United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S7-84-087 Sept. 1984
SERA Project Summary
Development of Criteria for
Extension of Applicability of
Low-Emission, High-Efficiency
Coal Burners: Fourth Annual
Report
R. Payne, P. L. Case, M. P. Heap, J. Lee, C. N. McKinnon, P. Nelson, and D. W.
Pershing
This report summarizes technical pro-
gress during the fourth year of effort on
EPA Contract 68-02-2667. NO, and
SOx emission characteristics of two
low-NOx distributed-mixing burners
were tested with three coals in a large
water-tube simulator furnace (50-70 x
10* Btu/hr firing rate). Increasing burn-
er load, burner zone stoichiometry, or
overall excess air increased NO, emis-
sions. Staging was limited by increases
in CO emissions and problems with
flame stability at burner zone stoichio-
metries below 0.6. The feasibility of
using dry sorbents injected directly into
the furnace for SO2 emission control
was investigated. Sorbenttype[Ca(OH)2
vs. CaCOa] and injection location had a
small effect on sulfur capture. Fuel
sulfur content also affected sulfur cap-
ture. Sulfur captures on the order of 40-
50 percent were obtained at Ca/S
molar ratios of 2.0.
Parametric studies of the sulfur cap-
ture process were performed in a spe-
cially constructed Boiler Simulator Fur-
nace fired at 1.0 x 10s Btu/hr. Under
fuel-lean conditions, thermal history
exerted controlling influence on sulfur
capture. Firing and heat extractio'n rates
strongly influenced sulfur capture due
to their effect on thermal history. Sor-
bent location, burner zone stoichio-
metry, overall excess air, and tertiary air
velocity had secondary effects on cap-
ture. Capture under fuel-rich conditions
was also investigated. Significant reten-
tion of sulfur in the solid phase and a
gaseous sulfur species distribution con-
siderably different than that expected
from equilibrium calculations were ob-
served. Sorbent injection location and
method affected sulfur capture under
fuel-rich conditions.
This Project Summary was developed
by EPA's Industrial Environmental Re-
search Laboratory. Research Triangle
Park, NC. to announce key findings of
the research project that is fully docu-
mented in a separate report of the same
title (see Project Report ordering infor-
mation at back).
Introduction
This report describes the progress on
EPA Contract 68-02-2667, "Development
of Criteria for Extension of Applicability of
Low-Emission, High Efficiency Coal Burn-
ers," from October 1,1980, to October 1,
1981. The program, initiated in October
1977, was structured to provide data for a
wider design base in applying the dis-
tributed mixing burner (DMB) as a means
of controlling the emissions of NO*. In this
context, the program's main objectives
were to:
• Expand the fuel capability of low NOX
burners to include the major types of
-------
solid fossil fuels projected for use by
the utility industry.
• Explore additional burner concepts and
configurations that show potential for
improving the emission and thermal
performance of pulvenzed-coal burn-
ers.
• Determine the effects of multiple burn-
er configurations that are encountered
in utility boilers.
• Directly compare the experimental
burners being developed here and the
current state-of-the-art for commer-
cially available coal burners.
• Provide testing in support of planned
application of the burner technology.
To meet these objectives, the program
was divided into seven tasks; see Figure
1. These tasks were designed to provide
experimental data relative to the DMB.
The work moves progressively from
bench-scale studies of the basic process-
es, through single burner studies at a
range of pilot scales, to multiple burner
configurations and a comparison of DMB
emission and operating characteristics
with current commercial technology. The
base program provides also for technology
transfer and review of progress by the
industry.
During the past reporting period, the
main focus of the program has shifted
somewhat, and renewed attention has
been directed toward dry sorbent SO,
control technology. The use of sorbents to
control emissions of SO, is not new and
was extensively studied in the late 1960's
and early 1970's in a series of develop-
ment and demonstration projects. Al-
though the pilot plant studies showed
promise, the results could not be dupli-
cated in full scale systems. This work,
together with results from recent pilot
scale studies, was reviewed; some of the
more recent data suggest that burner
conditions necessary to minimize NO, are
also favorable for sulfur capture by in-
jected sorbents. This has led to the
concept of a combined N0,/S0, control
strategy and to the need for a definition of
the conditions under which the emission
of both pollutants can be minimized
without significant impact on overall
combustion system performance. A modi-
fied program plan approach for Contract
68-02-2667, which extends the original
goals to include SO, control, is shown in
Figure 2. The experimental portion of this
program includes both bench scale stud-
ies to define optimum conditions for
sulfur capture, and pilot scale studies
with real burner systems to optimize and
evaluate the potential for combined
NO,/SO, control.
