The Effects of Oxy-firing Conditions on Gas-phase
Mercury Oxidation by Chlorine and Bromine
Topical Report
Reporting Period Start Date: April 2009
Report Period End Date: June 2010
Principal Authors: Paula A. Buitrago and Geoffrey D. Silcox
Issue date: October 2010
DOE Award Number: DE-NT0005015
Project Officer: David Lang
University of Utah
Department of Chemical Engineering
50 S. Central Campus Drive, MEB 3290
Salt Lake City, Utah
i
DISCLAIMER
This report was prepared as an account of work sponsored by an agency of the United
States Government. Neither the United States Government nor any agency thereof, nor
any of their employees, makes any warranty, express or implied, or assumes any legal
liability or responsibility for the accuracy, completeness, or usefulness of any information,
apparatus, product, or process disclosed, or represents that its use would not infringe
privately owned rights. Reference herein to any specific commercial product, process or
service by trade name, trademark, manufacturer, or otherwise does not necessarily
constitute or imply its endorsement, recommendation, or favoring by the United States
Government or any agency thereof. The views and opinions of authors expressed herein
do not necessarily state or reflect those of the United States Government or any agency
thereof.
ii
ABSTRACT
Bench-scale experiments were conducted in a quartz-lined, natural gas-fired reactor with
the combustion air replaced with a blend of 27 mole percent oxygen, with the balance
carbon dioxide. Quench rates of 210 and 440 K/s were tested. In the absence of sulfur
dioxide, the oxy-firing environment caused a remarkable increase in oxidation of mercury
by chlorine. At 400 ppm chlorine (as HCl equivalent), air-firing results in roughly 5
percent oxidation. At the same conditions with oxy-firing, oxidation levels are roughly 80
percent. Oxidation levels with bromine at 25 and 50 ppm (as HBr equivalent) ranged
from 80 to 95 percent and were roughly the same for oxy- and air-firing conditions.
Kinetic calculations of levels of oxidation at air- and oxy-conditions captured the
essential features of the experimental results but have not revealed a mechanistic basis for
the oxidative benefits of oxy-firing conditions. Mixtures of 25 ppm bromine and 100 and
400 ppm chlorine gave more than 90 percent oxidation. At all conditions, the effects of
quench rate were not significant. The presence of 500 ppm SO2 caused a dramatic decline
in the levels of oxidation at all oxy-fired conditions examined. This effect suggests that
SO2 may be preventing oxidation in the gas phase or preventing oxidation in the wetconditioning system that was used in quantifying oxidized and elemental mercury
concentrations. Similar effects of SO2 have been noted with air-firing. The addition of
sodium thiosulfate to the hydroxide impingers that are part of wet conditioning systems
may prevent liquid-phase oxidation from occurring.
iii
TABLE OF CONTENTS
DISCLAIMER .................................................................................................................... ii
ABSTRACT ....................................................................................................................... iii
LIST OF FIGURES ............................................................................................................ v
LIST OF ABBREVIATIONS ............................................................................................ vi
EXECUTIVE SUMMARY .............................................................................................. vii
INTRODUCTION .............................................................................................................. 1
METHODS ......................................................................................................................... 1
RESULTS AND DISCUSSION ......................................................................................... 5
Effects of Oxy-firing on Gas-phase Mercury Oxidation by Chlorine and Bromine....... 6
Effects of SO2 ............................................................................................................... 10
CONCLUSIONS............................................................................................................... 11
REFERENCES ................................................................................................................. 13
iv
LIST OF FIGURES
Figure 1. Sketch of the homogeneous mercury reactor (Fry et al., 2006).
Figure 2. Temperature profiles in the homogeneous mercury reactor.
Figure 3. Mercury analysis system (Fry et al., 2006).
Figure 4. Extents of oxidation by bromine and chlorine with air-firing and no SO2.
Figure 5. Extents of oxidation with oxy-firing, chlorine, and no SO2.
Figure 6. Kinetic calculations showing predicted effects of oxy-firing with chlorine in the
absence of SO2. Chlorine concentrations are as ppm HCl equivalent.
Figure 7. Oxidation of elemental mercury with oxy-firing, bromine, and no SO2.
Figure 8. Mercury oxidation with oxy-firing, mixtures of bromine and chlorine, and no
SO2.
