EPA/600/7-85/005
February 1985
FLUE GAS CONDITIONING
by
Shui-Chow Yung, Ronald G. Patterson,
Benjamin L. Hancock and Seymour Calvert
A.P.T., Inc.
5191 Santa Fe Street
San Diego, CA 92109
EPA Contract No. 68-02-2628
EPA Project Officer: Leslie E. Sparks
Air and Energy Engineering Research Laboratory
Office of Environmental Engineering and Technology
Research Triangle Park, NC 27711
AIR AND ENERGY ENGINEERING RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NC 27711
-------
TECHNICAL REPORT DATA
fPh asi rraJ luuiuriHun on the revent be fore completing/
1 REPORT NO
EPA/600/7-85/005
2
3 -RECIPIENT'S ACCESS^** NO.
PBS 5 I 739 1 2 /IS
J title and subtitle
Flue Gas Conditioning
& REPORT DATE
February 1985
£. PERFORMING ORGANIZATION CODE
7 AUTHORIS)
Shui-Chow Yung, Ronald G. Patterson, Benjamin L.
Hancock, and Seymour Calvert
8 PERFORMING ORGANIZATION REPORT NO
9 PERFORMING ORGANIZATION NAME AND ADDRESS
Air Pollution Technology, Inc.
5191 Santa Fe Street
San Diego, California 92109
io program element no.
11. CONTRACT/GRANT NO
68-02-2628
12. SPONSORING AGENCY name and address
EPA, Office of Research and Development
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND F*E R1OD COVERED
Final; 3/79 - 3/82
14. SPONSORING AGENCY CODE
EPA/600/13
IS supplementary notes aeerl project officer is Leslie E. Sparks. Mail Drop 61. 919/
541-2458.
i6 abstract The report gives results of a survey of available flue gas conditioning
agents and user experience. Many existing chemicals have been used as conditioning
agents in power plants or have been studied in the laboratory as potential agents.
The particle collection efficiency of an electrostatic precipitator (ESP) for coal-fired
power plant flue gas cleaning depends on the electrical properties of the fly ash,
among other things. Flue gas conditioning refers to the addition of chemicals to the
flue gas for modification of fly ash properties and/or electrical conditions in the
ESP to improve the collection efficiency of ESPs. Conditioning is usually used to
upgrade existing ESPs.
17. KEY won OS ANO DOCUMENT ANALYSIS
¦*. DESCRIPTORS
b. IDENTIFIERS/OPEN ENOED TERMS
i. COSATi 1 it'Id Croup
Pollution Dust
Flue Gases
Treatment
Fly Ash
Electrostatic Precipitators
Particles
Pollution Control
Stationary Sources
Conditioning Agents
Particulate
13 B 11G
21B |
14 G
131
13. UlSTRifaUTlON STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21 NO OF PAGES
iao
20 SECURITY Class (This page)
Unclassified
77 PRICE
EPA Form 2220-1 (t-73)
i
-------
NOTICE
This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation for use.
ii
-------
ABSTRACT
The particle collection efficiency of an electrostatic
precipitator (ESP) for coal-fired power plant flue gas cleaning
depends on the electrical properties of the fly ash/ among other
things. Flue gas conditioning refers to the addition of
chemicals to the flue gas for modification of fly ash properties
and/or electrical conditions in the ESP and thus improve the
collection efficiency of ESP's. It is usually used for upgrading
existing ESP's.
Many existing chemicals have been used as conditioning
agents in power plants or have been studied in the laboratory as
a potential agent. This report presents the results of a survey
of available agents and user experience.
iii
-------
Page Intentionally Blank
-------
TABLE OF CONTESTS
Page
Abstract iii
Figures v
Tables Vi
Acknowledgment viii
1 SUMMARY AND CONCLUSIONS 1
2 INTRODUCTION 17
3 SULFUR TRIOXIDE CONDITIONING 22
4 AMMONIA CONDITIONING 47
5 AMMONIUM COMPOUNDS 6 9
6 ORGANIC AMINES 85
7 ALKALI CONDITIONING 89
REFERENCES 95
LIST OF FIGURES
Paye
Figure 1. Resistivity as a function of the 32
concentration of injected sulfur trioxide
Figure 2. Ash resistivity versus triethylamine 87
concentration
Figure 3. Fractional efficiency for three performance 93
tests
-------
1.
2.
3.
4.
5.
6.
7.
8.
9.
10
11
12
13
14
15
16
17
18
19
20
21
LIST OF TABLES
Page
Possible Mechanisms Whereby a Conditioning 3
Agent May Affect Precipitability
Flue Gas Conditioning Agents and 5
Manufacturers
List of Past and Present Flue Gas Conditioning 8
Users
Properties of Sulfur Trioxide 23
Equilibrium Partition of Sulfur Oxides 24
Ash Resistivities for Users of SOs Conditioning 28
Ash Resistivities for Several Users of SOi 29
Conditioning
Field Determinations of Acid Partition In Flue 35
Gas Conditioning
Changes in Particle Sulfate Composition with 36
Sulfur Trioxide Conditioning
List of Past and Present Users of Sulfur 39
Trioxide Conditioning
Estimated Costs of a Sulfur Burning SO1 Flue 44
Gas Conditioning
ESP Performance for Several Selected Users 45
of SOi Conditioning
Properties of Ammonia 48
ESP Performance with Ammonia Conditioning 51
ESP Performance with Ammonia Conditioning 56
Partition of Added Ammonia Between Gas 58
and Particles
Changes in Particle Composition Due To 60
Ammonia Conditioning
Change of Particle Number Concentration with 61
Ammonia Conditioning
Economics of Ammonia Conditioning 63
List of Past and Present Users of Ammonia 65
Properties of Ammonium Sulfate Compounds 71
vi
-------
LIST OF TABLES (Cont'd.)
Table 22. Equilibrium Partial Pressures of Ammonia 72
and Sulfuric Acid in Equilibrium with
Ammonium Salts at Various Temperatures
Table 23. Weight Losses by Volatilization of Ammonium 72
Sulfate or Ammonium Bisulfate at a Heating
Rate of 4.0°C/MIN
Table 24. ESP Conditioning with Ammoniura-Sulfate 75
Compounds
Table 25. ESP Performance with Proprietary Compounds 77
Table 26. Fate of Ammonium Compound Conditioning Agents 82
Table 27. List of Past and Present Users of Ammonium 84
Compound and Proprietary Compounds
Table 28. ESP Performance with Sodium Conditioning 92
vii
-------
ACKNOWLEDGMENT
The following people supplied information for this report:
Danny Douglas
Nicole Letendre
Mary Roman
Eric Goerss
Vincent Albanese
James M. Brennan
Harold Fletcher
Charlen Spooner
Tom Hale
R. E. Wilbur
Timothy Gustafson
Jack Roehr
R. J. Mudd
Dr. Ralph Altman
Mr. Steve Biro
Jim Tice
Tim McKenzie
John Tytle
H.G. Brines
Robert Bisha
Frank Miholis
Jacob Katz
Dr. E. B. Dismukes
William R. Elliott
Florida Power Co.
New England Power Svc. Co.
Pacific Power & Light
Ohio Edison
Nalco
Detroit Edison
Iowa Public Service Co.
Colorado-Ute Electric Assoc.
Colorado Springs/ DPU
Wahlco Inc.
EPRI
Cleveland Electric Alluminating
Pennsylvania Electric Co.
Tennessee Valley Authority
Public Svc. Co. Colorado
Central Illinois Light
Commonwealth Edison
Precipitator Technology, Inc.
Southern Research Instittue
Northern Indiana Public Svc. Co.
viii
-------
Section 1
SUMMARY AND CONCLUSIONS
FLOE GAS CONDITIONING
Flue gas conditioning refers to the addition of chemicals to
the flue gas in a coal-fired power plant for modification of fly
ash properties and/or electrical conditions in the electrostatic
precipitator (ESP) and thus improve the collection efficiency of
ESP's. It is usually used to retrofit existing ESP's whose
performance has deteriorated/ or which are operating below design
efficiency.
The collection of fly ash in an ESP involves the
precipitation of the ash followed by its successful removal:
first from the collection plates and then the hoppers. For a
precipitator of given size and operating under fixed conditions#
the collection efficiency of the ESP is affected by the following
parameters:
1. the electric field strength and ion density in the
precipitation zone,
2. the adhesive and cohesive properties of the fly ash/ and
3. the average particle size and size distribution.
A conditioning agent may operate by affecting some or all of
these factors. The ash resistivity is important because it can
affect both "1" and "2".
CONDITIONING MECHANISMS
A conditioning agent may influence the ESP collection
efficiency through one or more of the following mechanisms:
1. Adsorbs on surface of fly ash and reduces surface
resistivity.
2. Adsorbs on fly ash and changes the adhesion and cohesion
properties of ash.
3. Increase ultrafine particle concentrations for space charge
enhancement.
4. Increase electrical breakdown strength of flue gas.
5. Increase the mean particle size.
1
-------
6. Change the acid dew point in the flue gas.
The effects of these mechanisms on ESP performance are
described in Table 1.
FLUE GAS CONDITIONING AGENTS
Many chemicals have been used as conditioning agents in
power plants or have been studied in the laboratory as a
potential conditioning agent. Table 2 shows a list of these
chemicals and their principal conditioning mechanisms.
Sulfur Trioiide
Sulfur trioxide (which readily becomes sulfuric acid when
water is present) is the most widely used conditioning agent in
the U.S. It is a natural component of flue gas from fossil fuel
combustion. It is hygroscopic and has a very low volatility so
it condenses easily. When adsorbed or condensed on the ash, it
forms a layer of conductive solution on the ash surface to reduce
the ash resistivity.
For conditioning, SOj is produced by one of the following
four processes:
1. vaporization of a sulfuric acid solution,
2. vaporization of liquid sulfur trioxide,
3. vaporization of liquid sulfur dioxide and oxidization to
sulfur trioxide over a vanadium pentoxide catalyst.
4. burning liquid sulfur in air to produce sulfur dioxide and
then oxidize to SOj.
Amnion i n
Ammonia is a vapor at room conditions and has a critical
temperature of 132°C. Above this temperature ammonia exists as a
single phase and cannot boil or condense. Therefore,
condensation on the fly ash would not be expected to occur with
ammonia above 132°C, although adsorption could take place.
Because of the high volatility of ammonia, it is injected in
the vapor form under its own vapor pressure. In Australia,
ammonia is also injected in solution form.
2
-------
TABLE 1. POSSIBLE MECHANISMS WHEREBY A CONDITIONING AGENT MAY
AFFECT PRECIPITARILITY (CASTLE, 1900)
Effect of
Conditioning Agent
Electrical Result
Mechanical Result
Effect on
Ef f iclency
Comments
Adsorbs on surface
of fly ash and
reduces surface
resistivity
olncreases the
nagnltude of the
precipitation
field
oReduces the voltage
drop In the dust
layer
oDelays the onset
of back corona
olncreases the
sparkover voltage
+ +
Useful for high resistivity
dusts:
oIncreases charging and
precipitation field strength
oReduces the electrical
adhesion on the wall and thus
Improves the effectiveness of
rapping
~ Keauces tne
electrical ad-
hesion effect on
the wall
* or -
Beneficial for high resistivity
dusts. If used with low or medium
resistivity dusts, further lowering
of adhesion forces could lead to
reentrainment losses.
Adsorbs on fly ash
and changes
coheslveness or
'stickiness*
Aids agglomeration
and Increases mean
particle size
+
Size enhancement may occur Indepen-
dently of resistivity change and
thus improve migration velocity.
oLarger size fraction also aids
removal by raopina
Dust layer on wall
becomes more cohesive
oCohesive dust layer tends to shear
off collecting plate with less
re-entrainment losses
Dust layer has
stronger adhesion to
wall
+ or -
oStronger adhesion is an advantaqc
for low resistivity dusts
oCould be a disadvantage for high
resistivity dusts
Increases particle
concentration due
to presence of
•fines" (i.e.,
particulate
reaction products)
Reduces ion density
(and thus current)
due to space charge
suppression
+ or -
The current reduction could reduce
charging effectiveness
On the other hand, the lower current
density will alleviate field re-
duction problems caused by the
voltage drop through a high resistance
dust laver
olncreases collection
field strength due
to space charge
enhancement
+
Space charge Increases the field
strength near the collecting electrode
olncreases sparkover
voltage
+
A slight Increase in sparkover
voltage usually results from Increased
space charge
-------
TABLE 1 continued
Effect of
Conditioning Agent
Electrical ReGult
Mechanical Result
Effect on
Efficiency
Comment
Increases electrical
breakdown strength
of flue gas
Increases the magnitude
of the precipitator
field:
olncreases sparkover
voltage
oDelays onset of
back corona
~ +
The breakdown characteristics of flue
gases are very sensitive to minor
concentrations of electro-negative
species and to surface conditions of
the dust layer. This can be Independent
of fly ash resistivity.
Neutralization of
acid in flue gas
Decreases acid dew-
point. This reduces
surface "tracking"
on hi-gh voltage
insulators a 1 lowing
higher voltages to
be applied
~ +
With some high sulfur coalSr the
sulfuric acid concentration in the
flue gas is so high that the acid
dew point may be above the flue gas
temperature. This nay result in acid
condensation on support Insulators.
KFY ~ ~ Indicates'strong tendency to increase efficiency.
~ Indicates tendency to increase efficiency.
- Indicates tendency to decrease efficiency.
Ref.: Castle, G. S. P. "Mechanisms Involved in Fly Ash Precipitation in the Presence of Conditioning Agents
A Review." IEEE Trans. Ind. Appl. FA-I6: 297-302 (1980).
-------
TABLE 2. FLOE GAS CONDITIONING AGENTS
AND CONDITIONING MECHANISMS
Condition
SO 3 r (HiSO *)
Resistivity modification
NH 3
Adhesion and cohesion
improvement
Space charge enhancement
Ammonium Compounds
(SOiOH)NHj (sulfamic acid)
Space charge enhancement
Resistivity modification
(NHJ 2S0»
NH»HSO»
(NH 2)jCO
(NHJ zHPO*
Organic Amines
(CH aCH?)jN (Triethylamine) Resistivity modification
(CHj)jN (Trimethylamine)
(CeH 11)NH, (Cyclohexylamine)
Alkali Compounds Resistivity modification
Na iSO *
Na ?CO j
Proprietary Compounds
Apollo LPA-30
Apollo LPA-40
Apollo LPA-50
Koppers "K"
Resistivity modification
Space charge enhancement
5
-------
nniiim Compounds
Conditioning with ammonium compounds offers a more
convenient method of injecting ammonia. The commonly used
ammonium compounds are sulfamic acid/ ammonium sulfate/ and
ammonium bisulfate. If injected upstream of the air pre-heater,
these compounds could decompose to ammonia and sulfuric acid and
may provide a combination of the effects of ammonia and sulfuric
acid conditioning.
Organic amines have been studied in the laboratory and pilot
scale ESP's as possible flue gas conditioning agents. Currently/
there are no industrial users. Of all the amines/ triethy1amine
has received the most attention. It is an organic nitrogen
compound and is highly soluble in water. It behaves similar to
ammonia/ but is a substantially stronger base. The melting and
boiling points of triethy1 amine are -115°C and 90°C. It
primarily decomposes to ammonia/ hydrogen cyanide/ nitrogen
dioxide/ and nitric oxide at temperatures above 340°C (650°F).
DrY Alkali
When the SOi concentration is low or when the temperature is
above 200°C/ the ash resistivity is indirectly related to the
alkali metal content in the ash and reduction of ash resistivity
by increasing the alkali metal concentration has been tried. Of
the many alkali metal salts/ sodium salts are the most commonly
used conditioning agents because of their availability and
relatively low cost. The widely used sodium compounds are sodium
carbonate and sodium sulfate. Sodium chloride has been tried in
the laboratory and found to be effective. However/ it is not
used because it can lead to corrosion of metal equipment.
The mechanism for sodium conditioning depends on how the
sodium is applied. If a sodium compound is injected into the
boiler along with coal/ it will decompose and the sodium is bound
in the ash. The sodium will increase the conductivity and lower
the ash resistivity the same way as natural sodium.
6
-------
If the sodium compound is co-precipitated with the ash, the
compound trapped in the space between the particles on the dust
layer offers an additional conductive path for charge
dissipation.
Most of the proprietary chemicals are ammonium compounds
with minor additives, such as surface active agents. Therefore,
these chemicals perform similarly to ammonium compounds.
HiKCPllanpnn
Several metal oxides, such as iron oxide and vanadium oxide,
have been investigated as possible conditioning agents (Kanowski
and Coughlin, 1977). Iron and vanadium oxides are claimed to
catalyze the reaction of SO2 to S0» and thus increase the
quantity of SOj present in the flue gas. This claim has not been
substantiated in the literature.
RESULTS
Table 3 shows a list of past and present flue gas
conditioning users in the U.S.A. Summaries of user experiences
are presented in the following sections.
Sulfur TrinTiriP
SO 3 conditioning is limited to cold side ESP's. The roost
common injection location is between the air preheater and the
electrostatic precipitator inlet. The temperature at the point
of injection and in the precipitator should be above the sulfuric
acid dew point of the gas after addition.
The dosage for SOj injection normally is in the range of 5
to 30 ppmv; but can be as high as 70 ppmv. The required dosage
will depend on the composition of the ash surface, i.e., whether
it is acidic, neutral, or basic. If the ash has a large amount
of alkaline compounds, a higher dosage of SOj is needed because
7
-------
TABLE 3. I.IST OF PAST AND PRESENT FLUE CAS CONDITIONING USERS
fnmp«ny
Plant Nflmf
Boiler Number
FCC Licensor
FGC Aapnt
Alabama Power Co.
Barry
4
Wahlco
Sulfur
trloxlde
ACPSCO Appalachian Power
Cabin Creek
AEPSCO Appalachian Power
Kanawha River
ACPSCO Columbus t Southern Ohio
Electric
Conesville
4
Arizona Public Service Co.
Pour Corners
4
Apollo
LPA-445
Baltimore Gas t Electric
H. A. Wagner
3
Apollo
Central Illinois Uqht
Duck Creek
Apollo
Central Illinois Light
E. D. Edwards
1,2.3
Wahlco
Sulfur
trioxide
Central Illinois eight
R. S. Wallace
7,8,9,10
Wahlco
Sulfur
trloxlde
Cincinnati Gas & Electric Co.
W. C. Beckjord
1,2,4
Wahlco
Sulfur
trioxide
City of Colorado Springe DPU
Martin Drake
1,5
ReEearch-Cottrell
Sulfur
trioxide
Cleveland Electric Illuminating
Co.
