co
00
o
CQ
Cu
J
<
O
u
s
o
OS
w
2
O
t—i
co
CO
W
co
D
O
W
co
<
O
b
O
J
o
£ <*>
h Ph
25
Sd
q2
w
Z CQ
° A
H-. Q
h w
< N
N —
~ Q
CC H-4
W D
hJ
(J
^ Q
PS m
C ^
^ 05
i-i-K t—4
V fc
CO
C
X) 03
cd
0 a
4->
PS
CD
bO
¦O Sh
•»
C .rH
c
CU >
o
o
CO
a> .
r-
•rH
c ctf
®
JQ
C3 -rH
1—<
O
> SH
PS
W T3
Sh
C
•
-
-------
PB 198 413
CHARACTERIZATION AND CONTROL OF
GASEOUS EMISSIONS
FROM
COAL-FIRED FLU IDIZED-B ED BOILERS
INTERIM REPORT
OCTOBER. 1970
FOR
DIVISION OF PROCESS CONTROL ENGINEERING
NATIONAL AIR POLLUTION CONTROL ADMINISTRATION
ENVIRONMENTAL HEALTH SERVICE
PUBLIC HEALTH SERVICE
DEPARTMENT OF HEALTH. EDUCATION. AND WELFARE
POPE, EVANS AND ROBBLNS
CONSULTING ENGLNEERS
A DIVISION OF PERATHON INCORPORATED
NATIONAL TECHNICAL
INFORMATION Ss^VICE
v«-
INTERIM REPORT
On
CHARACTERIZATION AND CONTROL OF
GASEOUS EMISSIONS
FROM
COAL-FIRED FLU IDI ZED-BED BOILERS
to
DIVISION OF PROCESS CONTROL ENGINEERING
NATIONAL AIR POLLUTION CONTROL ADMINISTRATION
ENVIRONMENTAL HEALTH SERVICE
PUBLIC HEALTn SERVICE
DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE
October 197 0
by
E. B. Robison, A. H. Bagnulo,
J. W. Bishop, and S. Ehrlich
POPE, EVANS AND ROBBINS
Consulting Engineers
A DIVISION OF /PERATHON INCORPORATED
Alexandria, Virginia 22314
-------
TABLE OP CONTENTS
1. Summary
1.1 Test Procedures
1.2 Sulfur-Dioxide Emission
1.3,Sulfur Trioxide Emission
1.4 Hydrocarbons Emission
1.5 Oxides of Nitrogen Emission
1.6 Particulate Emission
2. Conclusions
3. Recommendations
4. Introduction
4.1 Operating Variables
4.2 -Additive (Sorbent) Variables
4.3 Coal Composition
5. Equipment and Procedures
5.1 Pilot Scale Combustor, FBC
5.2 Full-Scale Boiler Module, FBM
5.3 Method of Sorbent Feed
5.4 Instrumentation
5.5 Materials
6. Results of Pilot Scale (FBC) Tests - Sintered Ash Bed
6.1 Sulfur Dioxide Emission with Coarse Additives
6.2 Sulfur Dioxide Emission with Finely Divided
Sorbents
6.3 Tests for Independent Effects of Bed Temperature,
Bed Depth, Sorbent Particle Size, Sorbent Distri-
bution, and Superficial Velocity
6.4 Sulfur Trioxide Emission
6.5 Hydrocarbons Emission
6.6 Oxides of Nitrogen Emission
6.7 Particulate Emission
6.8 Operation at Reducing Conditions
6.9 FBC Operation with a Limestone Bed
7. Results of Boiler Module (FBM) Tests
7.1 Sulfur Dioxide Emission
7.2 Hydrocarbons Emission
7.3 Nitrogen Oxides Emission
7.4 Effects of Fly-Ash Recirculation and Steam
Inje ction
7.5 Particulate Emission
F=>0.= =; AMD
TABLE OK CONTENTS (Continued)
Page
1
2
8. Discussion of FBC and FBM Test Results
9. Economic Analysis
^ 9.1 General
9.2 Basis of Performance Estimates
9.3 Performance Data j
® 9.4 Capital Requirements for Equipment ]
9.5 Annual Operating Costs 1
9.6 Comparison with Costs for Alternative Methods ]
® 9.7 Conclusions J
H APPENDIX A - Enclosures ^
APPENDIX B - FBM and FBC Test Data E
APPENDIX C - Sulfur Balance Data C
17
17
18
18
24
32
37
44
46
46
54
64
73
73
77
79
80
&1
96
100
100
10 5
105
106
ROBBING
-------
LIST OF FIGURES
Figure Page
1 Schematic of Fluidized-Bed Boiler 14
2 Fluidized-Bed Column (FBC) Construction Detail -
Front View 19
3 Fluidized-Bed Column (FBC) Construction Detail -
Side View 20
4 Air Distribution Grid Button 21
5 FBC Test System 22
6 FBC. Air and Exhaust Gas Ducting Showing
Sampling Points 23
7 FBC Cooling Coil Viewed from Above 25
8 Fluidized-Bed Module (FBM) Internal Construction 26
9 Schematic of the FBM Showing the Arrangement of
Pneumatic Coal and Additive Feed Tubes 28
10 Schematic Layout of the FBM Test System 29
11 FBM Test System 31
12 Schematic of the No. 1 Sorbent Feed System for
the FBC 33
13 FBC Operating with No. 1 Sorbent Feed System 34
14 FBC Operating with No. 2 Sorbent Feed System 35
15 Schematic of the No. 2 Sorbent Feed System
for the FBC 36
16 Vertical Cross Section of FBC Showing Sorbent
Feeders and Feed Points 38
17 FBM Sorbent Feed System 39
18 Analytical Instrumentation 40
19 Schematic of Gas Transfer System for Continuous
Monitoring of Sulfur Dioxide, Nitric Oxide and
Hydrocarbons 41
20 Schematic of the FBM Gas Sampling System 43
21 Reduction of Sulfur Dioxide Emission from the
FBC Burning a 4.5% S Coal with Coarse 1337
Dolomite Addition 50
22 Reduction of Sulfur Dioxide Emission from the
FBC Burning a 4.5% S Coal with Coarse 1359
Limestone Addition 51
23 Interpolation of 10-inch Bed Depch Data for
Comparison with 7-inch Bed Depth Data 53
iii
POFE EVANS AMD
LIST OF FIGURES (Continued)
Figure Page
24 Comparison of Sulfur Dioxide Reductions
Observed with the Coarse and Fine Sorbent
ar" a 4.5% Sulfur Coal 56
25 Sulfur Dioxide Reduction with Fine Sorbent
Addition to the FBC Burning a 4.5% Sulfur
Coal 60
26 Sulfur Dioxide Reduction with Fine Sorbent
Addition to the FBC Burning a 2.6% Sulfur
Coal 63
27 Variation in Sulfur Dioxide Reduction with
Sorbent Particle Size, Bed Depth and Bed
Temperature 68
28 Emissions from FBC Test No. 73 Burning a
Medium Sulfur Coal and Injecting -325 Mesh,
1359 Limestone on 1, 2 and 4 Sides 70
29 Hydrocarbons Variation with Flue Gas Oxygen
Concentration in the F^C Operation 74
30 Typical Variation in Nitric Oxide Concentration
with Oxygen Content in the Flue Gas from the
FBC 76
31 Measured Values of Nitric Oxide Concentration
in the Flue Gas at 3% Oxygen and Various Bed
Temperatures shown with Theoretical Equilib-
rium Values for the Temperature - 02 Content
Regime 7 8
32 Emissions Monitored in FBC Test 113 Burning a
Medium Sulfur Coal in a Fluidized-Bed of
1359 Limestone 84
33 Variation of Sulfur and Calcium in Bed and
Fly Ash During FBC Test 113 86
34 Emissions from the FBC Burning a 3% Sulfur Coal
in a Limestone Bed with Mild Reducing Conditions
and Regeneration 87
35 Emissions During FBC Test 114 Burning a Medium
Sulfur Coal in a Fluidized-Bed of 1359 Limestone 91
36 Variation in Bed and Fly hs^ Sulfur and Calcium
During Test 114 92
37 Sulfur Dioxide Reduction with Sorbent Addition
to the FBM Burning a 4.5% Sulfur Coal 99
38 Sulfur Dioxide Reduction with Sorbent Addition
to the FBM Burning a 2.6% Sulfur Coal 102
J —El 'AK5 Fee
-------
LIST OF FIGURES (Continued)
Figure Page
39 Variation m Hydrocarbons Concentration
with Flue Gas Oxygen Content in the FBM 103
40 FBM - Variation of Nitric Oxide Concentration
with Flue Gas Oxygen Content 104
41 FBM Collector Emission. Cumulative Frequency
Distribution of Particle Count 108
42 Estimated Reduction in Sulfur Dioxide Emission
by Addition of 1359 Limestone at Various
Stoichiometric Ratios 117
43 Ratio of Sulfur Emission to Sulfur Input
vs Limestone Flow Rate for the 4.5% S Coal 118
44 Ratio of Sulfur Emission to Sulfur Input
vs Limestone Feed Rate for the 2.6% S Coal 119
45 Schematic Arrangement of the Sorbent Feed
System for a Full Scale Fluidized-Bed Boiler 122
46 Total Fly Ash Rates for Full Load Operation
of a 250,000 Lb Per Hr Fluidized-Bed Boiler
with Limestone Addition (4.54 S Coal) 124
47 Total Fly Ash Rates for Full Load Operation
of a 250,000 Lb Per Hr Fluidized-Bed Boiler
with Limestone Addition (2.6% S Coal) 125
48 Estimated Total Cost of Converting a High
Sulfur (4.5%) Coal to a Lower Sulfur Coal
Equivalent by Limestone Addition to a
250,000 Lb Per Hr Fluidized-Bed Boiler 130
49 Estimated Total Cost of Converting a Medium
Sulfur (2.6%) Coal to a Lower Sulfur Coal
Equivalent by Limestone Addition to a
250,000 Lb Per Hr Fluidized-Bed Boiler 131
v
POPE EVANS AND ROHE1NS
LIST OF TABLES
Table
I
IT
III
IV
V
VI
VII
VIII
IX
X
XI
XII
XIII
XIV
XV
Pa9e
Sulfur Dioxide Reductions Observed with
Addition of Coarse (-7 +14 Mesh) 1337 Dolomite
to the FBC Burning a 4.5% Sulfur Coal 47
Sulfur Dioxide Reductions Observed with
Addition of Coarse (-7 +14 Mesh) 1359 Limestone
to the FBC Burning a 4.5% Sulfur Coal 49
Sulfur Dioxide Reduction Observed with Fine
Sorbent Addition to Combustion of a 4.5% Sulfur
Coal in the Fluidized Bed 58
Sulfur Dioxide Reduction Observed with Fine
Sorbent Addition to Combustion of a 2.6% Sulfur
Coal in the Fluidized Bed 61
Data Summary for S02 Reduction vs. 1359 Lime-
stone Particle Size, Bed Depth and Temperature 66
Data Summary for the Superficial Velocity Tests 7 2
Data Summary for Operation at Reducing Conditions 8 2
Data Summary for the 1359 Limestone Bed Tests 89
Sulfur Dioxide Reductions with Addition of Coarse
(-7 +14 Mesh) 1337 Dolomite to the FBM Burning a
4.5% Sulfur Coal 97
Data Summary for Injection of -3 25 Mesh Sorbents
into the FBM Burning a 4.5% Sulfur Coal 98
Data Summary for Injection of -325 Mesh Sorbents
into the FBM Burning a 2.6% Sulfur Coal 101
Summary of Capital Cost Components for Limestone
Addition Per Boiler. 500,000 Lfc/Hr Steam Plant
Consisting of Two 250,000 Lb/Hr Coal-Fired,
Fluidized-Bed Boilers 121
Operating Cost Ingredients Summary for Fine
Limestone Injection in a 500,000 Lb/Hr Fluidized-
Bed Boiler Plant 128
An.ual Operating Cost Data lor Fine Limestone
Injection 129
Cost of Reducing a High Sulfur Coal to Equivalent
0.7% Sulfur Coal in a Fluidized-Bed Boiler, Com-
pared with Cost of Purchase of Low Sulfur Coal in
the Chicago Area 133
-------
LIST OF ENCLOSURES
APPENDIX A
Enclosure Page
1 FBC Specifications A-l
2 FBM Specifications A-3
3 Derivation of the Bed Depth Relationship
from Balance of Heat and Mass A-5
4 Schematic Cross Section of the #2 Sorbent
Feed System A-7
5 Schematic of Sulfur Trioxide Condenser A-8
6 Isokinetic Probe for Particulate Emissions
Determinations A-9
7 Schematic Arrangement of FBC Air and Flue Gas
Ducting Showing Gas Sampling Points and Thermo-
couple Locations A-10
8 Schematic Arrangement of the FBM Test System
Showing Gas Sampling Points and Thermocouple
Locations A-ll
9 Analysis of Hydrocarbons Analyzer Calibration
Gas A-12
10 Sulfur Dioxide Calibration Gas Analysis A-13
11 Nitric Oxide Calibration Gas Analysis A-14
12 Calibration Curve for Sulfur Dioxide Infra-
red Analyzer A-15
13 Calibration Curve for Nitric Oxide Infra-
red Analyzer A-16
14 Analyses of "Perfect Eight" Unwashed 4.5%
Sulfur Coal A-17
15 Analyses of "Perfect Eight" Washed 2.6%
Sulfur Coal A-18
16 Analysis of Sorbents After Ignition A-19
17 Terminal Velocity and Minimum Fluidization
Velocity vs. Particle Diameter A-20
18 Estimation of Elutriating Particle Size for
the 1359 Limestone A-21
19 Nitric Oxide Equilibrium Concentrations for
the Fluidized-Bed Environment A-22
20 Particle Size Distribution of 1359 Limestone
Bed Before and After Fluidized-Bed Combustion A-23
POPE. EVANS AND ROBBINS
LIST or ENCLOSURES. (Continued)
Enclosure Page
21 Integrated Sulfur Balance for Sorption-
Desorption of the Limestone Bed During
FBC Test 114 >-24
22 Emissions During FBC Test 63 for Sulfur
Trioxide Emission Burning a 4.5% Sulfur
Coal A-28
23 Emissions During FBC Test 101 Burning a
Medium Sulfur Coal in a 1359 Limestone
Bed A-29
24 Emissions During FBC Test 102 Burning a
Medium Sulfur Coal in a 1359 Limestone
Bed A-30
25 Emissions During FBC Test 104 Burning a
Medium Sulfur Coal in a 1359 Limestone
Bed A-31
26 Emissions During FBC Test 105 Burning a
Medium Sulfur Coal in a 1359 L'-nestone
Bed A-32
27 Emissions During FBC Test 106 Burning a
Medium Sulfur Coal in a 1359 Limestone
Bed A-33
28 Emissions During FBC Test 107 Burning a
Medium Sulfur Coal in a 1359 Limestone
Bed A-34
29 Emissions During FBC Test 108 Burning a
Medium Sulfur Coal in a 1359 Liirestone
Bed A-35
30 Emissions During FBC Test 109 Burning a
Medium Sulfur Coal in a 1359 Limestone
Bed A-36
31 Emissions During FBC Test 110 Burning a
Medium Sulfur Coal in a 1359 Limestone
Bed A-37
32 Emissions During FBC Test 111 Burning a
Medium Sulfur Coal in a 1359 Limestone
Bed A-38
33 Emissions During FBC Tesc 112 Burning a
Medium Sulfur Coal in a 1359 Limestone
Bed A-39
viii
POPE EVANS AND ROBBINS
-------
LIST OF ENCLOSURES.' (Continued)
Enclosures Page
34 Emissions During FBC Test 113 Burning a
Medium Sulfur Coal in a 1359 Limestone
Bed A-4 0
35 Emissions During FBC Test 114 Burning a
Medium Sulfur Coal in a 1359 Limestone
Bed A-41
36 Emissions During FBC Test 115 Burning a
Medium Sulfur Coal in a 1359 Limestone
Bed with Change in Gas Velocity A-42
37 Emissions During FBC Test 116 Burning a
Medium Sulfur Coal in a 1359, Limestone
Bed with Regeneration A-43
38 Emissions During FBC Test 117 Burning a
Medium Sulfur Coal in a 1359 Limestone
Bed with Regeneration A-44
39 Emissions During FBC Test 118 Burning a
Medium Sulfur Coal in a Limestone Bed
with Bed Reducing A-45
40 Emissions During FBC Test 119 Burning a
Medium Sulfur Coal in a Limestone Bed
with Bed Reducing A-46
41 Emissions During FBC Test 120 Burning a
Medium Sulfur Coal in a 1359 Limestone
Bed with Ash Recirculation A-47
42 Emissions During FBC Test 75 Burning a
Medium Sulfur Coal with Injection of
-325 Mesh 1359R Limestone Above and at
Base of Bed at Constant Ca/S Ratio A-48
43 Emissions During FBC Test 76 Burning a
Medium Sulfur Coal with Addition of
-325 Mesh 1359R Limestone with Change in
Superficial Velocity A-49
44 Emissions During FBC Test 77 Burning a
Medium Sulfur Coal with Injection of
-325 Mesh 1359R Limestone with Change in
Superficial Velocity A-50
45 Proposed Design Arrangement for Minimizing
Particulate Emission Control Costs A-51
ix
POPE. EVANS AND P.OBBINS
1
1, SUMMARY
Pope, Evans and Robbins, in a continuing project* spon-
sored by the National Air Pollution Control Administra-
tion, has monitored air pollutant emissions from the
combustion of coal in a fluidized bed under a compara-
tively large number of different conditions. Efforts
were made to reduce emissions of oxides of sulfur by
the use of limestone-based sorbents and to determine the
conditions most favorable for the reduction. The major
test variables and ranges are summarized as follows:
Coal Type: Medium and high sulfur
Bed Temperature: 1500°F to 1900°F
Bed Depth: 6 to 20 inches
Bed Material: Sintered ash and limestone
Flue Gas Oxygen Content: 0.5 to 5%
Superficial Gas Velocity: 6 to 14 fps
Sorbent Type: A dolomite designated 1337 and a
limestone designated 1359
Sorbent State: Raw, hydrated and precalcined
Sorbent Particle Size: -7 to -325 mesh
Fly-Ash Recirculation: Full Range (0% to 80%)
Method of Sorbent Feed: Pneumatic feed with the
coal, pneumatic feed remote from
the coal feed and premixed with
the coal
The tests were conducted on both pilot-scale and full-
scale test units. The pilot scale fluidized-bed
combustor, designated the FBC, contained a rectangular
bed 12" x 16". The full scale unit, designated the FBM,
contained .a rectangular bed -v.20" x 72" and constituted
one half cell of a full-scale multicell boiler concept.
Emissions of sulfur dioxide, nitric oxide and hydro-
carbons were monitored continuously with periodic
samples taken for measurement of particulates and wet-
test determination of SOjj and NO*. When conditions most
favorable for air pollution control were established on
a pilot scale, the conditions were reproduced in tests
with the Fluidized-Bed Boiler Module (FBM).
This report describes the results of experiments carried
out between November 1967 and August 1969.
-------
2
1.1 TEST PROCEDURES
The FBC test program was begun with combustion of the
high sulfur coal in a sintered ash bed and addition of
sorbents in a particle size approximating the bed parti-
cle size (-7 +14 mesh). The 1337 dolomite and 1359
limestone*, in the raw and precalcined state, were injec-
ted at varying rates and operating conditions, i.e., bed
temperature and excess air. The superficial velocity
was held in the range of 12 — 14 feet per second for all
tests except two designed for this parameter.
The program was continued with the use of finely divided
sorbents in the sintered ash bed again with the high
sulfur coal. The decision to employ a smaller sorbent
particle size was based on increasing evidence that the
desulfurization reaction was limited by product shell
formation. Reducing the particle size increases the
surface-to-mass ratio and, in turn, the sorbent reac-
tivity.
The fine sorbents, both the dolomite and limestone, were
injected in the raw and calcined states, ground to a
-325 mesh particle size. A third state, the hydrate,
was studied because of its natural occurrence in a -325
mesh particle size. The test procedures involved,
principally, changes in bed temperature, sorbent feed
rate, and ash recirculation. The excess air was held
constant at a level which effected a 3% oxygen content
in the flue gas, a minimum value found necessary to
control hydrocarbons emission. The superficial velocity
was held in the 12 - 14 fps range.
The effect of varying the method of sorbent feed was
investigated. Three injection methods were studied
with lime hydrate fed at rates varying over the 1-3
stoichiometric range. The methods are distinguished
as follows:
a. Pneumatic injection of the sorbent at a single port
with the coal after having been mixed with coal in
the coal feed line.
b. Pneumatic injection into the fluidized bed at two
Designations established by Bituminous Coal Research, Inc.,
an affiliate of the National Coal Association, as follows:
1337 - 53% calcium carbonate and 46% magnesium carbonate?
1359 - 97% calcium carbonate.
3
ports remote from the coal feed port.
c. Premixing the sorbent and coal in the coal hopper.
The tests were conducted at the bed temperatures found
to be most favorable (1500°F - 1600°F). The excess air
and superficial velocity were restricted as before.
The two-port feed system was extended subsequently to
four-port feed in a study of sorbent distribution in
the bed.
A medium sulfur coal was tested with the dolomite and
limestone sorbents in the raw state, ground to -325
mesh, and as the hydrate. Bed temperature and feed
rates were varied for comparison of the response to
that observed with the high sulfur coal. Excess air
and superficial velocity were again held constant.
An investigation was conducted using the medium sulfur
coal to determine the independent effects of sorbent
particle size, bed depth and bed temperature. A cooling
coil inserted in the bed provided a variable heat trans-
fer surface for independent temperature control. Raw
1359 limestone with close cut particle sizes in the range
of -325 to -12 mesh was injected into beds 10 and 18
inches deep.
The last FBC investigation involving the use of a sin-
tered ash bed concerned the effect of reducing super-
ficial velocity (to 6 feet/sec) and the comparative
effect of sorbent injection above the bed.
The feasibility of burning coal in a fluidized bed of
limestone was demonstrated. A medium sulfur coal was
burned in the FBC containing a bed of 1359 limestone
initially in the raw state. Operating conditions, i.e.,
bed temperature and excess air, were varied for the
effect on emissions, sorption of sulfur in the bed,
subsequent desorption, calcination and bed loss. Heat
transfer measurements were made in the bed for compari-
son with values determined in the sintered ash bed.
Tests conducted in the full scale unit, FBM, were
devoted to the use of fine sorbents with combustion of
the medium and high sulfur coals in a sintered ash bed.
Emissions were monitored with injection of 1337 dolo-
mite and 1359 limestone in the raw state, ground to
-325 mesh particle size, and as the hydrate. The prin-
cipal variables were the sorbent feed rate and ash
recirculation. The temperature was held generally in
the range of 1500°F - 1600°F, and the flue gas oxygen
-------
4
at 3%. The superficial velocity was held in the 12- 14 fps
range for all tests. One test was conducted to ascertain a
possible correlation between nitrogen content in the coal
. and nitrogen oxides emission.
1.2 SULFUR DIOXIDE EMISSION
1.2.1 Reduction with Coarse Sorbents
Emission of sulfur, in the form of sulfur dioxide, from the
combustion of high sulfur coal in a sintered ash bed was
found to vary from 90% to 95% of the input sulfur. When
raw sorbents were injected into the bed in a relatively
coarse particle bize (-7 +14 uiusli) , the sulfur dioxide
emission was reduced more effectively with the 1337 dolo-
mite than with the 1359 limestone at the same Ca/S molar
feed ratios. The dolomite produced a reduction of 34% at
a ratio of 1.44, for a utilization of 37.4%, whereas the
limestone utilization was limited to 20.7%. The tendency
of the dolomite to decrepitate in the bed may have con-
tributed to its higher reactivity. Utilization is defined
as the percentage of input calcium which combines with
sulfur. The magnesium contained in the dolomite was as-
sumed to be inert.
The reduction in sulfur dioxide emission was found to im-
prove somewhat with increase in oxygen content in Lhe flue
gas. Near reducing conditions in the bed were found to
result in less effective sulfur capture. The reduction in
sulfur oxides was found to be more favorable at bed tempera
tures of 1500°F - 1600°F than at 1800°F when using the dolo
mite. With the coarse limestone addition, the effect of
Ded temperature was not well defined althuugh Uie reduction
in sulfur oxide emissions improved somewhat with increase
in bed temperature. Both sorbents precalcined by the sup-
plier were found to be less effective than the raw stone
under similar test conditions.
1.2.2 Reduction with Finely Divided Sorbents
a) Effect of fine grinding
FBC test results with -325 mesh sorbents indicated an
improvement over the coarse sorbent performance in
both sulfur dioxide reduction and sorbent utilization.
The performance was markedly improved in the case of
the 1359 limestone tests with the high sulfur coal;
this raw limestone fed at a Ca/S ratio of 1.5 indicated
an increase in utilization from 18% to 37% with the re-
duction in particle size. At the same stoichiometric
ratio, the 1337 dolomite utilization increased from
38% to 46% when the particle size was reduced from
-7 +14 mesh to -325 mesh. The tests conducted to de-
termine the independent effects of particle size, bed
5
depth, and bed temperature indicated that desulfuriza-
tion is strongly dependent on sorbent particle size
for a medium sulfur coal. Under similar test condi-
tions using the 1359 limestone, a 78% reduction in
sulfur dioxide emission observed with a -325 mesh
particle size was decreased to a 48% reduction when
the particle size was increased to 100 mesh. Reduc-
tions were even less with particle sizes larger than
100 mesh.
b) Effect of hydrating and precalcining
Performance of the fine raw sorbents in terms of sulfur
dioxide reduction at various Ca/S ratios was found to
be about the same as the corresponding hydrate. When
the hydrate of the 1337 dolomite was injected at a Ca/S
ratio of 2.0 burning the high sulfur coal, the reduction
in sulfur oxide emissions was 80% to 85%. The most
favorable single reduction was 88%, observed at a stoi-
chiometric ratio of 1.8 with this hydrate. Injection of
the 1359 limestone hydrate produced an 80% reduction at
a Ca/S ratio of 2.6. These results were found in the
FBC with a 10-inch deep bed operating at 1500°F to 1600°F,
3% oxygen in the flue gas, a -325 mesh particle size and
a superficial gas velocity of 12-14 fps.
The prccalcincd, finely divided, sorbents were found to
be considerably less reactive than the raw or the
hydrated sorbents.
c) Effect of sorbent type
The results indicate the dolomito to be more offoctivo
than the limestone when the stoichiometric ratio is
based on the calcium fraction of the dolomite only
(51% CaC03), but was less effective on a total sorbent
weight basis. The limestone containing 97% CaC03 would
be the more economical of the two sorbents in terms of
sulfur removal per unit weight of sorbent when the cost,
per ton of stone, is comparable.
d) Effect of coal S content
Percentage reduction in sulfur dioxide emission, as a
function of stoichiometric feed rate, was approximately
the same in the FBC for both the high sulfur coal
(4.5% S) and the medium sulfur coal (2.6% S). Under
the most favorable conditions, burning the 4.5% S Cual
and injecting the finely divided, raw, 1359 limestone,
utilization was found to be 40%, 33% and 20% at Ca/S
stoichiometric ratios of 1.0, 2.0 and 3.0 respectively.
Comparable utilizations were indicated in tests in the
larger FBM.
POPE. EVANS AND HOBBINS
-------
6
e) Effect of sorbent feed method
The tests to determine the most advantageous method of
sorbent feed failed to point up a significant advantage
for any one of the three tested, although in general
the best results were observed with a "two point"
feeder. This feeder provided pneumatic injection of
the sorbent into the bed at two points remote from the
coal feed port. Increasing the number of feed ports
in the FBC to four did not improve the sulfur capture.