Figure 9. Mercury oxidation by chlorine and bromine with oxy-firing in the presence of
SO2.
v
LIST OF ABBREVIATIONS
Br
Bromine
Cl
Chlorine
ESP
Electrostatic precipitators
HCl
Hydrochloric acid
Hgp
Particle-bound mercury
Hg++
Oxidized mercury
Hgo
Elemental mercury
HgCl2 Mercuric chloride
NO
Nitric oxide
NO2
Nitrogen dioxide
SCR
Selective catalytic reduction
SLPM Standard liters per minute
SO2
Sulfur dioxide
vi
EXECUTIVE SUMMARY
The fate and speciation of mercury in oxy-coal combustion processes is of particular
interest because trace amounts of mercury lead to the embrittlement and cracking of
aluminum heat exchangers that are used in the cryogenic separation and compression of
carbon dioxide. Mercury exists in three forms in coal-derived flue gas: particle-bound
(Hgp), oxidized (Hg++), and elemental (Hgo). The oxidized form is desirable because it is
more readily removed by adsorption on solids and is more readily captured by absorption
in wet flue gas desulfurization scrubbers. The adsorbed, particle-bound mercury is
removed by electrostatic precipitators (ESPs) or fabric filters.
Oxidized mercury is primarily mercuric chloride (HgCl2) and its formation depends on
the halogen content of the coal and the existence of catalytic surfaces that enable
oxidation. The latter include unburned carbon and SCR reactors. This report focuses on
the homogeneous, gas-phase oxidation of mercury by bromine and chlorine under oxyfiring conditions. There is a dearth of oxy-fired, mercury-oxidation results for gas-phase
and heterogeneous conditions.
Oxy-coal experiments conducted in a 1.5 MW facility (Imada et al., 2010) showed 75 %
mercury removal across the ESP with oxy-firing and 63 % with air-firing. The
temperature of the ESP and the sulfur content of the coal were key factors affecting
removal. Lower temperature and lower sulfur content led to higher removal. The recycle
of flue gas in oxy-coal systems led to higher mercury concentrations at the inlet to the
ESP.
Bench-scale experiments were conducted at the University of Utah in a quartz-lined,
natural gas-fired reactor with the combustion air replaced with a blend of 27 mole percent
oxygen, balance carbon dioxide. Quench rates of 210 and 440 K/s were tested. In the
absence of sulfur dioxide, the oxy-firing environment caused an increase in oxidation of
mercury by chlorine. At 400 ppm chlorine (as HCl equivalent), air-firing results in
roughly 5 percent oxidation. At the same conditions with oxy-firing, oxidation levels are
roughly 80 percent. Oxidation levels with bromine at 25 and 50 ppm (as HBr equivalent)
ranged from 80 to 95 percent and were roughly the same for oxy- and air-firing
conditions. Kinetic calculations of levels of oxidation at air- and oxy-conditions captured
the essential features of the experimental results but have not revealed a mechanistic
basis for the oxidative benefits of oxy-firing conditions. Mixtures of 25 ppm bromine and
100 and 400 ppm chlorine gave more than 90 percent oxidation. At all conditions, the
effects of quench rate were not significant.
The presence of 500 ppm SO2 caused a dramatic decline in the levels of oxidation at all
oxy-fired conditions examined. This effect suggests that SO2 may be preventing
oxidation in the gas phase or preventing oxidation in the wet-conditioning system that
was used to quantify elemental and oxidized mercury concentrations. Similar effects of
vii
SO2 have been noted with air-firing. The addition of sodium thiosulfate to the hydroxide
impingers that are part of wet conditioning systems may prevent liquid-phase oxidation
from occurring.
An evaluation of several potential pathways that gave rise to the positive effects of oxyfiring on oxidation did not reveal an understanding of the mechanism. It is possible that
the replacement of N2 by CO2 results in higher collision efficiencies in the key, ratecontrolling reaction for oxidation:
Hg + Cl + M = HgCl + M
(1)
In other words, CO2 is more effective at removing energy from the HgCl transition state
complex than N2. It is also possible that the decreased levels of NO and NO2 that occur
with oxy-firing remove a reaction that is consuming Cl atoms (Bozzelli, 2010), e.g.,
NO + Cl = NOCl
(2)
The kinetic calculations do not consider the identity of the third body, M, in Reaction 1
and hence this explanation cannot be supported by the calculations. Calculations in which
Reaction 2 is removed do not increase the level of oxidation at air-firing conditions.