Ashtabula
5
Wahlco
Sulfur
trioxide
Cleveland Electric Illunlnatlng Co.
Avon Lake
9
Wahlco
Sulfur
trloxlde
Cleveland Electric Illualnatlng
Co.
Eastlake
5
Wahlco
Sulfur
tr loxlde
Cleveland Electric Illunlnatlng
Co.
Lake Shore
18
Wahlco
Sulfur
trioxide
Colorado-Ute Electcic Assoc.
Hayden
1,2
Apollo
Comnonvealth Edison
Crawford
7,8
Wahlco
Sulfur
trioxide
Commonwealth Edison
Flsk
19
Wahlco
Sulfur
tr loxlde
Commonwealth Edison
Jollet
3,4,5,6
Wahlco
Sulfur
trloxlde
Commonwealth Edieon
Jollet
71,72,81,82
Comnonwealth Edison
Powerton
51,52
Wahlco
Sulfur
trioxide
Commonwealth Edison
Waukegan
IS,16,17,8
Wahlco
Sulfur
trloxlde
Commonwealth Edison
Will County
4
Wahlco
Sulfur
trioxide
Comnonwealth Edison/Indiana
State Line
1-1,1-2,1-3
Wahlco
S\ilf ur
trioxide
Comnonwealth Edison/Indiana
State Line
1-4,1-5,1-6
Commonwealth Edison/Indiana
State Line
2-1,2-2,2-3,3,4
Consumers Power Co.
B. C. Cobb
1,2,3,4,5
Wahlco
Sulfur
trioxide
Consumers Power Co.
J. C. Weadock
7,8
Wahlco
Sulfur
trioxide
Detroit Edison
Conners Creek
15,16
Wahlco
Sulfur
trioxide
Detroit Edison
Harbor Beach
1
Wahlco
Sulfur
trloxlde
Detroit Edison
Honroe
1,2
Wahlco
Sulfur
trloxlde
-------
TABLE 3. Continued
CaoBAax_HAige
Plant Name
Detroit Cdlson
Pennsalt
Detroit Edison
Port Huron
Detroit Edison
Trenton Channel
Duke Power
Belews Creek
Duke Power
Marshall
East Kentucky Rural Electric Power Coop.
W. C. Dale
Florida Power
Crystal River
Georgia Power
Harllee Branch
Gulf Power Co.
Scholc
Iowa Public 6ervlce Co.
G. W. Heal
Lansing Board of Water t Electric Light
Erlckson
Montana Power Co.
J. E. Corette
New England Power Co.
Sales Harbour
New England Power Co.
Brayton Point
New Jersey Gas t Electric
Mercer
New York State Electric t Gas
Goudey
New York State Electric 4 Gas
Greenldge
Northern Indiana Public Service Co.
D. H. Mitchell
Northern Indiana Public Service Co.
D. H. Mitchell
Ohio Edison Co.
Edgewater
Ohio Edison Co.
Gorge
Ohio Edison Co.
W. H. Saounls
Ohio Edison Co.
W. H. Sanguis
Pacific Power t Light
Brldger
Pacific Power t Light
Centralla
Pennsylvania Electric Co.
Front 6treet
Pennsylvania Electric Co.
Keystone
Pennsylvania Power t Light Co.
Brunner Island
Pennsylvania Power t Light Co.
Montour
Pennsylvania Power t Light Co.
Montour
Pennsylvania Power t Light Co.
Sunbury
Public Service Co. of Colorado
Arapahoe
Boiler Number FCC Licensor FGC
5
7,8,9A
1.2
UOP Sulfur trioxlde
Wahlco Sulfur trioxlde
Wahlco Sulfur trioxlde
Research-Cot tre 11 Sulfur trioxlde
3,4
2
3,4
Wahlco
Apollo, Nalco
Apollo
Sulfur trioxlde
2,4
1
1
11,12
4,5,6
4,5.6,11
1,2,3,4,5,6,7
1,2
9,10
1.2
1.3
1.2
3.4
1.2,3,4
Wahlco
Wahlco
Apollo
Nalco
Nalco
Apollo
Wahlco
Wahlco
Wahlco
Apollo
Dusco
Dueco
Wahlco
Apollo
Wahlco
Wahlco
Wahlco
Apollo
Wahlco
Wahlco
Sulfur Trioxlde
Sulfur trioxlde
LPA-40
LPA-40
"Sulfur trioxlde
Sulfur trioxlde
Sulfur trioxlde
Sulfur trioxlde
'Ammonia
Sulfur trioxlde
Sulfur trioxlde
Sulfur trioxlde
LPA-402A
Sulfur trioxlde
Sulfur trioxlde
-------
TABLE 3. Continued
£am&Anx_liAD£
Plant Name
Public Service Co. of Colorado
Public Service Co. of Colorado
Public Service Co. of Colorado
Public Service Co. of Colorado
Public Service Electric t Gas, Nev Jersey
Public Service Electric I Gas, New Jersey
Salt River Project
South Carolina Public Service Authority
Tampa Electric Co.
Tanpa Electric Co.
Tennessee Valley Authority
Tennessee Valley Authority
Tennessee Valley Authority
Tennessee Valley Authority
Tennessee Valley Authority
TUGCO Dallas Pover and Light
TUGOO Dallas Power and Light
Toledo Edison
UGI Corp. Lurerne Electric
Upper Peninsula Generating Corp.
Upper Peninsula Generating Corp.
Utah Power and Light
Virginia Electric t Pover
Wisconsin Electric Power Co.
Wisconsin Power t Light
Cameo
Cherokee
Comanche
Valmont
Hudson
Nercer
Hayden
Jefferies
Big Bend
F. J. Gannon
Bull Run
Gallatin
Kingston
Shawnee
Widows Creek "B"
Big Brown
Monticello
Bayshore
Hunlock Creek
Presque Isle
Presque Isle
Naughton
Yorktown
Pleasant Prairie
Colunbla
Boiler Number
FCC Licensor
FGC Agent
2
1,2,3,4
1,2
5
2
3,4
5,6
1
4
5
10
7,8
1,2
1,2,3,4,5,6
1,2,3,4,5,6
3
1,2
1
Lodge-Cottrell Sulfur trloxide
Nalco
Apollo
Apollo LPA-40
Apollo LPA-40
Apollo LPA-40
Ammonia
Ammonia
Ammonia
Apollo
Apollo
Nalco
UOP Sulfur trloxide
Apollo Sulfur trloxide
Wahlco Sulfur trloxide
Wahlco Sulfur trloxide
Apollo
Wahlco Sulfur trloxide
-------
the alkaline compound will react with, or neutralize, the
condensed sulfuric acid.
SOj conditioning is effective in reducing ash resistivity
and improving the ESP performance in those cases where particle
resistivity is the limiting factor in ESP performance. In
general, an addition rate of 20 pprav can lower the resistivity by
two orders of magnitude (from about 10* to 1010 Ohm-m to 10' to
101 Ohm-m).
There are cases where SOi conditioning has not been
effective. The reason could be among the following:
1. The conditioner supply malfunctioned.
2. The ESP performance is limited by other phenomena besides
ash resistivity.
3. The ash resistivity may already be satisfactory.
4. The temperature may be so low that acid condensation occurs
at the injection point before the S03 is mixed with the flue
gas.
5. The temperature is much higher than the acid dew point.
Even though SO3 conditioning can improve the particle
collection efficiency of the ESP, it can increase the emissions
of sulfuric acid mist and particulate sulfate compounds. The
emission rate of added SOj is higher for acidic ashes and high
gas temperatures.
Due to increased particle collection by the ESP, the plume
opacity is usually lower with conditioning. However, an "acid
plume" could be formed if the sulfur trioxide dosage and gas
temperature are too high.
Operating Problems
Interviews with several users of sulfur trioxide
conditioning revealed the following categories of operating
difficulties:
1. Corrosion of injection lines.
2. Deactivation of catalysts in the S02 to SOj converter.
3. Over conditioning (resistivity lowered too much).
11
-------
Economics
The estimated capital and operating costs of a sulfur
trioxide flue gas conditioning installation as of December* 1982
are 55.15/kW and 0.105 mills/kWh/ respectively.
Ammonia is injected either before or after the air preheater
in vapor or liquid form. Most of the users inject it downstream
of the air preheater to avoid plugging of the preheater with
deposits of ammonia reaction products.
The ammonia injection dosage is about 15 to 20 ppmv. In
most situations, ammonia can improve the ESP performance.
However/ the way in which ammonia affects the performance of
ESP's is not completely understood. The effectiveness of ammonia
appears to depend mainly upon the initial ash resistivity, flue
gas composition/ and temperature.
The ability of ammonia to alter resistivity is not clear.
Ammonia conditioned resistivities can be less than, the same as,
or greater than the unconditioned values. The temperature of the
precipitator has great effect on resistivity modification by
ammonia. Since the critical temperature for ammonia is 132°C,
condensation of ammonia is not expected to occur above this
temperature. Therefore, ammonia injection will not change the
ash resistivity above this temperature unless it is sorbed as a
reaction product as NHsSCh.
The improvement in ESP performance upon ammonia injection is
more likely due to space charge enhancement and improvement in
cohesive force between the ash particles. The injected ammonia
reacts with the existing sulfuric acid vapor, forming a fume of
fine salt particles of ammonium bisulfate and sulfate. This fume
creates a large surface area for collecting electrons. These
charge carriers have a lower mobility than electrons and allows
for a more stable corona,,a higher electrical field strength
without breakdown, and a higher specific power.
12
-------
Ammonia is sometimes injected along with sulfur trioxide.
Sulfuric acid condensation in the presence of ammonia (or
ammonium bisulfate and sulfate) forms viscous surface deposits
which increase the ash cohesivity and reduce particle
reentrainment from the collection plates.
The emissions caused by ammonia injection are minimal. A
significant part of the ammonia reacts with nitrogen oxides to
form elemental nitrogen. Reaction between NHi and SOi also
decreases the SOs emission.
Operating Problems
Ammonia conditioning could have the following operation and
maintenance problems:
1. Plugging of injection nozzles.
2. Leakage and freezing of injection lines.
3. Dust build up on discharge electrodes of the ESP.
Economics
The capital and operating costs of ammonia conditioning, in
December, 1982 U.S. Dollars, are $0.21/kW and 0.022 mills/kwh,
respectively.
Ammonium roipniindfi
Ammonium compound injection rate is in the range of 0.25 to
1.0 g/kg of coal burned. They are injected in solution form
either upstream or downstream of the air preheater. Upstream is
the preferred location because it offers long residence time and
high temperature to vaporize or decompose the agent. However,
upstream injection could cause plugging of the air preheater.
Ammonium compounds improve the ESP performance through the
mechanisms of resistivity modification and space charge effect,
but are not as effective as SOi in lowering the ash resistivity.
Results show that they can lower the ash resistivity by about
half an order of magnitude.
The injection of ammonium compounds causes increased sulfate
and ammonia emissions. A particulate sulfate emission rate about
13
-------
20 iig/m1 and 1,500 ng/m3 was measured without and with
conditioning, respectively (Patterson et al., 1979C).
Problems
The most common operating problem with ammonium compound
conditioning agents has been the plugging of the air preheater by
deposits of reaction products when the agent is injected upstream
of the air preheater.
Economics
There is no information on capital costs for an ammonium
compound conditioning systems. The operating costs range from
0.024 to 0.052 mills/kWh (December, 1982 U.S. Dollars).
Organic Amines
Even though triethylamine is less volatile than ammonia, it
can be injected in the vapor phase as well as in the solution
form. Because it decomposes at high temperature, it is injected
downstream of the air preheater.
The mechanism of triethylamine is not fully understood. The
most likely mechanism is ash resistivity reduction. A pilot plant
study by Brown et al. (1978) showed the resistivity decreased
from 3 x 10* Ohm-m without conditioning to 5 x 107 ohm-m with a
triethylamine dosage of 60 ppm. An independent pilot plant study
by Bickelhaupt et al. (1978) showed similar results. With an
injected concentration of 25 ppm of triethylamine, the
resistivity decreased one to two orders of magnitude in the
temperature range of 100°C to 150°C. It is more effective with
lower temperatures, less basic ash composition, and greater
concentration of the agent.
Triethylamine has only been studied in the laboratory and in
pilot plants. There are no industrial users, so no economic data
and user experience are available.
14
-------
Dry Alkali
Sodium conditioning, unlike the other conditioning agents,
is not limited to cold side BSP's. It can be added to the boiler
along with coal or into the flue gas just ahead of the ESP. It
can be applied either in solution form or in dry powder form.
The most important parameter which affects its effectiveness is
mixing of the sodium salt and the fly ash. To be effective, the
sodium must either be incorporated into all the ash particles or
co-precipitated with the ash on the precipitator plates so it
yields well mixed deposits.
When the sodium is applied in dry powder form for co-
precipitation# there may be difficulties in obtaining well mixed
deposits. Lederman et al. (1979) applied the sodium in solution
form and claimed that uniform coating of sodium salts on ash
particles was obtained. When the agent is injected for co-
precipitation with the ash, the dosage is 2% to 5% of the solids
as NazO. The co-precipitated sodium compound should have a
particle size distribution comparable to that for the ash.
In a pilot study, Gooch et al. (1981) added sodium compound
to the coal supply prior to pulverization as a means of
supplementing the sodium content of the fly ash. They speculated
that complete decomposition and volatilization of the
conditioning agent occurred in the boiler and subsequent
condensation of the sodium compound was uniformly distributed and
became an integral part of the fly ash surfaces.
Sodium is effective in reducing the fly ash resistivity if
the sodium is mixed well with the ash. In situ resistivity
measurements of co-precipitated ash showed the resistivity
decreased from 2.1 x 101B Ohm-m without conditioning to 3.7 x 10*
Ohm-m when conditioned with a 1.0 to 1.5% concentration of sodium
carbonate as sodium oxide (Schliesser, 1979 a,b). A reduction of
resistivity from 1 x 1010 to 1 x 101 Ohm-m was measured by Gooch
et al. (1980) when the sodium oxide content of the ash was
increased to 2.5% from the inherent 0.3%.
15
-------
Problens
Industrial users have only limited experience with sodium
conditioning; therefore, operational problems are not well
documented. However, there is one potential problem associated
with the addition of sodium compound to the coal. The sodium may
cause ash slagging and boiler fouling. In a properly operated
system this should not be a problem.
Economics
The capital costs for a liquid solution conditioning system
is about $1.55 to 3.10/kW installed. The operating costs,
excluding depreciation, is about 0.03 mills/kWh.
CONCLUSIONS
Of the many agents available, S03 is the roost commonly used.
SOj is effective in reducing the ash resistivity and will improve
the ESP performance if particle resistivity is the limiting
factor. The conditioning mechanisms of ammonia, ammonium
compounds, and organic amines are not fully understood and the
effectiveness of these compounds is not consistent.
Flue gas conditioning appears to be an acceptable and the
least expensive option for upgrading the ESP performance for
collecting high resistivity fly ash. However, before deciding on
flue gas conditioning, the reasons for poor ESP performance
should be determined. The poor performance could be due to
factors other than high resistivity. Once the cause has been
determined to be resistivity and conditioning has been chosen for
retrofit, the conditioning system should be designed and operated
with extreme care to avoid the harmful emissions due to
conditioning agents.
16
-------
Section 2
INTRODUCTION
CAUSES FOR UNSATISFACTORY ESP PERFORMANCE
ESP's have proved reliable, economic, and effective at
controlling particle emissions from coal-fired utility boilers.
Sometimes their performance has been unsatisfactory because of:
1. Coal composition change
2. More stringent particle emission regulations
3. Unstable electrical conditions
4. Changes in boiler and associated equipment operating
conditions
4
5. Insufficient collection area
6. Poor gas distribution and maintenance.
ESP's are usually designed for boiler burning a specific
coal. When a different coal, such as Western low sulfur, low
alkali coal is burned, the ESP performance can change with
changes in the following ash and flue gas properties:
1. Ash resistivity
2. Coal ash content
3. Ash composition
4. Ash particle size distribution
5. Moisture and fuel sulfur content
Ash Resistivity
The preferred ash resistivity, p, is in the range of 1 x 10*
to 1 x 1010 Ohm-m. When particles of high resistivity (p > 1 x
1012 Ohm-m) are collected on the collection plates of an ESP, an
insulating layer of ash forms there. Due to its high
resistivity, charge dissipation is impeded by this layer. The
voltage drop across this ash layer is increased and the field
strength in the intere1ectrode region is decreased
correspondingly. Thus, the particle charging rate falls, the
driving force for the particle migration falls. Additionally
17
-------
deposits are strongly electrostatically bound to the plates and
are difficult to dislodge.
Trying to overcome the decreased field strength by raising
the applied voltage to the discharge electrode can lead to the
occurrence of back corona# which is the local breakdown or
ionization of the gas molecules. Back corona occurs when the
local field on the interstitial gas in the dust layer climbs to
an excess of 1,000 to 2,000 kV/m. When back corona occurs,
positive ions will stream out of the dust layer, quench the
negative charges on the incoming particles, charge the particles,
and force the particles back away from the plate. The back
corona discharge can propagate through the interelectrode gas to
the corona wire when the applied voltage is raised even higher.
To prevent back corona, the operating voltage is lowered which in
turn lowers the particle charging level.
If the ash resistivity is below 10* ohm-m, charges dissipate
readily and the voltage drop across the dust layer is low. The
adhesion force on the collected particles is therefore low. The
particles are more susceptible to being reentrained in the flue
gas.
Ash Content
If the ash content of the coal is increased, then the mass
of ash per unit of energy is increased and the ash burden
entering the ESP will be higher.
Ash Composition
Depending on the composition of the ash, it may be
conductive. The presence of sodium, lithium, and potassium in
the ash increases its volume conductivity (Bickelhaupt, 1974).
The particle surface composition can aid agglomeration of
particles on collision with each other, or in keeping the
particles on the plate from being reentrained (Tassicker, 1975).
10
-------
Particle Size Distribution
The particle size distribution and mean particle size play a
direct role in the efficiency at which particles can be
collected. As the particle size decreases? particle migration
velocity decreases and so does the collection efficiency of the
ESP.
Another factor in the particle size distribution is the
chemical make-up of the size fractions. Usually particles
smaller than 5 jim and especially the sub-micron segment are
formed by condensation and are more spherical with minimal
surface porosity. While these fine particles have greater
surface area for a specific weight fraction? they also appear to
sorb less of the condensible gases? especially SOi (Katz, 1979).
Fine particles may also allow the collection surface deposit to
increase in depth by the greater cohesion force of the particles
(Katz, 1979).
