000
$41,000/year
3 Ib/hr
$161,000
$28,000/year
8.2 Ib/hr
$204,000
$32,000/year
3.5 Ib/hr
$13.7 x 106£/
$3.1 x 106/year£/
645 Ib/hr
$12.1 x 106£/
$2.1 x 106/ye
1,036 Ib/hr0-/
.4.x 106-/
$2.4 x 106/year£/
683
a/ Quantity emitted depends on control device installed;
b/ Control device required to meet standard is a fabric filter.
c/ Based on fact that approximately 65% of electric arc furnaces are
already equipped with fabric filters'.
d/ Control device selected to meet standard is 95.8% efficient electro-
static precipitator.
e/ Control device selected to meet standard is 98.2% efficient electro-
static precipitator.
110
-------
Table 23. INCREMENTAL ANNUALIZED COSTS FOR ELECTRIC ARC FURNACES TO MEET FINE
PARTICLE EMISSION STANDARDS
Emission Standard
10% Opacity
5% Opacity
Annualized Cost ($/Year)
Incremental Annualized
1 Cost ($/Year)
Control System
95.8% Efficient electro- , , 28,000
static precipitator
Model Furnace 'All Furnaces Model Furnace All Furnaces
• 2.1 x 106
98.2% Efficient electro-
static precipitator
All furnaces install Fabric filter
BICD
32,000
41,000
2.4 x 106
3.1 x 106
28,000
4,000
9,000
2.1 x 106
0.3 x 106
0.7 x 106
-------
Table 24. CONTROL EQUIPMENT COSTS FOR CEMENT PLANT ROTARY KILNS
Model Plant
Industry Total
I. Source description
Capacity
Production
Fine particle emissions
(a) Uncontrolled
(b) Present
II. Best installed control
system?-/
Installed cost
Annualized cost
Fine particle emissions
III. 10% Opacity!/
Installed cost
Annualized cost
Fine particle emissions
IV. 5% Opacity^/
Installed cost
Annualized cost
Fine particle emissions
3 x 106 barrels/year 550 x 106 barrels/year
400 x 10 *> barrels/year
1,325 Ib/hr
$462,000
$102,000/year
11 Ib/hr
$440,000
$67,000/year
138 Ib/hr
$481,000
$74,000/year
67 Ib/hr
243,000 Ib/hr
44,300 Ib/hr
$70.3 x 1
$15.5 x 106/year£'
2,000 Ib/hr
$67.5 x
$10.2 x I06/year£/
25,200
$73.2 x 1
$11.2 x HP/year^/
12,200 lb/hr£/
a/ Quantity emitted depends on control device installed.
b_/ Control device required to meet standard is a fabric filter.
£/ Based on fact that approximately 17% of'kilns are already equipped
with fabric filter.
d/ Control device selected to meet standard is 98.7% efficient electro-
static precipitator.
e/ Control device selected to meet standard is 99.4% efficient electro-
static precipitator.
112
-------
Table 25. INCREMENTAL ANNUALIZED COSTS FOR CEMENT KILNS TO MEET FINE
PARTICLE EMISSION STANDARDS
Incremental Annualized
Annualized Cost ($/Year) Cost ($/Year) .-•-
Emission Standard
10% Opacity
5% Opacity
Control System Model Kiln All Kilns Model Kiln All Kilns
98.7% Efficient electro- 67,000
static precipitator
99.4% Efficient electro- 74,000
static precipitator
10.2 x 106 67,000 10.2 x 106
11.2 x 106 7,000 1 x 106
All kilns install Fabric filter
BICD
102,000 15.5 x 106 28,000 4.3 x 106
U>
-------
Reduction in fine particle emissions from all cement kilns that might be
achieved by implementation of fine particle emission standards ranges from
72.5% for 10% opacity to 95.5% for the best installed control system.
Hot-mix Asphalt Plants - There are about 3,000 asphalt paving plants in
the U.S. ranging in capacity from 75 to 300 tons/hr. A plant with a
capacity of 150 tons/hr was selected as the basis for the estimates of
control costs. Only the rotary dryer which is the major source of emis-
sions was included in the cost estimates.
Tables 26 and 27 present the cost estimates for the rotary dryer. The
opacity regulations are not as restrictive as the standard based on the
utilization of the best installed control device—a fabric filter system.
Currently, approximately 1670 of the asphalt dryers are already equipped
with fabric filters.
The installed costs for each standard, as shown in Table 26, reflect the
fact that the opacity standards are not as restrictive as .the best installed
control system. However, the annualized costs for the opacity standards
are higher than that for the best installed control system. Therefore,
the choice of control equipment that might be installed to meet the opacity
standards would depend on whether selection is based on installed cost or
annualized cost.
Ferroalloy Furnaces - Three different types of furnaces are currently used
to produce ferroalloy materials. Since the furnaces are quite different
both with regard to their operation and emissions, individual models were
used for each type of furnace. Tables 28 to 30 summarize the cost esti-
mates for ferroalloy furnaces. Because of inadequate data on effluent
characteristics from the individual furnaces, it was not possible to de-
termine control equipment requirements for opacity regulations. As a
result, only the costs associated with the standard based on the best
installed control device are shown in Tables 28 to 30.
For the closed electric furnace, available data indicate that all exist-
ing furnaces are already equipped with the best installed control device.
Currently only about 40% of the hooded open furnaces are equipped with the
BICD (fabric filter), and installation of the BICD on all of this type of
furnace would result in an 81% reduction in fine particle emissions. Ap-
proximately 20% of existing unhooded open furnaces are equipped with fabric
filters, and installations of this device on all sources would reduce fine
particle emissions by about 52.5%.
114
-------
Table 26. CONTROL EQUIPMENT COSTS FOR HOT-MIX ASPHALT PLANT ROTARY DRYERS
Model Plant
I. Source description
Capacity
Production
Fine particle emissions
(a) Uncontrolled
(b) Present
II. Best installed control
systemk/
Installed cost
Annualized cost
Fine particle emissions
III. 107o Opacity!/
Installed cost
Annualized cost
Fine particle emissions
IV. 57o Opacity!/
Installed cost
Annualized cost
Fine particle emissions
150 tons/hr
90,000 tons/year
840 Ib/hr
a/
$53,000
$10,600/year
6.6 Ib/hr
$21,500
$19,800/year
60 Ib/hr
$29,700
$29,700/year
50 Ib/hr
Industry Total
700 x 106 tons/year
350 x 106 tons/year
3.26 x 106 Ib/hr
257,000 Ib/hr
$173 x 106£/
$34.6 x 106/year£/
25,600 Ib/hr
$28 x 106®/
$26 x 106/year£/
200,000 Ib/hr£/
$97 x 106&/
$97 x 106/yearfi/
167,000 lb/hr£/
a/ Quantity emitted depends on control device installed.
b_/ Control device required to meet standard is a fabric filter.
£/ Based on fact that approximately 16% of dryers are already equipped
with fabric filters.
d/ Control device selected to meet standard is 94.870 efficient wet scrubber.
e/ Based on fact that approximately 16% of dryers are already equipped
with fabric filters and assumption that 507o of the present cyclone
plus wet scrubber systems will meet 107o opacity standard.
f/ Control device selected to meet standard is 97.47o efficient wet scrubber.
£/ Based on fact that approximately 167o of dryers are already with fabric
filters and assumption that few of the existing cyclone plus wet
scrubber systems will meet 570 opacity standard.
115
-------
Table 27. INCREMENTAL ANNUALIZED COSTS FOR HOT-MIX ASPHALT PLANT ROTARY
DRYERS TO MEET FINE PARTICLE EMISSION STANDARDS
Emission Standard
10% Opacity
5% Opacity
Control System
94*8% Efficient wet
scrubber
97.4% Efficient wet
scrubber
All dryers install Fabric filter
BICD
Annualized Cost ($/Year)
Model Dryer All Dryers
19,800
29,700
10,600
26 x 106
97 x 106
34.6 x 106
Incremental Annualized
Cost ($/Year)
Model Dryer All Dryers
19,800
9,900
26 x 106
71 x 106
-------
Table 28. CONTROL EQUIPMENT COSTS FOR FERROALLOY FURNACES (CLOSED
ELECTRIC FURNACE)
Model Plantg/
Industry Total
I. Source description
Capacity
Fine particle emissions
(a) Uncontrolled
(b) Present
II. Best installed control
system0./
Installed cost
Annualized cost
Fine particle emission
17,000 kw 1,025,000 kw
(13,600 tons/year) (820,000 tons/year)
1,535 Ib/hr
b/
92,000 Ib/hr
11,300 lb/hrd/
$15,000
$3,000/year
38 Ib/hr
d/
11,300 Ib/hr
III. 10% Opacity - Insufficient effluent property data prohibited deter-
mination of control efficiency required.
IV. 5% Opacity - Same as 10% opacity.
a/ Operating hours - 8,000 hr/year
b_/ Quantity emitted depends on control device installed.
£/ Control device required to meet standard is a disintegrator.
d_/ MRI survey in 1970 indicated that all plants are already equipped
with disintegrators.
117
-------
Table 29. CONTROL EQUIPMENT COSTS FOR FERROALLOY FURNACES (HOODED OPEN
ELECTRIC FURNACE)
I. Source description
Capacity
Fine particle emissions
(a) Uncontrolled
(b) Present
II. Best installed control
system?./
Installed cost
Annualized cost
Fine particle emissions
Model Plants/
Industry Total
13,500 kw 1,350,000 kw
(10,800 tons/year) (i;080,000 tons/year)
1,295 Ib/hr
b/
'-••• 129,500 Ib/hr
84,100 Ib/hr
$188,000
$42,000/year
158- lb/hr£/
$11.3 x 106 d/
$2.5 x lOfy
15,800 Ib/hrl/
III. 10% Opacity - Insufficient effluent property data prohibited deter
mination of control efficiency required.
IV. 5% Opacity - Same as 10% opacity.
a/ Operating hours - 8,000 hr/year.
b_/
£/
d/
Quantity emitted depends on control device installed;1
Control device required to meet standard is a fabric 'filter.
Based on fact that approximately 40% of hooded electric furnaces are
already equipped with fabric filter or disintegrator system.
e/ Based on assumed capture efficiency of hooding system of 90%.
118
-------
Table 30. CONTROL EQUIPMENT COSTS FOR FERROALLOY FURNACE (UNHOODED OPEN
FURNACE)
Model Planta/
Industry Total
I. Source description
Capacity
Fine particle emissions
(a) Uncontrolled
(b) Present
II. Best installed control
system?./
Installed cost
Aanualized cost
Fine particle emissions
10,000 kw
(8,000 tons/year)
960 Ib/hr
b/
750,000 kw
(6 x 105 tons/year)
72,000 Ib/hr
60,800 Ib/hr
$750,000
$168,000/year
384 lb/hr
-------
Rotary Lime Kilns - Rotary lime kilns range in capacity from 50 to 560
tons/day. A kiln with a capacity of 250 tons/day was selected as the
basis for the estimate of control costs. Only the rotary kiln, which is
considered to be the major source of emissions from lime plants, was in-
cluded in the cost estimates.
Tables 31 and 32 present the cost estimates for the rotary kilns. The
opacity regulations are not as restrictive as the standard based on
utilization of the best installed control device--a fabric filter.
The installed costs for each standard, as shown in Table 31, reflect the
fact that the opacity standards are not as restrictive as that of the best
installed control system. However, the annualized costs for the opacity
standards are higher than that for the best installed control system.
Therefore, the choice of control equipment that might be installed to meet
the opacity standards would depend on whether selection is based on in-
stalled cost or annualized cost.
«
Municipal Incinerators - Individual furnaces used for municipal incinera-
tion range in size from 15 to 300 tons per 24 hr. An incinerator with a
capacity of 10 tons/hr (240 tons/24 hr) was selected as the basis for the
estimate of control costs.
Tables 33 and 34 present the cost estimate for municipal incineration.
The opacity regulations are not as restrictive as the standard based on
the use of the best installed control device—a 99.0% efficient electro-
static precipitator. Few incinerators in the U.S. are presently equipped
with electrostatic precipitators but several are installing or planning
to installed them. ,;:
Iron Foundry (Cupolas) - The cupola furnaces used in the laron foundry in-
dustry range in size from 1 ton/hr to 40 tons/hr or larger* A cupola with
a capacity of 10 tons/hr was selected as the basis for the} estimate of
control costs.
Tables 35 and 36 present the cost estimates for the foundry cupolas. They
also indicate that the opacity regulations are not as stringent as the
standard based on equipping the foundry with the best installed control
device (a fabric filter). However, the annualized costs for opacity stand-
ards are higher than that for the best installed system. Therefore, the
choice of control equipment that might be installed to meet the opacity
standards would depend on whether selection is based on installed cost or
annualized cost.
Primary Aluminum-Electrolytic Cells - It is difficult to characterize the
primary aluminum industry and the different sources and processes.
120
-------
Table 31. CONTROL EQUIPMENT COSTS FOR ROTARY LIME KILNS
I. Source description
Capacity '
Production
Fine particle emissions
(a) Uncontrolled
(b) Present
II. Best installed control
sy'stemb/
Installed cost
Annualized cost
Fine particle emissions
III. 10% Opacity regulation!/
Installed cost
Annualized cost
Fine particle emissions
IV. 5% Opacity regulation!/
Installed cost
Annualized cost
Fine particle emissions
Model Plant
Industry Total
250 tons/day 74,400 tons/day
87,500 tons/year 18.5 x 106 tons/year
255 Ib/hr
a/
$70,000
$14,000/year
4.0 Ib/hr
$42,000
$39,500/year
56 Ib/hr
$58,000
$60,500/year
24 Ib/hr
75,800 Ib/hr
22,000 Ib/hr
$12.6 x 106£/
$2.5 x 106/year£/
1,200 Ib/hr
$7.5 x 10%/
$7.1 x I06/year£/
10,500 lb/hr£/
$10.4 x 106£/
$10.8 x 106/yeare/
4,800 lb/hr£/
£/ Quantity emitte'd depends on control device installed.
b_/ Control device required to meet standard is a fabric filter.
£/ Based on fact that approximately 40% of kiln capacity is already
equipped with fabric filters.
d/ Control device selected to meet standard is 97.0% efficient wet
scrubber.
e/ Based on 40% present application of fabric filters and assumption
that only a few plants are equipped with wet scrubbers that meet
the required efficiency. :
f/ Control device selected to meet standard is 98.7% efficient wet'
scrubber.
121
-------
Table 32. INCREMENTAL ANNUALIZED COSTS FOR ROTARY LIME KILNS TO MEET FINE
PARTICLE EMISSION STANDARDS
Emission Standard
10% Opacity
5% Opacity
Control System
All kilns install Fabric filter
BICD
97.0% Efficient wet
scrubber
98.7% Efficient wet
scrubber
Annualized Cost ($/Year)
Model Plant All Plants
14,000
39,500
60,500
2.5 x 106
7.1 x 106
Incremental Annualized
Cost ($/Year)
Model Plant All Plants
14,000
25,500
2.5 x 106
4.6 x 106
10.8 x 106 46,500 8.3 x 106
-------
Table 33. CONTROL EQUIPMENT COSTS FOR MUNICIPAL INCINERATORS
Model Plant
Industry Total
I. Source description
Capacity
Production
Fine particle emissions
(a) Uncontrolled
(b) Present
II. Best installed control
system—'
Installed cost
Annualized cost
Fine particle emissions
III. 10% Opacity regulation!/
Installed cost
Annualized cost
Fine particle emissions
IV. 57o Opacity regulation®/
Installed cost
Annualized cost
Fine particle emissions
10 tons/hr
80,000 tons/year
45.6 Ib/hr
a/
$170,000
$27,000/year
1.5 Ib/hr
$116,000
$19,500/year
10.3 Ib/hr
$136,000
$23,000/year
5.7 Ib/hr
3,300 tons/hr
18 x H)6 tons/year
15,000 Ib/hr
13,100 Ib/hr
$55.0 x 106£/
$8.9 x 106/year£/
490 Ib/hr
$38.0 x 106£/
$6.4 x 106/year£/
3,400 Ib/hr
$45.0 x 106£/
$7.6 x 106/year£/
1,900 Ib/hr
a/ Quantity emitted depends on control device installed.
b_/ Control device required to meet standard is a 99.07» efficient electro-
static precipitator.
c_/ Based on fact that few, if any, municipal incinerators are presently
equipped with electrostatic precipitators.
d/ Control device selected to meet standard is 93.17o efficient electro-
static precipitator.
e/ Control device selected to meet standard is 96.270 efficient electro-
static precipitator.
123
-------
Table 34. INCREMENTAL ANNUALIZED COSTS FOR MUNICIPAL INCINERATORS TO MEET
FINE PARTICLE EMISSION STANDARDS
Emission Standard
10% Opacity
5% Opacity
All incinerators
install BICD
Control System
93.1% Efficient electro-
static precipitator
96.2% Efficient electro-
static precipitator
99.0% Efficient electro-
state precipitator
Annualized Cost ($/Year)
Model Plant All Plants
Incremental Annualized
Cost ($/Year)
Model Plant All Plants
19,500 6.4 x 106 19,500 6.4 x 106
23,000
27,000
7.6 x 106
8.9 x 106
3,500
7,500
1.2 x 106
2.5 x 106
I-1
to
-------
Table 35. CONTROL EQUIPMENT COSTS FOR IRON FOUNDRY CUPOLAS
I. Source description
Capacity
Production
Fine particle emissions
(a) Uncontrolled
(b) Present
II. Best installed control
systemk/
Installed cost
Annualized cost
Fine particle emissions
III. 10% Opacity regulation!/
Installed cost
Annualized cost
Fine particle emissions
IV. 5% Opacity regulation^/
Installed cost
Annualized cost
Fine particle emissions
Model Plant
10 tons/hr
10,000 tons/year
17.5 Ib/hr
a/
$60,400
$12,100/year
0.27 Ib/hr
$27,600
$22,400/year
4.5 Ib/hr
$34,500
$32,800/year
3.0 Ib/hr
Indus try Total
Unknown
13.1 x 106 tons/year
22,900 Ib/hr
13,100 Ib/hr
$79.1 x 106£/
$15.8 x 106/year£/
356 Ib/hr
$36.2 x I06e/
$29.4 x 106/year£/
5,900 Ib/hr
$45.2 x I06e/
$43.0 x 106/yeare/
4,000 Ib/hr
a/ Quantity emitted depends on control device installed.
b_/ Control device required to meet standard is a fabric filter.
