United States
Environmental Protection
Agency
Environmental Sciences
Research Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S3-83-109 Jan. 1984
&EPA Project Summary
Emission Characterization of
Major Fossil Fuel Power Plants
in the Ohio River Valley
Gary Baker, Paul Clarke, Richard Gerstle, Wade Mason, and Mark Phillips
This study characterizes the atmos-
pheric emissions from five major coal-
fired power plant units in the Ohio River
Valley between Portsmouth, Ohio, and
Louisville, Kentucky. This characteriza-
tion provides data that are representa-
tive of the boiler fuel emission control
combinations of the current power
plant population as well as those
scheduled to go on line before the end
of 1983. The Environmental Protection
Agency will use these data to determine
the impact of existing and planned
power plants on air quality in this river
valley.
This Project Summary was developed
by EPA's Environmental Sciences
Research Laboratory, Research
Triangle Park, NC. to announce key
findings of the research project that is
fully documented in a separate report of
the same title (see Project Report order-
ing information at back).
Introduction
To better understand the relative con-
tribution to the ambient air quality attri-
butable to emissions from coal-fired
power plants, the Environmental
Protection Agency (EPA) is conducting a
two-phase concurrent study in the Ohio
River Valley. The first phase involves the
selection of representative coal-fired
boilers that typify emissions from units
currently in use and those to be brought
on line prior to the end of 1983. This
phase will provide source emission data
for the second phase of the project: an
ambient monitoring placement study to
be conducted over a 3- to 5-year period.
This particular study presents the results
of the first phase.
The information presented in this
report is representative of emissions from
power plants operating in the Ohio River
Valley. The results of this report should
not be used to estimate emissions for
power plants located outside the Ohio
River Valley Region.
In the source emission study, five rep-
resentative units were selected for
testing. The constituents measured were
total filterable particulates, particulate
sulfate, sulfuric acid mist (H2SO4), sulfur
dioxide (SO2) particulate and gaseous
chlorides, and particulate and gaseous
fluorides (sampled by manual methods);
and sulfur dioxide (S02), carbon dioxide
(C02), nitrogen oxides (NO and NOx), and
oxygen (O2) (sampled by continuous
monitors). Particulate fractions were
calculated by particulate sizing and
elemental analysis was conducted for
particle size fractions.
This report presents the selection
strategy utilized to choose the five test
sites, a description of the plants tested,
the sampling and analytical techniques
employed, and a summary of the data
obtained. In addition, the pollutant
emissions for the study region were
projected for those power plants
presently operating and those which will
come on line by the end of 1983.
Site Selection
The EPA had determined that the study
area would consist of the stretch of the
Ohio River from Portsmouth, Ohio, to
Louisville, Kentucky. This area was
chosen because it probably represents
the most active region of coal-fired power
plants in the country. This region has 45
coal-fired power plant boilers in operation
-------
and another 17 units either under con-
struction or planned. Several common
characteristics were found when boilers
were grouped according to the following
age categories:
Group I:
25 years or older
Group II: 10 to 25 years old
Group III: less than 10 years old
Further, the relative size of the units
in these age categories was similar.
Group I ranged from 69 to 215 MW (113
MW average); Group II from 150 to 225
MW (192 MW average); and Group III
from 277 to 610 MW (469 MW average).
Five potential test sites were identified.
One unit was selected from Group I and
two units each were selected from
Groups II and III. The unit selected from
Group I typifies boilers of that era, and the
four units selected as representative of
Groups II and III are typical of those size
categories. One unit from Group II and one
from Group III are equipped with control
devices for controlling paniculate emis-
sions only. The other two units in Groups
II and III are equipped with full paniculate
and S02 control devices. The lettering
system used to designate each plant that
participated in the project indicates the
order in which the plants were tested.
Plant A, which was selected from
Group III, has a rated nameplate
generating capacity of 560 MW and was
placed into service in 1970. This Babcock
and Wilcox unit has an opposed-fired
burner configuration and is equipped
with a Buell weighted wire electrostatic
precipitator (ESP) to control paniculate
emissions.
Plant B, which was selected from
Group I, has a rated nameplate
generating capacity of 125 MW and was
placed into service in 1954. This Babcock
and Wilcox unit has a front-fired burner
configuration and is equipped with a
retrofit Research Cottrell ESP installed in
1973 to control paniculate emissions.