Moisture Content
The hydrogen content and the amount of bound and free water
in the coal determine the moisture content of the flue gas. High
moisture, in conjunction with very low temperature, can improve
the electrical conditions in the ESP and lower the ash
resistivity. Water condensation and adsorption in the pores and
on the surface of the ash particles leaches soluble compounds
from the particles to form a layer of conductive or cohesive
solution on the particle surface.
Puel Sulfur Content
The sulfur in the coal is oxidized in the boiler to sulfur
dioxide and sulfur trioxide. As the gas cools, sulfur trioxide
associates with water vapor to form sulfuric acid vapor. This
sulfuric acid condenses with the moisture in the flue gas to form
a conductive layer on the particles, lowering the resistivity.
The SO] concentration in the flue gas depends on the fuel
sulfur content, furnace operating conditions, and amount of metal
19
-------
impurities, such as vanadium and iron, for converting SO* to SOi
catalytically in the boiler.
Boiler and Associated Bfniipment Effects
Changes in the ratio of air to coal (excess air) changes the
flue gas flow rate, temperature, and the properties of the fly
ash produced. More air dilutes the particles and increases the
flue gas flow rate, which in turn reduces the specific collection
area. Oxidizing conditions in the boiler raise the melting
temperature and viscosity of the fusible components of the fly
ash (especially in wet-bottom boilers).
Flue gas temperature affects the electrical properties of
the ash. Measurements have shown a variation in resistivity with
gas temperature with a maximum around 150°C to 200°C (300°F
to 400°F). High sulfur coal with low exit gas temperature
produces low resistivity ash which may be too low for the ESP.
On the other hand, low sulfur coal with high exit gas temperature
results in high resistivity ash. For coals with less than 1%
sulfur, the ESP should be operated at a temperature below 145°C
(290°F). When coal sulfur content is higher than 2.5% the ESP
should be operated at a temperature above 200°C (400°F) (Katz,
1979).
If the boiler is run over capacity, the temperature and
amount of flue gas will be affected which in turn affects the ESP
performance as discussed above. Deterioration of the performance
of a control device upstream of the ESP (such as a cyclone
battery or other mechanical collector) can cause substandard
performance of the precipitator.
Insufficient Collector Area
The size or capacity of an electrostatic precipitator is
expressed in terms of the specific collection area (SCA) which is
the ratio of collection plate area to the flue gas volume
flowrate. The units are m,/(m,/s) or ft*/l,000 acfm.
Precipitators built for efficiencies on the order of 90% for high
sulfur coal have had specific collection areas of 20 to 40
20
-------
m2/(m'/s). These same precipitators when used on lower sulfur
coal ash have had much poorer performance. New precipitator
installations are being designed to operate on lower sulfur
fuels, and to meet higher performance standards. ESP's are
presently designed with a SCA greater than 100 m'/m'/s to meet
the New Source Performance Standards (NSPS).
METHODS FOR IMPROVING ESP PERFORMANCE
There are several methods for upgrading the ESP performance.
^ The options are:
1. Add collection plate area to the existing ESP to overcome
poor performance.
2. Use a wet electrostatic precipitator to minimize
reentrainment and keep the plate clean.
3. Increase or lower the gas temperature in the ESP.
4. Add chemicals to modify the fly ash or the electrical
conditions in the ESP.
For older ESP's, flue gas conditioning is often the most
cost effective method of increasing the ESP performance. Several
chemicals, such as sulfur trioxide, ammonia, ammonium compounds,
organic amines, and dry alkalis, are presently used or proposed
as conditioning agents. This report presents the results of a
survey of flue gas conditioning agents and user experience.
21
-------
Section 3
SULFUR TRIOXIDE CONDITIONING
CHEMICAL PROPERTIES
Sulfur trioxide is the most common flue gas conditioning
agent used by power plants in the United States. It is a
colorless, transparent, and stable liquid at room temperature and
atmospheric pressure. It has a low vapor pressure. While it is
not particularly corrosive in closed storage, it is highly
hazardous if released in bulk.
If it is allowed to cool and solidify, three forms of SO»
(alpha, beta, and gamma) may be formed. The lowest melting form
(gamma) is preferred for handling, but is unstable. The higher
melting solid (the alpha form) may form in the presence of
moisture (Schrader, 1970). When exposed to humid air, the vapor
quickly forms a dense fog of sulfuric acid (Archer, 1972).
The active conditioning agent in so-called "sulfur trioxide
conditioning" is sulfuric acid. At the temperatures and
humidities found at the air preheater outlet, sulfur trioxide is
almost completely hydrated to sulfuric acid. Table 4 shows some
important properties for SOi and sulfuric acid. Table 5 shows
the equilibrium partition between SO*, SOi, and HtSO*.
HjSO* has a low vapor pressure at room temperature and a
strong affinity for water. Dissolved in water, at moderate
concentrations it almost completely dissociates to two hydrogen
ions and a sulfate ion. The hydrogen ions are small and mobile
in solution. Their large number and mobility combine to make
aqueous sulfuric acid solution electrically conductive.
The low vapor pressure of sulfuric acid means that strong
solutions may be in equilibrium with very dilute vapor phase
concentrations. In addition, the particle surface curvature
increases the equilibrium vapor pressure (Nair and Vohra, 1975).
Therefore, small acid drops can exist above the boiling tempera-
ture of a flat surface of the same acid composition. The low
volatility and the high conductivity of sulfuric acid are
responsible for its effectiveness in lowering fly ash resistivity
in ESP's.
-------
TABLE 4. PROPERTIES OF SULFUR TRIOXIDE*
Property Sulfur Trioxide Sulfuric Acid
Melting Point# °C 10.6
Gamma form (unstable) 16.8
Beta form (unstable) 35.8
Alpha form 62.2
Vapor Pressure, kPa
25°C 35.3 0.702 x 10-*
44.8°C 101.3 3.62 x 10-1
218.3°C** 8,470.0 24.9
27 2 °C*** 101.3
Density, kg/mJ
Liquid (20°C) 1,920.0 1,830.5
Vapor (20°C, 101.3 kPa) 3.57 4.08
Latent Heat, kJ/kg 533.0 802.0
Heat Capacity, J/(kg-K)
Average, 20-30°C 322.0 384.0
*References: Schrader (1970), Perry & Chilton (1972), Graitro
and Vermeulen (1964).
~~Critical Point
***Aqueous sulfuric acid forms an azeotrope at 326°C, 101.3 kPa
and 98.5 wt % H2SO*. The normal boiling temperature shown
is for pure liquid HjSO*.
23
-------
TABLE 5.
EQUILIBRIUM PARTITION OF SULFUR OXIDES9
(Dismukes, 1976)
Temperature#
°C
Relative concentrations# % of sulfur compounds
SOi SO, HiSO»b
1,400
1,000
800
600
500
400
300
200
100
99.7
97.6
88.7
42.2
12.3
1.6
0.1
0.0
0.0
0.3
2.4
11.3
57.7
87.3
94.8
59.3
6.5
0.0
0.0
0.0
0.0
0.1
0.4
3.6
40.6
93.5
100.0
a. Calculated from the data in JANAF Tables assuming
concentrations of oxygen and water vapor equal to 4%
and 10% by volume respectively.
b. The maximum absolute concentration of each compound is
sharply limited below 300°C as a result of the
condensation of the predominant compound, HiSO», in a
binary H,SO*-HiO liquid mixture.
24
-------
CONDITIONING MECHANISM
Sulfuric acid reduces particle resistivity in electrostatic
precipitators by forming a conductive layer on the surface of the
particles. This layer forms by adsorption or condensation of
sulfuric acid and water on the surface of the particles and in
the crevices between particles on the collection plate.
CONDITIONING METHODS
Four processes are common for producing the sulfur trioxide
for conditioning (Archer/ 1972):
1. Vaporization of a sulfuric acid solution,
2. Vaporization of liquid sulfur trioxide,
3. Vaporization of liquid sulfur dioxide and oxidization to
sulfur trioxide over a vanadium pentoxide catalyst.
4. Burning liquid sulfur in air to produce sulfur dioxide and
then oxidize it to SOj.
Other methods/ such as stripping of oleum or in-situ
oxidation of S02 in combustion gas are not commonly used.
Acid Vaporization
Sulfuric acid is heated to greater than 315°C (600°F). The
vapor is then diluted with air and injected into the flue gas. A
direct-fired heater is used to produce hot air with a temperature
in the range between 315°C (600°F)f where the acid may condense,
and 600°C (1,100°F), where it dissociates. The hot air then
contacts the acid in a vaporizer, which can be a packed tower or
any other suitable direct contact heat-exchanger. The air flow
rate is chosen to supply all the necessary enthalpy for
vaporizing the maximum acid flow rate. The gas mixture from the
vaporizer is maintained at a temperature above the acid dew point
up to the injection point. Typical temperatures in the system
are:
Hot air to vaporizer 570°C (1,050°F)
Gas from vaporizer 450°C (850°F)
Gas at injection point 400°C (750°F)
25
-------
Catalytic Conversion of SO*
The catalytic conversion system includes an air heater, a
SOi generator, and a catalytic converter. The air heater is a
direct fired heater. SO» is produced either by burning liquid
sulfur in air or by vaporizing liquid sulfur dioxide. The SO*
gas is then mixed with hot air to obtain a maximum SO 2
concentration of about 8%.
The converter is a catalytic chamber with suitable
catalysts, such as vanadium pentoxide. The inlet gas mixture at
the catalytic chamber is at about 430°C to 450°C (800°F to
8500F). The oxidation of SOj to SOa is an exothermic reaction.
Therefore, multiple passes with interstage cooling are employed
to keep the temperature down and achieve good conversion
efficiency.
The SO] rich stream from the converter has a temperature
range from 450°C to 600°C (850°F to 1,100°F), which is much
higher than the acid dew point. Thermal insulation on the
injection duct will keep the injection temperature safely above
the acid dew point even with moisture present.
Vaporization of Liquid SOi
Liquid SOj is metered into a vaporizer and then diluted with
air. Dilution increases the volume for easier handling. The
mixture must be conveyed in heated lines to the point of
injection and is injected into the gas stream ahead of the
precipitator.
Injec
The most common injection location is between the air
preheater and the electrostatic precipitator inlet, for the
following reasons.
1. Injection upstream of the air preheater increases the chance
of acid condensation and corrosion there.
2. Injection into a high temperature region can lead to the
decomposition of sulfuric acid or sulfur trioxide into
sulfur dioxide which is not an effective conditioning agent.
3. This position gives adequate residence time for
26
-------
conditioning. SoRI (1972) determined that the fly ash could
be satisfactorily conditioned in as little as 1 second after
introduction of sulfur trioxide.
Southern Research Institute (SoRI) (1972) did tests on
boilers which had mechanical collectors (batteries of cyclones)
between the air preheater and the electrostatic precipitator.
They found injection before or after the mechanical collector to
be equally effective.
The temperature at the point of injection should be above
the sulfuric acid dewpoint of the gas after addition (Reese and
Greco, 1968) to prevent pre-mature condensation. The vapor
should be allowed to mix with the flue gas and diffuse to the
surface of the fly ash particles before condensation occurs.
RESULTS
Effects on Particle Resistivity
Table 6 and 7 show the ash resistivities measured with and
without sulfuric acid conditioning. Figure 1 shows a summary of
the effect of SO» conditioning on fly ash resistivity. In
general, SOi conditioning is effective in reducing ash
resistivity and improving the ESP performance if particle
resistivity is the limiting factor. The dosage for S03
conditioning is normally in the range of 5 to 30 ppmv# but can be
as high as 70 ppmv. An addition rate of 20 ppmv can lower
resistivity by two orders of magnitude.
The effectiveness of sulfuric acid in reducing particle
resistivity in ESP's depends on the gas temperature and ash
composition. It is les6 effective at reducing resistivity at
high temperatures (> 200°C) for two reasons. First, it is above
the acid dew point such that there is either less adsorption or
no condensation. Second, bulk or volume resistivity of the fly
ash falls with temperature in the fashion of an Arrhenius
relation (Bickelhaupt, 1974). At high temperatures, volume
conduction becomes more important than surface conduction. The
conductive sulfuric acid layer enhances surface conduction only.
The composition of the ash surface has an important effect
on the amount acid required for conditioning. If the ash has a
27
-------
TABLE 6. ASH RESISTIVITIES FOR USERS OP SO, CONDITIONING
Utility Plant-
City of Colorado Springs Drake
Dept. Pub. Util.
Commonwealth Edison
Detroit Edison
Iowa Public Service
Public Service Co
Crawford, Fisk,
Jolietr Powerton#
State Line# Waukegan,
Will County
SO) Dosage
ppmy
<60
5 to 30
15
Ash Resistivity. Ohm-m
without
SOi.
5 x 10#
1010 to 1011
Connors Creek, Harbor
Beachf Marysville#
Pennsalt, Port Huron,
Trenton Channel
Monroe 15
(with 11 ppm NHi)
107 to 10*
Neal
Arapahoe
20
14 to 18
10
i o
with
_SQi
6.5 x 10«
10*
107 to 10*
10«
(5xl07 to 10*
9 x 107
2xl010 to 1011 2x10* to 10*
-------
TABLE 7. ASH RESISTIVITIES FOR SEVERAL USERS OF SOj CONDITIONING
Reference
Utility, Plant
Fuel S03 Coll.
Sulfur Dosage Ash Resistivity SCA Effic.
wt % ppmv Ohm-m m2/(m,/s) %
Trevor
et al.
(1963)
CEGB, Kincardine
(Scotland, UK)
0.5
0
5
10
15
0.4 to 2 x 1010
0.25 to 1 x 1010
1 to 5 x 109
0.6 to 2.7 x 10»
86
95
97.5
99
Coutaller
Richard
(1967)
France
0.5
0
15
20
94.5
99.0
99.4
Schrader
(1970)
Germany
0.8 to
0.9
14 to 26
96.7
98.4
to
vo
Southern
Research
Institute
(1972)
Pub. Svc. Co,
Colorado
Cherokee 2
Utility X,
Unit 4
0.6
0
13
27
0
10
18 to 20
1.6 x 10*
1.4 to 5.6 x 10*
2.5 to 5 x 10*
4.5 x 10»
2.3 x 10»
PSC Colorado,
Cherokee 3
0.47 to
0.54
0
6
13 to 17
26
33
44
1 to 2 x 1010
9.3 x 10*
7.8 to 10 x 10*
5.6 to 6 x 10*
5 to 6.3 x 10»
6.6 x 10*
PSC Colorado,
Arapahoe 4
0.49
0
6
12
18
3.8 x 1010
5.7 x 10*
3.2 x 10*
1.9 x 10*
-------
TABLE 7. ASH RESISTIVITIES FOR SEVERAL USERS OF SO, CONDITIONING (Contd)
Reference Utility, Plant
Fuel SOj Coll.
Sulfur Dosage Ash Resistivity SCA Effic,
wt % ppmv Ohm-m m2/(mVs) %
SoRI
(1972)
(cont'd)
Green &
Landers
(1974)
Utility Y, 0.59 to
Unit 6 0.64
PSC Colorado 0.6 to
Arapahoe, Cameo, 0.7
Cherokee
0
4
8
16
0
18to25
2 x 1010
0.7 to 2.0x10*
1.3 to 2.1x10*
0.1 to 2.5x10*
10
11
17 to 43
54 to 94
73 to 96
White
Calgary Pwr., Ltd.
0.25
0
2
x 1010
25
(1974)
Sundance, Wabamum
10
3
x 10*
20
4
x 10*
18
30
1
xlO*
Cook
Commonwealth
0.5 to
0
24
to
(1975)
Edison, State Line
0.9
20
30
35
40
50
Klipstein
Unnamed
0
10* to 1010
(1975)
16
30
39to55
2
to 7 x 107
2 to 3
0
Cook &
Commonwealth
0.5 to
0
24
to
Trykowski
Edison, State Line
0.9
40
71
97.5
79 to 89
88
94
92
94 to 96
96
78 to 81
84 to 89
94
94 to 96
95 to 97
86 to 88
95.1 to 97
(1976)
-------
TABLE 7. ASH RESISTIVITIES FOR SEVERAL USERS OF SO, CONDITIONING (Contd)
Reference Utility, Plant
Fuel SOi Coll.
Sulfur Dosage Ash Resistivity SCA Effic.
wt % ppmv Ohm-m m'/lm'/s) %
Dismukes
& Gooch
(1977)
Iowa PS, Neal
0.6
0 5.7 to 8.3xl010
25 0.3 to 1 x 10*
39 91.3
99 to 99.2
Brown
et al.
(1978)
Pilot Scale
0
5
13
64
10*
10'
10*
85
91
99
98 to 99
Patterson
et al.
(1979)
Unnamed
0.8
1.1
0
32
1.7 x 10*
4.7 x 10*
36
80.2
94.9
Guiffre Pennsylvania Pwr 1
(1980) & Light, Montour
& Brunner Island. 2.2
0
19to25
0
5 x 10*
35 78 to 80
94,5to94.8
35 93.6
-------
\
¦\ \
\ '
~
\
\
PLANT 5 (135 C, BASIC ASH)
'PLANT 3 (110°C, BASIC
V* ASH)
6 (
Pr>.
7 ^ ,
&ASxc
ASH)
10
15
20
25
30
35
40
CONCENTRATION OF S03 INJECTED, ppm
Figure 1. Resistivity as a function of the concen-
tration of injected sulfur trioxide.
(Dismukas, 19 76)
32
-------
large amount of alkaline compounds which react with or neutralize
sulfuric acid, then a non-conductive salt layer will be formed.
For basic ashes, a higher S0> dosage is needed so that excess
acid can be adsorbed or condensed on top of the salts.
Emissions Caused By Sulfur Trioxide Conditioning
The major potential emissions caused by sulfur trioxide
conditioning are sulfuric acid mist and particulate sulfate
compounds. The emission rate will depend on the ESP performance,
ash composition, and the gas temperature. Conditioning with SOi
leads to the condensation of acid and the formation of sulfate on
fly ash particles. Depending on the degree of improvement in the
ESP particle collection efficiency with conditioning, an increase
in sulfate emission may result.
The amount of alkaline compounds, specifically those of
calcium, sodium, and magnesium, affect the split between free
acid (as hydrated vapor or as adsorbed or deposited liquid) and
sulfate bound to the particles. Chemical adsorption on, or
reaction with the particle surface can decrease the amount of
free acid present. Since sulfuric acid vapor is not collected by
the ESP to any appreciable extent and the bound sulfates are,
the more acid that the particles accept, the smaller the stack
emission of acid will be. Fly ash surfaces which are alkaline
favor such acceptance. Acidic ash surfaces (high AliOj, low
alkali) hinder such acceptance, and therefore are subject to
large stack emission of acid (Dismukes, 1976; Dismukes and Gooch,
1977).