£/ Based on the fact that only a small percentage (< 5%) of cupolas are now
equipped with fabric filters.
d/ Control device selected to meet standard is 94.2% efficient wet scrubber.
e/ Based on information indicating that few plants are equipped with control
devices that meet the required efficiency.
f/ Control device selected to meet standard is 97.1% efficient wet scrubber.
125
-------
Table 36. INCREMENTAL ANNUALIZED COSTS FOR IRON FOUNDRY CUPOLAS TO MEET FINE
PARTICLE EMISSION STANDARDS
Emission Standard
All cupolas install
BICD
10% Opacity
5% Opacity
Control System
Fabric filter
94.2% Efficient wet
scrubber
97.1% Efficient wet
scrubber
Incremental Annualized
Annualized Cost ($/Year) Cost ($/Year)
Model Plant All Plants Model Plant All Plants
12,100
15.8 x 106 12,100 15.8 x 10
6
22,400 29.4 x 106 10,300
32,800 43.0 x 106 20,700
13.6 x 106
27.2 x 106
to
-------
However, the electrolytic cells are one of the major particulate sources
within the industry and are the only ones on which enough information was
available to attempt estimation of costs for the emission standards con-
sidered in this program.
Primary aluminum plants may range in size from 150 to 700 tons/day, but
these usually consist of several "potlines" with each potline made up of
several hundred cells. Information obtained in the study indicated 15
cells might be ducted to the same control device. Therefore, a 15-cell
potline producing an estimated 3,000 tons/year of aluminum was selected
as the basis for the estimate of control costs.
Table 37 presents the cost estimates for the best installed control sys-
tem. The Alcoa 398 process was used as the best installed system, and
it was assumed to be equivalent to a fabric filter for cost purposes. How-
ever, the capture efficiency for the associated hooding systems was as-
sumed to be 95%.
Information was not available on characteristics of the effluent from the
electrolytic cells that would permit estimation of control device effi-
ciencies required to meet the opacity standards.
Primary Copper - In the U.S., 14 smelters represent 98% of the U.S.
capacity and these range in size from 48 tons/day up to 1,000 tons/day.
The configuration of these smelters is quite varied. A 230 ton/day
smelter was selected as the basis for this study to estimate the partic-
ulate control costs. Although a significant number of these plants may
have equipped some sources with S02 removal or recovery processes, the
costs of these processes and emissions from them have not been considered.
The 230 ton/day model plant consisted of one roaster, one reverberatory
furnace, and two convertors. It was assumed that two identical control
devices would be installed to control emission from the above sources and
that the average particle size distribution was 25% < 3 u.
Table 38 presents the cost estimates for the best installed control system-
assumed to be a 99.7% efficient electrostatic precipitator. It was also
assumed that none of the plants are now equipped with control systems (or
S02 removal processes) that would meet the required efficiency of 99.770.
Information was not available on characteristics of the effluents that
would permit estimation of control device efficiencies required to meet
the opacity standards.
127
-------
Table 37. CONTROL EQUIPMENT COSTS FOR PRIMARY ALUMINUM-ELECTROLYTIC CELLS
Model Plant
Industry Total
I. Source description
One potline (15 cells)
Production
Fine particle emissions
(a) Uncontrolled
(b) Present
II. Best installed control
systemk/
Installed cost
Annualized cost
Fine particle emissions
3,000 tons/year 4.0 x 10^ tons/year
29.1 Ib/hr
a/
$71,200
$16,000/year
1.5 Ib/hr
48,500 Ib/hr
16,200 Ib/hr
$95.0 x
$21.3 x I06/year£/
2,000 Ib/hr
a/ Quantity emitted depends on control device installed.
b_/ Control device required to meet standard is the Alcoa 398 control
system, which was considered equivalent to a fabric filter for
estimation purposes.
£/ Based on fact that few plants ate now equipped with the specified con-
trol device.
d/ Emissions based on estimated capture efficiency of 95%.
128
-------
Table 38. CONTROL EQUIPMENT COSTS FOR PRIMARY COPPER PLANTS
Model Plant
Indus try Total
I. Source Description
Production
Fine particle emissions
(a) Uncontrolled
(b) Present
II. Best installed control
systemS/
Installed cost
Annualized cost
Fine particle emissions
76,000 tons/year 1.7 x 10° tons/year
1,450 lb/hr£/
b/
$532,000
$84,000/year
16.7 lb/hra/
32,400 lb/hr£/
7,400 lb/hr£/
$11.9 x
$1.9 x 106/yearl/
374 lb/hr£/
a/ Emissions based on assumption that uncontrolled particle size is 25%
< 3 11.
b_/ Quantity emitted depends on control device installed.
£/ Control device selected to meet standard is 99.7% efficient electro-
static precipitator. S02 removal process costs and emissions have
not been considered.
d_/ Costs are based on assumption that plants are not now equipped with
control that would meet the required efficiency (99.7%).
129
-------
ESTIMATED REDUCTIONS IN FINE PARTICLE EMISSIONS
Implementation of emission standards based on particle size can achieve
significant reductions in the fine particulate burden entering the atmo-
sphere from stationary sources. The extent of the reduction depends upon
the type(s) of emission standards that are selected for implementation.
Table 39 presents a summary of estimated reductions in fine particle
emissions from selected industrial sources which might result from the
implementation of the opacity or BICD emission standards. The time re-
quired to achieve the installation of requisite control equipment on all
the sources of fine particulate pollutants is difficult to determine.
If one assumes that fine particle emission standards are proposed, adopted,
and implemented in the 1975-1980 period, then it seems reasonable to assume
that total compliance could be achieved by 1985.
The emission figures presented in Table 39 are based on current production
rates. Production rates have generally increased for industrial sources
as a function of time. Changes in production rates must be included in
calculations when an attempt is made to estimate fine particle emission
levels which may result from control efforts in future years. As a part
of a previous study, MRI made projections of fine particle emissions using
two different methods.ILL/ Table 40 presents the results of these projec-
tions for the same group of sources listed in Table 39.
The projection of emissions by Method I assumed that there would be no
change in the net control for a source. This assumption results in an
increase in emissions in proportion to increases in production capacity.
Production figures were projected by the same methods outlined in Ref. 92
and were used to proportionally increase the current mass of fine particle
emissions.
Method II takes into account the increase in production, but two assump-
tions are also made:
1. All sources would be controlled by 1980, i.e., application of control
will reach 100% by 1980.
2. Increased utilization of the most efficient control devices would
continuously increase the efficiency of control on fine particles so
that by the year 2000 it would be equivalent to controlling all sources
with fabric filters. The fabric filter was selected as a reference
standard of performance because it is one of the most efficient devices
currently available for collecting fine particulates. It seems realistic
to assume that improvements in the other control devices and possible
130
-------
Table 39. ESTIMATES OF FINE PARTICLE EMISSIONS AS A FUNCTION OF EMISSION STANDARD
LO
Estimated Fine Particle Emissions
Uncontrolled
I.
II.
III.
IV.
V.
VI.
VII.
VIII.
Source
Coal combustion
Iron and steel
A. Sinter machines
B. Basic oxygen furnace
C. Electric arc furnace
Cement plants, rotary kilns
Asphalt plants, dryers
Ferroalloy plants
A. Closed electric furnace
B. Hooded open electric
furnace
C. Unhooded open electric
furnace
Lime plants, rotary kilns
Municipal incinerators
Iron foundry, cupola
Lb/Hr
796,000
7,100
400,000
27,400
243,000
3,260,000
92,000
129,500
72,000
75,800
15,000
22,900
Ton/Year
3,184,000
28,400
1,600,000
109,600
972,000
1,956,000
368,000
518,000
288,000
303,200
37,500
22,900
Present
Lb/Hr
243,000
1,400
43,600
3,600
44,300
257,000
11,300
84,100
60,800
22,000
13, 100
13,100
Ton/Year
972,000
5,600
174,400
14,400
177,200
154,200
45,200
336,400
243,200
88,000
32,800
13,100
BICD Standard^./
Lh/Hr
15,000
64
2,560
645
2,000
25,600
11,300
15,800
28,800
1,200
490
356
Ton/Year
60,000
256
10,240
2,580
8,000
15,360
45,200
63,200
115,200
4,800
1,225
356
10% Opacity Standard 5% Opacity Standard
Lb/Hr
5,050
722
1,540
1,036
25,200
200,000
NCS/
NC
NC
10,500
3,400
5,900
Ton/Year Lb/Hr
20,200 2,510
2,888 354
6,160 800
4,144 683
100,800 12,200
120,000 167,000
NC NC
NC NC
NC NC
42,000 4,800
8,500 1,900
5,900 4,000
Ton/Year
10,040
1,416
3,200
2,732
48,800
100,200
NC
NC
NC
19,200
4,750
4,000
a/ Not calculated.
b_/ BICD-Best installed control device (not necessarily the highest efficiency device available, but rather the best that is generally being installed
at present time).
-------
Table 40. PROJECTIONS OF FINE PARTICLE EMISSIONS FROM INDUSTRIAL SOURCES
(106 Tons/Year)
1.
2.
3.
4.
5.
6.
7.
8.
Source ,
Stationary combustion
A. Coal
1. Electric utility
2 . Industrial
Iron and steel
A. Sinter machines
B. Basic oxygen furnace
C. Electric arc furnace
Cement plants, rotary kilns
Hot mix asphalt plants, dryers
Ferroalloy electric furnaces
•Lime plants, rotary kilns
Municipal incinerators
Iron foundries, cupolas
Method
I
II
I
II
I
II
I
II
I
II
I
II
I
II
I
II
I
II
I
II
I
II
1970
1.046
0.976
0.133
0.127
0.006
0.006
0.188
0.177
0.015
6.014
0.188
0.175
0.168
0.158
0.155
0.131
0.103
0.095
0.040
0.037
0.013
0.012
1975
1.201
0.883
0.150
0.111
0.007
0.006
0.293
0.235
0.018
0.010
0.227
0.164
0.205
0.158
0.166
0.074
0.134
0.094
0.051
0.037
0.015
0.010
1980
1.374
0.731
0.170
0.084
0.008
0.005
0.454
0.305
0,021
0.004
0.273
0.140
0.250
0.152
0,180
0.008
0.174
0.086
0.061
0.031
0.016
0.006
Year
1985
1.600
0.653
0.170
0.064
0.009
0.004
0.515
0.277
0.023
0.004
0.319
0.125
0.305
0.141
0.198
0.009
0.189'
0.071
0.071
0.027
0.017
0.005
1990
1.864
0.528
0.170
0.043
0.009
0.003
0.581
0.237
0.026
0.004
r
0.372
0.101
0.375
0.117
0.219
0.010
0.206
0.053
0.081
0.021
0.017
0.003
1995
2.110
0.335
0.166
0.023
0.010
0.002
0.618
0.170
0.029
0.003 .;
0.421
0.062
0.460
0.074
0.241
0.011
0.311
0.044
0.095
0.-014
0.018
0.002
2000
2.388
0.090
0.164
0.002
0.010
0.000
0.654
0.094
0.034
0.003
0.476
0.016
0.563
0.018
0.267
0.012
0.466
0.016
0.109
0.003
0.019
0.003
-------
development of new devices would allow the efficiency of control devices,
in the year 2000, to match the performance capability of the best devices
that are presently available. Details and examples of the calculations
for both methods are given in Ref. 61.
The two projection methods represent two extremes: no improvement in con-
trol versus stringent control. An intermediate position is more probable,
but projection of an intermediate situation requires information regarding
the percent application of each type of control device on specific sources
in the future years. Of course, this information is not available, and
to project emissions, gross assumptions would have to be made regarding
future equipment utilization.
ECONOMIC IMPACT OF FINE PARTICLE EMISSION STANDARDS ON SELECTED SOURCES
A detailed economic evaluation of the economic impacts of emission stand-
ards involves far more than simply making cost, revenue and capital
projections over a specified time period. A thorough feasibility analy-
sis must give careful attention to how all the different parts fit to-
gether and react with each other, in much the same way as a critical
path analysis describes the interactions among all the activities in a
complex construction project.
Just as in a natural ecosystem, many of the elements in the economic sys-
tem are closely interrelated, even though—just as in the ecosystem—the
interrelationships may not be instantly obvious, and even when they are,
they may be hard to express in quantitative or absolute terms.
The accompanying illustration (Figure 19) shows, in a general sense, some
of the major microeconomic factors that should be considered in the economic
analysis of the impacts of air quality control programs on a given firm or
industry.
Air[pollution control standards will primarily influence unit costs. Unit
costs include both variable and fixed elements. The variable costs, con-
sisting mainly of labor and materials, are related to volume; any change
in volume is accompanied by a proportionate change in the direct costs.
Fixed costs, consisting mainly of capital-related charges and overhead
expenses, are primarily related to the amount of capital invested in the
operation and remain constant over a wide range of operating conditions.
In total, unit costs generally decline as volume increases, as the fixed
costs are spread over a broader base. The imposition of air pollution
control standards will affect the firm's costs of production by increasing
its capital investment requirements, and by raising its direct manufacturing
133
-------
RETURN ON
INVESTMENT
Figure-19.
Interrelationships among factors affecting the economics
of .a firm
134
-------
costs by whatever amount is required to operate and maintain the new air
pollution control facilities.
Figure 20 illustrates a possible structure for analyzing the economic im-
pact of new air pollution control requirements on a specific firm. The
specified control efficiency and plant size (as reflected by its prdduc-
tion volume) dictate the size and type of control equipment needed, which
in turn determine the capital and operating costs associated with the
control program.
The total cost of control represents an incremental cost "of production,
which can either be absorbed by the firm or passed cm to its consumers
by means of increased prices. Should prices be increased, however, sales
are likely to decrease; if costs are absorbed, volume may remain constant,
but profit margins will suffer. Regardless of the economic impact on the
firm, then,-tits way of doing business will be affected.
t s-'
Table 41 shows an example of the economic impact of air pollution control
requirements on an electric generating plant, as reflected in its costs
of power production. Again, the control efficiency determines the control
cost, which results in higher costs of energy production, ultimately to
be passed along to the consumer. As is evident from the control costs
shown in Table 41, control becomes increasingly expensive as the effi-
ciency requirements increase. This is particularly significant in
certain industries, where any price increase will cause a major decline
in sales, with customers either cutting back on their consumption^ shift-
ing to other producers, or substituting other products.
An in-depth analysis of the economic impact of fine particle emission
standards on individual industry segments was outside the scope of this
study. The economic impact analysis conducted during this program was
confined to the determination of the increase in production costs for
specific industries. An example of the procedures used to calculate the
increase in production costs associated with the fine particle emission
standards is presented in the next section.
Example of Calculation of Economic Impact
The financial and operating characteristics of a typical coal-fired steam-
electric generating plant were developed to use as a model for determining
the economic effects of different control strategies for fine particulates.
Based on the compiled statistics of privately owned electric utilities in
the United States, generating costs are equally divided between capital
charges and production expenses, with capital charges averaging 15% of
135
-------
Specified Control
Control Efficiency Equipment
1
i 1
Capital Operating Control
Requirements Costs Credits
i
i i
Sources of Annual Cost Net Operating
Capital of Capital Cost
1 1
Total Cost Production
of Control Volume
1
Incremental
Unit Cost
Absorb Increase
Costs Prices
1
4 * 1
Product Competitive Price Elasticity
Substitution Factors of Demand
1 i '
1
Economic Impact Sales'
on Firm: Volume
• Profit
• Profit Margin
• Return on
Investment
'
Figure 20. Structure for microeconomic impact analysis
136
-------
Table 41. EXAMPLE OF ECONOMIC IMPACT OF AIR POLLUTION CONTROL REQUIREMENTS ON ELECTRIC POWER
GENERATING COSTS
u>
Control Method
None
Louver
Medium efficiency
cyclone
High efficiency
cyclone
Multiple cyclones
Electrostatic
precipitator
Fabric filter
Fabric filter
Fabric filter
Control
Efficiency
(%)
0
58.6
65.3
84.2
93.8
99.0
99.7
99.8
99.9
Total Annual Cost ($1,
Electric
Production
9,010
9,010
9,010
9,010
9,010
9,010
9,010
9,010
9,010
Pollution
Control
0
100
96
156
168
602
852
1,008
1,190
000)
Total
9,010
9,110
9,106
9,166
9,178
9,612
9,862
10,018
10,200
Net
Cost
(C/kwh)
0.935
0.945
0.945
0.951
0.952
0.997
1.023
1.039
1.058
Incremental
Cost
(C/kwh)
--
0.010
0.000
0.006
0.001
0.045
0.026
0.016
0.019
-------
the utilities net plant investment. Table 42 summarizes the pertinent
financial characteristics of the 400-MW model steam generating plant burn-
ing 1.2 million tons of coal annually to produce 2.4 billion kilowatt-
hours of electric energy.
As shown in Table 42 the total annual cost of service for this typical
plant comes to $36 million, or $0.015/kwh of net generation, exclusive of
emission control equipment.
Virtually any desired level of particulate emission control can be achieved
at a price. Table 43 shows the estimated control costs associated with
various control efficiencies, ranging from the best installed control de-
vice (BICD), an electrostatic precipitator with 99.0% efficiency to 99.9%
efficiency, and including the 10% opacity level (99.66% collection effi-
ciency) and the 5% opacity level (99.83% efficiency).
