SEPA
United States
Environmental Protection
Agency
Environmental Sciences Research
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S7-81-092 July 1981
Project Summary
Secondary Formation
Products in Power
Plant Plumes
W. D. Balfour, L 0. Edwards, and G. K. Tannahill
Source and airborne measurements
were made in October 1979, at the
TV A Widows Creek Steam Plant in
northeastern Alabama. Primary emis-
sions from the flue gas stacks were
determined including total mass, par-
ticle size distribution, SO2, NO», and
SOa/H2SO4. An instrumented fixed
wing airplane was used to acquire
plume data at various altitudes and
downwind distances. These data were
used to establish secondary pollutant
formation rates.
Sulfate transformation rates were
measured between 4.3 percent hr~1
and 0.6 percent hr~1 for a well-defined
plume at 1.1 hr and 2.2 hr plume ages
respectively. A chemical element
balance around the scrubber indicates
that as much as 40% of the total mass
exiting the scrubber was generated
from entrained liquor. This accounts
for 90% of the unit's sulfate emissions.
These primary emissions account for
between 4 and 17% of the total surfates
measured downwind in the plume.
This Project Summary was devel-
oped by EPA's Environmental Sciences
Research Laboratory, Research Tri-
angle 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
ordering information at back).
Introduction
The production of primary and sec-
ondary pollutants has been studied
extensively in plumes of unscrubbed
power plants, but only preliminary work
has been done to determine effects of
SO: scrubbers on downwind plume
chemistry. There has been increased
political pressure to reduce ambient
sulfate levels by reducing sulfur dioxide
emissions, the precursor to secondary
sulfate particles, from coal-fired power
plants. The principal strategies for
reducing SOz emissions are the conver-
sion to low sulfur coal or the installation
of wet or dry SOz scrubbers in the plant
stacks. Since the use of scrubbers
allows burning higher sulfur coal, SOz
scrubbing is likely to become more
common in the future, as utilities use
the more commonly available and less
expensive higher sulfur coal.
In the past five years several studies
have been conducted to determine the
rate of sulfur dioxide to sulfate aerosol
conversion in coal-fired power plant
plumes. These studies indicated a
conversion rate between 0 and 10% hr"1
for unscrubbed plants.
There is still uncertainty as to the
formation mechanisms involved in the
gas-to-particle conversion. Some studies
obtained good correlation between
formation rates and solar radiation,
indicating the importance of gas phase
free radical reactions. One of the main
mechanisms for S02 oxidation, the
reaction with hydroxyl radicals, the rate
having been established in the range
0.4 - 2.7% hr'1. Since the OH radical is
formed photochemically, the conversion
rate should show some dependence on
solar intensity.
-------
S02 oxidation may occur also on
particle surfaces or in water droplets.
Liquid phase oxidation may predominate
under cloudy or high humidity condi-
tions. Furthermore, metal salts in fly ash
act as catalysts for S02 oxidation in
water droplets. Therefore, the oxidation
rate may be affected by any droplets
produced by the scrubber and emitted
from the stack.
Thus, the formation mechanism and
the conversion rate in a scrubbed power
plant plume may be different from those
in an unscrubbed plume. This study
addresses the questions of the nature of
the primary emissions and the gas-to-
particle conversion rates determined for
a utility boiler equipped with a wet S02
scrubber.
To answer these questions, concur-
rent source and plume measurements
were made. Source characterization
measurements that were made included
the following: EPA Method 5 for total
mass, in-stack cascade impactor runs
for particle size distribution, controlled
condensation measurements for S03/
H2S04, continuous instrumental moni-
toring for S02, NOx/NO and 02, and
collection of scrubber slurry and coal
samples for elemental analysis.
The airborne measurements utilized
the Pennsylvania State University Aero-
commander 680E airplane equipped as
follows: Environment One condensa-
tion nuclei counter, Thermo Systems
3030 Electrical Aerosol Analyzer, MRI
1550B integrating nephelometer, TECO
Model 43 SO2 analyzer, Meloy OA350
ozone monitor, and TECO Model 14D-E
NO/NOx analyzer. In addition, temper-
ature, dew point, aircraft position,
altitude and navigation were monitored
routinely.
Two types of filters were used to
sample plume particulates. Pretreated
quartz filters (Gelman Micro-Quartz)
were used with a high-volume air
sampler to obtain aersol samples for
sulfate and cation analysis by ion chroma-
tography and inductively coupled argon
plasma emission spectrometry. A tandem
filter pack was used to obtain nitrate and
nitric acid samples. The first filter was
Teflon followed by a nylon filter (Ghia
Corp.).