The precipitator temperature also affects the sulfate
emissions. A temperature well above the acid dew point in the
precipitator allows the acid to remain in the vapor. The acid
vapor will nucleate and condense to form acid particles once it
exits the stack and mixes with ambient air. If the precipitator
temperature is close to or below the sulfuric acid dew point,
then the injected SOi may still form a fog of sulfuric acid drops
as well as condensing on the particles. Fog formation reduces
available acid for conditioning and causes acid mist emission.
Table 8 shows the results of field tests for the fate of
33
-------
injected acid. The plant tested by Dismukes and Gooch (1977)
(Iowa PS, Neal 2) had a basic ash. The precipitator temperature
was low. Dismukes and Gooch showed a decrease in total sulfate
emission with flue gas conditioning.
The plants tested by Patterson et al. (1979a) and Dismukes
(1976) had an acidic ash and a higher precipitator temperature
than the plant sampled by Dismukes and Gooch. They reported an
increase in stack sulfate emissions with conditioning.
An increase of sulfate emissions from conditioning can be
expected when the gas temperature approaches or is above the
sulfuric acid dewpoint in the precipitator and the ash is acidic
in character and does not absorb or accept the injected sulfuric
acid. Sulfate emissions from conditioning are less for ash which
is basic and has a high alkali content than that for acidic ash.
The effect of flue gas conditioning on particle composition
is shown in Table 9. The sulfate content of the fly ash increased
with sulfuric acid conditioning. The increase depends on S03
dosage and is less when the sulfur-to-ash weight ratio is less.
SO a conditioning may either increase of decrease the plume
opacity. Many plants achieve an opacity of 20% or less with SO]
conditioning due to increased particle collection by the ESP.
However, an acid plume could be formed if the SOj dosage and the
gas temperature are too high. Prudent operating procedure is to
operate at a conditioning level where SOj or HiSO* is below the
acid plume formation level.
Effpc.t- On Parfciclg Size ni utrlhutlnn
Patterson et al. (1979a) measured the particle size
distributions before and after conditioning. No detectable
difference occurred because of:
1. The mass of SO 3 added is small in comparison with the ash
(<1.0%) ;
34
-------
TABLE 8. FIELD DETERMINATIONS OF ACID
PARTITION IN FLUE GAS CONDITIONING
Reference
Temp.
°C
Injected
SO j
ppmv
SO >
ppmv
Inlet
Part.
Sulfate SO*
rag/Nra1 ppmv
Outlet
Part.
Sulfate
mg/Nm3
Dismukes (1976) 125 0 1.5
14 6
Dismukes & 130 0 0.4 68 0 16
Gooch (1977) 25 1.8a 123 0.9 6
Patterson 145 0 2.2 0.9 1.0 0.4
et al. (1979a) 32 10.9 5.4 8.1 2.5
Note: 1 ppmv SOj = 4 mg/Nm1 of SOT"
a: The SO] injection rate is not a measured value and
thus a large difference between injected and inlet SOE
may be found. Also some SO3 can be lost to dust and
ducts.
-------
TABLE 9. CHANGES IN PARTICLE SULFATE COMPOSITION
WITH SULFUR TRIOXIDE CONDITIONING
Added Particulate
S03 Mass Cone.
Reference Utility, Plant ppmv mg/Nm3
SoRI PSC Colorado 0 1,100 to 5,000
(1972) Arapahoe 4 6 1,100
12 1,100
18 1,100
PSC Colorado 0 2,200 to 6,000
Cherokee 2 13 2,200
27 2,200
PSC Colorado 0 6,000
Cherokee 3 13 900
26 900
33 900
44 900
Utility X
0
10
18 to20
17 ,300
Utility Y
0
4
8
16
7,000
Dismukes &
Gooch
(1977)
Iowa PS
Neal 2
0 6,000 to 6,300
25 6,900
Patterson Unnamed
et al. (1979a)
0 2,100 to 3,100
32 2,100 to 4,200
*Total of Free and Combined Sulfate
Total
Sulfate*
wt %
1 to 1.5
2.23
2.5
2. 97
0.29 to
1.28
1.42 to
1. 53
1.76 to
1.89
0.57 to
0.77
0.9
1.09
1.00
1.12
0.17
0.31
0.38
0.24 to
0. 27
0.2 9 to
0.35
0.34 to
0. 41
0.3 6 to
0.51
0.76 avg
1.43 avg
0.05-0.07
0.14-0.18
36
-------
2. Deposition of acid on the surface of the fly ash is only a
few molecular layers thick.
The earliest sulfur trioxide conditioning systems used
liquid SOi as a raw material. Because of its volatility,
hygroscopicity, and phase instability, it is hazardous to handle.
The sulfuric acid solution vaporization method has corrosion
problems. It is also difficult to turn down or shut down a
sulfuric acid vaporizer without plugging the injection lines.
The systems which start with SO* or elemental sulfur have
proved most reliable. Sulfur dioxide is less hazardous than S0Jf
but is the most expensive source of synthetic SOj (Archer, 1972;
Dalmon and Tidy, 1972). When elemental sulfur is available, it
is the most economical source of SOi.
In general, S02 and sulfur-based conditioning systems have
offered availabilities in excess of 90 to 95%. Interviews of
several users of S03 conditioning revealed the following
categories of operating difficulties:
1. Corrosion.
2. Catalyst deactivation.
3. Over conditioning.
Sulfuric acid condenses in the lines leading to the
injection nozzles if the wall temperature is below the acid dew
point. These lines are prone to plugging during periods of
slowdown, shutdown, or cold weather. The use of purge air in
these lines is recommended during shut down periods.
Periodic changing of the SO* to SOi converter catalyst has
been found necessary at one utility. The 6ulfur trioxide content
of the injected gas decreases over about a year of operation.
SOj concentration and ESP performance are restored after catalyst
replacement. Catalyst deactivation may be responsible for the
failure of sulfur trioxide conditioning in other installations.
It is possible to over-condition the ash. The presence of
too much sulfuric acid in the flue gas can (Reese and Greco,
1968) :
37
-------
1. Lower the resistivity so much that the ash is reentrained
from the collecting plates or bounces off them.
2. Form a solution between the particles on the plates which
binds them together and makes them difficult to dislodge
from the plate.
USER EXPERIENCE
The list of known past and present users of sulfur trioxide
conditioning given in Table 10 were contacted for their
experience. Summaries of user experience from those users who
replied to our inquiries/ appear below with other published
information.
Colorado Springs Department of Pnhlic
The DPU has used a Research-Cottre11 sulfur trioxide
conditioning system on the Martin Drake plant for seven years.
The efficiency of particle collection in the ESP increased 50%
over the unconditioned case. The fly ash resistivity decreased
by one order of magnitude.
Central Til Light
Central Illinois Light installed Wahlco sulfur trioxide
conditioning systems at their Wallace plant in 1976. and they
installed another Wahlco system at their Edwards plant in 1978.
Commonwealth Edison Company
Commonwealth Edison has installed 28 Wahlco sulfur burning
trioxide flue gas conditioners at various plants. The oldest
were installed in 1974. Test results from State Line units 1, 3
and 4 were reported by Cook and Trykowski (1976). The SOi dosage
was about 40 ppmv. Compliance with emission limit was achieved
at all but one of the units tested. That unit, number 3, had to
be derated. Efficiencies of 96 to 97% were reported. The
corresponding unconditioned efficiencies were about 85 to 88%.
38
-------
TABLE 10. LIST OF PAST AND PRESENT USERS OF SULFUR UUOXIDE CONDITIONING
Company Name
Plant Name
PGC Licensor
Alabama Power Co.
Barry
Wahlco
Central Illinois Light
E. D. Edwards
Wahlco
Central Illinois Light
R. S. Wallace
Wahlco
Cincinnati Gas & Electric Co.
W. C. Beckjord
Wahlco
City of Colorado Springs DPU
Martin Drake
Research-Cottrell
Cleveland Electric Illuminating Co.
Avonlake
Wahlco
Cleveland Electric Illuminating Co.
Eastlake
Wahlco
Cleveland Electric Illuminating Co.
Lake Shore
Wahlco
Coirmonwealth Edison
Crawford
Wahlco
Comnonwealth Edison
Fisk
Wahlco
Ccnrnonwealth Edison
Joliet
Wahlco
Coirmonwealth Edison
Powerton
Wahlco
Comnonwealth Edison
Waukegan
Wahlco
Ccnmonwealth Edison
Will County
Wahlco
Comnonwealth Edison/Indiana
State Line
Wahlco
Consumers Power Co.
B. C. Cobb
Wahlco
Consumers Power Co.
J. C. Weadock
Wahlco
Detroit Edison
Conners Creek
Wahlco
Detroit Edison
Harbor Beach
Wahlco
Detroit Edison
Monroe
Wahlco
Detroit Edison
Marysville
Wahlco
Detroit Edison
Pennsalt
UOP
Detroit Edison
Port Huron
Wahlco
Detroit Edison
Trenton Channel
Wahlco
Duke Power
Belews Creek
Research-Cottrell
East Kentucky Rural Electric Power Coop.
W. C. Dale
Wahlco
Icwa Public Service Co.
G. W. Neal
Wahlco
Lansing Board of Water & Electric Light
Erickson
Wahlco
New York State Electric & Gas
Goudey
Wahlco
New York State Electric & Gas
Greenidge
Wahlco
Northern Indiana Public Service Co.
D. H. Mitchell
Wahlco
References
Cook (1975)
Cook and
Trykcwski (1976)
Brennan and
Reveley (1977)
Dismukes and
Gooch (1977)
-------
TABLE 10. LIST OF PAST AND PRESENT USERS OF
Company name
Plant Name
Ohio Edison Co.
Pennsylvania Electric Co.
Pennsylvania Power & Light Co.
Pennsylvania Power & Light Co.
Pennsylvania Power & Light Co.
Public Service Co. of Colorado
Public Service Co. of Colorado
Public Service Co. of Colorado
W. H. Sammis
Front Street
Brunner Island
Montour
Sunbury
Arapahoe
Cameo
Cherokee
Tennessee Valley Authority Bull Run
UGI Corp. (Luzerne Electric) Hunlock Creek
Upper Peninsula Generating Corp. Presque Isle
Utah Power and Light Naughton
Wisconsin Electric Power Co. Pleasant Prairie
TRIOXIDE CONDITIONING (Cont'd)
ZQC Licensor
Wahlco
Wahlco
Wahlco
Wahlco
Wahlco
Lodge-Cottrell
UOP
Wahlco
Wahlco
Wahlco
References
Guiffre (1980)
Anon (1970)
SoRI (1972)
Green and
Landers (1974)
White (1974)
Brines and
Reveley (1978)
Midkiff (1979)
SoRI (1972)
-------
Detroit-. Edison
There are 17 Wahlco sulfur dioxide conditioning systems in
the Detroit Edison network. Brennan and Reveley (1977) reported
the following test results. At Marysville #11 there was a 10-
fold decrease in particle penetration through the electrostatic
precipitator with addition of 20 to 30 pprav sulfur trioxide.
They meet the regulations with low sulfur coal.
Iowa Public Service Company
Iowa Public Service uses a Wahlco sulfur-burning sulfur
trioxide conditioning system at George Neal, unit 2. Dismukes
and Gooch (1977) reported the results of a performance test. In
the last three years Iowa Public Service has attained:
1. A 100-fold decrease in particle resistivity.
2. A decrease of particle penetration from 9% to 1%.
They only have minor problems with the conditioning systems.
The units run very close to the compliance limit and an outage of
one transformer-rectifier set causes them not to comply.
Northern Indiana Puhlic Sprvire Company
NIPSCO has four Wahlco sulfur trioxide conditioning units,
one (on Dean H. Mitchell, unit 6) has been in use since April
1980. They had satisfactory performance on Mitchell 6 for the
first 8 months. Then intensified sparking and back corona were
noticed. They switched to a low sodium coal and decreased the
sulfur trioxide dosage from 40 ppm to 8 ppm. They have had
satisfactory performance since then.
Ohio Edison
Ohio Edison has four Wahlco units at W. H. Saramis, units 1,
2, 3, 4. These became operational in 1981. Ohio Edison reports
satisfactory performance on high resistivity ash when resistivity
is the limiting factor in ESP performance.
Pennsylvania Electric Company
Penelec has been using sulfuric acid injection at their
Front Street station for the last ten years. Acid injection is
41
-------
used as required to attain compliance. They have experienced
abnormally severe maintenance requirements.
Pennsylvania Power and Light
PP&L had five Wahlco flue gas conditioning units installed
in 1978 and 1979. They were installed at the following six
locations:
Brunner Island 1 and 3
Montour 1 and 2
Sunbury 3 and 4
Testing (Guiffre, 1980) indicated that they could burn any coal
and be in compliance at Brunner Island and Montour with a
conditioning level of 25 ppm. Guiffre reported an increase in
ESP efficiency from 90 to 95 percent and efficiency at the
Montour station of 99.6%. The sulfur trioxide systems have good
reliability.
Public Service Company of Colorado
PSCC is a pioneer in utility flue gas conditioning with
sulfur trioxide. They have seven units which use sulfur trioxide
or sulfuric acid.
Arapahoe 1. Unit 1 of the Arapahoe plant was converted to run on
coal instead on natural gas. A new electrostatic precipitator
was added with a Wahlco sulfur trioxide conditioning system. The
system has been operating since 1977 and attains compliance with
25 ppm sulfur trioxide (Brines and Reveley, 1978).
Arapahoe 4. A UOP sulfuric acid vaporizer was installed on
Arapahoe 4 in 1970 (Anon, 1970). There were many problems with
the acid delivery system. SoRI (1972) reports a 100-fold
decrease in ash resistivity with 6 ppm HiSOt. Green and Landers
(1974) report that even with conditioning, the ESP performance
did not meet the revised guarantee. The observed efficiency was
77% when the guaranteed efficiency was 87%. The DOP system was
replaced with a Wahlco conditioning system in 1979.
Cherokee 2. In 1971 PSCC installed a Lodge-Cottre11 liquid
sulfur trioxide vaporizer on their Cherokee plant, unit 2 boiler.
SoRI (1972) reported a 10,000-fold decrease of particle
42
-------
resistivity with 13 ppmv of sulfur trioxide conditioning. Also
the electrical behavior of the ESP was better with conditioning.
Green and Landers (1974) reported that the conditioned particle
collection efficiency was 95.2% which is better than the
guaranteed conditioned efficiency of 94.2%.
Problems. Green and Landers (1974) reported the following
problems with PSCC's sulfur trioxide systems:
1. Condensation of acid in the ducts.
2. Ash buildup at the injection nozzles.
ECONOMICS
Table 11 shows the estimated costs of a sulfur trioxide flue
gas conditioning installation as of December 1982. The capital
cost is about $5.15/kW and the operating cost is about 0.105
mills/kWh.
SUMMARY
Effectiveness
Sulfur trioxide (or its active equivalent, sulfuric acid) is
the most widely used flue gas conditioning agent because it has
been proven effective in those cases where particle resistivity
is the limiting factor in electrostatic precipitator performance.
High resistivity ash is most often the result of switching from
high sulfur to low sulfur and low alkali coals.
Tables 6/ 7, and 12 show that the addition of SO a reduces
ash resistivity and increases ESP efficiency. Most of the power
plants described above previously used high sulfur coal but now
use low sulfur coal. They also use cold-side (<180°C)
precipitators. Such utilities have had the most problems with
poor precipitator performance (viz. Borsheim, 1977; Cook and
Trykowski, 1976; Cragle, 1976; Dismukes and Gooch, 1977;
Klipstein, 1975).
Cases of non-effectiveness of sulfuric acid conditioning
stem from the following:
43
-------
TABLE 11. ESTIMATED COSTS OF A SULFUR-BURNING SOi
FLUE GAS CONDITIONING UNIT.
Capital Costs (Installed), $/kW
Conditioning Eqpt. $ 4.12
Dedicated Power ($1000/kW) 1.03
Operating Costs, mills/kWh
Sulfur at $125/Mg 0.044
Electricity at ($0.05/(kWh) 0.061
Basis: 500 MW coal-fired utility.
December, 1982 prices
40 ppmv design dosage
20 ppmv operating dosage
37.9% Plant thermal efficiency
[9,000 BTU/(kWh) Heat Rate]
23 MJ/kg Coal Heating Value
Source: Wahlco, Inc.
-------
TABLE 12. ESP PERFORMANCE FOR SEVERAL SELECTED USERS OF SOj CONDITIONING
Efficiency
SCA without with
Utility Plant Fuel Sulfur m,/(m3/s) SOj SOj
Commonwealth Edison
Detroit Edison
Crawford, Fisk
Joliet, Powerton
State Line, Waukegan
Will County
Connors Creek,Harbor
Beach, Marysville,
Pennsalt, Port Huron,
Trenton Channel
Monroe
0.5
0.5 to 0.7
2.3
(Blend)
24 to 26
87
37
85 to 90
97.5 to 98.5
98
99.8*
Iowa Public Service Neal
Public Service Co Arapahoe
Colorado
0.6 to 0.7
39
91
99
0.6 to 0.7 20 to 30 60 to 86** 14 to 50
(mg/Nm3)
~Combined Ammonia and Sulfur Trioxide Conditioning
~~Reference: Green and Landers (1974)
-------
1. The conditioner supply may malfunction. The lines may plug
or the SOi conversion may fall.
2. The precipitator performance may be limited by other
phenomena besides resistivity. The transformer/rectifiers
may be undersized. The superficial velocity may be too
high causing reentrainment. The velocity distribution may be
too irregular.
3. The temperature may be so low that acid condensation occurs
\
at the injection point.
4. The temperature may be so high (>200°C) that surface
conduction is not important.
5. The ash resistivity may already be satisfactory (approx. 10*
Ohm-m).
Problems
It is possible to over-condition the ash. The presence of
too much sulfuric acid in the flue gas can:
1. Lead to acid particle nucleation before mixing with the flue
gas.
2. Lower the resistivity so much that the ash is reentrained
from the collecting plates or bounces off them.
3. Form a solution between the particles on the plates which
binds them together and makes them difficult to dislodge
from the plate.
Most SOi conditioning systems have given 95% or higher
availability. The most reliable designs have been the ones which
use elemental sulfur and those which use sulfur dioxide even
though it is not commonly used.