Estimated annual control costs vary from $241,500 at the 99.0% efficiency
level to $465,000 at 99.9% efficiency. Here, annual charges are estimated
at 15.0% of the total installed cost of the control equipment, the same
rate applied to the utility's other plant facilities. This, it should be
pointed out, is the lowest capital charge rate that would conceivably
apply to the control equipment, and a rate in the 17 to 20% range is
likely to be applied to this type of air pollution control equipment.
At control efficiencies in the 99.0 to 99.9% range and with capital charges
computed at 1570 of installed equipment costs, the total impact of air pol-
lution control costs on the net cost of electric generation is almost
negligible, as shown in Table 44. Here it can be seen that net costs per
kilowatt-hour ($0.015 without emission controls) increase only to $0.0152
even at the 99.9% control level. At 10% opacity, total control costs
amount to just $0.00013/kwh, and at 5% opacity they increase only to
$0.00016/kwh.
Thus, the net cost difference between zero control and control at the 5%
opacity (or 99.83% efficiency) level amounts to an increase of 1.3% in
terms of the cost of generating electrical energy under the prescribed con-
ditions. By way of comparison, O'Connor and Citarella reported that the
total annual cost of particulate air pollution for a 600-MW steam-electric
plant ranges from 0.7 to 2.0% of the total annual cost of power.^-L'
Control Costs in Selected Industries
The costs of fine particulate control associated with appropriate control
levels—BICD or specified opacity levels—have been estimated for the in-
dustries representing the main sources of fine particulate emissions.
138
-------
Table 42. TYPICAL FINANCIAL CHARACTERISTICS OF 400-MW ELECTRIC
GENERATING PLANT^/
Net plant investment at $300/kw $120,000,000
Operating revenues at $0.015/kwh 36,000,000
Operating expenses
Fuel - 1.2 x 106 tons at $7.00/ton 8,400,000
Other production expenses at $0.004/kwh 9.600,000
Total operating expenses 18,000,000
Capital charges]*/ at 15% of plant investment 18.000,000
Total cost of service $ 36,000,000
a/ Net annual generation = 2.4 x 10^ kwh; load factor = 68.5%.
b/ Includes depreciation, interest, A&G expenses, ad valorem and income
taxes, and return on investment.
139
-------
Table 43. ESTIMATED CONTROL COSTS FOR 400-MW ELECTRIC GENERATING PLANT
Installed Cost Annual
Emissions
- . i
Controlled
m
99.00
99.50
99.60
99.66
99.70
99.80
99.83
99.90
Emissions
Uncontrolled
(7.)
1.00
0.50
0.40
0.34
0.30
0.20
0.17
0.10
of Control
Equipment
($)
1,030,000
1,280,000
1,355,000
1,400,000
1,430,000
1,530,000
1,570,000
1,670,000
Operating and
Maintenance Cost
($>
87,000
100,000
105,000
110,000
114,000
132,000
143,000
215,000
Annual
Capital
Charges
($)
154,500
192,000
203,250
210,000
214,500
229,500
235,500
250,500
Total
Annual
Control Cost
' r$)
3
241,500
292,000
308,250
320,000
i. 328, 500
^361-, 500
378,500
465,500
Table 44. EFFECT OF CONTROL COSTS ON COST OF ELECTRIC POWER
Emissions
Controlled
(7.)
No control
99. 00*/
99.50
99.60
99,66k/
99.70
99.80
99.83£/
99.90
Total Cost
of Power
Generation
($/Year)
36,000,000
36,000,000
; 36,000,000
36,000,000
36,000,000
36,000,000
36,000,000
36,000,000
36,000,000
Total Annual
Control Cost
($)
0
241,500
292,000
308,250
320,000
328,500
361,500
378,500
465,500
Total Annual
Cost
($)
36,000,000
36,241,500
36,292,000
36,308,250
36,320,000
36,328,500
36,361,500
36,378,500
36,465,500
Cost/kwh
m
0.01500
0.01510
0.01512
0.01513
0.01513
0.01514
0.01515
0.01516
0.01519
a/ BICD.
b/ 107. opacity.
£/ 57. opacity.
140
-------
Capital charges have been computed at four different levels, thus encom-
passing the entire range of possibilities for a given plant. The specific
levels selected represent the minimum depreciable life established by
the Internal Revenue Service for approved pollution abatement equipment;
the maximum expected useful service life of typical combustion equipment
for which control facilities are required; and two intermediate values,
reflecting the probable economic life ranges of the particulate control
devices being installed.
The four capital charge rates used in converting the installed cost of
particulate control systems to an equivalent annual amount are made up
as follows:
Straight-Line Average Total Annual
Economic Depreciation Interest Other Capital Capital Charge
Life Year Rate (%) at 8% Related Expenses (%) Rate (%)
5 20 5 5 30
10 10 5 5 20
15 7 5 5 17
20 5 5 5 15
The average interest is computed at 8% of the undepreciated account bal-
ance, and, added to the straight-line depreciation rate, constitutes the
"capital recovery rate" used to amortize an investment over a specified
time period. Other capital-related expenses include insurance costs, ad (
valorem taxes, and administrative expenses.
Total annual costs, including both capital charges and direct operating
expenses, were computed for each model plant, at selected control levels,
for each capital charge rate. These annual costs were then related to
the model plant's production rate, resulting in a net annual control cost
per unit of production. This amount, related to the value of the end
product, provides a realistic measure of the economic impact of a given
control strategy under the prescribed conditions.
Table 45 summarizes the estimated total annual control costs for model
plants in selected industries. Estimated costs are presented for the
BICD, 10% opacity and 5% opacity emission standards. Table 46 presents
the estimated costs associated with the BICD standard for both the
model plants and the total industry.
141
-------
Table 45. EFFECT OF CONTROL CRITERIA ON CONTROL COSTS IN SELECTED INDUSTRIES
to
Source
i'x..
Coal-fired electric plant
Municipal .incinerator
Cement plant (rotary kiln)
Asphalt plant (rotary dryers)
Iron and steel
(a) Basic oxygen furnace
(b) Electric arc furnace
(c) ;. Sintering (i&ndbox)
Lime plant (rotary kiln)
Iron foundry cupola
Control Fine Particle
Criteria Control Efficiency (%)
BICD 98.13 '
19% Opacity ,. 99.36
5% Opacity 99.68
BjgD
107. Opacity
5% Opacity
BICD
10% Opacity
5% Opacity
BICD
10% Opacity
5% Opacity
BICD
10% Opacity
5% Opacity
BICD
10% Opacity
5% Opacity
BICD
10% Opacity
5% Opacity
BICD,
10% Opacity
5% Opacity
BICD
10% Opacity
5% Opacity
96.71
77.41
87.50
99.17
89.58
94.94
99/21
92.86
94.05
99.70
99.82
99.91
97.64
93.54
97.24
99.10
87.76
94.05
" I/.
98.43
78.04
90.59
98.46
74.29
82.86
Model Plant Total Annual Control Cost ($1,000)
Annual Production at Specified Capital Charge Rate
Rate ' 0.15 0.17 0.20 .<*' 0.30
2.4 x 109 kwh ' 241.5
320.0
- 378.5
80,000 tons 35.5
25.3
29.8
3.0 x 10^ bbls 125.1
89.2
98.1
90, 000. tons 13.3
20.9
31.2
1.0 x 106 tons 211.0
227.0
255.0
100,000 tons 50.1
36.1
42.2
1.46 x 106 tons 277.0
184.5
204.5
87,500 tons 17.5
41.6
63.4
10,000 tons 15.1
23.8
34.5
262.1
348.0
409.9
38.9
27.6
32.5
134.3
98.1
107.7
14.3
21.3
31.8
230.2
247.0
2 74'. 2
53.7
39.3
46.3
297.0
199.9
221.5
18.9
42.4
64.6
16.3
24.3
35.2
293.0
390.0
457.0
44.0
31.1
36.6
148.2
111.4
122.1
15.9
22.0
32.7
259.0
277.0
303.0
59.2
44.1
52.4
327.0
223.0
247.0
21.0
43.7
66.3
18.1
25.2
36.3
396.0
530.0
614.0
61.0
42.7
50.2
194.4
155. f
170.2
21.2
24.1
35.6
355.0
377.0
399.0
77.4
60.2
72.8
427.0
300.0
332.0
28.0
47.9
72.1
24.2
27.9
39.7
-------
Table 46. TOTAL ANNUAL COST OF ACHIEVING BICDS/-LEVEL CONTROL OF FINE PARTICULATES IN SELECTED INDUSTRIES
CO
...t-.
Source
Coal-fired electric plant
• f- *- •
Municipal incinerator
Cement plant (rotary kiln)
Model Plant
Production
Rate/Year
2.4 x 109 kwh
80,000 tons
3.0 x 106 bbls
Total Annual Control Cost for
Model Plant at Specified
Capital Charge Rate ($1,000) "
0.15 0.17 0.20 0.30
241.5 262.1 293.0 396.0
35.5 38.9 44.0 61.0
125.1 134.3 148.2 194.4
Total Annual
" Industry
Production
516 x 109 kwh
18 x 106 tons
400 x 106 bbls
-Total Annual Control Cost for
Industry at Specified Capital
Charge Rate ($106)
0.15 0.17 0.20 0.30
51.9 56.4 63.0 85.1
8.0 8.8 9.9 13.7
16.7 17.9 19.8 25.9
Asphalt plant (rotary
dryers)
Iron and steel
(a) Basic oxygen furnace
(b) Electric arc furnace
(c) Sintering (windbox)
Ferroalloy
(a) Unheeded open electric
90,000 tons
1.0 x 106 tons
100,000 tons
1.46 x 106 tons
13.3
furnace
(b) Hooded open electric
furnace
(c) Closed electric
furnace
Lime plant (rotary kiln)
Iron foundry cupola
8,000 tons
10,800 tons
13,600 tons
87,500 tons
10,000 tons
205.5
51.4
3.8
17.5
15.1
14.3
15.9
21.2 350 x 106 tons 51.7 55.6 61.8 82.4
211.0 230.2 259.0 355.0 50 x 106 tons 10.6 11.5 13.0 17.8
50.1 53.7 59.2 77.4 21.5 x 1Q6 tons 10.8 11.5 12.7 16.6
277.0 297.0 327.0 427.0 54 x 106 tons 10.2 11.0 12.1 15.8
205.5 220.5 243.0 318.0 600,000 tons
15.4 16.5 18.2 23.9
51.4 55.2 60.8 79.6 1.08 x 106 tons 5.1 5.5 6.1 8.0
4.1 4.5 6.0 820,000 tons 0.2 0.2 0.3 0.4
18.9 21.0 28.0 , 18.5 x 106 tons 3.7 4.0 4.4 5.9
16.3 18.1 24.2 13.1 x 106 tons 19.8 21.4 23.7 31.7
a/ See footnote (b), Table 1, for definition of BICD.
-------
Table 47 summarizes the estimated annual costs of controlling fine partic-
ulate emissions via the BICD standard in selected industries, expressed in
terms of appropriate production units for each industry. Table 48 presents
the unit values of the output from selected industries which were used to
develop the data in Table 47.
As is evident from the data in Table 47, the economic impact of .fine par-
ticulate control will vary substantially from industry to industry. For
example, at the 20% capital charge rate (reflecting a 10-year equipment
life) applied to BICD-level control of coal-fired electric generating
plants, the $0.00012/kwh cost represents < 1.0% of a typical utility's
total cost of service. For municipal incinerators, though, control costs
could easily add 8 to 10% to the total waste disposal costs. Control
costs for cement plants and asphalt plants will fall gnerally in the 1.5
to 4.0% range, while various metallurgical operations in the ferroalloys
industry may require control expenditures ranging anywhere from < 0.5% to
> 20% of the value of their products.
144
-------
Table 47. SUMMARY OF FINE PARTICULATE CONTROL COSTS AT BICD-LEVEL CONTROL
($/Unit of Production)
U1
Source
Coal-fired electric plant
Municipal incinerator
Cement plant (rotary kiln)
Asphalt plant (rotary dryers)
Iron and steel
(a) Basic oxygen furnace
(b) Electric arc furnace
(c) Sintering (windbox)
Ferroalloy
(a) Unheeded open electric furnace
(b) Hooded open electric furnace
(c) Closed electric furnace
Lime plant (rotary kiln)
Iron foundry cupola
Primary aluminum (electrolytic cells)
Primary copper
Annual Produc-
tion Rate for
Model Plant
2.4 x 109 kwh
80,000 tons
3.0 x 106 bbls
90,000 tons
1.0 x 10^ tons
100,000 tons
1.46 x 106 tons
8,000 tons
10,800 tons
13,600 tons
87,500 tons
10,000 tons
3,000 tons
76,000 tons
Fine
Particle
Control
Efficiency
98.13
96.71
99.17
99.21
99.70
97.64
99.10
60.00
87.80
97.52
98.43
98.46
94.85
98.85
Capital Charge Rate
0.15
0.00010
0.444
0.042
0.147
0.211
0.501
0.190
25.688
4.759
0.276
0.200
1.512
6.52
1.455
0.17
0.00011
0.486
0.045
0.159
0.230
0.537
0.203
27.563
5.107
0.298
0.216
1.613
6.99
1.595
0.20
0.00012
0.550
0.049
0.177
0.259
0.592
0.224
30.375
5.630
0.331
0.240
1.814
7.71
1.805
0.30
0.00017
0.763
0.065
0.236
0.355
0.774
0.292
39.750
7.370
0.441
0.320
2.418
10.08
2.505
-------
Table 48. UNIT VALUE OF OUTPUT FROM SELECTED INDUSTRIES
Industry or Product
Electric energy
Units
Ki lowat t-hour
Municipal refuse incineration Ton
Cement
Asphalt
Pig iron
Steel
Lime
Barrel
Ton
Ton
Ton
Ton
Unit Cost or Value ($)
0.015
10.00
4.32
23.50
78.16
187.26
18.00
146
-------
SECTION
BENEFIT/COST RELATIONSHIPS FOR FINE PARTICULATE CONTROL
INTRODUCTION
Particulate emissions from a single point source will generally disperse
over a wide area at low concentrations. The specific area over which the
dispersion occurs will depend on a variety of meteorological and topographi-
cal factors, as will the particulate concentrations experienced at any given
point within the area.
The economic effects of airborne particulate matter are a function of the
particulate concentration, the area over which the dispersion has occurred,
the economic values exposed to various concentrations, and the rate of
economic loss associated with the interaction between the pollutants and
the economic values.
Five categories of economic loss or economic damage can be attributed to
the presence of air pollutants. These include effects on: (1) human health;
(2) animals; (3) vegetation; (4) materials; and (5) aesthetics. Consider-
able research effort has been devoted to identifying the economic effects
of air pollution in each of these areas, but very little has been accom-
plished in quantifying the effects. Because of the lack of reliable
quantitative data on effects, only a generalized benefit/cost analysis was
conducted during this program. The following sections of this chapter
present the methodology utilized for the benefit/cost analysis and the
main results of the analysis.
DETERMINATION OF ECONOMIC DAMAGE ATTRIBUTABLE TO A SPECIFIC SOURCE
The approach used to allocate quantitatively the economic damages attribut-
able to a given emission source is based on the following assumptions:
1. Potential economic damage is related to the quantity of particles dis-
charged into the atmosphere.
147
-------
2. The damage potential is highest near the emission source, decreasing
as the distance from the source increases.
3. The total economic damage will be the integrated effects of pollutant
concentrations, interaction coefficients, and exposed economic values
throughout the entire dispersion area.
4. This integrated total economic damage will be the same as if it were
calculated on the basis of an average pollutant concentration, affecting
an average economic value, at an average rate of interaction.
5. The particulate concentrations at which the economic damage will be
evaluated are 75 ug/m3 and 60 ug/m^. The former concentrations correspond
to the primary ambient air standard for particulates, while the latter is
the secondary ambient air standard.
6. A point source having any specified emission rate (Ib/hr) of particu-
lates with an average atmospheric residence time of "t" (hr) will generate
a sufficient quantity of particles to produce a particulate concentration
of 75 and 60 ug/m3 in a given volume of air.
7. Given the volume of air having a particulate concentration of 75 and
60 ug/m3, and further given the geometric shape of the volume containing
that concentration, the area of influence can be calculated.
8. Lacking specific meteorological data, a hemisphere provides a reasonable
representation of the atmospheric diffusion of particulate matter from a
stationary point source.
9. The area of influence is defined as the area over which a single emis-
sion source could cause and sustain a particulate concentration of 75 and
60 ug/m3, assuming uniform dispersion throughout a hemisphere, with the
point source located at its center.
10. The economic loss factor, or damage intensity rate, is defined as the
total annual economic damage incurred per square mile within the area of
influence that would be experienced as a direct result of continuous ex-
posure of all economic values within the area of influence to a sustained
particulate concentration of 75 and 60 ug/m3.
11. The total economic damage allocable to a point source is the product
of the area of influence (mile2) and the damage intensity rate ($/mile2).
148
-------
12. The economic benefits resulting from pollution control measures rep-
resent, the difference between the total economic damage incurred from an
uncontrolled emission and the total economic damage incurred from the re-
duced level of pollutants discharged from the controlled source.
13. Economic values and loss factors are assumed to remain constant through-
out the area influenced by a specific pollutant source. However, the area
of influence decreases as the quantity of particles discharged decreases,
and the total economic losses will decrease in direct proportion to the area
of influence.
14. The area of influence varies with the 2/3 power of the quantity of
particles discharged. Thus, an order-of-magnitude increase in particulate
emissions would increase the area of influence, and also the total economic
damage, by IQ0-66?, Or 4.63 times its previous level. Similarly, a 90%
control efficiency on a particulate source would decrease the area of in-
fluence and total economic loss to 21.4% of its former level.