Additional possible reactions involving NO, Cl, and HgCl include (Bozzelli, 2010)
NO + HgCl = NOCl + Hg
HCl + NO = NOCl + H
Cl + NOCl = Cl2 + NO
The decreased levels of NO at oxy-firing conditions would diminish the importance of
these reactions in removing Cl radicals from the mercury oxidation process and prevent
the destruction of the vital intermediate, HgCl. A fundamental understanding of the
experimental and theoretical observations in this study is still incomplete.
viii
INTRODUCTION
The fate and speciation of mercury in oxy-coal combustion processes is of particular
interest because trace amounts of mercury lead to the embrittlement and cracking of
aluminum heat exchangers that are used in the cryogenic separation and compression of
carbon dioxide. Mercury exists in three forms in coal-derived flue gas: particle-bound
(Hgp), oxidized (Hg++), and elemental (Hgo). The oxidized form is desirable because it is
more readily removed by adsorption on solids and is more readily captured by absorption
in wet flue gas desulfurization scrubbers. The adsorbed, particle-bound mercury is
removed by ESP’s or fabric filters.
Oxidized mercury is primarily mercuric chloride (HgCl2) and its formation depends on
the halogen content of the coal and the existence of catalytic surfaces that enable
oxidation. The latter include unburned carbon and SCR reactors. This report focuses on
the homogeneous, gas-phase oxidation of mercury by bromine and chlorine under oxyfiring conditions. There is a dearth of oxy-fired, mercury-oxidation results for gas-phase
and heterogeneous conditions.
Oxy-coal experiments conducted in a 1.5 MW facility (Imada et al., 2010) showed 75 %
mercury removal across the ESP with oxy-firing and 63 % with air-firing. The
temperature of the ESP and the sulfur content of the coal were key factors affecting
removal. Lower temperature and lower sulfur content led to higher removal. The recycle
of flue gas in oxy-coal systems led to higher mercury concentrations at the inlet to the
ESP.
METHODS
The homogeneous mercury reactor is shown schematically in Figure 1 along with
associated equipment. The reactor consists of a 50-mm OD x 47-mm ID quartz tube (132
cm in length) located along the center of a high-temperature Thermcraft heater. The tube
extends 79 cm below the heater, is temperature controlled, and has a quartz sampling
section attached at the bottom with a capped end. The peak gas temperature in the
electrically heated zone was about 1080C. The reactor was operated with two
temperature or quench profiles: 210 and 440 K/s. The former will be referred to as low
quench (LQ) and the latter as high quench (HQ). The air-fired profiles are given in Figure
2 and were obtained by adjusting the current supplied to the heaters surrounding the
bottom 79 cm of the quartz tube. The effect of oxy-firing on temperatures and
temperature profiles was not determined although temperatures should be modestly lower
because of the higher heat capacity of CO2 (37.3 J/mol-K at 300 K) relative to that of N2
(29.1 J/mol-K at 300K).
A 300-W, methane-fired, premixed burner made of quartz glass supplied realistic
combustion gasses to the reactor. All reactants were introduced through the burner and
1
passed through the flame to create a radical pool representative of real combustion
systems. The burner provided 3.7 SLPM of combustion gases. To study the effects of flue
gas components such as SO2, NO, NO2, HCl, and HBr, different concentrations of these
or related species were introduced through the burner.
A Tekran 2537A Mercury Analyzer, coupled with a wet sample conditioning system
designed by Southern Research Institute (SRI), provided measurement of total and
elemental mercury in the exhaust gas and is shown in Figure 3. Sample gas was pulled in
two streams from the last section of the quartz-lined reactor into a set of conditioning
impingers. In the standard configuration of the conditioning system, one stream was
bubbled through a solution of stannous chloride to reduce the oxidized mercury to
elemental form, followed by a solution of sodium hydroxide to remove acid gases. This
stream was analyzed to give the total mercury concentration in the sample. The second
stream was treated with a solution of potassium chloride to remove oxidized mercury
species, followed by a solution of sodium hydroxide for acid gas removal. Sodium
thiosulfate was added to the KCl impinger (Cauch et al., 2008) to prevent the liquidphase oxidation of elemental mercury by Cl2 and Br2. The KCl-washed stream was
analyzed to give the elemental mercury concentration in the sample. Oxidized species
were calculated by the difference between total and elemental mercury concentrations. A
chiller removed water from the sample gas and each stream was intermittently sent to the
analyzer.
