1GCEA^SPECIAL REPORT
SECONDARY BENEFITS OF
EMISSION CONTROL SYSTEMS USED
TO MEET NSPS FOR UTILITY BOILERS
MAY 1980
By
L. L. Scinto
TRW, Inc.
Contract No. 68-02-3138
EPA Project Officer: W. H. Ponder
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK. N.C. 27711
-------
1GCEA^SPECIAL REPORT
SECONDARY BENEFITS OF
EMISSION CONTROL SYSTEMS USED
TO MEET NSPS FOR UTILITY BOILERS
MAY 1980
By
L. L. Scinto
TRW, Inc.
Contract No. 68-02-3138
EPA Project Officer: W. H. Ponder
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK. N.C. 27711
-------
DISCLAIMER
This report has been reviev^ed by the Industrial Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publica-
tion. Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
-------
ABSTRACT
A study of airborne emissions from controlled utility .boilers was made
to characterize the performance of emission control systems in removing
non-criteria pollutants. Both particulate control and flue gas desulfuri-
zation systems were evaluated. Non-criteria pollutants considered were
trace elements, organic compounds, and primary sulfates.
Removal efficiencies and emission factors for non-criteria pollutants
were reported for a flue gas desulfurization system and various particulate
control devices.
Emissions of toxic trace elements from controlled and uncontrolled
bituminous coal-fired boilers were estimated for various types of control
devices.
A preliminary assessment of the environmental impact of controlled and
uncontrolled trace element emissions from pulverized bituminous coal-fired
dry bottom boilers was made. It was concluded that control of emissions to
the level of current new source performance standards would result in sub-
stantial reduction in the adverse health effects of airborne emissions from
electric power generation.
This study was performed- under the Conventional Combustion Environmental
Assessment program, Contract No. 68-02-3138, by TRW, Incorporated, under the
sponsorship of the U.S. Environmental Protection Agency. The work was com-
pleted as of May 1980.
-------
CONTENTS
Abstract . iii
Figures v
Tables vi
1. Introduction 1
Background 1
Organization 3
2. Conclusions 5
Secondary Benefits of Particulate Control 5
Secondary Benefits of Flue Gas Desulfurization 6
Environmental Impact Assessment 7
3. NSPS for Utility Boilers 8
Combustion Processes. .... 8
Applicability of NSPS 9
4. ECS Performance 12
Particulate Control Systems 13
Flue Gas Desulfurization Systems 30
References 35
Appendices
A. Background Data 39
B. Particulate ECS Data 47
C. FGD System Data 56
D. Novel Emission Control Systems ; .. 60
E. Environmental Impact Analysis 65
-------
FIGURES
Number Page
. 1 Trace element emissions from a pulverized coal-fired 16
boiler .......
2. Trace element emission factors for western coal-fired
power plants using cyclones and wet scrubbers for emission
controls 19
3. Trace element emission factors for western coal-fired
power plants using cyclones and wet scrubbers for emission
controls , 20
4. Trace element emissions from two western coal-fired power
plants 23
5. Performance of hot- and cold-side electrostatic
preci pita tors 24
6. Performance of a cold-side ESP on a coal-fired power plant . 25
7. Effect of FGD system on trace element emissions. ...... 32
D-l. Particulate control by novel devices 61
-------
TABLES
Number Page
1. Standards of Performance for Electric Utility Steam
Generators - 40 CFR Part 60. .. 10
2. Actual Trace Element Collection Exhibited by Cyclone
Separator 15
3. Trace Element Collection Efficiencies Exhibited by Venturi
Scrubbers - Coal-fired Power Plants. . l. . . . 17
4. Trace Element Collection Efficiencies Exhibited by
Electrostatic Precipitators - Coal-fired Power Plants. ... 22
5. Controlled and Uncontrolled Benzo(a)pyrene Emissions From
a Pulverized Coal-fired Power Plant , 27
6. Controlled Benzo(a)pyrene Emissions From Coal-fired Power
Plants 28
7. Effect of Electrostatic Precipitator (ESP) on PSE From
Combustion Sources 29
8. Trace Element Collection Efficiency Exhibited by FGD
System - Coal-fired Power Plant 31
9. Reduction of SOa and S04= Emissions by FGD System During
Coal Firing 34
10. Reduction of Fluoride Emissions by FGD System During Coal
Firing . 34
A-l. Average Trace Element Concentrations in Coal ... 41
A-2. Average Trace Element Concentrations of Residual Oil .... 43
A-3. Partitioning of Elements in Coal Combustion Residues .... 45
A-4. Trace Element Enrichment Factors for Coal-fired Utility
Boilers Equipped With Particulate Emission Controls. .... 46
B-l. Size Distributions for Controlled and Uncontrolled
Particulate Emissions From Utility Boilers 48
B-2. Efficiencies of Particulate Removal by Control Devices for
Various Size Fractions 50
B-3. Estimated Trace Element Emission Factors From Combustion of
Bituminous Coal 53
C-l. Controlled and Uncontrolled Trace Element Emissions From a
Coal-fired Utility Boiler Equipped with FGD System 57
C-2. Comparison of Trace Element Removal Efficiencies of an SOg
Scrubber on a Coal or Fuel-Oil Fired Industrial Boiler ... 58
vi
-------
TABLES (Continued)
Number Page
D-l. Performance of Novel Fine Participate Control Devices to
Conventional Venturi Scrubbers . 62
D-2. Tabulated Data From Nahcolite Test Program ... 64
E-l. Environmental Assessment of Uncontrolled Emissions From
Pulverized Coal-fired Dry-bottom Boilers . 67
E-2. Environmental Assessment of Controlled Emissions From
Pulverized Coal-fired Dry-bottom boilers .......... 68
E-3. Environmental Assessment Summary 68
vii
-------
SECTION 1
' INTRODUCTION
The primary benefit of an emission control system (ECS) is the removal
of criteria pollutants from the boiler flue gas stream. In increasing
numbers, utilities are turning to the use of particulate and/or gaseous
ECS's to comply with emission reductions required by New Source Performance
Standards (NSPS) promulgated by the U.S. Environmental Protection Agency
(EPA). Control devices such as wet scrubbers, electrostatic precipitators
(ESP's), and fabric filters (baghouses) are being used to mitigate par tic u-
late emissions. Flue gas desulfurization (FGD) systems are used to control
gaseous sulfur dioxide emissions.
In addition to the primary benefits of sulfur dioxide and particulate
removal systems, secondary benefits are obtained by the application of
control technologies to utility boilers. Besides being able to mitigate
emissions of criteria pollutants required by NSPS, emission control systems
have the ability to remove potentially hazardous air pollutants from boiler
flue gases. Conventional ECS's can remove a substantial portion of such
non-criteria pollutants as trace elements, organic compounds, and primary
sulfates from combustion source emissions.
This paper will examine the ability of particulate control and FGD
systems to mitigate emissions of non-criteria pollutants from utility
boilers. Data from the open literature and from the Conventional Combus-
tion Environmental Assessment (CCEA) data base will be presented to quantify
removal efficiencies and emission factors for sources equipped with various
ECS's.
BACKGROUND
In considering the emission of non-criteria pollutants from combustion
sources, and the .ability of ECS's to mitigate these emissions, it is helpful
to understand the origins and characteristics of the emitted pollutants.
Trace element emissions arise from the presence of these elements in the
-------
fuel. The trace element emissions from fossil fuel-fired utilities, equipped
with modern pollution control devices, consist primarily of submicron
particles. A large body of recent work has shown that the smaller fly ash
particles resulting from combustion may be considerably enriched in several
toxic trace elements. The most widely accepted model for trace element
enrichment in fly ash formation involves the volatilization of these elements
during combustion, followed by condensation or adsorption over the available
matrix material {composed primarily of the nonvolatile oxides of Al, Mg, and
Si). The smaller particles, which show the highest concentration of several
trace metals, are not as efficiently collected by pollution control devices
as are larger fly ash particles. These smaller particles, enriched in poten-
tially toxic trace metals, also have the highest atmospheric mobilities and
are deposited preferentially in the pulmonary and bronchial regions of the
respiratory system.
Emissions of organic compounds, particularly polycyclic organic matter
(POM), originate largely from inefficient mixing of fuel and oxidant during
combustion rather than from strictly the fuel itself. Gaseous POM is formed
during combustion by pyrosynthetic and pyrolytic reactions, and condenses
or is adsorbed on available ash surfaces in cooler parts of the combustion
process. At stack temperatures, a significant portion of the POM may still
be in the gaseous form, and can be removed by wet scrubbing devices. Parti-
culate POM. may be removed by particuVate control devices.
Primary sulfate emissions (PSE) arise both because of the presence of
sulfur in the fuel and because of improper combustion conditions. With
fuels of high sulfur content burned with a high level of excess oxygen,
fuel sulfur is oxidized to not only SOg, but to higher oxides such as SO^,
HgSO^, and metallic sulfates (MSOJ, The proportion of sulfate emitted as
gas, liquid, and particulate varies from source to source, and thus the
efficiency of different types of ECS's in reducing PSE also varies.
A thorough discussion of the health effects of non-criteria pollutant
emissions is beyond the scope of this report. However, when appropriate,
Maximum Acute Toxicity Effluent (MATE) values for the health effects of air
emissions have been used to quantify the potential adverse effects of these
emissions on the human population. Numerical values of the MATE correspond
-------
to the concentration of a substance estimated to cause minimal adverse
effect in a healthy receptor exposed once or intermittently for short time
periods. Discharge severity (DS) has also been used in this report to
characterize health effects. DS is a simple index of the potential hazard
of a single substance in a discharge. It is used in the early stages of an
environmental assessment to (1) identify the substances in a waste stream
which are potentially the most harmful; (2) evaluate the need for further
characterization of the discharge or development of improved control tech-
nology; or (3) present the results of waste stream chemical analyses in
terms of health effects in a manner which facilitates review and interpre-
tation. DS is a unitless quantity equal to the discharge concentration of
*
a particular substance divided by its MATE value .
ORGANIZATION
This report deals specifically with utility boilers. The applicability
of NSPS to utility boilers is discussed in Section 3. A brief description
of combustion processes and a summary of regulations is included.
Data on the ability of various ECS's to reduce non-criteria pollutant
emissions are presented and analyzed in Section 4. The data are predomi-
nantly from studies on coal-fired utility boilers rather than units fired
with oil or gas. Coal-fired sources have been more extensively characterized
because they represent the largest source of non-criteria pollutant emissions,
and because they hold promise as being the major contributors to U.S. energy
self-sufficiency in the near future. Separate subsections deal with parti-
culate controls and FGD systems, and present data on the ability of these
ECS's to remove specific classes of non-criteria pollutants.
Several appendices are included at the end of the report. They contain
supplemental data pertinent to discussions in the text, and are presented
as follows:
In calculating DS, MATE has been equated to the Discharge Multimedia
Environmental Goal (DMEG) for the same substance.
-------
Appendix A - Background Data.
Appendix B - Particulate ECS Data.
Appendix C - FGD System Data.
f Appendix D - Novel.Emission Control Systems.
t Appendix E - Environmental Impact Analysis.
The major conclusions of the report, based on data herein, follow the
Introduction as Section 2.
-------
SECTION 2
CONCLUSIONS
Emission control systems are capable of removing a large portion of
the potentially hazardous pollutants produced from fossil-fuel combustion.
In addition to removing criteria pollutants such as SOg, NOX, and particu-
lates, emission controls remove non-criteria pollutants. Such pollutants
include trace elements, organic compounds, and sulfates.
In the U.S., coal combustion is responsible for the majority of annual
mass emissions of non-criteria air pollutants from power generating facil-
ities. Emissions of non-criteria pollutants from oil combustion are less,
because both emission factors and energy consumption are less for oil
combustion than for coal. Combustion of natural gas has a negligible con-
tribution to non-criteria pollutant emissions to the atmosphere.
In assessing the secondary benefits of emission control systems, both
particulate control and flue gas desulfurization systems have been consi-
dered. Although the quantity of relevant source test data quantifying
non-criteria pollutant collection efficiencies and emission factors is
meager, sufficient data exist with which to estimate average emissions.
Conclusions based on the existing emissions data base are presented below
for particulate control and flue gas desulfurization systems. Further
conclusions based on emission estimates are also presented, in order to
provide a quantitative approximation of the secondary benefits of emission
controls.
SECONDARY BENEFITS OF PARTICULATE CONTROL
Control systems which most efficiently collect fine particulates
provide the greatest reduction in emissions of non-criteria
pollutants, because these species tend to concentrate on fine
parti cu-lates.
s For dry particulate control systems, volatile trace metals
(particularly Hg, Sb, and Se), organics, and halogens are the
least efficiently collected non-criteria pollutants, because
these species are often present in the vapor phase.
-------
§ Mechanical collectors exhibit poor removal efficiencies for
trace elements. Collection efficiencies for individual
trace elements generated from coal combustion range from
3-9521. Emission factors for trace elements with low Maximum
Acute Toxicity Effluent concentrations (for health effects
of air emissions, MATE less than one milligram per cubic
meter) range from about 1-1000 pg/J heat input.
t Electrostatic precipitators are normally able to collect
greater than 90% of any given trace element generated from
coal combustion'(with the exception of Hg, Sb, Se, and the
halogens). Emission factors for trace elements with low
MATE concentrations range from about 0.1-100 pg/J.
Electrostatic precipitation has been shown to reduce primary
sulfate emissions from coal and. oil combustion by approxi-
mately 50%.
A combination of mechanical collectors and an electrostatic
precipitator has been shown to collect 2-71% of the benzo(a)-
pyrene emitted from a coal-fired utility boiler.
§ Trace element collection efficiencies for wet scrubbers used
on coal-fired utility boilers approach the values for electro-
static precipitators for some elements. Compared to mechanical
collection, wet scrubbing results 1n order of magnitude
decreases in emission factors for at least half of the low-
MATE compounds identified in coal fly ash.
Fabric filters installed on coal-fired combustion sources have
been found to. have collection efficiencies for Hg and Se at least
as high as for electrostatic precipitators. Under optimum condi-
tions, a fabric filter has removed 100% of the Hg 1n coal from the
flue gas. However, As and Be collection efficiencies for fabric
filters appear to be less than for electrostatic precipitators.
SECONDARY BENEFITS OF FLUE GAS DESULFURIZATION
From 4-97% of low-MATE trace element emissions were collected
by a limestone slurry sulfur dioxide scrubber operating at a
coal-fired utility. With the exception of As, removal effi-
ciencies exceeded 65%. Emission factors after the control
device ranged from less than 1-1000 pg/J for low-MATE trace
elements.
The concentration of polycyclic organic matter was reduced
from 18.7 micrograms per cubic meter to below the detection
limit by the use of the limestone slurry scrubber.
- Primary sulfate emissions can sometimes be reduced by wet
scrubbing, however, under some operating conditions, reentrain-
ment of scrubber liquor can occur to reduce or negate any
emission reductions.
-------
t Fluoride emissions were reduced by limestone slurry scrubbing
by more than 62%.
ENVIRONMENTAL IMPACT ASSESSMENT
t The discharge severity of low-MATE trace elements from uncon-
trolled combustion of pulverized bituminous coal in dry-bottom
boilers is approximately equal to the discharge severity of
sulfur dioxide and nitrogen oxides. If control to the level
of NSPS is practiced, the low-MATE trace element discharge
severity is only about 1% of that for sulfur dioxide plus
nitrogen oxides.