It must be recognized that the economic losses computed in this manner are
allocated rather than actual; actual losses would be incurred over a much
larger area at a much lower rate. However, the total economic damage should
be essentially comparable.
Computing the Impact Area
The radius of the hemispheric dispersion model (see Figure 21) is a func-
tion of the emission rate and the residence time in the atmosphere of the
particles emitted. Equations (6) and (7) present the expressions for the
radius of the hemisphere for ambient concentrations of 75 and 60 ug/m3}
respectively.
R = 0.0867 (Et)l/3 for 75 ug/m3 (6)
R = 0.0954 (Et)1/3 for 60 ug/m3 (7)
where R is the radius in miles, E the emission rate in Ib/hr, and t the
particle residence time in the atmosphere in hours.
The area of influence or impact area for each concentration level is given
by Eq. (8) or Eq. (9).
149
-------
S = Point Source of Pollutant
t
Emissions
R = Radius of Impact Area
2 = Area pfj Economic Impact ot
Critical Concentration
Figure 21. Simplified hemispherical pollutant dispersion model
150
-------
A = 0.0236 (Et)2/3 for 75 ]ig/m3 -(8)
A = 0.0286 (Et)2/3 for 60 jig/m3 (9)
where A is the impact area expressed in square miles.
Control Costs
Control costs are known to vary inversely with control efficiency. The
control cost/control efficiency relationship, furthermore, is believed to
be exponential, in that the control cost per unit of control efficiency in-
creases with increasing control efficiency. Thus, the cost of achieving
each additional percent of control efficiency costs more than did the last
percent improvement, and less than the next.
The general cost efficiency relationship for particulate control systems
has this form: '"
C = 0.8(1/U)°-3 (10)
where C is the annualized cost per CFM of control system capacity for achiev-
ing a specified level of control efficiency, with U being the percent of
total emissions remaining uncontrolled.
Using this equation, a 90% control efficiency would be expected to cost in
the neighborhood of 0.8(1/10)0.3, or $0.40/CFM of capacity. Similarly, a
99% control efficiency would cost 0.8 (1/1)°-3, or $0.80/CFM; and increas-
ing the efficiency to 99.9% would raise control costs to 0.8(1/0.I)0-3, -.
or $1.60/CFM.
Economic Damage
The economic damage associated with exposure to a 75 or 60 Jig/m3 concentra-
tion of particulate will depend on the specific characteristics of the area
affected, with population density the most important single factor.
The total economic damage to humans, animals, vegetation, materials, and
aesthetics attributable to sustained exposure to atmospheric pollutants
may range from less than $10,000/mile2 in sparsely populated rural areas,
to more than T$l million/mile2 in populous urban and industrial locatipns-.
In densely populated urban areas, effects on human health are by far the
most important; in sparsely populated rural areas, effects on animals and
151
-------
vegetation are more economically significant; and in moderately populated
areas encompassing both urban and rural characteristics, materials and
aesthetics are the chief recipients of air pollution damage.
Some effects, then, are related primarily to population density, while
others are more dependent on the characteristics of the- area exposed to
the pollutants. Data are not available, however, to permit accurate
quantification of the economic effects of air pollution in any specific
category. The estimates presented here are based on a review of the
literature, and are intended for illustrative purposes only.
Human Health Damage - The economic effect of a 75 and 60 jig/m3 fine partic-
ulate concentration on human health is estimated at $20 and $16/person/year.
These cost figures take into account premature death and losses-"incur red as
a result of air pollution related diseases such as bronchitis, lung cancer,
respiratory ailments, cardiovascular illnesses, and cancer. Particulate
pollutants are known to be a major cause of bronchitis, an important con-
tributor to lung cancer and respiratory diseases, and a minor factor in
cancer and cardiovascular diseases. The total economic cost of these five
diseases is reported to be $13.7 billion annually,-of whicli some 20% might
be attributed to air pollutants of all types.2£/ Assuming that $1.0 bil-
lion in annual health damage results from ambient particulates, the damage
can be estimated at about $5.00-$5.50/year/persoh/20 iig/m3. A 75 ug/m3
concentration, therefore, would result in damage of about $20/person/year,
while a 60 ug/m^ concentration wo'uld result in damage of about $16/persori/
year. •
Animal Health Damage - Fine particulates at the concentrations considered
here would be expected to cause little economic damage to animals, even in
predominantly agricultural areas. Cattle population in rural areas may
range from about 200/mile^ when the animals are on pasture, to as many as
60,000/mile2 in confined feedlots. However, cattle on feed are generally
kept for no more than 4 months, so air pollutants at the prescribed level
would have relatively,little impact.
Vegetation Damage - Particulate pollutants,^except for specific types (e.g.,
fluorides), are unlikely to have any significant impact on agricultural ;i:
crops or otherx vegetation.. In farming areas,_ the total va,lue of crops ex-
posed to pollutants would seldom exceed $100,QOO/mile2,, a,nd .the detrimental
effect of particulates on crop yields would.be less than 5%. The-maximum ....
expected economic loss, then, would be less than $5,000/mile2 in exclusively
agricultural areas, and would -decline with increasing population as the
value of exposed vegetation decreases.
152
-------
Materials Damage - Particulate damage to materials is almost entirely in
the form of soiling, and actual economic losses are incurred only when the
frequency (or tost) of cleaning is increased. Since the value of materials
exposed to damage (or economic loss) by soiling is closely related to popu-
lation, the resulting damages would be expected to increase with increasing
population density. It is unlikely that actual losses due to soiling at
the prescribed fine particulate concentrations of 75 and 60 ug/m^ would
amount to more than $10/person/year, and materials damage would more likely
be in the neighborhood of $4/person/year.
Aesthetic Damage - Aesthetic damage is the least quantifiable category of
economic losses suffered as a result of air pollutants, but is probably
second in importance only ,to health effects. The main aesthetic effect of
particulates in the atmosphere is related to visibility. Defining the
economic value of various levels of visibility, though, is a perplexing
problem.
There is considerable overlapping between aesthetic and materials damage,
especially in the area of soiling, in that soiling of materials does not
generally cause physical damage to most materials, but does impair human
enjoyment or appreciation of them. It is this aesthetic effect that pro-
vides the incentive to clean the dirty materials.
Aesthetic damage is related to both population and area; to both the num-
ber of people affected and to the area over which the effects are experi-
enced. On a per capita basis, then, aesthetic damages will decline as
population density increases, while in total they will rise with increas-
ing population. For the purposes of this study, aesthetic damages at the
prescribed fine particulate concentration were assumed to range from about
$50/person in sparsely populated areas down to around $8/person in congested
urban areas.
TOTAL ECONOMIC DAMAGES
The preceding discussion presented a methodology for assessing the economic
damage attributable to particulate pollutants emitted from a specific
source. To apply this procedure to a specific urban^ rural, or mixed loca-
tion with the objective of determining total economic damage from sources
in that area and the relative contribution of each aource, it is necessary
to have^data on factors such as (1) number and type of sources in a specific'
location Ci-e'» emission inventory), rates of emission from specific sources,
and economic values exposed to particulate pollutants. Detailed analyses
for specific locations were outside the scope of the current program, and
only a generalized analysis of total economic damage as a function oftpopu-
lation densities was performed.
153
-------
Table 49 summarizes the estimated economic damages resulting from fine par-
ticulate emissions for areas exposed to a concentration of 75 ug/m3 with
population densities ranging from 10 to 50,000 persons/mile2. Here, the
damages have been grouped into three categories because of the difficulties
in distinguishing between several closely related areas. The three cate-
gories are: (1) human health; (2) animals and vegetation; and (3) mate-
rials and aesthetics.
Total estimated economic damages per square mile resulting from a con-
tinuous ambient fine particulate concentration of 75 ug/m3 are shown in
Table 49 to range from $5,800/mile2 at a population density of 10 people/
mile2, up to $l,602,000/mile2 at a population density of 50,000. Annual
per capita damages incurred over the same range vary from $580 down to
$32/person
These estimated damages are plotted, both in total and on a per capita
basis, in Figure 22. Also shown in Figure 22 are population densities
typical of different geographic regions and metropolitan areas, ranging
from less than 12 people/mile2 over the State of Arizona, to some 75,000
people/mile2 in the Manhattan Borough of New York City. Average popula-
tion density in the Continental United States is 50 people/mile2.
A typical suburban community consisting of single-family residences will
have around 4,000 to 5,000 people/mile2, while central city areas may
run two or three times that density. Completely rural areas often have
only 20 to 30 inhabitants/mile2-
About 70 to 75% of the total U.S. population lives in urban areas, with the
rest in rural areas. Applying these percentages to typical combination of
central city, suburbs, and outlying rural areas gives a composite popula-
tion density as follows:
40% of population at density of 20,000 people/mile2
+ 35% of population at density of 5,000 people/mile2
+ 25% of population at density of 50 people/mile2
= 100% of population at average density of 196 people/mile2
The economic losses suffered over the U.S. as a result of the particulate
pollutants would be in the $12,000 to $14,000/mile2 range, averaging about
$70/person annually.
154
-------
Table 49. ESTIMATED DAMAGES RESULTING FROM FINE PARTICLE POLLUTION AT
AN AMBIENT CONCENTRATION OF 75 ug/m3
Annual Economic Damage
Fine Particulate
Population
Dens ity/mi^
10
20
50
100
200
500
1,000
2,000
5,000
10,000
20,000
50,000
Human
Health
200
400
1,000
2,000
• 4,000
10,000
20,000
40,000
100,000
200,000
400,000
1,000,000
Animals and
Vegetation
5,000
4,600
4,100
3,750
3,450
3,100
2,900
2,700
2,450
2,300
2,150
2,000
0
Attributable to 75 Jig/m
Concentration ($/mi )
Materials and
Aesthetics
600
1,120
2,100
3,900
6,800
14,500
25,000
44,000
92,500
160,000
280,000
600,000
Total
Damage
5,800
6,120
7,200
9,650
14,250
27,600
47,900
86,700
194,950
362,300
682,150
1,602,000
Annual
Per Capita
Damage
580
306
144
97
71
55
48
43
39
36
34
32
155
-------
01
a\
Okla.
P.V., Minneap.
Brooklyn
N.H.. Ind. Ohio Minn. City Dallas Ks. .
U.S.
Ariz. Col. Ks. Avg.
1,000,000
100,000
a>
&
I
8
UJ
~s
o
10,000
1,000
1 1. 1, t,
I I I I I II
Cal.
K.C.,
Chicago N.Y.C.
Wash., N.Y.
V
N
X
Md. Ct. R.I. Phoenix Mo. Seattle D.C. Cit
ili ii nil MIi.i
~l i ii mi I I ii i 1111 r~7
Manhattan
N.Y.C.
_J
Jill
fnjio.ooo
Total Damage
($/Yr)
Per Capita Damage
($Ar)
"I I I I I I I 11 I I I I I I I 11 I I I I I I I I I I I I I I I
1,000
V
a>
a
a
a.
a
U
100
10
10
100,000
100 1,000 10,000
Population Density - People/Mi*
Figure 22, Economic damage resulting from continuous exposure to 75 ug/m^ fine particulate concentration
-------
Table 50 presents a comparison of the estimated damages resulting from
fine particle pollution at ambient concentrations of 75 and 60 ug/m3.
The damage at 60 ug/m3 is 80% of the estimated damage at 75 ug/m3.
COST/BENEFIT RELATIONSHIPS
Once the control cost/control efficiency relationships have been established
and the economic damage functions defined, a benefit/cost table can be con-
structed for any given set of conditions. Then, from the benefit/cost
table, incremental benefit-cost relationships can be defined. The point
at which incremental benefits and incremental costs are equal, or where
total economic costs are a minimum, represents the economic optimum con-
trol level.
Ultimately, the information thus developed will provide a basis for expres-
sing the optimum control efficiency as a function of the economic damage
intensity rate. For example, a control level of 99.7% might be appropriate
if the damage potential were $1.0 million/mile2, while a higher efficiency
could be justified in more densely populated areas, and a lower efficiency
could be tolerated in areas less susceptible to economic damage.
Calculation of cost/benefit relationships will be illustrated using a coal-
fired power plant as an example. Figure 23 illustrates the cost-vs-
efficiency relationships for fine particulate control on a 400-MW coal-
firetf electric generating plant, over the 99.0 to 99.9% control efficiency
range. As shown in this figure, total annual control costs (with capital
charges included at a 20% annual rate) rise from $148,000 at 90% effi-
ciency, to an estimated $580,000 at 99.9% efficiency.
Table 51 shows the extent of impact of the remaining fine particulate
emissions from the power plant at various levels of control efficiency,
employing the techniques described earlier in this chapter. The eco-
nomic impact area thus calculated describes the extent over which the
prescribed emissions will create and sustain a fine particulate concen-
tration of 75 or 60 ug/m3. This area of influence (in mile2) multiplied
by the damage incurred per square mile at the specified population density
(from Table 49) gives the total damage resulting from the specified emis-
sion source.
Tables 52 and 53 bring together the cost of control data from Figure 23
and the damage resulting from the uncontrolled fraction of the fine par-
ticulate emissions at populations densities of from 50 to 500 people/mile2,
thus developing the total economic cost of fine particulate emissions under
the prescribed conditions.
157
-------
Table 50. COMPARISON OF ESTIMATED DAMAGES RESULTING FROM FINE PARTICLE
POLLUTION AT AMBIENT CONCENTRATIONS OF 75 AND 60 jig/m3
($/mile2)
Population
Density/mi^
10
20
50
100
200
500
1,000
2,000
5,000
10,000
20,000
50,000
Economic
Total
5,800
6,120
7,200
9,650
14,250
27,600
47,900
86,700
194,950
362,300
682,150
1,602,000
q
Damage at 75 Ug/m
Per Capita
580
306
144
97
71
55
48
43
39
36
34
32
Economic
Total
4,640
4,896
5,760
7,720
11,400
22,080
38,320
70,080
155,960
289,840
545,720
1,281,600
q
Damage at 60 ug/m
Per Capita
464
245
115
77
57
44
38
35
31
29
27
26
158
-------
0.1
0.2
0.4 0.5 1.0 1.5 2.0
Uncontrolled Fine Particle Emissions (%)
_J 1 1 1 L_
3.0
5:0
7.0
ia.o
99.9
99.8 99.7 99.6 99.5 99.0 98.5 98.0
Fine Particle Control Efficiency (%)
97.0
95.0 93.0
Figure 23. Fine particulate control costs for 400-MW coal-fired electric generating plant
90.0
-------
Table 51. EXTENT OF IMPACT OF UNCONTROLLED FINE PARTICLE EMISSIONS FOR
400-MW COAL-FIRED ELECTRIC PLANT
Fine Particle
Control
Efficiency
(%)
90.0
91.0
92.0
93.0
S 94.0
95.0
96.0
97.0
98.0
99.0
99.9
Uncontrolled
Fine Particle
Emissions
(%)
10.0
9.0
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.1
Uncontrolled
Emissions
Qb/hr)
370.0
330.0
296.0
259.0
222.0
185.0
148.0
111.0
74.0
37.0
3.7
Fine Particulate
Loading at 100 hr
Residence Time
(lb)
37,000
33,300
29,600
25,900
22,200
18,500
14,800
11,100
7,400
3,700
370
Economic
Impact Area
at 75 ug/m3
(ml2)
26.30
24. 15
22.66
20.73
18.70
16.56
14.27
11.78
8.99
5.66
1.22
Economic
Impact Area
at 60 ug/m3
(mi2)
31.76
29.60
27.37
25.04
22.59
20.01
17.24
14.23
10.86
6.84
1.47
-------
Table 52. ECONOMIC COSTS OF CONTROL AND DAMAGE CAUSED BY FINE
PARTICIPATE EMISSIONS AT VARIOUS POPULATION
DENSITIES AND CONTROL EFFICIENCIES
Fine Particle
Control
Efficiency
m
90.0
91.0
92.0
93.0
94.0
95.0
96.0
97.0
98.0
99.0
99.9
Annual
Control
Cost
( $1.000)
148
158
169
181
195
211
230
254
289
349
580
Total Annual Cost ($1,000) at Given Population Density for 75 ug/nr
50/miz , .
Damage
Cost
189.4
173.9
163.2
149.3
134.6
119.2
102.7
84.8
64.7
40.8
8.8
Total
Cost
337.4
331.9
332.2
330.3
329.6
330.2
332.7
338.8
353.7
389.8
588.8
100/miz
Damage
Cost
253.8
233.0
218.7
200.0
180.5
159.8
137.7
113.7
86.8
54.6
11.8
Total
Cost
401.8
319.0
387.7
381.0
375.5
370.8
367.7
367.7
375.8
403.6
591.8
200/miz 500/miz
Damage
Cost
374.8
344.1
322.9
295.4
266.5
236.0
203.3
167.9
128.1
80.7
17.4
Total
Cost
522.8
502.1
491.9
476.4
461.5
447.0
433.3
421.9
417.1
429.7
597.4
Damage
Cost
725.9
666.5
625.4
572.1
516.1
457.1
393.9
325.1
248.1
156.2
33.7
Total
Cost
873.9
824.5
794.4
753.1
711.1
668.1
623.9
579.1
537.1
505.2
613.7
-------
Table 53. ECONOMIC COSTS OF CONTROL AND DAMAGE CAUSED BY FINE PARTICULATE
EMISSIONS
Fine Particle
Control
Efficiency
(%)
90.
91.
92.
93.
£ 94.
95.
96.
97.
98.
99.
99.