The experiments were performed with different dopants added through the burner. Before
adding these, the baseline mercury level at the furnace outlet was checked using a
material balance. Because of limited time and materials, the experiments were performed
only once.
Table 1 gives the baseline, air-fired flue gas composition for the experiments on a dry
basis. The gas composition was not intended to duplicate the flue gas in coal-fired power
plants; the intent of this work was to study reactions of mercury and common flue gas
species in a well-controlled system. All species that were added to the reactor (SO2, Cl2,
Hg0, and Br2) passed through the flame. Their subsequent speciation depended on flame
chemistry as well as on the temperature profile in the reactor. The effects of oxy-firing on
flue-gas composition were not independently measured with gas analyzers as they were
for the air-fired results in Table 1. For the oxy-fired experiments, the air was replaced
with a mixture of 27 percent oxygen and 73 percent CO2, and the amount added was
calculated to give the same oxygen level, 1.0 percent, as in the air-fired tests. This
mixture composition, 27 percent O2 and 73 percent CO2, was chosen to match typical
industrial conditions.
At the oxy-fired conditions, nitrogen still entered the system through the bottled Cl2, Br2,
and SO2 gases that were added through the burner. The bromine bottle was 3000 ppm Br2
in air and the chlorine bottle was 6000 Cl2 in air. Sulfur dioxide was also 6000 ppm in air.
The concentration of NOx was not measured and NO and NO2 were not added to the
reactor because their effects on mercury oxidation in parallel, air-fired tests were
2
negligible. Under oxy-firing conditions, the temperature profiles and gas compositions in
the reactor were not measured and probably differ somewhat from those shown in Figure
2.
Compressed
Air
Filters
Pressure
Regulator
Flashback Arrestor
PS Analytical
Mercury Cal
Gas Generator
Solenoid Valve
Support
Tee &
Blowoff
Nozzle
Mass Flow Controllers
Cl2/
Air
Cal
Gas
SO2/
Air
Cal
Gas
Purge
Nozzle
Thermcraft
Heater
Rotameter
CH4
UV
Detector
NO
Inswool
O2
Analyzer
Data
Acquisition
Temperature
Controllers
Heat
Tape 2
NO
Analyzer
CO
Analyzer
Heat
Tape 1
Heat
Tape 3
Mercury
Analysis
System
Heat
Tape 4
CO2
Analyzer
Figure 1. Sketch of the homogeneous mercury reactor (Fry et al., 2006).
3
1200
Temperature, oC
1000
Low Quench
~210 K/s
800
600
High Quench
~440 K/s
400
200
0
0
1
2
3
4
5
6
7
8
Time, sec
Figure 2. Temperature profiles in the homogeneous mercury reactor.
PS Analytical
Mercury Cal
Gas Generator
Mercury Reactor
HgT
Tekran 2537A
Mercury Analyzer
Hg0
SnCl2
KCl
Intermittent
Hg0 and HgT
Chiller
NaOH
4 Port Sampler
NaOH
Peristaltic Pump
NaOH
SnCl2
KCl
NaOH
Fresh
Reagents
To NOx, CO2,
CO and O2
Analyzers
Hg0
Vacuum
Pump
Waste
Reagents
NaOH
SnCl2
KCl
NaOH
HgT
Figure 3. Mercury analysis system (Fry et al., 2006).
4
Table 1. Air-fired flue gas compositions (dry basis).
Species
O2
H2O
CO2
NO
SO2
HCl*
HBr**
Hg0
Concentration
0.8 vol%
16.5 vol%
7.7 vol%
30, 500 ppmv
0-500 ppmv
0-500 ppmv
0-50 ppmv
25 µg/Nm3 (1
atm and 0C)
*Assuming all chlorine added as HCl
**Assuming all bromine added as HBr
The differential equations describing the gas-phase chemical kinetics of mercury
oxidation were integrated using CHEMKIN and compared with results from a stiff
integrator, REKS, developed by Reaction Engineering International. The two sets of
calculations agree, but only those from REKS are shown below because the University of
Utah had more experience with REKS and CHEMKIN provided somewhat inconsistent
results. CHEMKIN and REKS include modules for the simulation of plug-flow (PFR)
and completely stirred tank (CSTR) reactors. The calculations performed in this study
used a CSTR to represent the burner and a PFR to represent the heat and quench sections
of the University of Utah facility. The temperature profiles shown in Figure 2 were used
as input and the feed streams to the CSTR were chosen to match the experimental
conditions.