Control of emissions from pulverized bituminous coal-fired dry-
bottom boilers to the level of current NSPS affords a 99.5%
reduction in the low-MATE trace element discharge severity,
but only a 54% reduction in sulfur dioxide plus nitrogen oxides
discharge severity.
-------
SECTION 3
NSPS FOR UTILITY BOILERS
Standards of performance for new stationary sources of air pollution
have been promulgated by the EPA for utility boilers. These new source
performance standards (NSPS) apply to very specific categories of utility
sources. In this Section, these sources and their emission characteristics
are described briefly. The regulations pertaining to these sources are
summarized and some of the specifics of their applicability are presented.
COMBUSTION PROCESSES
According to estimates by the National Electric Reliability Council
(NERC), coal combustion accounted for 64.8% of the fuel energy consumed in
external combustion for electricity generation in the U.S. during 1978 (1).
Petroleum and natural gas accounted for 21.7% and 13.6%, respectively.
Since potentially hazardous emissions of non-criteria pollutants are gener-
ally greatest for coal (on a per joule basis), coal combustion represents
the largest source of these pollutants.
Emissions of two important classes of non-criteria pollutants - trace
elements and POM - are associated largely with particulate emissions. Since
particulate emissions arise from ash in the fuel, emissions of trace elements
contained in ash are high from coal combustion, moderate from oil combustion,
and essentially nil for gas combustion. Similarly, the relatively poor
combustion achieved with coal leads to higher POM emissions than from oil
or gas.
The characteristics of emitted particulates from coal and oil combustion
differ substantially. These differences are important in analyzing the
effectiveness of control devices. Coal combustion produces more particulate,
but oil combustion generates smaller fly ash particles which are not as
efficiently removed by control devices. The amount of ash retained in the
boiler as bottom ash versus the amount emitted as fly ash depends on firing
configuration. For bituminous coal combustion (the major source of U.S.
-------
electric power), the average fraction of coal ash collected as bottom ash
(Eg) for different boiler types ranges from 0.2 for pulverized dry-bottom
boilers, to 0.35 for pulverized wet-bottom boilers, to 0.865 for cyclone
boilers (2). That is, of the total ash in coal, 80% is emitted as fly ash
from pulverized dry-bottom boilers, 65% from pulverized wet-bottom boilers,
and 13.5% from cyclone boilers. A more detailed characterization of com-
bustion source emissions, related specifically to trace element emissions,
is given in Appendices A and B.
APPLICABILITY OF NSPS
The NSPS for utility boilers, which comprise 40 CFR Part 60 Subpart Da,
are shown in Table 1. These regulations were promulgated June 11, 1979
(44 FR 33580) and incorporate some changes from the Subpart Da regulations
as proposed September 19, 1978 (43 FR 42154).
In general, the Subpart Da regulations apply to electric utility steam
generating units which: (1) are capable of firing over 73 MW (250 million
Btu/hr) heat input of fossil fuel (alone or in combination.with any other
fuel), and (2) for which construction or modification is commenced after
September 18, 1978. The facility of which the unit is a part must: (1) be
physically connected to a utility power distribution system, and (2) be
constructed for the purpose of selling (a) greater than 25 MW electrical
output and (b) more than one-third of its potential electrical output
*
capacity .
Exclusions
The following sources are excluded from all or part of the applicabili-
ty of the Subpart Da regulations:
Any source modified to accommodate other (nonfossil) fuels
shall not come under the applicability of Subpart Da.
Any source modified from gas or liquid fuel to accommodate
any other fuel (fossil or nonfossil) shall not come under
the applicability of Subpart Da.
Potential electrical output capacity is defined as 33% of the heat input
rate.
-------
TABLE 1. STANDARDS OF PERFORMANCE FOR ELECTRIC UTILITY
STEAM GENERATORS - 40 CFR PART 60
Affected Facility
Coal-fired boilers
(and coal-derived.
fuels)
Pollutant
Participate
Opacity
Emission Level
13 ng/J (0.03 Ib/million Btu)
20% (27% for 6 min/hr)
Monitoring Requirement
Ho requirement
Continuous
*«
Oil or gas-fired
boilers
SO
N0x-rt
Anthracite,
Bi tumi nous,
and Lignite**
Subbituminous
coal
Coal-derived
fuels and
shale oil
- >1ore_than 252
coal refuse
Particulate
Opacity
SO,
NO - oil
NO;;- gas
520 ng/J (1.20 Ib/million Btu)t
and 90% reduction*, except 70%
reduction when emissions are
less than 260 ng/J (0.60 lb/
million Btu)*
260 ng/J (0.60 To/million Btu)
220 ng/J (0.50 Ib/million Btu)
220 ng/J (0.50 Ib/million Btu)
Exempt
13 ng/J (0.03 Ib/million Btu)
20% (27% for 6 min/hr)
340 ng/J (0.80 Ib/mill1on Btu)
and 90% reduction* or
86 ng/J (0.20 Ib/million Btu)
(no reduction requirement)
130 ng/J (0.30 To/million Btu)
86 ng/J (0.20 Ib/million Btu)
Continuous compliance
Continuous compliance
No requirement
Continuous «
Continuous compliance
Continuous compliance**
Continuous compliance
Continuous monitors are used to determine excess emissions only, unless noted as "continuous
compliance".
* For SRC-1 an 85% reduction requirement applies (24-hour average).
* Percent reduction requirement does not apply to facilities firing 100% anthracite, resource
recovery facilities firing less than 25% fossil fuel (90-day average), or facilities located
in noncontinental areas.
30-day roll inn average, except where noted.
^Commercial demonstration permits are available for: SRC-1 (SOg) 1.20 Ib/million Btu and 80S
reduction (24-hour); FBC ($02) 1.20 Ib/million Btu and 85% reduction; Coal liquefaction (NOX)
0.70 Ib/nrillion Btu.
**If more than 25% lignite which was mined in North Dakota, South Dakota, or Montana is fired
in a slag tap furnace, the standard is 0.80 Ib/million Btu.
10
-------
t Resource recovery facilities which derive greater than
25% of their heat input (quarterly basis) from nonfossil
fuels are exempt from the sulfur dioxide regulations
contained in Subpart Da.
a Facilities burning only anthracite coal are exempt from
the percent reduction sulfur dioxide requirements of
Subpart Da, but must meet the 520 ng/J emission limit.
Other Aspects of NSPS
In addition to setting standards of performance, NSPS regulations:
Define "emergency conditions" of control system mal-
function for the purpose of determining compliance during
such malfunction, and
t Allow the granting of commercial demonstration permits,
waiving some parts of the NSPS for a period of time to
allow for an initial full scale demonstration of a plant
utilizing new technology.
11
-------
SECTION 4
ECS PERFORMANCE
A review of the literature of air pollution control technology was
conducted to determine the effectiveness of conventional control devices
in reducing the emission of non-criteria pollutants from utility boilers.
Results of this review are presented as controlled and uncontrolled emission
factors (mass per unit heat input) for the source, and collection efficiency
of the control device. Emission rates (mass per unit time) are not general-
ly reported, since they are dependent on source size, and can be accurately
estimated-from-emi-ssion-factors. Non-criteria pollutants considered were
mainly trace elements, but also included organics (including POM), primary
sulfates (SOj and S0^~), and fluoride.
Control .systems-considered in this evaluation included both particulate
control and FGD systems. Particulate control devices included cyclones,
mechanical precipitators (multiclones), wet scrubbers, ESP's, and fabric
filters. Because of limited data, only one type of FGD system was evaluated.
The data base for non-criteria pollutant emissions from controlled
utility boilers is not extensive. Only a limited number of studies have
focused on emissions of these pollutants, and of these studies, only a few
have reported on the performance of control devices. The major sources of
data for this report are summarized in Appendix B.
In this section actual source test data are presented on the perfor-
mance of particulate control and FGD systems in utility boiler applications.
Since the number of sources for which data are available is minimal, the
data presented-should be interpreted as being sound, but statistically in-
adequate. In an attempt to lend statistical validity to these results,
supplementary data in Appendix A have been used to estimate emission factors
for trace element emissions from controlled utility boilers. The calcula-
tions and assumptions used for these estimates are given in Appendix B. The
statistical significance of the emission estimates are discussed in terms of
the adequacy or inadequacy of the data base.
12
-------
One further qualification should be made concerning the data presented
in this Section. Because collection efficiency of a control device is
dependent on inlet particle concentration, ECS configuration plays a part
in determining control device efficiency. Thus, the collection efficiency
for a wet scrubber operating on a boiler flue gas stream may not equal the
efficiency of the same device on a flue gas stream that has undergone
electrostatic precipitation. Control device configuration may also affect
enrichment factors (defined in Appendix A). These facts will affect the
quality of emission estimates derived from any data presented in this report.
PARTICULATE CONTROL SYSTEMS
Reduction of non-criteria pollutant emissions, especially toxic trace
elements, is enhanced if a particulate control system can efficiently remove
the small particles on which these pollutants are concentrated. Four con-
ventional technologies that have been used in the utility industry to control
particulate emissions are discussed below. A brief assessment of each
technology precedes the available source test data for that technology. A
summary of novel devices is included in Appendix D.
Mechanical Collection
Technology Assessment
Cyclone separators are the most common mechanical collectors used by
utilities to mitigate particulate emissions. Cyclones can efficiently
*
remove particles of approximately 5 urn diameter at dust loadings of 2-200
3 3
g/m (1-100 gr/ft ) with only a moderate pressure drop. Because of their
ability to handle high dust loadings, cyclones are best suited for use as
precleaning devices for other control processes. However, their inability
to collect fine particles (<3 pm), which are associated with adverse health
effects, is a major disadvantage. Typical mass collection efficiency for
these devices range from 20-30% for cyclone boilers (coal and oil), 40-60%
for pulverized coal-fired boilers, and 70-85% for stoker-fired boilers (3).
Throughout this' report, particle size or diameter will be in terms of the
equivalent aerodynamic diameter.
13
-------
High efficiency cyclones (mechanical precipitators, multiclones) offer
greater collection efficiency, but at higher pressure drop. Collection
efficiencies range from 30-40% for cyclone boilers, 65-75% for pulverized
coal-fired boilers, and 85-95% for stoker-fired boilers (3).
Trace Element Emissions
The ability of cyclone separators to remove trace elements has not been
well characterized. Trace element collection efficiency for a cyclone
separator was reported by Radian Corporation (4) for a unit installed on a
250-MW cyclone boiler firing North Dakota lignite. Results of the study
are presented in Table 2.
Emission factors for 18 trace elements from an Eastern coal-fired
power plant ..are shown_in F-igure 1. Data are from Cowherd et al (5), for a
125-MW pulverized dry-bottom unit. Emissions are reported as picograms of
trace element per joule heat input in coal (pg/J) .
Wet Scrubbers
Technology Assessment
A common type of wet scrubber designed for particulate removal is the
venturi. This device can attain a total mass collection efficiency of 99%
or more. Collection efficiency for large particles is high, but decreases
with decreasing particle size. Energy input has a strong influence on
performance. At higher energy inputs (pressure drops), collection effi-
ciencies for fine particles increase. Other types of wet scrubbers include
cyclonic, spray tower, and packed bed designs. Wet scrubbers are often
designed to remove not only particulates but sulfur dioxide as well.
Trace Element Emissions
The results of two studies on the impact of venturi scrubbers on trace
element emissions from coal-fired power plants were reported by Radian (6).
Trace element collection efficiencies for these tests are presented in Table 3.
Throughout this report, numerical values reported for emission factors in
pg/J are approximately equal to hourly emissions in grams for a source
with potential electrical output capacity of 92 MW.
-------
TABLE 2. ACTUAL TRACE ELEMENT COLLECTION
EXHIBITED BY CYCLONE SEPARATOR*
Trace
Element
Al
As
B
Ba
Be
Ca
Cd
Co
Cr
Cu
F
Fe
Hg
Mg
Mn
Mo
Ni
Pb
Sb
Se
V
Zn
Collection
Efficiency, %
66.0
75.3
31.4
95.4
84.3
54.8
44.0
45.1
27.7
56.8
25.3
54.2
3.2
61.0
66.8
24.9
18.6
30.0
7.4
33.1
36.2
39.4
Source: Data from Radian (4) for a
coal-fired power plant with cyclone
boiler firing North Dakota lignite.
15
-------
PRRTICULRTE CONTROL BY TWO FOUR-BRNK. MULTICLONE UNITS
10'
CL
LJ
U
_J
As Ba Be Ca Cd Cl Co Cr
Cu F Hg
CLEMENT
Mn N1 Pb Sb Sn V Zn
Figure 1. Trace element emissions from a pulverized coal-fired boiler.
-------
3. " TRACE- ELEMENT "COLtfefiON-EFFieiENCIES EXHIBITED-BY-
VENTURI SCRUBBERS - COAL-FIRED POWER PLANTS
Element
Al
As
B
Ba
Be
Br
Ca
Cd
.. Cl
Co
Cr
Cu
F
Fe
Hg
K
Li
Mg
Mn
, . Mo .
Na
Ni
. Pb
.. Sb
Se
Si
Sn
Sr
Th
U
V
Zn
Collection Efficiencies,
Venturi
99.5
96.3
NDt
99.5
ND
ND
99.7
ND
98.4
>99.8
96.1
ND
ND
>99.9
ND
>99.9
ND
99.9
99.9
98.4
99.9
ND
ND
95.5
85.0
ND
ND
99.8
>99.9
99.5
99.2
99.3
Percent
Venturi
99.7
92.1
93.6
99.0
99.2
ND
99.0
92.3
ND
96.9
88.9
99.3
98.0
99.2
12.6
- ND
ND
98.5
ND
52.8
ND
95.0
98.0
99.3
97.8
ND
ND
ND
ND
96.0
97.0
97.4
Source: Radian (6).
ND - No data.
17
-------
Figures 2 and 3 present data from Radian (4) and Ondov et al. (7) on
trace element emission factors for utilities firing western coal. Wet
scrubber data are from four tests on boilers firing pulverized coal, and
equipped with venturi scrubbers. Data for cyclone firing, also from Radian
(4), are from a single test of a cyclone boiler firing North Dakota lignite,
and equipped with cyclone collectors. MATE values used in this report to
categorize non-criteria pollutants describe the concentration of contaminants
in source emissions to air that will not evoke significant or irreversible
harmful responses in humans, provided exposures are of limited duration.
V
MATE'S are taken from del and and Kingsbury (8) and reflect most recent
changes (September 1979) made to this document.
Data for collection of non-criteria pollutants by a venturi scrubber
operated.Jn FGD mode are presented in the section on FGD systems and in
Appendix C.
Electrostatic Precipitators
Technology Assessment---
ESP's are effective in controlling emissions of both large and small
particulates from combustion sources. Overall collection efficiency of
high-performance ESP's can exceed 99.5%. Studies have shown ESP efficiency
reaches a minimum for particle diameters of approximately 0.5 vm, but still
exceeds 90% for these particles. In addition to high collection efficiency,
another advantage to ESP's is the low pressure drop across the device.
However, ESP's have the disadvantages of high capital costs and varia-
tions in efficiency with changing fly ash resistivity and flue gas flow rate.