46
-------
Section 4
AMMONIA CONDITIONING
Ammonia conditioning has been used successfully in
Australia. There the coals have very low sulfur (0.3 to 0.7
weight percent) and produce high resistivity ash (as high as 1012
Ohm-ro). Because of the success in Australia much research has
been done on ammonia conditioning in the U. S., yet there are few
users here.
CHEMICAL PROPERTIES
Ammonia is a vapor at room conditions. Table 13 shows the
properties of ammonia. The critical temperature is 132°C. Above
this temperature ammonia does not boil or condense. Therefore*
condensation on the fly ash would not be expected to occur with
ammonia above 132°C, although physical adsorption might.
Because of the high volatility of ammonia, it is injected in
the vapor form under its own vapor pressure. In Australia,
ammonia is also injected in solution form (McLean, 1976;
Tassicker, 1975; Watson, 1976; Walker and Lamb, 1978). A
concentrated ammoniacal liquor (CAL) by-product is available from
steel making operations. CAL contains about 17% ammonia (as
ammonia) by weight, both free and in the form of various salts,
notably chloride and bicarbonate.
CONDITIONING MECHANISM
The way in which ammonia affects the performance of electro-
static precipitators is not completely understood. Dismukes
(1975) reports the following effects:
1. Space charge enhancement. Ammonia reacts with the sulfuric
acid vapor present in the gas, forming a fume of fine salt
particles of (NHt^SO* and NH»HSO*. The electrons and ions
from the corona collide with and reside on the fine
particles. The mobility of these charged particles is
less than the ions, therefore the current density falls.
47
-------
TABLE 13. PROPERTIES OF AMMONIA*
Melting Point/ °C -78
Vapor Pressure, kPa
-33 °C 101.3
25°C 1010
132.4°C** 11,300**
Density, kg/m3
Liquid (20°C) 610
Vapor (20°C, 101.3 kPa) 0.71
Latent Heat (at -33°C) kJ/kg 1373
Heat Capacity/ J/(kg-K)
Avg/ 20-30°C 670
*Reference: Perry and Chilton (1972)
**Critical State
48
-------
The "gas phase resistivity" increases# thus the voltage drop
across the interelectrode gap rises and the voltage drop
across the dust layer falls. The field in the
interelectrode space is raised and the chance of electrical
breakdown in the dust layer is decreased. This allows
operation at higher applied voltages.
2. Increased cohesiveness. Sulfuric acid condensation in the
presence of ammonium sulfate or bisulfate leads to
adsorption of acid and salts to water on the surface. These
surface deposits are viscous and sticky# which increases
cohesion between neighboring particles in the dust layer on
the collection plates and reduces reentrainment.
Because of the uncertainty of such things as natural SOt
concentrations in the gas# quantification of the above effects
has been unpredictable.
INJECTION LOCATION
Ammonia can be injected either before or after the air
preheater. At Colorado Ute Electric Association# Hayden 1 and 2#
it is injected upstream of their hot side ESP. With injection at
this location the air preheaters require washing roughly twice a
year to avoid plugging with deposits of reaction products.
Reese and Greco (1968) tried ammonia injection upstream of
the air preheater at several plants at TVA. They experienced air
preheater plugging. The plugging problem was eliminated by
injecting the ammonia downstream of the air preheater. The
precipitator efficiency due to this relocation of injection
location was about the same.
Baxter (1968) also used ammonia injection upstream of the
air preheater. If the temperature in the air preheater were
above 200°C no plugging occurred. Cooler temperatures brought
plugging.
In Australia# where ammonia is the most popular conditioning
agent, ammonia was injected downstream of the air preheater at
all the ammonia conditioning installations (Watson, 1976).
49
-------
RESULTS
Particle Resistivity
The success of sulfur trioxide conditioning with
resistivity related raalperformance prompted investigations of
resistivity modification by ammonia. Tables 14 and 15 show a
summary of the results.
The ability of ammonia to alter resistivity is not clear.
Ammonia conditioned resistivities can be less than, the same as,
or greater than the unconditioned values. The temperature of the
precipitator has great effect on the sensitivity of the ash to
resistivity modification by ammonia. Ammonia is more effective
at low temperature (<110°C).
In two cases/ Stations "CI" and "C2" reported by Baxter
(1968) and at Detroit Edison Monroe 1 and 2, the addition of
ammonia raised the resistivity. In the case reported by Baxterr
the temperature was low and the resistivity before conditioning
was very low because of adsorbed or condensed sulfuric acid. The
injected ammonia neutralizes the acid and thus increases particle
resistivity.
The case of Detroit Edison-Monroe is not so clear. This
plant uses a medium sulfur blend of high sulfur Ohio coal and low
sulfur Kentucky coal. Simultaneous ammonia and sulfur trioxide
injection is used at Monroe. The addition of sulfur trioxide
lowers the resistivity to about 10s Ohra-ra, which is too low and
can lead to increased particle reentrainment. The addition of
ammonia alone or in conjunction with sulfur trioxide raised the
particle resistivity at Monroe.
The data of Baxter (1968), McLean (1976)), Bickelhaupt et al.
(1978), and Brown et al. (1978) show ammonia conditioning de-
creases the particle resistivity. McLean shows one case with a
four order of magnitude decrease in resistivity with ammonia
conditioning. Bickelhaupt and coworkers found that only acidic
ash responded to resistivity modification by ammonia, and then
only at low temperatures (<110°C). The work of Bickelhaupt et
al. did not take into account the effect of sulfur oxides on this
process. Brown et al. (1978) found that the resistivity of the
50
-------
TABLE 14. ESP PERFORMANCE WITH AMMONIA CONDITIONING
Fuel
NH j
Ash
Spec. Coll.
Coll.
Utility,
Sulfur
Dosage
Resistivity
Area
Effic.
Temp
Reference
Plant
wt %
ppmv
Ohm-m
m2/(m3/s)
%
°C
Baxter
Station
"A"
Low
0
10x#
204
(1968)
15
10»
85a
Station
"B"
Low
0
1010
149
15
5x10#
40a
Station
"CI"
Med.
0
5x10*
143
15
10*
70a
Station
"C2"
Med.
0
5x10*
143
15
10*
30a
Station
"D"
Med.
0
3x10*
204
15
5x10*
60a
Reese
TVA,
2.5 to
0
20 to 24
57
132
Greco
Widow1s
Creek
3.5
5
80 to 86
(1968)
Colbert
7.5
84 to 89
15
92
25
92
75
97
Dalmon
Laboratory
LOW
0
0.095*>
150
Tidy
Sulfur
50
0.123°
(1972)
Oil
180
0.180®
0C
0.136®
200c
0.14
^Percent decrease of particle concentration over unconditioned case
bEffective migration velocity, m/s
cWith 300 ppmv SOf
-------
TABLE 14. ESP PERFORMANCE WITH AMMONIA CONDITIONING (Contd)
Fuel
NH,
Ash
Spec. Coll.
Coll.
Utility,
Sulfur
Dosage
Resistivity
Area
Effic.
Temp.
Reference
Plant
wt %
ppmv
Ohm-m
m*/ (mVs)
%
°C
Dalmon
Laboratory
Low
10d
0.151b
Tidy
Sulfur
180
0.166
(1972) cont'd
Oil
10e
0.145°
180e
0.161
McLean
ECNSW
0.3 to
0
1011
0.06b
(1976)
(Australia)
0.4
40 to
55
0.08
Mobile Pilot
0.36
0
3x1012
30 to 72
0.07b 116
to 149
ESP
40
«10#
0.09
0.3 to
0
1011
0.04b
0.4
55
0.08b
0.6 to
0
2xl01J
0.05b
0.7
50
0.06b
0.4
0
>101J
0.037b
17
0.06b
Southern
TVA,
3.59
15
2x10*
25
124
Research
Widow's
0
4x10*
127
Institute
Creek "B" 7
7
3xl07
135
(1972)
7
3xl07
137
15
2x10*
139
15
2xl07
141
15
5x10*
142
0
3x10*
143
15
6xl07
143
^Effective Migration Velocity
, m/s
With
100 ppmv HC1
eWith
300 ppmv SOf and
100 ppmv HC1
-------
TABLE 14.
ESP PERFORMANCE WITH AMMONIA CONDITIONING (Contd)
Ui
u>
Reference
Potter &
Paulson
(1974)
Di smukes
(1975)
Tassicker
(1975)
Ashton
(1976)
Utility,
Plant
CSIRO
(Australia)
Pilot ESP
TV A,
Widow1 s Creek
Bull Run
Gallatin
ECNSW
(Australia)
Pacific Pwr &
Light, Bridger
Fuel NH,
Sulfur Dosage
wt %
0.5
0.9
3.5
1.2
4
0.4
0.6 to
0.7
ppmv
0
20
0
10
0
20
0
7
0
20
0
>0
>0
>0
>0
0
>0
Ash Spec. Coll.
Resistivity Area
Ohm-m m*/(m3/s)
45
4x109
4x10 *
1x10®
3x10s
3xlOe
4x10*
4x10*
3x10*
20 to 24
28
Coll.
Eff ic.
%
92
96
90
98
87
>99
Unchanged
350?
Temp.
°C
120
132
127
143
70
60 to 80
97 to 9 8.5
99 to 99.5
99.5
52 to 60
70 to 80
95
110
55 to 60 98.9 to 99.1
No Effect
No Effect
No Effect
No Effect
Watson
(1976)
ECNSW 0.6
(Australia)
0
>0
74
94
97 to 98
116
^Particle Concentration, mg/Nm3
-------
TABLE 14. ESP PERFORMANCE WITH AMMONIA CONDITIONING (Contd)
Fuel NHj Ash Spec. Coll. Coll.
Utility, Sulfur Dosage Resistivity Area Effic. Temp.
Reference Plant wt % ppmv Ohm-m ml/(m3/s) % °C
Bickelhaupt Laboratory Hi #1 0 lxlO10 0 110
et al. 37 3x10* 110
(1978) 100 to 120 4x10' 112
0 2xl010 144
100 to 120 9x10* 144
0 4x10' 352
100 to 120 4x10* 352
Laboratory Hi #2 0 8x10* 112
37 7x10* 110
100 to 120 3x10* 112
ui 0 2x10* 144
100 to 120 10* 144
0 3xl07 352
100 to 120 3xl07 352
Lo #1 0 6xl07 70
100 to 120 8x10* 70
0 4x1010 110
37 4xl010 110
100 to 120 7x10* 112
0 8xlOl0 144
0 107 352
100 to 120 107 352
Lo #2 0 2x10* 110
37 2x10* 110
118 2x10* 110
-------
TABLE 14. ESP PERFORMANCE WITH AMMONIA CONDITIONING (Contd)
ui
ui
Buel
nh,
Ash
Spec. Coll.
Coll.
Utility,
Sulfur
Dosage
Resistivity
Area
Effic.
Temp.
Reference
Plant
wt %
ppmv
Ohm-m
m2/(m3
/s)
%
°C
Brown
Pilot Plant
0.3
0
3x10*
85
et al.
5
2
to 5x10*
87
(1978)
50
1
to 5x10*
88
to 90
78
3
to 7x10*
Walker
ECNSW
0.3 to
0
0.2to0.4^
90tol25
Lamb
(Australia)
0.7
20to30
0.4to0.6°
90tol30
(1978)
0
0.2to0.3°
162to210
20to30
0 . 3to0.6
150to205
Katz
Case "B"
2.3
0
26
71.6
119
(1979)
2.8
0
26
63.4
139
2.8
10
26
84
139
2.3
0
26
98
163
2.3
32
26
96.5
163
Case "C"
1.73 to 0
42
96.1
132
1.76
12
42
99.2
132
4.5
42
99.4
141
Castle
Australia
Low
0
45
91
to 92
(1980)
20
96
-------
TABLE
Utility
Colorado-Ute
Elect. Assoc.
Detroit Edison
Plant
Hayden
Monroe 1,2
Tennessee Valley Widow's
Authority Creek "B" 7
ESP PERBORMANCE WITH AMMONIA CONDITIONING
Specific
Buel Added Ash Collection Coll.
Sulfur NHj Resistivity Area Effic.
% wt ppmv Ohm-m mV(mJ/s) %
0.5 0 97.8
100 99.4
2.3
(Blend)
11 to 12
11 to 12
(15 ppmv SO»)
3x10*
0.3 to 6x10*
0.5 to 1x10*
37 97.5 to 98.
99.8
3 to 3.5
10 to 15
25
75
85 to 90
-------
ash decreased at first with small amounts of ammonia, but
addition of more ammonia Bhowed no change.
Two tests of the Tennessee Valley Authority's plants showed
little or no effect of ammonia conditioning on ash resistivity.
Southern Research Institute (1972) reported that the
resistivities correlated with temperature but not with ammonia
dosage at TVA's Widow's Creek plant with high sulfur coal.
Dismukes (1975) showed no effect of ammonia conditioning at the
same plant on both high and low sulfur coal, and at the Bull Run
and Gallatin plants on medium and high sulfur coals.
Spafford et al. (1979) did experiments on the fate of
ammonia in a laboratory ESP and a flue gas train setup without
fly ash. A summary of the data is shown in Table 16. They found
the following:
1. A significant part of the ammonia (20 to 30 ppm out of 63
ppm added) reacted with nitrogen oxides. The reaction
product is probably elemental nitrogen.
2. In the presence about 30 ppmv SO i, a stoichiometric amount
of the ammonia reacted with sulfuric acid to form ammonium
bisulfate aerosol. Practically all of this aerosol was
collected by the ESP. The effective ion mobility in the ESP
decreased (a space charge effect).
3. In the presence of low amounts (1 to 2 ppm) of SOi, the
ammonia neutralized the sulfuric acid, but no change in
electrical conditions was noticed.
The Southern Research Institute (1972) and Dismukes (1975)
performed field tests to determine the destination of the ammonia
injected at TVA plants. Their data also appear in Table 16. The
tests of SoRI (1972) found negligible ammonia in the fly ash.
They also could not account for all the ammonia injected. The
recoveries were less than 50%.
In the test described by Dismukes (1975), there was
a reasonable material balance on the ammonia. Almost all of the
ammonia ended up in the solid phase before it reached the
57
-------
TABLE 16.
PARTITION OF ADDED AMMONIA BETWEEN GAS AND PARTICLES
Added
Gas Phase
Particle
Temp.
NH,
SOi
NH,
SO,a
NB s
Reference
°c
ppmv13
ppmv*3
ppmvb
mg/Nm3b
mg/Nm,b
Southern
142
0
15
0
70to80
0
Research
135
7
13tol5
0.2
80tol20
<2
Institute
140tol43
15
lOtol8
0.4
88to97
<2
(1972)
Dismukes
132
0
5
0. 6C
33
<1.5
(1975)
10
1
0.8
30
6.7
127
0
2
<0.3
7
1
<0.3
132
0
11
<0.1
117d
-------
electrostatic precipitator. There was also marked reduction of
the SOj entering the ESP.
The results of Dismukes (1975) and Spafford et al. (1979)
reveal that the sulfur trioxide vapor emission from the plant
would be decreased with the addition of ammonia and that most of
the ammonia and all of the sulfate particles were removed in the
ESP.
Effect on Particle Composition
A summary of the investigations of particle composition
change is given in Table 17.
The Southern Research Institute (1972) analyzed ash from the
inlet of the ESP in TVA Widow's Creek and found the sulfate
contents correlated with temperature only, not with added ammonia
concentration. The ammonia in the ash was negligible.
Dismukes (1975) analyzed the ash from several TVA plants
and found no change in pH with ammonia conditioning. Contrary to
SoRI's findings at the same plant earlier (1972)/ Dismukes also
found significant increase in the particle ammonia content with
ammonia conditioning.
Baxter (1968) determined the equilibrium acidities of ash
from several facilities which use ammonia conditioning and found
the ash was acidic in all cases. It seems that ammonia always
raises the pH of the ash.
Effect: on Particle Size Digfcrihufcion
The effect of ammonia conditioning on particle size
distribution is not known because the outlet particle size
distributions with and without conditioning have not been
published. Dismukes (1975) reported the formation of fumes at
the inlet of the precipitator with ammonia injection. The total
number of nuclei in the 0.005 to 1.0 nm diameter range increased
about 3 times with ammonia conditioning (Table 16). Spafford et
al. (1979) speculated that the aerosol is ammonium bisulfate.
59
-------
TABLE 17. CHANGES IN PARTICLE COMPOSITION
DUE TO AMMONIA CONDITIONING
Researcher
Baxter
(1968)
Inlet Particle Properties
Injected
Utility
Station "A"
NHj Loading SOT"
ppmv rag/Niti* % wt
0
15
NH t
% wt
PH
5.5
7
Station "B'
0
15
5
6
Station "C1
0
15
3.5
4.5
Station "D"
0
15
3
4
Southern
Research
Institute
(1972)
TVA, Widow's
Creek
15
9,400 0.73to <0.02 11.0
0.88
9,400 0.87to <0.02 10.5to
1.29 10.9
9,400 0.86to <0.02 9.5to
1.21 10.6
Disniukes
(1975)
TVA
Widow's Creek
0 15,200
10 16,700
0.22
0.18
1.24
<0.01
0.04
<0.01
5.2
5.4
10.0
20
Bull Run 0
7
Gallatin 0
20
1.31 0.12 10.0
0.31 <0.01 4.5
0.33 0.05 4.4
1 <0.01 8.6
1.4 0.21 8.6
60
-------
TABLE 18. CHANGE OF PARTICLE NUMBER CONCENTRATION WITH
AMMONIA CONDITIONING (Dismukes, 1975)
Nucleus
Ammonia Concentration
Plant ppmv m"1
TVA Bull Run 0 2.4X101
7 9.8x101
Widows' Creek 0 1.2x10*
20 3.1x101
Gallatin 0 2.2X101
20 6.1x101
61
-------
Operating Prnhlpms
Investigators in Australia (Walker# 1977, Watson, 1976;
McLean, 1976) reported the following operating and maintenance
problems:
1. The normal problems associated with pressurized gases, such
as: leaks, vaporizer freezing in cold weather, freezing of
water in the lines.
2. The injection nozzles may plug during periods of extended
shutdown because of buildup of dust and ammonium carbonate
deposits. The use of purge air prevents this buildup.
3. Deposits are noticed on the discharge electrodes of the ESP.
They are "sausage-like shapes or spikes of hard dust"
(Walker and Lamb, 197 8).
Effect on Plume Opacity
Dismukes (1975) reported oscillation in the opacity at the
TVA-Gallatin plant before conditioning with ammonia. The opacity
oscillation followed the rapping cycle of the ESP and slowly
vanished after injection of ammonia.