The sorption efficiency was considerably less when a
sorbent stream directed into the bed was suddenly di-
verted to a feed port above the bed. These results
might have been anticipated -- a fluidized bed is a
good mixer; injection above a fluidized bed is similar
to injection into a conventional boiler.
f) Effect of bed temperature
Tests made to determine the effect of bed temperature
showed the sorbents to be more effective at the lower
end of the operating range (1550°F). Sulfur dioxide
reductions of 78% and 24% were observed at respective
temperatures of 1550°F and 1800°F.
g) Effect of bed height
At 1500°F a reduction of 73% with a 10-inch deep bed
increased to 78% with an 18-inch deep bed.
h) Effect of superficial velocity
Tests conducted with successive lowering of the super-
ficial gas velocity but with injection of fine limestone
at a constant Ca/S ratio did not show a significant im-
provement in sulfur control despite the decrease in
velocity.
i) Effect of fly ash recirculation
Recirculation of fly ash with fine sorbent injection
improved the sulfur control in some instances but the
results were inconsistent.
1.2.3 Reduction with the Use of Limestone Beds
The tests conducted with a medium sulfur coal burning ir a
bed composed of 1359 limestone indicated that the emission
of sulfur dioxide could be controlled almost completely for
a period of 2 to 3 hours with the favorable sorption condi-
tion, i.e., 1550°F temperature and 3% 02 in the flue gas.
When the breakthrough of sulfur dioxide becomes significant,
most of the sulfur may be driven out of the bed by increas-
ing the bed temperature and lowering the oxygen concentration.
The bed thus "regenerated" could be reused for sulfur control
7
by reverting to operation under the sorption conditions.
During the regeneration phase, sulfur dioxide concentra-
tions as high as 8.1% were observed, a value some 30
times the untreated gas concentration. A cyclic process
for carrying out the sorption and desorption on a contin-
uous basis was devised but remains undeveloped.
The bed remained active after two cycles of sorption and
regeneration. Additional work is indicated to establish
the reactivity over a number of cycles and for a number
of stones. Bed attrition rates were found to be high
during calcination (5% to 7% of initial calcium charge
lost per hour) but lower during sorption and regenera-
tion (3% and 4% per hour respectively).
Measurement of the overall heat transfer coefficient in
the limestone bed indicated the same value (47 Btu/ft2hr°F)
observed in the sintered ash bed.
1.3 SULFUR TRIOXIDE EMISSION
Average values of sulfur trioxide concentrations observed
in the flue gas from the process were found to be 30 to 50
ppm in a field of 3800 ppm sulfur dioxide. The sulfur tri-
oxide invariably disappeared when a sorbent material was
injected. None was observed with the limestone bed tests.
1.4 HYDROCARBONS EMISSION
The fluidized-bed combustor can be operated with as little
as 5% excess air without evolution of smoke, but hydro-
carbons concentration in the flue gas may be as much as
1500 ppm (methane) at this excess air level. The test data
show that hydrocarbons emission is sharply dependent on
oxygen content in the flue gas determined by the excess air
rate. An excess air rate of 17% was necessary to burn up
hydrocarbons in the FBC, while 24% was required for the FBM.
These values correspond respectively to 3% and 4% oxygen in
the flue gas.
The heat loss incurred by increasing the excess air from
5% to 17% is approximately 0.8% of the input energy based
on a flue gas exit temperature of 400°F. The heat recovered
from complete combustion of the hydrocarbons is about 0.9%
of the input energy. These results indicate that operation
with less than 17% excess air would not be advantageous in
terms of thermal efficiency, whereas operation at 17% excess
air has the obvious advantage cf lower hydrocarbons emission.
The 17% excess air rate was considered minimum for the bed
operation. During a few tests with reducing conditions in
the bed, sufficient air was added overbed to complete hydro-
carbons combustion and to result in 3% oxygen in the flue gas.
A measurable concentration of carbon monoxide does not ap-
pear in the flue gas at a 3% or higher oxygen concentration.
POPE EVANS -A.NJD ROBBINS
-------
8
5 OXIDES OF NITROGEN EMISSION
Emission of nitric oxide from the FBC was found to
increase with oxygen content in the flue gas. Values
of 320 ppm at 1% oxygen increased to 440 ppm at 5%
oxygen. A -typical value of 380 ppm at 3% oxygen_
corresponds to ^0.30 pounds NO per million Btu (MBtu)
input.
The nitric oxide concentrations do not correlate with bed
temperature. This would be expected since the measured
values are well above those predicted by thermodynamic
equilibrium. This result suggests the presence of local
temperatures higher than the measured bed temperature or
that the nitrogen content in the coal plays a role. In-
frequently the emission may rise to 0.38 lb per MBtu
with no increase in oxygen concentration.
Comparison of infrared determinations for NO and wet
tests for N0X indicate that nitric oxide (NO) is the
dominant oxide of nitrogen. Oxides of nitrogen other
than NO, determined by difference, were found to vary
in the range of 10 to 30 ppm.
Emission of nitric oxide from the FBM was observed to
be less than the level found with the FBC at the same
flue gas oxygen content (3%) and temperature. The
average emission of NO from the last sixteen FBM tests
was 0.22 pounds per MBtu, a value equivalent to
¦v275 ppm concentration. In general, nitric oxide emis-
sion was not affected by addition of sulfur control
sorbents.
.6 PARTICULATE EMISSION
Particulates passing the FBC cyclone collector represented
about 10% of the fly ash input without fine sorbent addi-
tion. When the fine sorbent was added, the particulate
emission rate was increased (from -*-2.0 to M.O lb/hr), but
the percentage of the total input that was emitted remained
the same. Most of the fine sorbent was collected in the
cyclone.
Somewhat higher collection efficiencies (95%) were
found with the FBM collector. One sample of the fly
ash discharged to atmosphere curing fine sorbent injec-
tion was analyzed for particle size. The analysis
showed that 90% of the particulate emitted was smaller
than 5 microns.
9
2. CONCLUSIONS
The results obtained from the test program thus far and
the economic study led to the following conclusions:
a. Emission of sulfur dioxide from the combustion of
coal can be reduced to cur-ently acceptable levels
by burning the coal in a fluidized bed and injecting
finely divided limestone into the bed. A 4.5%
sulfur coal can be converted to an equivalent 1.0%
sulfur coal with the injection of -325 mesh 1359 raw
limestone at a rate of 27 lb/100 pounds of coal,
equivalent to a stoichiometric ratio of 1.9. A 2.6%
sulfur coal can be converted to the 1% equivalent
with addition of 10 lb/100 pounds of coal, equiva-
lent to a stoichiometric ratio of 1.2.
b. Limestone injection equipment involves a compara-
tively low capital investment, approximately
$220,000 for a 500,000 lb/hr boiler plant contain-
ing two 250,000 lb/hr boilers.
c. The cost of reducing sulfur dioxide emission to the
equivalent 1% sulfur coal is estimated to be
$.54 per ton of coal for the 2.6% sulfur coal and
$1.06 per ton for the 4.5% sulfur coal with the use
of -325 mesh 1359 raw limestone at the rates indicated
above where limestone is available at $2.05 per ton.
These are incremental costs based on the assumption
that the plant is built with air pollution control
in mind. Improvement in costs will depend largely
on an economical method of increasing the sorbent
utilization. Possibility for improvement exists in
the use of limestone beds in a cyclic process or in
the processing of partially reacted stone to expose
the unreacted core.
d. For a once through process, grinding to a fine parti-
cle size (-325 mesh) appears necessary for the 1359
limestone which is very durable in comparison with
the dolomite. Fine grinding should be beneficial
with other limestones, but perhaps not necessary if
the stone tends to decrepitate in the bed. The 1359
lime hydrate, which occurs naturally in a fine size,
is as reactive as the finely ground raw stone but at
$3 5.00 to $20.00 per ton ir much more costly.
e. Utilization of the finely ground raw limestone for
sulfur control varies in the range of 40% - 33% at
stoichiometric ratios of 1 to 2. Slightly higher
sorbent utilization is indicated for the 1337
POPE EVANS AND ROBB1NS
-------
10
dolomite if utilization is based on the calcium frac-
tion alone. On a total weight basis, however, the dolo-
mite removes less S02 per pound of stone fed than the
limestone. Utilization appears to be limited by product
shell formation even with the fine sorbent particles.
f. When a medium sulfur coal is burned in a bed made up
entirely of -10 + 20 mesh 1159 limestone, 99% of the
sulfur dioxide is captured initially. The sulfur diox-
ide emission rises with time; after 2 to 3 hours, the
capture rate may drop to 90%. The emission would be
expected to rise steadily in time until the capture rate
becomes negligible.
To maintain a high capture rate, the stone must be
either replaced or regenerated. By raising the bed
temperature and decreasing the oxygen concentration,
90% or more of the sulfur may be driven out of the
spent stone and the stone thus regenerated. Makeup of
the bed to replace attrition losses was indicated to
be ^5% per hour of operation.
During the regeneration phase, sulfur dioxide concen-
trations as high as 8.1% were observed--some 30 times
the untreated gas concentration. The high concentra-
tion should facilitate sulfur recovery or scrubbing,
if this is desired.
g. Sulfur trioxide emission is completely eliminated by
limestone injection or by combustion of the coal in a
limestone bed. This would permit coal-fired boilers
to be designed with lower flue gas temperatures than
is normally permitted when low temperature corrosion
is a problem.
h. Emission of oxides of nitrogen from the FBM was found
to average 0.22 pounds per MBtu input at 17% excess
air. Values reported for conventional coal-fired boil-
ers of similar capacity vary from .31 to 2.2 lb/MBtu.
NOx emission is not affected by limestone injection
and is higher than predicted by thermodynamic equilib-
rium at the measured bed conditions. _Emissions from
the FBC were somewhat higher (.30 lb/MBtu). NOx emis-
sion increases with increasing excess air and may be
decreased by operating with reducing conditions in the
bed.
i. Emission of hydrocarbons from the fluidized-bed combus-
tion process can be controlled effectivley with ^24%
excess air based on FBM test results. This rate is
favorable in comparison to values of 40% to 50% excess
air commonly employed in conventional boilers. Carbon
monoxide was not detected in the flue gas with 24%
excess air.
POPE, EVANS AND ROBBINS
11
RECOMMENDATIONS
For continued research to improve the air pollution
control capability of the fluidized-bed combustion
process, the following measures are recommended:
a. Improvement in sulfur emission control without
added cost necessarily implies increased sorbent
utilization. The present utilization limit of
40% reflects the theoretical potential for improve-
ment. Increasing sorbent utilization would seem
to require a method of gaining access to the cal-
cined core of the sorbent or a repeated use of the
product shell area of the sorbent particle in a
cyclic sorption-regeneration operation.
The spent sorbent particle can probably be broken
down by hydration because of the heat generated in
the process and the naturally fine state of the
product hydrate. This breakdown was observed dur-
ing the test program when spent sorbent particles
were dropped into water or exposed to humid air.
The application to the fluidizec. -bed boiler would
involve wetting the spent sorbent fly-ash mixture
with a minimum amount of water at a point down-
stream from the dust collector and then reinject-
ing the mixture into the bed. Another technique
which might give access to the core is grinding of
the spent sorbent before reinjection.
b. The investigation of combustion in a limestone bed
should be continued as a means of increasing the
effective sorbent utilization for possible applica-
tion in industrial or utility-size boilers. Opti-
mum concentration of sulfur dioxide in the off-gas
during the sorbent regeneration phase should be
determined for its bearing on sulfur recovery.
c. While emissions of NO may be somewhat lower than
from conventional boilers, they are still present.
Therefore, methods for reducing NO emissions should
be sought.
The possibility of finding an inexpensive sorbent
which acts as effectively on NO as limestone does
on sulfur oxides appears remote. Unlike sulfates
and sulfides, most nitrates, nitrites and nitrides
are not stable at the bed-operating temperature.
The systems study by Esso Research provides a
valuable checklist of methods that might find appli-
cation either as an in-situ control process or
-------
12
on the stack gas. These include catalytic decom-
position, catalytic reduction, adsorption, absorp-
tion and modification of operating conditions.
Recommended for evaluation are lowering the oxygen
gradient between the bottom and the top of the bed
by recirculation of flue gas, operation of the bed
under slightly reducing conditions, and reduction
of the oxygen partial pressure at the base of the
bed by combustion of a premixed hydrocarbon fuel,
such as natural gas. Reduction of NO emissions of
50% have been obtained under certain operating con-
ditions indicating a potential for NO control via
fluidized-bed combustion.
POPH. FVAr.'S
bobbins
13
INTRODUCTION
Pope, Evans and Robbins in 1965 undertook a program
sponsored by the Office of Coal Research, United States
Department of the Interior, to develop low cost, high
capacity, coal-fired boilers. In comparison with oil
and r-as-fired units, the conv ntional coal-fired boiler
suffered a competitive disadvantage in higher capital
cost for any given steam capacity. The primary aim of
the program was to improve the economic position of coal
as a boiler fuel. A report on the boiler development
program will be published by the Office of Coal Research.
The problem of increasing steam capacity while reducing
the capital cost (furnace size) necessitated an increase
in combustion intensity, i.e., heat release per unit
volume and also an increase in heat transfer rate to
reduce heat transfer surface requirements at the high
volumetric heat release rates. These requirements
demanded a new approach in coal combustion technology.
The concept of fluidized-bed combustion provided the
most promising area of investigation as the basis for
this new approach.
Early test results indicated that fluidized-bed combus-
tion afforded order-of-magnitude increases in both com-
bustion intensity and heat transfer rates. From these
results it was predicted that a coal-fired, railroad
transportable, multicell boiler capable of producing
250,000 pounds of steam per hour was feasible. Develop-
ment and testing of a full-scale, single cell of a
multicell boiler concept has been in progress since
1967.
The fluidized-bed combustion principle is illustrated
in Figure 1. Crushed coal is injected into a bed of
granular, inert material which is fluidized by air
flowing upward through the .bed. The coal particles
are dispersed rapidly in the bed because of its turbu-
lent motion and burned with oxygen supplied by the
fluidizing air. Most of the ash residue accompanied by
a fraction of carbon is blown out of the bed and
entrained in the gas stream.
Rapid oxidation of the coal particles gives rise to com-
bustion intensities (heat relecse rates) as high as
350,000 Btu per hour per cubic foot of bed volume.
The high heat transfer rate permits rapid removal of
heat through the walls surrounding the bed. This, in
turn, permits control of bed temperature to a compara-
tively low 1600°F despite the rapid heat release.
F^OPE. EVANS AMD ROBBINS
-------
14
FLUE GAS TO
FIGURE 1. SCHK?1^TIC 07 FLUIDIZED- 3ED BOILER
POPE EVAN S AMI3 RCSeiNS
15
Measured overall heat transfer coefficients of
45 Btu/ft2hr°F result in an average heat flux of
'WO,000 Btu/ft^hr. This flux is such that a sizable
fraction of the heat released in the bed can be con-
verted to steam energy with a relatively small amount
of heat transfer surface surrounding the bed. The
total heat transfer surface in the boiler can thus be
markedly reduced.
The fluidized-bed combustion principle, therefore,
makes it possible to decrease the capital cost of coal-
fired boilers by increasing the steam capacity per unit
volume of furnace. Other advantages include the fact
that the coal need not be cleaned. Poor quality fuels,
having high ash fractions, can be burned in a fluidized-
bed combustor. In addition, the coal need not be pulver-
ized but merely crushed. These factors would reduce both
coal and coal preparation costs.
The low bed operating temperature (1600°F) should reduce
boiler tube corrosion and fireside ash deposition. The
uniform temperature distribution throughout the bed
should reduce the possibility of tube distortion from
local, high thermal stresses. The low -xcess air require-
ment showed promise of increased thermal efficiency over
conventional coal-fired boilers.
Principal disadvantages are the higher fan power re-
quired because of the pressure drop across the bed and
air distributors, and piping and control requirements
that possibly may be more complex than those employed
With conventional boilers. Operation with some coals
would require makeup of bed material. The present
state-of-the-art requires that the coal be single-
screened to preclude buildup of large inert particles
in the bed.
It appears impractical to attempt to burn coal com-
pletely in one pass through a fluidized-bed combustor.
Special methods must therefore be employed to insure
high levels of combustion efficiency. One such method,
the Carbon-Burnup Cell, is now under intensive study.
The potential disadvantage of high dust loadings and
subsequent erosion from ash recirculation may also be
overcome through the use of the Carbon-Burnup Cell.
The combustion principle and the performance character-
istics pointed up a number of potential advantages for
air pollution control. Among these was the fact that
the random motion of the bed particles could provide
an ideal environment for contacting limestone with
-------
16
sulfur oxides in the flue gas. Sulfur emission control
by injection of limestone into boiler flue passes has
been demonstrated by others, but effectiveness of the
method was limited apparently by formation of a sulfate
shell around the injected particles which prevented fur-
ther reaction. The possibility existed that the fluid-
ized bed could provide not only the gas-solids contact-
ing and the residence time for the desulfurization re-
action but could erode a product shell and continuously
expose unreacted surface. Bench scale studies of this
reaction by others also indicated that the bed operating
temperature range would be favorable for sulfur capture.
The low bed operating temperature was felt to be a char-
acteristic favorable for the control of nitrogen oxides
emission. Thermodynamic considerations and experience
with other combustion processes indicated that nitrogen
oxides emission increases with rise in flame temperature.
Operating at a temperature of 1600°F, the fluidized-bed
boiler was felt to have a clear advantage over conven-
tional boiler systems which burn coal at temperatures
of 2500°F and higher.
A further potential advantage, noted earlier, is the
fact that the coal could be burned at near stoichio-
metric air rates without visible smoke in the flue gas
discharged to atmosphere. This meant that smoke emis-
sion could be eliminated without loss in thermal effi-
ciency which necessarily follows the use of excess air
for smoke control.
In November 1967, the National Air Pollution Control
Administration (NAPCA) of the United States Department
of Health, Education, and Welfare initiated an air pollu-
tion test program through an interagency transfer of
funds from NAPCA to OCR. The test program was undertaken
to characterize the pollutant emissions from the combus-
tion of coal in a fluidized bed and to assess the poten-
tial of fluidized-bed combustion for air pollution con-
trol .
The test program entailed initially the investigation
of the operating variable effects on emissions and the
effect of injecting sorbent materials (limestone and
dolomite) into the fluidized bed of inert material.
Subsequently the investigation .fas expanded to include
the use of sorbent material as the bed material.
The variables are itemized in the discussion below.
POPE EV.Ais.-S AND ROBBINS
17
4.1 OPERATING VARIABLES
a. Bed temperature
1500°to 1900°F
b. Bed depth, static
6 to 20 inches
c. Bed composition
sintered ash and limestone
d. Air rate
superficial velocity 6 to 14 fps
e. Fuel rate
required to match superficial
velocity 6 to 14 fps
f. Ash recirculation
full range (0% to 80%)
4.2 ADDITIVE (SORBENT) VARIABLES
a. Sorbent type
limestone, dolomite and a
natural mine additive
b. Sorbent state
raw, calcined and hydrated
c. Sorbent feed rate
stoichiometric ratio 1 to 3
d. Sorbent particle size
-7 +14 to -325 mesh
e. Method of sorbent injection
f. Water (or ^team) injection
4.3 COAL COMPOSITION
a. Ash content
7.2 and 10.7 wt. percent
b. Sulfur content
4.5, 3.0, and 2.6 wt. percent
-------
18
5. EQUIPMENT AND PROCEDURES
5.1 PILOT SCALE COMBUSTOR, FBC
Initial tests were conducted in a pilot scale combustor,
designated the FBC. The FBC consisted of a rectangular
combustion space, 12" x 16", enclosed by an air distri-
bution grid at the bottom, and waterwalls around the
periphery as shown in Figures 2 and 3. Air was passed
into a plenum below the grid, through the grid buttons
and into the combustion chamber where it fluidized the
bed material and provided the combustion oxygen. Coal,
crushed to pass through a 1/4" screen, was injected
through a port at the base of the bed.
The air distribution grid contained a matrix of grid
buttons mounted in a mild steel plate. The buttons were
fabricated in stainless steel and designed to direct the
air slightly downward toward the grid plate. This down-
ward flow tended to eliminate stagnant areas around the
buttons and provided cooling air for the grid plate. A
cross section of a button is shown in Figure 4.
The bed material consisted generally of sintered coal ash
crushed and screened to a mesh size of -7 +14. The bed
was heated to coal ignition temperatures with a premix gas
burner flame directed downward onto the bed as shown in
Figure 2. The ignition procedure involved fluidizing the
bed material with minimum air flow, raising the bed tempera-
ture to 800°F and then injecting coal until the combustion
was self sustaining. About 10 minutes is required for
ignition. The bed temperature was monitored with a
number of thermocouples spaced vertically in the combus-
tor. Kaowool seals were provided to prevent flue gas
leakage out of the system. Specifications for the FBC
are presented in Appendix A, Enclosure 1. The coal feed
rate to the unit was approximately 110 lb/hr for an
energy input of 1.35 x 106 Btu/hr.
The FBC test system is shown in a photograph, Figure 5,
and schematically in Figure 6. Combustion products from
the FBC were passed through a heavy gauge welded seam
duct, through an induced draft fan, through a dust col-
lector and on to atmosphere. The slanted configuration
of the duct between the FBC and the induced draft (I.D.)
fan was intended to provide gas cooling without causing
wall surface temperatures to fall below the dew point
of sulfur trioxide ^360°F. This was accomplished since
the flue gas temperature declines in this region from
•vl300°F to ^800°F. The control damper provides a vari-
able back pressure on the system to create a slightly
19
FIGURE 2. FLuIDISED-oED COLUMN (?BC) CONSTRUCTION
DETAIL -- FROZIT VlEv'
POPE EVANS AND ROBBINS
-------
20
FIGUrtS 3. FLUIDIZED-EED COLUMN (FEC) CONSTRUCTION
DETAIL - SIDE VIE1./
21
FIGURE 4. AIR DISTRIBUTION GRID EUTTON
POPE E\^NS A-InID ROSSIVS
-------
FIGURE 6. FBC AIR AND EXHAUST GAS DUCTING SHOWING SAMPLING POINTS
FIGURE 5. FBC TEST SYSTEM
-------
24
positive pressure on the FBC and thus prevent infiltra-
tion of air at the hood connection.
Combustion air was provided from an external fan to
reduce the noise level in the test area. The air flow
rate was monitored by a pitot tube in the long entrance
duct E.id a gate valve in the line provided air flow
control to the unit. The coal feed rate was controlled
by a variable speed drive on the coal feed screw. Fly
ash collected was discharged into bags or recirculated
into the FBC as indicated in Figure 6. Locations of
thermocouples arc described in Section 5.4-Instrumenta-
tion.
The temperature of the bed during operation of the FBC
(or FBM) depends for the most part on the bed depth
vhioh governs the total transfer surface at the water
walls. This dependency created a problem in determin-
ing the separate effects of temperature and depth. The
problem was solved by insulating the periphery of the
bed and installing an internal cooling coil as shown iii
Figure 7. The bed temperature was then adjusted at
various depths by raising or lowering the coil. This
mode of temperature control was also used in the lime-
stone bed tests. The temperature vs. depth tests are
discussed in Section 6.3 and thr limestone tests in
Section C . 9.
5.2 FULL- SCALE BOILER MODULE, t'BM
The full-scale boiler module, designated the FBM, is a
boiler unit capable of generating steam under pressure.
In this unit the fluidized bed is contained in a rectan-
gular enclosure in which each wall is a row of vertical
boiler tubes seal-welded so as to form a gas-tight enclo-
sure. The FBM represents one half cell of the multicell,
fluidized-bed packaged boiler concept developed under
the OCR project. Two modules placed back to back would
comprisp one cell, A number of cells placed side by side
without intervening insulation would make up the full-
scale boiler.
A cutaway sketch of the FBM is provided in Figure 8.
The fluidized-bed cross section is ^18 x 72 inches,
roughly seven times the FBC crors section. The bed is
currounded by vertical water tubes which extend from the
grid plate to the overhead drum. No other tubes are
plared in the bed. The water tubes are joined together
bv a steel webbing and are backed by insulation. i'luo
gas from the bed passes between the tubes at llie tup of
the unit and around the steam drum.
25
Preceding page blank
FIGURE 7. FBC COOLING COIL AND INSULATING SLEEVE
VIEWED FROM ABOVE
POPE. EVANS AND BOBBINS NOT REPRODUCIBLE
-------
SNiaaOH C3NV SNVAH :3dOc2
27
The combustion space is accessible through a water-
cooled panel at the front of the unit. The panel
contains a premix gas burner used to fire the bed.
The burner directs a flame downward onto the front
of the bed. Two pneumatic feed ports are provided
below the access panel, one for the coal feed tube
and tve other for the additive or fly-ash feed tube.
The tubes are extended into the bed area to discharge
the solids at points shown in Figure 9.
From a plenum at the base of the unit, air is directed
upward through a grid and into the bed area. The grid
consists of a mild steel plate containing buttons of
the same spacing and design used in the FBC operation.
The bed material used in the FBM tests was the same
-7 +14 mesh sintered ash. The static bed depth varied
from 12 to 20 inches. Thermocouples were mounted
throughout the bed as shown in Figure 8. Detailed
specifications of the FBM are presented in Appendix A,
Enclosure 2.
In operation, the bed is raised to the ignition point
of coal by use of the gas burner. Combustion of the
coal begins in the vicinity of the turner flame and
propagates rapidly throughout the bed. Firing with
a coal input of 800 lb/hr, the FBM produces 200 psig
steam at the rate of 5000 lb/hr. The energy not
absorbed by the waterwalls leaves this test rig as
hot products of combustion. In a commercial unit,
the energy of these gases would be extracted in a
conventional gas-to-surface convection bank.
A schematic drawing of the FBM test system is shown in
Figure 10. Air from an external forced-draft fan
passes through the air preheater (or bypass) and into
the FBM plenum. Coal feed is controlled by the rota-
tion of a star feeder which drops the coal into a
pneumatic feed tube at the injection port. A supply of
coal is maintained automatically in a small hopper
above the feeder by screw feed from a larger hopper.
Sorbent materials were screw fed to the injection port
at a rate controlled by a variable speed screw drive.
Ash recirculation is accomplished by pneumatic trans-
port of fly ash from the dust collector through a star
feeder control.
Flue gas from the FBM is mixed with ambient air in the
ducting above the unit to reduce temperature before it
enters the air preheater. As the flue gas passes
through the air preheater, a portion of the fly ash
-------
saaru aaaj aAixiaav onv
ivod oiivwnatM do iHawaotwHHV shj. dnimohs waa am ao diivkshds • 6 annoi-i
waisxs xs3i wsj shi jo oiivwaHDS "oi annou
xovit
-------
Si-r.SBOH GNV 5NVA3 adOd
(S*£T + ZL) (UZT * ,,91)
"9 iZ = ("^rfTTixyTs^Trx'TzTj'uZi = ?p
' Zp qoxpaad oq uoxqEnba aqq buxAxddv
• (s>{UBq aqnq uaaMqaq aouBqsxp aqq) S'iT = Zn ZL = ZX
'Z paq 'WSJ SIR UI -„ZT = TP 'ajnqEJtaduiaq paq .JoOSSt
qB puB „ZI = Tm ' „9T = rT }Bq} os I paq aq D3.J aq} qarr
•waj PUB oa.1 aqq oq IX3M Ax^xej saxxdds dxqsuoxqBxaJ sxqi,
"Z pus x spaq aqq 30 AxaAxq
-osdssi qqpxM pus qqBuax 'qqdap aqq. die « puB x 'P a.xaq«
(Z« + ?x)t"IT rP
('A + lx)3«ZI = ZP
:uoxqBxaa
axduixs aqq swoxxoj uoi;D3S ssoio paq ut afiuBqo q^XM
quauiaaxnbaa qqdap paq aqq suoT^ipuoa quaxBAxribe aapun
•£ a:ms
-oxoua 'v xxpuaddv ux paquasajd sx dxqsuoxqBxai aiR «oqs
oq uoxqEnba aouexEq qBaq aqq JO uoxqBAXaap v "4UEq.s
-uoo qda5( sx oxqsj sxqq jx suoxsuamxp quaxajjxp BuxAEq
suiaqsAs ux auiBs aqq AxaqsuixxoaddB aq XTTM aanqBaaduiaq
paq aqq a^EJ xanj puB a;Ei jxe uaAf& b qqxM 'BaiE UBxd
paq oq. aOBjans Buxxooo 30 OTq.Ejr auiEs aqq uxBquxeui oq
A-iBssaoau sb« qqdap jaqsajiB E asnsoaq sqsaq ogj aqq ux
UBqq sqsaq wa.1 aqq ux aaq-Bajfi sb« qqdap paq 6ux5(io« aqj,
•afiuEqo uoxqxpuoo Buxq-sxado
UB JiaqjB UOTqxpUOO aqEqS-ApB3qS M9U 8 JOJ pa«OXX® se«
'qssax qB 'saqnuxui o£ jo poxjiad v "paAaasqo uoxssxuia uo
qoajja aqq pus paxxBA uoxqxpuoo BuxqBaado xaqqo auios 30
paq.Bxq.xux sbm paaj quaqxos aqq aqsqs ApBaqs q.v • sbB
anxj aqq ux apxxoxp jnjxns Pue uaBAxo 30 sanx^A qusqs
-uoo paqBOxpux xazAxBUE HI Z0S aiR PUB xaq.aH ^aXTe3 3fq
uaq« pauinsss sbh aqBqs ApBaqg "paxxBAa^d suox^xpuoo
aqBqs-ApEaqs XT^un aariqEaaduiaq paq paxxsap aqq qs uoxq.