0
0
0
0
0
0
0
0
0
0
9
Annual
Control
Cost
($1.000)
148
158
169
181
195
211
230
254
289
349
580
o
Total Annual Cost ($1,000) at Given Population Density for 60 ue/nr
50/miz
Damage
Cost
182
170
157
144
130
115
99
82
62
39
8
.9
.5
.7
.2
.1
.3
.3
.0
.6
.4
.5
Total
Cost
330.9
328.5
326.7
325.2
325.1
326.3
329.3
336.0
351.6
388.4
588.5
100/mi2
Damage
Cost
245.2
228.5
211.3
193.3
174.4
154.5
133.1
109.9
83.8
52.8
11.3
Total
Cost
393.2
386.5
380.3
374.3
369.4
365.5
363.1
363.9
372.8
401.8
591.3
200/mi2
Damage
Cost
362.1
337.4
312.0
285.5
257.5
228.1
196.5
162.2
123.8
78.0
16.8
Total
Cost
510.1
495.4
481.0
466.5
452.5
439.1
426.5
416.2
412.8
427.0
596.8
500/mi2
Damage
Cost
701.3
653.6
604.3
552.9
498.8
441.8
380.7
314.2
239.8
151.0
32.5
Total
Cost
849.3
811.6
773.3
733.9
693.8
652.8
634.7
568.2
528.8
500.0
612.5
-------
Here it can be seen that the total economic cost of pollution is, in every
case, high when the control level is low (at 90% in the example); costs
then decrease as controls increase, until the optimum control level (the
point of minimum total cost) is reached; then increase as the benefits of
control can no longer justify their costs. These total economic cost-vs-
control efficiency relationships are plotted in Figure 24 for the four dif-
ferent population densities.
As would be expected and is clearly evident in Figure 24, a high popula-
tion density (with its correspondingly high damage cost/mile2) justifies
a higher emission control efficiency than does an area populated at low
density.
In the example, the optimum control efficiency for a population density
of 500/mile2 is 99.1%; for 200/mile2, 97.9%; for 100/mile2, 96.0%; and at
50 people/mile2, 94.0% control is economically justified. These optimum
control efficiencies are plotted against population densities in Figure
25.
Using this approach—refined, hopefully, with more reliable cost and bene-
fit data—the optimum control efficiency can be readily determined for any
particulate source or combination of sources at any location, employing
local characteristics in the analysis. The approach should be equally
valid in appraising the economics of control for other air pollutants,
providing the relevant data can be developed.
163
-------
1000
800
o
+- o
,92
— 600
II
o o
° i j
ii i \j
1 &
S o
< E
_ o
o Q
400
200
0.5
I
1.0 2.0 3.0 4.0" 5.0
Uncontrolled Fine Particle Emissions (%)
7.0
500
CN
I
0)
Q.
200 £
100
50
10.0
c
0)
Q
c
o
D
a.
Figure 24. Total annual economic costs of fine particulate control at various population densities
-------
lOO.Or-
98.01
£
"o
£
0)
96.01
ui
0)
c
O
U
E
94.01
| 92.0|
"o.
O
90.01
30
50
i I
100 200
Population Density - People/Mi^
500
Figure 25. Optimum fine particle control level at various population densities
-------
SECTION X
OVERALL FEASIBILITY OF EMISSION STANDARDS BASED ON PARTICLE SIZE
INTRODUCTION
Our analysis of the implications of emission standards based on particle
size has identified some technical deficiencies that will limit the type
of standards that can be proposed and implemented in the near future.
However, because the technical deficiencies relate primarily to the lack
of data on particle size distributions of effluents and control equipment
fractional efficiency, there appear to be no insurmountable technical
obstacles to emission standards based on particle size.
r
The economic impact of fine particulate control was found to vary sub-
stantially from industry to industry. Estimates of costs associated with
the control of fine particulates varied from less than 1.0% up to 20% of
the value of the product. The variation in economic impact suggests that
it will probably be necessary to consider less restrictive standards for
industries that experience a significant adverse economic impact.
The types of emission standards that appear feasible based on existing
technical and economic realities are discussed in more detail in the fol-
lowing section.
FEASIBILITY OF SPECIFIC TYPES OF STANDARDS
Although our analysis of various formats for,emission standards based on
particle size was performed in the context of general regulations that
might be applied uniformally to all sources, a more realistic approach
would be to tailor the emission standard to specific sources of fine par-
ticulate pollutants. Tailoring of standards would permit a greater degree
of flexibility in an overall control plan for fine particulates, and would
acknowledge the differences in the importance and difficulty of control
of individual sources.
166
-------
The exact format of the emission standard(s) that could be proposed and im-
plemented for specific sources will be limited by: (1) collection effi-
ciency in fine particle size range of available control equipment, and (2)
availability of source compliance monitoring techniques. Our conclusions
regarding the overall feasibility of the various formats for emission
standards investigated in this study are presented in the next subsections.
Opacity Regulations
Regulation of plume opacity provides a viable method of reducing the emis-
sion of fine particulates from stationary sources. Opacity standards ap-
pear to be the easiest to implement of the alternatives analyzed in the
current program. The data base for determining control equipment perfor-
mance requirements necessary for compliance with opacity standards is more
extensive than that available for other alternative emission standards.
Our analysis of the control equipment efficiency required for compliance
with a 10% or 57, opacity regulation indicated that for most sources the
opacity regulations would not impose collection efficiencies exceeding
existing equipment capability. Even with allowance for inadequacies in
some of the data used to determine the collection efficiencies needed to
attain 10% and 5% plume opacity for specific sources, compliance with
stringent opacity regulations would not generally require installation of
control equipment that exceeds the efficiency of the best control device
currently in use on that source.
Commercial transmissometers are available for measuring the in-stack opacity
of particulate emissions. Although additional field testing needs to be
conducted to define the performance of transmissometers on several different
types of sources, transmissometers provide a suitable, demonstrated tech-
nique for instrumental evaluation of the compliance of a source with opacity
regulations. With completion of additional field testing, transmissometers
should rapidly achieve the status of "off the shelf" technology.
Since the opacity standards would generally require the installation of
control equipment that does not exceed the performance capability of the
best available technology, the estimated economic impact of opacity stand-
ards is less severe than that for other alternatives analyzed in this study.
The extent of reduction in the emission of fine particulates is also lowest
of the alternatives studied for the same reason.
167
-------
Regulation Based on Best Installed Control System
An emission regulation which specifies that a source cannot emit fine par-
ticulate in a quantity exceeding that which would be emitted from the same
source equipped with the best control device currently being installed on
that source is obviously feasible. A regulation of this type merely
specifies that ^11 sources must be equipped with the best control system
currently in use for the specific source category.
The major unknown regarding the use of the BICD concept as an emission
standard is the degree of emissions reduction that can actually be achieved.
As noted in Appendix A, only a limited amount of reliable data is available
on the fractional efficiency characteristics of control equipment. In order
to determine the actual collection efficiency of control equipment in the
fine particle size range, extensive field testing will be required with
reliable sampling techniques.
A standard requiring installation of the best control system currently in
use will have nearly the maximum economic impact on industrial sources.
The economic impact analysis presented in Chapter 6 clearly demonstrates
this observation. Maximum reduction in fine particle emissions would also
be achieved by such an emission standard.
Mass Emission Regulations
A mass emission standard specifying the maximum amount of particulate less
than a given particle size which can be emitted from a source is the most
direct approach to regulating the emission of fine particulates. An almost
unlimited number of emission standards might be formulated using restriction
of mass emissions.
The concept of potential-emission rate offers an interesting basis for emis-
sion standards based on particle size. Regulations based on this concept
would establish emission limits which vary with the pollution potential of
the source, e.g., limitation of the mass rate of fine particle emissions
in pound per hour as a function of potential-emission rate of fine particu-
lates, also in pound per hour. Regulations of this type could be tailored
to control (1) specific sources of fine particulate pollutants; (2) specific
size ranges within the fine particulate range;'or (3) specific components
of the fine particulate stream emitted by a given source.
In general, mass emission regulations are a feasible vehicle for reducing
the emission of fine particulates, and they are the most flexible of all
the types of standards considered in this study. Because of the inherent
168
-------
flexibility afforded by the use of mass emission rates as a basis for fine
particle emission standards, the extent of reduction in emissions for
specific sources can be carefully selected and the technical and economic
impact can be tailored to fit the realities of a specific industrial cate-
gory.
To obtain maximum utilization of a mass emission regulation, it will be
necessary to broaden the data base on fine particulate emission rates and
control device fractional efficiency.
169
-------
SECTION XI
REFERENCES
1. Strauss, W., Industrial Gas Cleaning, Pergamon Press (1966).
2. Air Quality Criteria for Partlculate Matter, NAPCA Publication No.
AP-49, January 1969.
3. Esmen, N. A., and M. Corn, "Residence Time of Particles in Urban Air,"
Atmospheric Environment, 5, 571-578 (1971).
4. Corn, M., "Particle Size: Relationship to Collector Performance,
Emission Standards, and Ambient Air Quality," presented at 2nd
International Air Pollution Conference, International Union of Air
Pollution Prevention Associations, Washington, D.C., December 1970.
5. "Deposition and Retention Models for Internal Dosimetry of the Human
Respiratory Tract," Task Group on Lung Dynamics, Health Physics, 12,
173-207 (1966).
6. Findeisen, W., "Tiber das Absetzen kleiner in der Luf t suspendierten
Teilchen in der menschlichen Lunge bei der Atmung," Arch. Ges.
Physiol., 236, 367-379 (1935).
7. Hatch, T. F., and Paul Gross, Pulmonary Deposition and Retention of
Inhaled Aerosols, Academic Press, New York (1964).
8. Amdur, Mary 0., and Morton Corn, "The Irritant Potency of Zinc Am-
monium Sulfate of Different Particle Sizes," Ame r. Indus t. Hyg.
Assoc. Journal, 24, 326-333 (1963).
9. Hemeon, W. G. L., "The Estimation of Health Hazards from Air Pollution,"
Arch. Indust. Health, U, 397-402 (1955).
10. Amdur, M. 0., "The Respiratory Responses of Guinea Pigs to Sulfuric
Acid Mist," Arch. Ind. Health. 1J3, 407-414 (1958).
170
-------
11. Nader, J. A., et al., "Location and Mechanism of Airway Constric-
tion after Inhalation of Histamine Aerosol and Inorganic Sulfate
Aerosol," Inhaled Particles and Vapors, !!_, C. N. Davies (ed.),
Pergamon Press, London (1967).
12. Amdur, M. 0., and D. Underbill, "The Effect of Various Aerosols on
the Response of Guinea Pigs to Sulfur Dioxide," Arch. Environ.
Health. 16, 460-468 (1968).
13. Dautrebande, L., "Microaerosols," Academic Press, New York (1962).
14. Dautrebande, L., "Bases experimentales de la protection centre les
gas de combat," J. Duculot, Belgium (1939).
15. Dautrebande, L., and R. Capps, "Studies on Aerosols. IX. Enhance-
ment of Irritating Effects of Various Substances in the Eye, Nose
and Throat by Particulate Matter and Liquid Aerosols in Connection
with Pollution of the Atmosphere," Arch. Intern. Pharmacodyn, 82,
505-527 (1950).
16. Dautrebande, L., J. Shaver, and R. Capps, "Studies on Aerosols. XI.
Influence of Particulate Matter on Eye Irritation Produced by
Volatile Irritants and Importance of Particle Size in Connection
with Atmospheric Pollution," Arch. Intern. Pharmacodyn, 85, 17-48
(1951). ~~
17. LaBelle, C. W., J. E. Long, and E. E. Christofano, "Synergistic Ef-
fects of Aerosols. Particulates as Carriers of Toxic Vapors,"
Arch. Ind. Health, II, 297-304 (1955).
18. Dalhamn, T., and L. Reid, "Ciliary Activity and Histologic Observa-
tions in the Trachea after Exposure to Ammonia and Carbon Particles,"
Inhaled Particles and Vapors, 2^, C. N. Davies (ed.), Pergamon
Press, London (1967).
19. Boren, H. C., "Carbon as a Carrier Mechanism for Irritant Gases,"
Arch. Environ. Health, 8, 119-124 (1964).
20. Gross, P., W. E. Rinehard, and R. T. DeTreville, "The Pulmonary
Reactions to Toxic Gases,M Am. Ind. Hyg. Assoc. J., 27, 315-321
(1967).
171
-------
21. Pattle, R. E., and F. Burgess, "Toxic Effects of Mixtures of Sulphur
Dioxide and Smoke with Air," J. Pathol. Bacteriol.. 7$, 411-419
(1957).
22. Salem, H. , and H. Cullunbine, "Kerosene Smoke and Atmospheric Pollu-
tants," Arch. Environ. Health, £, 641-647 (1961).
23. Frank, N. R. , M. 0. Amdur, and J. L. Shittenberger, "A Comparison of
-the Acute Effects of S0£ Administered Alone or in Combination with
NaCl Particles on the Respiratory Mechanics of Healthy Adults,"
Intern. J. Air and Water Poll., 8, 125 (1964).
24. Burton, G. G., M. Corn, J. B. L. Gee, C. Vassallo, and A. Thomas,
"Absence of Synergistic Response to Inhaled Low Concentration Gas-
Aerosol Mixtures in Healthy Adult Males," presented at Ninth Annual
Air Pollution Medical Research Conference, Denver, Colorado (1968).
25. Toyama, T., "Studies on Aerosols. I. Synergistic Response of the
Pulmonary Airway Resistance of Inhaling Sodium Chloride Aerosols
and S02 in Man," Japan J. Ind. Health, 4, 86 (1962).
> ' >
*. i
26. Nakamura, K. , "Response of Pulmonary Airway Resistance by Interaction
of Aerosols and Gases in Different Physical and Chemical States,"
Japan J. Hyg., !£, 322 (1964).
27. Toyama, T. , and K. Nakamura, "Synergistic Response of Hydrogen
Peroxide Aerosols and .Sulfur Dioxide to Pulmonary Airway Resis-
tance," Ind. Health, 2, 34 (1964).
28. Speizer, F. E. , "An Epidemiological Appraisal of the Effects of
Ambient Air on Health: Particulates and Oxides of Sulfur,!1 JAPCA.
19_, 647-656 (1969).
29. Winkelstein, W. , "The Relationship of Air Pollution to Cancer of the
Stomach," Arch. Environ. Health, 18_, 544-547 (1969).
30. Zeidburg, L. D., R. M. Hagstrom, H. A. Sprague, and E. Landau,
"Nashville Air Pollution Study. VII. -Mortality from Cancer in
Relation to Air Pollution," Arch. Environ. Health, 15, 237-248
(1967).
/
31. Zeidberg, L. D., R. J. M. Horton, and E. Landau, "The Nashville Air
Pollution Study. V. Mortality from Diseases of the Respiratory
System in Relation to Air Pollution," Arch. Environ. Health, 15,
214-224 (1967).
172
-------
32. Fletcher, C. M., B. M. Tinker, I. D. Hill, and F. E. Speizer, "A
Five-Year Prospective Field Study of Chronic Bronchitis," preprint
(presented at the llth Aspen Conference on Research in Emphysema,
June 1968).
33. Douglas, J. W. B., and R. E. Waller, "Air Pollution and Respiratory
Infection in Children," Brit. J. Prevent. Soc. Med., 2£, 1-8 (1966).
34. Lunn, J. E., J. Knowelden, and A. J. Handyside, "Patterns of Respi-
ratory Illness in Sheffield Infant School Children," Brit. J.
Prevent. Soc. Med., 21, 7-16 (1967).
35. Watanabe, H., "Air Pollution and Its Health Effects in Osaka, Japan,"
preprint (presented at the 58th Annual Meeting, Air Pollution Con-
trol Association, Toronto, Canada, 20-24 June 1965).
36. Bates, D. V., "Air Pollutants and the Human Lung," American Review
of Respiratory Disease, 10J5, 1-13 (1972).
37. Connor, W. D., and J. R. Hodkinson, Optical Properties and Visual
Effects of Smoke Stack Plumes, PHS Publication No. 999-AP-30,
Cincinnati, Ohio (1967).
38. Junge, C. E., Air Chemistry and Radioactivity, Academic Press, New
York (1963).
39. Middleton, W. E. K., Vision Through the Atmosphere, University of
Toronto Press, Toronto (1952). :
40. Charleson, R. J., H. Horvath, and R. F. Pueschel, "The Direct Mea-
surement of Atmospheric Light Scattering Coefficient for Studies
of Visibility and Air Pollution," Atmos. Environ., l_, 469-478
(1967).
41. Noll, K. E.j P. K. Mueller, and M. Imada, Atmos. Environ.. 1, 501
(1967). ~~
i
42. Nicholson, B. R., "Visibility Effects of Various Atmospheric Pollu-
tants," New .Mexico Environmental Improvement Agency (1971).
43. Gates, D. M., "Spectral Distribution of Solar Radiation at the Earth's
Surface," Science. 151, 523-529 (1966).
173
-------
44. McCormick, R. A., and D. M. Baulch, "The Variation with Height of
the Dust Loading Over a City as Determined from the Atmospheric
Turbidity," JAPCA, 12, 492-496 (1962).
>
45. Robinson, N., Solar Radiation. Elsevier, Amsterdam, London, and
New York (1966).
46. Landsberg, H., Physical Climatology, 2nd Edition, Gray, Dubois,
Pennsylvania, 317-326 (1958).
47. Steinhauser, F., 0. Eckel, and F. Sauberer, "Klima and Bioklima von
Wien," Wetter und Leben, £ (1955).
48. Meetham, A. R., Atmospheric Pollution; Its Origins and Prevention,
Pergamon Press, New York (1961).
49. Mateer, C. L., "Note on the Effect of the Weekly Cycle of Air Pollu-
tion on Solar Radiation at Toronto," Intern. J. Air Water Pollution,
4, 52-54 (1961).
50. Robinson, G. D., "Long-Term Effects of Air Pollution - A Survey,"
Center for the Environment and Man, Inc., Hartford, Connecticut,
June 1970.
51. Cobb, W. E., and H. J. Wells, "The Electrical Conductivity of Oceanic
Air and Its Correlation to Global Atmosphere Pollution," Journal of
Atmospheric Sciences, 27^ 814-819, August 1970.