Changes in the opacity oscillations were not clearly evident
until approximately 30 min after startup of the conditioning
system. After two hours the opacity reached steady-state. When
injection of ammonia stopped, the opacity increased and the
oscillations from rapping returned after 2 hr.
Effect
According to Dismukes (1975), photomicrographs of the fly
ash showed particles bridged by a feathery material thought to be
ammonium sulfate. Cascade impactor studies indicated that the
conditioned particles bounced less.
ECONOMICS
Reese and Greco (196 8) reported the installation costs for
ammonia injection facilities at TVA. The costs were updated to
December, 1982 U.S. Dollars based on Chemical Engineering Plant
Cost Index and are presented in Table 19. The capital and
62
-------
TABLE 19. ECONOMICS OF AMMONIA CONDITIONING
Capital Costs, $/kW
Conditioning Equipment $0.16
Dedicated Steam Facilities 0.05
at $240,000/(Mg/hr)
Operating Costs, mills/kWh
Ammonia at^$325/Mg $0,018
Steam at at $12.30/Mg 0.004
Basis: 550 MW coal-fired utility
December, 1982 prices
100 ppmv design dosage
50 ppmv operating dosage
37.9% Plant thermal efficiency
(9000 BTU/(kWh) Heat Rate)
23 MJ/kg Coal Heating Value
References: Reese and Greco, 1968;
63
-------
operating costs are $0.21/kW of plant capacity and 0.022
mills/kWh.
USER EXPERIENCE
Table 20 is a list of ammonia conditioning users. A summary
of user experience is presented in the following paragraphs.
Cleveland Electric Tllaminating Co. (CET)
Ammonia conditioning has been used at the CEI Avon Lake and
East Lake plants for about four years. The coal is a variable
blend of high sulfur Ohio coal and low sulfur Pennsylvania coal.
The precipitators were guaranteed for 99.5% efficiency and have
SCA (specific collection areas) of 37 m2/(m,/s) (190 ft*/l,000
acfm) at Avon Lake and 44 m2/(m,/s) (225 ft2/l,000 acf) at East
Lake.
Ammonia conditioning is used as required to obtain
compliance with the particulate regulations. East Lake complies
with the particulate regulations only part of the time. They are
forced to limit coal feed to the boiler (therefore reducing power
production) to meet the compliance requirements. Avon Lake is
not in compliance/ even with ammonia conditioning. CEI plans to
add another precipitator in series at Avon Lake.
. In general CEI would rather not depend on flue gas
conditioning at their plants in order to meet the pollution
regulations. The precipitators they are planning to install on
future projects will have a minimum SCA of 120 m,/(m,/s) (600
ft2/lP000 acf), especially if low sulfur coals are burned.
Colorado-Ute Electric Association
Colorado Ute has used ammonia at their Hayden units 1 and 2
for three years. These are hot side ESP's. Their dosage ratio
is 1 ton of ammonia per 2,700 MW-hr (about 100 ppmv). The
particle mass collection efficiency has improved from 97.8% to
99.4%. Their fuel is 0.5% sulfur coal.
64
-------
TABLE 20. LIST OF PAST AND PRESENT USERS OF AMMDNIA
Company Name
Plant Name
j2GC Licensor
References
Central Illinois Public Service
Cleveland Electric Illuminating Co.
Cleveland Electric Illuminating Co.
Colorado Ute Electric Assoc.
Columbus & Southern Ohio Electric
Detroit Edison
Ohio Edison Co.
Pacific Pcwer & Light
Pennsylvania Electric Co.
Tennessee Valley Authority
Tennessee Valley Authority
Ifexas Utility Generating Co.
Avon Lake
East Lake
Hayden
Gonesville
Monroe
W. H. Sanmis
Jim Bridger
Keystone
Gallatin
Widows Creek
Monticello
Research-Cottrell
Re search-Gottrell
Research-Cottrell
Research-Cottrell
Flakt/Carborundum
"B"
Locklin
et al. (1900)
Ashton (1976)
Katz (1979)
Reese and
Greco (1968)
Dismukes (1975)
Research-Cottrell
-------
Detroit Edison uses ammonia and sulfur trioxide on Monroe
units 1 and 2. The fuel is blended high and low sulfur coalsf
averaging 2.3% sulfur. Detroit Edison meets the opacity
requirements when using the combined conditioning agents. Monroe
units 3 and 4 burn the same coalr but their precipitators have
30% higher specific collection area and do not need conditioning.
Ohio Edison uses ammonia on their Sammis plant, units 5, 6
and 7. In general ammonia is a useful conditioning agent for low
temperature applications (below 150°C) in flue gases from medium
sulfur coal.
Pari fir. Powpr and Light
Preliminary testing on the Pacific Power & Light, Jim
Bridger Plant showed that ammonia conditioning had no effect on
the precipitator performance. The fuel is 0.6% sulfur Wyoming
coal.
Pennsylvania Electric
Penelec uses ammonia intermittently at Keystone units 1 and
2. They use a medium sulfur coal (1.5 to 2.0% Sulfur). The gas
temperature at the precipitator is low (approx. 120°C) and non-
uniform. Ammonia is used when the flue gas temperature is very
low (winter). With a dosage of 4 to 6 ppmv the efficiency is
raised from 96.1% to 99.4% (Katz, 1979). The electrical
conditions in the precipitator are improved with ammonia. No
sparking is noticed without conditioning; slight sparking is
noticed with conditioning. Penelec has problems with dust
buildup on the plates and hopper plugging with ammonia
conditioning.
TVA has been working with ammonia flue gas conditioning for
-------
twenty years. Reese and Greco (1968) and Disraukes (19760
reported the results of earlier work.
Today the TVA uses ammonia only at Widows Creek "B" unit 7.
The fuel is 3 to 3.5% Sulfur coal. The temperature at the
precipitator is low (less than 150°C). The precipitator is small
(approx. 25 mVmVs specific collection area at full load). With
10 to 15 ppm of ammonia conditioning the particle collection
efficiency is raised from 70% to 85 or 95%. This still is not
high enough to meet compliance requirements. A scrubber has been
added to meet the particle regulations and to remove sulfur
oxides. The electrostatic precipitator and ammonia conditioning
systems are still used to protect the induced draft blowers.
SUMMARY
Effectiveness
The data in Tables 14 and 15 show ammonia conditioning to be
effective in increasing the precipitator collection efficiency.
This increase in efficiency does not correlate with any change in
resistivity of the fly ash.
Unlike sulfur trioxide conditioning, effects of ammonia
conditioning occur almost immediately after injection starts.
Sulfur trioxide conditioning must penetrate the whole inventory
of ash on the collection plates to show any effects, which takes
on the order of 8 to 24 hr. Better electrical conditions (higher
voltages, lower currents, higher specific power) are noticed
immediately on addition of ammonia. Decreased oscillations in
the stack opacity are noticed in 30 min to 3 hr.
The magnitudes of the response times indicate the following:
1. The immediacy of the improvement of electrical conditions
bears out the space-charge effects discussed by Dismukes
(1975) which reduce the voltage drop across the dust layer
and increase the voltage drop in the free space.
2. The slower effect on opacity suggests that part of the fly
ash inventory on the collection plates is made more adhesive
by the ammonia and as a result, reentrainment is reduced.
67
-------
Ammonia has a low but non-zero electronegativity (White,
1953). Absorption of the free electrons created in the corona
zone is improbable/ but possible. This absorption of the elec-
trons to form ions of lower mobility may explain the effect of
ammonia on precipitator electrical conditions.
The existence of the space-charge effect depends on the
presence of efficient charge carriers in the gas. The formation
of aerosol ammonium bisulfate fumes (or sulfate, depending on the
ammonia-sulfuric acid stoichiometry) creates a large surface area
for collection of electrons. These charge carriers will have a
lower mobility than electrons (or molecular ions, for that
matter). A lower ion mobility allows a more stable corona,
higher electrical field strength without breakdown, and larger
specific power.
PrnhlpmR
Plugging of the air preheater is noticed if ammonia is
injected upstream of that device. Plugging of the injection
nozzles may occur if purge air is not used during shutdowns.
Deposits may build up on the discharge electrodes.
68
-------
Section 5
AMMONIUM COMPOUNDS
Conditioning with ammonium compounds offers a more
convenient way of injection of ammonia. The commonly used
ammonium compounds are sulfamic acid, ammonium sulfate, and
ammonium bisulfate. Once injected, these compounds could
decompose to ammonia and sulfuric acid and therefore, may provide
a combination of the effects of ammonia and sulfuric acid
conditioning.
Many proprietary flue gas conditioning agents contain
ammonium compounds. The experience with these agents is also
presented in this section.
CHEMICAL PROPERTIES
Sulfamic acid (NH2S020H) can be considered the anhydride of
ammonium bisulfate, an ammonium-sulfate compound with NHj to S03
stoichiometry of 1:1. In the crystalline state, it is present as
a bipolar ion, +HNH2SO»~. Its ease of ionization makes it quite
soluble in water. Since it is bipolar the molecules are bound
tightly in the crystal, making its vapor pressure low. The
properties of sulfamic acid are presented in Table 21.
Sulfamic acid decomposes to produce SO* and NHi when
injected into the flue gas. Dismukes (1974) reported that above
380°C the sulfamic acid decomposes according to the following
equation:
NH 2SOt OH—> 1/6 SOj + 5/6 S02
+ 1/2 H20 + 5/60 N2 + 2/3 NH,
+ 1/6 NOt (1)
The decomposition rate is slow and therefore long residence time
is needed.
At temperatures around 200°C and in the presence of water,
sulfamic acid may be hydrolyzed to ammonium bisulfate.
69
-------
NH2S020H + H20 —> NHtOSOiOH
(2)
This would be an effective compound for coating the surface of
the particles because ammonium bisulfate is liquid above 144°C to
147 °C.
Ammonium sulfate
The solubility, acidity, and decomposition temperature for
ammonium sulfate (NH*)tSO» are shown in Table 21. The partial
pressures of ammonia and sulfuric acid above ammonium sulfate are
given in Table 22. The rate of vaporization of ammonium sulfate
are shown in Table 23.
Ammonium sulfate is a weak acid in solution, but is highly
soluble. As a crystal it has a low vapor pressure. These are
similar to the attributes that make sulfuric acid effective.
In addition, the presence of ammonia in equilibrium affects the
ion mobility in the gas phase and may make the fly ash particles
more cohesive.
Ammonium sulfate will decompose to NH3 and S0» at elevated
temperatures. Spafford et al. (1979, 1980) give the following
decomposition reaction for ammonium sulfate at 650°C:
(NH *)2SO4 —> 2 NH3 + HzO + SO3 (3)
On cooling to 160°C Spafford et al. found that liquid
ammonium bisulfate was the major recombination product. Further
cooling to 90°C produced ammonium sulfate.
AflUBQfll Uin bisulfate
The properties of ammonium bisulfate are presented in Tables
21 through 23. Like ammonium sulfate, ammonium bisulfate
decomposes before it boils at atmospheric pressure. The
decomposition products are one mole each of ammonia, sulfur
trioxide, and water for one mole of ammonium bisulfate.
70
-------
TABLE 21. PROPERTIES OF AMMONIUM SULFATE COMPOUNDS
Solubilities
g/kg
pH at Mel ting Pt.,
25°C for °C
Q .1 mol/L
Decomposition
Temperature,
°C
Sulfamic Acid
H2NSO2OH
250
1.25
205
323-467
Ammonium Sulfate 790
(H»NO)2S02
5.5
510
(Closed system)
23 5
Ammonium bisulfate 1,000
H»NOS02OH (25°C)
1.6
144-147
<200
-------
TABLE 22. EQUILIBRIUM PARTIAL PRESSURES OF AMMONIA
AND SULFURIC ACID IN EQUILIBRIUM WITH AMMONIUM
SALTS AT VARIOUS TEMPERATURES3
Temp,
(NH»)
2 SO k
NH*HSO
°C
[NH3
]r
atm
tHzSOk
], atm
[NH,] -
fH*SO»
120
4.40
X
10"'
2.20
X
10~'
1.58
X
10~7
144
3.32
X
10~7
1.66
X
10"»
1.26
X
10"8
175
3.28
X
10~s
1.64
X
10~s
1.27
X
10"s
200
1.67
X
10"5
8.36
X
10~s
5.33
X
10'5
225
7.22
X
10"s
3.61
X
10~5
2.17
X
10"*
300
2.72
X
10"3
1.36
X
10~3
6.98
X
10~»
a Dismukes, 1974.
TABLE 23. WEIGHT LOSSES BY VOLATILIZATION OF AMMONIUM
SULFATE OR AMMONIUM BISULFATE AT A HEATING
RATE OF 4.0 °C/MINa
Temperature, Cumulative
Compound ^C weight loss, %
(NH»)2 SO* 200 1
250 2
300 15
350 100
NHtHSOt 200 1
250 3
300 8
350 25
400 100
a Dismukes, 1974.
72
-------
Ammonium bisulfate is highly soluble in water and is a good
electrolyte. It is a strong acid and is non-volatile. Ammonia
bisulfate decomposes easily and exibits many of the effects of
both sulfuric acid and ammonia conditioning. It may improve
electrical conditions by decreasing ion mobility and decrease
rapping losses through greater particle cohesion. It can also
decrease the particle resistivity.
CONDITIONING MECHANISM
Ammonium compounds improve ESP performance through one or
more of the following mechanisms:
1. Decrease resistivity
2. Increase space charge
3. Increase cohesivity.
Resistivity
Ammonium bisulfate and ammonium sulfate are non-volatile
acids like sulfuric acid. They can form layers of conductive
solution on the surface of the fly ash collected on the plates
and lower the electrical resistivity of the ash layer.
Space charge
The decomposed ammonia and sulfuric acid recombine to form
fine particles which accept the electrons from the corona
discharge, which form ions of lower mobility. The decreased
mobility balances the voltage drop in the interelectrode region
between the gas phase and the dust layer. With an adequate
electric field in the interelectrode region, the charged
particles can migrate faster.
Cohesiveness
The condensation of viscous liquid or the adsorption of
reactive compounds on the surface of the fly ash particles
increases their affinity for each other on the collecting plates.
Greater cohesion between the particles decreases the amount of
73
-------
the particles which become reentrained when the plates are
cleaned by rapping.
Conditioning with anhydrous ammonia has been shown to
increase fly ash cohesiveness (Dismukesf 1975). it is thought to
react with sulfuric acid to form some compound or compounds which
bridge the fly ash particles together. Conditioning with
ammonium sulfate compounds is expected to do the same.
INJECTION LOCATION
Ammonium compounds in solution can be injected either
upstream or downstream of the air preheater.
1. Upstream of the air preheater. This position offers a
temperature of 600°C or greater and a residence of several
seconds.
2. Downstream of the air preheater. The temperatures
downstream of the air preheater are lowf about 120°C to
200°C. Residence time usually is less than 1 second.
In most cases the agents are injected upstream of the air
preheater because long residence times and high temperatures are
required to vaporize or decompose them. The major disadvantage
of injecting at this location is that the air preheater may be
plugged and corroded by the reaction products of the conditioning
agents.
RESULTS
Results for conditioning wih ammonium compounds are
summarized in Table 24. Table 25 shows the results for
proprietary chemicals.
Effect on Particle Reaistiviy
Dalman and Tidy (1972) and Brown et al. (1978) showed that
ammonium sulfate, sulfamic acid, and ammonium bisulfate are
effective in reducing the ash resistivity because these compounds
can either form a conductive layer of solution on the surface of
the particles or decompose to produce SOj. Landham et al. (1981)
determined the effects of Apollo LPA-40 and generic ammonium
74
-------
TABLE 24. ESP PERFORMANCE WITH AMMONIUM-SULFATE COMPOUNDS
Reference
Utility
Plant
ruel
Sulfur
Htt
Agent
Doeagc
ppmv
fieBlst-
lvlty
ohm-cm
Specific
Coll Area
m'/tm'/s)
Coll,
Effi-
ciency
t
Temper-
ature
•c
Dalmon
Labora-
Low (NH.),SO.
0
7
•
o
H
K
0.095b
Tidy
tory
Sulfur
55 to 60° 1
to
2
* 10»
0.125 to 0.13b
(1972)
Oil
67"
3
X
10'
0.1 4b
125®
3
X
109
0.12b
140® 2
to
3
x 10«
0.165b
NH.BSO,
110®
9
X
10*
0.125b
127®
4
X
10*
0.13 5b
NH tSOaON
55a
1
X
10"
0.142b
85®
1
.5
X 10'
0.142b
90®
0.13b
Station "A"
1 Sulfamic
0
90
Pilot
Acid
0.25c to 0.75c
96 to 99
DlBmukes
AEPSCo
0.8 Sulfamic
0
95.3
(1974)
Appalachian
Acid
u
©
95.1
Power
o.e
0
78.6
Cabin Creek
u
o
89.7
0. 9C
84.6
1.0
0
95.3
0.65c
95.1
1.0
0
68.9
0.25c
88.2
o
n
87.6
a
b
c
maol/kg on the ash
effective nlgratlon velocity, m/s
g/kg coal
-------
Fuel
Utility Sulfur
Reference Plant Wt%
TABLE 24. continued
Reslst-
Dosage lvlty
Agent pprav ohm-cm
Specific Effi- Temper-
Coll Area ciency ature
m'/tm'/s) * °c
Disoukes
(1974)
Station *B*
Pilot
Station "C"
Pilot
Station *D*
Pilot
Station "B"
Pilot
Unnamed
2.5
1.0
0.5
2.5
Ammonium
Sulfate
Ammonium
Sul fate
Ammoni um
Sulfate
Ammonium
Bisulfate
Ammonium
Sulfate
0
15
0
15
0
15
0
15
0
15
10*
10*
10"
10'
10"
270c
120^
300'
160*"
58'
16<"
270'
110'
50
150
150
130
150
140
-J
CT>
Brown
et al.
(197 8)
Pilot
Plant
0.3
Sulfamic
Acid
0
5
8 to 10
11
65
73
64
98
0.5 to 3 x
10li
1 to 8 x
1 to 3 x
10'
10'
3 x 10'
10'
10"
10'
10«
Ammonium
0
3 x
10*
Sulfate
3
to 4
0.4 to
1 x 10'
5
to 10
2 to
O
H
X
00
20
to 30
7 to
©
X
t-*
©
91
4 x
10*
Ammonium
0.