-snqraoo aqq 6uxzxxxqBqs puE paq aq^ Bux^xuBx peAxOAUX
suoxqBxado W3J pus oaj a^q- H"4°q ^03 saxnpaooid qsa^ aqi
*XX ajnBxj se paquasajd sx uiaqsAs aqq
30 qdBxfcoqoqd v -xaqiuBqo uoxqsnqaioo aq-q ux aanssajd
Xoaq.uoo oq. Buxqonp aqq ux papxAOid sx a a dure p eousxsq
V •aaaqdsourqB oq uaqq pus uej qjB^p paonpux ue oq
qonp 6uox E qfinoaqq sttox? sb6 aqq aoqoaxx00 wojj
-^addoq Joqoaxx°o aqq oqux paj rta^os sx aaddoq qsa^J
aqq ux qsB aqq 'uoxqBxnoaxoax Buxjrta 'uiBaxqsuKop joq
-oaxx°° auooxqxnui b ux pauxEqax sx qsB Ax J aqq ?o 'HXni:?
aqj, -uMoqs aaddoq aqq ux paqoaxx03 sx pus qno sdojp
0€
-------
32
In the FBM experiments it was found that a 24" deep bed
did operate at 1550°F firing the same coal at the same
rate, per unit bed plan area, as in the comparable FBC
experiment. It is not suggested that the relation would
apply to scale-up of very large beds.
Emissions were monitored in the FBM tests with the same
procedures used in the FBC tests. Emissions were moni-
tored without sorbent addition, with coarse sorbent
addition, and with fine sorbent addition. Most of the
tests were performed using fine, sorbent addition, low
bed temperature and a 3% oxygen concentration in the
flue gas, which are the conditions found in the pilot
scale to favor sulfur dioxide control. The sorbents
used included both the 1337 dolomite and 1359 limestone
ground to a -325 mesh particle size and the hydrated
forms of these which occur naturally in a -325 mesh
size. Precalcined sorbents were not tested in the FBM
because of poor performance in the FBC tests. Ohio #8
Pittsburgh Seam coal, washed and unwashed, was used in
the tests except for one test involving a low sulfur
E. Kentucky coal.
Recirculation of fly ash was employed as a test condi-
tion by feeding the fly ash from the collector to the
sorbent injection port as shown in Figure 8. The rate
determined by the feeder was 80% - 90% of the input ash.
In two tests, steam was injected into the inlet air at
approximately 400 lb/hr.
.3 METHOD OF SORBENT FEED
Three methods of sorbent feed were employed on the FBC
during the course of the test program. The first in-
volved screw-feeding the sorbent into the pneumatic
line used to carry the coal feed into the unit. This
method, pictured in Figures 12 and 13, employs a long
inclined screw feeder and a variable speed drive. The
assembly was designated the #1 feeder system.
A second system, designated the #2 feeder, was fabri-
cated for injecting the sorbent at two points remote
from the coal-feed injection port. The feeder system
consisted of a lock hopper for the sorbent mounted on
a short screw feeder as shown in Figures 14 and 15.
The outlet of the feeder was connected to a pneumatic
feed system which divided the sorbent flow between two
injection tubes. The orientation of injection ports is
shown in Appendix A, Enclosure 4.
The lock hopper in this system was necessary to counter
the static pressure at the bottom of the bed. In the
#1 feeder system and the coal feed system, this pressure
differential is borne effectively by the inclined screw.
POPE. EVANS AND ROBBINS
33
INCLINED
FIGURE 12. SCHEMATIC OF THE NO. 1 SORBENT FEED SYSTEM FOR THE FBC
POPE. EVANS AND ROBBINS
-------
3iaionaoMd3H ion
a
G
S
o
•o
g
a
55
o
6
§
W
8
CO
K
cn
£3
iue|q aSed Suipaoajj
Preceding page blank
FIGURE
14. FBC OPERATING WITH NO. 2 SORBENT FEED SYSTEM
-------
36
COLUMN (FBC)
FIGURE 15. SCHEMATIC OF THE NO. 2 SORBENT FEED SYSTEM FOR THE
POPE, EVANS AND ROBBINS
37
A third method of sorbent feed involved premixing the
sorbent with the coal and feeding the mixture through
the coal feed screw. A brief series of tests was con-
ducted to determine the most effective system for sulfur
dioxide control.
A combined feeder system was developed in an effort to
study the effect of distribution based on the favorable
results of the #2 feeder. The system consisted of a
four-point injection configuration, shown in Figure 16,
with provision for controlling the sorbent flow into one
two or all four sides of the FBC without change in sor-
bent mass flow.
An attempt to achieve the ultimate in sorbent distribu-
tion was made by injecting the fine material into the
inlet air duct for distribution through the grid buttons
Although the sorbent particle size is much smaller than
the button port diameter, the sorbent agglomerated and
plugged the buttons rapidly.
In the FBM test series only one method of sorbent feed
was used—that of feeding the sorbent into the bed at
two points opposite the two coal feed ports. This feed
arrangement is shown in the schematic of Figure 9 and
in the photograph of Figure 17.
5.4 INSTRUMENTATION
Emissions of sulfur dioxide, nitric oxide and hydro-
carbons were monitored continuously with the instrumen-
tation pictured in Figure 18. Infrared analyzers
CBeckman 215) were used to monitor sulfur dioxide and
nitric oxide. Hydrocarbons were detected with a flame
ionization analyzer (Beckman 109A) using methane as the
reference gas. The signal output of each of these units
was displayed on strip chart recorders shown at the
right side of Figure 18.
The gas transfer system used with these analyzers is
sketched in Figure 19. The system permitted recheck-
ing of calibrations on any of the three units at any
time during the test by switching from sample gas to
reference and zero gases at the rotameter valves. The
sample gas was drawn from the flue gas stream through a
sintered stainless steel filter and conditioned to re-
move water. The sample gas was again filtered before
entry into the analyzers to prevent possible contamina-
tion of the optical cells and the hydrogen burner.
POPE. EVANS AND ROBBINS
-------
SNiaaoH qnv SNVAa 'adOd
SJ.Kicd aaaa asv snsaaaj
.iKaaHOS snimohs ohj m Rouoas ssoho TifDiitiSA -91 booou
8£
-------
SYMBOLS:
r~~\
( I CALIBRATION, REFERENCE AND
V, / COMBUSTION GAS SOURCES
® C03TR0L VALVE
FIGURE 15. SCHEMATIC OF GAS TRANSFER SYSTEM FDR CONTINUOUS MONITORING OF SULFUR DIOXIDE,
UTRIC OXIDE AND HYDROCARBONS
-------
42
The FBC gas sample was drawn into the instrument room
from the FBC exhaust duct which extended overhead. In
sampling the FBM flue gas, special precautions were
necessary because of infiltration of dilution air in the
duct above the unit and the poor instrument response
which would result from drawing a small sample a long
distance (>60 feet) from unit to instrument room. A
system was devised to draw a large gas sample from with-
in the FBM (at the gas passage around the steam drum),
pass it through a dust collector, and then through a
loop above the instrument room. The sample tube was a
3" pipe with sections screw-fitted and welded. The
system was driven with an I.D. fan located at the dis-
charge to atmosphere. A schematic drawing of the system
is shown in Figure 20.
Periodic samples were taken from the flue gas to deter-
mine SO3, S02, and NOx by wet chemical analysis. The
sulfur oxides analytical system consisted of a hydrogen
peroxide absorption train preceded by a sulfur trioxide
condenser shown in Appendix A, Enclosure 5. The sulfur-
ic acid in each part of the system was determined by
titration with barium perchlorate using thorin as the
indicator. The nitrogen oxides analytical system con-
sisted of the standard phenoldisulfonic acid procedure*
using a Beckman Model B spectrophotometer for optical
density measurement.
Particulate emissions were monitored with an isokinetic
probe system shown in Appendix A, Enclosure 6. The
probe design permits equalization of internal and exter-
nal static pressures to match the sampling velocity
with the stream velocity. Locations of sampling points
in the FBC and FBM test systems were indicated in
Figures 6 and 10 respectively.
A Bailey oxygen analyzer (Type OC1530A) was used as an
operating device to indicate the oxygen concentration in
the flue gas. During a test period, the air input rate
was held constant and the coal rate adjusted to main-
tain the oxygen concentration at the desired value. The
Bailey instrument was calibrated periodically with O2,
N2 and C02 mixtures and found to be very reliable. The
flue gas oxygen was also verified using the standard
Orsat technique which determined also carbon dioxide and
carbon monoxide.
Temperatures in the bed and at various other points in
the system were recorded on a Honeywell Multipoint
recorder. A multiple switch panel was used to connect
POPE EVANS AND ROBBTNS
SNiaaoa cjnv snws aaod
-------
44
the recorder input to either the FBC or FBM systems as
required. Locations of thermocouples in the systems
are indicated in Appendix A, Enclosures 7 and 8.
The infrared analyzers and the hydrocarbon analyzers
were calibrated with gas mixtures supplied by vendors.
The concentration of the active components in the cali-
bration gases was checked after delivery to the labora-
tory. i The methane mixture was analyzed by the National
Bureau of Standards—a report is shown in Appendix A,
Enclosure 9. This gas, containing 1265 ppm CHi,, was
used to calibrate a second methane mixture before it
was depleted.
The sulfur dioxide calibration gas was analyzed with a
peroxide absorption train. Gas concentrations of 3906
ppm and 25 30 ppm were used in the program. Analysis of
the first calibration gas supply indicated a value of
2530 ppm as shown in Enclosure 10. Analysis of the
nitric oxide calibration gas is shown in Enclosure 11.
The output signal of the infrared sulfur dioxide ana-
lyzer varies in a nonlinear manner with SO2 concentra-
tion. The calibration curve provided with the instru-
ment was checked by precision dilution of the known
calibration gas. The curve was found to be correct
except for a slight deviation at the low end of the
range. The calibration curve and check points are
shown in Appendix A, Enclosure 12. The calibration
curve was used without correction since the deviation
is not more than 1% of full scale.
The calibration curve for the nitric oxide I.R. analyzer
is shown in Appendix A, Enclosure 13. The contribution
of water vapor to the signal output is significant with
this analyzer." The water vapor correction determined by
the supplier (180 ppm) was checked by testing a dry gas
in the analyzer for comparison with a moist sample. A
correction of 200 ppm was noted and incorporated in the
data reduction. The range of this unit is 0-1000 ppm NO.
5.5 MATERIALS
5.5.1 Coals¦ Two coals selected for the test program con-
slated of an unwashed high sulfur coal containing 4.5%
sulfur and 10.7% ash, and the same coal after washing.
The rfashed product contained 2.6% sulfur and 7.2% ash.
The coal was mined from the #8 Pittsburgh seam at the
Georgetown mine, Cadiz, Ohio. Proximate and ultimate
analysis of each coal is shown in Appendix A, Enclo-
sures 14 and 15. A comparatively high content of iron
oxide in the ash is reported. One other coal, an East
Kentucky, Pike County, low sulfur and low nitrogen coal,
POPE EVANS AND HOBB1NS
45
was used in one test to compare the effect of
nitrogen content on nitric oxide formation.
5.5.2 Sorbent Materials. Two limestone-based SO2 control
additives were studied ill the raw, calcined and
hydrate forms with a range of particle sizes from
-7 +14 to —325 mesh. These additives consisted of a
dolomite containing about 53% calcium carbonate and
46% magnesium carbonate (designated 1337) and a lime-
stone containing 97% calcium carbonate (designated
1359). Analyses of these are given in Appendix A,
Enclosure 16. The dolomite was supplied by the
Dolite Company, Gibsonburg, Ohio, and the limestone by
the M.J. Grove Company, Frederick, Maryland.
5.5.3 Bed Material. For the most part, the starting bed
material consisted of sintered coal ash ground to a
-7 +14 mesh. The sintered ash was procured from a
local deposit and from the operation of the FBM in
previous work. On occasion, the ash was obtained from
the Anacostia power plant located nearby. Attempts to
fluidize heavier bed materials such as limestone
pointed up the need for special consideration.
The particle size range was selected to facilitate
fluidization during the light-off procedure and also
to preclude the possibility of serious elutriation
losses at the operating bed temperature and super-
ficial velocity. The flue gas velocity-particle size
range for various material densities is shown in
Appendix A, Enclosure 17. The particle density of
the sintered ash is ^120 lb/cu ft and normal operating
superficial velocity 12-14 ft/sec.
The FBC was operated successfully with a bed of the
high calcium limestone (1359). Details of this effort
are discussed in Section 6.9.
POPE EVANS AND ROBBINS
-------
A-l
ENCLOSURE 1. FBC SPECIFICATIONS
1. Air Supply
Two centrifugal fans in series for 300 cfm at 30" w.g.
connected to a smooth 4" diameter conduit 20' long.
Air flow is controlled with a ^ate valve and monitored
with pitot pressure, static pressure and temperature
measurements.
2. Plenum
Mild Steel, 1/4" thickness, 21" x 18" x 12" outside
dimensions with 8" diameter air inlet.
3. Water Column
Mild Steel, 1/4" thickness, 24" x 20" x 36" outside
dimensions with 16" x 12" x 36" inside dimensions.
A. Wall on inlet air side contains:
a) One nominal 3" diameter pipe for lightoff
burner
b) One nominal 1" diameter instrument port.
B. Left wall (facing air inlet) contains:
a) One nominal 2" diameter pipe with valve for
removal of bed material.
b) Eight nominal 1" diameter instrument ports at
various levels.
c) One nominal 1" diameter water outlet.
d) One nominal 2" diameter pressure relief port.
C. Right wall (facing air inlet) contains:
a) ,One rectangular 2" x 1" coal feed port.
b) One nominal 3/4" diameter cooling water inlet.
D. Wall opposite the air inlet contains:
Three nominal 1-1/2" diameter ports.
POPE EVANS AND BOBBINS
A-2
ENCLOSURE 1. (Continued)
4. Air Distribution Grid
The grid contains 130 stainless steel air distribution
buttons spaced on 1-1/4" centers each containing eight
drilled ports, .087" diameter. The air is discharged
downward at an angle of 15° to the horizontal.
5. Water-cooled Hood
The hood is a truncated pyramid 24" x 20" at ths
bottom and 17" x 17" at the top with a height of
24" and a flue opening 12" diameter. Material is
110 gauge mild steel. One 4" diameter observation
port is provided with 1" diameter water ports and a
2" diameter pressure relief port.
6. Flue System
From the FBC-1 hood, the flue system is run in 12"
diameter #10 gauge steel pipe to the induced draft
fan. From the fan the pipe is continued at 6"
diameter again #10 gauge steel. All connecting
sections are welded.
7. Dust Collector
The collector contains two 8" diameter centrifugal
collector units with a dust hopper, rotary feeder
and a valve for fly ash removal.
POPE EVANS AND ROBBINS
-------
ENCLOSURE 2. FBM SPECIFICATIONS
Air Supply
One centrifugal fan at 2500 cfin at 50" w.g. connected to
12-inch square duct which expandc to ful? width of
plenum at inlet. Air is controlled by means of a damper
and monitored by an orifice.
Plenum
Mild steel, V thickness, 72" x 20V x 12" inside
dimensions with a 6' x 1' air inlet.
Boiler Construction
a. Single 20" steam drum
b. Dual 6" lower headers
c. 23s" risers on 4" centers for side walls
d. 4" downcomers (external)
e. 5'4" distance from grid to uninsulated bottom of steam drum
f. Combustion space = 53 ft3
g. Projected heating surface = 80 ft2
h. Average direct contact surface = 30 ft2
i. Boiler capacity = 5000 lbs/hr excluding convection
heat transfer; 7000 lbs/hr including convection heat
transfer
j. 8.75 ft2 of bed area
k. Heat release rate: 800,000 to 1,200,000 Btu/'ft2hr
1. Pressure rating: 300 psi design, 200 psi normal operating
POPE. EVANS AND ROBBINS
A-4
ENCLOSURE 2. (Continued)
4. Air Distribution Grid
The grid contains 815 stainless steel air distribution
buttons spaced on centers, each containing eight
drilled ports, .087" diameter. The air is discharged
downward at an angle of 15" to the horizontal.
5. The flue system is fitted with three air infiltrators
for temperature quenching, and a two-pass, 104 tube
(1" x 6'), 600° air preheater; this is followed by a dust
collector, which exits to a 16" duct. The system is
drawn by a 4000 cfm, 5" w.g. static pressure, induced
draft fan.
6. Dust Collector and Fly Ash Reinjection
The dust collector contains twelve 10-inch diameter
centrifugal collector units with a dust hopper, a
4" Allen-Sherman-Hoff rotary feeder for fly-ash reinjection
and a valve for fly-ash removal.
7. Coal Input
700 - 900 lbs per hour
POPE. EVANS AND ROBBINS
-------
A-5
ENCLOSURE 3. DERIVATION OF THE BED DEPTH RELATIONSHIP
FROM BALANCE OF HEAT AND MASS
Heat balance on the system may be expressed as follows:
(1) K,G0 AH = GoCpm(TB - Tj + hAg(TB - T,)
Heat release = flue gas loss + heat removed.
where Go - mass flow of coal and air through the
system lbs/hr
AH = heat content of fuel BTU/lb
K = constant
Cpm = mean heat capacity of the flue gas
Tg. = bed temperature °F
To = reference temperature °F
h = radiant plus convective heat transfer
coefficient BTU/hr ft2 °F
= effective bed cooling surface ft2
Tw ¦= cooling surface wall temperature °F
Equation (1) may be restated as follows:
(2) K2UACPAH = uAcpCpmCTB - To) + hAg(TB - Tw)
where u = superficial velocity ft/sec
Ac = bed cross section ft2
p = gas density lbs/ft3
by dividing (2) by the 1st term and rearranging
(3) Ws(TB " Tw>_ , CPm(TB " To>
K2uA pAH L K2AH
C I
The first terra is the fraction of the total heat which is
removed from the bed and is seen to be constant for fuel
type and bed temperature.
POPE, EVANS AND ROBBINS
A-6
ENCLOSURE 3. (Continued)
If u, h and TB are to be constant for two systems of
different size, then the ratio of cooling surface to
bed crosj-sectional area must be constant •
The effect on bed depth is seen from the area ratio:
,As d(2 1+ 2w) . .
(4) = constant
c
for beds of varying dimension
(5)
di liWi(12 + W2)
37 ~ 12W2 (li + Wj)
where d, 1, and w are the respective depth, length and
width of the two systems
This analysis assumes that the effective bed cooling surface
is proportional to the bed depth. This is not strictly true,
since radiation losses in a vertical direction from the bed
are independent of bed depth. Another source of error is
that use of the linear dimensions, 1 and w, does not account
for the additional heat transfer surface of the round tubes
which actually make up the walls of the fbm. At minimum bed
temperature the respective bed depths are 10-12 inches and
20-24 inches.
POPE EVANS AND ROBBINS
-------
SMIHBOH QNV SMWVJ3
£-V
A-S
WATER INLET
140°F TO 190°
FRIT-MEDIUM POROSITY
S02 TO PEROXIDE A3SOP.BERS
AND VOLUME :1EASUR3!1ENT
ENCLOSURE 5. SCHEMATIC OF SULFUR TRIOXIDE CONDENSER
POPE EVANS AND BOBBINS
-------
A-9
^ ^ ^ ^ ^ V ^ ^ ^ V V ^ ^ ^ V ^ \\ \\ ¦
• DUCT
-f-
§2Z&^
0
n
a co
o
si
O H
H
o o
> ^
iSi >
s z
»d
O M
PO
i-3
s >
W 3
2 a
8?
o a
G M
TJ
se
w
tr*
O O
o a
> o
n n
M H
O 25
Z O
W
OT-V
-------
w
z
o
c-1
o
W
G
73
W
J* VI
z o
o a
s n
o >
o »
c 55
~a >
F z
W O
w
si
> »-3
w O
O *1
z
(A »-3
a:
M
*¦3
w
?
93
O
s:
H
Z
O
o
>
in
z
o
o
z
H
<5)
-------
A-13
ENCLOSURE 10. SULFUR DIOXIDE CALIBRATION GAS ANALYSIS
The gas was passed slowly thru two absorption columns
each containing a solution of 1.5% H202 freshly prepared.
Sample volume was determined by change in pressure in a tank
of known volume and corrected for standard condition. The
solutions were then boiled to remove peroxide and titrated
with 1/10 N sodiun. hydroxide. Four tests were rndde with the
following results:
Test
No.
Tank
Volume*
liters
Pressure,
mmHg
Initial Final
A P,
mmHg
Gas
Temp.
°F
Sample Volume
, corrected to
liters
1
34.71
113 315
202
70
9.24
2
34.71
315 535
220
70
10.05
3
34.71
535 725
190
70
8.69
4
34.71
130 345
215
70
9.85
Titration of the solutions with 0.0985N sodium hydroxide
yielded the following data and computed results:
Test
No.
NaOH Solution
Scrubber #1
Volume, ml.
Scrubber #2
Blank
S02 Concentration
ppm
1
19.10
0.70
0.0
2530
2
20.70
0.76
0.0
2530
3
17.50
0.65
0.0
2520
4
20.70
0.80
0.0
2540
Avg 2530
The sodium hydroxide solution was standardized against a
potassium acid phthalate solution.
Concentrations were computed from the relation:
S02 ppm ¦= 12.05 (V)N
where V = titer volume ml corrected for blank
N = titer normality (Equivalents/liter)
Vg = sample volume at STP-liters
POPE EVANS AND ROBBINS
A-14
ENCLOSURE 11. NITRIC OXIDE CALIBRATION GAS ANALYSIS
Analysis of the nitric oxide calibration gas was made with
the Phenol-disulfonic acid procedure. Four gas samples were
taken in flasks containing a small quantity of H202 solution
and allowed to stand overnight. The solutions were processed
as required and the absorbances of the final solution read on
a Becknan Model B spectrophotmet-er. The concentrations were
determined from a calibration curve prepared by similar treat-
ment of KNO3.
The following data were taken during the test.-:
Sample
No.
Flask
Volume,
ml
Pressure,
mmHg
Initial Final
AP,
mmHg
Temp.,
°F
Volume
at STP
, (70 °F)
ml
1
1970
32 763
731
70
1900
2
1972
31 759
728
72
1885
3
1969
35 759
724
75
1860
4
1972
33 758
725
75
1870
Sample Absorbance Equivalent Concentration
No. % mg NO? ppm NO
1 25.0
2 25.0
3 24.7
4 24 .7
2.83 780
2.83 786
2.77 780
2.77 776
Avg 780.5
The concentration of NO was determined from the relation:
ppm N02 (or NO) = (5.24^x 105)(C)
s
where C = concentration of NO2 mg
Vg = gas sample volume 70°F and 760 mmHg
POP£ HVAMS Al-JD ROBBINS
-------
A-15
BECKMAN INSTRUMENTS, INC.
MODEL NO. 215 A
APPLICATION: SULFUR DIOXIDE, RANGE: 0-5000 PPM BY VOLUME
AMPLIFIER NO.: 200065, DETECTOR NO.: 1243 A
ZERO GAS: NITROGEN, CALIBRATION PitESSURE: ATMOSPHERIC
SPAN GAS: ANALYZED CYLINDER ^2650 PPM S02
CALIBRATION PRESSURE: ATMOSPHERIC
SAMPLE PRESSURE: ATMOSPHERIC
X
CALIBRATI0
N CHECK
IOOO 2000 3000 4000
PPM S02 IN N2BY VOLUME
5000
ENCLOSURE 12. CALIBRATION CURVE FOR SULFUR DIOXIDE
INFRARED ANALYZER
POPE, EVANS AND ROBBINS
A-16
BECKMAN INSTRUMENTS, INC.
MODEL NO. 215 A
APPLICATION: NITRIC OXIDE, RANGE: 0-1000 PPM BY VOLUME
AMPLIFIER NO.: 200066, DETECTOR NO. 1436 A
ZERO GAS: NITROGEN, CALIBRATION PRESSURE: ATMOSPHERIC
SPAN GAS: ANALYZED CYLINDER -v9 00 PPM NO
CALIBRATION PRESSURE: ATMOSPHERIC
SAMPLE PRESSURE: ATMOSPHERIC
ENCLOSURE 13. CALIBRATION CURVE FOR NITRIC OXIDE
INFRARED ANALYZER
POPE. EVANS AND ROBBINS
-------
A-17
ENCLOSURE 14. ANALYSES OF "PERFECT EIGHT" UNWASHED
4.5% SULFUR COAL
Source: Pittsburgh #8 Seam, Georgetown Mine
Cadiz. (Harrison City) Ohio
% Weight
ULTIMATE ANALYSIS
As Rec1d
Dry
Moisture
6.01
Carbon
66.21
70.45
Hydrogen
4.57
4.86
Nitrogen
2.50
2.66
Chlorine
0.05
0.05
Sulfur
4.45
4.73
Ash
10.73
11.42
Oxygen (diff)
5.48
5.83
100.00
100 .00
PROXIMATE ANALYSIS
As Rec'd
Dry
% Moisture
6.01
% Ash
10.73
11.42
% Volatile
36.49
38.82
% Fixed Carbon
46.77
49.76
100.00
100.00
BTU
12157
12934
% Sulfur
4.45
4.73
SULFUR FORMS
% Pyritic Sulfur
2.92
% Sulfate Sulfur
0.08
% Organic Sulfur
1.73
% Total Sulfur
4.73
FUSION
REDUCING ATMOSPHERE
Initial Def. (ID)
1980°F
Softening (H==w)
2125 °F
Softening (H==1/2W)
2160°F
Fluid Temp. (FT)
2270°F
ASH ANALYSIS
Silica (SiO,)
43.64%
Iron Oxide tFe-O,)
25.68%
Titania (TiO,)
xxxxx
Alumina (A1_0,)
25.02%
Manganese Oxiae (Mn,0.)
xxxxx
Lim? (CaO) J 4
2.06%
Magnesia (MgO)
trace
Alkalies (Na^O / K^O by diff.)
2.36%
Sulfur Trioxide (SO.)
1.24%
Phosphorous pentoxiae (P20,.)
xxxxx
100.00%
POPE, EVANS AND ROBBINS
A-18
ENCLOSURE 15. ANALYSES OF "PERFECT EIGHT" WASHED
2.6% SULFUR COAL
Source: Pittsburgh #8 Seam, Georgetown Mine
Cadiz. (Harrison City) Ohio
Size Consist
As Received Dry IV' Modified at Mines
Moisture 5.00%
Volatile 37.30
Fixed Csrbcr 50.50
Ash 7.20
100.00%
BTU 13,000
Sulfur 2.60%
39.30%
53.10
7 .60
100.00%
IV
1*"
3/4'
3/5'
1/8'
x iy
x 3/4"
x 3/B"
Y 1/8"
x 0
lb .75%
44 .85
21.89
11.19
3.32
13,680
2.80%
ASH ANALYSIS
ULTIMATE ANALYSIS
Silica (SiO-
(Fe203)
(Ti02)
(ai2o3)
(Mn304)
Iron Oxide
Titania
Alumina
Manganese Oxide
Lime (CaO)
Magnesia (MgO)
Alkalies (Na^)
Sulfur Trioxide (SO^)
Phosphorous pentoxide
/ K2° feff
• )
(p2o5)
43.u 4%
25.68
xxxxxx
25.02
xxxxxx
2.06
trace
2.36
1.24
xxxxxx
100.00%
As Rec'd.