RESULTS AND DISCUSSION
Typical U of U, air-fired, gas-phase oxidation results with the high quench profile and a
baseline mercury concentration of 25 µg/m3 are shown in Figure 4. Bromine is clearly a
superior oxidant at these conditions with oxidation levels ranging from 30 to 80 percent
at bromine concentration from 20 to 50 ppm bromine (as HBr equivalent). The extent of
oxidation by chlorine ranges from 2 to 7 percent at 100 to 500 ppm chlorine (as HCl
equivalent). The results in Figure 4 were obtained in the absence of SO2.
5
Figure 4. Extents of oxidation by bromine and chlorine with air-firing and no SO2.
Effects of Oxy-firing on Gas-phase Mercury Oxidation by Chlorine and
Bromine
Replacement of the air with a mixture of 27 % oxygen, balance carbon dioxide, has a
dramatic effect on the ability of chlorine to oxidize mercury. Figure 5 shows gas-phase
oxidation levels with oxy-firing of 70 to 80 % at 100 and 400 ppm chlorine (as HCl
equivalent). Oxidation levels with bromine at 25 and 50 ppm (as HBr equivalent) range
from 70 to 100 %. The results in Figure 5 were obtained in the absence of SO2.
6
Figure 5. Extents of oxidation with oxy-firing, chlorine, and no SO2.
The surprisingly high levels of oxidation by chlorine are compared to kinetic calculations
in Figure 6. The two air-firing curves are for the high and low quench rates and show
levels of oxidation that are similar to those in Figure 4. The two oxy-firing curves show
predicted levels of oxidation ranging from 60 to almost 100 percent.
After investigating several potential pathways that give rise to the positive effects of oxyfiring on oxidation, the kinetics are still not understood. It is possible that the replacement
of N2 by CO2 results in higher collision efficiencies in the key, rate controlling reaction
for oxidation:
Hg + Cl + M = HgCl + M
(1)
In other words, CO2 is more effective at removing energy from the HgCl transition state
complex than N2. The kinetic calculations do not consider the identity of the third body,
M, in Reaction 1 and hence an explanation in terms of the efficiency of third-body
collisions cannot be supported by the calculations.
It is also possible that the decreased levels of NO that occur with oxy-firing diminish the
significance of a reaction that is consuming Cl atoms (Bozzelli, 2010), e.g.,
NO + Cl = NOCl
(2)
Calculations in which Reaction 2 is removed do not increase the level of oxidation at airfiring conditions.
7
Additional possible reactions involving NO, Cl, and HgCl include (Bozzelli, 2010)
NO + HgCl = NOCl + Hg
HCl + NO = NOCl + H
Cl + NOCl = Cl2 + NO
The decreased levels of NO at oxy-firing conditions would diminish the importance of
these reactions in removing Cl radicals from the mercury oxidation process and prevent
the destruction of the vital intermediate, HgCl. A fundamental understanding of the
experimental and theoretical results in Figures 5 and 6 is still not available.
Oxy-firing, high quench
Oxy-firing, low quench
Air-firing, high quench
Air-firing, low quench
Figure 6. Kinetic calculations showing predicted effects of oxy-firing with chlorine in the
absence of SO2. Chlorine concentrations are as ppm HCl equivalent.
Figure 7 shows gas-phase oxidation levels with oxy-firing and bromine as a function of
quench rate and bromine concentrations. The levels of oxidation are insensitive to quench
rate and range from 75 to 100 percent as bromine concentration is increased from 25 to
50 ppm (as HBr equivalent).
8
Figure 7. Oxidation of elemental mercury with oxy-firing, bromine, and no SO2.
The effects of mixtures of bromine and chlorine with oxy-firing on oxidation are shown
in Figure 8. The presence of 25 ppm bromine (as HBr equivalent) and chlorine
concentrations from 100 to 400 ppm (as HCl equivalent) resulted in higher levels of
oxidation that when just bromine or chlorine was present. Changing the chlorine
concentration from100 to 400 ppm had no effect on the levels of oxidation.
9
Figure 8. Mercury oxidation with oxy-firing, mixtures of bromine and chlorine, and no
SO2.