The latter problem has led to adoption of two stage ESP configurations. For
high resistivity fly^ash typical of low-sulfur western coals, the efficiency
of ESP's is reduced. ESP's located downstream of the air heaters in the
normal cold-side configuration suffer most from this variation in resis-
tivity. Newer designs have incorporated ESP's in a hot-side configuration
upstream of the air heaters, which allows operation of optimum fly ash
resistivity. However, hot-side ESP's may encounter resistivity problems
when the coal is-.low in sodium.
18
-------
103
0.
z
!r 10
a
I
u
EMISSION FACTOR, P ICOGRflMSxJOU
L °o S-
y
X
/
y
/
/
*
f
f\
f\
f
ff
f f
t\
ff
f f
f
t
f t
ff
f\
f .
.
$
%
\
'/
V
'/
^
.,
r
f'f
f'f
f'
f'f
r[
»
rt
f
f'f
r'f
/
f[
f\
I
I
1
~
T
f _
,«,
/^
(**
,[
/t
/|
/^
ff
'" As Ba Be Cd
ELEMENTS WITH MrVTE
f t-
f ,
^*
F*|
,»'<
f\
r*_i
,*,
Se Tl
1
V
/
/
V
y
1
t
r*[
f
f\
f\
f\
f\
f\
f _
r»|
f\
f i
f
f\
f\
,»_
f
f'f
f*,
f
r
\l
Figure 2. Trace element emission factors for western coal-fired
power plants using cyclones and wet scrubbers for
-emission controls (4, 7).
The four plants using scrubbers were all fired with pulverized coal; the
one plant using cyclones was cyclone fired. Maximum acute toxicity
effluent (MATE) values from (8) are for source emissions to air based on
health effects.
19
-------
to* ^
»-
Q_
FXISSION FACTOR, P ICOORAMS/'JOU.E HEAT IN
I .i s * . ^
A
\
V,
4
^
1
1
~
?7
f .
f ,
f .
f
f 4
f
f .
f .
f t
f 4
f ,
f
f
f
f
f .
t .
f 4
f .
f .
(
I
»
~
*"
f f*
f\4
f'f
1
I
F
r*
t* t
f\4
f'f4
f\>
f 1
f'f-
f'(
W
F
1
r^l
£
*
e
EL
-
rn
f i
r
r '
f
DC
1
NT
7.
%
i
C
S
'"
r.-
t
f 4
f
f -
f
f
r'f
HI1
L
[H
^
1
MRTE
.
X
1
>l
^
1*
^
9
^.1
M
""
EM
IU
H
iNT
.!(
1
n
Wf
~
T1
r -
^ .
f
f
f 4
t '
f
*f
H
CU
1
to
SIC K
N
1
^
^
^
!
re*
-i
i
f i
/ .
f
f
f
f
f
f .
f
f
f
f
f *
f t
f
f
f
f
f
f *
f '
f
1
S
1
n
f 4
f
f ,
t 4
f
f 4
f
S
C
.S
S
I
r
YCI
CR
CR
~
yn
f 4
f 4
f .
f
f .
f 4
f
f ,
f 4
''f
1C*
UB.
LA*.
z
E
»
*
1
n
fO
m
-
*
ff'
f 4
f 1
f 4
f t
f'f
'
'
'~
^
Figure 3. Trace element emission factors for western coal-fired
power plants using cyclones and wet scrubbers for
emission controls (4,7).
The four plants using scrubbers were all fired with pulverized coal; the
one plant using cyclones was cyclone fired. Maximum acute toxicity
effluent (MATE) values from (8) are for source emissions to air based on
health effects.
20
-------
Trace Element Emissions
Selected data from four different studies (4, 9, 10, H) Of Esp trace
element collection efficiency are presented in Table 4. Variations in
specific elemental collection efficiencies for the four ESP's tested reflect
differences in design capacity, operating conditions, and configuration.
In most cases, collection efficiencies are high, and in good agreement with
data from the other sources tested.
Mann et al (12) reported the results of a study of trace element
emissions from two tangentially-fired utility boilers firing western sub-
bituminous coal. Although both coals were of the same type, they differed
in moisture and ash content and heating value. In one plant, a hot-side ESP
operating at 370°C (700°F) was used for particulate control, and in the
other plant, a cold-side ESP operating at 120°C (250°F) was used. Total
(particulate and vapor) trace element emissions after the ESP's were
reported for 13 elements. These data are shown in Figure 4.
Trace element emission factors for pulverized western coal-fired power
plants are summarized in Figure 5. Only data for elements with low MATE
values (<1 mg/m ) are shown. Two sets of data each'for hot-side (4, 12)
and cold-side (7, 12) ESP configurations were plotted to indicate the
minimum (shaded bar) and maximum (dotted line) emission factor measured
for each element.
Lyon (13) has summarized the results of extensive studies on a coal-
fired power plant in Tennessee, performed by Oak Ridge National Laboratories.
Tests were conducted on a.290-MW cyclone boiler firing eastern bituminous
coal, and equipped with a cold-side ESP. Figure 6 shows the results of two
tests in which emission factors were computed after the control device.
Note that for selenium, analysis results for Test 1 performed with two
different methods yielded emission factors of 0.8 and 10 pg/j.
Organic Emissions
The ability of an ESP to collect compounds of POM turns on its efficient
removal of particulate on which the POM is adsorbed. However, POM emitted
in the gas phase cannot be removed by ESP's.
21
-------
TABLE 4. TRACE ELEMENT COLLECTION EFFICIENCIES EXHIBITED BY
ELECTROSTATIC PRECIPITATORS - COAL-FIRED POWER PLANTS
Element
AT
As
B
Ba
Be
Br
Ca
Cd
Cl
Co
Cr
Cu
F
Fe
Hg
K
Li
Mg
Mn
Mo
Ng
. Ni
P
Pb
Sb
Se
Si
Sn
Sr
Th
U
V
Zn
Total ash
Plant A*
99.2
99.9
94.7
99.9
97.6
NO
99.0
95.5
4.5
98.2
85.6
99.1
-92.3 -
98.6
0.0
ND
ND
99.0-
98.6
89.2
ND
78.5
ND
91.6
96.0
61.8
ND
ND
ND
ND
98.2
92.2
96.3
99.1
ESP Collection
Plant B*
NDf
ND
ND
ND
ND
ND
ND
98.8
ND
ND
99.8
ND
ND
99.6
ND
ND
ND
ND
100
ND
ND
99.7
ND
99.3
99.1
94.3
ND
ND
ND
ND
ND
99.9
99.6
99.7
Efficiency,
Plant C*
99.6
97.5
ND
99.5
ND
ND
99.6
.96.7
ND
99.3
98.6
99.3
ND
99.3
ND
99.4
ND
100
99.1
ND
ND
99.4
ND
96.6
77.5
95.7
ND
ND
100
99.3
98.6
98.7
98.2
99.5
Percent
Plant D*
98.9
88.5
ND
96.0
99.1
. 99.8
98.7
91.2
ND
97.5
96.2
ND
ND
98.7
ND
99.0
ND
98.8
98.4
94.9
ND
ND
ND
' 94.5
92.3
92.3
ND
ND
ND
98.7
96.3
96.3
93.7
97.0
Plant A - 350 MW unit, Western coal, hot-side ESP (4); Plant B -
105 MW unit, Eastern coal (9); Plant C - 290 MW unit (10); Plant
D - 625 MW unit, cold-side ESP (11).
ND - No data.
22
-------
MC I'K'LCll' I U-V'OKS
OK' EMISSION CONTROL
ro
CO
HOT-SI DC
-->- COLD-SIDC
a: in0 -
u
Cd Cr Cu
Mn Hg Ni Se T1 U Zn
Ash As
Figure 4. Trace element emissions from two western coal-fired power plants (12).
-------
WESTERN CORL-FIRED POWER PLRNTS FIRING PULVERIZED CORL
ro
-p»
J
H-
Q.
Z
t-
cn
i Id2'
u
D
o
\
1 f\
^
EMISSION FRCTOR,PICOGRfih
i
\ o o
_ O i
1 t . 1
_t - .'4 J
r * *
xx
XX
xx
XX
XX
xx
xx
xx
XX
XX
XX
XX
xx
XX
xx
xx
x /
x>
xx
XX
XX
xx
XX
XX
xx
xx
XX
XX
xx
XX
XX
xx
XX
XX
xx
XX
xx
XX
xx
XX
XX
xx
"»« ""j - L-II'J - - -
HOT
(MIN)
COLD (MIN)
HOT
(MRX)
COLD (MflX)
f" "1
r" i"
F; i
2 i h.
X . J7x~,
^7x KZ pr?
xx xx p^xx^ PX>
n^pii
XX
xx
XX
xx
xx
xx
XX
xx
i
r * "
i§
I
77
XX
xx
xx
XX
xx
XX
xx
xx
xx
xx
XX
XX
xx
xx
Xx
!
1
xx
XX
X/
XX
xx
xx
XX
xx
xx
XX
'X
xx
XX
xx
XX
XX
XX
XX
xx
XX
XX
XX
I
r J
XX
-i
i !
1 t
I
I
T
xx
xx
XX
xx
XX
x>
XX
XX
XX
XX
Xx
^
7~7\
XX
XX
X /
x'x
xx
XX
XX
xx
XX
XX
Xx
"T
xx
xx
XX
xx
XX
Xx
X
fj
^x1
XX
xx
XX
xx
X>
xx
xx
xx
XX
xx
XX
XX
xx
xx
xx
XX
xx
xx
XX
XX
XX
Xx
As Ba Be Cd Co Cr Cu Hg
ELEMENT
N1
Sb
Se
Figure 5. Performance of hot- and cold-side electrostatic precipitators (4, 7, 12).
-------
CYCLONE BOILER FIRING ERSTTRN BITUMINOUS CORL
10'
ro
en
\-
\
i/i
en
iz
o
o
u
u
cr
o
CO
U
10
-1
!
1
'/
XX
/ s
1
v/.
//
f /
f /
//
/ /
/ /
f /
//
As
f /
f /
f /
f /
//
\
f /
' /
' /
//
v/
//
f /
/
/v
f /
//
XX
/x
Y
'
v/
Ba
Cd
Co
]
I
f /
' /
//
/ /
/ /
/ /
TfST I
TfSt 2
f/
f /
f /
//
/
/
/
x
'&
Cr
Cu N1
CLfMCNT
Pb
Sb
Se
1
XX
/
//
XX
XX
Figure 6. Performance of a cold-side ESP on a coal-fired power plant (13).
-------
In his
(PAH) emissions, Hangebrauck (14) reported data for emissions from a utility
boiler firing pulverized coal. Data for benzo(a)pyrene (BaP) emissions
before and after emission controls are presented in Table 5. BaP collection
efficiency ranged from 2-71%. Table 6 presents BaP emission data from six
coal -fired power plants of different design, using mechanical collectors
and/or ESP's. Note that nearly all controlled and uncontrolled BaP emis-
3
sions reported in Tables 5 and 6 exceed the MATE value of 20 ng/m . It
should be noted, however, that ESP's designed and built since the Hangefarauck
study exhibit more efficient parti cul ate removal, and thus are likely to
collect more PAH emissions than other units.
Primary Sulfate Emissions-
Data on the effect of electrostatic precipitation on primary sulfate
emissions (PSE) of sulfur trioxide/sulfuric acid and metallic sul fates are
presented in Table 7 (15, 16, 17). Two points are notable:
Percent reduction in particulate sulfates (MSCty) and total
PSE- is generally higher for coal-fired units than for oil-
fired units. This is most likely due to the smaller mass
median diameter of oil ~fly ash.
, . Percent reduction in SOa/HzSCty is less than for MS04 since
the former are present in flue gas primarily as gases and
liquid droplets. Much df the reduction in acid aerosol is
probably due to removal of H.2S04 adsorbed on parti cul ates.
Increases in SQj/HgSC^ emissions from ESP-controlled systems have been
noted in the literature, and often are accompanied by unexpected decreases
in S02 across the ESP. This observation was noted by McCurley and DeAngelis
(17) in reporting PSE data on a coal-fired industrial source equipped with
a hot-side ESP. They suggest two potential conversion mechanisms of SOg to
S03 based on the input of energy from the ESP to the combustion gases via
the corona discharge (electrical arcing across the ESP electrodes).
« Arcing in a precipitator may cause localized "hot spots" in
which the conversion of S02 to $03, and/or S04 would occur
quite rapidly, as the temperature is a dominant rate control-
ling factor. Since the gases are already hot in comparison
to those encountered in an ESP in a cold side configuration,
it is plausible that this additional heat input could cause
the observed results.
26
-------
r\>
f
t
TABLE 5. CONTROLLED AND UNCONTROLLED BENZO(a)PYRENE EMISSIONS
FROM A PULVERIZED COAL-FIRED POWER PLANT*
Test
Number
1
2
2
3
4t
4t
5t
5t
Sampling
Point*
B
B
A
A
B
A
B
A
M9/S
21 i
9.4
8.1
8.4
18
17
130
39
BaP Emission
3**
ng/Nm
no
50
42
42
130
120
930
270
Rates
ng/kg coal
1.3
0.57
0.51
0.49
1.5
1.4
12
3.3
BaP Emission Factor
pg/J heat input
i
0.046 r
0.021 j
0.018 1
t
j
0.021
0.053
0.052
0.42 1
0.12 1
* Data are from Reference (14) for a vertically-fired dry bottom furnace.
t Test numbers 4 and 5 were conducted at 75% load, as opposed to Tests 1,2 and 3, which were
conducted at full load. .
* B = sampling point before ash collector; A = sampling point after multiple cyclone and ESP.
** Nanograms BaP per cubic meter of flue gas at 21°C, 1 atmosphere. .
-------
TABLE 6. CONTROLLED BENZO(a)PYRENE EMISSIONS FROM COAL-FIRED POWER PLANTS (14)
Unit Number*
1
1
1**
1**
Control
Methodt
MC
MC
MC
MC
+ ESP
+ ESP
+ ESP
* ESP
Fuel Rate
kg/s
. 16.4
16.9
12.1
11.6
ug/s
8.1
8.4
17
39
BaP Emission Rates
3* !
ng/Nm gg/kg coal
42
42
120
270
0.51
0.49
1.4
3.3
BaP Emission Factor
pg/J heat input
0.018
0.018
0.052
0.12
ESP
ESP
13.1
12.5
6.2
7.8
39
48
0.46
0.62
0.016
0.020
MC + ESP
14.3
53
320
3.7
0.13
ro
CO
4tt
4
4tt
MC
MC
MC
2.2
2.7
2.3
9.0
1.7
1.4
330
57
46
4.2
0.64
0.62
0.13
0.021
0.020
ESP
ESP
14.9
16.8
170
39
730
170
11
2.4
0.35
0.072
6**
6**
6**
MC
MC
MC
2.3 . 1.6 ,
2.3 <1.0
2.3 <1.0
58
<36
<33
0.68
<0.42
<0.42
0.023
<0.014
<0.014
t
*
Units 1,2,3 and 4 fired pulverized coal; units 5 and 6 fired crushed coal. Unit 1 = vertically-fired,
dry bottom; Unit 2 = front wall-fired, dry-bottom; Unit 3 » tangentially-fired, dry bottom; Unit 4 =
opposed-downward-inclined burners, wet bottom; Unit 5 eyelone-fired, wet bottom; Unit 6 = spreader
stoker, travelling grate. All units operated at 100X load unless otherwise noted.
MC multiple cyclone; ESP » electrostatic preclpitator
Nanograms BaP per cubic meter of flue gas at 21°C, 1 atmosphere.