Moisture 5.00%
Carbon
Hydrogen
Nitrogen
Oxygen
Sulfur
Ash
70.36
5.17
0.39
8.73
2.55
7 .20
Dry
74 .06%
5.44
1.03
9.19
2.66
7.62
100.00% 100.00%
FUSION TEMPERATURE OF ASH
Reducing Oxidizing
Atmosphere Atmosphere
Initial Deformation 2,020°F 2,365°F
Fusion (Softening) 2,120°F 2,440°F
Fluid Temperature 2,240°F 2,530°F
Hardgrove Index 58-61
Free Swelling Index 4-1/2
rCPE HC5B1NS
-------
A-19
ENCLOSURE 16 . ANALYSIS OF SORBENTS AFTER IGNITION
DOLOMITE (1337) LIMESTONE (1359)
CONSTITUENT * By Wt. % By Wt.
CaO
Si02
55 97
43 1.2
MgO
Fe2Oi 0.33 0.22
0.92 1.07
A1203 0.15 0.29
LOSS ON
CALCINATION 47.4 43.6
Analysis provided by the National Air Pollution Control
Administration
POPE. E-VAMS AND ROfeBINS
A-20
100.
10.0 —
I.OO
0.10
ENCLOSURE 17.
PARTICLE DIAMETER
TERMINAL VELOCITY AND MINIMUM FLUIDIZATION VELOCITY
VS PARTICLE DIAMETER
POPE: EA/AMS AlsJD ROBBINS
-------
A-21
ENCLOSURE 18. ESTIMATION OF ELUTRIATING PARTICLE SIZE
FOR THE 1359 LIMESTONE
The smallest particle size that would be retained in
the bed was estimated from the intermediate law which
is applicable for the test conditions, i.e., Reynolds
number between 2 and 500. The particle size follows
the relation1
ft""
0875
Where:
Dp is the particle diameter in inches
the superficial gas velocity ft/sec taken as 14.
JX the gas viscosity, lb/ft sec taken as 2.9 x 10-5
^ the gas density, lb/ft3 taken as .020 at 1600°F
^ the gravitational constant, 32.2 ft/sec2
^ the particle density, lb/ft3 taken as 162.0
Accordingly:
6Z.7
(32.zV Uez - o.02)
C.7IM
« 0.011 in.
For this particle size the Reynolds number is:
u . Dpuet . (,0022/u>i(t'0(o.&a = 179
* M 2.9" lo"s
which value falls in the applicability range of the law
Adapted from Leva, Max: "Fluidization," McGraw Hill
Book Company, Inc., New York, 1959
POPE. EVANS AND BOBBINS
A-2 2
-ENCLOSURE 19. NITRIC OXIDE EQUILIBRIUM CONCENTRATIONS
FOR THE FLUIDIZED-BED ENVIRONMENT
The equilibrium constant, K, defined as
is related to the free energy change, AG, and is given
directly in the JANAF tables
T, °K
T, °F
Log,o K
K
1100
1520
-3.633
0.00023
1200
1700
-3.275
.00053
1300
1880
-2.972
.00107
1400
2060
-2.712
.00194
1500
2240
-2.487
.00319
With air at 1500°K, (2240°F) for example,
PPM (NO) = 106 P[jo = 106(pN2 • PoJ0"5K =
106 (0.2 x 0.8)0 " 5 (.0032) = 1276 ppm
Similarly, T, "F Equi. NO ppm
1500 92
1700 222
1880 429
2060 775
If is reduced to 0.05 corresponding to a
possible FBM condition, is reduced by a
, , ,o. f
factor I.OS/.2.\ or to !ialf the value shown abov
-------
A-2 3
ENCLOSURE 20. PARTICLE SIZE DISTRIBUTION OF 1359 LIMESTONE
BED BEFORE AND AFTER FLUIDIZED-BED COMBUSTION
POPE. EVANS AND ROBBING
A-24
ENCLOSURE 21. INTEGRATED SULFUR BALANCE FOR SORPTION-DESORPTION
OF THE LIMESTONE BED DURING FBC TEST 114
For the absorption phase the total sulfur absorbed by the bed is
S = Input sulfur - fly ash loss - emission
B
°r SB = /GcScdt " 'SFGFdt ' k/MCS02Gcdt
where: S„ = total sulfur in the hed lbs
' " ' B
G = coal rate, lb/hr
c
S = sulfur content in coal,lb/lb
c
t = time, hours
SF = sulfur content in fly ash,lb/lb
Gp = fly-ash rate,lb/hr
dry mole flue gas .. n„_6
k = constant = lb coal x 10
^S02 = concentration of S02 in flue gas, ppm
_M = molecular wt. of sulfur = 32
The sulfur retained in the bed during the 4.28 hour
absorption period was computed as follows:
(A) Sulfur input = = 63.0 x ^qq x 4.28 = 8.35 lbs
(B) Fly-ash loss = ^/GpS^dt = 16 x x 4.28 = 1.23
(C) Emission loss = •fcs02dt = -3^6 (32) (63) (3200) _
° 106
2.11 lbs
POPE EVANS AND ROBB1NS
-------
A-25
ENCLOSURE 21. (Continued)
The integral ''^sc^dt rePresents area urdcr the curve
during the absorption period.
The sulfur retained in the bed is
Sfl = A - (B + C)
(D) = 8.35 -(1.23 + 2.11) = 5.01 lbs
5 01
% retained in bed = jpyg- = 60.2%
During the regeneration phase (t = .65 hours), the sulfur
loss from the bed is:
Recovered sulfur = emission -(input - fly ash)
S_ = kMG
K C
,t t t
•'Ccn ¦ /G S dt + /G
o S02dt 0 c c o
FSFdt
(E) Emission sulfur = '326—<32) *63) (7950) = 5>25 xbs
10 6
The value 7950 ppm hours was determined from the area
under the concentration curve (Figure 2) during regenera-
tion.
POPE EVANS AND ROBBING
A-26
ENCLOSURE 21. (Continued)
•l rt q
(F) input = 63 (j^j£.)(.65) = 1.27 lbs
(G) Fly ash = 16 (x^t)(-65) = .21 lbs
IH) Sulfur recovered isE-F+G=5.25-1.27 +21=
4.19 lbs
(I) Sulfur retained in bed-after regeneration
= bed mass x Sc =49 (^g-) = -49 lbs
(J) Total of H and I =4.68 lbs
Percent of sulfur recovered from bed
A 1 Q
= , x 100 = 89.8%
4 . D o
Sulfur unaccounted for = A + F-(B + C+ E+ G + I)
= 8.35 + 1.27 - [1.23 + .21 + 5.25 + 2.11 + .49]
= .33 lbs
% unaccounted = 3^2 = 3-5%
POPE . EVAN S -A. Ki D ROBB1NS
-------
,0.9
in a:
S t-J
? CL
v
in 3
z 1
0 oi
ir>
33 r-
0 u'
a
ffi
z
m
0.8
5 0.7
x
o
y o,6
a:
0.5
a
2
<
X
in
0.3
0.2
0 I
£ 1700
J±l u. IS00
,3 ° 1500
in
ID
LU
q
X
o
a:
D
WET TESTS
NO^ = 475 ppm
SO^ = 3800 ppm
START ADDITION OF
& 1610 ppm
1
1
SO 3
0
ppm
SO 3
5
ppm
15 5
ppm ppm
39
ppm
SO3
0
ppm
"\
- 2
EXTENDED
STATE SUL
TEST WITHOUT SORBENT
FUR TRIOXIDE LEVEL
ADDIr
riON
TO DE
TEHMI
NE ST
EADY
_. .
— nt
n
v2 _
TEST PERIOD , HOURS
ENCLOSURE 22.
EMISSIONS DURING FBC TEST 63
BURNING A 4.5% SULFUR COAL
FOR SULFUR TRIOXIDE EMISSION
-------
ENCLOSURE 24. EMISSIONS DURING FBC TEST 102 BURNING A
MEDIUM SULFUR COAL IN A 1359 LIMESTONE BED
ENCLOSURE 23.
TEST PERIOD, HOURS
EMISSIONS DURING FBC TEST 101 BURNING A
MEDIUM SULFUR COAL IN A 1359 LIMESTONE BED
-------
ENCLOSURE 26. EMISSIONS DURING FBC TEST 105 BURNING A
MEDIUM SULFUR COAL IN A 1359 LIMESTONE BED
z
en
>
'i
0
3
0
tt
ffi
z
ffi
ENCLOSURE 25.
2 3
TEST PERIOD , HOURS
EMISSIONS DURING FBC TEST 104 BURNING A
MEDIUM SULFUR COAL IN A 1359 LIMESTONE BED
-------
ENCLOSURE 28. EMISSIONS DURING FBC TEST 107 BURNING A
MEDIUM SULFUR COAL IN A 1359 LIMESTONE BED
r
O °-8
o 07
X
o
2 06
"05
o
<
o 04
§03
CD
a
<
g 0.2
a.
0.1
111
!— '
O
CO
UJ
O
X
o
o
o:
WET TESTS
et
q N0x = 390 ppra
~ S0X = 250 ppm
S03 = 0 PPm
BED WEIGHTS, LBS
INITIAL 75 (CaC03)
ADDED 25 (CaC03)
FINAL 58
COAL SULFUR = 2.95%
EQUIVALENT S02
FOR SULFUR
— 0
h' - ¦ 0
(300
1700
Z 3
1600
¦ U 2
500
.[
2 3
TEST PERIOD, HOURS
ENCLOSURE 27. EMISSIONS DURING FBC TEST 106 BURNING A
MEDIUM SULFUR COAL IN A 1359 LIMESTONE BED
-------
WET TESTS
O NOx =410 ppm
~ SO}
so, =
900 ppm
0 ppm
BED WEIGHTS, LBS
INITIAL 80 (CaC03)
ADDED 21 (CaC03)
FINAL 54
COAL SULFUR
2.95%
EQUIV
AXiENT
S02
INPUT
NO
»9
ADD
a LBS 1359—^
"7
-"add
V- —
*13 LBS 13!
<
l r-i_
so2
1±
HC
Oo
" \
TEMP
~
TEST PERIOD , HOURS
ENCLOSURE 30.
EMISSIONS DURING FBC TEST 109 BURNING A
MEDIUM SULFUR COAL IN A 1359 LIMESTONE BED
,0.9
o 08
2
Ul
O
X
o
07
y o.6'
q:
h
2 0 5
o
2
<
o 0.4-
X
• CO
§03
CD
a:
<
g 0.2
tr
Q
I 1000
£ U. 1700
Q° 1600
W
m 15001
2 3
TEST PERIOD, HOURS
ENCLOSURE 29. EMISSIONS DURING FBC TEST 108 BURNING A
MEDIUM SULFUR COAL IN A 1359 LIMESTONE BED
-------
5-'
2
ffl
>
2
0
»
0
til
ffl
2
V
ENCLOSURE 32.
TEST PERIOD , HOURS
EMISSIONS DURING FBC TEST III BURNING A
MEDIUM COAL IN A 1359 LIMESTONE BED
,09
08-
0.7 f
o 0.6
0.5
0.41
h- u.
WET TESTS
I 0 3-
0
02
z
1800
1700
1600
Q NO^ = 4 50 ppm
0 SO = 1100 ppm
SO, =
ppm
BED WEIGHTS. LBS
INITIAL 80 (CaC03)
ADDED 39 (CaC03)
FINAL 63
COAL SULFUR = 3.02%
-EQUIVALENT S02 INPUT
¦Of
3- =
2-
3
-NO
l~\
-ADD, 15 LBS 1359
I
iADD-24 LBS 1359|
SO
HC
¦Oi-
TEMP.
TEST PERIOD , HOURS
ENCLOSURE 31. EMISSIONS DURING FBC TEST 110 BURNING A
MEDIUM SULFUR COAL IN A 1359 LIMESTONE BED
-------
0.9
9
— 0
—0
1600
o 5
BOO,
iJ 4'
1400
° 3
1300
2
T! 6
WET TESTS
o
~
NO
X
SOx
so,
0 PPM
EQUIVALENT
S02 FOR
SULFUR IN
COAL FEED —
BED WEIGHTS, LBS
INITIAL H (CaC03)
ADDED 62 (CaC03)
FINAL (EST.) 60
COAL SULFUR = 3.05%
3 -4
TEST PERIOD, HOURS
^ (
rm
t
y
NO
\
s
O
|S02
\
\
*
-
HC
\
57~0
\
* \
*
----
TEMP.
" *
i
0/
1
ENCLOSURE 34.
EMISSIONS DURING FBC TEST 113 BURNING A
MEDIUM SULFUR COAL IN A 1359 LIMESTONE BED
ENCLOSURE 33. EMISSIONS DURING FBC TEST 112 BURNING A
MEDIUM SULFUR COAL IN A 1359 LIMESTONE BED
-------
.0 9
08
0.7
o 06
-0 5'
o
<
- 04-
tfi
r- 0 3
o
CO
a:
5 0.2
o i
ce
Q
£ o it
-7-. 6
='0
1=0
1700
o 4
ISOO
J 3
1500
° ?
1400
1
WET TESTS
BED WEIGHTS, LBS
o
~
NO
260 PPM
S0X = 175
SO, = 0 PPM
INITIAL
ADDED
FINAL !
~~Ti (CaC03)
67 (CaC03)
64
EQUI' 'ALENT S02
FOR SULFUR
IN COAL FEED
COAL SULFUR = 2.95%
NO HYDROCARBONS DATA
AIR
NO
SO
V-
RATL AND COAL RATE SET FOR SUPERFICIAL VELOCITY OF:
_a
¦12 FPS
8 FPS
DECREASE FROM
12 FPS TO 9 FPS
^ TEMP r
I 2 3 4 5 6 7
TEST PERtOO, HOURS
ENCLOSURE 36. EMISSIONS DURING FBC TEST 115 BURNING A MEDIUM SULFUR
COAL IN A 1359 LIMESTONE BED WITH CHANGE IN GAS VELOCITY
ENCLOSURE 35. EMISSIONS DURING FBC TEST 114 BURNING A MEDIUM SULFUr.
COAL IN A 13 59 LIMESTONE BED WITH REGENERATION
-------
ENCLOSURE 38. EMISSIONS DURING FBC TEST 117 BURNING A MEDIUM SULFUR COAL
IN A 1359 LIMESTONE BED WITH REGENERATION
ENCLOSURE 37. EMISSIONS DURING FBC TEST 116 BURNING A MEDIUM SULFUR
COAL IN A 1359 LIMESTONE BED WITH REGENERATION
-------
ENCLOSURE 40. EMISSIONS DURING FBC TEST 119 BURNING A MEDIUM
SULFUR COAL IN A LIMESTONE BED KITH BED REDUCING
.^CLOSURE 39. EMISSIONS DURING FBC TEST 118 BURNING A MEDIUM SULFUR COAL
IN A 1359 LIMESTONE BED WITH CHANGE IN FLUE GAS 02 CONTENT
-------
.0.9
O °-8
o 0.7 4
/
WET TESTS
o
~
NO.
-EQUIVALENT S02 INPUT
START SORbENT
FEED ABOVE
THE BED
1.75 Ca/S RATIO
412 PPM
x - 1040 PPM
s°3 = 0 PPM
SO
CHANGE SORBENT
FEED TO BASE OF
THE BED AT
1 7 5 Ca/S RATIO
COAL SULFUR
3.0%
ENCLOSURE
2 3
TEST PERIOD , HOURS
42. EMISSIONS DURING FBC TEST 75 BURNING A MEDIUM SULFUR COAL WITH INJECTION OF
-325 MESH 1359R LIMESTONE ABOVE AND AT BASE OF BED AT CONSTANT Ca/S RATIO
ENCLOSURE 41. EMISSIONS DURING FBC TEST 120 BURNING A MEDIUM SULFUR
COAL IN A 1359 LIMESTONE BED WITH ASH RECIRCULATION
-------
ENCLOSURE 44. EMISSIONS DURING PBC TEST 77 BURNING A MEDIUM SULFUR COAL' WITH INJECTION
01 -325 MESH 1359R LIMESTONE WITH CHANGE IN SUPERFICIAL VELOCITY
ENCLOSURE 43. EMISSIONS DURING FBC TEST 76 BURNING A MEDIUM SULFUR COAL WITH ADDITION
OF -325 MESH I359R LIMESTONE WITH CHANGE IN SUPERFICIAL VELOCITY
-------
SN1SBOH CIN v Sl'-iVAH 3dOo
rd
2S
n
r
o
a
M
0
3
O
o
o
cn
M
a
a
M
O
z
>
?3
2
Ci
w
s
w
2
»-3
o
25
O
>
»
H
M
O
G
>-3
M
W
Z
w
w
o
z
n
o
o
o
o
w
•-3
0 © 0
o
s s
s *
o v n
— a u>
TS-
V
APPENDIX B
TEST DATA FBC AND FBM
F>OPE EVANS AND ROBB1NS
-------
TABLE B-l rm TEST CONDITIONS AHO DATA
Injcetic
lb/hr
ci- Carbon ' R«c11
Couaty
II Sulfur
h mii
17 S
16 2
1) I
16 S
} I 0.0
E %y to
Ohio IS S«U
unwa«h©».
1640
1600
1500
7 400
7400
7400
3400
1550
IOC*
unvaahad
« S» Sulfur
10 11 Aah
Cas a&^ile syiu
urn** a had
4.St Sulfur
10.7* Ajh
7600
7(00
7600
7600
1.14
1.60
2 30
14 2
14 5
14 5
1 J 0 0
Ohio II Saaa
unwashed
4 SI Sulfur
10.71 Aah
JI00
1000
2400
2900
J 2 0.0
3 4 0.0
» 0 0.0
11.6 44 0
22.4 416
20.0 4] 4
Loat Iijnitlo© - olaeardad
74M
7400
7J00
7408
1.11
!>
3»00
2100
2150
2(00
1J.I
14 •
14 2
3.0 0.0
2.2 O.O
>1 2
45 0
3f .0
15.4
26.5 42 1
15.0 41 I
'Natural aitw Uaattoa
7 2% CaCO>
Ittu *dd«d to «lr Lol*t
Ohio II Bmb
7t00
7100
3»00
2100
2100
1100
14 0
15 3
IS 4
J*.2
42.0
51.0
65.2
SO 0
46.7
42 0
Ohio IS Baas
vaatod
2.6% Sulfur
7 2% Aab
11*0
1100
1770
7100
7100
7100
7100
2100
2450
2SOO
2)00
16 S
15.1
15.2
14.«
SI.7
S4.0
41 2
• gta Oj TATLttLoi
Isjactlod tMt
Ohio IS S«a
vashad
2 (I Sal for
7800
7100
7100
2200
1650
1300
SS 0
21.4 47.0
27.5 43.5
* 450 lb/hr StMB to «ir
vMh«d
2.61 Suitor
7 2* Aah
1680
1CS0
1600
0 2250
1.17 1000
l.)7 850
*4S0 Lb/hr luu addod to Air L&lat
Ohio IS Ba«a
210 274
Ohio tl S«M
waabtd
2.6% tulfur
1600
1600
1600
1600
35.0 40 3
Ohio II IMI
unvaahad
4 S% Solfur
10.7% uh
1SS0
1SS0
1520
7400
7400
7400
Ohio IS Daaa
uaw*ah«d
4.5k o-ifn
10.71 A*h
1660
1660
isao
7400
7400
7400
3750
1350
1100
61 0
37 I 42.5
37 J 37.1
Ohio II !»¦
jQ
~n-
-------
b-;
7A9LB »"1 (Cgnt.inu«d)
Ifj^ C.I j I TriJ
2.6* Sulfur
7 2* Aah
Ohio IB B«an
J 6t bwifur
»io «B S«u
w»(b«d
2 4\
7 21 Mh
IkOO
16 70
LlMttoat idtntifaution and nus^siie? ayatea of
fclLuainona Coal >a»**reti, Zne
C • aalclsad by •urpLUf, a - hydxatad by tuppJier,
:&orb«nt Utilisation dofln«4 *•
7 400
7*oa
7400
7483
CPDS d«alqut«« Phanoldlaulfonic Acid Hotrod
eC«a analyila by I-fr«r»d AMlyiar
•flv aah not recirculated ualaaa Indicated by "V«
15 5 2 9 0 C
SI J
o*«
57 }
11 1
O.I
14 I 11 0.0
38 «
36 0
)~ 1
ft
5
-------
8-3
Cc«_ Tyt-o
And
"lue Gi«
0*yg-;n
-------
B-4
TA6L1 >-2 (Coptlnuad)
S*» Aftalyala i Concentration In volu— parcant of W aa ind i
Coal Type
aad
Ccmpoaltlos
n 1
Ttit
Cond
Coal
lata,
Lb/hr
a
cacbona.
oraat
Analyst*
ru
Aah Data
Taat
Bad
Ha tar Lai
Papth,
iacbaa
Tr:q> ,
(Uta,
{Coitrol),
and
Rate,
lo/hr
Aalio.
Ca/B
!£4
da
SO j ,
uo.
MO,.
t
(1 *
Catboo
Cocteat
Lb/hx
PP»
Oj
CO
*
4
Lies* Ife/fcs
JtaaarM
rec
1»
Ohio *• liu
uewaihad
4.51 eulfur
10 7« Ash
MO
•
IK
119
110
110
1540
1510
1410
1120
1120
1100
J100
3 9
3 0
3 0
3.0
IB.2
30 1
30.1
0
1.15
1.93
1.13
J900
3300
2150
2350
39*0
2250
37
(2
250
2S0
220
133
uo
125
as
1)
10
90
15 0
15.4
14.4
14 2
3.9
3.a
2.1
2.a
0.2
15 4
37
40
13.4
14 0
20.7
laa
roc
20
Ohio 41 ku
4.St Cultai
10.71 Alb
Mb
•
110
112
110
110
1740
1700
17 tO
1450
1**0
1070
1070
1040
1070
1070
3.0
3 0
3.0
3.0
) 0
1359ft
17.4
24 -
32 2
12 I
1.15
1.47
; 0
2.0
3100
3350
2750
2500
3*00
3110
31(0
3120
10
40
210
250
220
220
310
309
210
70
as
9v
70
14 •
15.4
15.1
;s a
11.0
2 4
3 2
3.1
3 0
11 a
24 4
34 2
39 5
10 2
It 1
17.1
19.7
S3 3
45 0
51.1
nc
21
Ohio IB ftaaa
uwMhad
• 5* 8u>t-»*
10 71 Mb
Mb
•
110
115
112
112
120
1>30
1520
1530
1500
1S2C
1100
1100
1130
1100
1100
J.O
3 0
3 0
3.0
2.0
'.359B
14.1
35.*
33.0
35.0
0
~
1.57
2.0
3 0
3100
3)00
3>20
J900
2I0:
37J0
3200
3960
0
0
150
70
10
0
0
225
240
210
200
2J0
00
40
40
•0
CO
13 7
15.2
15.*
3.1
2.4
3 3
13.2
33.1
33.4
31.3
14 7
14 1
12 a
15 t
<0 9
54.0
•V.3
Tm
rec
22
Ohio H 8*as
uwanhad
«.t! Snltox
10 71 Mb
Mb
7
2
lit
112
111
1650
1450
1540
1100
1100
1L00
3 0
J.O
1.0
S'5.
0
14.3
n s
1.0
1 *
3100
3)00
3726
35(0
310*
2510
120
54
31
320
320
279
113
114
1»
• 0
ao
30
13 1
15.2
15 *
2 9
2.3
3.0
IS 4
30.8
15.4
14.2
(0 1
59 7
45 4
nc
23
Ohio to Mia
untaabad
4.5% Suitor
10 7» Jkah
Mh
•
2
101
107
103
101
101
1*20
1(00
15*0
lilO
1510
loao
1010
1010
1040
1010
3.0
3.0
3.0
3.0
3.0
1337C
U.l
10
32.(
21
1 21
2 17
2 S4
2.17*
4110
SIM
J10C
3950
1720
3550
1910
120
137
330
300
240
2*0
2«0
M0
131
UJ
40
40
40
40
UO
15
15 2
14 «
15
14.2
2.8
2.8
3.3
3.0
31.1
33.4
11
59
24.8
14.4
20
37.2
33
33 0
25.7
27.0
r«i
Tn
T«»
r«a
Taa
'hut injactad lata rpaa
nc
14
Ohio II l«n
10.7« Mb
Mb
7
1
2
If
>7
¦7
101
101
16B0
1480
1440
1(10
1*40
1150
1120
u'.a
1150
1150
3.0
3.0
3.0
3.0
2 0
1337C
17.2
2t.2
34.0
1.13
I 73
2.2*
)tU
2910
1110
22S0
3510
2150
2240
0
0
0
410
320
210
2*0
2(0
(7*
U3
535
130
•0
40
20
ISO
19
14
\l *
14
14 «
2.a
3.4
3.8
2.a
3.5
11.3
31 3
35.4
31.4
14.4
K.O
15.7
42 4
33 9
11.4
Taa
Taa
'htit injaetad Lb to faa apat
>a. •orbant fa
ad data b
ntwilaUa
FBC
25
Ohio M San
SOMIhtd
4 s» s«irw
10 71 Mb
M
I
3
12»
120
12)
1750
1720
17M
1010
10U
1010
1.0
1.0
3.0
1337C
31.4
19.8
9
1-19
1.4
3450
3)50
3U0
3000
*150
100
70
50
200
101
49
40
14.1
15.1
2.8
2 ¦
a.a
12.9
• '.1
47 9
37 9
14.3
nc
Ohio II l*a
oovajtMd
4.5* tultu
10 7« Mb
Mb
I
2
4
12)
114
123
11»
1710
1700
1700
1115
1100
1100
1010
1120
3 0
3.0
3 0
3.0
1339C
13 7
23.4
32.2
0
1.5
1.3
3.4
3100
3)00
2150
2)70
370
3)9
214
40
40
«0
40
13 9
15
14 a
14.1
3 2
2.9
1
3.2
13.2
33.4
37 5
11 4
10.3
11.0
«3 7
40 3
39
34.4
nc
27
¦tartbarn
». Va coal
5k Solfur
Mb
•
1
2
3
1)0
125
121
1M0
1510
1550
10*0
10*0
1041
9.0
] 0
3.0
ML*
JS.3
10
0
1.3
1.4
4)00
3110
3790
41 SO
lis
210
310
210
30S
120
•0
90
12.(
19
15.5
3.8
2.9
2.4
a. a
>4
7.4
23 4
44 5
S3 5
47 9
•Mtoral aiaa llaaatoaa
711 CaCO,
17(0 1100 | « 0 0 49M 42W 230 ]M US }j| 14.J J.B 0 0 - 31.1
1T« UN 5.6 SC.* lS.t .t UM 1» 3M il.fc J.6 ft ».J 11.4 33.5
1710 UM ].» -7 *14 43.a 1.75 J»M JW UC U.l 3.8 0 11.7 T.I 21 2
1*49 110* 3.6 42.4 1.75 )M0 2lfr t0 is j.| 0 j| * 1».J SI ]
roc lojaotioa of Bydrata ifito Plaea - tottcaw Ploggad
3* to Dtut tkUiMS
nc mail ol 113* 9j4rat* with Coal - Coal F»«J«r Plogftl
SO HO
nc Frail of 1359 Bjdtata *1U Oo*l - Coal Flonad
31 wo Oat* obtained
Ohio It M
uafaabari
4.31 Sal for
10.71 Ask
240 2tS
lit#
1MI
mo
1540
)0M
.1050
10M
1050
13.4
13.4
25.9
MOO
2300
11 SO
1401
1U 370
31.0
44.0
41.2
37.5 30 0
42 31.t
21.4 M 0
k). 1 and Bo. 2 ratdti*
4.51 laUi
10.74 Mfe
{!/¦• ¦ 0)
Q
1010
10SO
ion
1010
1.2c
1.3*
2.55
3550
1700
1100
3440 11*
u7i i«2
13 5
14 t
14 5
52.2
SI 3
74.0
44.5
44.3 » 4
37 9 HI
31.4 57.4
n~
-------
B-S
TABLE B-2 ICoatlBUfd)
, Concentration In vo t uae urctnt or i
Co* J Type Bad Teat Coal 8ed Mr Oxygen Typo Injection Solo 1 R Wot Teata i » PCS Bydro- Or ml tion Utilita- Carbon Reelr-
and Bed Depth, Cond Rate, Tenp Rate, (Control), and Rale, futlo, iQ} ¦ 50], S0>, no, CO,. carbonj, '¦¦Itiii, > (1 * ), tlon,4 Content cull- taiaalot
Onpoaltlon Materia] tnchea Bo. lb/hr *r Ib/hr Slia* Ib/hr Ca/S Pt« pp> ppe pt*c ppo C5J 5, C5 tioo* Ib/hr
Ohio »l Sea*
Ohio ••
¦•(bad
(.51 lolfur
10.7* Aah
Frcmiitd wlUt coal 10 poonda 1 Leeatone par
100 poutd* coal
1420
15*0
1(30
mo
1110
loao
1010
loao
1010
1010
1110
1120
17*0
1M0
1M0
1010
;zso
2004
3400
32S0
340 310
J100
3200
2000
IS I
14 9
14.1
11.0
-11. s~
15 *
14.0
14 0
FBC
19
Oblo II lua
nibid
2 44 rulfur
7 24 Aab
Ask
10
1
2
1
10S
102
104
1430
14M
1470
low
low
10 so
J.o
3.0
3.0
13*41
-3M
0
ii.a
24.4
0
S. 2
4.1
2000
SO
300
240
•
311
MO
IK
270
140
SO
SO
SO
so
1« 1
14.2
IS 4
3.2
3.1
1.1
0
0
0
94.0
as o
ia.«
19.4
32 a
31.S
Bo 1 Additive Feeder
Ho 2 Additive Fwdu
rac
44
Oblo «• tMl
2.44 Sulfur
7.It Aah
Aah
11
1
2
1
4
Ui
1M
in
1SI0
1400
1410
itse
KM
loso
10S0
1.0
1.0
3-0
3.0
nsfa
0
24 0
17 2
13.0 •
0
3 4
2.4
1.1
2200
270
SSO
740
1970
410
146
340
344
344
140
130
140
W
so
w
so
w
14 4
15 0
14.4
1.1
1.1
2.7
•
•
0
0
as o
74.0
44.5
23 6
2a a
11.2
52 0
24.9
39.4
21.0
1.9
4 2
no. 1 Add 1 LIT* rndax
Mo 2 Additive fMdar
Ti —I ibiI flU coal. 11.0 po
100 pounda oo*l
i
!