The technology which has the potential
to provide simultaneous control of NO,
Program Planning
Task 1
Program Definition
— Fuels Selection
— Experimental Plan
^— Measurements Protocol
Experimental Effort
Task 2
Fuel Screening Experiments
Task 3
Single-Burner Experiments
Task 4
Multiple-Burner Experiments
Tasks
'omparison to Current Technology
- Key Program Elements
Technology Transfer
Task 6
Industry Coordination
Design Guidelines
Task 7
Data Analysis and Criteria
Development
Figure 1.
Program structure for developing criteria for extending applicability of low-
emission/high-efficiency coal burners.
and SO, has been dubbed LIMB—lime-
stone injection into multistage burners.
FigureS shows the LIMB process, simpli-
fied. Coal, combustion air, and the sorbent
(usually some form of calcium carbonate)
are injected into the furnace. The sorbent
may be mixed with the coal or injected
with one of the combustion air streams
(e.g., staging air). Subsequent events can
be considered to take place in three
regions:
1. Particle Heating—a short residence
time region where the coal and
sorbent are heated, typically at
rates of 104 K/sec.
2. Fuel-Rich—a reducing region, de-
ficient in oxygen.
3. Fuel-Lean—an oxidizing region
where the coal is burned out and
the combustion products are cooled
before entering the convective sec-
tions.
The oxidation of fuel-bound nitrogen
(producing fuel NO) accounts for 80
percent of the total NO,formed during the
combustion of pulverized coal. The mini-
mization of fuel NO formation requires
that volatile fuel nitrogen species are
prevented from reacting with oxygen and
forming NO. During combustion the vola-
tile fuel nitrogen species (XN) can follow
two general paths:
• Fuel-lean—
XN+02~NO (1)
• Fuel-rich—
XN-HCN-N-NH-NHa -NH3 (2)
N + NO-Nz + O (3)
Reaction 2 summarizes the nitrogen shift
reactions that produce HCN or NH3 under
fuel-rich conditions. Reaction 3 is the
most probable path forming N2 in pulver-
ized coal flames. All fuel-bound nitrogen
is not volatile, and the coal char contains
nitrogen which can be oxidized. Since
char burnout must be under oxidizing
conditions, NO production from char
nitrogen is unavoidable. Consequently, it
is important that the conditions in the
heating and fuel-rich zones promote the
evolution of fuel nitrogen species.
The oxidation of both organic and
inorganic coal sulfur produces the S02
emissions from pulverized coal combus-
tion. The ultimate gas-phase sulfur spe-
ciation depends on the temperature and
gas-phase stoichiometry. In the fuel-rich
region sulfur will exist as H2S, COS, or
CS2, and in the fuel-lean regional most al
the sulfur will exist as S02. An injectet.
-------
1
r
Sms
Concept t
|/V0,/SO,<
1 *
II Scale
development
•
Combustion Modification
ol System Evaluation
I Cor
cept/Developmenn
i
ft.
Pilot
Concept D
1
Scale
evelopment
•
i + *
Gas-Phase Experiments
(Argon Furnace)
Measurement Systems Dev.
I/A Contacting
Ca/S
Flame Temperature
F/A Ratio
*
Define Sulfur/ _
Sorbent Chemistry
Systems Analysis
[
U
<
Preliminary
* Conceptual
Analysis
Conceptui
—Sorbent
Boiler Sys
—Reliabili
New low f
—SO, Cor
Extensive
Engineerin
Analysis
Screening Experiments
BSF
Inheren
Operati
Input Vi
1
t Capture
ig Variables
Properties
iriables
P
Recommended Burner
m •» « Concepts for Pilot —
Scale Study
Feed Systems
terns Analysis
ty/Operability
I/O, Boiler Systems
itrol
i
>
^ Recommended Boiler
Systems Concepts
/Concept Optimization
ISMS
II 10x106Btu/Hr
r*1 Scale Burner
1 1 Evaluation
\ u*
\
\IWS
icted Concepts
02)
Desigi
MlnNO
*
| \60-120 x 10s Btu/Hr^
U Scale Burner J
| Evaluation \
L Two Concepts or
Design Conditions
1
•
LSpec
— * Sys
Field Dt
Existing low rVO,
Burner Systems
Steini
WIPCl
B/Wi
Other
Appr
7 Conditions for
^Max S Capture
nuller
J Prototype
•JRB
s as
opriate
Potential Near- Term
Demonstration ofNOt/SO,
Control
'
ification of
'monstrations
I Extensive Engineering System Design
I —2 Systems (Minimum)
' Engineering Cost Estimates
Figure 2, Program plan approach—combined NO^/SO^ control.