52. Air/Water Pollution Report, 7 May 1973.
53. Watt, K. E. F., "Tambora and Krakatau: Volcanoes and the Cooling of
the World," Stanford Research Quarterly, December 1972.
54. "Restoring the Quality of Our Environment," report of the Environ-
mental Pollution Panel, President's Science Advisory Committee,
The White House, Washington, D.C., 111-131, November 1965.
55. Changnon, S. A., "The LaPorte Weather Anomaly—Fact or Fiction,"
Bulletin of American Meteorological Society. 49^ (1), 4-11 (1968).
56. Hobbs, P. V., et al., "Cloud Condensation Nuclei from Industrial
Sources and Their Apparent Influence on Precipitation in
Washington State," Journal of Atmospheric Sciences. 27_, 81-89
(1970).
174
-------
57- Aynsley, E., "How Air Pollution Alters Weather," New Scientist,
66-67, 9 October 1969.
58. Schaefer, V. J., "The Inadvertent Modification of the Atmosphere by
Air Pollution," Bulletin of American Meteorological Society, 50^
(4), 199-206 (1969).
59. Warner, J., "A Reduction in Rainfall Associated with Smoke from Sugar
Cane Fires - An Inadvertent Weather Modification," J. Applied
Meteorology, 7, 247-251 (1968).
60. "Particulate Pollutant Systems Study, Volume I - Mass Emissions,"
Midwest Research Institute, EPA Contract No. CPA 22-69-104,
1 May 1971.
61. "Particulate Pollutant Systems Study, Volume II - Fine Particle Emis-
sions," Midwest Research Institute, EPA Contract No. CPA 22-69-104,
1 August 1971.
62. "Particulate Pollutant Systems Study, Volume III - Handbook of Emis-
sion Properties," Midwest Research Institute, EPA Contract No. CPA
22-69-104, 1 May 1971.
63. Goldberg, A. J., "A Survey of Emissions and Control for Hazardous and
Other Pollutants," EPA/OR&D internal report, November 1972.
64. Hidy, G. M., and S. K. Friedlander, "The Nature of the Los Angeles
Aerosol," Proceedings of the Second International Clean Air Congress,
H. M. England and W. B. Beery, Editors, Academic Press, New York
(1971).
65. Duncan, L. J., "Analysis of Final State Implementation Plans—Rules
and Regulations," The Mitre Corporation Report MTR-6172, Rev. 1,
July 1972.
66. Feldman, P. L., and D. W. Coy, "Comparison of Computed and Measured
Opacities: Lignite-Fired Boilers," Research Cottrell, Bound Brook,
New Jersey.
67. Callaghan, D. J., "Detailed Background Information for Modification
of Regulation 2 Regarding Particulate Emissions," (BAACD internal
memo).
175
-------
68. Ensor, D. S., and M. J. Pilat, "Calculations of Smoke Plume Opacity
from Particulate Air Pollutant Properties," J. Air Poll. Control
Assoc.. 2U No. 8, 496-501 (1971).
69. Greco, J., and W. A. Wynot, "1971 Operating and Maintenance Problems
Encountered with Electrostatic Precipitators," American Power Conf.
Proc.. 33, 345-353.
70. "Scurbbing System Removes Submicron Particulates," Chem. Eng.,
20 September 1971.
71. Teller, A. J., "A Fresh Look at the Technology of Particulate Re-
moval Via Scrubbing," Eng. Mining J., April 1971.
72. "A New Process for Cleaning and Pumping Industrial Gases—The Aero-
netics System," U. S. Patent No. 3,613,333.
73. Private Communication, Mr. H. E. Gardenier, Vice President, Aero-
netics, Inc., September 1972.
74. Sem, G. J., et al., "State of the Art: 1971-Instrumentation for
Measurement of Particulate Emissions from Combustion Sources,"
Thermo-Systems, Inc., EPA Contract CPA 70-23, July 1972.
75. Sem, G. J., et al., "Monitoring Particulate Emissions," Chem. Eng.
Prog., 67_9 No. 10, October 1971.
76. Presentation to Bay Area Air Pollution Control District Advisory
Council by Bay Area League of Industrial Association, Inc.,
September 1970.
77. Internal Memo of Chief Administrative Officer, Bay Area Air Pollution
Control District, 22 July 1970.
78. Texas Air Control Board, "Instrumental Method for Measurement of
Transmittance," 14 September 1972.
79. McKee, H. C., "Instrumental Method Proves Superior for Control of
Visible Emissions," paper 73-245, presented at the 66th Annual APCA
Meeting, Chicago, Illinois, 24-28 June 1973.
80. Bentner, H. P., "Measurement of Opacity and Particulate Emissions
with the Lear Siegler On-Stack Transmissometer," paper 73-169, pre-
sented at the 66th Annual APCA Meeting, Chicago, Illinois, 24-28
June 1973.
176
-------
81. Yocom, J. E., "Problems in Judging Plume Opacity—A Simple Device
for Measuring Opacity of Wet Plumes," J. Air Poll. Control Assoc.,
13;(1), 36-39 (1963).
82. Bird, A. N., J. D. McCain, and D. B. Harris, "Particulate Sizing
Techniques for Control Device Evaluation," paper 73-282, pre-
sented at 66th Annual APCA Meeting, Chicago, Illinois, 24-28 June
1973.
83. Dorman, R. G., "Dust Sampling," Filtration and Separation, November-
December 1968.
84. Badzioch, S., "Correction for Anisokinetic Sampling of Gas-Borne
Dust Particles," J. Inst. Fuel. March 1960.
85. Logan, T. J., et al., "Experimental Investigation of Isokinetic and
Anisokinetic Sampling of Particulates in Stack Gases," paper pre-
sented at the 65th Annual Meeting of the AICHE, New York, November
1972.
86. Jackson, M. L., "Particle-Molecule Collection by Sonic Flow Im-
pingers," paper presented at the 65th Annual Meeting of the Air
Pollution Control Association, Miami Beach, Florida, June 1972.
87. Control Techniques for Particulate Air Pollutants, U.S. Department
of Health, Education, and Welfare, Washington, D.C. (1969).
88. Sargent, Gordon D., "Dust Collection Equipment," Chemical Engineering,
January 1969.
89. O'Connor, J. R., and J. F. Citarella, "An Air Pollution Control Cost
Study of the Steam-Electric Power Generating Industry," APCA
Journal, 20(s), 283-288 (1970).
177
-------
SECTION XII
GLOSSARY OF TERMS
A = Area
b = Extinction coefficient
c = Annualized cost
Cc = Average efficiency of control equipment
Ct =' Percentage of the production capacity on which control equip-
ment has been installed
E = Emission rate, tons/year
ef = Emission factor (uncontrolled), pounds/ton
g(r) = Mass distribution function
I = Intensity Of transmitted light
I0 = Intensity of incident light
_ Specific particulate volume
Extinction coefficient
L = Visual range
M = Mass concentration
m = Refractive index of the particles relative to air
n(r) = Size frequency distribution, number of particles of radius r
per volume per Ar
P = The production rate, tons/year
178
-------
) = Particle light extinction efficiency factor, the total light
flux scattered and absorbed by a particle divided by the
light flux incident on the particle
r = Particle radius
t = Time
R = Radius
T = Plume transmittance
W = Mass concentration of particles in exhaust stream
a = Size parameter, 2nr/X
X = Wavelength of light
p = Particle density
179
-------
SECTION XIII
APPENDICES
CONTROL TECHNOLOGY FOR FINE PARTICLES
INTRODUCTION
The technology applicable to the control of fine particles is the same as
for any participate emissions, but with interest centered on the higher
efficiency equipment. Currently available devices include electrostatic
precipitators, wet scrubbers, fabric filters and afterburners. New con-
trol devices are under development with the aim of removal of fine parti-
cles at lower cost or more efficiently. Characteristics of both the
conventional devices and some new devices and control concepts are de-
scribed in the following sections of this appendix.
CURRENTLY AVAILABLE CONTROL TECHNOLOGY
The control devices currently available for controlling fine particle
emissions are generally the same as those usually considered for control-
ling total particulate emissions, that is, electrostatic precipitators,
fabric filters, scrubbers and afterburners. However, to date there has
been little incentive for manufacturers to investigate the ability of
their equipment to remove fine particles.
Currently, control equipment is rated mainly by one parameter, namely,
overall mass efficiency. Specification of control equipment efficiency
by overall performance is inadequate with respect to fine particle emis-
sions. Penetration (i.e., 1-mass efficiency) in specific size ranges is
a more revealing term for rating control equipment performance. Correla-
tions of penetration vs particle size are called "fractional efficiency
curves." Table A-l presents typical data on average collection effi-
ciencies for various particle sizes and various particulate control
equipment.!./
Midwest Research Institute, under contract from EPA, has been conduct-
ing several studies involving the control of fine particulate emissions
180
-------
Table A-l. AVERAGE COLLECTION EFFICIENCIES FOR VARIOUS PARTICLE SIZES
AND VARIOUS PARTICIPATE CONTROL EQUIPMENT*/
Efficiency. %, in Micron Ranee
Type Collector
Baffled settling chamber
Simple cyclone
Long -cone cyclone
Multiple cyclone
Overall
58.6
65.3
84.2
74.2
0-5
7.5
12
40
25
5-10
22
33
79
54
10-20
43
57
92
74
20-44
80
82
95
95
X.M.
90
91
97
98
(12-in. diameter)
Multiple cyclone 93.8 63 95 98 99.5 100
(6-in. diameter)
Irrigated long-cone 91.0 63 93 96 98.5 100
cyclone
Electrostatic precipitator
Irrigated electrostatic
precipitator
Spray tower
Self-induced spray
scrubber
Disintegrator scrubber
Venturi scrubber
Wet impingement scrubber
Baghouse
97.0
99.0
94.5
93.6
98.5
99.5
97.9
99.7
72
97
90
85
93
99
96
99.5
94.5
99
96
96
98
99.5
98.5
100
97
99.5
98
98
99
100
99
100
99.5
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
a/ Data based on standard silica dust with the following particle distribution:
Particle Size Range Microns Percent by Weight
0-5 20
5-10 10
10-20 15
20-44 20
> 44 35
181
-------
from stationary sources. A part of this effort has involved the evaluation
of information that is available on the fractional efficiency measurements
of different control devices.2/ Our analysis indicated that a major por-
tion (over 957o) of the data currently available on the particle size of
particulates emitted from industrial sources has been obtained by sampling
and particle sizing techniques that are not suitable for the particle size
range < 2 u. As a result, only a meager quantity of accurate data is avail-
able on particle size in the < 2 u size range for effluents from uncontrolled
or controlled sources. As a result, the ability of control devices to col-
lect the < 2 u particles is ill-defined.
Figures A-l to A-4 illustrate some of the available fractional efficiency
data.2_/ Log-probability coordinates were used in order to magnify the ef-
ficiency relationship for the smaller particles. All available information
relating to specific type of device, overall efficiency, operating condi-
tions (such as pressure drop or water rate, etc.), and sampling and analy-
sis techniques are included. Figure A-l illustrates data for electrostatic
precipitators, while Figure A-2 presents fabric filter fractional efficiency.
Wet scrubber characteristics are given in Figures A-3 and A-4. The data for
each type of device varied over a wide range, but this variation is not
surprising considering the variations in types and design of devices, test-
ing procedures, operating conditions and particle size analysis techniques.
Since the data for each type of device did vary over a wide range, the
curves were examined with the objective of drawing general curves that
would represent low, medium, and high overall efficiencies for each type
of device. The data for each type of control device were carefully as-
sessed to determine the general fractional efficiency curves shown in
Figure A-5.2/
Performance Capabilities of Current Equipment
The fractional efficiency curves described in the preceding section show
that even though the control devices can achieve overall mass efficiencies
in excess of 99%, their ability to remove fine particles is often sig-
nificantly less.
Variations in the usual control equipment designs may increase the cap-
abilities of these devices. One such variation is the wet electrostatic
precipitator in which water removes the collected material and often makes
possible the use of electrostatic precipitation on sources where it would
otherwise not be suitable. Preconditioning of effluent and injection of
NHq or other gases has also been used to change the resistivity of the
dust to permit better collection in electrostatic precipitators.
182
-------
99.99
00
CO
ELECTROSTATIC PKCIPITATOSS .
u
O
o
u
10.0
100.0
MRTlClf DIAMtm - MICKONS
Figure A-l. Fractional efficiency data for electrostatic precipitators
-------
Symbol Identification for Figure A-l
D Stairmand, Dry ESP
Stairmand, irrigated ESP
Well Conditioned Dust, Particle Sizing by Electron Microscope
K Wet ESP, High Resistivity Dust, Particle Sizing by Electron Microscope
O No Information Except 99.9 % Efficient
£3 No Information Except 99.9 % Efficient
V P'lot Scale ESP on Coal-Fired Boiler; Gas Velocity 2.62 Ft/Sec
A "Typical" Curve
184
-------
00
tn
99.9°
99.9
99.£
99.0 -
95.0
90.0
1 - 1
1 - \
I I | [
O
50.0
20.0
-I 1 1 T
FABRIC FIITE"
0.05
0.1
0.2
0.5
1.0
2.0
5.0
1C.O
20.C #.
Z
O
50.0
O
*-
u
tJ
6
u
90.0
95.0
99.0
99.8
99.9
-I 1 1—L_L_LJ99.99
100
PARIICU DIAMETER - MOONS
Figure A-2. Fractional efficiency data for fabric filters
-------
Symbol Identification for Figure A-2
35 Shakes, Precoated, Monodisperse Aerosol, 1-4 In. A
Loaded, Precoated, Monodisperse Aerosol, 1-4 In. AP
1C No Information Except 99.9 + % Efficient.
£3 No Information Except 99.9 + % Efficient
Data Using Cascade Impactor, Shake Type Collector, Very Low AP
Data Using Cascade Impactor, Ultrajet Type Collector, 10-12 In. A P
186
-------
99.99-
00
0.01
0.01
SCRUBBERS
10.0
0.05
0.!
C.?
0.5
1.0
2.0
S.O
10.0
20.0
90.0
95.0
99.0
99.8
99.9
-g_l_U99.99
100.0
u
50.0 ~
4W
O
0
a
PARTICLE DIAMETER • MICRONS
Figure A-3. Fractional efficiency data for wet scrubbers
-------
Symbol Identification for Figure A-3
Stairmand, Gravity Spray Tower, Less Than 1 In. AP
Stairmand, Impingement Scrubber
Stairmand, Orifice Scrubber
Stairmand, Venturi Scrubber
Spray Tower, Soluble Dust (Na^/SO^), Optical Counter, Liquid Flowrates
Approximately 0.03 Ft^/Mirx
LJ Venturi Scrubber "Typical" Curve
t3 Venturi Scrubber, H3PO4 Plant Mist, Approximately 30 In. Ap
188
-------
99.95
00
ID
I 1 1 L_l_L_U99.99
10.0
PAITIClf OIAMITEI • MICHONS
100.0
Figure A-4. Fractional efficiency data for wet scrubbers
-------
Symbol Identification for Figure A-4
LJ Data Using Cascade Impactor
W 4-6 In. H2O AP, 4 gpm/1,000 cfm
Q 4-6 In. H2O Ap, 3.3 gpm/1,000 cfm
4-6 In. H2O Ap, 3.0 gpm/1,000 cfm
£3 Perforated Plate Scrubber, 3 Plates, 2,750 cfm, Vane Separator, Water Recycled,
12 In. AP
Perforated Plate Scrubber, 3 Plates, 2,750 cfm, Venturi Separator, Water Recycled,
12 In. A,P
Perforated Plate Scrubber, 3 Plates, 2,750 cfm, Venturi Separator, Fresh Water,
12 In. AP
Asphalt Plant, Wet Scrubber, Low Ap
Venturi Scrubber (1967), Throat Velocity 17,800 Ft/Min
Orifice Scrubber (1967), 6 In. AP
O Wet Centrifugal Scrubber (1967), 3.5 In. AP
O Venturi Scrubber, 30 In. AP, H3PO4 Mist
A Stairmand, Spray Tower, 1 In. AP, 18 Gal/1, OQO cf.
Stairmand, Venturi Scrubber, 6 In. Throat, 3,500 cfm Gas
190
-------
99.99
0.01
i I 1—i—i—r—n
fAHTICU DIAMETER - MICRONS
99.99
Figure A-5. Extrapolated fractional efficiency of control devices
191
-------
One of the largest users of electrostatic precipitators has been the coal-
fired power plants. Electrostatic precipitators in this service process
very large volumes of flue gases at temperatures usually in the range of
300-600°F. Design efficiencies may be as high as 99.7% but the particle
size of the flyash emissions is comparatively large.
Wet scrubbers are in wide use in many types of sources and there have been
a multitude of variations in design. These devices have generally not been
used for controlling sources containing a large percentage of fine parti-
cles except for higher pressure drop Venturi scrubbers. Venturi scrubbers
are capable of high overall mass efficiencies (> 997o) even on sources con-
taining a higher percent of fumes such as basic oxygen furnaces in the
steel industry. More recently, variation in scrubber design for pressure
drops on the order of 4-10 in. of H20 have shown high efficiencies. Ex-
amples of this are the marble bed scrubber and the turbulent contact ab-
sorber. These types of scrubbers have undergone increased development in
connection with S02 removal processes for power plants, etc.
Fabric filters are usually considered as the highest efficiency device and
have been demonstrated to have efficiencies in excess of 99.9%. Variations
in design of fabric filters include fabric material and cleaning method.
The cleaning method may also permit variations in air to cloth ratio which
range from as low as 2 cfm/ft^ up to 25 cfm/ft2.
For many sources, as the applied control device efficiencies are increased,
the emissions during periods when the control devices are out of operation
or when their operation is impaired become quite significant. Reliability
and maintenance requirements are briefly reviewed in the next section.