i
1
rac
Ohio M Seta
vatbad
2 41 Gulfux
7.24 Aah
Aab
4
1
2
1
4
102
103
104
1710
1710
1090
1100
1090
lose
low
1034
1-0
3.0
3-0
J.0
13S9H
12 •
4
14.S»
0
2.0
1.4
2 4
2200
ISOO
1700
2100
1440
1410
13S0
140
140
>40
Ml
so
so
so
14 1
1S.0
1S.0
14.4
2.4
3.0
1 •
a.*
•
0
0
0
31 a
22.7
IS.9
14 2
13 9
«4.a
21 4
29.a
19.9
Bo 1 Add111va Feeder
Ho 2 Additive Fm4u
•Fraaiiced with coal. 14.S po
100 pound* ml
and a llMatese par
FBC
41
Otiio M lua
vaabed
J C4 fellfur
7 24 Alb
Aab
12
1
2
1
4
10*
101
110
1*70
ISIS
1600
1510
low
10M
10S0
1 0
3-0
3.0
3.0
1137B
O
14.0
19. 4
12.7*
0
1.4
1.14
2000
100
soo
1000
19S0
710
11
336
110
330
1M
so
so
so
so
IS 0
14.4
14.4
14 7
3.9
3 2
2.4
2.4
0
0
0
0
0
40.0
74 0
SO 0
42. a
44.9
43 0
4a.2
31 3
29 3
21 a
1.2
2.4
No 1 Additive Feeder
¦o 2 Additive Feeder
¦ProaLnd wltb ooel. 12 7 po
100 poocda coal
Mil 1 Lee atone per
rac
4»
Oblo *¦ Mas
«ub*4
2.4* sulfur
7.24 Aab
Aab
4
1
3
4
100
101
112
1710
1740
1770
17M
loso
loso
loso
1040
3-0
3.0
3-0
3.0
11371
0
17 3
14 0
12 7»
0
1.4
1 44
1 14
23S0
1400
1SS0
iao
lao
340
3ao
404
so
30
»
IS 0
14.7
14 2
7.4
3.0
3.4
J 2
0
0
0
0
0
40 4
31 0
11 0
2S.2
24 3
29 2
23 9
22 4
39 4
19 S
Wo. 1 AAJitive Feeder
No. 2 Additive Feeder
'FrMlad witii coal. 12 7 po
104 pourde coal
uade liaeatooe per
rac
10
Ohio 10 9eaa
weahed
2 44 Gal fur
7 24 Aab
Aab
12
1
2
1
1»S
101
110
1460
1S70
1510
1S70
loso
loso
loso
loso
) 0
3 0
3 0
1337B
0
14 7
ia 4
14 0*
0
1.3S
1 SS
1.44
1900
aoo
4S0
910
1BI0
140
490
910
40
370
370
37C
370
19S
so
50
so
so
14.5
14.5
15.0
3.4
2.9
3.0
B
0
0
0
S7.9
76 4
51.0
43 S
49.0
3S.0
46.a
S3 S
31.4
34 9
Bo 1 Additive Feeder
tto 1 Additive Feeder
•PTeninerJ with coal 14 0 po
100 poucde coal
und* 1 Li»eatone par
rac
Ohio M *--ir
waabed
2.44 Solfax
7 21 Aah
Alb
12
1
2
1
4
109
109
1544
ISM
1400
1400
low
low
loso
loso
) •
J 0
3.0
3.0
-32S
0
IS.6
19 0
14.0*
0
1.4
1.44
2100
1100
700
1110
2010
1040
SSO
1100
11
3ac
IK
no
iao
34S
so
so
so
14 1
IS.2
15 2
IS 4
3.4
1 0
3. J
0
0
0
0
0
47 S
44.7
47 5
34 0
41 7
32 8
Ho 1 Additive Feeder
mo 2 Additive Feeder
¦Preaind wltb coal 14 poua
100 pounla coal
da ILeeataue per
rvc
42
Ohio II ff-f
wsehed
2 44 Salfur
7 24 mk
Aab
12
1
2
1
4
111
10}
111
15*0
1570
1400
1400
loso
loso
loso
low
3 0
3 0
1.0
3-0
1337B
-325
0
21 a
19.4
u.i*
0
1.9
1.4
1.4S
2000
700
400
900
1190
490
440
•10
44
374
373
375
37S
19a
so
50
so
1S.0
m!i
1S.0
2 9
3 1
2.9
3 2
0
0
0
0
70 0
SS 0
431S
33.3
1.9
¦a. i Adsitive reader
no. 2 Additive Feeder
'frtaiod wltb coal. 11 po
100 potuda rnal
-------
task
B-*
4-2. (Ceritiouari)
Caa
Analva
titration Ln
parceot
*« Indlc
Static
rlue Caa
3ortM»rit Data
BO]
CmI Typa
sad
tajscn
Typo
tola
X
Vat
TMtl
1 R.
Sorb«nt.
A»*l £M
la
and
Coopoaltion
B*d
Dapth,
lachoa
Cond
Rate,
lb/hr
totp,,
Rata,
lb/hr
(Control),
and
SllB*
lb/hr
Ca/S
SO,.
SO] ,
BO,
WO*.
Hydro-
carbon*.
tnalyala
lion
II fi 1 .
Racir
PPO
CO)
o>
CO
tioe*
lb/hr
art.
rtc
M
Ohio II Mu
nah«J
3 It Sulfur
7 24 JUh
Atb
11
1
2
3
110
UC
1570
ists
1565
1050
1050
1050
3 0
3.0
1351ft
*51 -200
0
27 7
1<
3
2
2150
1100
1300
200C
910
1320
95
140
MO
1*0
50
SO
50
14 <
14.2
15.2
2
3
2
0
J)
41 9
39 5
16.3
19.7
1 1
2 9
•pTealiad with coal. K.O pounda 1U
100 fwni» coal
pa,
roc
55
Ohio it Mm
vaah«4
2.(1 Sulfur
7 2% Aab
*»b
10
1
2
4
UO
117
in
1610
1UD
1SI0
1S30
1050
1050
1050
1050
1 0
2 0
>0
itona
0
0
0
2100
2400
2110
2050
2400
2320
2290
2110
0
0
52
no
>10
390
420
1530
200
50
0
It.5
15 9
IS. 1
14 0
2
4
4
0
0
0
0
"
1.1
1 *
rac
sc
Ohio (I &¦«->
unvaahad
4 j* Sal fur
10 71 A»S
Aab
10
1
2
in
in
1513
1540
1050
1QS0
3 0
13371
-325
35 Q
0
1 12
3550
17S0
3450
1700
175
399
140
50
50
15 5
3
* 1
49 0
43 «
2.0
5.1
nc
it
Ohio It t*n
D»nh*3
4.st auirur
10 7« Aah
Uk
10
1
nt
113
1590
1570
1050
1050
3 0
37.*
0
1 25
3550
1400
3540
1400
32
3
JM
1*0
340
50
50
15 .
3.1
0
100
3100
3100
3BOO
1*00
3100
3710
3*90
3740
1*10
5
15
5
39
0
WB
«40
UO
*40
475
40
40
40
40
40
IS.5
15.7
15 4
15.4
15 S
2.5
2.5
2
2
2
2.4
0
0
0
0
0
35. *
41.0
45.0
*3.0
47 0
41 0
23.0
a.«
90j T*n
rac
M
Ohio «• B*m
umuM
4.St Sulfas
10.It lab
Aah
12
1
3
121
120
117
1S50
use
1010
1010
low
3.0
3.0
3.0
1359C
-325
0
25.0
14.2
0
2
1.5
3750
2200
25<0
J70C
2210
2*21
21
0
0
<00
«oo
<00
190
40
40
40
15 2
2
41.2
30 I
15 I
21 1
1.9
3.2
2.9
»BC
*5
Ohio *1 hN
unvaabad
l.ot sulfur
10.71 JUh
Aab
10
1
2
4
119
113
111
lit
1540*
1540
1540
lose
10*0
10SO
1040
3.0
3.0
3.0
3.0
13591
-12 *14
-14 *1*
-1*
0
21 0
>1.0
21.0
0
2 <**
3.4
2.<
2500
1170
1170
1700
2390
1*70
0
340
340
340
340
10
SO
>0
50
0
21.0
21.0
34.0
10.1
10.1
13.1
70 0
70.0
<5 0
•Tu+nratva vuiad terlag tot •
**Ma Taat for daacTlptm cf toti.
rac
(i
Ohio II Cm*
unvaahad
1 Ot Sulfur
10.71 Aah
Aab
"
1
2
)
«
113
113
lit
in
1550*
1550
1550
1550
1010
1011
1010
1010
3.1
3 1
3.1
1359«
-20 *30
•40 *50
-325
0
2«.l
24.2
21 2
0
2
2
2
asoo
1700
1140
7*0
]IM
1*10
1700
<50
400
ISO
*00
«
-40 *50
-325
0
14.«
K.O
K.7
0
2.5**
2.3
2
2450
1250
1*00
550
7450
100
340
440
440
50
50
50
50
0
49.0
34 <
77 0
19 4
13.1
30 I
<5.7
43 I
51 I
•TDBKritu* mm ftrltd tan.
•'SM'fnt fox diacrlptiai of 1 — La .
PBC
it
Ohio || Sua
UDHttflld
2.Ill Sulfur
10.71 hah
Aflb
10
1
a
4
12
13
- tl
12
1710*
nio
1100
17*0
120
120
<20
3.0
3.0
3.0
3.0
1359V
-325
-40 *50
-20 *30
0
IS.5
K.I
1C.2
2 »"
2 1
3 7
23S0
1700
2230
2150
340
140
310
340
SO
SO
50
SO
0
31 0
12 5
15 5
11 9
4 5
5.7
51.2
57 I
42 <
52 5
•TMptntora waa r&rlad tela) t*st.
**Ma Tait for daaarlptioa of ta«ta.
rac
<9
Oblo || 9«a
imuM
2 Olt Sulfur
10.It Aah
Aab
11
2
3
5
(9
*5
IS
<5
<5
i'to
17(0
1704
1730
420
120
cto
*20
*20
3.0
3.0
3.0
3 0
3.0
1359*
•20 *30
-40 *50
•100 *200
-325
0
17 0
1*.4
17.0
15 I
1
2 7 ••
2.7
2
1M0
2150
2230
1700
1700
440
410
10-400
400
so
50
50
50
0
15.1
11.0
32 0
32 0
S.t
4 2
11.9
12 1
44.1
53.9
K.O
47.2
5< 2
* la^iaiatara vai ftrltd asrLof Lest.
••Saa Tan foT daacTlptioo or tamtm .
rac
Ohio *1 Bttf
3 021 Sulfur
10 Tl Ash
Aab
10
10
10
11
11
1
3
3
4
5
Si
<2
45
t)
<9
1120*
15S0
1120
1SS0
1100
<20
*20
<20
(20
<20
3 0
3 0
3.0
3.0
3.0
1359*
-100 *200
0
16.*
17. <
17. <
1* 1
2 ••
2.1
2
2
2530
17S0
2100
1250
1900
410
4BO
450
440
4*0
50
SO
50
50
0
30 0
13.7
50.0
23.3
10.7
4.1
17.9
4.3
50.0
50 1
51.0
•Tuiiaraturo ou varlad dnrLoa Ml .
•*8oa Tost for daacrlptioo of tana.
PBC
11
OTliO II 8MB
anvaahod
1 Ot Solfur
10 ?t Ash
Aab
IS
1
7
4
*
it
*5
ti
45
11
USO*
1(50
1430
1470
11*0
*20
<20
<20
<20
<20
2.1
2 9
I >
3.0
3 0
1359*
-20 *30
•100 *200
-325
0
1* .3
16.3
1< 3
2 < **
2 Jt
2
2
2500
1950
2100
2000
1350
1400
370
ISO
310
140
370
375
50
50
SO
so
50
0
22 0
11 <
22 5
11.0
I 5
• I ?
11 I
54.3
44.5
37.0
42.0
49 0
"Tanparatora au rarlad dnlaj laat,
**&aa Tut for daacriptloe of laata.
10 '»
-------
T&BLE >-?
(Coatinuad)
CmI Typa
and
CMpoaltlon
Co* I
Rate,
lt»/Kr
pjo ppc-
Cirbon Rocir-
Content cul*- Baiaaion,
« t ion* lb/hr
Ohio IS Saaa
vaahad
J.04* Sulfur
WOO*
71.S 1(00
10 5 1640
1510
1SS0
15S0
1S<0
1050
ioso
1050
ioso
1050
IPSO
1050
1050
1050
iSIr
II 2
if. 2
1* •
11.5
21.7
20 0
20.2
2400
1510
56 0
IS 0
74 0
21 0
28.0
21 0
_24.7_
lltira r**4 1 alda (CtBd. 2)
lltiva rMd Z lidoa (Ceod J)
litlva Pa«d 4 ildu (Cond 4)
Paed 2 aid** ap^oaita (Cood i]
. i.i i*t.io : etfoiiw (i wun
Obld II Shi
/Wabad
/J.0% Sulfur
/10.7% Aab
15«0
liio
1570
1050
1050
1050
2610
1100
1040
>olBt of lajactloo nnrf
ihmn Md M (late |u apaca) durtnv Cond 2
Halt of M (Md dulif Coed. 1
Ohio la Uu
14 9 3 1
Superficial Valocitj (
Ohio la hta
Mflwd
1 05% Sailor
1S20
1510
1530
15W
1519
For Taata 101-110
SO] Data partaia t«
alngla point 1b tla
55.1
41 1
56 0
Superficial Valocity C
*£•• Appndlx A, BticU
"Pluidliad Bad of Liaa
tranalant ic Mtnn <
•S*« Appandi* ft. taeloaon 14.
'•PlaldlMd Bad of UautdM. mo aorbaat croaa flow Talta «ri
aatsr* aad r*a«lta ara praiaatad in graphical for*
•Sm Appandli *i bcloaen XI.
"Pluidliad Bad of 11—al iit i Wo aorbaat croaa flow Taat* vara
uualait 1b aator* and tunlt# *fa preaaatad La fnphlol fora.
•Boa Appendix a, edcIomt* 2».
'•rlulditad Bad of Llaaatoca. So aorbanl croaa flow Taata wara
traaaiaat is aatara tod rani t a ara prasaatad in graphical for*.
Ohio II fill
vcibad
2 95% Sulfur
1M0 -
17«
1520 -
1«00
1400 -
1710
45 5*«
41 I
JJ 6
*S«a AppendLi a, tMleasn IT.
'•Muldirad Bad of r IaaaliUM . ao aorbaat croia flow Taata wara
tranaiast is natura aad rvaalta ara praaaatad In graphical fora.
107 mt
LMO
3 0% Balfor
420 420
15 I
J3 *
)1 J
>7.9
'Sm Kfparfli A, tseioam II.
"rluldltod Bad of Llaa*tM*. mo aorbaot croaa flo« Taata wara
traaaiaat in utan aad nsalta an praaaatad la fraphlcal fora
Oblo II Swa 135t»
waabad -10 *20
J.05* Gulfor
-1/4*0
1500 -
1510
1720
1720
'Sm Appoadli A, Eaeloaor* N
"Ploidiaad Bad of Liaaatxxta. Bo aorbaat croaa floa. Taata vara
traaaiaat La oatnr* aad rtnlu ara praaaatad La graphical fen
1470 -
1*20
1*70 -
1610
1600 -
1420
1100
1100
1100
fcr*aala_.
• ¦ Iba 13591
•13 Iba 13591 addad
utixa, 'mo aorbaat croaa flew Tfata Mi
and ntolu ut praaaatad in graphical foi
t «aq of coidiLlu
I of emdltlae
Obio It tMl
vaahad
j, 42% Saifur
-1/4 a 0
"'Fluidiiad Bad of
Ohio »l 8a«a
3 Mt Sulfur
'tao Appeadl* a, Bclonra ]l.
**Plaidliad Bad of Liaastooa Bo aorbaat c
traaaiaat li oatnr* and raaalta ara praaa
Addad 60 Iba fly a«b
A)
1450 -
1540
1S40 -
1319
1100
_uee_-
54 o
47 6
18 H-
¦Saa Appeatiii k, bkIsmt* 11.
>*rioldliad Bad of LlaaatoM. Bo aorbaat croaa flow. Taata wara
traaaiaat La aatara aod raaulta axa praaaatad In frapblaal for*.
i lbs of b*d in 5- lb ij ii. i ¦ — trf
-------
»-l
CM Aritlyala. CoIX
TABU i-2 (Coetinsad)
ration In rota— pre am or t
Coal Typa
aoa
Coapoaitlon
Coal
Rata,
lb/hr
Sorbtnt. pats
lnJ«ciion
lb/hr
Ca/S PP"°
PPO PP=
),
Sorbcnt
Ulihia-
tion.d
* 113
Ohio <¦ 1MB
waahad
J.OSI tolfur
13S9R
-10 *20
10
1
2
3
14
S3
<0
1410 -
1(00
450
*50
»5o
3 0 ••
3 0
3.0
400 0
14.3
11 2
17 2
3.1 0 " ••
3.0 '• '•
53. 5
42-1
49 I
¦Jm AppTnllJ 1, n«.lr—la >4.
"Ploldiiad Bad of rf aniaa to aert
42 Lb* 13391 idM
toat creat tiae Taut* wars
prvMAt«d Id grapfalctl fw.
roc
114
Ohio II Baaa
W*lh*d
J.09% Sulfur
1359*
-10 <20
10
1
2
3
s?
*¦»
7S
1490 -
1900
620
720
720
3.0 •• "
3.0
J 0
•» • 0 0
215
2S5
240
11.1
17.0
17.0
2 • 0 •• ••
2.9 •• "
3 0 *" ••
41 7
44.?
C2 3
44.9
49 lb« 1359ft iMal
*S*« Appaadii A, Eoc>oacr* 35.
"Flaldiiad tod of T i — h i— *o aori
truiliat La satara ol ntfalu «ra
toal eroaa flow Tarta vara
praautad lo «raptil«al fora
pbc
11)
Ohio II toaa
«20
10
1
2
3
47 4
47 4
17SO
isao
1510
480
500
3.0 •• '•
3.0
3.0
ns o
276
210
1< 0
14.0
3.0 0 •• ••
3 1 *• ••
42 7
40.4
67 lba VlStB avail
Cat raloelty daeraa»art few 1-2 to 1
•las '1T***,'» *» 34.
'*Plal£„^ad to* "( LlMtaa, la mot
triaalaat La natiaa ad raaalti ara
(pa
43 5
baat eroaa flew. TaitA wara
praaantad Id graphical fora.
rtc
ll«
Ohio M lama
3.02* Solfor
13391
-10 *20
10
1
2
3
41
tl
15&0
1930
1610
(SO
tiO
(SO
3 0 •* **
1.0
230
222
350
250
240
11.2
14 1
17 2
3 0 0 ••
1.2 2 ••
J 1 0 ••
55 1
It 9
52 «
"}« AfiputfU A, tnr.1o»Tw J7.
"Floidliad tod of UMtm. to ao*1
tranaiaat La Mtsra ad waalu ara
40 lba bad ¦atari*! " 1
Mnt eroaa fit*. Twit* nara
praaaAtad in graphical fora
roc
117
Ohio It 8*a
waaftad
J ®7i SmUoi
1359S
-10 *20
10
1
2
3
57
S<
•2
1460
2000
lt«0
1600
~50
(SO
*40
*50
3 0 •• "
.}
3 0
•• 0 0
219
210
440
260
430
11.7
11.2
17 •
11 O
2 9 0 •• ••
.2 4 ** ••
2 9 0"
1 2 •• ••
44 «
44 €
SI 2
57.6
>B*a Appendix A, beliMB* M-
40 lba bad oatatUU adtal
••riuldiiad tod of Linertma to aor
bant croai t low Taat* vara
(traaantad Ld qrt^Uctl (on.
rac
ill
Ohio II Sau>
waahtM
3 H* Sulfur
13S»*
-10 *20
10
2
3
4
• 0
47
(I
79
1310
1930
isse
1930
450
t.5 0
(50
s»0
3 0 •• ••
.2
5.0
.2
5 5*
359
3t0
140
410
310
50
420
0
420
u %
14.2
11 7
13.1
1,1 « " •*
o .1 •• ••
4.1 0
.1 .2 *• »•
55 1
*1.1
CS-0
50.9
•8*a AppodLa A, ¦— 17.
**Pl«ldlMd tad of UaMtn. to ml
tranaiaac lo utm and iwlU «r«
Mat cro«« flow. Taati
praaa^tad Id graphical fora.
PBC
119
Ohio II Saaa
«aab*d
3 Olt Sulfur
13S9I
10
1
20C0
1418
2000
1J10
450
450
450
(50
3.0 •* ••
319
%so
11 I
1.1 0 •* ••
31 2
'hi kffttdix A. tuili'i—¦ M-
"riuldliaa tod o( "-*¦¦¦- to tart
tdulist Id niear* «ad naslU ara
40 lba 1359S addad
>aot croai fuw. Taau wwra
praaantad Id graphical fora.
3
4
1 0
3.0
310
j«a
It 9
11 «
9 0.2 •• ••
i i « " "
43-9
52 <
PBC
120
Ohio II lw
vast) *4
3 01* Sulfur
133**
-10 *20
10
1
(0
1700
1530
<50
450
4.0 •* '*
3 0
310
J10
>100
20.5
15 2
It 5
2 9 0 •• ••
3 2 M •*
3 2 **
54 3
59 1
11.1
54.7
'Im Appaodia A, Baclo—«i 41 -
40 lba 1359K addad
"riuldliad tod of Lleeetooa to aer)
truslant Id ttaton ad tmlu ara
Met croa* flow. Taata wara
praaaatad la graphical form.
\lM»ceM 14«ntloua Cc«t toaaarch, lac.
C • ealclaad by auppliar. B * hvdratad by aoppliar,
R • raw itoaa
1 9 Standard llao* alia
Otillfticn daflnad at!
lOtil nation - |ufgK?S^ir lUtlo?*^
4Gj« i
•fly <
I by Infrared Aaalyx
rKl«nl*t«' nl«a
by 'T*i*
&
-------
APPENDIX C
SULFUR BALANCE DATA FBC AND FBM
POPE E^.NS
-.2-JO RCB31NS
5IiST
SBOH OfNTV S MVA3
cj to H1
m cr> o co
o o t-1 o
it' U NJ H
:8^a!
w W H
I—1 H' t—1 (O
^ U W H
Ji»U» W H
t-3
£ n> '-d
CJ
one
to
• rt n
o
^ o
to M
O 3
• &
to
t-1 w o
o* ao
H CJ
W H-
N,W 0)
C"> I
2 w c
o ui
V- M
CO 0 m
(t3C
e n
W H -J CO
4* Cn 00 00
W U WH
K) O in CO
45k U) CJ \->
I—1 O O CO
W U N H
CO O CO CO
W W W H
U1 —I si
O yi C* Ul WHHH' ff! ^ CD W ^ O W ^
00 00
o> cn
L71 CO
o o
CO CD
M O
*3
\
310>
•ft 0>
CO (T> W
ft ST
C
W ^3
C H-
•
•
•
• •
•
•
«
•
•
•
•
03
o
o
o
O
o
o
o o
o o
o
o
o
o
O
o
O
o
o
o
o
o
o
o
o
o
0
l-t>
4*
to
<_n
to
CO
CJ
ui to
cn cn
<7\
1—*
o\
cr>
a\
to
CJ
cn
cn
h-"
to
cn
cn
rt
C ft)
si
Cj
CJ
CJ
CO
CO
a\ j—1
cn i—1
u>
si
o
CO
UI
CO
to
CJ
00
cn
si
4*
si
to
H-
O
3
h w
3*
F
W
vo^o
fo un co to
-j 4* *x>
W O CD lb
sun h A
I—' 1—1 I—'
^ A CJ UI
Ul 4^ 00 to
J—1 4*
-j o
CJ CD O CO
Ui itk H
C© 4k 4* to
h-1 UI ui 4*
to CO M 4*
•sip W U>
O J5k u> to
CJ ui si A CO CJ
n] H J* en jk (O
Cj cj cj cj
si nI -J vj
o to o o
CJ CJ CO CJ
SlvlsJvJ
O h-» O O
Cj cj Cj Cj
si sj nJ -si
o o o o
cj cj cj Cj
si si sj sj
CJ CJ CJ CJ
si si si s]
Cj Cj
si si
CJ CJ
si sj
T-0
0* ^ M c
(15 • 3 H
CJ CJ
¦"-s. cnH3 Hh
SIdP C C
si si
r+ H
CD W
rt
C
G
Ir" 3
cr wp
1
in c o
\ o
CJ
2! k> o
-Ck to
C C
CD H 3
rt rt
C 0
-------
SNISQOti OSNVAS "3dOci
^3
Z n> ^
0 w t3
• rt O
O
2S O
0 3
• D*
F CO
CfHO
01 3 n
N» H-
S|u> cn
co c
a h h
fto Ml
e 3 e
H
m
rr 3"
C
o o o o o o o
s] >J IO ^ --J 0"N NJ
O KJ K) -4 H W H
1 W'fl
H c M
D> !-•*<
0 hh
ftC t»
M- H 0)
O 3"
3
cr co ij
w c H
\ j-"^5
S|H»
C 0J
w h cd
rt 3*
c
o
o
3
3
C
o
Cli
U> LJ tJ U)
U» U> U>
-0 -J -J
U) (J u u
-~J -J -«J *o
o o o o
CO
c
Hfil O
ht)3 ft
c a &
n m
"d
V o» a
2|cn as
3* H-
to O
ft 3
C
El ~ co
cr ahc
W • 3 H
\ U1»0 H»
S|t#> C c
rt H
to co
6 §
to tn ir
K C o
3|h- n
Hi O
wee
rt 1 3
C if
C-3
TABLE C-2. FBC SULFUR BALANCE DATA
FBC Data: Rates in pounds per hour
TEST NO.