sorbent can capture sulfur in both the
fuel-rich and fuel-lean regionsfsee Figure
3). In the fuel-rich region the reduced
sulfur species can react with sorbent
directly, producing CaS:
CaCOa + H2S - CaS + H20 + C02 (4)
under fuel-lean conditions SO2 reacts
with the calcined sorbent, producing
CaS04:
CaO + S02+ 1/202-CaS04 (5)
Rates of both reactions depend on the
gas-phase sulfur concentration which
benefits the fuel-rich capture because
the reactants are more concentrated (if
the sulfur content of the char is mini-
mized). Fuel-rich capture could have the
further advantage -that calcium sulfide
may not block the pore structure, allowing
greater utilization of the limestone.
Before the fuel-lean capture process
can take place, the sorbent must be
calcined. The reactivity of the calcine
appears to depend most strongly on peak
temperatures. If temperatures are too
high the material is dead-burned, result-
ing in a dramatic reduction in the surface
area and a loss of reactivity. Consequent-
ly, the thermal history experienced by the
sorbent particles is critical, and it may be
advantageous if they are excluded from
the heat release zone where peak tem-
peratures are highest. Regeneration of
SOZ can occur: the sulfide is more stable
at higher temperatures than the sulfate
but, if capture occurs under fuel-rich
conditions, the sulfide must be converted
-------
Input
Partially Calcined
Sorbent
Figure 3. Simplified process description of LIMB.
to sulfate in the fuel-lean region without
decomposing the CaO and S02.
The sulfur capture process results in
particulate emissions which potentially
consist of: fly ash, partially calcined
sorbent, calcined sorbent with varying
degrees of reactivity, and partially sul-
fated sorbent. The composition and prop-
erties of the solid emissions will depend
on coal and sorbent properties as well as
the LIMB process.
In the present reporting period of the
contract, pilot scale studies have been
carried out with a reduced scale (50 x 106
Btu/hr) version of the Steinmuller Staged
Mixing Burner to compare DMB perform-
ance and current commercial technology.
A field prototype version of the DMB
burner system has also been extensively
tested under EPA Contract 68-02-3127.
Both burner tests also provided an op-
portunity to investigate sorbent injection
in unmodified (for SO, control) systems.
While the experimental data indicate
some potential for significant SO, removal
under low NO, conditions, the pilot scale
results provide little further understand-
ing of the basic mechanisms involved in
sulfur capture, or of the ideal conditions
required for process optimization.
Most experimental effort during this
period was on characterizing the S02
capture process using bench-scale (up to
1.0 x 106 Btu/hr firing rate) facilities.
During the bench scale studies, two
experimental approaches were taken.
The first attempted to remove some of the
complexities associated with coal com-
bustion and examined sorbent/sulfur
interactions in a totally gas-phase system.
In the second, studying coal combustion
under furnace conditions which simulate
the temperature/time history of a com-
plete boiler system, it was necessary to
construct a new Boiler Simulator Furnace
(BSF) facility. Experimental results from
Fly Ash
both phases of the bench-scale experi-
mental program are presented and discussed.
Analytical requirements for this pro-
gram were extensive, and previous ex-
perience has shown the need for carefully
controlled conditions when sampling for
sulfur species in the presence of active
sorbents. These requirements necessitat-
ed the development of several new sam-
pling and analytical capabilities at EER,
and these are described. Some conclu-
sions (although still of an interim nature)
drawn from the results of the work are
presented.