,76d
3 x
10*
Sulfate
99.92
d
e
particle concentration mg/Nm'
percentage reduction of particle emissions over unconditioned case
-------
Reference
Sparks
(1976)
Utility
Plant
Pennsylvania
Power t Light
Montour
Fuel
Buifur
Wt«
1.5
TABLE 25. ESP PERFORMANCE WITH PROPRIETARY COMPOUNDS
Coll,
Resist- Specific Effl-
DOBage lvity Coll Area ciency
ppmv ohm-cm m'/tm'/s) »
0.38"
10*
3 to 5 x 10'
35
97
99.6
Temper-
ature
•c
13 6 to 147
Dlsmukes
(1974)
New Jersey
Gas t Elec
Mercer
0.56
1.42
0
0.25c
0.38c
0. 5C
2. 0C
0
0. 6C
95
26b
3 9b
22t
22t
27t
30b
Cragle
(1976)
Pennsylvania
Power t Light
Montour 1
Montour 2
2.5
1
1
2.4
1.3
1.3
0
0
0.4e
0
0
1.4e
41
41
43 to 86'
258 to 387f
172 to 215f
260' 138 to 154
1200 to 5100f
1 650 to 860f
McNlnch S. Carolina 0.97 0
(1976) Pub Svc Auth to 0.6
Jefferles 3(4 1.15 0.6
cm'/kg coal
ng/J boiler duty
g/kg coal
LPA-40 formulation was mainly sulfamic acid until 1974,
cm'/kg coal. 1 gal/ton Is 4.17 cm'/kg.
ng/J boiler duty. 1 lbm/MMBTu ¦ 430 ng/J.
at reduced load (611 of full),
at 86t of full load
a
b
c
d
e
f
g
h
344*9
155f«
129
fh
then it was changed to ammonium sulfate (Cragle, 1976).
-------
TABLE 25. continued
Reference
Boraheln
(1972)
Preacey
et al.
(1979)
Otillty
Plant
Montana
Pove r
Corvette
Arlxona
Pub Svc Co
Four Corner!
ru«l
6ulfur
Ntt
1.2
Do 8 a 9e
PP"v
0®
0.29®
0.54®
0.4J
0®
0.4®
Reaiat-
lvlty
Ohra-cm
Bennett 6
Koker (1978)
II
3.2 to
3.9
0
0.83®
12
0.}
0
0.29'
-J
00
1}
0.8 to
l.S
0
0.63®
0.63d I 0.42'
Bennett &
Kober
(1978)
I 4
1.3 to
1.7
0.6
to
1.0
0.31® I 0.42®
0
0.21®
0.42®
0.42® t 0.42®
crn'/kg coal
nq/J boiler duty
at 148 KW load
at 158 KM load
at 163 KM load
at 173 HW load
opacity, percent
Coll,
Specific Effl- Temper-
Coll Area ciency ature
ra'/fn'/a) I *c
108f 1
50 to 80l)
80fk
68 to 73fl
95.2 121
96.6
27.4 98.8 131
99.5
69 22 to 30m 366
12m
34 161f 127
116'
45f
34 925f 133
39f
37 293d 16B
176d
155d
86d
-------
Reference
Otlllty
Plant
Fuel
Sulfur
Ht*
Landhan
et al.
(1981)
Montana
PWE
Corvette
Tampa
Elect.
Gannon
0.7
0.7-1.0
1.1-1.2
Patterson Unnamed 0.9-1.6,2
et al.
(1979 b)
Patterson
et al.
(1961)
Tampa
Elect.
Gannon
1.0 to 1
TABLE 25. continued
Coll,
Resist- Specific Effi- Temper-
Dosage ivlty Coll Area ciency ature
ppmv ohm-cm m'/lm'/s) % *c
0 3 * 10' 28 77.3 130
0.4d 10" 97.5
0 2 x 10' 65 to 68 99.90 160
to
0.66d 3 x 10' 170
0.29d 3 x 10* 99.96
0 32 to 35 99.7 107 to
•42d I 0.31d 99.9 117, 141
0 98.9 147 to 165
0.63d 99.5
-------
sulfate on ash resistivity in full scale installations and found
both reduce resistivity (Table 25).
Pressey et al. (1979) and Patterson et al. (1979b) reported
the field test results of Apollo LPA-455, which is a solution of
diammoniuro hydrogen phosphate. They found that LPA-445 has no
effect on particle resistivity.
It seems that in roost cases ammonium compounds can lower the
ash resistivity. However* it is not as effective as SOj. With
ammonium compounds# the resistivity decreases by about half an
order of magnitude.
Cobesiveness
Ash cohesiveness is usually indirectly measured by the
magnitude of the opacity decrease during rapping. An opacity
trace with smaller periodic spikes can be attributed to decreased
reentrainment due to rapping.
Patterson et al. (1979b) show an opacity meter trace for a
plant using Apollo LPA-445 conditioning. The baseline (no
conditioning) opacity was 15-18%. The spikes corresponding to
rapping ranged from 40% to greater than 60% opacity. With LPA-
445 conditioning the inter-rapping opacity was 12 to 13% and the
peaks of the rapping spikes were between 14 and 21% opacity.
Therefore LPA-445 is effective in increasing ash cohesiveness.
This could be due to the phosphates in the agent which form
polymers on decomposition. The coprecipitation of these polymers
or the coating of the fly ash surface with the polymers could
increase the cohesivity.
No information was reported for the other ammonium
compounds.
Pressey et al. (1979) performed a simultaneous impactor,
optical, and condensation nucleus sampling at the Four Corners
station of the Arizona Public Service which used Apollo LPA-445
conditioning agent. They found a significant increase in the
numbers of fine particles. The condensation nucleus count was
80
-------
1010 to 101* m-' with conditioning versus 10* m-3 without. The
optical particle measurements showed a shift in size from small
particles (0.3 - 2 nm) to larger ones (2-3 nm) when conditioning
was shut off.
Rial unions Caused by fche Agents
Table 26 shows the partition of injected ammonium compound
as reported by Landham et al. (1981). The data were collected in
Gannon Station of Tampa Electric. It appears that the ammonium
sulfate is distributed between the gas phase and particulate
phase entering the ESP. Most of the ammonia produced is
collected in the ESP.
Patterson et al. (1981) reported sampling results at the
same plant. They reported an increase in the particulate sulfate
and iron compound emissions with Apollo LPA-40 conditioning
agent. The particulate sulfate emission was about 20 p.g/nt¦ and
1,500 ng/m3 for unconditioned and conditioned operation. The
sulfate particles were emitted mostly in the submicron size
range.
Ammonium phosphates are known to decompose to ammonia and
either phosphorus pentoxide or to various condensed polymeric
forms of phosphate. Spafford et al. (1979) found that Apollo
LPA-445 decomposed mainly to gaseous ammonia and condensed
phosphates. Some unreacted orthophosphate ion was also detected.
The major recombinations products at 160°C and 90°C were
distributed among ammonium phosphate, sulfate and nitrate salts.
Phosphine (PH3), a toxic gas, was not found to be a reaction
product.
The most common operating problem with ammonium compound
conditioning agents has been plugging of the air preheater by
deposits of condensation products when the agent is injected
upstream of the air preheater.
81
-------
TABLE 26. FATE OF AMMONIUM COMPOUND CONDITIONING AGENTS
Reference
Temp
*C
Injected ppmv
NH. SO,
Location Gat Phase Particle Pha6e Conditioning
at ESP NH¦ SOi Nil! SO, Agent
00
to
Landham 16") 12.2
et al.
19 61
(Gannon)
16^ 12.8
16^ 10.6
Landham
et al.
(1961)
(Cot re tte)
133
6.8
7.4
6.1
6.4
^. 3
3.4
Inlet
Outlc-l
Inlet
Outlet
Inlet
3.6 2.4
0.2 0.9
2.2 1.2
0.2 1.9
2.6 0.8
Outlet 0.3 0.6
2.6 <0.-j
Inlet
Outlet
3.7 Inlet 2.4 0.7"i
Outle t
0.7
4 . *> 2.8
0.2 <1
2.2 6.6
0.1 <1.7
7.3 2.9
0.1 <1
0.73 12
0.21
0.33 12
0.02 0.9V
LPA - 40 Injected
at 540°C
LPA - 40 injected at
at 17 9® C
(NH.)iSO. injected
at 17 5° C
-------
ECONOMICS
No information on capital investment is available. Landham
et al. (1981) reported the following operating cost for
conditioning with Apollo LPAt-40:
Patterson et al. (197 9b) showed costs for Apollo LAC-51B and LPA-
445.
USER EXPERIENCE
Table 27 shows a list of past and present users of ammonium
compounds and proprietary chemicals. No user experience was
reported.
Because of air preheater plugging problems, the following
utility companies have discontinued the use or testing of these
chemicals.
Colorado Ute Electric Association - Hayden
Northern Indiana Public Service - Mitchell
Florida Power - Crystal River
Pennsylvania Power and Light - Montour
Montana Power
(Corrette)
Tampa Electric
(Gannon)
$0,042 mills/kwH
(1981)
$0,024 mills/kwH
(1982)
LAC 51B
LPA 445
$0,052 mills/kwH (1982)
$0,031 mills/kwH
33
-------
TABLE 27. LIST OF PAST AND PRESENT USERS OF AMMONIUM
COMPOUND AND PROPRIETARY CHEMICALS
Company Uame
Plant Name
Licensor
References
Arizona Public Service Co.
Four Corners
Apollo
Pressey et al.
(1979)
Baltimore Gas & Electric
H. A. Wagner
Apollo
Locklin
et al.
(1977)
Central Illinois Light
Duck Creek
Apollo
Colorado Ute Electric Assoc.
Hayden
Apollo
Floricfa Power
Crystal River
Apollo
Georgia Power
Harllee Branch
Apollo
Montana Power Co.
J. E. Corette
Apollo, Nalco
Borsheim (1977)
New England Fewer Co.
Brayton Point
Nalco
New England Power Co.
Salem Harbor
Nell CO
Locklin
et al.
(1980)
New York State Electric & Gas
Goudey
Apollo
Northern Indiana Public Service Co.
D. H. Mitchell
Apollo
Ohio Edison Co.
Edgewater
Dusco
Ohio Edison Co.
Gorge
Dusco
Ohio Edison Co.
W. H. Sammis
Apollo
Pacific Power & Light
Centralia
Pennsylvania Power & Light Co.
Montour
Locklin
et al.
(1980)
Pennsylvania Power & Light Co.
Brunner Island
Locklin
et al.
(1980)
Public Service Co. of Colorado
Valmont
Nalco
Public Service Electric & Gas, New Jersey
Hudson
Apollo
Public Service Electric & Gas, New Jersey
Mercer
Apollo
South Carolina Public Service Authority
Jefferies
Apollo
McNinch
(1976)
Icinipa Electric Co.
Big Bend
Locklin
et al.
(1980)
Tampa Electric Co.
F. J. Gannon
Apollo
Texas Utility Generating Co.
Big Brown
Apollo
McGraw
(1980)
Toledo Edison
Bayshore
Nalco
Locklin
et al.
(1980)
Virginia Electric & Power
Yorktown
Apollo
Locklin
et al.
(1980)
-------
Section 6
ORGANIC AMINES
CHEMICAL PROPERTIES
Organic amines, such as triethylamine and cyclohexylamine,
have been studied in the laboratory and pilot scale ESP's as
possible flue gas conditioning agents by researchers in Australia
and the United States, e.g. Potter and Paulson (1974), Collin
(1978), and Bickelhaupt et al. (1978). Currently there are no
industrial users.
Of all the amines, triethylamine, (C2Hs)zN, has received the
most attention. It is an organic nitrogen compound and is highly
soluble in water. It behaves similar to ammonia, but is a
substantially stronger base than ammonia. The melting point and
boiling point of triethylamine is -115°c and 90°C, respectively.
It decomposes extensively to ammonia, hydrogen cyanide, nitrogen
dioxide, and nitric oxide at temperatures above 345 0 C (650°F)
(Spafford et al., 1979).
Conditioning Mechanism
The mechanism of triethylamine conditioning is not fully
understood. Potter and Paulson (1974) mentioned that
triethylamine increases particle size by agglomeration of
suspended particles. This is a speculative statement based on
measured ESP performance rather than on actual size distribution
measurements.
Based on available data, it seems that triethylamine
improves the voltage-current characteristics by means of reducing
the ash resistivity. Triethylamine is adsorbed on the surface of
fly ash and forms a conductive layer (Bickelhaupt et al., 1978;
Collin, 1978).
Injection Location
Since there are no industrial UBers of triethylamine, no
preferred injection location has been established. Spafford et
al. (1979) recommended, based on results from the thermal
85
-------
decomposition experiments, that the gas temperature at the
injection point be below 370°C. Bickelhaupt et al. (1978) found
that the effectiveness of triethylamine increases with decreasing
gas temperature. Therefore, it seems that the logical injection
location is after the air pre-heater.
Even though triethylamine is less volatile than ammonia, it
can be injected in the vapor phase as well as in solution form.
Dosage ranges from 20 to 140 ppm.
RESULTS
Effects on Particle Resistivity
Brown et al. (1978) measured in-situ the ash resistivity
with a point-to-plane probe in the flue gas from a pilot-scale
boiler which burned western Alberta bituminous coal with 0.3%
sulfur, 13% ash, and over 30% low reactivity roacerals. They
found that triethylamine is effective in reducing the ash
resistivity. The resistivity decreased from 3 x 10' ohm-m
without conditioning to 5 x 107 ohm-m with a triethylamine dosage
of 60 ppm (Figure 2).
The results of a laboratory study by Bickelhaupt et al.
(1978) showed similar conclusions. With an injected
concentration of 25 ppm of triethylamine, the resistivity
decreased one to two orders of magnitude in the temperature range
of 102°C to 150°C for specific ashes. It is more effective with
lower temperatures, less basic ash composition, and greater
concentration of the agent.
Emissions Canned hy the Agent
Because there are no industrial users, emissions caused by
organic amine conditioning have not been determined.
Spafford et al. (1979) analyzed the thermal decomposition
products of triethylamine. They found the compound decomposes
extensively at 650°C, but remains largely intact molecularly at
lower than 370°C. It is capable of reacting with either sulfur
dioxide or sulfuric acid to form bisulfite, sulfite, bisulfate,
86
-------
e
7-
Hi
t-
10
M
40
60
100
120
140
(C2Hs)3N CONCENTRATION ppm
Figure 2. Ash resistivity versus triethylamine
concentration (Brown et al., 1978)
87
-------
or sulfate salt. One of the detected decomposition products is
the carcinogenf N-nitrosodiethylamine.
ECONOMIC AND USER EXPERIENCE
Tri ethyl amine has only been studied in the laboratory and in
pilot plants. There are no industrial users. Therefore, no
economic data and user experience are available.
88
-------
Section 7
ALKALI CONDITIONING
The combustion of many Western coals produce high
resistivity ash which is difficult to collect. Analysis of the
ash revealed that the ash is low in alkali. Bickelhaupt (1974)
found the resistivity of fly ash is inversely proportional to the
amount of sodium and lithium present in the ash. Therefore,
attempts have been made to decrease the ash resistivity by
increasing the alkali concentration in the ash.
Of the many alkali compounds, sodium salts are the most
commonly used conditioning agents because of their availability
and relatively low cost. The widely used sodium compounds are
sodium carbonate and sodium sulfate. Sodium chloride has been
tried in the laboratory and found to be effective. However, it
is not used because it can lead to metal corrosion.
Alkali earth compounds, including dry limestone and magnesia
have also been studied. These compounds react with excess
sulfuric acid to enhance the space charge effect.
CONDITIONING MECHANISM
If the sodium compound is co-precipitated with the ash, the
compound trapped in the space between the particles on the dust
layer offers an additional conductive path for charge
dissipation.
If the sodium compound is injected in the boiler along with
coal, sodium is bound in ash and it affects resistivity the same
way as natural sodium (Gooch et al., 1982).
INJECTION LOCATION
Sodium conditioning, unlike the other conditioning agents,
is not limited to cold side ESP's. It can be added to the boiler
along with the coal or into the flue gas just ahead of the ESP.
In the first case, sodium oxide from the injected compound is
incorporated in the ash as a resistivity-lowering constituent.
When added to the flue gas, it is co-precipitated with the ash,
89
-------
serving as a conductive species in a mixture with the relatively
nonconductive fly ash.
The sodium compound can be applied either in solution form
or in dry powder form. No matter where and how sodium
conditioning is applied/ the most important parameter which
affects its effectiveness is mixing of sodium salt and fly ash.
To be effective/ the sodium must either be incorporated into all
the ash particles or co-precipitated with the ash on the plates
yielding well mixed deposits.
When the sodium is applied in dry powder form for
coprecipitation/ there may be difficulties in obtaining well
mixed deposits as have been discovered by Gooch et al. (1980,
1982). Lederman et al. (1979) applied sodium in solution form to
a hot-side ESP and claimed that uniform coating of sodium salts
on ash particles was obtained.
Gooch et al. (1981) added sodium compound to the coal supply
prior to pulverization as a means of supplementing the sodium
content of the fly ash. They speculated that complete
decomposition and volatilization of the conditioning agent
occurred in the boiler and subsequent condensation of the sodium
compound should be uniformly distributed and become an integral
part of the fly ash surface.
As a coal additive/ the dosage is such that the resulting
alkali concentration (expressed as Na»0) is about 1 to 2% by
weight. When the agent is injected for co-precipitation with the
ash/ the dosage is 2 to 5% of solids as NaiO. The co-
precipitated sodium compound should have a particle size
distribution comparable to that for the ash.
RESULTS
Effects on Parfciele Resistivity
Sodium is effective in reducing the fly ash resistivity if
the sodium is mixed well with the ash. Schliesser (1981)
demonstrated the effectiveness of sodium carbonate dry powder as
a conditioning agent on a pilot precipitator cleaning and the
cold-side slipstream from a commercial boiler. In situ
90
-------
resistivity measurements have shown a decrease in resistivity
from 2.1 x 1010 Ohm-m without conditioning to 3.7 x 10* Ohm-m
with a 1.0 to 1.5% concentration of sodium carbonate as sodium
oxide.
Gooch et al. (1980) reported pilot field test results on the
co-precipitation of sodium carbonate powder with high resistivity
fly ash in a cold-side ESP. In situ resistivity of the ash was
reduced from 1 x 10l# to 1 x 10* Ohm-m when the sodium oxide
content of the ash was increased to 2.5% from the original 0.3%.
However, in a later report on the same study Gooch et al.
(1982) concluded that coprecipitation of fly ash and injected dry
sodium carbonate powder in a cold-side ESP is ineffective in
lowering the resistivity. The powder was injected at 160°C and
the mass median diameter of the powder was 30 jim. Gooch et al.
(1982) speculated that the following two factors may be
responsible for the ineffectiveness of cold-side conditioning
with dry sodium carbonate powder:
1. The presence on the plates of a residual layer of high
resistivity ash that can not be removed by rapping or
washing.