46
Additive
Test Condition
Additive St. Ratio
Suifur input
Sulfur emission
Sulfur in fly ash
Sulfur retained in bed
Input less output
0.0
2.72
2 . 27
0 . 38
0.07
3 .6
2.57
0.44
1.30
0.80
0.03
1359 H
Coal Sulfur Contej.t
2.50
2.6
2 .77
0.65
1.62
0.40
0.10
2.1
2.75
0.84
1.34
0 . 40
0.17
TEST NO.
47
Test Condition
Additive St. Ratio
Sulfur input
Sulfur emission
Sulfur in fly ash
Sulfur retained in bed
Input less output
0.0
2.65
2.40
0.24
0.01
2.0
2 .70
1.64
0.57
0.35
0.14
Additive
1359 H
Coal Sulfur Content 2.60
1.4
2.75
1.91
0.76
0.17
- 0.09
2.6
2.78
1.52
1.04
.17
0.05
TEST NO. 48
Additive 1337 H
Test Condition
Additive St. Ratio
Sulfur input
Sulfur emission
Sulfur in fly asli
Sulfur retained in bed
Input less output 0.15
0.0
2.70
2.23
. 32
1.4
2.58
0 . 84
1.08
0.70
0 . 04
1.6
2.75
0.67
1. 55
0.35
0.13
Coal Sulfur Content 2.50
4
1.16
2-62
1.16
1.18
0.35
-0.02
P~=E EVANS roe:
-------
C-4
TABLE C-2. (Continued)
TBC Data: Rates in pounds per hour
TKST NO. 49 Additive 1337 H
Coal Sulfur Consent 2.60 %
Test Condition 1
2
3
4
Additive St. Rctio 0
1.6
1.4
1.16
Sulfur input 2.62
2.73
2.73
2.80
Sulfur emission 2.S2
1.54
1.53
1.53
Sulfur in Ely ash *22
.73
.95
1.05
Sulfur retained in bed
.35
.20
. 14
Input loss output -0.12
0.13
.05
.08
TEST NO. 50 Additive 1337 H
Coal Sulfur Content 2.50 %
TesL Condition 12 3 4
Additive St. Ratio 0 1.3 1.5 1.46
Sulfur input 2.56 2.62 2.58 2.67
Sulfur emission 2.10 .92 .61 1.11
Sulfur in fly ash 0.32 1.15 1.41 1.14
Sulfur retained in bed ~ -^0 *^0 *30
Input less output O.I4 -0.15 - .04 0.13
"TEST NO._51 Additive 1337 H
Coal Sulfur Content 2.60 %
"Test Condition 1 2
Additive St. Ratio - 1.4
Sulfur input 2.52 2.46
Sulfur ei.iis:? i 0:1 2.19 1.25
Sulfur in fjy anii .22 .80
Sulfur lvUiir'jd in ~ *40
Input ]csu oulpul. 0.11 0.01
POPE E.VA.NIS X-.l-JO R0"3:F~:M£.
3
4
1.6
1.46
2.72
2.62
.88
1.27
1.40
1.14
.30
.15
0.14
0.06
i3=:.-!S.
TABLE C-2. (Continued)
FBC Data: Rates in pounds per hour
TEST NO. 52 Additive 1337 H
Coal Sulfur Content 2
Test Condition
1
2
3
4
Additive St. Ratio
0
1.9
1.6
1.65
Sulfur input
2.58
2.67
2.70
2.68
Sulfur emission
2.15
0.71
0.68
0.98
Sulfur in fly ash
0.38
1.20
1.55
1. 30
Sulfur retained in bed
-
o
r-
o
0.40
0 . 30
Input less output
0 .05
0.06
0. 17
0. 10
TEST NO. 53 Additive 1337 H
Coal Sulfur Content
Test Condition 1 2 3 _4
Additive St. Ratio 0 1.6
Sulfur input 2.68 2.60
Sulfur emission 2.35 1.53
Sulfur in fly ash -36 .57
Sulfur retained in bed - .60
Input less output -0.03 -0.10
TEST NO. 54 Additive 1359 R
Coal Sulfur Content 2.
Test Condition 1 2 3
Additive St. Ratio 0 3.0 2.0
Sulfur input 2.72 2.80( 2.72
Sulfur emission 2.39 1.28 1.51
Sulfur in fly ash 0.28 0.78 0.70
Sulfur retained in bed ~ 0.60 0.40
Input less, output 0.05 0.14 0.11
-------
C-6
TABLE C-2. (Continued)
FBC Data: Rates in pounds per hour
TEST NO. 56
Additive 1337 R
Test Condition
Additive St. Ratio
Sulfur input
Sulfur emission
Sulfur in fly ash
Sulfur retained in bed
Input less output
0
4.72
3.96
.59
0.17
1.12
4.53
1.93
1.80
0.70
0.10
Coal Sulfur Content 4.40 %
POFE. E'v'ANS AND ROBBINS
TEST NO. 57
C-7
TABLE C-2. (Continued)
FBC Data: Rates in pounds per hour
Additive 1337 R
Test Condition
Additive St. Ratio
Sulfur input
Sulfur emission
Sulfur in fly ash
Sulfur retained in bed
Input less output
0.00
4.80
4.15
.40
0.25
1.30
4.80
1.68
2.36
0.60
0.16
Coal Sulfur Content 4.3 %
TEST NO. 58
Additive 1337 R
Test Condition
Additive St. Ratio
Sulfur input
Sulfur emission
Sulfur in fly ash
Sulfur retained in bed
Input less output
0.0
4.75
4.22
.45
0.08
1.60
4.75
1.20
2.55
0.70
0.30
Coal Sulfur Content 4.3 %
TEST NO. 59
Additive 1337 R
-Test Condition
1
2
Additive St. Ratio
0.0
2.3
Sulfur input
2.83
2.83
Sulfur emission
2.82
0.83
Sulfur in fly ash
.15
1.44
Sulfur retained in bed
-
0.40
Input less output
-.14
0.16
Coal Sulfur Content 2.6 %
POPE EVANS AND ROBBINS
-------
C-8
TABLE C-2. (Continued)
FBC Data: Rates in pounds per hour
TEST NO¦ 60 Additive 1359 r
Coal Sulfur Content 2.6 %
Test Condition
1
2
Additive St. Ratio
0.0
2.0
Sulfur input
2.85
2.85
Sulfur emission
2.65
1.11
Sulfur in fly ash
.1
1.35
Sulfur retained in
bed "
.35
Input less output
0.10
0.04
TEST NO. 61 Additive 1359 R
Coal Sulfur Content %
Test Condition
1
2
Additive St. Ratio
0.0
1.25
Sulfur input
5.35
5.10
Sulfur emission
4 .75
2.39
Sulfur in fly ash
.38
1.93
Sulfur retained in
bed
.70
Input less output
0.22
0.08
TEST NO. 62 Additive 13 59 R
4 .
Coal Sulfur Content
Test Condition
1
2
Additive St. Ratio
0.0
1.6
Sulfur input
5.00
5.00
Sulfur emission
4 .42
1.90
Sulfur in fly ash
.42
2.40
Sulfur retained in bed
-
.55
Input less output
0.16
0.15
POPE. E>ZANS AND ROBETNS
C-9
TABLE C-2. (Continued)
FBC Data: Rates in pounds per hour
TEST NO. 63 Additive 1359
Coal Sulfur Content 4.^ %
Test Condition
1
2
Additive St. Ratio
0.0
1.6
Sulfur input
4.90
4.90
Sulfur emission
4.50
2.03
Sulfur in fly ash
.30
2.20
Sulfur retained in
bed -
.55
Input less output
O
O
0.12
TEST NO. 64 Additive 1359 R
Coal Sulfur Content ^*
Test Condition
1
2
3
Additive St. Ratio
0.0
2.6
1.5
Sulfur input
5.30
5.30
5.30
Sulfur emission
4.60
2.70
3.14
Sulfur in fly ash
.30
2.00
1.60
Sulfur retained in
bed -
.50
.30
Input less output
0.40
0.10
0.26
eva>"= Axr,
-------
C-10
TABLE C-2.
FBC Data:
(Continued)
FBC Test No. 106
Test Condition
1
2
3
Flue gas cutput
1.70
.27
1.25
Fly ash output
.41
.37
.33
Bed retention
.95
2.4
1.6
Total output
3.06
3.04
3.18
Ikiput
3.12
3.12
3.12
Input-output
.06
.08
-.06
FBC Test No. 107
Test Condition
1
2
Flue gas output
.60
3.75
Fly ash output
.35
.39
Bed retention
2.20
-.90
Total output
3.15
3.24
Input
3.16
3.16
Input-output
.01
-.08
FBC Test No. 108
Test Condition
1
2
3
Flue gas output
1.45
2.5
2.7
Fly ash output
.42
.28
.32
Bed retention
1.1
.28
.08
Total output
2.97
3.06
3.10
Input
3.13
3.16
3.13
Input-output
.16
.10
.03
*Tests 106-120 run with bed of 1359 limestone calcined in
place. Rates are in pounds of sulfur per hour.
POPE. EVANS AIJD ROBBING
C-ll
TABLE C-2. (Continued)
FBC Data: Rates m pounds per hour*
FBC Test No. 109
Test Condition
1
2
3
Flue gas output
.9
.9
.82
Fly ash output
.25
.19
.23
Bed retciiticn
1.9
2.0
2.2
Total output
3.05
3.09
3.25
Input
3.18
3.18
3.18
Input-output
.13
.09
-.07
FBC Test No. 110
Test Condition
1
2
3
Flue gas output
1.02
1.28
.81
Fly ash output
.32
.25
.29
Bed retention
1.7
1.6
2.0
Total output
3.04
3.13
3.00
Input
3.18
3.18
3.18
Input-output
.14
.05
.18
FBC Test No. Ill
Test Condition
1
2
3
Flue gas output
2.52
2.52
2.52
Fly ash output
.02
•45
;39
Bed retention
.50
• 02
.02
Total output
3.04
2.99
2.93
Input
3.17
3.17
3.17
Input-output
.13
.18
.24
"5 —
Tests 106-120 run with bed of 1359 limestone calcined in
place. Rates are in pounds of sulfur per hour.
A M-J PCBSINS
-------
C-12
TABLE C-2. (Continued)
FBC Data: Rates In pounds per hour*
FBC Test NO. 112
Test Condition
1
2
3
Flue gas output
2.74
2.1
1.8
Fly ash output
.17
.36
.32
Bed retention
.2
. b
.8
Total output
3.11
2.96
2.92
Input
3.00
3 .00
3.00
Input-output
-.11
.04
.08
Tests 106-120 run with bed of 1359 limestone calcined in
place. Rates are in pounds of sulfur per hour.
C-13
TABLE
C-2.
(Continued)
FBC
Data:
Rates
in pound
s per
hour*
FBC Test 113
Test time hrs
1
2
3
4
Flue gas output
0.0
0.08
.6
.7
Fly ash output
0.35
0 .28
.32
.38
Bed retention
1.56
1.64
.90
.80
Total output
1.91
2.00
1.82
1.88
Input
1.98
1.98
1.98
1.98
Input - Output
.07
-.02
. 16
.10
FBC Test 114
Test time hrs
1
2
3
4
*
4.3
Flue gas output
0.0
.39
0.85
1.15
10.4
Fly ash output
0.27
.25
0 .35
.27
.30-
Bed retention
1.75
1.47
.7
.5
-8 .8
Total output
2.02
2.11
1.90
1.92
1.8
Input
1.95
1.95
1.95
1.95
1.95
Input - Output
-.07
-.16
.05
.03
.15
*
Regeneration
FBC Test 115
Test time hrs
1
2
3
4
5
6
Flue gas output
.08
.16
.63
.45
-
.8
Fly ash output
.26
.40
.34
.14
-
.25
Bed retention
1.65
1.33
.84
.99
-
.44
Total output
1.99
1.89
1.81
1.58
-
1.49
Input
1.95
1.95
1.95
1.38
1.38
Input - Output
-.04
.06
.14
.20
. 11
*Tests 106-120 run with bed of 1359 limestone calcined in
place. Rates are in pounds of sulfur per hour.
-------
C-14
TABLE C-2. (Continued)
FBC Data: Rates in pounds per hour*
FBC Test 116
Test Time hrs
1
2
3
**
4.2
5
6
Flue gas outpi't
0.00
.08
.42
17.5
-
.16
Fly ash output
.45
.37
.30
.27
-
.26
Bed retention
1.48
1.41
] .20
1
o>
O
-
1.57
Total output
1.83
1.86
1.82
1.77
1.99
Input
1.96
1.96
1.96
1.96
1.96
Input-Output
.13
.10
.04
.19
-.03
FBC Test 117
Test Time hrs
1
2
3
**
4.4
5
6
Flue gas output
0.00
0 .00
.16
25.5
.41
-
Fly ash output
.42
.54
.40
.6
.24
-
Bed retention
1.47
1.40
1.40
-24.4
I—'
00
-
Total output
1.89
1.94
1.96
1.7
1.83
Input
1.85
1.85
1.85
1.85
1.85
Input-Output
-.04
-.09
-.11
.15
.02
FBC Test 118
Test Time hrs
*±
1
2
3
4
4.8**
6
Flue gas output
20 .5
.16
.47
.37
21.5
-
Fly ash output
.2
.25
.22
.15
.15
-
Bed retention
-19 .0
1.34
1.05
1.30
-19 .8
-
Total output
1.7
1.75
1.74
1.82
1.85
Input
1.8
1.80
1.80
1.80
1.80
Input-Output
.1
.05
.06
-.02
-.05
* Tests 106-120 run with bed of 1359 limestone calcined in
place. Rates are in pounds of sulfur per hour.
* *
Regeneration
POPE, EVANS AKD RCB3INB
C-15
TABLE
C-2.
(Continued)
FBC
Data :
Rates in pounds per hour*
FBC Test 119
Test Time hrs
1
2
3 4
5
6
Flue gas output
.19
.65
.20
.25
.62
Fly ash output
.30
.26
.14
.20
.18
Bed retention
1.41
.90
1.44
1.30
.96
Total output
1.90
1.81
1.78
1.75
1.76
Input
1.88
1.88
1.88
1.85
1.88
Input-Output
-.02
.09
.10
.10
.12
FBC Test 120
Test Time hrs
1
2
3
Flue gas output
.00
.05
.50
Fly ash output
.36
.25
.02 (Recircu
lation)
Bed retention
1.56
1.60
1.52
Total output
1.92
1.90
2.04
Input
2 .02
2.02
2.02
Input-Output
.10
.12
-.02
*Tests 106-120 run with bed of 1359 limestone calcined in
place. Rates are in pounds of sulfur per hour.
= OPE EVA. 'S ANO R03fe!>
-------
C-16
TABLE C-3. FBM SULFUR BALANCE DATA
FBM Data: Rates in pounds per hour
TEST NO. 17
Additive 1359 H
Test Condition
Additive 8t. Ratio
Sulfur input
Sulfur emission
Sulfur in fly ash
Sulfur retained in bed
Input less output
0.0
32.2
26.3
1.6
2.3
.72
32.2
20.5
9.3
2.8
- . 4
Coal Sulfur Content 4-3 *
.84
34 .2
19.0
10.5
2.2
2.5
TEST NO. 20
Additive 1337 H
Test Condition
Additive St. Ratio
Sulfur input
Sulfur emission
Sulfur in fly ash
Coal Sulfur Content 2.6 %
0.0
19.0
17.4
1.0
Sulfur retained in bed —
0.6
Input less output
1.17
23.0
15.1
7.0
1.2
-0.3
1.46
23.4
12.6
8.7
.6
1.5
TEST NO. 21 Additive 1337 H
Coal Sulfur Content 2-5 *
Test Condition
1
2
Additive St. Ratio
0.0
1.37
Sulfur input
22.3
24 .0
Sulfur emission
19.7
9.1
Sulfur in fly ash
.8
10.4
Sulfur retained in
bed
2.5
Input less output
1.8
2.0
F=OF>E. EVANS AT-JD ROHB1NS
C- 17
TABLE C-3. (Continued)
FBM Data: Rates in pounds per hour
TEST NO. 22
Additive 1337 H
Test Condition
Additive St. Ratio
Sulfur input
Sulfur emission
Sulfur in fly ash
0.0
23.1
22 .4
.3
Sulfur retained in bed
Input less output 0-4
1.46
22.9
7.0
11.8
2.8
1.3
Coal Sulfur Content ^-5 *
TEST NO.
23
Test Condition
Additive St. Ratio
Sulfur input
Sulfur emission
Sulfur in fly ash
0.0
20.8
20.3
.8
Sulfur retained in bed —
Input less output -0.3
2.4
20.8
6.8
11.9
1.8
0.3
Additive 1337 R
Coal Sulfur Content
2.5 %
TEST NO.
24
Additive 1337 R
Coal Sulfur Content 4.3%
Test Condition 1
Additive St. Ratio 0.0
Sulfur input 18.9
Sulfur emission 18.2
Sulfur in fly ash .8
Sulfur retained in bed —
Input less output -0.1
2.4
19 .9
4.0
13.6
1.8
0.5
2.2
19.9
4.4
13.5
.9
1.1
-------
C-18
TABLE C-3. (Continued)
FBM Data: Rates in pounds per hour
TEST NO. 25
Test Condition
Additive St. Ratio
Sulfur input
Sulfur emission
Sulfur in fly ash
0.0
35.0
3C.5
1.8
Sulfur retained in bed —-
Input less output
2.7
1.8
35.2
9.7
21.6
2.8
1.1
Additive
1337 R
Coal Sulfur Content 4.3 %
TEST NO. 26 Additive 1337 R
Coal Sulfur Content %
Test Condition
1
2_
3
Additive St. Ratio
0.0
1.7
1.9
Sulfur input
39.6
39.6
39.6
Sulfur emission
34.5
12.6
10.4
Sulfur in fly ash
1.7
21.0
24.O
Sulfur retained in
bed —
4.5
2.7
Input less output
3.4
1.5
2.5
TEST NO. 27 Additive 1359 R
4.3 *
Coal Sulfur Content
Test Condition
1
2
Additive St. Ratio
0.0
2.0
Sulfur input
33.2
33.2
Sulfur emission
28.3
7.9
Sulfur in fly-ash
2.1
17.8
Sulfur retained in bed"
—
5.6
Input less output
2.8
1.9
POPE, EVANS AMD ROB3INS
C-19
TABLE C-3. (Continued)
FBM Data: Rates in pounds per hour
TEST NO. 28 Additive 1359 R
Coal Sulfur Content 2.
Test Condition
1
2
3
Additive St. Ratio
0.0
2.4
2.2
Sulfur input
20.6
20.3
20.3
Sulfur emission
20.9
6.8
7.7
Sulfur in fly ash
0.8
11.0
11.1
Sulfur retained in
bed —
1.8
.9
Input less output
-1.1
0.7
0.6
TEST NO.
29
Additive 1359 R
Test Condition 1
Additive St. Ratio o.O
Sulfur input 30.9
Sulfur emission 27.2
Sulfur in fly ash 2.4
Sulfur retained in bed —
Input less output 1.3
1.7
31.0
11.0
14 .2
4.5
1.3
Coal Sulfur Content *-¦
2.0
31.0
7.7
17.6
3.6
2.1
TEST NO. 30 Additive 1359 H
Coal Sulfur Content
Test Condition
1
2
3
Additive St. Ratio
0.0
1.4
1.8
Sulfur input
22.4
22.6
22.6
Sulfur emission
O
22.0
11.0
8.9
Sulfur in fly ash
.6
9.0
11.5
Sulfur retained in
bed —
2.0
1.2
Input less output
-0.2
0.6
1.0
POPE EVANS AND ROBBINS
-------
C-20
TABLE C-3. (Continued)
FBM Data: Rates in pounds per hour
TEST NO.
Additive 1359 H
Coal Sulfur Content 2.6 %
Test Condition
Additive St. Ratio
Sulfur input
Sulfur emissio\^
Sulfur in fly ash
Sulfur retained in bed
Input less output
1
2
3
o
o
1.3
1.6
20.8
20.8
20.8
20.0
10.3
8.8
0.2
8.2
9.5
-
1.6
0.8
0.6
0.7
1.7
TEST NO. 32 Additive 1359 R
Coal Sulfur Content 2.6 %
Test Condition
1
2
3
Additive St. Ratio
0.0
1.6
1.8
Sulfur input
18.7
19.0
19.0
Sulfur emission
18.5
7.5
6.6
Sulfur in fly ash
0.2
9.3
10.5
Sulfur retained in bed
-
1.6
0.8
Input less output
0.0
0.6
1.1
POPE EVANS AND ROBB1NS
S6 / f/
BIBLIOGRAPHIC DATA
SHEET
T. Report No.
APTD-0655
4. Title aod Subtitle
Characterization and Control of Gaseous Emissions From Coal-
Fired Fluidized-Bed Boilers
3. Recipient's Accession No.
5* Report Date
October 1970
7. Awhof(s)
E. B. Robison, A. H. Bagnulo, J. W. Bishop, and S. Ehrlich
8. Performing Organization Repc.
No.
Performing Organization Name tod Addresa
Pope, Evans and Robbing
Consulting Engineers
A Division of Perathon Incorporated
Alexandria, Virginia 22314
10. Project/Tnak/Vork Uou No.
II. Contract/Gram No.
12. Sponsoring Organization Name and Address
Division of Process Control Engineering
National Air I-ullutlon Control Administration
Environmental Health Seivice - Public Health Service
Department of Health. Education, and Welfare
Washington. D. C.
13. Type of Report & Period
Covered
IS. Sopplecseatary Nocea
16. Abstract*
j Results are prooonted from a test program to characterize the air pollution
emissions from the combustion of coal in a fluidized bed combustion and to assess
the potential of fluidized-bed combustion for air pollution control. These
emissions were monitored under a comparatively large number of different conditions
Efforts were made to reduce emissions of oxides of sulfur by the use of limestone-
based sorbents and to determine the conditions u,ost favorable for the reduction.
Emissions of sulfur dioxide, nitric oxide and hydrocarbon were monitored continuously
with periodic samples taken for measurement of particulates and wet test deter-
mination of ScO and Nps;. When conditions most favorable for air pollution control
were established on a pilot scale, the conditions vere reproduced in tests with the
fluldlzed-bed boiler module.(s ^
If. Key Vorda and Document Analysis. 17a. Descriptor*
Fluidized bed processors
coal
Combustion gases
Limestone
Sulfur Dioxide
Nitrogen dioxide
Hydrocarbons
17k. Ueetificrs/Open-Eoded Tertna
17e- COSATI Field/Group 13/B
18. Availability Statement
Unlimited
19. Security Class (This
Report)
Unclassified
21-"NoTof Pages
80
-------
46
RESULTS OF PILOT SCALE (FBC) TESTS - SINTERED ASH BED
1 SULFUR DIOXIDE EMISSION WITH COABSE ADDITIVES
Initial tests to investigate the S02 control potential
of the fluidized-bed corabustor were carried out with
the 1359 limestone and 1337 dolomite sorbent materials
ground and screened to a -7 +14 mesh, a size roughly
that ox the bed material. This size was selected in an
attempt to increase the residence time of the particle
in the bed and thus increase the sulfur capture. The
sorbents were used in the raw state and as calcined by
the supplier. The effects of bed temperature, bed
depth, sorbent feed rate and excess air (as determined
by the oxygen content in the flue gas) were investigated
initially in order to determine the optimum operating
conditions for sulfur retention. Three tests were con-
ducted with reducing conditions in the bed. 90% to 95%
of the input sulfur is emitted as sulfur dioxide without
sorbent addition.
The reductions in SO2 emissions observed in the FBC
with the coarse 1337 dolomite are shown in Table I and
the corresponding data for the 1359 limestone in Table II
Sulfur dioxide reductions observed in the FBC tests are
plotted as a function of stoichiometric ratio in
Figures 21 and 22 respectively. Sorbent utilization
percentages are given in Tables I and II; these are
obtained by calculating the average portion of calcium
in the sorbent feed which reacts with sulfur. Stoichio-
metric ratios were computed on the basis of 4.5% sulfur
in the coal and the calcium content of the sorbent. The
magnesium fraction in the dolomite was assumed to be
chemically inert. The stoichiometric ratio, designated
the Ca/S ratio in the tables, is the ratio of moles of
calcium in the sorbent fed to moles of sulfur in the
coal.
Comparison of results, presented in Tables I and II,
indicates that the dolomite is more effective in sulfur
capture than the high calcium limestone, based on the
calcium alone, and ignoring the magnesium fraction.
Sorbent utilization values of up to 35 to 40% were
observed with the dolomite whereas the limestone util-
ization was limited to a maximum of about 20%. One
contributing factor may have been the friability of
the dolomite. The dolomite tended to decrepitate dur-
ing calcination in the bed and was elutriated. The
limestone, on the other hand, tended to build up in
the bed. The dolomite, in breaking up, could have ex-
posed more surface per unit mass for the sulfur reac-
tion and combined with sulfur before leaving the bed.
The limestone, in retaining its particle size, would
expose less reactive surface per unit mass.
POPE. EVANS AND HCBBINS
SNIHHOH CINV onca:1 TUOci
*1
*1
*1
*1
GO
ffl
a)
w
to
to
a
to
CO
to
O
O
n
O
n
O
n
O
0
O
1—1
M
<£)
Cn
a\
(_n
U>
ro
O
~3
a it
o w
• rt
SO W » B) W »
qi cu Q) oj tu
« « t, € £ t,
?o ?c
Qi Q)
S Z
» 50
& QI Q>
S S £
50 50 50
Q) OJ QJ
a ?
0
50 50 50
0
Q> Cu Cu
cn
* z z
rt 0
CD 3
H-
r+
0)
M 1—1
H
1—1
H
1—' I-1
O O
O
0
0
O
0 0
VO
O"! 0> CTl
-J
-J
-J
-O
1—' M
(-¦
H1
t—•
\->
M M
M
1—' 1—' 1—•
M
H-*
M
M
h-*
1—'
ON CN
-O
CD
CO
CD
Cn
cn
c»
-vj
00
cn
-O
CO
CO
ro to
O
O
ON
CO
CO 0
0 cn co
cn
cn
O
cn
O
0
0
0 0
O
O
O
O
0 0
0
000
0
0
O
0
O
0
0
to ro
to
Co
u>
CJ u>
to
ui co co
i-1
1—1
CO
t-"
Co
r-»
0 0
0
O
O
0
0 0
O
000
0
0
O
O
O
0
0
M M
l-»
I-*
H
M O
M
to to M
h->
H
M
I-4
M
<71 O
0
-J
h-"
£» —J
O
UlOP
to
I-1
ro
ro
h-»
I-*
O O
0
O
O
O
VO
cr\ .c.
0
O
0
C3
O
O
£*
.£*
to
0
to
W
cn
CO
U>
Co
u>
to u>
ro
HHW
co
CO
CO
ro
CO
CO
co
LO
H1
O
CO
0 0
M
Cn vo f—'
H»
|-«
to
to
ro
cn Co
cn
Cn tsj
CO
to
1—
co
to
to
W -J
ro
CO
H*
0
to
Ln vj
to
to
cn
CO
to
Co
O
• •
•
•
•
•
• •
•
• ¦ •
•
•
•
•
•
•
O
O CO
V-1
co co
to to to
to
to
CJ
V-4
ro
3
«£>
to
o\
to
-j vo
cn
ON
O
to
to
00
M
cn
cr
00
cn
CO
to
«sj
ro
CO
• <+ Oi
3*
~3
o (0 00
*0 Cu
oc
r* (B
?d o
om
W
H- 3
O rr
3
Lf
-------
TABLE II. SULFUR DIOXIDE REDUCTIONS OBSERVED WITH ADDITION OF COARSE (-7 +14 MESH)
1359 LIMESTONE TO THE FBC BURNING A 4.5% SULFUR COAL
Test
Mo.