** 75* load
tt Fly ash reclrculated.
-------
TABLE 7. EFFECT OF ELECTROSTATIC PRECIPITATOR (ESP) ON PSE
FROM COMBUSTION SOURCES
Percent Reduction 1n Emissions
System S03/H2S^4 RsT^ Total PSE Reference
Oil-fired utility. Cold
side ESP. MgO addition and 3 61 46 15
recirculation of ash from
ESP.
Coal -fired utility. - 58 16
Coal-fired utility. Cold on QQ -n
side ESP. 20 " ' 17
Corona discharges also have been shown to produce ozone (03)
which could-readily react with S02 to yield SOs and 02- This
second mechanism has been postulated previously to explain
the apparent conversion of N2 to NO in an ESP.
Fabric Filters
Technology Assessment-
The greatest advantage of fabric filters (baghouses) is their ability
to efficiently remove submicron particles regardless of changes in coal ash
properties. Disadvantages lie in the temperature and chemical limitations
of fabric materials. Pressure drop is usually higher than for ESP's, but
energy consumption is about equivalent to an ESP operating at high efficiency
on high resistivity ash.
Bradway and Cass (18) reported on extensive studies of a baghouse
installation on a small coal-fired utility. In 22 tests (half at normal
and half at abnormal operating conditions), the average total particulate
mass collection efficiency of the baghouse was 99.84% at an outlet loading
q
of 0.0039 g/m (0.0017 gr/dscf). Collection efficiency for submicron
particles was 99% at 0.1 pro, 99% at 0.5 ym, and above 99% at 1.0 pro. No
statistical difference in performance was found between normal and abnormal
operating conditions.
29
-------
Trace Element Emissions
Yeh et al (19) reported that a baghouse was superior to an ESP in re-
ducing toxic trace element emissions from a coal-fired combustion source.
Under average conditions, the percent of trace elements entering in the coal
which was not collected in the boiler (as bottom ash) or the baghouse was 9*
for As, 23% for Be, 45% for Cd, and 37% for Pb. Under optimum conditions,
0-18% of the Hg and 9-13% of the Se in the entering coal were emitted. Ensor
(20) reported that greater than 99.9% removal of Al, Ca, Cu, Fe, Pb, Ni, Si,
U, and Zn were achieved by a baghouse installed in a coal-fired utility. Trace
i
element removal data from an EPRI study (RP 1130-1) have not yet been published.
FLUE GAS DESULFURIZATION SYSTEMS
No comprehensive studies of non-criteria pollutant emissions from FGD
systercsinsta-l-led-on-utll-ity-boilers have been published in the literature.
Data presented in this section and in Appendix C are derived from studies
performed under the Emissions Assessment of Conventional Combustion Systems
(EACCS) program.
Leavitt et al (21) studied the performance of a limestone slurry SO-
scrubber installed on a bituminous coal-fired cyclone boiler at a utility
in Kansas. The unit was rated at 874 MW (gross) electrical output. The
fuel fired was low grade subbituminous coal of local origin, containing " * '
approximately 25% ash and 52-6% sulfur with a heating value of 20.9 to 22.6
GJ/kg. The FGD unit was a two-stage venturi-absorber, designed for simul-
taneous S02 and particulate removal.
Trace element collection efficiencies for the FGD unit during a single
test are presented in Table 8. An overall removal efficiency of 94% was
obtained for these trace elements. Figure 7 shows the relationship between
controlled and uncontrolled trace metal emission factors.
Organic emissions were also characterized before and after the FGD
unit. Removal efficiency averaged 14% for C,-Cg organics, 72% for Cy-C-ig
organics, and 63% for >C,g organics. Specific organic compounds identified
at the scrubber inlet included aliphatic hydrocarbons, substituted benzenes,
ethylbenzaldehyde, dimethy!benzaldehyde, 2,6-pereriden-dione-4-one, 2,6-
dimethyl-2,5-heptadion-4-one, and the methyl ester of a long chain acid.
These organics were present at levels of 0.2 to 20 ug/m . With the exception
30
-------
TABLE 8. TRACE ELEMENT COLLECTION EFFICIENCY EXHIBITED
BY FGD SYSTEM - COAL-FIRED POWER PLANT*
Trace Element Removal Efficiency, %
Al
As
Be.
Ca
Cd
Co
Cr
Cu
Fe
Hg
Mn
N1
Pb
Sb
Se
Sr
V
Zn
97.7
4.08
91.4
95.9
88.6
93.2
90.8 \
84.2
96.8
94.0
78.6
97.3
73.6
65.4
76.2
91.7
89.4
80.0
Data from Leavitt (21), for a two stage venturi-
absorber S02 scrubber installed on a cyclone boiler
firing Eastern bituminous coal and rated at 874 MW
(gross).
31
-------
ro
CORL-r rR[;f> FOWCR H RNT r fR !UO >-;!G)i-<;,Ul ! i.!f-. [:RjT['fvJ fit TUMINOUf, CORL
10'' -
1
-}
C:~
~r *
i-4
a1 'd"
i r
1'
u-
7j
o
~>
\ ' 0 "
V
a:
a
u
u
C.
%
L<
O
a:
L.
b 'C'
1- 4
LI
LO
»- 4
v "
LJ
f r/-. » -\
/" r , ' ' » I./ *.. i J
/ : : ' W! "!: ' C."|/5
^ r/i ^
£ ^ ^.-
' / */ ; " v/
^ 1 ^ P : 1
x ?x fyl yi ry
/
/
/
''/
\
'/
/
/,
/
y
y
/
/
n K
;' ^
' 4 K y
' - rTT*"1 ^
^ L
i||
' < X
'' i X
' J /
' ^
;j x;
X J '-/^
'A S/
' V,
\ i
' ?
;j ^
'« X
J .. -, /
/ f y I ^r
1 '/^ ^
^ 1 *-^ ^x
] ^j //
A
'] y fy y
;i 1
1
'l f
^
t /,
j y,
' < X
* i ^.
- j /
-;; ^
** j ./^
''-1 ^
. i/*/ ₯}*/''.'
% t $ JO
' */ X > X y. »-y» v,
' ' I"T/I X* r / ' ' X* " /^ X
'J f/1 '^ ^^ ^ ^ t^
'j Y Y'} 7' V)( Yf\ YX 7
'\ */ s/'. 6r ''/ //'. S/-T* s/
' J ^ y« /s Y *' f \ /**' */''** */
* Xy* */ ' */ f * *' /
' A */ * s '' */ y * f
''<'' xT' yi' y''' '' x'' ^' y^
'i x ' x- xt-' />.., "/ X' x
' x * x' *" ^ ' _/T ' * s ' * / f *
''A '/'A '/'. ry/i r/l'-i r/'.'-l v ' iv
;i ^
/ *
',\ W V/ "/
:! L 0 0 ^ ^
' ! X -N X/ (-y // . X
^ y-i yj ix "/I y
/' y- K/J yi 'As
' V - !>4r. iyj '/., V
/ y'/ x''1 yj y -^ v
/ / - y '- ' yi y ' v
' .v . y .-. VL. y .' v
. y. V/. Vx] y-' V
' y ' S^ -: i ^ ' ''' ' x^
' 7 ' ~ / ' i 7 ' 7 ''> x/
X- X/j X- X'< X
;] ^; x,;j xj; ^,;| y:
/> ' //' \ Sjf //''( ^/
k
»
j
t
'
'
'
f
/
/_
Al As Be* Ca Cd Co Cr Cu Hg Mn Ni Pb Sb Se Sr V Zn
f :>'/>' r
l.t « "U. « 1
*
Actual emission measured at 0.67 pg/J.
Figure 7. Effect of FGD system on trace element emissions.
-------
of ethylbenzaldehyde, substituted benzenes, and aliphatic hydrocarbons,
none of the other orgam'cs were identified at the scrubber outlet. Several
polycyclic organics were also identified at the scrubber inlet, at concen-
trations ranging from <0.03 to 9 yg/m (several orders of magnitude less
than their MATE values).* They included naphthalene, substituted naphtha-
lenes, biphenyl, and substituted biphenyls. None of these compounds of POM
were identified at the scrubber outlet.
Primary sulfate emissions were substantially reduced by the FGD unit.
Removal efficiencies, for S03 and S0^~ are shown for two tests in Table 9.
Fluoride emissions were also measured at the scrubber inlet and
outlet. Removal efficiency for fluoride was greater than 62-84*, as shown
in Table 10.
Other studies have focused specifically on the ability of wet SOg
scrubbers to remove primary sulfates. Homolya and Cheney (22) reported
results of an extended series of measurements at a coal-fired utility boiler,
equipped with~a-wat-limestone FGD unit. Concurrent inlet and outlet samples
were taken across the FGD and analyzed for total sulfate. In 10 of 32
measurements, outlet sulfate concentrations exceeded those at the inlet,
probably due to scrubber liquor entrainment. Of the remaining 22 measure-
ments, sulfate removal ranged from 5% to'56%, and averaged 25%. About 85%
of the emitted sulfate consisted of free H2S04> which was in the gas phase
prior to scrubbing, but was converted to aerosol by penetrating the slurry
and demisters (both maintained below the acid dewpoint). In another study
at.a coal-fired power plant, Delumyea and Zee (23) tested the performance of
an FGD system consisting of turbulent contact absorber (TCA) modules utilizing
thiosorbic lime, installed downstream of a cold-side ESP. They found in a
series of ten measurements that the sulfate removal efficiency of the TCA
ranged from 27% to 52%, with an average removal efficiency of 37%. In one
case (not included in calculating average removal efficiency) outlet sulfate
concentration exceeded inlet sulfate concentration.
k O
Total concentration of POM measured = 18.7 vg/m . Detection limit was about
0.03
33
-------
TABLE 9. REDUCTION OF S03 AMD S04= EMISSIONS
BY F6D SYSTEM DURING COAL FIRING (21)
Sulfur
Compound
<
so/t
Emission
Inlet
99.4
53.7
15.5
18.2
Factor, nq/J
Outlet
17.2
1.9
5.1
2.4
Removal
Efficiency,
Percent
82.7
96.5
67.1
86.8
Mole % of Total
Sulfur Species
in Flue Gas
2.1
1.7
0.3
0.5
Level 2.
f Level 1 .
TABLE 10. REDUCTION OF FLUORIDE EMISSIONS BY FGD SYSTEM
DURING COAL FIRING (21)*
Fluoride Emission Factor. ng/J
Inlet'Outlet
Removal Efficiency,
0.37
0.9
<0.14
<0.14
>62
>84
Level 1.
34
-------
REFERENCES
1. 8th Annual Review of Overall Reliability and Adequacy of the North
American Bulk Power Systems. National Electric Reliability Council.
Princeton, N.J., August 1978.
2. Shih, C. et al. Emissions Assessment of Conventional Stationary
Combustion Systems. Volume III Electricity Generation External
Combustion Sources. EPA-600/7-79-029d, TRW Incorporated, October
1979. (draft)
3. Compilation of Air Pollution Emission Factors. Third Edition,
including Supplements 1-7. U.S. Environmental Protection Agency,
Publication No. AP-42, August 1977.
4. Radian Corporation. Coal-Fired Power Plant Trace Element Study,
A Three Station Comparison. Volumes I-IV. PB-257 293, Austin,
Texas, 1975,
5. Cowherd,-.C.. Jr., et al. Hazardous Emission Characterization of
Utility Boilers. Midwest Research Institute, Kansas City, Missouri,
1975.
,6. Edwards, L.O., et al. Trace Metals and Stationary Conventional
Combustion Sources. Radian Corporation, Austin, Texas, March 1979.
(draft).
7. Ondoy, J.M., R.C. Ragaini, and A.H. Biermann. Comparisons of
Particulate Emissions from a Wet Scrubber and Electrostatic
Precipitator at a Coal-Fired Power Plant. American Chemical
Society Division of Environmental Chemistry Preprint, 16(2):200-203,
- 1976.
8. Cleland, J.G. and G.L. Kingsbury. Multimedia Environmental Goals
for Environmental Assessment. Volumes I and II plus September 1979
update. EPA-600/7-77-136, U.S. Environmental Protection Agency,
Research Triangle Park, N.C., November 1977.
9. Lee, R.E. Jr., et al. Concentration and Size of Trace Metal Emissions
From a Power Plant, a Steel Plant, and a Cotton Gin. Environmental
Science and Technology, 9(7):643-647, 1975.
10. Lyon, W.S. and J.F. Emery. Neutron Activation Analysis Applied to
the Study of Elements Entering and Leaving a Coal-Fired Steam Plant.
International Journal of Environmental Analytical Chemistry,
4(2):125-133, 1975.
35
-------
11. Ondov, J.M., et al. Elemental Emissions from Coal-Fired Power Plant:
Comparison of a Venturi Wet Scrubber System with a Cold-Side Electro-
static Precipitator. UCRL-80110, Lawrence Livermore Laboratory,
Livermore, California, 1977.
12. Mann, R.M., et al. Trace Element Study of Fly Ash Emissions from Two
Coal-Fired Steam Plants Equipped with Hot-Side and. Cold-Side Electro-
static Precipitators for Particulate Removal. EPA-908/4-78-008,
Radian Corporation, Austin, Texas, 1978.
13. Lyon, W.S. Trace Element Measurements at the Coal-Fired Steam Plant.
CRC Press, Cleveland, Ohio, 1977.
14. Hangebrauck, R.P., D.J. von Lehmden, and J.E. Meeker. Sources of
Polynuclear Hydrocarbons in the Atmosphere. Public Health Service,
U.S. Department of Health, Education, and Welfare, Cincinnati, Ohio,
1967. 44 pp.
15.- Dietz, R.-N.,-R.F.--Wieser, and L. Newman. Operating Parameters Affect-
ing Sulfate Emissions from an "Oil-Fired Power Unit. In Workshop
Proceedings on Primary Sulfate Emissions from Combustion Sources,
Volume II. EPA-600/9-78-020b, U.S. Environmental Protection Agency,
Research Triangle Park, N.C., August 1978. pp. 239-270.
16. Homolya, J.6., H.M. Barnes, and C.R. Fortune. A Characterization of
the Gaseous Sulfur Emissions from Coal- and Oil-Fired Boilers. In:
Proceedings of the 4th National Conference on Energy and the Environ-
ment. Cincinnati, Ohio, October 1976.
17. McCurley, W.R. and D.G. De Angelis. Measurements of Sulfur Oxides
from Coal-Fired Utility and Industrial Boilers. In: Workshop Pro-
ceedings on Primary Sulfate Emissions from Combustion Sources,
Volume II. EPA-600/9-78-020b, U.S. Environmental Protection Agency,
Research Triangle Park, N.C., August 1978. pp. 67-85.
18. Bradway, R.M. and'R.W. Cass. Fractional Efficiency of a Utility
Boiler Baghouse, Nucla Generating Plant. PB-246 641, GCA Corpora-
tion, Bedford, Massachusettes, August 1975. 148 pp.
19. Yeh, J.T., et-al. Removal of Toxic Trace Elements from Coal Combus-
tion Effluent Gas. PERC/R1-76/5, U.S. Energy Research and Develop-
ment Administration, Pittsburgh, Pennsylvania, 1976. 21 pp.
20. Ensor, D.S., R.G. Hooper, and R.W. Scheck. Determination of the
Fractional Efficiency, Opacity Characteristics, Engineering and
Economic Aspects of a Fabric Filter Operating on a Utility Boiler.