Results
Source Characterizations
The TVA Widows Creek Steam Plant is
located 20 miles northeast of Scottsboro,
Alabama and 30 miles southwest of
Chattanooga, Tennessee. The plant has
six wall-fired 135MW boilers (Units 1 -6)
ducted into a single 1000 foot stack.
Two additional tangentially fired 557MW
boilers (Units 7 and 8) are each ducted
into individual 500 foot stacks. Each
unit, except Unit 7, is equipped with an
electrostatic precipitator with a design
efficiency of 99.5 percent for fly ash
removal. The precipitator on Unit 7 has a
design efficiency of 90 percent.
Ammonia (10 ppm) is injected into the
flue gas to enhance collection efficiency
for Unit 7. A limestone wet scrubber has
been retrofitted downstream of the
precipitator on Unit 8; the unit has a
design efficiency of 80 percent for S02
removal. Units 1 -6 burn a 0.8% sulfur
coal, Unit 7 a 2.6% sulfur coal and Unit 8
a 4.1% sulfur coal.
A summary of the source character-
ization data is listed in Table 1. More
extensive listings are available in the
full report. Table 1 lists the range of
values obtained during the testing pro-
gram. The Project Report lists resultson
a daily basis. Inlet and outlet measure-
ments across the scrubber or Unit 8 al-
lowed a mass balance. Pertinent data is
listed in Table 2. The scrubber can con-
tribute mass through entrainment of
scrubber liquor high in suspended or
dissolved solids. This mass contribution
can be calculated using the concept of
chemical element balance. If one
assumes that mass flow of species j
from the scrubber, 0°ut, is composed of
material from the fly ash and scrubber
liquor entrainment, then the mass
Table 1. Summary on Source Characterization Data
Units 1-6
Unit 7
UnitS
Unit 8B
(Inlet)
EPA method 5 (gm/m3)
80s + 803 (ppm)
Chemical Composition
(% of a sample's total mass)
so;
Ca++
AI+++
/Wo++
T/***
Particle size distribution
(% Mass less than 2 iim)
Controlled Condensation
HaSOt/SOa (ppm)
SOa (ppm)
1.52-3.20
1663-1882
0.002-0.007
0.370-0.933
0.268-1.208
1.54-4.63
0.091-0.318
0.075-0.247
7.9-15.3
0.047-1.18
1568-2242
0.009-0.021
113-270
<0.001
25.70-26.20
3.49-7.32
3.95-4.13
0.349-0.478
0.400-0.451
20.7-53.2
0.34-0.92
258-269
5.61
2051
0.003
0.812
0.660
2.28
0.106
0.136
10.9
2.51-2.69
2135-6300
Instrumental Gas Analysis
SOt (ppm)
NO (ppm)
NOz (ppm)
Oz (ppm)
383-408
246-262
0-19
10.5-11.5
1543-1901
138-191
4-13
7.2-8.5
212-242
75-82
13.5-13.5
—
-------
Table 2. Mass Balance Across S02 Scrubber Unit-8
Total Ca** Mg** f Al*
so;
Mass Flow
Into Scrubber
(g/min), 0,ln 188,500 4090 660 19 14.743 1056 5090
Mass Flow Out
of Scrubber
(g/min), 0,°"* 314 20.5 1.42 0.056 12.9 1.75 64.8
Scrubber
concentration
(g/l),C — 0.700 0.140 <0.001 0.003 0.001 1.725
balance can be written as:
0r=p*r+c,v
where \ = mass flow (g/min) of spe-
cies j out of scrubber and
into the stack.
P = penetration fraction of
species j through scrub-
ber
\n- mass flow (g/min) of
species j into scrubber
Cj = concentration of j in
scrubber liquor (g/l)
V = volume rate of scrubber
liquor entrained (l/min)
A multiple variable regression analysis
was performed on a set of six equations
(one each for Ca, Mg, Al, F, Ti, 804) with
three independent (tf"', 0jln and Q) and
two dependent (V and P) variables. The
resultant values that best fit the data
were:
V= 34.3 ± 2.8 1/min
P = 0.00097± 0.00035
The results indicate fly ash penetra-
tion through the scrubber is small,
perhaps 0.1 percent.
Based on the above values, the mass
flows of each species as well as the total
can be calculated. Table 3 summarizes
these results.
The predicted mass flow representing
primary fly ash penetration is 183
gm/min. If the scrubber liquor mass
flow is calculated by difference, the
contribution is 131 gm/min. When the
contribution is calculated by summing
the individual species contribution, the
Table 3. Calculated Mass Flow Rate from Scrubber Unit
Total Ca** Mg** F~ Al*
Ti*
so;
Mass from
Penetration
(gm/min)
Mass from
Entrainment
(gm/min)
183
>86.6*
3.97
23.98
0.64
3.36
0.018
0.34
14.30
0.10
1.02
0.024
4.94
59.10
*This is the sum of the, species measured.