Plant C, which was selected from
Group II, has a rated nameplate
generating capacity of 163 MW and was
placed into service in 1958. This
Combustion Engineering unit has a
tangential-fired burner configuration.
The paniculate emission control system
consists of two ESP's in series. The
newer retrofit Research Cottrell ESP was
installed in 1975.
Plant D, from Group III, has a rated
nameplate generating capacity of 411
MW and was placed into service in 1962.
This Babcock and Wilcox unit has an
opposed-fired burner configuration. The
air pollution control equipment consists
of an American Air Filter (AAF) rigid
frame ESP that was installed in 1978.
After passing through the ESP, the flue
gas enters a carbide lime mobile bed flue
gas desulfurization (FGD) system, which
was also installed in 1978 by AAF.
Plant E, from Group II, has a rated
nameplate generating capacity of 156
MW and was placed into sevice in 1962.
This Combustion Engineering unit has a
horizontal-fired burner configuration.
The air pollution control equipment
consists of a Research Cottrell weighted
wire ESP installed in 1962. After passing
through the ESP, the flue gas enters an
AAF lime slurry FGD system, which was
installed in 1976.
Sampling Methods
Both manual and continuous
sampling methods were used to obtain
emission data for the various pollutants.
During on-site testing, the power plants
operated in a normal manner with the
loads changing according to demand.
Tests were conducted over at least a 5-
day sampling period in an effort to obtain
emission data under a number of
operating conditions. Manual sampling
was performed at the outlets of the air
pollution control devices only.
Particulate emissions were
determined by EPA Method 5. Samples
were collected at a filter and probe
temperature of approximately 120°C.
Filterable particulates consist of the
material caught in the probe and filter
portion of the EPA Method 5 sampling
train. The material (condensibles)
captured in the impinger section of the
sampling train was not included. Stack
gas flow rates determined during the
Method 5 tests were used to calculate the
mass emission rates of all the pollutants
sampled.
Particle size distributions were
determined with an Andersen 2000 Mark
III in-stack cascade impactor. This
impactor has eight stages and a backup
filter, with particle size cutoffs ranging
from 0.5 to 15 /urn. A standard EPA
Method 5 control module was used to
maintain isokinetic sampling conditions.
The Andersen impactor was also used to
obtain paniculate samples in various
particle size fractions suitable for
elemental analysis by X-ray fluorescence
(XRF). To determine the particle size
distributions, the impactor was run with
all eight stages plus a glass fiber backup
filter. To obtain samples for elemental
analysis, the impactor was assemblec
with a Zefluor filter and a sample on e
substrate suitable for XRF analysis. Al
particle size samples were obtained at <.
single sampling point located in the stacl
at a point of average gas velocity.
The Controlled Condensation Systerr
(CCS) was utilized to simultaneously
collect and differentiate paniculate
sulfate, chloride, fluoride, sulfur trioxide
(SO3 as H2SO4), sulfur dioxide (SO2),
gaseous hydrogen chloride (HCI), anc
hydrogen fluoride (HF). The CCS
sampling train consists of a heated Vycoi
probe, a heated quartz filter that collects
the paniculate matter, a modifiec
Graham condenser that cools the gas
below the acid dewpoint to collect the
SO3 (H2SO4) vapor, and impingers thai
scrub the SO2, HCI, and HF from the flue
gas sample. Samples were obtained at a
single point located approximately one
meter from the stack wall.
Summary of Results
The primary purpose of this study is tc
provide a comprehensive characteriza-
tion of source emissions for powei
plants presently operating in the Ohi<
River Valley and for those that will come
on line by the end of 1983. The task was
accomplished by testing a selected group
of plants and developing emission factors
for specific paniculate (Table 1) anc
gaseous (Table 2) pollutants.
Particle size distribution (PSD) samples
were taken with an Andersen Mark III in
stack cascade impactor. All particle size
samples were obtained at a single
sampling point located in the stack at s
point of average velocity. Particle size
distribution samples were rur
simultaneously with the EPA Method E
tests. The impactor was not heatec
during the tests at Plants A, B, and C; ii
was heated at Plants D and E to prevem
moisture from condensing in the
impactor. The flow rate through the
impactor was maintained in the optimurr
range for all the sample runs.