EPRI/FP-297, PB-261 472, Metrology Research Incorporated, Altadena,
California, 1976. 206 pp.
21. Leavitt, C., Et al. Environmental Assessment of Coal- and Oil-Firing
in Controlled Utility Boilers. TRW Incorporated, Redondo Beach,
California, March 1979 (draft).
36
-------
22. Homolya, J.B. and J.L. Cheney. A Study of Primary Sulfate Emissions
from a Coal-fired Boiler with FGD. Journal of the Air Pollution Control
Association, 29(9): 1000-1004, September 1979.
23. Delumyea, R.D. and C.A. Zee. CCEASulfates Sampling and Analysis on
Utility FGD Unit. Prepared by TRW, Inc. under EPA Contract No. 68-02-
2613, Task 25, for U.S. Environmental Protection Agency, Research
Triangle Park, N.C. April 1980.
24. Tyndall, M.F., et al. Environmental Assessment for Residual Oil
Utilization - Second Annual Report. EPA-600/7-78-175, Catalytic
Incorporated, September 1978.
25. Klein, D.H., et al. Pathway of Thirty-Seven Trace Elements Through
Coal-Fired Power, PI ant. Environmental Science and Technology,
9(10): 973-979, 1975.
26. Schrag, M.P., editor. Fine Particle Emissions Information System
Users Guide. EPA-600/8-78-006, Midwest Research Institute, Kansas
City, Missouri, June 1978.
27. Shannon, L.O. and P.G. Gordon. Particulate Pollutant System Study.
Volume II - Fine Particulate Emissions. Midwest Research Institute,
Kansas City, Missouri, August 1971.
28. Weast, T.E., et al. Fine Particulate Emission Inventory and Control
Survey. EPA-450/3-74-040, PB-234 156, Midwest Research Institute,
Kansas City, Missouri, January 1974.
29. Cato, G.A., L.J. Muz to t, and R.E. Hall. Influence of Combustion
Modifications on Pollutant Emissions from Industrial Boilers. In:.
Proceedings of the Stationary Source Combustion Symposium, Volume
III. EPA-600/2-76-152c, U.S. Environmental Protection Agency,
June 1976. ,
30. Crawford, A.R., et al. The Effects of Combustion Modification on
Pollutants and Equipment Performance of Power Generation Equipment.
In: Proceedings of the Stationary Source Combustion Symposium,
Volume III. EPA-600/2-76-152c, June 1976.
31. Bolton, N.E., et al. Trace Element Measurements At the Coal-Fired
Allen Steam Plant. NSF-EP-43, Oak Ridge National Laboratory, Oak
Ridge, Tennessee, March 1973.
32. Kaakinen, J.W., et al. Trace Element Behavior in Coal-Fired Power
Plants. Environmental Science and Technology, 9(9):862-869, 1975.
33. Hillenbrand, L.J., R.B. Engdahl, and R.E. Barrett.. Chemical
Composition of Particulate Air Pollutants from Fossil-Fuel Combustion
Sources. "EPA-R2-73-216. Battelle Columbus Laboratories, Columbus,
Ohio, March 1973.
37
-------
34. Gorden, G.E., et al. Study of the Emissions from Major Air Pollution
Sources and Their Atmospheric Interactions. University of Maryland,
College Park, Maryland, Report for the period November 1, 1972 to
October 31, 1974.
35. Curtis* K.E. Trace Element Emissions from the Coal-Fired Generating
Stations of Ontario Hydro. Report No. 77-156-K, Ontario Hydro
Research Division, April 1977.
36. Ragaini, R.C. and J.M. Ondov. Trace Contaminants from Coal-Fired
Power Plants. UCRL-76794, Lawrence Livermore Laboratory, Liver-more,
California, September 1975. -
37. Ray, S.S. and F.G. Parker. Characterization of Ash from Coal-Fired
Power Plants. PB-265 374, Tennessee Valley Authority Power Research
Staff, Chattanooga, Tennessee, January 1977.
38. -Oglesby, S. Jr.-and-D. Teixeira. A Survey of Technical Information
Related to Fine-Particle Control. RP 259, Southern Research Insti-
tute, April 1975.
39. Leavitt, C., et al. Environmental Assessment of Coal- and Oil-Firing
in a Control-led..Indus trial Boiler, Volume III. EPA-600/7-78-164c,
TRW Incorporated, Redondo Beach, California, August 1978.
40. Drehmel, D.C. Fine Particle Control Technology: Conventional and
Novel Devices. Journal of the Air Pollution Control Association,
27{2):138-140, 1977.
41. Abbott, J.H. and D.L. Harmon. Concepts in Fine Particulate Control
in the Metallurgical Industry. U.S. Environmental Protection Agency,
Research Triangle Park, N.C. May 1979. 34 pp.
42. Knight, J.H. The Use of Nahcolite for Removal of Sulfur Dioxide
and Nitrogen Oxides from Flue Gas. Superior Oil Company Oil Shale
Department, October 1977.
43. Hangebrauck, R.P., D.A. Denny, and W.G. Tucker: Nomenclature for
Environmental Assessment Projects. U.S. Environmental Protection
Agency, Research Triangle Park, N.C., August 1979. 7 pp.
38
-------
APPENDIX A
BACKGROUND DATA
In this Appendix, data will be presented which are essential to a
detailed examination of non-criteria pollutant emissions from utility
boilers. Particular attention is paid to trace elements in coal as a
background for emission estimates.
ENERGY CONSUMPTION
Both current and future fuel consumption data are available from
estimates provided by the National Electric Reliability Council (NERC).
For 1978,'NERC (1) estimated total fuel consumption of 120.4 Tg of western
bituminous coal (2,698 PJ), 316.6 Tg of eastern bituminous and anthracite
coal (8,280 PJ), 31.1 Tg of lignite coal (477 PJ), 94.1 x 106 m3 of resi-
dual oil* (3,830 PJ),.and 62.9 x 109 m3 of natural gas (2,399 PJ) for
electricity generation external combustion sources.
According to NERC, coal combustion for electric power generation
consumed 11,455 PJ of energy in 1978. By far, the largest portion of this
was bituminous coal, which accounted for 95.6% of the total. Anthracite
and lignite accounted for the remaining 0.27% and 4.16%, respectively. In
the bituminous category, consumption was dominated by pulverized dry-bottom
boilers, which consumed 8,370 PJ of the total 10,949 PJ of bituminous coal
fired. Pulverized wet bottom boilers and cyclone boilers consumed 1,266 PJ
and 1,217 PJ, respectively. The remaining 94 PJ was consumed in stoker-
fired units.
CHARACTERISTICS OF COAL AND OIL
In order to estimate emissions from coal firing, the fuel must be
fully characterized. Essential information includes heating value (gross),
sulfur and ash content (as-fired), and trace element constituents.
Includes 2.9% distillate oil.
39
-------
Emission factor estimates in this report will be based on a heating
value for bituminous coal of 25,586 J/g (11,000 Btu/lb). National average
sulfur and ash contents used were 1.92% sulfur and 14.09% ash.
Table A-l is a summary of the trace element characteristics of various
U.S. coals by region. A large number of references contain coal trace
element data. Notable among these is the computerized National Coal Resources
Data System, which provides published coal analyses from both the U.S.
Geological Survey (USGS) and U.S. Bureau of Mines. Average trace element
concentrations are given in Table A-l for eastern bituminous coal, western
bituminous coal, bituminous coal (combined eastern and western bituminous
coal, in proportion to their consumption by electric utilities), North Dakota '
lignite, Texas lignite, and anthracite. The average values for eastern
bituminous coal-,-western bituminous coal, and North Dakota lignite are
weighted averages in accord with annual coal production by county. The
average values for Texas lignite and anthracite are averages of the trace
element data provided by USGS and unweighted by county.
The data base on the trace element content of residual oil is limited.
The most comprehensive data base is the one developed by Tyndall et al (24),
for which a composite oil analysis based on a weighted average of U.S.
crudes (domestic and imported) was used to characterize the trace element
content of residual oil feedstock. Average trace element concentrations of
residual oil from this data base are presented in Table A-2. Variations
in these average concentrations are not known. An average heating value of
43,760 J/g (146,000 Btu/gal) is assumed for residual oil.
BEHAVIOR OF TRACE METALS DURING COMBUSTION
Emissions of trace elements from oil-fired utility boilers can generally
be computed by assuming that all trace elements present in the oil are
emitted through.the stack . Based on this assumption and the data in Table
A-2, emissions of Ni, V, Be, Pb, Co, Cu, and P from oil combustion are of
-environmental concern.
Using this assumption an estimated emission factor can be computed for
each trace element in Table A-2. The estimated emission factor EFj for
element i with concentration C-j in oil is EF-j (ng/J) = 22.86 x C^ (ppm).
40
-------
TABLE A-1. AVERAGE TRACE ELEMENT CONCENTRATIONS IN COAL
of tk«
tl1.tr <»«>
«l>nU«(«l|
toll (M
oron (B)
Itrlw (M)
I«rr1lluri (8t)
81
Crtlut (tr)
furoplw ([»)
flwrtdt (F)
Inn (Ft)
Cjllltfi (&)
bdolfnlw (Cd)
fonwnlvl (£«)
Hifrlw (HI)
Htrcur, (Kg)
HoUltw (Ho)
lodlnt (1)
India. (1*1
IrldltM (I')
fount* (()
Unlk.nlui (1.)
Itlklui (H)
Lgttlli* (Iv)
0.14
11.02]
11.4
O.I
41.9
87.7
1.00
I.9S
10.1
1.476
0.42
16.2
1.064
8 79
11.7
2.34
1S.2
1 4j
O.SS
0.51
01 1
10.621
4,21
1.44
4.2*
1.09
0.21
0.20
I.4S
0.20
0.20
1,662
10.1
11.2
0.14
Hf»». torn
.00
267
1.8
.10
1.2
6.9
.0?
.00
i.i
1S2
.01
1.00
92
.17
7.0
.08
1.0
.04
.11
.01
7.4
1?)
.31
.41
.17
.0?
.01
.OS
.74
.05
.70
109
.ss
!oi
»*
17
M
69
4
75
64
78
1)
29
1*
18
24
10
77
80
2)
19
21
f
74
S7
)?
69
10
68
2)
45
9
19
II
JJ
64
49
70
Meiltr
TtlT '
W»
0.18
15.916
1.44
0.10
4S.Q
IDS
0.86
0.42
4.87
l».7*2
1.28
20.2
294
4.14
11.8
0.74
14.8
1.18
0.51
f.99
141
8.858
1.81
0.15
7.47
0.82
0.099
0.18
0.64
0.17
0.70
1,111
7.06
IS.O
0.06*
Dr«l*tlo»
of Ikt
Ht«n, pp«
.12
48S
.71
Mt
1.1
27
.17
.00
.76
1.787
.01
7.9
19
.48
.78
.17
!o?
.00
.71
4.1
266
.21
.00
.14
.11
.00
.00
.00
.01
M
47
.4?
7.1
.00
?.
'Tei«'-
Orvl'tto* *
ol tkt
Horth fahou llyiUt '_ 1t»«> llonltt . _ hi
"(»'«" ~" Jti'ndVrJ ' 'Kin ~5t«ndjr3 'STe'"
tl" Oevlttlon » pp> 0»UIIon * cp»
Ol tkt Of tkt
: Hri*. (*
19
29
41
2
SB
40
SO
4
18
29
2?
17
18
48
SO
11
S7
14
4
1*
M
79
48
'4
42
14
M
"4
14
10
2
28
41
2*
12
O.S8
12.170
8.62
0.10
S6.S
148
0.46
LSI
8 61
7.101
0.66
18.?
6S2
7.S7
2S.9
1.9?
15.1
1.4S
O.S4
7 79
107
10.1)7
4.10
I.2S
1 7*
lioi
0.18
0.20
1.2)
0.18
0.20
1.510
9.28
?».?
0.17
.00
264
1.8
.00
t.i
8.9
.07
.00
I.I
ISI
.01
.99
91
.17
7.0
.08
1.0
.08
.10
.01
?.*
171
.11
.47
.1?
.06
.01
.OS
.78
.05
.00
108
.SS
9.S
.01
S6
6)
12$
8
I7S
104
126
17
47
61
45
41
48
120
1)0
17
111
IS
11
4)
86
6*
111
14
110
17
75
11
1)
21
8
61
10S
7?
12
0.049
4.416
S.24
0.06?
64.7
soi
0.11
0.40
0.77
11.707
0.35
1).)
63.S
1.14
7.S?
1.7)
11.1
O.SO
O.U
O.I?
77.$
4.749
1.8?
0.11
0.7*
0.74
0.044
0.14
0.36
0.072
0.062
MS
1.74
l.M
0.049
Hem. pp>
0.01
S??
0.19
0.00
10
8*
0.08
0.7)
0.4?
i./m
0.18
S.6
)?
0.09
1.7
0.8*
0.04
0.06
0.0)
o.os
*.4
S66
0.14
0.11
0.07
0.1?
0.0?
o.o?
0.19
0.00
0.00
171
0.84
7.2
0.01
10 ' 0.18
9 M.I36
7 ' 1.0
7 ' - '
10 714
10 124
10 1.14
1 " ,
7
1
1
1
1
9.447
0.26
.44
<240
7.9
20.4 .
..
24. S
..
..
4*. 9 ,
4.140'
Ml
3.54
».
Q.22
..
-.
.»
7
« 512
10 20.9
1 11.7
1
**** ppo
0.01
BU
o.i
..
U
i)
0.11
.*
S20
0.01
»
o.s
I.S
1.8
..
*
^ }
751
O.U
.
0.95
.«
0.01
.
.«
.,
,.
H
1.8
1.7
4
2?
24
^.
29
2?
?»
._
2?
2*
24
24
74
79
29
,. '
.
18
21
I*
1)
24
..
.«
..
..
2?
28
24
O.I)
20.172
1.65
-0.87
8.84
M.6
1.)?
0.15
..
1.04?
0.19
46.9
404
9.11
IS.*
10'*
2 02
1.21
0.88
81. S
4,161
S.5»
I.I?
1.49
102
O.U
O.M
<0.4I
-0.lt
2.118
20.*
20.2
1.19
te«l
SI
2
S)
19
3)
SO
5)
..
S)
19
22
12
S)
S)
19
2?
so
S)
8
,.,
50
50
5)
4«
5)
50
. ConU.unJ .
tl IW m«t*r of 5tt> of o>U. A i»t ol i)«tf wjr r«prfl m «nr«o* ol * muter of 4«U poloM er to*rll*ft t tlaglr
feptndln) on tkt rtftrtnce sourct.
K «ol (O*patt4 bKiutt tht only 1«t« >tlltb1t tre til for to4l fn» tfc* \UM count/.
point.'
-------
TABLE A-1. (Continued)
IN)
Tract llwwt
tkvMilm )
SdllllM (M)
ftlobli* (M>)
MM4/»ti» (M)
llcktl (il)
Osilw (Oil
tofphgriis {)
tlld (P»)
ftllldlu* (N)
rVtltodjvlMi IrY)
mtUM in)
tubltflw (U)
Ihtaluo IIU)
RhotfiM in.)
tutktnluc (!)