Pilot-Scale Results—IMOX and
SOX Control
Two low-NO, distributed-mixing burn-
ers were tested in the large watertube
simulator (LWS) to determine the effect of
various operational and fuel related pa-
rameters on NO, emissions. The results
-------
obtained with these burners in the LWS
are compared with results obtained in
other facilities in Figures 4, 5, and 6. In
general, even though absolute levels of
NOX emissions were different in the dif-
ferent facilities, effects of burner zone
stoichiometry (Figure 4), overall excess
air (Figure 5), and load (Figure 6) are
similar. Effects of these operating pa-
rameters were also independent of coal
type, although again the absolute levels
of NOX emissions were different for the
different coals in the same facility (the
LWS). In general, increased load, excess
air, and burner zone stoichiometry (de-
creased staging) all increased NOX emis-
sions for all the coals in the facilittes
compared. NO* emission characterization
of the EPA prototype distributed mixing
burner was performed under EPA Con-
tract 68-02-3921 and is described in
detail in that contract report. In general,
the effects of burner zone stoichiometry,
excess air level, and load on NOX emis-
sions were similar to those observed with
the Steinmuller burner; again, baseline
NO, emissions for the different coals
were different and NOX emission levels
for the same coals were slightly different
for the two burners.
In addition to the characterization of
NO. emissions, a series of experiments
were carried out with both burners to
determine the potential for SO2 removal
by direct injection of calcium-based sor-
bents. These studies were carried out
without modifying either burner and only
at conditions which were optimized for
-a
o
I
A
o
600
500
400
300
200
700
I
I
I
I
LWS Furnace
(50 x 10e Btu/hr)
_1_
I
0.5
0.6
0.7 0.8
5/?B
0.9
1.0
Figure 4. Effect of burner zone stoichiometry on /VO« emissions—comparison of data,
Steinmuller burner.
NO, emissions. No attempt was made to
optimize the firing system for sulfur
capture. Results of the sorbent injection
tests in the two burners indicate that
sorbent type [Ca(OH)2 or CaC03] and
injection location (with fuel or staging air)
had small effects on the sulfur capture in
either burner. The effect of sorbent
location was more pronounced with the
prototype burner and appears to be a
result of changes in sorbent thermal
history. The data also indicate that sulfur
capture increases with coal sulfur con-
tent. Figure 7 summarizes results from
studies with the EPA prototype and
Steinmuller burners in several furnaces.
Results from Steinmuller tests in the
LWS and in the IFRF furnace are included,
along with data from the prototype burner.
Comparing the LWS results using the
prototype and Steinmuller burners with
Indiana coal, shows that the prototype
burner resulted in higher sulfur capture
even though the burners were operated
at comparable conditions. This perhaps
indicates the influence of burner design
conditions. The LWS results using a
prototype burner with Utah coal and the
low sulfur coal IFRF test results show
good general agreement. This indicates
that burner effects may be unimportant
for low sulfur coals; since these results
are from burners which were not opti-
mized for SOz capture, they should be
treated with caution.
The good agreement between the LWS
data for Indiana coal and the IFRF data
with the high sulfur coal is somewhat
surprising in light of the higher furnace
temperature (1000°C vs. 600°C in the
LWS) employed at IFRF.
In general, the pilot-scale tests indicate
a lack of more definitive assessments of
the parameters controlling SO2 capture
via dry sorbents. Of help would be studies
focused on the potential for application of
this technology in field boilers, and the
optimum burner design and operating
parameters required to maximize sulfur
capture without affecting NO, emissions
or boiler efficiency. Further studies of
sorbent injection for SOz control are
planned as a continuation of the 68-02-
2667 contract effort. Several pilot-scale
furnaces using two burners will be tested
in the next year.
Bench-Scale Studies of SO2
Capture by Dry Solvent
Injection
The major experimental effort during
this reporting period was characterization
of the SOz capture process at bench-
-------
600
500
400
I
g
300
200
700
X Saar«>
«*»
LWS Furnace
(50 x 106 Btu/hr)
I
I
1.0
1.4
Figure 5.
1.1 1.2 1.3
Total Stoichiometric Ratio (SR^
Effect of excess air on NO Demissions — comparison of data, Steinmuller burner.
scale. The effort included: development of
sampling and analytical systems for meas-
uring SOs and determining sulfur cap-
ture; design, construction, and testing of
an experimental facility, the Boiler Simu-
lator Furnace(BSF); and investigating the
effect of various combustion and opera-
tional parameters on sulfur capture in the
BSF. The measurement of sulfur species
in combustion products containing active
sorbents introduces several problems
related to sample acquisition. A "phase
discrimination" probe, which minimizes
gas/solid contacting after sample extrac-
tion, was designed, constructed and
tested. Methods for measuring sulfur
species concentrations in fuel-rich com-
bustion gases were developed as were
methods for solids analysis (including
sorbent, ash, and coal composition and
surface areas).