Effects of SO2
The results in Figures 4 to 8 were obtained in the absence of SO2. At 500 ppm SO2, the
encouraging oxy-fired results in Figures 4 to 8 are replaced with the results in Figure 9.
The ability of chlorine and bromine to oxidize mercury drops dramatically. A
mechanistic explanation of these results may involve gas-phase reactions between the
halogen atoms and SO2. A second explanation involves aqueous-phase reactions that are
interfering with the conditioning system and the analysis of mercury by the atomic
fluorescent analyzer used in this study (Tekran 2537A Mercury Analyzer). It is believed
that the aqueous reactions can be controlled by adding reducing agents such as sodium
thiosulfate to the sodium hydroxide impinger solutions. The results in Figure 4 to 8 may
be due to oxidation of elemental mercury in the hydroxide impingers. The KCl impinger
solution is already treated with sodium thiosulate (Cauch et al., 2008) to prevent halogens
from oxidizing mercury. Sulfur dioxide also acts as a reducing agent; hence, the
hypothesis that the results in Figure 9 are an artifact of the wet conditioning system.
10
Figure 9. Mercury oxidation by chlorine and bromine with oxy-firing in the presence of
SO2.
CONCLUSIONS
Bench-scale experiments were conducted in the University of Utah’s quartz-lined, natural
gas-fired reactor with the combustion air replaced with a blend of 27 mole percent
oxygen and 73 percent carbon dioxide. In the absence of sulfur dioxide, the oxy-firing
environment caused an increase in oxidation of mercury by chlorine. At 400 ppm
chlorine (as HCl equivalent), air-firing results in roughly 5 percent oxidation. At the same
conditions with oxy-firing, oxidation levels are roughly 80 percent. Oxidation levels with
bromine at 25 and 50 ppm (as HBr equivalent) ranged from 80 to 95 percent and were
roughly the same for oxy- and air-firing conditions. Kinetic calculations of levels of
oxidation at air- and oxy-conditions captured the essential features of the experimental
results but have not revealed a mechanistic basis for the oxidative benefits of oxy-firing
conditions. Mixtures of 25 ppm bromine and 100 and 400 ppm chlorine gave more than
90 percent oxidation. At all conditions, the effects of quench rate were not significant.
The presence of 500 ppm SO2 caused a dramatic decline in the levels of oxidation at all
oxy-fired conditions examined. This effect suggests that SO2 may be preventing
oxidation in the gas phase or preventing oxidation in the wet-conditioning system that
was used in quantifying oxidized and elemental mercury concentrations. Similar effects
of SO2 have been noted with air-firing. The addition of sodium thiosulfate to the
hydroxide impingers that are part of wet conditioning systems may prevent liquid-phase
oxidation from occurring.
11
These results imply that oxy-firing conditions may greatly enhance the gas-phase
oxidation of mercury by chlorine and emphasize the importance of preventing liquidphase oxidation of elemental mercury in wet conditioning systems. It is recommended
that sodium thiosulfate be added to the hydroxide impingers that are typically part of wet
conditioning systems. Additional bench-scale testing is needed to verify the positive
effects of oxy-firing on gas-phase mercury oxidation and the negative effects of SO2.
Small, pilot-scale testing with pulverized coal is needed to explore the importance of
surface effects and catalysis by unburned char.
12
REFERENCES
1 Bozzelli, J. W. Department of Chemistry and Environmental Science, New Jersey
Institute of Technology. Private communication. 2010.
2 Cauch, B.; Silcox, G.; Lighty, J.; Wendt, J.; Fry, A.; Senior, C. Confounding Effects
of Aqueous-Phase Impinger Chemistry on Apparent Oxidation of Mercury in Flue
Gases. Environ. Sci. Technol. 2008, 42(7), 2594–2599.
3 Fry, A.; Cauch, B.; Lighty, J. S.; Silcox, G. D.; Senior, C. L. Experimental Evaluation
of the Effects of Quench Rate and Quartz Surface Area on Homogeneous Mercury
Oxidation. Thirty-First Symposium (International) on Combustion. The Combustion
Institute: Pittsburgh, PA, 2006.
4 Imada, N., H. Kikkawa, K. Kobayashi, N. Oda, “Study of Mercury Behavior in Flue
Gas of Oxy-fuel Combustion,“ The 35th International Technical Conference on Clean
Coal & Fuel Systems, Clearwater, Florida, June 6 to 10, 2010.
13