Antimony (So)
Sctmflw (Scj
Sflfftluv (S*)
Slllto* (SI)
Smrlw (])
Tin (>)
KrtMtlM (Sr)
T«i>t«lm (I«)
TtrbluB lib)
Itllurlui (1«|
Tnarlui (Tt)
Tltnilw (fll
IhjIIlK (Tl)
Urinlw |u)
VtAAdlW (V)
Iwwttn («)
Utrlw (I)
Ttttrtl* (Tb)
Hoc (i.)
ZlrCOAlM (If)
i«Ur
«
405
i!7I
470
S.7)
S.12
11.1
t.t
It!
4.41
0.1
2.11
0.1
10.4
0.2
0.1
0.1
1.47
S.I4
1.01
11.800
I.M
1.11
77.*
0.10
0.11
Q.tl
1.45
lit
0.17
O.It
l.ll
17.1
O.M
o!st
3ft 9
S1.I
Orrljtlon
ol tto
KM», op*
10
1.1
.il
IS
1.4
t.t
l.t
.10
M
.41
.10
.00.
Il'
.20
.10
.10
.07
.45
.12
4*2
.10
.44
1.4
.01
.Ot
.01
.to
27
.00
.00
.11
4.0
.14
.11
.02
1.S
t.t
t tfcittr
* ft*
IS 1.474
77 St.*
S* 4.01
M I.IOS
17 4.10
M 11.*
17 14.8
i O.t
41 til
71 .17
I .021
10 .8*
i .1
it .41
1 .2
1 .1
i .1
M .SI
II .»0
41 .54
IS 1 ,1S8
24 .40
K .71
11 1S1
24 .54
22 .41
10 .081
14 .40
40 4*8
10 .41
t .011
ts .si
71 t .7
tl .11
4* .50
11 .12
7* 34.8
it 47.1
tl tte
Ntm. pp>
tl
4.1
.7*
20
.IS
.00
.IB
K«
.11
K
.00
«c
.4*
8C
K
K
.0)
!ot
S22
.M
.45
1)
.02
.04
.00
.M
.02
.00
.55
1.1
.01
.54
.01
t.t
7.1
t BUtMtnout North tfckott limit* l*«l« U»ilU tothrleltt
Htu SttnArtf Net* Stmdiri) HTM {UnJtrJ Httn Sttiijtrd
II pea Br.l.tlon H ft* OnUllon rfm Or>l«tloi> 8 pp> OoUtlon
ol th* i of th* of Ik* Of the
*». M" ! «». PP» !». »» IW». pp>
tl r 1.120 2* 62 2.211 54 ( 2.041 Ili 27 751 M SI
47 M.8 1.1 124 44.7 17 10 111 If 27 71.1 7.) i)
4S S.7I .it 104 t.it 0.02 2.11 8.2* 27 7.40 8.12 il
28 I4S 14 14 1.411 407 Ml 18 27 Mi 108 S)
IS S.3J 1,4 ft .SB 0.07 157 0 It 24 1.4S 0.11 47
10 . 7.17 t.t it .01 .0 tl.t S 1 4 11.4 l.{ 20
SI '17.* 1.2 1)0 .to .«f 1 11.* 87 2* 17.1 1.4 SI
1 . 0.2 .09 t .062 .00 - -
-------
TABLE A-2. AVERAGE TRACE ELEMENT CONCENTRATIONS OF RESIDUAL OIL
U>
Trace
Element
Vanadium
Nickel
Potassium
Sodium
Iron
Silicon
Calcium
Magnesium
Chlorine
Tin
Aluminum
Lead
Copper
Cadmium
Cobalt
Rubidium
Titanium
Manganese
Chromium
Barium
Zinc
Phosphorus
Molybdenum
Arsenic
Selenium
Uranium
Antimony
Boron
Concentration,
ppm
160
42.2
34
31
18
17.5
14
13
12
6.2
3.8
3.5
2.8
2.27
2.21
2
1.8
1.33
1.3
1.26
1.26
1.1 ,
0.90
0.8
0.7
0.7
0.44 '
0.41
Trace
El emeht
Gallium
Indium
Silver;
Germanium
Thallium
Zirconium
Strontium
Bromine
Fluorine
Ruthenium
Tellurium
Cesium
Beryllium
Iodine
Lithium
Mercury
Tantalum
Rhodium
Gold
Platinum
Scandium
Bismuth
Cerium
Tungsten
Hafnium
Yttrium
Niobium
Concentration,
ppm
0.4
0.3
0.3
0.2
0.2
0.2
0.15
0.13
0.12
0.10
0.1
0.09
0.08
0.06
0.06
0.04
0.04
0.03
0.02
0.02
0.02
i 0.01
0.006
; " 0.004
0.003
0.002
0.001
Source: Tyndall et al (24).
-------
Analysis of trace element emissions from coal-fired utility boilers
has indicated that for certain of the trace elements, there are definite
differences in the concentrations of these elements between the fly ash and
bottom ash fractions, and between the fly ash at the inlet to control devices
and the suspended particles in the stack gas. Most of the studies agreed
that-trace elements are distributed into the various fractions of coal com-
bustion residue according to definite partitioning patterns. The three
main classes of partitioning behavior observed are (25):
Class I. Elements which are approximately equally concentrated
in the fly ash and bottom ash, and are not volatilized to a
great extent during combustion,
Class II. Elements which are enrivhed in the fly ash relative
to their concentrations in the bottom ash, due to volatilization
during combustion. . . .
e Class III. Elements which are discharged to the environment in
the gas phase.
According to Klein et al (25), results from the study conducted at
Tennessee Valley Authority's Allen Steam Plant indicated partitioning of
the elements into the three classes as discussed above and shown in Table
A-3. Of 38 elements analyzed, 20 elements were found to belong to Class I,
9 elements were found in Class II, and 3 elements in Class III. Six
elements - chromium, cesium, sodium, nickel, uranium, and vanadium - could
not be definitely assigned to a class but appeared intermediate between
Class I and Class II. Selenium exhibits behavior intermediate between
Class II and Class III. In examining the results from other studies, it
is noted that the enrichment behavior of the trace elements are generally
consistent, despite the differences in the furnace and coal types, sampling
and analysis procedures. Generally the Class II elements are heavily
concentrated on the smaller fly ash particles.
In the calculation of trace element emission factors, it is more
.convenient to use the concept of enrichment factors. This is because trace
element emissions are dependent on the trace element content of coal, the
boiler firing configuration, size of the boiler, as well as the efficiency
of particulate control devices, among other factors. The use of enrichment
factors enables direct comparison and compilation of trace element emission
44
-------
TABLE A-3. PARTITIONING OF ELEMENTS IN COAL COMBUSTION RESIDUES
Class
Class I
Concentrated
fly ash and
Class II
Enriched in
equally between
bottom ash
fly ash
Al
Ba
Ca
Ce
As
Cd
Cu
Co
Eu
Fe
Hg
Ga
Mo
Pb
Element
K
La
Mg
Mn
Sb
S
Zn
Rb
Sc
Si
Sm
Sr
Ta
Th
Ti
Class III
Discharged in the gas phase
Hg
Cl
Br
Source: Klein (25).
data on a normalized basis, and facilitates the computation process. For
the purpose of this report, the overall enrichment factor (ER) is defined
as the ratio of the concentrations of an element and aluminum in stack fly
ash, divided by the corresponding ratio in coal. Thus,.
ER,
where: C-j = concentration of element i, yg/g
CAT = concentration of aluminum, ug/g
subscript s = stack fly ash
subscript c - coal.
Aluminum is selected as the reference element because it is known to exhibit
about the same concentration in fly ash and bottom ash, and among fly ash
particles of different sizes. Enrichment factors, as defined above, are
dependent on the collection efficiency of control devices. Since enrichment
for the volatile Class II species is more pronounced in the finer particu-
lates, enrichment factors for boilers equipped with high efficiency control
devices are correspondingly higher. Table A-4 summarizes enrichment factor
data obtained from many studies reported in the literature.
45
-------
TABLE A-4. TRACE ELEMENT ENRICHMENT FACTORS FOR COAL-FIRED UTILITY BOILERS
EQUIPPED WITH PARTICULATE EMISSION CONTROLS*
O>
Trac
El erne
Alt
As
B
Ba
Be
Br
Ca
Cd
C1
Co
Cr
Cu
F
Fe
Hg
K
LI
Mg
Kn
Ho
Na
Hi
f
Pb
Sb
Se
Si
Sn
Sr
Th
U
V
In
e No
Control
nt Mean
Enrichment s(x)
Factor, x
1.00
1.9
.-
.79
..
..
0.80
2.0
»
1.7
2.0
2.6
.
0.96
..
1.3
«
1.3
1.0
-
1.6
2.1
2.6
3.8
1.7
0.97
1.2
1.1
1.5
2.4
2.2
0.6
..
.-
..
..
0.05
0.1
».
0.1
0.1
0.7
BW
0.32
..
0.1
-.
0.1
. 0.03
--
0.1
0.6
,-
0.7
0.9
0.4
--
..
«
0.07
0.6
0.3
Mechanical Precipttator
No. of
Data
Points
..
2
1
2
2
2
2
2
-w
2
._
2
..
2
2
-.
2
2
»-
2
2
2
1
_.
1
1
2
2
2
Mean
Enrichment .s(x)
Factor, x
1.00
3.23
0.48
0.82
0.21
3.40
0.90
1.2
0.86
0.87
.
»_
*
.*
0.43
2.55
.
1.1
._
1.45
6.03
1.23
._
0.63
0.99
..
..
1.26
1.54
..
2.34
0.35
0.01
..
2.30
0.37
0.06
,0.23
._
._
..
0.35
0.35
^0
0.15
siis
0.87
._
0.48
»-
0.34
0.58
No. of
Data
Points
3
2
2
**
2
3
1
2
2
^_
w.
»
2
2
2
2
3
2
2
1
»
2
3
Mean
Enrichment
Factor, 5
1.00
4.36
2.23
0.85
3.37
1.10
3.64
1.56
3.21
2.22
1.23
1.11
1.29
1.53
1.52
2.95
1.13
5.14
0.99
8.08
12.48
16.0
1.0
4.95
1.36
0.88
0.98
1.22
1.69
ESP
S(x)
1.48
1.50
0.24
1.51
0.41
1.31
0.38
1.28
0.80
^_
0.41
0.32
1.12
0.29
0.21
1.18
0.33
2.47
0.22
2,77
4.93
8.17
3.17
0.43
0.34
0.18
0.29
0.34
Ho. of
Data
Points
12
4
8
10
^
7
11
13
11
10
w
12
._
6
2
6
14
9
7
12
2
13
B
8
1
8
6
4
6
12
13
Wet
Mean
Enrichment
Factor, x
1.00
19.05
18.3
2.75
1.55
m .
2.53
31.35
.»
5.00
26.15
2.50
2.05
.*
1.33
0.12
4.4
1.80
65.7
60.6
99.7
10.4
11.4
8.75
12.9
1.0
70.3
6.70
0.14
6.1
11.3
9.90
Scrubber
s(x>
.
'17.05
«
1.65
0.3S
..
1.97
29.65
.
3.80
24.25
0.10
.
0.85
.-
0.37
--
-.
..
58.3
«
..
..
5.3
4.45
2.40
«
--
5.60
.
-
4.50
Tio. of
Data
Points
._
2
1
2
2
.-
2
2
, .*
2
2
2
.
2
-
2
1
2
2
2
2
Enrichment factor is defined as the concentration ratio of element I to the reference element
emitted 1n fly ash, divided by the corresponding ratio In coal.
Altuiinun is the reference element. Therefore, by definition, the enrichment ratio Is unity.
-------
APPENDIX B
PARTICULATE ECS DATA
Data are presented in this Appendix on the collection efficiency of
various ECS's for different sizes of fine particles, and on estimated
emissions of trace metals from bituminous coal combustion. These data
supplement information in Section 4 of the text.
FINE PARTICLE EMISSIONS: .... ...
The data sources for fine particulate emissions are rather limited.
Four sources were used for this compilation. The Fine Particle Emissions
Information System (FPEIS) (26), a computerized data base maintained by
the Environmental Protection Agency, provided three sets of data on parti -
culate size distributions from pulverized coal-fired dry-bottom boilers,
plus data on stokers-with fly ash reinjection., Shannon et al (27) and
Weast et al (28) of Midwest Research Institute (MRI) reported efficiencies
of particulate control devices as a function of particle size, and particu-
late size distributions'for cyclone boilers fired with bituminous coal.
Cato et al (29) reported particulate size distributions for pulverized coal-
fired boilers, spreader stokers without fly ash reinjection, and oil-fired
boilers. Crawford et al (30) reported particulate size distributions for
three pulverized coal-fired boilers: two tangential and one horizontally
opposed.
For five types of utility boilers, the particulate emission factors
for four size fractions (less than 1 ym, 1-3 ym, 3-10 ym and greater than
10 ym, where the particulate sizes represent the aerodynamic particle dia-
meters) are presented in Table B-l. The total uncontrolled particulate
loadings have been calculated assuming a national average of 14.09 percent
ash in the feed coal, except for the spreader stoker with fly ash reinjec-
tion. This value was calculated using the AP-42 emission factor. These
total particulate loadings were multiplied by the size distributions for
each category to obtain the emission factor for each size fraction.
47
-------
TABLE B-1. SIZE DISTRIBUTIONS FOR CONTROLLED AND UNCONTROLLED
PARTICULATE EMISSIONS FROM UTILITY BOILERS
CD
Combustion
System
Pulverized Bituminous
Coal-fired Dry- bottom
Boiler
Bituminous Coal -fired
Spreader Stoker With
Fly Ash Re inject Ion
Bituminous Coal-fired
Spreader Stoker
Without Fly Ash
Relnjectton
-
Bituminous Coal-fired
Cyclone Boiler
Residual 011 -fired
Boiler
Participate
Control Device
None
Cyclone
Hultlclone
Scrubber
Electrostatic Preci pita tor (ESP)
Venturl Scrubber
Fabric Filter
None
Cyclone
Multiclone
. Scrubber
Electrostatic Preci pita tor (ESP)
Venturl Scrubber
Fabric Filter
None
Cyclone
Hultlclone
Scrubber
Electrostatic Preci pita tor (ESP)
Venturl Scrubber
Fabric Filter
None
Cyclone
Hultlclone
Scrubber
Electrostatic Predpltator (ESP)
Venturl Scrubber
Fabric Filter
None
Cyclone
Hultlclone
Scrubber
Electrostatic PreclpUator (ESP)
Venturt Scrubber
Fabric Filter
1 '! I .<
! Emission Factor, ng/J
10pm
3737
1121
187
15
19
<8
<1.9
4017
1205
201
16
20
<8.0
<2
2963
889
148
12
15
<6
<1.5
254
76
13
1.0
1.3
-------
Uncontrolled particulate emissions from pulverized dry-bottom and
stoker units have similar size distributions, with about 1 percent by weight
of the particulates in the less than 1 urn size fraction and the bulk (78-90
percent) of the particulates in the greater than 10 ym fraction. For the
cyclone boiler, a greater proportion of fine particulates is emitted - 8
percent by weight are in the less than 1 ym fraction and only 34 percent
by weight are in the greater than 10 ym fraction. Oil firing produces much
fewer and finer particulates than coal firing, with 35 percent by weight
in the less than 1 ym fraction and only 13 percent by weight in the over r
10 ym fraction.