The bench-scale facility constructed for
these studies can duplicate the thermal
history of the solid particles (coal and
sorbent) and the products of combustion
in a pulverized coal-fired boiler. The
facility consists of three major compo-
nents:
A radiant furnace, a horizontal refrac-
tory-lined cylinder which simulates
the firing zone. Removable cooling
tubes provide heat extraction.
A post-flame cavity which simulates
the space between the firing zone and
the superheater of a conventional
boiler.
A convective section, cooled by banks
of air-cooled stainless steel tubes,
which simulates the superheater, air
heater, and economizer.
a Schematic of the Boiler Simulator
Furnace is shown in Figure 8.
A parametric screening study of the
effect of combustion and operation
variables on sulfur capture by sorbents
directly injected into the furnace was
planned; the first portion was com-
pleted. The results of program defini-
tion experiments were used to specify
design criteria for the BSF and param-
eters for the first set of screening
studies. Parameters investigated in-
cluded: firing rate, heat extraction
rate, sorbent injection location, overall
excess air, burner zone stoichiometry,
and tertiary air velocity. One sorbent
(Vicron 45-3, a large-grain calcite
limestone) and one coal (a medium
sulfur [1-2 percent], bituminous Indi-
ana coal) were used for all studies.
The studies established that under
fuel-lean conditions, thermal history
exerts a controlling influence on sulfur
capture. Results shown in Figure 9
showtotal sulfur capture as a function
of calcium-to-sulfur ratio for three
thermal conditions. Furnace temper-
atures are also shown for the uncooled
and most heavily cooled conditions.
The effect of thermal history on
capture in the various furnace zones
was further analyzed. Capture was
measured at the exit of the radiant and
post-flame sections as well as at the
furnace exit. Results summarized in
Figure 10 show sulfur capture in the
radiant and post-flame sections of the
BSF as a function of Ca/S-'S'O'z in each
zone. Three conditions are shown:
high firing rate (with and without
radiant zone cooling) and low load
without cooling. Reducing the firing
rate reduced the temperatures and
increasedtheresidencetimes. Cooling
the radiant zone decreased peak tem-
peratures with little effect on the
residence times. Radiant zone cooling
dramatically increased capture; the
low-load conditions gave relatively
poor capture in the radiant zone, but
both reduced-temperature cases (cool-
ed and low-load) gave approximately
the same capture in the post-flame
section.
Other parameters(mcludmg sorbent
location, burner zone stoichiometry,
overall excess air, and tertiary air
velocity) had secondary effects, which
-------
600
500
400
o
300
200
700
We/her Burners (At
Constant Boiler
Load)
Saar Coal
Boil*'
LWS
I
\
I
I
60
80
120
140
Figure 6.
100
Load % Nominal
Effect of load on /VO» emissions—comparison of data, Steinmuller burner.
depended to some extent on other
variables, on sulfur capture.
Sulfur capture under fuel-rich condi-
tions was evaluated using both Indiana
coal and propane/H2S as fuels. Fuel-
rich conditions were obtained by firing
the burner with a sub-stoichiometric
amount of air, running the entire
radiant furnace fuel-rich, and adding
the remainder of the combustion air at
the base of the radiant furnace. Sulfur
capture appears to be proportional to
sulfurspeciesdrivingforcetothe 1 /2-
power. The increase in sulfur species
concentration under fuel-rich condi-
tions (due to reduced dilution) should
benefit sulfur capture. Figure 11 sum-
marizes data on sulfur species distri-
bution as a function of stoichiometric
ratio based on equilibrium calculations
(free energy minimization) and on
actual experimental measurements of
the gas phase in the BSF near the end
of the fuel-rich zone. The equilibrium
calculations indicate that (for stoichio-
metries below 0.95) H2S is the only
species of importance. However, data
from propane/H2S and Indiana coal
measurements demonstrate that a
wide spectrum of reduced sulfur spe-
cies are present under these condi-
tions. With the propane/air flames,
S02 was significantly more important
at rich conditions than would have
been anticipated from equilibrium
calculations, and both COS and CSz
were nonnegligible. The data in Figure
11 indicate that with the Indiana coal
LWS
O Prototype Burner—Indiana Coal
O Prototype Burner—Utah Coal
0 Steinmuller Burner—Indiana Coal
IFRF
80
1.09% S Coal
2.42% S Coal
70
S 60
S. 50
I
<§" 40
I
30
20
10
i
01 2345
Ca/S Molar Ratio
Figure 7. Comparison of Steinmuller and
DMB data—limestone injected
in tertiary ports
the gas-phase species distribution
was non-equilibrium and that a signif-
icant amount of coal sulfur remained
in the solid phase. These results
emphasize two problems associated
with sulfur capture under fuel-rich
conditions: SC>2 concentrations may
be nonnegligible, and a large fraction
of the sulfur may remain in the solid
phase. Tests will be carried out in the
next year with propane/H2S in an
attempt to separate t he effect of sulf ur
evolution from the coal, and capture
from the gas phase.