Limestone
State
Bed
Depth
In.
Bed
Temp.
°F
Flue Gas
O2 %
Ca/s
Ratio
S02 Concen. ppm
Initial Final
% S02
Reduction
% Limestone
FBC 19
Raw
8
1540
3.0
1.15
3900
3300
15.4
13.4
Raw
8
1540
3.0
1.93
2850
27.0
14.0
Raw
8
1540
3.0
1.931
2350
40.0
20.7
FBC 20
Raw
8
1700
3.0
1.15
3800
3350
11.8
10.2
Raw
8
1740
3.0
1.47
2750
27.6
18.8
Raw
8
1650
3.0
2.0G
2500
34.2
17.1
Raw
8
1650
3.0
2.00
2300
39.5
19.7
FBC 21
Raw
8
1520
3.0
0.90
3800
3300
13.2
14.7
Raw
8
1530
3.0
1.57
2920
23.1
14.7
Raw
8
1500
2.0
2.00
2900
23.6
U.8
Raw
8
1520
3.0
2.00
2600
31.5
15.8
FBC 22
Raw
8
1650
3.0
1.00
3900
3300
15.4
15.4
Raw
8
1540
3.0
1.90
2720
30.8
16.2
FBC 26
Calcined
8
1700
3.0
1.50
3800
3300
13.2
8.8
Calcined
8
1700
3.0
2.30
2950
22.4
9.7
Calcined
8
1700
3.0
3.4 0
2370
37.5
11.0
Recirculation
TABLE I. (Continued)
Test Dolomite
No. State
Bed Bed
Depth Temp.
In. °F
Flue Gas CA/S SO^Concen. PPm % SO2 % Sorbent
O2 % Ratio Initial Final Reduction Utilization
FBC
23
Calcined
8
1600
3.0
1.282
4180
2850
31.8
24.8
Calcined
8
1580
3.0
2.17 2
2700
35.6
16.4
Calcined
8
1580
3.0
2. 542
2050
51.0
20.0
Calcined
8
1580
3.0
2.17 2;3
1720
59.0
27.2
FBC
24
Calcined
7
1680
3.0
1.132
3500
2930
16.3
14. 4
Calcined
7
1680
3.0
1.732
2410
31.2
18.0
Calci ned
7
1680
3.0
2.26 2
2250
35.6
15.7
FBC
25
Calcined
6
1720
3.0
1.15
3650
3350
8.5
7.4
Calcined
6
1720
3.0
1.60
3150
12.9
8.1
Ash Recirculation
With water injection
-------
80
90
SUPERFICIAL VELOCITY: 12-14 FPS
BF-J TEMPERATURE: AS SHOWN
COAL: OHIO #8 PITTSBURGH SEAM,
UNWASHED, 4.5% S
EXCESS AIP: 3% 02 IN FLUE GAS
BED: SINTERED ASH,
8" STATIC DEPTH
100
1.0 2.0
Ca/S STOICHIOMETRIC RATIO
3.0
FIGURE 22.
REDUCTION OF SULFUR DIOXIDE EMISSION FROM THE FBC
BURNING A 4.5% S COAL VITH COARSE 1359 LIMESTONE ADDITION
SYMBOL SORBEHT
in
>'
7
in
¦>
z
0
a
o
a
o
X
o
k!
D
W
70
80
90
100
TEST CONDITIONS:
SUPERFICIAL VELOCITY: 12-14 FPS
BED TEMPERATURE: AS SHOWN
COAL: OHIO #3 PITTSBURGH SEAM,
UNWASHED, 4.5% S
EXCESS AIR: 1-4% 02 IN FLUE GAS
BED: SINTERED ASH
7-10" STATIC DEPTH
BED TEMP.,
°F
1500
1600
1700
1800
1580
1700
C = CALCINED,
R = RAW
SORBENT SIZE:
-7+14 MESH
1.0 2.0
Ca/S STOICHIOMETRIC RATIO
3.0
FIGURE 21. REDUCTION OF SULFUR DIOXIDE EMISSION FROM THE FBC
BURNING A 4.5% S COAL WITH COARSE 1337 DOLOMITE ADDITION
-------
52
Sulfur retention and sorbent utilization are seen to
increase slightly with flue gas oxygen content. An
increase was observed in FBC Test 5, 10 and 14 in
which other factors were held fairly constant. In
Test No. 5, conducted at a bed temperature of 1800°F
and a Ca/S ratio of 1.1 the sorbent utilization was
increased from 15.4% to 21.0%, when the oxygen content
was increased from 1% to 3%. Test No. 10 was initiated
with reducing conditions ir, the bed, i.e., with less than
stoichiometric air passing through the bed. The balance
to make up the 1% oxygen concentration in the flue gas
was supplied hy overbed air. The sorber.t utilization
increased from 20.4 to 26.8% when the bed condition was
changed from reducing to oxidizing. The results of
Test No. 14 indicated an increase in sorbent utilization
from 12.2 to 16.8% with increase in oxygen although the
improvement may have been partly due to 60°P drop in bed
temperature. This result-is reasonable inasmuch as
oxygen is required to retain sulfur in a more stable
form according to the relation:
CaO + S0Z t j 02 ~ CaSOj,
In all subsequent FBC tests the oxygen concentration in
the flue gas was maintained at 3% to improve sulfur
capture but more importantly to limit hydrocarbons emis-
sion, as discussed in Section 6.5.
Sulfur retention and sorbent utilization increase with
decrease in bed operating temperature to the lower end
of the operating range. This effect is evident from
the results of FBC Tests 5, 6 and 11 for the 1337 dolo-
mite. Under otherwise similar conditions the sorbent
utilization changed from 21.0 to 24.2 to 32.5 for re-
spective temperatures of 1800<>F, 1680"F and 1550°F. A
similar effect is noted in Table II for the 1359 lime-
s tone.
The effect of bed depth is less well defined because of
variation in other parameters. Interpolation of reduc-
tions and Ca/S ratios for Test No. 13 as shown in
Figure 23 indicates a reduction of 45% at a ratio of
1.2 with a 10-inch bed. Also shown is a reduction of
39% observed in Test No. 6 conducted at this ratio and
a 7-inch bed depth. The effects of bed depth and tem-
perature were similar with injection of fine sorbents
as indicated in Section 6.3.
POPE EVAMS AND BOBBINS
53
o
10
20
o
H
5 60
D
a
w
* 70
N
O
w
80
90
100 0 0.5 IO 1.5 ZO
Ca/S STOICHIOMETRIC RATIO
FIGURE 23. INTERPOLATION Or 10-INCH BED DEPTH DATA
FOR COMPARISON WITH 7-INCH BED DEPTH DATA
POPE EVANS AND ROBBIMS
-------
54
The calcined 1337 dolomite was less effective than the
raw stone as recorded in Table I for FBC Tests No. 23,
24, and 25. The performance of the 1359 limestone in
the -7 +14 particle size was likewise poor as shown in
Table II. A deep mined limestone from Northern West
Virginia was tested with a 5% sulfur coal from the same
mining area. The limestone, containing 72% CaCOj, and
screened to the -7 +14 mesh size, effected an S02 reduc-
tion of 36% at a Ca/S ratio of 1.4 (25.4% utilization).
The test conditions were 1550°F bed temperature and
3.0% O2 in the flue gas. Data for this test (FBC 27)
and others are summarized in Appendix B. Sulfur bal-
ances are shown in Appendix C.
6.2 SULFUR DIOXIDE EMISSION WITH FINELY DIVIDED SORBENTS
The investigation of sulfur dioxide emission control by
sorbent injection was redirected to the use of fine
sorbents in an effort to increase the reactive surface
of the sorbent for greater desulfurization.
The tests were conducted in the FBC with the following
considerations with respect to variables:
Sorbent Type: Two sorbents were tested, the
1337 dolomite and the 1359 limestone.
Sorbent State: The raw stone of each of the
two sorbents was ground to a -325 moch particle
size for the test series. Both sorbents were
also tested in the hydrated form which is com-
mercially available in a -325 mesh particle
size. Both the calcium and magnesium fractions
of the dolomites were hydrated. The sorbents
are designated 1337R, 1359R, 1337H and 1359H
to distinguish the raw and hydrated forms
respectively. One test was run with precalcined
limestone designated 1359C.
Sorbent Particle Size: The effect of variation
in particle size from -12 to -325 mesh was tested
with the 1359R limestone. Except for these tests,
reported in Section 6.3, the sorbent particle size
was -325 mesh.
Sorbent Feed Rate: The Forbent feed rate was
varied in the range of 1 to 3 stoichiometric ratio
based on the calcium content of the sorbent and
the sulfur content in the coal.
POPE EVANS AND ROBBINS
55
Sorbent Feed System: Three .methods of sorbent
feed were employed as discussed in some detail
in Section 5.3. These were (1) addition of
sorbent at the coal feed port 181 Feeder],
(2) injection of sorbent at two points away from
the coal feed port _[£2 FeederJ and (3) premixing
the sorbent and coal in the hopper. The 12 Feeder
system was modified for four-point feed in a test
of sorbent distribution. One test was conducted
with sorbent injection above the bed for compari-
son .
Coal Sulfur Content: The tests were conducted
with Ohio 18 Pittsburgh seam coal, unwashed and
washed containing respectively 4.5% and 2.6%
sulfur.
Flue Gas Oxygen Content: The oxygen content m
the flue gas was held constant at 3% since previ-
ous results, noted in Section 6.5, indicated this
value to be minimum for control of hydrocarbons
emission. Higher values contribute to loss in
thermal efficiency.
Bed Temperature and Depth: The bed temperature
was varied in the range of 1500°F to 1800°F to
investigate the temperature effect with fine sor-
bent particles. The bed depth was adjusted to
the greatest value consistent with bed temperature
A test acrico was conduotcd to invactigata tho in-
dependent effects of bed temperature depth and
particle size. The series is discussed in Section
6.3.
Ash Recirculation: Fly ash was recirculated on a
number of tests as a final test condition. The
rate was 80% of the collected ash.
Superficial Gas Velocity: The tests were con-
ducted with the superficial gas velocity held
constant within the range of 12 to 14 fps in
most tests. The effect of superficial velocity
was investigated as discussed in Section 6.3.
The test results indicated a marked improvement in
sulfu- dioxide reduction and sorbent utilization with
the fine sorbent as compared to the coarse sorbents
under similar conditions. The improvement was most pro-
nounced with the use of the 1359 limestone. A compari-
son is presented in Figure 24 showing the effect of
particle size change with both sorbents in the raw state
Test conditions included a 1500°F - 1600°F temperature
POPE EVANS AND ROBHINS
-------
CN V SNVAZl 3dOd
SULFUR DIOXIDE REDUCTION, %
57
range, 3% oxygen in the flue gas, a 34 fps superficial
velocity and a A.52 sulfur coal in each case.
The reductions observed with the fine sorbents, both
raw and hydrated, while burning the 4.5% sulfur coal
are shown in Table III. One tG&t of precalcined 1359
limestone is included. The reductions are plotted as
a function of Ca/S ratio in Figure 25 for bed tempera-
ture in the range of 1500°F - 1600°F. The trends show
the 1337 dolomite, again, to be more reactive Lhan the
1359 limestone when the magnesium fraction of the dolo-
mite is considered inert. The trend indicates further
that the hydrated forsn of the sorbents is as reactive
as the corresponding fine raw sdrbent. The .most favor-
able single observation was mad& with the dolomite
hydrate, designated 1337H. The reduction was 88% at a
Ca/S ratio of 1.8, the corresponding utilization being
47.2%. The average dolomite utilization based on the
trend line containing both the hydrate and raw form
data is indicated to be M5%.
The data trend in Figure 25 for the 1359 fine limestone
indicates a similar reactivity for both the hydrate and
raw forms, a lesser reduction th..n observed with the
dolomite (65% at a Ca/S ratio of 1.8) and utilization
decreasing with increasing Ca/S ratio. The utilization
varies from 40% at a ratio of 1.0 to 28% at a ratio of
3.0. By comparison, the coarse stone utilization
(Table II) did not exceed 20%.
The precalcined form of the 1359 limestone, designated
1359C in Table III and Figure 25, was less effective
than the same stone in the raw or hydrated forms. This
result may have been due to the possibility that the
supplier's conditions for calcination may not have pro-
duced as "soft" a calcine as the 1500°F fluidized-bed
environment. The limestone was calcined by the supplier
to optimize hydration, but the conditions were reported
to be proprietary and were not released.
Sulfur dioxide reductions observed with injection of
sorbents during combustion of a 2.6% sulfur coal are
summarized in Table IV. The percent reductions, as a
function of the Ca/S, ratio, were approximately the same
for this medium sulfur coal as for the 4.5% sulfur coal.
The po.'nts, taken at bed temperatures in the range of
1500°F - 1600°F, are plotted m Figure 26.
POPH EVANb AND .--eCBBINS
-------
TABLE Hi. (Continued)
Bed Bed Flue
esi_ Depth Temp. Gas Ca/S Feed SO? Concen. ppm
In- °F 02 % Sorbent Ratio System Initial Final Reduction Utilization
No
FBC
41
FBC
10
10
10
1560
.1600
1600
3.0
3.0
3.0
1359H
1359H
1359H
3.40
2.24
2 .10
#1
#2
PREMIX
3450
350
1000
900
90.0
71.0
74.0
26.5
31. 7
35.2
42
FBC
9
9
1640
1600
3.0
3.0
1359H
1359H
1.65
2.80
#2
PREMIX
3350
1150
600
65.5
82.0
39.6
29.3
56
FBC
10
: 580
3.0
1337R
1.12
#2
3550
1750
49.0
43.6
57
FBC
10
] 570
3.0
1337R
1.25
#2
3550
1400
60.7
43.5
53
FBC
10
1570
3.0
1337R
1.57
Tr 2
3600
1000
72.2
46 .0
61
FBC
10
1540
3.0
1359R
1.25
#2
3550
1900
46.5
37.3
62
FBC
10
1550
3.0
1359R
1.6
#2
3550
1500
57,8
36.0
64
12
1550
3.0
1359C
2.6
#2
3750
2200
41.2
15. S
12
1550
3.0
1359C
1.5
#2
3750
2560
30.8
21.1
TABLE III.
SULFUR DIOXIDE REDUCTION OBSERVED WITH FINE SORBENT ADDITION
TO COMBUSTION OF A 4.5% SULFUR COAL IN THE FLUIDIZED BED
rest
\To.
Bed Bed
Depth Temp.
In. cf"
Flue
Gas
O2 % Sorbent
Ca/S Feed S02 Concen. ppm % SO2 % Sorbent
Ratio System Initial Final Reduction utilization
32
a
1580
3.0
1359H
1.10
#1
3650
2200
40.5
36.8
TBC
33
9
1600
3.0
1337H
1.05
#1
3750
2000
46.7
44.5
FBC
38.4
39
9
1540
3.0
1337H
1.38
#1
3400
1600
53.0
9
15S0
3.0
1337H
1.87
jf2
400
88.2
47.2
9
1540
3.0
1337H
1.17
PREMIX
1650
51.2
43.6
FBC
40
7
174(1
3.0
1337H
1.55
#1
3550
1250
64.7
41.8
7
1720
3.0
1337H
2.08
#2
650
81.9
39.2
7
1760
3.0
1337H
1.17
PREMIX
1800
49.2
41.1
FBC
32
8
1580
3.0
1359H
1.10
#1
3700
2200
40.5
36.8
FBC
35
9
1560
3.0
1359H
1.04
41
3600
2200
39.0
37.5
9
1560
3.0
1359H
1.10
#2
1950
46.0
42.0
9
1560
3.0
1359H
2.15
#1 + #2
1400
61.2
28.5
FBC
36
9
1540
3.0
1359H
•1.16
#1
3550
1700
52.2
45.0
9
1580
3.0
1359H
1.36
#2
1700
52.2
38.4
9
1580
3.0
1359H
2.55
#1 + #2
800
77.5
30.4
FBC
37
8
1640
3.0
1359H
1.34
#1
3400
2000
41.2
30.8
8
1620
3.0
1359H
1.34
#2
2250
34,0
25.2
FBC
33
8
1760
3.0
1359H
1.75
#1
3500
1470
57.9
33.1
8
1720
3.0
1359H
1.45
#1
1780
49.2
33,9
8
172C
3.0
1359H
1.28
#1
1800
48.8
38.0
8
176C
3.0
1359H
1.16
#1
2030
42.0
36.2
-------
TABLE IV. SULFUR DIOXIDE REDUCTION OBSERVED WITH FINE SORBENT ADDITION
TO COMBUSTION OF A 2.6% SULFUR COAL IN THE FLUIDIZED BED
3ed Bed Flue
Test Depth Temp. Gas Ca/S Feed SO; Cone, ppm % SO^ % Sorbent
. In. °F °2 % Sorbent Ratio System Initial Final Reduction Utilization
>
Z
0
£
0
0
s
7
F3C
47
?3C
43
rsc
49
rBC
50
r3c
51
11
1600
3.0
1359H
3.60
#1
2200
270
87.7
2 4.4
11
1640
3.0
1359H
2.60
#2
550
75.0
28.3
11
1650
3.0
1359H
2.10
PREMIX
760
65.5
31.2
6
1780
3.0
1359H
2.00
#1
2200
1500
31.8
15.9
6
1300
3.0
1359H
1 .40
*2
1700
22.7
16.2
6
1800
3.0
1359H
2.60
PREMIX
1400
36.3
13. 9
12
1505
3.0
1337H
1.40
#1
2000
800
60.0
42.8
12
1600
3.0
1337H
1.60
#2
500
75.0
46.9
12
1590
3.0
1337H
1.16
PREMIX
1000
50.0
43.0
6
1780
3.0
1337H
1.60
#1
2350
1400
40.4
25.2
6
1770
3.0
1337H
1.46
u
1550
34.0
23.3
6
1790
3.0
1337H
1.16
PREMIX
1550
34.0
29.2
12
1570
3.0
1337H
1.35
#1
1880
860
54.2
40.1
12
1580
3.0
1337H
1.55
#2
490
73.9
47.7
12
1570
3.0
1337H
1.46
PREMIX
910
51.6
35.4
12
1580
3.0
1337H
1.40
#1
2100
1100
47.5
34.0
12
1600
3.0
1337H
1.60
#2
700
66.7
41.7
12
1600
3.0
1337H
1.4 6
PREMIX
1110
47.5
32 . 5
u
0
tl
p]
><
Z
01
>
z
0
XI
0
ffl
s
z
01
STONE CONDITION;
c = CALCINED
H = HYDRATE
K = RAW
FIGURE 25.
1.0 2.0
Ca/S STOICHIOMETRIC RATIO
SULFUR DIOXIDE REDUCTION WITH FINE SORBENT ADDITION TO THE FBC
BURNING A 4.5% SULFUR COAL
-------
SYMBOL SbRBENT MESH SIZE
100
Ca/s STOICHIOMETRIC RATIO
FIGURE 26.
SULFUR DIOXIDE REDUCTION WITH FINE SORBENT ADDITION TO THE FBC
BURNING A 2.6% SULFUR COAL
table IV. (continued)
Bed
Bad
Flue
Test
Depth
Temp.
Gas
Ca/S
Feed
SO2 Cone,
. ppm
% S02
% Sorbent
No.
In.
Op
O, %
Sorbent
Ratio
System
Initial
Final
Reduction
Utilization
FBC
53
11
1560
3.0
1337R
2.00
92
2050
550
73.5
36.7
11
1560
3.0
1337R
1.20
#2
2050
1060
48.0
40.0
FBC
59
10
1590
3.0
1359R
2.33
#2
2350
650
72.4
31.0
60
10
1550
3.0
1359R
2. 33
#2
2300
900
60.8
30.4
10
1550
3.0
1359R
1.20
#2
2300
1250
46.0
35.0
-------
66
TABLE V. DATA SUMMARY FOR S02 REDUCTION VS. 1359 LIMESTONE
PARTICLE SIZE, BED DEPTH AND TEMPERATURE
FBC Bed Bed
Test Depth Temp.
No. Inches °F
Particle
Size Stoich. SO? Cone. ppm * SO2
Microns Ratio Initial Final Reduction
65
10
1540
1680
2.6
2500
1870
28.0
1530
1410
2.6
2500
1870
28.0
1530
1000
2.6
2500
1700
34.0
66
10
1530
840
2.4
2500
1700
34 .0
1530
420
2.6
2500
1940
24.0
1550
44
2.6
2500
760
72.0
67
18
1520
840
2.5
2450
1250
49.0
1580
420
2.5
2450
1600
34.6
1550
44
2.6
2400
550
77.0
68
10
1770
840
2.6
2550
2150
15.5
1810
420
2.8
2500
2230
12.5
1770
44
2.7
2500
1700
31.0
69
18
1770
840
2.7
2500
2150
15.1
1750
420
2.6
2500
2230
11.0
1700
149
2.7
2500
1700
32.0
1750
44
2.5
2500
1700
32.0
70
10
1520
149
2.8
2500
1750
30.0
10
1850
149
2.8
2500
2100
13.7
18
1550
149
2.8
2500
1250
50.0
18
1830
149
2.8
2500
1900
23.3
71
10
1600
840
2.6
2500
1750
30.6
1620
840
2.6
1800
28 .2
1650
840
2.6
1950
22.0
1670
840
2.6
2050
18.0
10
1670
420
2.8
2580
2100
18.6
1690
420
2.8
2250
13 .0
10
1630
149
2.7
2580
2000
22.5
1620
149
2.7
1900
26 .2
1670
149
2.7
2100
18.6
10
1660
44
2.6
2620
1350
49.0
1670
44
2.6
1450
45.0
3'.a:j= and rob:?
67
TABLE V. (Continued)
C
st
0.
Bed
Depth
Inches
Bed
Temp.
°F
particle
Size
Microns
Stoich.
Ratio
SO2 Cone.
Initial
ppm
Final
% S02
Reduction
72
18
1700
840
2.6
2530
1800
29.0
1640
840
2.6
1650
34 .7
1610
840
2.6
1500
41.0
18
1660
420
2.7
2530
2000
21.0
1650
420
2.7
1900
25.0
1640
420
2.7
1850
26.9
18
1700
149
2.6
2530
1900
25.0
1640
149
2.6
1700
33.0
1610
149
2.6
1600
36.7
18
1660
44
2.8
2530
1200
52.5
1650
44
2.8
1100
56 .5
-------
64
The data trends indicate that the fine dolomite (133/)
is again more reactive than the fine limestone (1359).
The dolomite hydrate proved to be somewhat more reactive
than the fine, raw stone. Utilization of the fine raw
dolomite is 42% and 37% at respective Ca/S ratios of
1.0 and 2.0.
The 1359 limestone hydrate appears to be as reactive as
the fine raw stone when used with the 2.6* sulfur coal.
Utilization of the fine raw stone indicated by the trend
is 38%, 32% and 27% for respective Ca/S ratios of 1.0,
2.0, uiid 3.0.
A comparison of the method of sorbent feed into the FBC
failed to point up a clear advantage for any particular
method of sorbent feed although, in general, the most
favorable observations were made with the S2 feed system.
Test data for the series are summarized in Appendix B.
Sulfur balances are presented in Appendix C.
6.3 TESTS FOR INDEPENDENT EFFECTS OF BED TEMPERATURE, BED
DEPTH, SORBENT PARTICLE SIZE, SORBENT DISTRIBUTION, AND
SUPERFICIAL VELOCITY
A statistical experiment was conducted to establish the
separate effects of bed temperature, bed depth and sor-
bent particle size on the desulfurization reaction in
the fluidized bed. The information provided by the
experiment was intended to form a basis on which to es-
timate the necessity for fine grinding and to establish
the relative advantage of more massive beds which must
be supported by added fan power.
The experiment was conducted in the FBC after modifica-
tion to permit control of bed temperature with a movable
internal cooling surface. The modification is described
in Section 5.1. A sintered ash bed, sized -7 +14 mesh,
was fired with Ohio #8 seam washed coal which, in this
case, contained 3% sulfur. The 1359 limestone was se-
lected as the sorbent because of its apparent durability
observed in previous tests. The sorbent was injected
with the J 2 Feeder system described in Section 5.3. The
sorbent feed rate was controlled as closely as possible
to a stoichiometric ratio of 2.6, a ratio estimated to
yield an 80% SO2 reduction with the -325 mesh particle
size.
The 1359 limestone was prepared in seven sizes ranging
from 12 mesh to -325 mesh. These size groups were
-12 +14, -14 +16, -18 +20, -20 +30, -40 +50, -100 +200
and -325 U.S. Standard Mesh. The particle sizes repre-
sented by the largest screen size in these ranges
POPE. EVANS AND BOBBINS
65
correspond to 1680, 1410, 1000, 840, 420, 149 and
44 microns respectively. The first three were tested
at a single test condition for reference, while the
last four were tested over the range of temperature
and bed depth. The elutriation particle size, i.e.,
the smallest particle size that remains in the bed at
the 14 fps superficial gas velocity was 30 mesh. This
size was estimated from the intermediate law as shown
in Appendix A, Enclosure 18.
The bed depth was varied at two levels—10 inches and
18 inches. The bed temperature was varied in three
levels, one value at the extreme ends of the operating
range 1500°F - 1800°F and one intermediate temperature.
The results of the test, indicating the reduction in S02
with sorbent particle size, bed temperature and depth
are summarized in Table V and plotted in Figure 27.
The data trends suggest the following conclusions:
a. A sorbent ground to pass through a 200 mesh screen
can be expected to be much less effective than
sorbent ground finer so as to pass through a
325 mesh screen.
b. The reduction-particle size curve appears to pass
through a minimum reduction in the particle size
range of -40 +50 mesh (420 microns). Such a mini-
mum might occur from loss in bed residence time
without a compensating increase in reactive surface.
c. Increase in bed depth (and residence time) is less
effective with the -325 mesh particle than with
larger sizes. In every case the advantage of in-
creased residence time declines as the bed tempera-
ture is raised from 1550°F to 1800°F.
d. All particle sizes are more effective in sulfur
capture at bed temperatures of 1550°F than at 1800°F
This result is consistent with thermodynamic equilib
rium data reported by others3 and with performance
observed in the regeneration of limestone beds
(Section 6.9).
1Battelle Memorial Institute, "Fundamental Study of Sulfur
Fixation by Lime and Magnesia," June 30, 1966
POPE EVANS AND ROEBINS
-------
SNISBOH dMV SNVA3 HdOci
89
69
Other data are summarized in Appendix B. Emissions of
nitric oxide and hydrocarbons were the same as observed
in previous tests (Sections 6.5 and 6.6).
SORBENT DISTRIBUTION TEST
A test procedure was carried o'lt to investigate the
possibility of improving the desulfurization efficiency
by better distribution of the sorbent around the periph-
ery of the FBC. For the test procedure a second two-
point feeder was placed on the side of the FBC opposite
to the first as shown in Figure 16, The two feeders
were then connected to the FBC by pneumatic tubes which
would permit injection of sorbent on one, two and all
four sides. The feeders were calibrated precisely and
the feed rates adjusted to maintain a constant sorbent
feed into the bed as the number of injection points was
increased. The 1359 limestone, in a -325 mesh size, was
injected into the FBC at first one, then two, and then
four sides with a constant rate of 2.0 stoichiometric
ratio.
Emissions monitored during the test are shown in Figure
28. The results failed to show c.i improvement with in-
crease in the number of injection ports. At the end of
the test the Ca/S ratio was increased to 3 to check for
a possible defect in the instrumentation which might
have prevented a variation. The decline in SO2 emis-
sion at the higher ratio indicates normal functioning
of the instrument.
The results indicate that single-point injection in the
FBC is adequate to effect the optimum S02 reductions
for the bed volume. For the larger bed volume in the
FBM, the results suggest that distribution may not be a
problem. The two-point injection appeared to be ade-
quate in the FBM but a similar distribution test was
not made.