Emission factors from boilers equipped with particulate removal devices
(cyclones, multic!ones, scrubbers, electrostatic precipitators, Venturi
scrubbers, and-fabric filters) were calculated by using average efficiencies
of particulate removal for each size fraction as presented in Table B-2.
By comparing total particulate loadings from one boiler type with various
emission control devices, it is clear that fabric filters have the greatest
removal efficiency. In terms of the health effects of particulate emissions,
the size fractions with aerodynamic diameter less than 1 ym may be considered
the most important, since these particles are not removed by the upper
respiratory tract. For this size fraction, the high efficiency electrostatic
precipitators and. the fabric filters are the most efficient particulate
removal devices.
TRACE METAL EMISSION FACTOR ESTIMATES
The principal reference sources used to develop the trace element
emissions data base for bituminous coal-fired utility boilers were data
from the CCEA program and the following literature sources:
A study by Schwitzgebel et al of the Radian Corporation
to characterize trace element emissions from three coal-
fired utility boilers (4) - The units sampled include a
tangentially-fired 330-MW boiler with three venturi scrubbers,
a tangentially-fired 350-MW boiler with a hot side electro-
static precipitator, and a 250-MW cyclone boiler with a
mechanical cyclone for particulate control. The first two
plants were fired with Wyoming subbituminous coal and the
third plant with lignite coal. A material balance approach
for 27 elements was used to characterize the effluents around
the power plants.
49
-------
TABLE B-2. EFFICIENCIES OF PARTICIPATE REMOVAL BY
CONTROL DEVICES FOR VARIOUS SIZE FRACTIONS
Particulate
Control Device
Efficiencies of Participate Removal, %
<1 pro 1-3 ym 3-10 ym >10 ym
Medium Efficiency
Cyclone
Multiclone
Medium Efficiency
Scrubber
High Efficiency
ESP
Venturi Scrubber
Fabric Filter
0.25
11
26
96.5
71
96
12
54
77
98.25
99.5
99.75
50
85
98.0
99.1
>99.8
>99.95
70
95
99.6
99.5
>99.8
>99.95
Source: Reference 2.
A study conducted by Bolton et al of the Oak Ridge National
Laboratory on the Allen Steam plant (31) - The boiler sampled
was a 290-MW cyclone unit burning coal from Kentucky and
Southern Illinois, and equipped with an electrostatic preci-
pitator. Determinations were made for concentrations and
mass balances of 54 elements.
A study conducted by Kaakinen et al of the University of
Colorado on the Valmont Power Station (32) - The boiler
sampled was a 180-MW unit, equipped with a mechanical collector
followed by an electrostatic precipitator in parallel with a
wet scrubber. The samples collected were for all input streams
and all outfall streams. Chemical analysis data were available
for 18 elements, including three radionuclides.
A study conducted by Mann et al of the Radian Corporation (12)
- The units sampled were two 350-MW tangentially-fired boilers
using Wyoming subbituminous coal and equipped with electrostatic
precipitators. Mass balance data for 15 elements were available.
A study conducted by Klein et al of Oak Ridge National Laboratory,
also for the Allen Steam plant (25) - The concentrations and mass
flow rates of 37 elements were followed through the cyclone boiler.
50
-------
t A study conducted by Hillenbrand et al of Battelle Columbus
Laboratories for the Edgewater Power plant (33) - The unit
sampled was a pulverized coal-fired boiler equipped with an
electrostatic precipitator. Enrichment factors for 27 trace
elements across the electrostatic precipitator were reported.
e A study conducted by Gorden et al of the University of Maryland
on the Chalk Point Station (34) - Two 355-MW units firing
pulverized coal were sampled. The samples collected included
coal, bottom ash, fly ash from the economizer, fly ash from
the electrostatic precipitator, and fly ash suspended in the
stack gas. Analysis for 35 elements were performed. The
enrichment of an element in the suspended fly ash relative to
its concentration in the coal was determined.
A report by Curtis of the Ontario Hydro on trace element
emissions from the R.L. Hearn, Lakeview, Lambton and Nanticoke
stations (35) - The four stations sampled have a total of 24
boilers firing pulverized coal and a generating capacity of
9,200 MW. Data presented were based on a continuing program
for the measurement of 44 trace elements in coal, ash, and
stack gas. The boilers sampled were equipped with electro-
static precipitators of 98.6 percent average efficiency.
t A study conducted by Cowherd et al of the Midwest Research
Institute on the Widows Creek Power Plant (5) - The unit
sampled was a 125-MW, tangentially-fired boiler equipped with
a mechanical fly ash collector. Analysis and mass balances
for 22 trace elements were reported.
t A study conducted by Lee et al of the U.S. Environmental
Protection Agency (9) - The unit sampled was a 105-MW coal-
fired power plant in Illinois. Changes in concentrations of
12 trace elements across the electrostatic precipitator were
reported.
A study conducted by Ragaini and Ondov of the Lawrence Livermore
Laboratory (36) - The plant tested was a single tangentially-
fired unit fed with western coal and operating at 430 MW. The
unit used a cold side electrostatic precipitator with between
99.5 and 99.85 efficiency. Enrichment factors for 20 trace
elements were determined.
In addition to the above studies, three reports of a survey nature
provided an extensive review of trace element emissions from coal combus-
tion. These are the reports prepared by Ray and Parker of the Tennessee
Valley Authority (37), Oglesby and Teixerira of the Southern Research Ins-
titute (38), and the Radian Corporation (6).
51
-------
To estimate air emissions of trace elements from coal-fired utility
boilers, it is necessary to examine the data base from three aspects: (1)
the trace element content of coal consumed by utilities, (2) the fate of
f
trace elements during coal combustion, and (3) the effect of different
pollution control devices. The first two points were addressed in Appendix
A, where data on the trace element content of coal were presented, and an
overall enrichment factor defined to describe the fate of trace elements
during combustion.
With the use of enrichment factors, trace element emission factors can
be calculated from the trace element content of coal, the heating value of
coal, the fraction of coal ash produced as bottom ash, and the efficiency
of the control device for particulate removal. In equation form, the
emission factor EF^ for trace element i is calculated as:
(C.)c -
EF. = nTL-£- (1 - EB) (1 - En) ER. x 103
c
where: EFj -Demission factor for element i, ng/J
(Ci)c = concentration of element in i coal, yg/g
He = higher heating value of coal, J/g
EB = fraction of coal ash retained in the boiler as bottom ash
En = fractional total particulate collection efficiency of
control device n
ER.j = overall "enrichment factor for element i.
The values for (C^)c and ER^ are tabulated in Appendix A. Values for
Eg depend on boiler type and were discussed in Section 3 of the text.
For E , the following values have been assumed:
t For mechanical precipitators (multic!ones), E = 0.702
For ESP's, E = 0.9787
o For wet scrubbers, E = 0.996
Actual efficiency in the field is highly dependent on inlet particle loading
and size distribution, as well as other factors specific to the device (e.g.,
fly ash resistivity for ESP's).
Table B-3 is a summary of estimated trace element emission factors from
controlled and. uncontrolled combustion of bituminous coal. Only those trace
52
-------
TABLE B-3. ESTIMATED TRACE ELEMENT EMISSION FACTORS
FROM COMBUSTION OF BITUMINOUS COAL
Trsee
Element
No
Control
Devicet
Estimated Emission
Mechanical
Preel pita tar
Alone*
Factor, pg/J*
Electrostatic
Precl pita tor
Alone**
Wet
Scrubber
Alonett
As
Ba
Be
Cd
Co
Cr
Cu
H1
Pb
Sb
Se
V
As
Ba
Be
Cd
Co
Cr
Cu
N1
Pb
Sb
Se
V
PULVERIZED DRY BOTTOM BOILERS
520 265 25
3.700 662 89
1+* 7.3 2.2
41 21 1.7
400 64 7.9
1,620 290 55
1,200 121 . 23
1.200 183 62
570 94 39
140 68 10
140 30 28
2.500 390 27
PULVERIZED MET BOTTOM BOILERS
430 216 21
3.000 538 72
I 6.0 1.8
34 17 1.4
330 52 6.4
1.300 235 45
1,000 98 16
950 149 SO
460 77 31
120 55 8.1
110 24 23
2.000 320 22
CYCLONE BOILERS
21
54
0.19
2.8
4.8
85
4.8
227
10
1.3
4.2
48
19
51
0.18
2.6
4.5
79
4.5
213
9.6
1.2
4.0
45
As
Ba
Be
Cd
Co
Cr
Cu
N1
Pb
Sb
Se
V
88
620
I
7.0
68
270
210
200
96
24
23
420
" " 44.8" "" "
112
1.2
3.5
11
49
20
31
16
11
5.0
66
' 4.3
15
0.37
0.29
1.3
9.3
3.9
11
6.6
1.7
4.7
4.6
3.5
9.2
0.03
0.46
0.80
14
0.81
38
1.7
0.22
0.71
8.1
Estimated emission factors derived from weighted average Individual trace element
concentrations in bituminous coal; national average heating value of bituminous coal
as fired (25,586 J/g); mean or assumed total partlculate collection efficiencies of
various boilers and ECS's; and experimental mean enrichment factors for individual
trace elements across a boiler or control device.
Mean total partlculate collection efficiencies (mass of bottom ash per mass of ash In
Incoming coal) for the three boiler types are as follows, with number of data points
in parenthesis: pulverized dry bottom boilers « 0.20 (48); pulverized wet bottom
boilers 0.35 (91); cyclone boilers 0.865 (44).
Assured total partlculate collection efficiency 1s 0.702.
Assumed total partlculate collection efficiency Is 0.9787, probably low for
existing NSPS.
^Assumed total particulate collection efficiency Is 0.996.
**lnsuff1cient data.
53
-------
elements with MATE values less than 1 mg/m are shown. Only bituminous
coal combustion is considered because:
Coal combustion accounts for more trace element emissions
than oil combustion in the U.S.
8 Bituminous coal combustion accounts for about 95.6 % of
the energy consumed by utility boilers in the U.S., based
on 1978 data.
The data base for controlled trace element emissions from
bituminous coal combustion is more extensive than that for
oil firing or for anthracite or lignite coal firing.
It must be noted that the data-in Table B-3 are based on the use of a
single control device. The overall enrichment factor ER^ takes into account
both boiler firing type and emission control device n, where n = 1. For
facilities-utilizing-two-or-more ECS's in series, a different method of
calculating the emission factor EF.. must be used. The overall enrichment
factor ER.j may be expressed as the product of enrichment factors across the
boiler and all subsequent ECS's, i.e., .
where: ERjj = enrichment factor for element i across the boiler .
r^ = enrichment factor for element i across control device n
n = number of control device (total of N devices).
The value of r.. will vary depending on the inlet particle loading and size
distribution. The value of ER* is simply the enrichment factor for the no
control situation, from Table A-4. Using data reported in Table A-4, r^
for each control -device/boiler- type combination can be estimated. Then the
following general equation can be used to estimate EF. :
where: (n)n = enrichment ratio for element i across control device n.
Alternately, if the collection efficiency of control device n is known
specifically for element i,
54
-------
.ERg(l-EB)Ji [l -(£,)]
where: (E.) = fractional mass collection efficiency of device n for element
1 n. i (total of N devices).
Statistical evaluation of the data used in calculating the emission
factors in Table B-3 has led to the following conclusions concerning the
adequacy of the emissions data base for bituminous coal-fired utility
boi1ers:
For pulverized bituminous coal-fired dry bottom boilers
equipped with electrostatic precipitators, the existing data
base is inadequate for barium, beryllium, calcium, iron,
lithium, nickel, phosphorus, lead, and selenium, and adequate
for all-the-other-trace elements.
« For pulverized bituminous coal-fired wet bottom boilers
equipped with electrostatic precipitators, the existing data
base is inadequate for beryllium, calcium, Iron, lithium,
nickel,-phosphorus, and adequate for all the other trace
elements.
§ For bituminous coal-fired cyclone boilers equipped with
electrostatic precipitators, the existing data base is in-
adequate for beryllium, iron, lithium, nickel, and phosphorus,
and adequate for all the other trace elements.
0 The existing data base characterizing trace element emissions
from any type of bituminous coal-fired boilers equipped with
wet scrubbers is generally inadequate. Among the trace
elements, the existing data base is only adequate for cadmium,
copper, potassium, molybdenum, antimony, selenium, and zinc.
The existing data base characterizing trace element emissions
from any type of bituminous coal-fired utility boilers with
no emission control or equipped with mechanical precipitators
is generally inadequate due to lack of sufficient enrichment
ratio data.
t The existing trace element emissions data base for bituminous
coal-fired stokers is inadequate since no data are available.
55
-------
APPENDIX C
FGD SYSTEM DATA
Most data on FGD systems reported in the literature are for installa-
tions on coal-fired boilers. However, limited trace element data exist.
Data presented in this Appendix are intended to supplement data presented
in Section 4 of the text.
COAL-FIRED UTILITY SOURCE
Table C-l presents further data from the study by Leavitt et al (21)
reported in Section 4. Concentration of trace metals emitted to the
atmosphere-is, compared _to the MATE value to arrive at a discharge severity
for each species. The data indicate that the FGD system reduced the total
discharge severity of the flue gas stream (for the indicated elements) from
the uncontrolled level of 740 to 54, a 92.7% reduction.
COAL- AND OIL-FIRED INDUSTRIAL SOURCE
Leavitt et al (39) studied an industrial boiler equipped with dual-
alkali FGD system which was capable of firing either coal or oil. They
obtained data on the performance of the FGD system during both coal and oil
combustion. Trace element collection efficiencies exhibited by the FGD
system are presented in Table C-2. The overall removal efficiency for the
indicated elements was 99.5% for the coal-fired test. Of the 22 elements
reported, 18 exceeded their MATE value at the scrubber inlet, and only four
at the outlet. Oil firing resulted in only an 87% removal of trace metals,
due to the smaller median diameter of oil fly ash particles. Of the same
22 elements, 11 exceeded their MATE value at the scrubber inlet, and five
at the outlet.
Other conclusions of the study are summarized below, for both coal and
oil firing.
56
-------
TABLE C-1. CONTROLLED AND UNCONTROLLED TRACE ELEMENT EMISSIONS FROM
A COAL-FIRED UTILITY BOILER EQUIPPED WITH FGD SYSTEM
Concentration , mq/m
Element Scrubber Scrubber
Inlet Outlet
MATE for Ai r
(Health Basis)1"
Discharge Severity (PS)
Scrubber Scrubber
Inlet Outlet
Al
As
Be
Ca
Cd
Co
Cr
Cu
Fe
Hg
Mn
Ni
Pb
Sb
Se
Sr
V
Zn
**
TDS
TWDSft
132
0.98
0.021
49
5.1
0.19
1.3
1.2
401
0.095
0.70
2.0
11
0.78
0.37
0.46
0.78 .
105
3.0
0.94
0.0018
2.0
0.58
0.013
0.12
0.19
13
0.0057
0.150
0.054
2.9
0.27
0.088
0.038
0.083
21
10
0.50
0.002
13
0.100
0.100
0.500
0.100
1.0
0.050
5.0
0.015
0.15
0.50
0.20
10
0.050
5.0
13
2.0
11
3.8
51
1.9
2.6
12
400
1.9
0.14
130
73
1.6
1.9
0.046
16
21
740
504 ng/J
0.30
1.9
0.90
0.15
5.8
0.13
0.24
1.9
13
0.11
0.03
3.6
19
0.54
0.44
0.0038
1.7
4.2
54
22 ng/J
*
Analyses
* M » v> ^ i** i tMt
by Level
nf+it^f\ ^T ** w
2 procedures.