2. The ash and sodium carbonate deposit on the plate is in-
homogeneous.
Precipitator Perfnrnanrp
Table 28 and Figure 3 shows the reported ESP performance
with and without sodium conditioning. Based on these data,
sodium conditioning is effective in improving the ESP performance
when a uniform distribution of sodium compound is obtained. The
improvement in ESP colleciton efficiency is less with co-
precipitation of Na20» than with SOi conditioning.
Industrial users only have a limited experience with sodium
conditioning and operational problems are not well documented.
91
-------
TAffl£ 28. ESP FHIPORMMH WITH SODIUM SALT OCNDITCCNING
VD
M
Utility
Plant:
Wisconsin Power
Columbia Station
Public Service
Co. of Colorado -
Cananche Station
Unit #2
City of Colorado
Springs# Colorado -
Martin Drake Power
Plant, Unit #5
Gulf Power Company
Lansing Smith Plant
Units #1 and #2
Fuel
Sulfur f
MtS
Agent
15% Na2CO,
City of Colorado
Springs# Colorado
Martin Drake
15% Na2COj
Solution
Na2C03
Sodium
Sulfate
Added to
coal
Na2COj
Dosage
Temperature
Specific
Coll. Area
m2/(m3/s)
370 to 400°C
2 to 5%
of sodium as
Na20 in solids
2 to 5% wt.
of sodium as
Na20 in solids
0.5 to 3.5%
Na20 in ash
370°C
160°C
77.2
0.96% Na20 in
ash;
(2.1 kg Na2SO*/
100 kg coal)
1.9% NaxO in ash
1.5 to 3.5%
Nat in ash
350°C
63.8
160°C
65
65.1
Collection
Efficiency,
%
84
99.3
98.0
99.8
98.21
99.64
99.71
87.5
92.3
to 93.5
Reference
Lederman,
et al.
(1979)
Lederman,
et al.
(1979)
Gooch
et al.
(1980)
Gooch
et al.
(1981)
Gooch
et al.
(1982)
-------
100
BASELINE
10
1
SO 3
. 1
0
PARTICLE DIAMETER, MICROMETERS
Figure 3. Fractional efficiency for three per-
formance tests. (Gooch et al., 1981)
93
-------
However, there is one potential problem associated with the
addition of sodium compound to the coal. The sodium may cause
ash slagging and boiler fouling if the system is not operated
properly.
ECONOMICS
Lederraan et al. (1979) provided investment data for a liquid
solution conditioning system. The capital costs are about $1.55
to 3.00 per installed kW. The operating costs, excluding
depreciation, is about 0.03 raills/kWh.
94
-------
REFERENCES
Anon. SOj Injection to Aid Stack Cleanup? Elect. World, 17 3
( 26): 22-24 (1970).
Archer, W. E. Electrostatic Precipitator Conditioning
Techniques. Power Engrg., Zfi. (12): 50-53 (1972).
Baxter, W. A. Recent Electrostatic Precipitator Experience with
Ammonia Conditioning of Power Boiler Flue Gases. J. Air
Poll. Control Assoc., IS: 817-820 (1968).
Bennett, R.P. and A.E. Kober. Chemical Enhancement of
Electrostatic Precipitator Efficiency. Combustion, ££ (4):
27-29 (1978).
Bennett, R. P. and A. E. Kober. Chemical Enhancement of
Electrostatic Precipitator Efficiency. In: Symposium on the
Transfer and Utilitzation of Particulate Control Technology,
vol. I. pp. 113-120. F. P. Venditti, J. A. Armstrong, and
M. Durham, compilers. EPA-600/7-79-044a. United States
Environmental Protection Agency, Research Triangle Park,
North Carolina (1979).
Bickelhaupt, R. E. Electrical Volume Conduction in Fly Ash. J.
Air Poll. Control Assoc., .25: 251-255 (1974).
Bickelhaupt, R. E. Sodium Conditioning to Reduce Fly Ash
Resistivity. NTIS/PB-236 922. United States Environmental
Protection Agency, Research Triangle Park, North Carolina
(1974).
Bickelhaupt, R. E., E. B. Dismukes, and R. B. Spafford. Flue Gas
Conditioning for Enhanced Precipitation of Difficult Ashes.
EPRI FP-910. RP 724-1. Electric Power Research Institute,
Palo Alto, California (1978).
Borsheim, R. Fly Ash Conditioning Brings Particulate Emissions
into Compliance. Power Engrg., £1 (1): 48-50 (1977).
Brennan, J. H. and R. L. Reveley. Flue Gas Conditioning with SOs
to Improve Precipitator Performance. Proc. Amer. Power
Conf., Ail: 742-747 (1977).
Brines, H. G. and R. L. Reveley. Modification of Electrostatic
Precipitator Performance by Use of Fly Ash Conditioning
Agents. AS ME Paper 78-WA/APC-3 presented at the Winter
Annual Meeting, San Francisco, California, 10-15 December
197 8. American Society of Mechanical Engineers, New York,
New York (197 8).
95
-------
Brown/ T. D., K. Lee, and N. Sekhar. Modifications of
Electrostatic Precipitator Performance by Use of Fly Ash
Conditioning Agents. AS ME Paper 78-WA/APC-3 presented at
the Winter Annual Meeting, San Francisco, California, 10-15
December 1978. American Society of Mechanical Engineers,
New York, New York (1978).
Castle, G. S. P. Mechanisms Involved in Fly Ash Precipitation in
the Presence of Conditioning Agents — A Review. IEEE
Trans. Ind. Appl., IA-16; 297-302 (1980).
Collin, P. J. Surface Reactions Between Triethylamine and Fly
Ashes in Electrostatic Precipitators, in Proceedings of
International Clean Air Conference, White et al. (Ed) Ann
Arbor Science Publishers Inc., Ann Arbor, Michigan (1978).
Cook, R. E. Sulfur Trioxide Conditioning. J. Air Poll. Control
Assoc., 2S: 156-158 (1975).
Cook, R., and M. Trykowski. Sulfur Trioxide Injection at State
Line Station. Paper 2 in: Conference on Particulate
Collection Problems in Converting to Low Sulfur Coals, pp.
20-29. G. B. Nichols, compiler. EPA-600/7-76-016. United
States Environmental Protection Agency, Research Triangle
Park, North Carolina (1976).
Cragle, S. H. Operating Experience with ESP Conditioning in
Relation to an Electrostatic Precipitator Upgrading Program.
Paper 1 In: Conference on Particulate Collection Problems in
Converting to Low Sulfur Coals, pp. 3-19. G. B. Nichols,
compiler. EPA-600/7-76-016. United States Environmental
Protection Agency, Research Triangle Park, North Carolina
(1976) .
Dalmon, J. and D. Tidy. A Comparison of Chemical Additives as
Aids to the Electrostatic Precipitation of Fly-ash. Atmos.
Env., £: 721-734 (1972).
Dismukes, E. B. Conditioning of Fly Ash with Sulfamic Acid,
Ammonium Sulfate, and Ammonium Bisulfate. NTIS/PB-238 922.
United States Environmental Protection Agency, Research
Triangle Park, North Carolina (1974).
Dismukes, E. B. Conditioning Fly Ash with Ammonia. J. Air Poll.
Control Assoc., .25: 152-156 (1975).
Dismukes, E. B. Techniques for Conditioning Fly Ash. Paper 7 in:
Conference on Particulate Collection Problems in Converting
to Low Sulfur Coals. G. B. Nichols, compiler. EPA-600/7-
76-016. United States Environmental Protection Agency,
Research Triangle Park, North Carolina (1976).
-------
Dismukes, E. B. and J. P. Gooch. Fly Ash Conditioning with
Sulfur Trioxide. NTIS/PB-276 651. United States
Environmental Protection Agency, Research Triangle Park,
North Carolina (1977).
Dismukes, E. B. Flue Gas Conditioning in Coal-fired Power Plants
in the United States. In: Second US/USSR Symposium on
Particulate Control. F. A. Ayer, compiler. EPA-600/7-78-
037, pp. 107-122. United States Environmental Protection
Agency, Research Triangle Park, North Carolina (1978).
Gmitro, J. T. and T. Verraeulen, Vapor-Liquid Equilibrium for
Aqueous Sulfuric Acid. AlChe Journel, lfi: 740-746, (1964).
Gooch, J. P., R. E. Bickelhaupt, and L. E. Sparks. Fly Ash
Conditioning by Coprecipitation with Sodium Carbonate. In:
Second Symposium on the Transfer and Utilization of
Particulate Control Technology vol. I. F. P. Venditti, J.
A. Armstrong, and M. Durham, compilers. EPA-600/9-80-039a,
pp. 132-153. United States Environmental Protection Agency,
Research Triangle Park, North Carolina (1980).
Gooch, J. P., R. E. Bickelhaupt, G. H. Marchant, R. D. McRanie,
L. E. Sparks, and R. F. Altman. Improvement of Hot-Side
Precipitator Performance with Sodium Conditioning - An
Interim Report, J. Air Pollution Control Assoc., 21: 291-297
(1981).
Gooch, J. P., R. E. Bickelhaupt, R. S. Dahlin, G. H. Marchant,
and E. C. Landham. A field Demonstration Study to Evaluate
Sodium Carbonate Injection for Reducing Fly Ash Resistivity.
Draft of Final Report to EPA. Contract No. 6 8-02-2656, June
7, (1982).
Green, G. P. and W. S. Landers. Operating Experience with Gas
Conditioned Electrostatic Precipitators. Paper 8 in:
Proceedings, Symposium on Control of Fine Particulate
emissions form Industrial Sources, San Fancisco, California,
15-18 January, 1974.
Guiffre, J. T. Precipitator Upgrading and Fuel Control Program
for Particulate Compliance at Pennsylvania Power and Light
Company. In: Second Symposium on the Transfer and
Utilization of Particulate Control Technology, vol. I. F.
P. Venditti, J. A. Armstrong, and M. Durham, compilers.
EPA-600/9-80-039a, pp. 80-99. United States Environmental
Protection Agency, Research Triangle Park, North Carolina
(1980) .
Kanowski, S., and R. W. Coughlin.
Ash without Addition of SOj
Sci. Tech., 11: 67-70 (1977).
Catalytic Conditioning of Fly
from External Sources. Env.
97
-------
Katz, J., The Effective Collection of Fly Ash at Pulverized
Coal-fired Plants. J. Air Poll. Control Assoc., 15: 525-528
(1965).
Katz, J., The Art of Electrostatic Precipitation, Precipitator
Technology, Inc., 1979.
Landham, E. C. Jr., G. H. Marchant, Jr., J. P. Gooch, and R. F.
Altman. Field Evaluations of Ammonium Sulfate Conditioning
for Improvement of Cold-side Electrostatic Precipitator
Performance. Paper presented at the Air Pollution Control
Conference, Orlando, Florida, March, (1981).
Lederman, P. B., P. B. Bibbo, and J. Bush. Chemical
Conditioning of Fly Ash for Hot-side Precipitation. In:
Symposium on the transfer and utilization of particulate
control technology, vol. I. F. P. Venditti, J. A.
Armstrong, and M. Durham, compilers. EPA-600/7-044a, pp.
79-98. United States Environmental Protection Agency,
Research Triangle Park, North Carolina (1979).
Locklin, D. W., H. H. Krause, D. Anson, W. Reid. Electric Utility
Use for Fireside Additives. EPRI CS-1318. RP 1035-1.
Electric Power Research Institute, Palo Alto, California,
1980.
McNinch, M. 0. Chemotherapy Tames fly Ash from Low-Sulfur Coal.
Electric World, 185(2):32-33, 1976.
McGraw, M. G. Fly Ash Emission Problems Resolved. Elect.
World, 114 (8): 123-125 (1980).
Midkiff, L. A. Flue-gas Conditioning Upgrades Performance,
Cuts Down Size of Precipitators. Power, 123 (4): 99-101
(1979) .
McLean, K. J.. Factors Affecting the Resistivity of a
Particulate Layer in Electrostatic Precipitators. J. Air
Poll. Contorl Assoc., 2£: 866-870 (1976).
Nair, P. V. N. and R. G. Vohra. Growth of Aqueous Sulfuric Acid
Droplets as a Function of Relative Humidity. J. Aerosol
Sci., £: 265-271 (1975).
Patterson, R. G., P. Riersgard, R. Parker, and S. Calvert.
Effects of Conditioning Agents on Emissions from Coal-fired
Boilers. Test report no. 1. EPA-600/7-79-104a. United
States Environmental Protection Agency, Research Triangle
Park, North Carolina (1979a).
Patterson, R. G.r J. Long, R. Parker, and S. Calvert.
Effects of Conditioning Agents on Emissions from Coal-fired
Boilers. Test Report no. 2. EPA-600/7-79-104b. United
States Environmental Protection Agency, Research Triangle
Park, North Carolina (1979b).
98
-------
Patterson, R. G., P. Riersgard, R. Parker, and L. G. Sparks.
Flue Gas Conditioning Effects on Electrostatic
Precipitators. In: Symposium on the Transfer and
Utilization of Particulate Control Technology, vol. I. F.
P. Venditti, J. A. Armstrong# and M. Durham, compilers.
EPA-6 00/7-79-044a, pp. 169-178. States Environmental
Protection Agency# Research Triangle Park# North Carolina
(1979C).
Patterson# R. G.# R. Parker# H. Scott# and S. Calvert.
Effects of Conditioning Agents on Emissions from Coal-fired
Boilers. Test report no. 3. Contract 68-02-2628. United
States Environmental Protection Agency# Research Triangle
Park# North Carolina (1981).
Perry# J. H. and C. H. Chilton, eds. Chemical Engineers'
Handbook# 5th edition McGraw-Hill Book Company# New York#
New York (1972).
Potter, E. C., and C. A. J. Paulson. Improvement of
Electrostatic Precipitator Performance by Carrier-gas
Additives and its Graphical Assessment Using an Extended
Deutsch Equation. Chem. Ind. (London), 532-533 (1974).
Pressey, R. E., D. Osborn, and E. Cole. Flue Gas
Conditioning at Arizona Public Service Company Four Corners
Unit 4. In: Symposium on the Transfer and Utilization of
Particulate Control Technology, vol. I. F. P Venditti, J.
A. Armstrong, and M. Durham, compilers. EPA-600/7-79-044a,
pp. 179-204. United States Environmental Protection Agency,
Research Triangle Park, North Carolina (1979).
Reese, J. T., and J. Greco. Experience with Electrostatic
Fly Ash Collection Equipment Serving Steam-electric
Generating Plants. J. Air Poll. Control Assoc., IB: 523-528
(1968) .
Reissmann, H. How To Raise the Efficiency of Dry
Electrostatic Precipitators by Means of Gas Conditioning.
Berichte der Kernforschungsanlage Juelich, (Fed. Rep.
Germany) Jul-Conf-28. pp. 125-139 (1978).
Schliesser, S. Pilot Demonstration of Sodium Carbonate
Conditioning. EPA-600/7-79-107. United States Environmntal
Protection Agency, Research Triangle Park, North Carolina
(1979a).
Schliesser, S. P. Sodium Conditioning Test with EPA Mobile ESP.
In: Symposium on the Transfer and Utilization of Particulate
Control Technology, vol. I. F. P. Venditti, J. A.
Armstrong, and M. Durham, compilers. EPA-600/7-79-044a, pp.
205-240. United States Environmental Protection Agency,
Research Triangle Park, North Carolina (1979b).
99
-------
Schliesser, S. P., S. Malani, C. L. Stanley, and L. E. Sparks.
EPA Mobile ESP Hot-side Performance Evaluations. In: Second
Symposium on the Transfer and Utilization of Particulate
Control Technology, vol. I. F. P. Venditti, J. A.
Armstrong, and H. Durham, compilers. EPA-600/9-80-039a, pp
56-79. United States Environmental Protection Agency,
Research Triangle Park, North Carolina (1980).
Schrader, K. Improvemnt of the Efficiency of Electrostatic
Precipitation by Injection SO t into the Flue Gas.
Combustion, A2 (4): 22-28 (1970).
Southern Research Institute. A Study of Resistivity and
Conditioning of Fly Ash. NTIS/PB-212 607. United States
Environmental Protection Agency, Research Triangle Park,
North Carolina (1972).
Spafford, R. B., E. B. Dismukes, and H. K. Dillon. Analysis of
Thermal Decomposition Products of Flue Gas Conditioning
Agents. EPA-600/7-79-17 9. United States Environmental
Protection Agency, Research Triangle Park, North Carolina
(1979) .
Spafford, R. B., H. K. Dillon, E. B. Dismukes, and L. E. Sparks.
Analysis of Thermal Decomposition Products of Flue Gas
Conditioning Agents. In: Second Symposiumon the Transfer
and Utilization of Particulate Control Technology, vol. I.
F. P. Venditti, J. A. Armstrong, and M. Durham, compilers.
EPA-600/9-80-039a, pp. 202-223. United States Environmental
Protection Agency, Research Triangle Park, North Carolina
(1980).
Sparks, L. E. Electrostatic Precipitator Options for
Collection of High Resistivity Fly Ash. Paper 5 in:
Conference on Particulate Collection Problems in Converting
to Low Sulfur Coals. G. B. Nichols, compiler. EPA-600/7-
76-016, pp. 58-67. United States Environmental Protection
Agency, Research Triangle Park, North Carolina (1976).
Tassicker, 0. J. Some Aspects of Electrostatic Precipitator
Research in Australia. J. Air Poll. Control Assoc., 25:
122-128 (1975).
Trevor Busby, H. G., and K. Darby. Efficiency of Electrostatic
Precipitators as Affected by the Properties and Combustion
of Coal. J. Inst. Fuel (London), 2A (268): 184-197 (1963).
Walker, A. B. Operating Experience with Hot Precipitator on
Western Low Sulfur Coals. Palmer House, Chigaco,
Illinois. Presented at: American Power Conference, April
18-20, (1977).
100
-------
Walker, F. and N. Lamb. Recent Experience with Flue Gas
Conditioning on Power Boiler Precipitators in New South
Wales. Paper presented at the International Clean Air
Conference/ Brisbane/ Australia/ May (1978).
Watson, K. S. Australian Experience with Flue Gas Conditioning.
In: symposium on Particulate control in Energy Processes,
compiled by D. E. Blake. NTIS PB-260499. 1976.
White/ H. J. Electricl Resistivity of Fly Ash. Air Repair, 2
(2): 79-87 (1953).
White, H. J. Resistivity Problems in Electrostatic
Precipitation. J. Air Poll. Control Assoc./ 2Ai 314-338
(1974) .
101
------- |