Maintenance and Reliability Requirements of Current Equipment with Regard
to Control of Fine Particles
Increasingly stringent emission regulations will make it imperative that
control devices installed operate reliably. As a first requirement, the
control device and all associated equipment must be properly designed with
the proper materials of construction and adequate instrumentation to shut
the system down or bypass the control device in case of upsets. This is
necessary to prote'ct the control equipment; and prevent long and costly
downtimes for repairs.
Assuming that the equipment has been properly designed and proper materials
of construction used, regular maintenance and inspection schedules must be
followed in order to avoid unscheduled downtime for repairs and to prevent
deterioration in the operating efficiency of the system. Because the fine
192
-------
particle fraction of any effluent stream is the most difficult to collect,
it is this portion of the effluent which will most likely escape collection
if the equipment is not kept at its peak operating efficiency.
Each type of control device has its own characteristic problem areas that
most frequently occur, leading to decreased efficiency and lowering reli-
ability, if regular maintenance is not practiced. In the case of electro-
static precipitators the most frequently reported problem is breakage of
electrode wires which takes that associated section of the precipitator
out of operation. This may cause only a small decrease in efficiency but
the problem usually cannot be repaired without shutting down all or a "Large
part of the precipitator. References 3 and 4 report the results on a study
of the reliability of electrostatic precipitators of various design types
serving 51 power generating units of TVA. Figure A-6 indicates the items
that contributed to the average unavailability of nine classes of electro-
static precipitators on the TVA system over a recent year's operation._L'
Reference 3 reported that the overall weighted average availability was
92.6% for a 1-year period of operation.
Wet scrubbers are especially susceptible to corrosion and buildup problems.
Corrosion problems are best controlled by choice of construction materials
and inspection and repair during regular shutdowns. Buildups of particu-
late matter or other solids from the scrubbing solution are a problem that
may be more difficult to control than corrosion. Both problems of corro-
sion and buildup will result in decreased efficiency of particulate removal,
with the finer particles being the ones most likely to escape collection.
Operational reliability of fabric fibers is related to prevention of over-
heating or condensation. Any malfunction that causes either situation can
quickly ruin the entire set of bags in a fabric filter unit. This not only
requires costly replacement of the fabric but also takes the control device
out of service until new fabric can be obtained and installed.
i
Another problem, one that may be difficult to detect, is small holes inj the
bags. This can occur because of abrasive particulate matter, but it is
also often due to improper installation and tensioning of the bags. This
latter problem is preventable if regular and proper maintenance inspection
and repairs are carried out.
Few people would question the merits of a good maintenance program, but the
press of plant operations often cause such programs to be passed over. It
is often the nonproduction equipment, such as pollution control devices,
that are neglected. Increasingly stringent regulations and monitoring ef-
forts along with required incident reports and penalties are being implemented.
193
-------
23
22
21
20/7
/
7 -
UJ
_i
m 6
§5
i-
z
Ul
O 4
cc
ui
o.
e—a INSULATORS I iFLYAStm j| INTERNAL SHORTS
Illlll CONTROLS S I SLUICE SYSTEM B B WIRES
UNKNOWN
\
V
\ N
3 Z
^
v \
s \
\
\ 0
\ V
A
PRECIPITATOR
B C D E F
H
Figure A-6. Electrostatic precipitator unavailability
194
-------
The incident reports and penalties should help to promote regular mainte-
nance programs to keep the installed equipment operating at its peak with
resultant improvement in collection—especially of the fine particle frac-
tions.
NEW OR EMERGING CONTROL TECHNOLOGY
There are several possible avenues that might lead to improved control of
fine particulates. The more promising approaches can be grouped into three
broad categories: (1) development of new or novel particulate control de-
vices; (2) augmentation of commonly used collection mechanisms by additional
forces that do not approach zero in the ^ 2 ji size range; and (3) particle
agglomeration techniques.
Recent developments in each area are briefly summarized in the following
sections.
New Control Devices
A variety of devices which are claimed to have high collection efficiencies
in the fine particle range have been reported in the literature. Table A-2
presents a list of many of the devices recently reported in the literature.
For most of these devices supporting data for the claims of high efficiency
are meager, unavailable, or inconclusive. Extensive testing programs will
be required to determine the full potential of these devices.
One of the more promising new control devices is the ADTEC system. The
ADTEC system is a wet scrubbing system that operates on the conventional
Venturi collection mechanism of inertial impaction, but establishes the
requisite particle-droplet differential velocity by utilizing waste process
heat rather than external energy. On the basis of currently available in-
formation, this system appears to offer significant improvement in the col-
lection of fine particles, at modest energy consumption rates, where a
waste gas is available which contains a sufficient amount of thermal
energy. SiZ./
Control devices utilizing steam condensation also appear to offer promise
for improved collection of fine particulates. Condensation phenomena are
already utilized in many types of wet scrubbers in that the effluent gas
is contacted with water spray resulting in cooling and saturation of the
gas stream. However, the efficiency of these devices for removing fine
particles is not especially high unless considerable mechanical energy is
dissipated during the condensation, such as that which occurs in a Venturi
scrubber.
195
-------
Table A-2. NEW CONTROL DEVICES FOR PARTICULATE POLLUTANTS
Control Devices
Manufacturer(s) or Investigator(s)
ADTEC-wet scrubber
Aerodynamic immaculator
(wet scrubber)
PENTAPURETM IMPINGER
Nucleation scrubber
Condensation Scrubber
Granular bed filters
Fluid beds
Charged droplet scrubber
Space charge precipitator
Pulsed precipitator
Electret filters
Aronetics, Inc.
Tullahoma, Tennessee
Lone Star Steel Company
Lone Star, Texas
Purity Corporation
Elk Grove Village, Illinois
Teller Environmental Systems, Inc.
New York, New York
Oak Ridge National Laboratory,
APT, Inc.
University of Melbourne
Rex-Chainbelt, Inc., Ducon Company
Carnegie-Mellon University, City
University of New York
University of Idaho,
Oregon State University
TRW, Inc.
University of Washington,-MIT
University of California
Belco Pollution Control
Battelie Northwest
196
-------
The experimental work of Schauerl/ has demonstrated that condensation ac-
complished by the use of a diverging nozzle system can produce very high
removal efficiency, but this requires excessive quantities of steam.
Lancasters1 work2/ indicates that condensation phenomena offer a potential
means of improving the operational efficiency of existing wet scrubber
units but the technique of direct steam injection is inefficient and could
only be considered in situations where low pressure waste steam was avail-
able.
The inefficient usage of steam appears to be due primarily to inadequate
distribution of the steam between potential condensation nuclei. If the
efficiency of steam usage can be increased by the design of special con-
densation scrubbers which adequately distribute the condensed steam among
aerosol particles, the system could be competitive with conventional
scrubbers in normal applications. For the removal of ultrafine particles,
the condensation scrubber may prove to be superior.
Augmentation of Collecting Mechanisms
Theoretically, improvements in the control of fine particles might result
from better exploitation of particle collection mechanisms (e.g., dif-
fusiophonesis, thermophoresis). Such phenomena might be used advantage-
ously for collection of fine particles if devices can be designed to
utilize them without unreasonable high energy requirements.
A brief presentation of some of the theoretical aspects and experimental
results for these mechanisms is given in the succeeding sections. A more
in-depth analysis of these mechanisms, with emphasis on wet scrubber tech-
nology, has been carried out by Calvert, et al.10-12/ among others. Equa-
tions presented and developed by Calvert plus the many important references,
will be of interest to those who may want to continue investigations in
this area.
Thermophoretic Forces - Exploitation of thermal forces to improve the col-
lection of fine particles might be accomplished in one of two ways: (1)
design and construction of cleaners based entirely on particle deposition
by thermal forces; and (2) use of thermal forces as a contributing mecha-
nism in existing dust collectors.
From the standpoint of the collection efficiency of fine particulates,
thermal precipitators appear quite promising. However, consideration of
197
-------
the energy requirements presents a different picture. The power required
for thermal precipitation would be at least an order of magnitude greater
than that required for most other air cleaners. If waste thermal energy
is available from the gas stream that is being cleaned, then a thermal
precipitator is more attractive from an external energy consumption stand-
point. ' l
The deposition of particles from a hot gas in a cooled, packed bed is an
instance in which thermal forces may improve collection efficiency. Ex-
periments have shown that when a packed bed is initially cold, particle
collection is more nearly complete. Theoretical calculations indicate
that, since the passages in a bed are narrow, a temperature difference of
50°C might give rise to a temperature gradient of 1000°C/cm in the pass-
ages. Calculations show that this could result in the deposition of 98.8%
of the 0.1 u particles in a 9-in. deep bed.il/
Diffusiophoretic Forces - Diffusiophoresis might be a useful mechanism to
exploit in conjunction with other mechanisms for the removal of small
particulates from gas streams. Diffusiophoresis has the following ad-
vantages:
1. The fundamental mechanisms is independent of particle size and becomes
more important compared to other removal mechanisms for particles below
2 u.
2. The particulate removal efficiency that can be expected depends on
operating conditions and equipment design but can theoretically reach
100%. In a more practical situation, theory indicates that diffusio-
phoresis might account for more than 30% of the total collection effi-
ciency.
To exploit the mechanism of diffusiophoresis, the collecting device must
be designed such that one component in the gas phase is diffusing toward
a collecting surface. The most practical case would be the diffusion of
water vapor toward a surface.
198
-------
Agglomeration of Particles
The agglomeration or coagulation of particulates could be used as a step
in a sequence of operations aimed at controlling the emission of fine
particles. If sufficiently large particles can be produced, it may be
possible to use conventional low-cost techniques as the final collection
step. The movement of particles toward one -another may be brought about
solely by Brownian motion, but usually other influences play a major role,
e.g., turbulence of the fluid, gravitational forces, electrostatic forces,
and sonic forces.
The following sections discuss the nature of particle agglomeration with
and without the use of external forces, and indicate possible avenues
that might be utilized to enhance the control of fine particles.
Agglomeration in Absence of External Forces - When the movement of parti-
cles, leading to contact and agglomeration, can be accomplished only by
Brownian movement (diffusion), the process is called thermal coagulation.
The major disadvantage of thermal agglomeration is the long time required
to grow particles. The only variables which can be used to control
particle-particle collision rate are temperature, gas composition and
the particle concentration. The temperature and suspending gas composi-
tion do not appear to be useful variables.
The possibility of influencing the coagulation rate of particles suspended
in air by introducing a second gas or vapor has been explored in several
experimental programs, and the only clear-cut results seem to be those in
which the added vapor in some way affects the shape of the particles, or
in which vapor is actually being transferred between the vapor phase and
the particles so that appreciable gaseous diffusion occurs (i.e., dif-
fusiophoresis).
Onercould conceivably change the particle distribution and concentration
by seeding the particle suspension with large particulates to act as
"agglomeration sites," but this does not appear very useful based on ex-
perimental studies. Thus, thermal agglomeration is not a viable approach
to augmenting the ability of control systems to control fine particulate
emissions.
199
-------
Turbulent Agglomeration - Turbulence increases the relative velocities
among particulates which in turn increases the chance of particulate col-
lision. Theoretical studies on the coagulation of aerosols in turbulent
flow have been conducted by Levich,!!/ East and Marshall,JA/ Tunitskii,I-L/
Obukhov and Yaglon,lJI/ and Beal.il/
Experimental study of the rate of agglomeration in the presence of tur-
bulence is difficult because turbulence also accelerates the deposition
of particles on walls. The rate of deposition increases with particle
size so that the assessment of the course of agglomeration from the in-
crease in mean particle size in complicated. Experiments by Yoder and
Silverman are the only results that are subject to any rigorous analy-
sis.—' These investigators performed experiments to obtain data on the
deposition and agglomeration of particles in turbulent air flow. In
their experimental design, deposition and agglomeration were occurring
simultaneously, and their basic problem was to separate the two effects.
They did this by measuring both the total number concentration and the
fraction of particles which had agglomerated at the inlet and outlet of
their test section. By applying theoretical concepts, they could then
infer the separate effects of both deposition and agglomeration from the
measured parameters. They did not, however, make any direct observation
of either phenomenon.
Figures A-7, A-8, and A-9 present some of the results of Seal's theoreti-
cal calculations. As shown in Figure A-7, the "exact" solution of the
diffusion equation is asymptotic to the solutions based on either Brownian
motion or turbulent diffusion alone and does not differ very much from a
simple sum of these solutions. Figure A-8 presents the normalized co-
agulation constant, K = ffo , as a function of particle diameter, while
Figure A-9 illustrates the variation of the agglomeration rate constant
(Kno) as a function of particle size for a constant mass concentration
of 10-8 g/cm3.
Beal's analysis has several limitations. Only the steady-state solution
of the diffusion equation was obtained, only interactions among particles
of equal size were considered, and all particles were assumed to stick
together on impact. The last assumption is the most tenuous.
Figures A-7 and A-8 indicate that turbulence does not significantly en-
hance agglomeration for particles less than 0.2 u in diameter, but can
increase the coagulation constant of 0.5 u particles by about a factor of
10 and 1 u particles by a factor of 1Q2. The energy expended to accom-
plish this increase is about 2.7 x 107 dyne-cm-sec"1 or 0.004 hp/g.
200
-------
I I I i I I !l|
FLUID Alft
PIPE OlA.VLTER-IOCM
TEV,PEI
-------
I I I I IIIJ I I I I I 11>l I I I I 11 III
; FLUID «AIR
-_ PIPE DIAMETER -10 CM
I TEMPERATURE • 20* C
- PARTICLE DENSITY • 1.0 GM/CC
CONCENTRATION • ICT« GM/CC
I I I i ii nl I I I I I i II| I I I I 11 III I I I I I
I0'2 2 9 10"' 2 9 10° 2 9 10' 2 9
PARTICLE DIAMETER, pm
Figure A-9. Agglomeration rate constants for various
particle diameters and fluid velocities
202
-------
This is the equivalent to about 40 hp/1,000 ft3. If this energy is ex-
pended over a period of 1 min, a 0.5 u particle could be grown to about
2-3 u. This is an extremely high energy consumption, and, therefore,
turbulent agglomeration does not appear attractive for general use.
Agglomeration in Sonic Fields - Several different effects are responsible
for the enhanced agglomeration rate due to sonic forces including (1) col-
lection of the particles at the antinodes in the sonic standing wave (due
to radiation pressure), (2) hydrodynamic forces between the particles, and
(3) additional collisions due to the different vibrational amplitudes of
different sized particles.
The relative importance of the three mechanisms cited above has not been
determined. In fact, there is no assurance that other mechanisms are not
important. It is, therefore, not possible to construct a theoretical
model.
However, previous theoretical studies have developed equations which ex-
press the functional dependence of the three mechanisms.^./ From a
theoretical standpoint, all three mechanisms show a strong dependence on
the radii of the particles, with the forces due to radiation pressure and
hydrodynamics increasing with increasing particle size and the relative
vibrational amplitudes decreasing with increasing particle size. There
is also a strong frequency dependence in all the mechanisms and particle
agglomeration becomes negligible at low frequencies. There is, therefore,
an optimum frequency for acoustic agglomeration which varies with particle
size.
Coagulation by a sonic field has as its principal advantage its applic-
ability to any aerosol, including those comprised of submicron particles.
The principal disadvantage of sonic coagulation is its moderately high
energy requirements. A second major disadvantage is the low efficiency
of acoustic coagulators and their inability to handle highly dispersed
suspensions. Even with long residence times, sonic agglomerators which
incorporate inertial separators for particle collection, cannot treat
suspensions having particle loadings of < 0.5 to 1.0 grains/ft3, it is,
therefore, necessary to augment highly dispersed suspensions with a water
mist or other particles to increase the particle loading and obtain satis-
factory separation.
Water augmentation appears necessary to obtain the removal of a large
fraction of the particulates. If a pound of water per 1,000 cfm is used,
one would expect an increase in the power requirements of up to 1 hp/1,000
203
-------
cfm, depending on the device used to introduce the water and the mean
droplet size. The minimum energy requirements would then be 5.5 to 10
hp/1,000 cfm. Whether this is competitive with other devices capable of
removing fine particles remains to be seen.
Agglomeration of Charged Particles - One method of increasing the rate of
agglomeration of fine particulates is to add a bipolar charge, either with
or without an externally imposed field. To a limited degree, this occurs
in a standard electrostatic precipitator but not sufficiently to permit
efficient collection of fine particulates. With proper conditions the
large electrostatic forces between particulates can produce a large in-
crease in the rate of agglomeration of submicron particulates.
Space-charge precipitation is a method of removing particulates from gas
streams based on the migration of charged particles in their own space-
charge fields. However, particulate gas systems are usually not suffi-
ciently concentrated to provide adequate fields. It is, therefore,
necessary to add a cloud of charged drops in order to produce fields
which will remove the particles.
Basically, a space-charge precipitator consists of two parts: (1) a
short section where both the particles and the added drops are charged
by high voltage coronas; and (2) a section of grounded tubes or plates
on which the particles and drops are collected. In the collecting sec-
tion, the drops and particles migrate to the surfaces, where they coalesce
and flow from the prec.ipitator as a slurry.
Both theoretical and limited experimental studies on space-charge pre-
cipitation have been conducted at the University of California.19-21/
While the experiments may not accurately model a practical system, the
calculations and experiments done to date indicate that space-charge
precipitation may be a viable approach to the collection of particulates.
Energy requirements for a space-charge precipitator are relatively.un-i
certain. Calculations performed by investigators at the University of
California indicate a total energy requirement of about 0.5 hp/1,000 cfm
of gas; Based on a preliminary cost estimate^ space-charge precipita-
tion appears to be economically competitive with conventional electro-
static precipitation. Additional testing with a pilot-scale model on an
actual industrial source will be needed to allow precise analysis of the
potential of full-scale units.
204
-------
REFERENCES
1. McGraw, M. J., and R. L. Duprey, "Compilation of Air Pollutant Emis-
sion Factors," Environmental Protection Agency, Research Triangle
Park, North Carolina, April 1971.