* r**i * /M
A*TC i £ *%M <£»*#%**
t M ft «**k*Ml*^ Mh«« ft *+**f*
******* r* \ i
**
AH] calculation, with the follov/ing exceptions: Ca and Sr - AH«; Co
and Cr as the element; Zn as ZnO.
Discharge severity (DS) is defined as the ratio, of discharge concentra-
tion to MATE value.
Total discharge severity (TDS) = IDS for the indicated elements.
Total weighted discharge severity = z(DS x emission factor) for elements
with MATE <1 mg/m3 only.
57
-------
TABLE C-2. COMPARISON OF TRACE ELEMENT REMOVAL
EFFICIENCIES OF AN SOe SCRUBBER ON A
COAL OR FUEL-OIL FIRED INDUSTRIAL BOILER
Trace
Element
Al
As
B
Be
. Ca
Cd
Co
Cr
Cu
Fe
Hg
Mg
Mn
Mo
Ni
Pb
Sb
Se
Sr
V
Zn
Zr
Total
Removal Efficiencies,
Fuel -Oil
Fired Boiler
92
81
.... 93
Unknown
83 . : .
77
89
90
99
95
87
91
87
89
83
94
91
87
98
71
90
94
87
Percent
Coal -Fired
Boiler
99
97
88
98
99
99
99
95
99
"99
55
99
98
99
95
99
99
97
99
98
98
99
99
Note: SOp scrubber is a double alkali unit.
Source: -Leavitt et al (39).
58
-------
Coal Firing
Average SO^ removal was 97%.
NOX reductions of 0-24% were measured across the scrubber,
but may have been due to sampling phenomena.
Total particulate removal by the scrubber was 99.4%.
t S03 removal was 33%.
SO*" removal was 88%.
Total organic emissions were generally less than 9 ng/J and
these emissions appeared to'be primarily C-| to GS hydrocarbons
and hydrocarbons heavier than C]5« While uncontrolled
emission rates for Cy to C-jg, and C-jg and higher hydrocarbons
are low, emissions of these organics were further reduced by
21% and 85%,-respectively, in the scrubber unit. POM was not
found in the scrubber inlet or outlet at detection limits of
0.3 pg/m3.
Oil Firing
Average SOg removal was 97%.
Total particulate removal was between 75% and 84%.
SOg removal was 28%-29%, about the same as for coal firing.
S0*~ removal v/as about 60%."
Organic emissions were generally less than 5 ng/J and were
similar in character to coal-fired emissions. Approximately,
88% and 83% of Cj to C-jg, and C-js and higher hydrocarbons,
respectively, were removed by the FGD system.
59
-------
APPENDIX D
NOVEL EMISSION CONTROL SYSTEMS
Often an ECS cannot be strictly categorized as a scrubber, fabric
filter, or ESP. The system may differ significantly from conventional
technology. In these cases, the ECS may be considered a novel device.
Novel systems described in this appendix are designed for either fine
particulate removal, or simultaneous SOg and participate removal.
Novel Fine Parti oil ate ECS's
The following fine particulate ECS's have been tested by EPA as novel
devices (40, 41):
1. Sonic agglomerator (Braxton)
2. Steam-hydro scrubber (Lone Star Steel)
3. Dynactor scrubber (RP Industries)
4. Pentapure impinger (Purity)
5. ADTEC scrubber (Aronetics)
6. CHEAP (Andersen 2000)
7. Centrifield scrubber (Entoleter)
8. Gravel bed (Rexnord)
9. Electrostatic scrubber (APS)
10. Electrotube (APS)
11. FRP-100 low energy wet scrubber (Century Industrial Products)
12. Apitron (American Precision Industry)
13. Electrified bed (Particulate Control System)
14. Ionizing wet scrubber (Ceilcote)
A detailed description of these devices is beyond the scope of this report.
Notably, all these ECS'-S utilize water except for the gravel bed, Apitron,
and electrified bed. Test results for eight of the systems listed above
are presented in Figure D-l (40). Only the Steam-hydro, ADTEC, and elec-
trostatic scrubbers demonstrate greater than 90% removal (<10% penetration)
for 0.5 micron diameter particles. The Apitron shows greater than 90%
removal of 0.5 ym particles (41).
Table D-l compares the performance of seven novel ECS's to the per-
formance of conventional low (25 cm WG) and high (273 cm WG) pressure drop
60
-------
PENETrlflTION flS ft FUNCTION OF RERCOYNfWIC PRRTICLE OIWCTCR
/E;. BED
DY.N & CENT
ELECTSOSTfl
CHEW
RDTEC
PfWTICLE DlflNCTER. MICRONS
Figure D-l. Particulate control by novel devices.
Dynactor and centrifield scrubbers.
Electrostatic scrubber.
61
-------
venturi scrubbers. The Steam-hydro and ADTEC units show the best perform-
ance of the systems shown, being slightly better than the high pressure
drop venturi in removing fine particles. Unfortunately, both novel devices
require a large amount of enthalpy from waste heat to drive the scrubber.
When total energy use is considered, performance of these and other novel
devices is in line with conventional technology.
Simultaneous SOo/Particulate ECS
Baghouse application of dry sorbent technology appear to have a sound
technical basis. Various studies and test programs have resulted in designs
that can be effective at SO^ removal rates of 70% to 80% using nahcolite,
a naturally-occurring mineral form of sodium bicarbonate, as the sorbent
material. Removal rates of 85% to 90% are possible based on test data,
however, further work on system design is needed to attain these higher
efficiencies.
TABLE 0-1. PERFORMANCE OF NOVEL FINE PARTI.CULATE CONTROL DEVICES
TO CONVENTIONAL VENTURI SCRUBBERS (40)
Smallest Diameter
Device Collected at 50%
Efficiency, microns
Impinger 3.5
Gravel Bed 1.4
Venturi (25 cm WG) 0.7
Dynactor 0.65
Centrifield 0.6
CHEAF 0.53
Venturi (273 cm WG) 0.3
ADTEC 0.13
Steam-hydro <0.1
62
-------
During January and February of 1977, a pilot baghouse study was con-
ducted at Basin Electric Power Cooperative's Leland Olds Station in Stanton,
North Dakota. Wheelabrator-Frye Incorporated was responsible for the
equipment and conduct of the test. Superior Oil Company provided the
nahcolite, and Behtel Power Corporation coordinated the overall effort (42).
The Leland Olds Station burned lignite with an ash content of 8.35%,
sulfur content of 0.75%, and higher heating value of 6495 Btu/lb (15.07 x
10 J/kg). Nahcolite was injected downstream of the air preheater and was
collected, along with the coal ash, in 12 bags, each 11.5 inches (0.292 m)
in diameter and 30 feet (9.15 m) long. The bags were made of combination
fiberglass cloth with a silicone graphite finish. Total fabric area was
o
1080 square feet (100 m), and the unit was operated at a filter ratio of
3:1. Bags were cleaned by deflation and mechanical shaking.
A total of 81 runs were conducted during the program. Test results
relating SOg and NOX removal to stoichiometric ratio are shown in Table
D-2. The data show that simultaneous removal of SOg and particulates in a
properly designed-baghouse is practical.
63
-------
TABLE D-2. TABULATED DATA FROM NAHCOLITE TEST PROGRAM (43)
o>
Stolchlometrtc Removal Efficiency
Test
1
2
3
4
5
6
9
10
14
15
16
16A
17
18
19
19A
20
20G
22
22A
226
23
23G
26
27
Ratio
1.21
1.05
0.98
1.12
0.93
1.08
0.79
0.87
0.51
0.45
0.86
0.79
1.50
1.52
0.92
1.66
0.48
0.42
1.14
0.90
0.83
1.63
1.13
0.89
0.92
J*
33.5
37.4
47.6
55.7
50. 5
82.7
46.4
54.6
47.0
37.2
68.8
64.8
85.7
90.1
67.2
91.0
13.1
22.2
27.0
24.0
49.3
40.3
60.3
65.3
59.0
N0x
14.3
14.5
12.0
13.4
9.2
18.5
13.9
12.6
9.1
11.0
24.5
19.2
24.0
35.0
33.6
52.9
6.0
5,9
13.7
6.8
11.1
25.4
10.6
12.5
Paniculate
99.4
99.7
99.6
99.7
99.8
99.8
99.9
99.7
99.9
99.8
99.4
99.8
Feed
Utilization
X
27.5
35.9
48.4
50.2
55. 5
76.9
59.2
63.3
96.2
84.6
79.9
81.9
58.4
59.9
75.1
55.0
27.7
53.5
23.7
26.2
58.9
24.4
53.2
73.1
64.0
Inlet
ppffl
956
998
861
865
903
842
949
944
830
1708
1707
1091
967
1719
2422
2760
918
948
1726
894
1655
915
947
831
1013
Pl>fl
628
669
668
C40
612
622
581
601
584
573
531
562
554
557
529
560
434
511
539
487
379
514
516
540
521
Temp.
°F
297
297
291
301
289
288
289
291
288
294
289
292
289
291
296
292
291
288
294
289 .
294
292
295
286
292
Outlet
PPfl)
636
625
451
383
447
146
509
429
440
1073
533
384
138
170
794
248
798
738
1260
679
839
546
376
288
415
WO.
ppfl
538
572
588
554
556
507
500
525
531
510
401
454
421
362
351
264
408
481
465
454
388
457
385
483
456
dec
Load
fU
419
402
401
419
398
419
408
419
416
408
399
399
407
407
387
401
378
402
397
380
416
372
397
407
406
Lb Steam
Itr
2.29
2.35
2.27
2.35
2.29
2.36
2.32
2.37
2.36
2.29
2.25
2.24
2.31
2.31
2.29
. 2.38
2.00
2.40
2.35
2.12
2.35
2.15
2.30
2.42
2.37
Dates
1977
2/4
2/5
2/2
2/3
2/12-15
2/14-15
2/15-16
2/16
2/17
2/19-23
2/20-23
2/18
2/21-24
2/22-24
2/25
3/2
2/27
3/1
2/28
2/26-27
2/28
2/27-28
2/28
3/3
3/1
-------
. APPENDIX E
ENVIRONMENTAL IMPACT ANALYSIS
In order to determine the environmental impact of controlled versus
uncontrolled non-criteria pollutant emissions, a preliminary analysis has
been performed based on trace element emissions data in this report.
Terminology and methodology set forth by the EPA (43) has been used to
judge the environmental acceptability of air emissions front bituminous
coal-fired dry-bottom boilers. Emissions estimates from AP-42 (3) and
Appendix B of this report form the basis of the uncontrolled emissions data.
ControVTed-emtssions-are-based on NSPS for S0~, NO , and particulates from
c ** i
Section 3.
Uncontrolled SOg emissions (in g/kg coal) from pulverized coal combus-
tion are estimated~as 19-times the sulfur content (in percent) of the coal
(3). Using national average heating value and ash content of bituminous
coal (Appendix A), this results in an emission factor of 1426 ng/J. Un-
controlled NOX emissions, based on 9 g NOX (as NOgJ/kg coal (3), average
352 ng/J.
Uncontrolled trace element emissions from pulverized dry-bottom coal
combustion were computed for 11 elements with MATE values less than 1 mg/m
in Appendix B of this report. These elements are As, Ba, Cd, Co, Cr, Cu,
Ni, Pb, Sb, Se, and V. Uncontrolled particulate emissions, based on
national average ash content and heating value of bituminous coal, and
E^ = 0.2 for pulverized bituminous coal-fired boilers, is estimated at 4410
ng/J.
Controlled 502* NOX, and particulate emissions are based on the NSPS
..of 520 ng/J, 260 ng/J, and 13 ng/J, respectively. This results in an
average removal efficiency of 63.5% for S02, 26.1% for NOX, and 99.7% for
total ^articulates.
65
-------
Using the required removal efficiency of E = 0.997 for total particu-
lates, trace element emissions for this degree of control can be calculated.
Assuming this control is provided by an ESP, controlled trace element
emission factors (pg/0) can be calculated from data in Table B-3 by de- .
creasing the ESP emission factors in the table by the ratio of (1 - 0.997)/
(1 - 0.9787).
Data from Shin (2) provide a conversion factor from emission factor
2
(ng/J) to emission rate (mg/m ). For pulverized dry-bottom coal combustion,
the emission factor (ng/J) is equal to a constant B, times the emission
rate (mg/m ). -The value of B is 0.47751 ± 0.0095 (95% confidence .range of. ..
the mean).
Discharge severities (43) can now be calculated for controlled and
uncontrolled emissions of S0« and individual trace metals. The results are
presented in Tables E-l and E-2.
Table E-3 presents a summary of this preliminary assessment. The
data show that application of emission controls to meet NSPS is extremely
effective in reducing the severity of the air emissions"from pulverized
coal-fired dry bottom boilers. Since these sources burned 73% of all coal
consumed for electric power production in the U.S. 1n 1978 (1), these
results are indicative of the Immense secondary benefits of emission
controls.
Data in Table B-3 were calculated based on an ESP efficiency of 0.9787.
66
-------
TABLE E-1. ENVIRONMENTAL ASSESSMENT OF UNCONTROLLED EMISSIONS FROM
PULVERIZED COAL-FIRED DRY-BOTTOM BOILERS
Element Emission
Factor, ng/J
As
Ba
Cd
Co
Cr
Cu
Ni
Pb
Sb
Se
V
Total
S02
NOX (as N02)
Total
Grand Total
.52
3.7
.041
.40
1.62
1.2
1.2
.57
.14
.14
2.5
1426
352
Emission
Rate, mg/m^
1.09
7.75
.086
.838
3.39
2.51
2.51
1.19
0.293
0.293
5.24
2990
737
i
MATE Value,
mg/m^
.50
.50
.10
.10
.50
.10
.015
.15
.50
.20
.050
13
9
.Discharge
Severity
2.18
15.5
.86
8.38
6.79
25.1
168
7.95
.586
1.47
105
341
230
82
312
653
67
-------
TABLE E-2. ENVIRONMENTAL ASSESSMENT OF CONTROLLED EMISSIONS FROM
PULVERIZED COAL-FIRED DRY-BOTTOM BOILERS
Element Emission
Factor, ng/J
As
Ba
Cd
Co
Cr
Cu
Ni
Pb
Sb
Se
V
Total
S02
NOX (N02)
Total
Grand Total
.0035
.0125
.00024
.0011
.0077
.0032
.0087
.0055
.0014
.0039 '
.0038
520
260
Emission
Rate, mg/m3
.0074
.026
.00050
.0023
.016
.0068
.018
.012
.0030
.0083
.0080
1090
544
MATE Value,
mg/m3
.50
.50
.10
.10-
.50
.10
.015.
.15
.50..
.20
.05
13
9
Discharge
Seven ty
.015
.053
.0050
.023
.032
.068
1.2
.077
.0059
.041
.16
1.70
83.8
60.5
144.3
146
TABLE E-3. ENVIRONMENTAL ASSESSMENT SUMMARY
Uncontrolled
Trace Elements
S02 + NOX
Total
Discharge
Severi ty
341
312
653
% of
Total
52
48
__
Controlled
Discharge
Severi ty
1.7
144.3
146
% of
Total
1
99
__
% Reduction
in Discharge
Severi ty
99.5
53.8
77.6
68
------- |