The particle size distributions wen
examined for each plant tested to deter
mine what percentage of paniculate
emissions fell into the range of inhalable
and fine particulates. The inhalable par
ticulate (IP) range includes those particle;
less than 15 //m in diameter. The fine
paniculate (FP) range includes particles
Iessthan2.5/ym in diameter. Because the
cut-points of the different stages of ar
Andersen Mark III impactor do not include
15 fjm and 2.5 /jm, a computer prograrr
was used to calculate and extrapolate
these values. The different loads at whicl
-------
Table 1 Paniculate Emission Factors
Type of
Group Plant control
1 B ESP
II C ESP
II £ ESP and SO2
scrubber
III A ESP
ESP and SO2
III D scrubber
* Percentage of the paniculate
t Percentage of the paniculate
Table 2. Gaseous Emission
Type of
Group Plant control
1 B ESP
II C ESP
II E ESP and SO2
scrubber
III A ESP
III D ESP and SO2
scrubber
Total
paniculate
(mg/MJ)
33.2
490.0
25.8
85.0
40.7
emissions less than
emissions less than
Factors (mg/MJ)
SO2*
892
735
576
683
905
Inhalable
paniculate
range*
76.0
49.2
71.4
70.9
71.9
15 um in diameter.
2.5 um in diameter.
#2SO4
4.8
4.8
16.1
2.4
19.3
Fine
paniculate
range\
34.2
13.7
39.8
18.5
40.7
NO2
9.2
10.0
5.4
11.8
5.3
Paniculate
chlorides
(mg/MJ)
0.04
0.04
0.04
0.03
0.04
/VO*
278
214
355
508
802
Paniculate
fluorides
(mg/MJ)
0.04
O.O4
0.04
0.06
0.04
ttC/f
32.3
41.2
0.20
43.7
0.82
Paniculate
su/fates
(mg/MJ)
0.09
0.08
0.13
0.07
0.61
HF
3.0
3.3
0.10
3.8
0.09
* SO2 emission factors dependent on sulfur content of coal.
t HCI emission factors dependent on chlorine content of coal and type of emission controls.
Table 3. Percentage of Inhalable and Fine
Paniculate
Inhalable panic- Fine paniculate
ulates (<15 um), (<2.5 urn).
Plant percent percent
A
B
C
D
E
70.9
76. 0
49.2
71.9
71.4
18.5
34.2
13.7
40.7
39.8
the plants operated during testing did
not significantly affect the particle size
distributions. All PSD runs for each sepa-
rate plant were grouped together and the
average mean cumulative mass concen-
tration was determined for each unit. The
computer program extrapolates the best-
fit curve for the data supplied and
removes outliers. Table 3 presents a
comparison of the amount of paniculate
being emitted in the inhalable and fine
particulate ranges at each plant.
At each plant tested particle size
samples were run with the Andersen
Mark III impactor by varying the number
of filter stages directly in front of the
Zefluor backup filter. Four different
impactor sampling configurations were
used during the sampling period, with
two, four, six, or eight impactor stages
placed before the Zefluor backup filter.
The Zefluor backup filters used for the
different particle sizing runs were
submitted to NEA Laboratories for
elemental analysis. The filters for Plants
A, B, and C were analyzed for 22
elements by neutron activation analysis
(NAA) and X-ray fluorescence (XRF). The
filters for Plants D and E were analyzed
for these 22 elements and 4 additional
elements, P, Ga, Rb, and Sr. Only two
elements, CdandBa, were not detectable
at any concentration in the samples
analyzed.
The percentage of the total
concentration for each element was
determined. For Plants A, B, and C, the
matrix elements Si, Al, Mg, Fe, and K
comprised between 90 and 95 percent of
the total elemental concentrations in
most cases. For Plants D and E, the
primary elements detected were Al, Si, P,
S, Ca, and Fe, which comprised between
93 and 97 percent of the total elemental
concentration.
-------
Gary Baker, Paul Clarke, Richard Gerstle, Wade Mason, and Mark Phillips are
with PEDCo Environmental, Inc., Cincinnati, OH 45246.
Kenneth Knapp is the EPA Project Officer (see below).
The complete report, entitled "Emission Characterization of Major Fossil Fuel
Power Plants in the Ohio River Valley." (Order No. PB 84-120 328; Cost:
$22.00, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Sciences Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
AUS GOVERNMENT PRINTING OFFICE 1984-759-015/7266
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
BULK RAT
U.S. POSTA
PAID
Cincinnati, 0
Permit No. G
Official Business
Penalty for Private Use $300
------- |