Table 4.
Calculated Sulfate Formation Rates
Flight # (SOVSO2 + SOJxIOO
5 4.75
1.27
6 14.6
2.19
7 16.4
8.84
9 5.59
45
11 —
0.47
Plume Age (hrs)
1.1
2.2
0.6
1.2
0.5
0.6
0.2
0.5
0.2
0.5
Rate
Percent /»r~1
4.3
0.6
24.3
1.8
32.7
14.7
28
90
<0.3
0.9
The levels of nitrate did not occur at values significantly
above background
result is 110 gm/min. Agreement (15
percent) is well within the expected
experimental error.
From this analysis it can be concluded
that of the paniculate emissions from
Unit 8,40 percent are contributed by the
scrubber itself (131 gm/min 4- 131 +
183 gm/min). Also, greater than 90
percent of the sulfate emissions are
generated by the scrubber (59.1 -=-64.0).
Further analysis shows that the
entrained mass from the liquor is mainly
dissolved solids. This is evidenced by
the sulfate-aluminum-calcium ratios in
the participate matter, which is similar
to the solids dissolved and different
from suspended solids ratio. Thus it can
be expected that the scrubber generated
mass will contribute to the fine particle
size fraction.
Plume Characterizations
Plume characterizations were based
on results of plume cross-sectional
analyses downwind of the plant. The
distance was related to the travel time
or plume age based on wind speed
information furnished by the pilot balloon
operators.
A summary of the calculated sulfate
transformation rates is given in Table 4.
The results were determined as follows:
the sulfate values used in the numerator
were obtained from hi-vol samples and
were corrected for ambient background
values. They were expressed as equiva-
lent S02 volumetric concentrations. The
SOa values in the denominator were the
result of averaging measured SO: data
(one observation per 0.5 second) over
the period of hi-vol sampling. Plume
ages were based on the downwind
distance and wind speed.
Thus, no nitrate transformation rates
could be determined. Based on limita-
tions of sampling time and detection
limit, the minimum transformation rate
detectable for nitrate is 12 percent hr~1
Similar calculations for sulfate give a
value of 0.3 percent hr~1.
Several values listed in Table 4 are
suspiciously high. The plant is located in
a valley running approximately north to
south. The prevailing wind during the
study was from the west. Thus, after a
brief travel time the plume impinged on
the escarpment to the east. If significant
S02 deposition occurred, then the de-
nominator in the calculations would be
erroneously low and the resultant
transformation rates high. Many values
calculated were an order of magnitude
higher than the typical 1 -10 percent hr~1
reported by other workers.
US GOVERNMENT PRINTING OFFICE-1981 -757-012/7208
-------
Conclusions and
Recommendations
The primary findings from this study
are presented below.
The venturi scrubber besides being
any effective flue gas desulfurization
device (>90% removal) provided a high
degree of control for particulates (>99.8%
removal). Efficiency was lower for parti-
cles less than 2 /urn in diameter.
As a result of the scrubber collection
characteristics the emissions from the
scrubbed stack exhibit a higher mass
fraction of respirable particles (0.5) than
from the unscrubbed unit (0.1). In terms
of absolute emissions, the scrubbed
stack in an order of magnitude less than
the unscrubbed for particles less than 2
yum.
The scrubber tested did not efficiently
remove SOa/HzSC^ from the flue gas.
Analysis of the data showed that 40%
of the total mass exiting the scrubber is
generated from entrained liquor. This
accounts for 90% of the sulfate emis-
sions from the scrubber stack.
Sulfates emitted directly from the
thru stacks account for between 4 and
17% of the total sulfates measured
downwind.
Sulfate transformation rates were
measured between 4.3% hr~1 and 0.6%
hr~1 for well-defined plumes. Corrections
for primary sulfate emissions were not
significant.
Nitrate transformation rates could not
be measured directly and are estimated
to be less than 12%hr~1.
As a result of this study, recommen-
dations can be made in two areas: (1)
development of sampling and analytical
techniques and (2) considerations when
selecting a site for plume studies.
In the sampling techniques there is a
need for dependable real time contin-'
uous analyzers for sulfate and nitrate
aerosols.
Site selection requires consideration
of several parameters: history of the
unit, constant load operation of the
boilers, stable fuel source, preferably
the same for all units, and local terrain
effects.
This Project Summary was authored by H. M. Barnes, who is also the EPA
Project Officer (see below).
The complete report, entitled "Secondary Formation Products in Power Plant
Plumes," was authored by W. D. Balfour, L. O. Edwards, and G. K. Tannahill
of Radian Corporation, Austin, TX.
The above report (Order No. PB81-199 390; Cost: $9.50, 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
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
CHICAGO It. 60604
------- |