A further complication of the sulfur
capture process under fuel-rich condi-
tions is due to the impact of the
physical staging method, used to ob-
tain long residence times under fuel-
rich conditions, on the thermal charac-
teristics of the experimental system.
The data presented in Figure 12 was
obtained firing Indiana coal. The tem-
perature profiles were derived from
wall temperature measurements and
-------
Flame
Zone
(Radiant
Furnace)
Figure 8. Boiler simulator furnace—original configuration.
calculated residence times (based on
plug flow). It can be seen that first
stage stoichiometry had a dramatic
effect on the thermal history of the
reactants. Under rich conditions, the
radiant zone temperatures are signifi-
cantly reduced, and the residence time
is increased due to the decreased
mass flow rate. The temperature at the
entrance of the post-flame section,
where the second-stage air is added,
increases during burnout of the coal.
The magnitude of this increase will
depend on the primary-zone stoichio-
metry and on the inter-stage heat
removal. As shown in Figure 12, the
effect of first-stage stoichiometry on
sorbent utilization for Indiana coal at
rich-zone stoichiometries greater than
0.7 was slight. However, captures
decreased dramatically for fuel-rich
stoichiometries less than 0.7. This is
probably due to a combination of
regeneration during burnout, a reduc-
tion in sulfur species driving force
because of sulfur retention in the char,
and a decrease in the reaction rate
because of the lower temperatures in
the fuel-rich region.
Further parametric screening studies
at bench scale are planned. In the next
year: sulfur capture will be investigated
in the fuel-rich region of gas flames
(eliminating the char sulfur retention
complication) in the BSF; the effect of
thermal environment on sulfur capture
will be investigated in more detail; and
the effect of fuel and sorbent type on
sulfur capture will be investigated.'
8
-------
i i r
70 —
i
I
«
o
to
Additional -
Cooling
125.000
Btu/hr ,
i r
3000
2500
2000
1500
1000
500
1 2 3 4 c
Ca/S Molar Ratio
Date: 6/25. 7/1. 7/29/81
Rate: 800.0OO Btu/hr
Cooling: As Shown
SRf.- 0.2
SRe: 0.6
SR-,: 1.20
Residence Time, sec
Coal: Indiana
Ca/S Molar Ratio: As Shown
Sample Port: #3
Sorbent: CaCOa. 45-3
Injection Loc.: Tertiary Air
Figure 9. Influence of heat extraction rate—base case conditions.
[7 High load uncooled
O High load cooled
i Low load uncooled
60
•S
9)
t 40
Q)
20
— Radiant
~ Post-Flame
50
100
150 0
fSOl(Ca/S>
50
100
150
Figure 10. Effect of load and radiant zone cooling on sulfur capture.
-------
100
80
I
•3 60
& 40
20
(AJ
H£
-S03
SO2
H,S
0.6 0.8 1.0 1.2 0.6 0.8 1.0 1.2 0.6 0.8 1.0 1.2
SR SR SR
Figure 11. Effect of stoichiometry on sulfur species distribution. A, equilibrium calculations;
B, propane and H£; C, Indiana coal.
30
&* 20
§|
1$
i!
s?Ts 10
oo
o/ o
-SR,=0.65
2500
u.
o
9)
1500
-------
o
o 220
CO <0
1— J>
r~ur" rv
CD*-' "XX
o-co-x
OX3DO
X »-»
m c
m z
m
2
o
5'
Da
§2
0) ~
- o
O3
I
01
10
O)
oo
------- |