2. "Participate Pollutants System Study, Volume II - Fine Particle Emis-
sions," Midwest Research Institute, EPA Contract No. CPA-22-69-104,
August 1971.
3. Greco, J., and W. A. Wynot, "1971 Operating and Maintenance Problems
Encountered with Electrostatic Precipitators," Am. Power Conf. Proc.,
33i:345-353.
4. Greco, J., "Electrostatic Precipitators - An Operator's View," Proceed-
ings of Design. Operation and Maintenance of High Efficiency Particu-
late Control Equipment Specialty Conference, St. Louis, Missouri,
March 1973.
5. Shannon, L. J., et al., "Research and Development Program for Control
of Fire Particulate Emissions," Interim Report No. 1, EPA Project
No. R-801615, Midwest Research Institute, November 1972.
6. A New Process1 for Cleaning and Pumping Industrial Gases - The Aeronetics
System, U.S. Patent No. 3,613,333.
7. Private Communication, Mr. H. E. Gardenier, Vice President, Aeronetics,
Inc., September 1972.
8. Schauer, P. J., Ind. Eng. Chem., 43jl532 (1951).
9. Lancaster, B. W., and W. Strauss, "A Study of Steam Injection Into Wet
Scrubbers," Ind. Eng. Chem. Fundamentals. ^.£:362-369, No. 3 (1971).
205
-------
10. Calvert, S., et al., Scrubber Handbook (Contract No. CPA-70-95) APT,
Inc., Riverside, California, August 1972.
11. Goldshmid, J. Jr., D. Leith, and S. Calvert, "Flux Force Scrubbers—An
Engineering Analysis," presented at 164th ACS National Meeting,
New York, New York, 27 August 1972.
12. Strauss, W., Industrial Gas Cleaning, Oxford, Pergamon Press, Limited
(1966).
13. Levich, V., Dokl. Akad. Nauk.. SSSR, 99:809 (1954a).
't.
14. East, T. W. R., and J. S. Marshall, Q. Journal R. Met. Soc.. 80j26
(1954).
15. Tunitsky, N. N., Zh. Fiz. Khim.. 20:1136 (1946).
16. Obukhov, A., and A. Yaglon, Prikl. Mat. Mekh., 15:1 (1951).
17. Beal, S. K., "Turbulent Agglomeration of Suspensions," Aerosol Science,
.3:113-125 (1972).
18. Yoder, J. D., and L. Silverman, "Influence of Turbulence on Aerosol
Agglomeration and Deposition in a Pipe," Paper No. 67-33, 60th
Annual Air Pollution Control Association Meeting, Cleveland, Ohio,
13 June 1967.
19. Faith, L. E., S. N. Buktany, D. N. Hanson, and C. R. Wilke, Ind. Engr.
Chem. Fundamentals, 6:519 (1967).
20. Kostow, Lloyd P., "Design and Testing of Space-Charge Precipitators,"
M. S. Thesis, University of California, Berkeley, California,
6 March 1972.
21. Webber, M. E., "Experimental Studies on Space-Charge Precipitation,"
M. S. Thesis, University of California, Berkeley, California,
5 September 1969.
206
-------
APPENDIX B
EXAMPLE OF PROCEDURE FOR DETERMINING CONTROL
COSTS FOR COMPLIANCE WITH FINE PARTICLE EMISSION STANDARDS
INTRODUCTION
The determination of the control costs required for compliance with
various fine particle emission standards involves the following major
steps:
1. Development of model plant for each industry category.
2. Determination of control-device performance required to meet a
specific fine particle emission standard.
3. Determination of costs for model plant and industry for compliance
with specific fine particle emission standard.
Individual steps in the procedure will be illustrated in the following
sections using the coal-fired power plant category as an example.
SPECIFICATION OF CHARACTERISTICS OF MODEL PLANT
1. Boiler type: pulverized coal fired.
2. Boiler capacity: 400 MW (existing boiler range from < 25 MW up to
1,000 MW or more).
3. Coal usage rate: assume boiler operates at 100% capacity at all times,
Reference 1 states that a 400-MW plant would consume 150 tons/day.
4. Gas flow rate: Ref. 2 gives the following relation between boiler
capacity and gas flow rate:
1 MW = 2,800 acfm2-/
207
-------
Reference 3 indicates that a 400-MW plant would have a gas flow rate of
800,000 scfm. Assuming that the exit gas temperature is 300°F,2/ the gas
flow rate given by Ref. 2 can be converted to scfm as follows:
(2,800 *£*) (400 MW) (-) = 765,000 scfm (1)
MW 760
Since the gas flow rate from Refs. 2 and 3 are in good agreement, the figure
of 2,800 acfm/MW will be used in subsequent calculations.
5. Stack diameter: assume that the exit velocity of stack gas is 4,000
ft/min.
Cross-sectional area of stack = Gas Flow Rate
Exit Velocity
acfm
= r^soo) ~Mi"ir4oo MWI
4,000 ft/min ' ( )
= 280 ft2
Stack diamter = 4 Area)
rr / (3)
i"(4) (280)1
L TT J
= 18.9 ft = 5.76 meters
CONTROL DEVICE PERFORMANCE REQUIRED TO MEET PLUME OPACITY REGULATION
Equation (4) was used to calculate the allowable outlet grain loading for
a specific plume opacity.
W = -K p/L In |- (4)
Io
The efficiency of the control device was then determined from Eq. (5).
208
-------
Efficiency = Inlet Grain Loading - Outlet Grain Loading
Inlet Grain Loading
The inlet grain loading to the control device for the model plant was cal-
culated from the relation:
Inlet Grain Loading = (Emission Factor) (Coal Usage Rate)
Gas Flow Rate
(l90 1* r) (l50 £2S^°al) (y.OOO SSia)
\ ton of coal/ \ hr / \ l£_/
min 6
60 — 1.12 x 10b
hr nun
. 97 grain \ /460 + 300
acf ) \460 + 60 <
= 4.34 grains/scf
The allowable grain loading at the outlet of the control device for plume
opacities of 5% and 10% and the corresponding required control device ef-
ficiency are given in Table B-l.
Table B-l. SUMMARY OF CONTROL EQUIPMENT PERFORMANCE REQUIRED
TO MEET PLUME OPACITY REGULATIONS
(Coal-Fired Power Plant)
Overall Mass
Outlet Grain Loading Efficiency of
Plume Opacity Permitted (grain/scf) Control Device
10% 0.0149 99.66
5% 0.00728 99.83
209
-------
The main type of control device currently installed on coal-fired power
plants to control particulate emissions is the electrostatic precipitator.
The cost of an electrostatic precipitator for the model plant was deter-
mined from Figures 17 and/or 18, Chapter 6, and the cost for the total in-
dustry was determined using the following equation:
Total Cost = (Total Industry Production CaPacitv\ Lt fm Model plant\ (?)
\ Model Plant Production Capacity ] \ /
Table B-2 summarizes the calculation of the annualized costs required to
meet a 107» or 570 plume opacity standard.
CONTROL SYSTEM CORRESPONDING TO BEST INSTALLED CONTROL DEVICE
Electrostatic precipitators are the most efficient particulate control de-
vices currently being installed on coal-fired power plants. Reference 2 in-
dicates that the average design efficiency of electrostatic precipitators
installed on power plants in the 1970-1971 period was 997o (overall mass
efficiency) . The cost of this control device was determined from Figures
17 and/or 18, Chapter 6.
COMPARISON OF CONTROL EQUIPMENT COSTS
Table B-3 presents a comparison of the control costs required for compliance
with the two fine particle emission standards.
210
-------
Table B-2. ANNUALIZED COSTS FOR POWER PLANTS TO MEET 10% AND 5% PLUME OPACITY STANDARD
Opacity Control Annualized Cost
Standard Device Required Model Plant All Plants
10% Opacity 99.66% Electrostatic (a) $0.22/cfm/year /i50^000_JlW\ ($2>5 x 105/year) =
precipitator * ' $94 x 10^/year
(b) ($0.22/cfm/year) (1.12 x 106 cfm) =
$2.5 x 105/year
5% Opacity 99.83% Electrostatic (a) $0.27/cfm/year ^150,000 MW\ ($3>0 x io5/year) =
precipitator \ 40° m I $112 x 106/year
(b) ($0.27/cfm/year) (1.12 x 106 cfm) =
$3.0 x 105/year
-------
NJ
H1
NJ
Table B-3. ANNUALIZED COSTS FOR POWER PLANTS TO MEET 10% AND 5% PLUME OPACITY AND
BEST INSTALLED CONTROL STANDARD
Emission Regulation
Best installed control
device
107. Opacity
57, Opacity
Control System
99% Electrostatic
precipitator
99.66% Electrostatic
precipitator
99.83% Electrostatic
precipitator
Annualized Cost ($/Year)
Incremental Annualized
Cost ($/Year)
Model Plant All Plants Model Plant All Plants
190,000
250,000
300,000
71 x 106
94 x 106
112 x 106
60,000
110,000
23 x 106
41 x 106
-------
REFERENCES
1. "Control of Air Pollution from Electric Power Generators—A State-of-
the-Art Review," Research Cottrell, Inc., June 1969.
2. "Particulate Pollutant Systems Study, Volume III - Handbook of Emis-
sion Properties," Midwest Research Institute, EPA Contract No. CPA-
22-69-104, 1 May 1971.
3. Anonymous, "Basic Technology," Chemical Engineering-Deskbook Issue,
p. 172, 27 April 1970.
213
-------
APPENDIX C
PLUME OPACITY
The opacity of a plume is defined as one minus the transmittance of the
plume, i.e.,
Plume Opacity = 1 - Transmittance (1)
Plume transmittance can be calculated from Eq. (2) :ii?_/
r2
T = I^ = exp [- TT L / QE(Q',m)r2n(r)dr] (2)
rl
where,
or = Size parameter, 2nr/X,
r = Particle radius,
i
A. = Wavelength of light,
m = Refractive index of the particles relative to air,..,-,
n(r) = Size frequency distribution, number of particles of radius r
per volume per Ar,
QE(o-,m) = Particle light extinction efficiency factor, the total light
flux scattered and absorbed by a particle divided by the
light flux incident on the particle,
L = Plume width,
214
-------
I = Intensity of transmitted light, and
I0 = Intensity of incident light.
The light extinction efficiency factor Qjr for spheres, ellipsoids and
cylinders can be computed using the Mie equations. ft/ For pure scatterers
with typical refractive indices, Qg can vary from near zero for very small
particles, to about four when the particle diameter is near the wavelength
of light, and approaches a theoretical limit of two for very large parti-
cles. For spherical particles,
where,
W = Mass concentration of particles in exhaust stream,
p = Particle density, and
g(r) = Mass distribution function.
Substitution of Eq. (3) into (2) yields:
/^
[ - 3/4 ^ J ^ g(r)dr
T = exp [ - 3/4 g(r)dr] (4)
*1
By suitable rearrangements of the preceding equations, Pilat and Ensor?./
have developed a simplified equation to calculate the expected mass con-
centration for various values of plume transmittance (or opacity) , average
particle density, and plume diameter.
W = -K p/L In (I/I0) (5)
where,
_ Specific particulate volume
Extinction coefficient
215
-------
The parameter K is dimensionally similar to the volume surface characteris-
tic size described by Herdan (1960).I/ K is primarily a function of the
particle size distribution, refractive index, and to a lesser degree, the
wavelength of light.
Equation (A) can be used to obtain an understanding of the dependence of
opacity on various parameters, and thereby, some important implications
regarding the use of opacity as an indicator of fine particulate emission.
The effect of mass concentration, path length, and particle density on the
transmittance can be evaluated directly from Eq. (4).
If p is defined as WL/p, and I1 is 3/4 / (-:£) g(r)dr, then Eq. (4) becomes
/*
T = exp (-pi') (6)
For a constant I1, a change in (3 from g^ to $2 results in a change in trans-
mittance given by:
T2 - T (7)
Thus, the percentage change in transmittance caused by a change in W, L,
or p depends on the initial transmittance level. For example, a doubling
of W or of L or a halving of p will result in an 80% reduction of trans-
mittance of TI = 0.2 or a 20% reduction if TI = 0.8.1/
The optical properties of particles are described by the refractive index,
which is a complex number, m = n^-in£. The real part, n^, describes the
light-scattering properties and the imaginary part, n£, describes the light
absorption of the particulate material. The refractive index has a strong
influence on the evaluation of Qg and, hence, on the transmittance.
Figure C-l shows percent opacity plotted against-the mass median radius
for a hypothetical case. Four curves are plotted, each for a different
value of m from 1.33 to 3.0 with n£ assumed equal to zero. Thus, the
effect of the real part of the refractive index on opacity is illustrated.
The lower value of m = 1.33 represents water droplets in air. The m =
1.5 curve approximates the inorganic components of fly ash. The higher
valjues of m are typical of iron oxide, for example.!/
216
-------
100 r-
a.
o
Geometric Standard Deviation of Particle
Size Distribution, (7=2
0.1 1.0
MASS GEOMETRIC MEAN RADIUS RGW (MICRONS)
Figure C-l.
Effect of real part of refractive index on
opacity particle-size relationship!./
217
-------
It is important to note that all of the curves show a maximum opacity in
the 0.1-0.3 Jim range, the opacities falling off rapidly on either side of
the maximum. About 1 um the curves merge and opacity is independent of
the refractive indices. For finer particles, the higher values of re-
fractive index result in higher opacities. A plume composed of iron
oxide particles will, therefore, transmit less light than a fly ash plume
of identical characteristics. For the case where the plumes have identi-
cal characteristics, different levels of plume opacity would, in fact,
represent the same level of fine particulate emissions to the atmosphere.
Figure C-2 illustrates the effect of the imaginary part of the refractive
index, i.e., light absorption, on opacity. The four curves plotted rep-
resent refractive indices all with n^ = 1.5 but with n£ varying from
0 to 0.9. The effect is significant. For mass median radius less than
0.5 um, light absorption by fine particles plays a dominant role in de-
termining opacity. In fact, for the higher values of n2, the calculated
opacities become virtually independent of particle size. This has impor-
tant implications in the use of opacity as an emission indicator. High
opacities could be measured in a plume composed of very fine absorbing
particles whereas an identical plume of nonabsorbing particles might go
unnoticed.!./
Particle size distribution also exerts a strong influence on plume opacity
because of the fact that particles in different size ranges contribute un-
equally to the overall opacity. For nonabsorbing particles, a given mass
of particles less than 0.1 jam contributes less and less to opacity as the
particle size decreases. Above 1 um the same is true as size increases.
As a result, particulates which are monodisperse or nearly so, should be
expected to yield opacities which show a greater dependence on particle
size than more polydisperse distributions.±J
Figures C-3 and C-4 show the composite effects of some of the main param-
eters on the plume opacity as expressed in terms of the parameter K, de-
veloped by Pilat and Ensox.—' Figure C-3 shows the functional dependence
of K on size distribution parameters for a black aerosol, while Figure C-4
depicts the same relationships for a white aerosol.—'
218
-------
loo r-
10 h
u
2
o
»—
1.0 h-
Geometric Standard Deviation of Particle
Size Distribution , O" = 2
WL//) =
0.1
1
0.01
0.1 1.0
MASS GEOMETRIC MEAN RADIUS, RGW (MICRONS)
10
Figure C-2.
Effect of imaginary part of refractive index
on opacity particle-size relationship^-'
219
-------
10'
I
u
JO
10
10
Geometric standard deviation,
Refractive index= 1.96 0.66 i
Wave length of light=550 nm
•2
10 '" 10" .10° lo1 102
Geometric mass mean radius, r8W (microns)
i
Figure C-3. Parameter K as a function of log-normal size
distribution parameters for black aerosol—'
Geometric standard deviation
i
1
Refractive index =1.50
Wave length of light=550 nm
10'" 10"' 10° 10l 102
Geometric mass mean radius, r-gw (microns)
Figure C-4.
distribution parameters for a
Parameter K as a function of the log-normal size
white aerosol-!/
220
-------
REFERENCES
1. Feldman, P. L., and D. W. Coy, "Comparison of Computed and Measured
Opacities: Lignite-Fired Boilers," Research Cottrell, Bound Brook,
New Jersey.
2. Ensor, D. S., and M. J. Pilat, "Calculation of Smoke Plume Opacity
from Particulate Air Pollutant Properties," Journal of Air Pollu-
tion Control Association. 21:496-501, No. 8 (1971).
3. Hardan, G., Small Particle Statistics. Butterworths, London (I960).
*US. GOVERNMENT PRINTING OFFICE: 1974 546-318/337 1-3 221
-------
SELECTED WATER
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
1. Report No.
w
Feasibility of Emission Standards Based on Particle Size
L.J. Shannon, P.G. Forman, W. Park
Midwest Research Institute
425 Volker Blvd.
Kansas City, Missouri 64110
12. if-Sponsoring Organization Environmental Protection Aeencv
-»: i> -o j
.L.r *'
Environmental Protection Agency Report
Number EPA-600/5-74-007, March 1974.
5. R ortD;
e.
5. PF • f ormir, Orgar Won
68-01-0428
13. TypeofRepoi ^
Period Covered
Final
lo.
The technical and economic feasibility of particulate emission standards based on
particle size was assessed in this program. Specific attention was focused on
standards to regulate the emission of fine particulates—particulates below 2 u in
size. The program was divided into four major areas of effort:
1. Analysis of approaches for regulating fine particle emissions from stationary
sources.
2. Definition of technological and economic requirements necessary for imple-
mentation of emission standards.
3. Identification of benefits that would accru if control procedures for -fine
particulates can be implemented.
4. Assessment of overall feasibility of implementation of fine particle emission
standards.
17a. Descriptors
Particulates; Emission Standards
17b. Identifiers
I7c. COW P.P. field & Gro-jp
18 •}
19. Sr-urity Class.
(Report)
20. Security Class.
------- |