In a subsequent test the -325 mesh limestone was in-
jected above the bed for a comparison of the SO2 control
effectiveness with the inbed injection. Test conditions
were otherwise the same as employed in the distribution
tests. The coal and sorbent feed rates were held con-
stant as the sorbent feed was diverted from above the
bed to the base of the bed.
The results showed a marked loss in effectiveness of
capture when feeding the sorbent above the bed as com-
pared to the usual in or below the bed feeding.
POPE EVANS AND HOB3INS
-------
SNISQOTH CUMV SNVA3 5dO:
BEO TEMR,
°F
.EMISSION, LSS PER 1Q BTU IfiPUT
- HYDROCARSONS (HC) AND KiTRlC OXIDE (NO)
a
G
H W
z s
C-; H
M W
O CO
H H
M O
3 2
<7> cn
» ^
ui 50
to O
ui 3
3C T3
W to
co n
a
- ^
w
I—1 w
u» h3
i_n
«x» Z
O
It1 •
PI OJ
cn
H ®
O G
w !a
H
o z
z n
a c
d s
•t- cn
G
01 f
H
D G
M W
m
n
o
Oi
71
The results are summarized as follows:
Test Condition
No sorbent input
Sorbent, above bed
Sorbent, base of bed
Ca/S
Ratio
1.75
1.75
S02
Cone.,
PPm
2600
1750
1000
Reduc.,
%
29
62
Limestone
Utilization,
%
16.6
35. 4
Emission curves for this test (FBC 75) are presented
in Appendix A, and other data are summarized in Appen-
dix B.
SUPERFICIAL GAS VELOCITY TESTS
Superficial gas velocity is defined as the flue gas
velocity which would exist in the combustion unit at
the operating temperature without the fluidized-bed
material. This parameter is directly related to heat
release rote, a factor which marks a principal advan-
tage of the fluidized-bed combustion prococc ovor other
methods of firing. Operation of the fluidized-bed
boiler at less than maximum heat release rate (and maxi-
mum gas velocity) would not be beneficial unless an ad-
vantage with respect to sulfur emission control could
be demonstrated. This control should improve with in-
creased sorbent residence time afforded by a reduction
in gas velocity.
Tests were conducted in the FBC to investigate the ef-
fect on sulfur dioxide emission when the superficial
gas velocity was reduced from 13 to 6 feet per second
without change in the stoichiometric sorbent feed rate.
The finely divided limestone wac injoctod into tho bod
through the #2 feeder at a Ca/S ratio of 2.7. As the
coal and air rates were reduced to effect the lower
superficial velocities, the sorbent feed was reduced
in proportion to maintain the Ca/S ratio. A sintered
ash bed, 10 inches deep, was operated at 1550°F with
3% oxygen in the flue gas. The test was repeated at a
Ca/S ratio of 2.0, a position on the curve where the
sorbent utilization is greater. The test results, sum-
marized in Table VI, indicate little or no improvement
in S02 reduction or limestone utilization when the
superficial velocity was decreased. Emission curves for
these tests (FBC 76 and 77) are included in Appendix A.
POP'S. EA/AK'S AND ROBBINS
-------
SNreaoH qnv snlv/vh HdOd
I »"3
Z (I)
0 W
• f*"
fo CO «-•
n
<7» 00 O NJ M
U H Ol 03 CD
<
ft) Hi W
r, H- C
l o o *o
l 0 H- o
H- 0) hj
ft W 1
H 33 3 H
cr fu (D u>
\ ft 01 cn
3* (0 ft v©
H O 50
3
ft
U1 O U1
to N) to to
1 f?
0
• H-
o
sr
Co to to O
o*\ U*r i»n i»n
n
T3 O CO
"0 3 O
3 0 to
50
fD
a
d co
dp O O
ft to
U U) Co u
to to to to
G ,
ct H O
H- H- ft>
H 3 M
H-(T> O
dP N CD C
P ft H
(TOP)
ft
o
-------
74
1680°F
(400
™ I200
1000
800
600
400
200
BED DEPTHS INDICATED 3\' SY 'BOL
BED DEPTH
5 IMCKES
INCHES
10 INCHES
12 INCHES
SUPERFICIAL VELOCITY:-" 12-14 FPS
BED TEMPERATURE: 1530-li3S0°F
(AS SHOVTN)
COAL: OHIO H8 PITTSBURGH SEAM,
WASHED, 2.6% S, 7.2% ASH
BED: SINTEP.ED ASH
LO 2.0 aO1530°F 40
OXYGEN CONCENTRATION It; FLUE GAS, %
FIGURE 29. HYDROCAR30NS VAPIATION WITH FLUE GAS
OXYGEN CONCENTRATION IN TIT.F. FBC OPERATION
POPE EVANS AND ROBB1NS
75
bed depth have a negligible effect in comparison with
the oxygen concentration, and that hydrocarbons can be
limited to t-50 ppm at this value.
A flue gas oxygen content of 3% corresponds to an excess
air rate of approximately 17% above the stoichiometric
requirement. Any air in excess of the stoichiometric re-
quirement will result in a thermal loss chargeable to
the boiler since heat transfer surface cannot be econom-
ically provided to recover heat by cooling flue gas below
about 250°F. In addition, excess air removes heat from
the bed which must be recovered, in part, by convective
heat transfer surface which is less effective than in-bed
heat transfer surface. For this, as well as other reasons,
e.g. larger fans, increasing the excess air requirement
increases the capital cost of the boiler system.
The thermal loss due to excess air is partially compen-
sated for by the energy released by burning the hydro-
carbons to the 50 ppm level. So, for example, where the
excess air is increased from 5% to 17% and the flue gas
exits at 400°F, an efficiency loss of about 0.8% is in-
curred due to excess air while the hydrocarbons, assumed
to be methane, drop from 800 ppm to 50 ppm. The combus-
tion of the hydrocarbons releases an additional 110 Btu
per pound of coal fed for an efficiency gain of %0.9%.
In this example the optimum operating point might be
around 10-12% excess air, if maximum thermal efficiency
were the only goal. Most of the tests in the program were
made at 17% excess air, primarily because of the hydro-
carbon emission. Concentrations of 50_j?pm at this level
correspond to i*.02 lbs of methane per MBtu input. This
emission would appear to be favorable in comparison with
conventional boilers, but data on the latter operating at
the same excess air level are lacking. Further work on
coal feeding systems may provide a basis for lower excess
air operation without an increase in hydrocarbons.
Injection of sorbent materials into the bed does not
increase hydrocarbons emissions in steady-state opera-
tion. Sorbent injection at rates as high as 60 lbs per
100 lbs of coal failed to show a significant increase in
hydrocarbons emission.
Hydrocarbons generated by low excess air or reducing
conditions in the bed can be burned effectively by in-
jection of air above the bed in sufficient quantity to
make up the 3% oxygen content in the flue gas. This
result is discussed further in Sections 6.8 and 6.9.
Carbon monoxide concentrations in the flue gas from the
FBC may be as much as 0.5% at 1% oxygen content but are
negligible at the 3.0% oxygen level.
POPE, ETvWNS AND ROBBING
-------
76
TEST CONDITIONS:
SUPKUFICIAL VELOCITY: 1<1 FPS
BED TEMPERATURE: 17 50°F
BED: 8" STATIC DEPTH
COAL: OHIO #8 PITTSBURGH SF.AM,
UNWASHED
SORBENT FEED RATE: ZERO
500
g 400
Oi
300
200
100
1-0 2j0 3.0 4.0
OXYGEN CONTENT IN FLUE GAS, %
5.0
FIGURE 30. TYPICAL VARIATION IN NITRIC OXIDE CONCENTRATION
WITH OXYGEN COMTFNT IN THE FLUE GAS FRO:: TKE FBC
POPE EVANS AND ROBBINS
77
6 OXIDES OF NITROGEN EMISSION
Emission of nitric oxide from the FBC was found to vary
with the oxygen concentration in the flue gas as deter-
mined by the excess air rate. Nitric oxide in the flue
gas was found to increase from 320 ppm at 1.0% oxygen
content to 440 ppm at a 5.0% oxygen content. This varia-
tion is shown in Figure 30_together with the emission in
terms of pounds of NO per MBtu input. The emission at
3% oxygen content is 0.30 lbs per MBtu input.
In a number of tests conducted at 3% oxygen content in
the flue gas the nitric oxide concentration varied from
220 ppm to 470 ppm with no apparent correlation with
bed temperature. Data points observed when burning a
4.5% S, 2.5% N2 coal with 3% O2 in the flue gas are
shown in Figure 31. Theoretical curves are also pre-
sented in the figure to show the thermodynamic equilib-
briuin concentrations of nitric oxide that should exist
for the oxygen concentrations that exist across the
bed, i.e., 20% O2 in the inlet air and 3.0% O2 in the
flue gas, and for the range of temperatures investigated.
The shaded area in the figure is the area in which the
data would theoretically be expected to fall. F-ar the
method used to produce the theoretical curves see
Appendix A, Enclosure 19.
The figure shows that NO concentration should not ex-
ceed 100 ppm at a bed temperature of 1550°F. The fact
that concentrations of 300 to 400 ppm were observed
suggests the presence of local temperatures around the
coal higher than those observed by the bed thermo-
couples. Another possibility is that nitrogen in the
coal may play a role in the reaction. One test con-
ducted with two coals of different nitrogen contents
is discussed in the FBM test results (Section 7.3).
Nitric oxide emissions from the FBC appeared to be
unrelated to bed depth at the 3% oxygen concentration
level. Variation in bed depth during FBC Test 4 4
produced the following results:
Bed Depth 5 in. 8 in. 12 in.
02 Cone ,} % 123 123 123
NO Cone., ppm 280 340 380 305 360 400 360 370 380
As a rule, the use of sorbent materials was observed
to have little or no effect on nitric oxide emission.
Steady state concentration values were found to de-
crease and increase with sorbent injection. In two
instances, however, a definite reduction was observed.
POPE EVANS AND HOBSL-;;.
-------
78
BED TEMPERATURE, °F
FIGURE 31. MEASURED VALUES OF NITRIC OXIDE CONCENTRATION
IN THE FLOE GAS AT 3% OXYGEN AND VARIOUS BED
TEMPERATURES SHOWN WITH THEORETICAL EQUILIBRIUM
VALUES FOR THE TEMPERATURE - 02 CONTENT REGIME
EVANS -A1MID ROSeiutc
79
During FBC Test 18 conducted with the unwashed coal,
a sintered ash bed 14 inches deep and operating at
1760°F with 2% O2 in the flue gas, the NO concentra-
tion was reduced from 250 ppm to 60 ppm when 1337 raw
dolomite in a -7 +11 mesh ciac was injected through
the #1 feeder at a Ca/S ratio of 1.75. A careful exami-
nation of the instrumentation failed to roveal a defect
which might have caused the reduction. The reduction
was real but its cause undetermined. A similar effect
vas observed in FBC Test 25.
Although the discussion has been directed to the emis-
sion of nitric oxide, NO, the results are applicable
to total oxides emission, NC*. Tests tc determine all
the cnidea by the phenoldiculfonic acid procedure indi-
cated approximately the same concentrations as the
infrared absorption unit which ic conGitivc to NO only.'
Concentrations of the DJiidcG of nitrogen higher than
nitric oxide are estimated by difference to vary in
the range of 10 to 30 ppm.
On the average/ the nitrio ejtido omiccion from tho FBC
is approximately 0.30 lbs/MBtu input at the 3% oxygen
content in the flue gao. The corresponding concentra-
tion is 375 ppm.
.7 PARTICULATE EMISSION
>jost of the ash- frora the coal burned in the FBC was
elutriated as fly ash from the bed and collected in a
cyclone. The location of the cyclone in the test
assembly is shown in Figure 6. Isokinetic samples
taken do\m3tream of the oyclono indicated that up to
10% of the fly ash was discharged from the system.
This high particulate loss reflects principally the
poor collection efficiency of the toct cyclone with
the fine ash.
Klieu LIiO finely divided oorbento vere added to the
system, the particulate emission was increased.
Typical emission data with and without sorbent addition
are summarized as follows:
Computed
Ash Input
lb/hr
1359
Limestone
Input
lb/hr
Fly ash
Collected,
lb/hr
Fly ash
Discharged
lb/hr
12.8
12.9
12.6
12.9
0
0
21.4
28.0
22.0
23.2
41.0
43.4
1.5
2.4
3.9
4.9
POJPE. EVANS AND HOBBINS
-------
80
Although the particulate emission is increased by sor-
bent addition, the results show that the bulk of the
sorbent is retained in the collector despite the -325
mesh particle size.
The particulate emission from the FBM cyclone during
a, similar sorbent test was counted by microscope for
particle size distribution. The results are discussed
in Section 1. rj .
The fact that the total fly-ash rate is larger than
the computed ash ir.put, without sorbent addition, is
due to the presence of unburned carbon in the fly ash.
The fly-ash carbon content may vary from 45% to 60%.
When sorbent is added to the system, the fly-ash
carbon content is reduced to about 30% apparently from
dilution with the spent sorbent.
The energy lost from unburned carbon in the fly ash
amounts to about 10% of the input energy. Recovery
of this energy through the use of the Carbon-Burnup Cell
concept is now under investigation. The energy can be
recovered to some extent with recirculation of the fly
ash through the combustor. Recirculation of the fly
ash containing spent sorbent improved sulfur capture
in some instances but the results were inconsistent.
6.8 OPERATION AT REDUCING CONDITIONS
Three tests were made in the FBC with the bed at
slightly reducing condition. The reducing conditions
were produced by stabilizing the combustion at 1%
oxygen content in the flue gas and then decreasing the
air rate by 10% with constant coal feed. Since the
1% flue gas oxygen content corresponds to 5% excess
air (From Figure 29), an air rate reduction of 5%
would effect stoichiometric conditions. A reduction of
an additional 5% in the air rate produces a 5% defi-
ciency of oxygen in the bed. After the 10% air reduc-
tion, air was supplied above the bed to reestablish
the oxygen concentration in the flue gas at 1%.
For the effect on sulfur control, 1337R dolomite was
added at a Ca/S ratio of 1.1 during the reducing con-
dition. When the sulfur dioxide concentration dropped
to a lower steady-state level, the operation was re-
verted to the oxidizing condition without change in
the coal or sorbent feed rates. Nitric oxide and
hydrocarbons were monitored continuously.
POPE EVAN.= 4ND HOPBIMS
81
The results of the tests are summarized in Table VII.
Sulfur dioxide reduction is shown to improve with
the oxidizing bed (31.2% vs 21.4%) whereas NO reduc-
tion is favored by reducing conditions in the bed
(240 ppm vs 320 ppm concentrations). Hydrocarbons
concentrations appear to be greater with reducing
conditions, but the difference observed may have been
due tr. very small changes in tha oxygen content. The
rapid variation of hydrocarbons emission with flue gas
oxygen at the 1% level was discussed in Section 6.D.
Subsequent teats conducted at 3% O2 in the flue gas
indicated a more effective reduction m NO emission
with reducing conditions. At this oxygen level, hydro-
carbons can be consumed with overbed air. These points
are discussed in Section 6.9.
6.9 FBC OPERATION WITH A LIMESTONE BED
6.9.1 General
The FBC was operated with a bed consisting entirely
of 1359 limestone instead of inert ash. Emissions
were monitored from the comt stion of a washed
#8 Pittsburgh Seam coal in the bed, and the parameters
affecting sulfur retention were investigated. Removal
of sulfur retained in the bed was also studied. The
overall heat transfer coefficient was determined for
comparison with the value observed with the sintered
ash beds.
Initial attempts to fire a bed of limestone in the
FBC led to problems in bed temperature control. The
weight loss and endothermic heat requirement of calci-
nation and the rapid heat removal combined to create
an unstable situation. When the bed became calcined,
the bed temperature increased causing attrition losses.
The loss of bed in turn reduced the heat removal rate
and further increased the temperature. The operation
could probably have been stabilized by trial-and-error
addition of limestone. It was decided, however, that
an independent means of temperature control would solve
the problem and provide a desirable control capability
during the investigation. The independent temperature
control was effected with a sleeve installed in the
FBC to retard the heat transfer through the walls and
a mo/able coil installed in tne bed. This modifica-
tion was discussed in Section 5.1 and the coil and
sleeve arrangement shown in Figure 7. The bed tempera-
ture was controlled by adjusting the vertical position
of the coil m the bed.
-------
82
TABLE VII. DATA SUMMARY FOR OPERATION AT REDUCING CONDITIONS*
FBC Test No.
Bed Temp. °F
SO? Reduction^
Reducing Bed
Oxidizing Bed
MO Concentration, ppm
Reducing Bed
Oxidizing Bed
HC Concentration, ppm
Reducing Bed
Oxidizing Bed
1Addition of 1337R dolomite -7 +14 mesh at 1.1 ratio
with 4.5% Sulfur coal
* Flue gas oxygen content 1.0% for all conditions
See text for further description of test conditions.
6
1800
17.8
22.6
280
330
Erratic
500
9
1800
16.9
NA
220
NA
NA
NA
10
1750
21.4
31.2
240
320
560
435
POPE, EVANS AND ROBBINS
83
The sleeve and coil arrangement permitted the use of
deeper beds, i.e., 20" as compared to 10" in the
open unit. It was found convenient to ignite the bed
with a 10" depth and then add an additional quantity,
even though the whole of the bed could have been ig-
nited and stabilized. Positioning the coil provided
the fine adjustment of temperature.
The 1359 limestone was selected because of its appar-
ent durability and screened to a -10 +20 mesh particle
size. This partir.le size selection is somewhat
smaller than, the sintered ash bed size 1-7 +14 mesh)
because of the greater density of the raw limestone
C2.6 vs 1.8 specific gravity). The particle size
distribution is shown in Appendix A, Enclosure 20.
Typically 75 lbs of the limestone made up the original
charge with an additional 60 lbs added for an initial
raw bed weight of 135 lbs and a depth of 16 to 17
inches.
6.9.2 Sorption of Sulfur
The tests were made with Ohio #8 (Pittsburgh) Seam
washed coal containing about 3% sulfur. The coal was
fired at a rate of "*65 lbs per hour into a bed having
an initial weight of ^135 lbs. No sorbents were added
other than the bed limestone. The superficial velocity
was maintained at the same level employed in the sin-
tered bed operation, i.e., 12-14 fps. Concentrations
of sulfur dioxide, nitric oxide and hydrocarbons in the
flue gas were monitored continuously and spot samples
were taken for sulfur trioxide and oxides of nitrogen.-
All sorption tests were conducted at 3% oxygen in the
flue gas unless otherwise noted.
The bed operating temperature was found to be important
with respect to sulfur retention. At 1400°F the lime-
stone did not calcine and consequently did not retain
sulfur. At 1900°F the retention was minimal as expected
from previous work (Section 6.3, Figure 27). The
temperature range of 1500°F - 1600°F appeared to be
most favorable for sulfur sorption.
The results of sorption tests indicated that sulfur in
the coal could be sorbed almost completely for a period
of two to three hours after which time sulfur dioxide
began to appear in quantity in the flue gas. This be-
havior is illustrated in Figure 32 which shows the
emissions monitored during FBC Test 113, one of the
best of the program. The figure also shows the sulfur
input in equivalent sulfur dioxide emission. The bed
POPE EVANS AND P.CHEINS
-------
SNiaaO^d QMV SW VA3 HdCJ
C5
G
s
O M
gg
r CO
w
H M
Z O
> CO
hp) D
tT" G
G »
H H
a z
h n
N
w *1
?8
w
w ^
D M
w
O H
^3
vo w
ci
2 n
W 2
cn o
4
O >
2
M 3
c
It1
G
JO
85
temperature curve shows a variation m the range of
1500 F to 1600 F. The oxygen content was fixed at 3%.
The initial bed weight was 136 lbs and the depth
17 inches- Emission curves for other tests are pre-
sented in Appendix A.
The variation in sulfur and calcium contents of the
ber< for Test 113 is shown in Figure 33 together with
the sulfur and calcium contents of the fly ash. The
sulfur content of the bed had increased to 7.4 wt. %
at end of the test. This value indicates that lb% of
the bed limestone had been utilized in sulfur capture.
The increase in calcium content of the bed is due to
the weight lost in calcination.
The sulfur contents of the fly ash indicate that a
small part of the sulfur is retained in the fly ash.
The rate of sulfur flow in the system was indicated
to be the following:
SULFUR RATE, LBS/HR
Test Time, Hours
Flue gas output
Fly-ash output
Bed retention
Total output
Input
Input less
output
1
2
0.00
C . 08
0.35
0.28
1.56
1.64
1.91
2.00
1.98
1.96
0.07
-.02
_3 _4
0.60 0.70
0.32 0.38
0.90 0.80
1.82 1.88
1.98 1.98
0.16 0.10
The fact that sulfur in the fly ash remained relatively
constant suggests that this sulfur is contained in -the
fly-ash particle core and is not affected by the bed
reaction.
Attrition loss of the bed material was found to be
high during the calcination phase but comparatively
low afterward. During calcination, 5% to 7% of the
calcium in the bed was lost per hour. The calcium
loss during subsequent sorption was reduced to a rate
of 2% to 4% of the initial calcium charge. These
values approximate the loss during regeneration to be
discussed in Section 6.9.3.
Loss of unburned carbon in the fly ash during the
limestone bed tests indicated substantially the same
loss observed with the sintered ash bed, i.e., 9% to
12% of the input energy. Typical variation in fly-
ash carbon loss is shown in Figure 34. Heat transfer
measurements in the limestone bed indicated a coeffi-
cient of 47.0 Btu/ft'hr°F, about the same coefficient
observed in the sintered ash bed operation.
POPE EVANS AND ROBBIMS
-------
FIGUFS 34. EMISSTONS DURING FBC TEST 119 BURNING A MEDIUM SULFUR COAL IN
A LIMESTONE 3ED WITH MILD REDUCING CONDITIONS AND REGENERATION
POPE, EVANS AND ROBBINS
SM
«
D
Cm
.-5
D
W
70
60
50
40
I 30
H
a
S 20
CaM
SF
CaF
CURVE IDENTIFICATION:
CaF
S BED
CALCIUM
CaF
= FLY
ASH CALCIUM
SM
= BED
SULFUR
SF
= FLY
ASH SULFUR
2 3
TEST PERIOD, HOURS
FIGURE 33. VARIATION OF SULFUR AND CALCIUM IN BED AND FLY ASH DURING
FBC TEST 113
-------
88
Results of the stable limestone bed tests are summar-
ized in Table VIII showing the time observed for 20i
breakthrough of sulfur dioxide above the bed and other
data. The sorption of sulfur is shown, by the compari-
son, to be favored by increase in the bed mass (and
depth), increase in the oxygen content of the flue gas
and by a low bed temperature in the range of 1500°F to
16003F. At lower temperatures, sorption may be limited
by failure of the.bed to calcine. Reduced sorption at
the higher temperature is consistent with results of
previous tt>sts (Section 6.3). Recirculation of
fly ash did not appear to improve; the sorption rate,
and reducing conditions in the bed seriously lowered
the sorption efficiency of the bed. Lowering the super-
ficial velocity from 12 to 8 fps delayed the 20% break-
through as might be expected, since the input sulfur is
proportional to the superficial velocity.
6.9.3 Desorption of the Sulfated Bed
The difference in sulfur retention in the bed with
variation in temperature and oxygen level suggested
the possibility that sulfur retained at the favorable
conditions could be released by changing either tem-
perature or oxygen level or both. Desorption of
sulfur might effectively "regenerate" the bed for
further sorption.
The "regeneration" procedure was first carried out in
FBC Test 114 with increase in temperature only and no
change in the oxygen content of the flue gas. The
procedure involved sorption in a bed weighing 119 lbs
for four hours at a temperature of 1520°F. The tem-
perature was then increased to 1920°F.
A plot of emissions monitored during the test is
shown in Figure 35. Variation of calcium and sulfur
contents in the bed is shown in Figure 36. The re-
sults show that during regeneration sulfur was re-
leased from the bed at a rate sufficient to produce
an S02 concentration of 1.5% {15,000 ppm) above the
bed. At the same time, the sulfur content in the bed
decreased from 6% to ^0.8%. A rigorous sulfur balance
employing integration of the sorption, input and de-
sorption curves indicated that about 90% of the sulfur
sorbed in the bed was released during the period of
higher temperature.
The details of the sulfur balance are presented in
Appendix A; Enclosure 21.
POPE EVANS AND ROBBINS
»-3 "1
z ft to
own
z
m lh
CO oo
o o
|—. H- h- h-
u~l ,U Ul
ui ko © o
© © o o
*-¦ K
cr d
w Qi
o M
H ft tfl 3
5 TJ It
• (t a it
1o a
N til H
W C
o ft 03
3 1
i
1
Ul
.25
^ 0 CD Hi
V-»
—i
UI
UT
tt C »-t O
• vQ O 1-1
3" P»
X* N)
1 O
Of
O O
K) H>
00
tr 3
tt H Z
\ a o
g's
~3 o
G 3
O ft
*< H
o o
t—1 3
nj cn
3 0
Pj
•a
o © ¦ ¦
. . O o
I—' o Ct
D
cn
h- tr ft d* w
0
I—' D
cn
w
ft CA
ft
cn
O
(f< o n o
o
• ft
0
0
H- O
10
rt
n
HUM
3
cn u
n
h
Oi H
0
13
JU JD rg
o
<*> 0
13
13
T3
h
o
rt
h
ft
ft
3 ft
*0
CD ft ft H
•o-o
H-
H
0 H-
rt
o
o
in a 3" 0
ft rt
0
0
ft o
H-
t—*
3
hi n n 3
(V K>
3
3
3
0
ft
13 • O 1
yr o
O
3
10 < <7\ c to
' (D iQ ©
3
W
1
M
ro
1- ET 3* dp
o n
o
o cr
H ft
tn
O 0)
3 a
dP
i
ft
Sitn
ffl H
•-3 O
C 3
50 0
3
S e. ft
68
-------
WET TESTS
2 3
TEST PERIOD, HOURS
FIGURE 35. EMISSIONS DURING FBC TEST 114 BURNING A MEDIUM SULFUR
COAL IN A FLUIDI ZED-BED OF 1359 LIMESTONE
POPE, EVANS AND ROBBINS
TABLE VIII. (Continued)
FBC SOjinput
Test Equiv.
No. Ibs/MBTU
Init. Time for 20%
bed MasB Bed Flue Bed S02Break-
Initial Final Depth Gas Temp. through
lbs lbs iiu 02 % °F hrs«
NO HC Test
Emission Emission Condition
lbs/ffBTU lbs/MBTU Remarks
11B 4.7
(contin-
uation of"1
117)
; 4 . 7
119
120
4.7
4.7
4.7
4.8
4.9
4.3
4.8
4.8
86^
114
114
55
51
11
14
17
17
1 Final bed from 117
2
Added 27 lbs limestone
3.0
0.1
3.0
5.0
0.2
3.03
3.0
1.0
3.0
1600
1920
1550
1550
1930
1570
1570
2000
1520
3.0 1550
0.9
1.3
2.5
0.1"
1.5
2.9
0.28
0. 29
0.29
0.33
0.25
0.16
0.30
0.38
0.30
0.24
0.02
0.16
0.02
0.00
0.15
0.02
0.02
0.05
0,02
0.04
3 Made up with overbed air
<«
Time from start of reducing conditions
Sorption
Desorption
Peak S02 5.5%
Sorption
Sorption 20%
breakthrough
delayed 1.2 hre.
Desorption
Peak S02 6.0%
Reducing condi-
tion in bed
after 1 hr
sorption
Sorption
1st cycle
Desorption
Peak SO2 8.1%
Sorption
2nd cycle
Sorption with
80% a&h
recirculation
-------
SNIQSOiJ QMV SNVA3 3dOd
SULFUR,
O
G
»
H
z
o
tt
w
^3
CO
M
O
O
• M
Z
w
o
z
£6
93
This result appears to contradict thermodynamic data
which show calcium sulfate to be stable at 1900 0if
under oxidizing conditions. When chemical analysis
of the bed showed only small quantities of the more
unstable sulfite in the bed before regeneration, it
was concluded that local reducing conditions break
dovn the sulfate according to the following relation:
Ca S04 + CO Cor H2) -+ Ca 0 + S02 + C02 (or H20)
From the sulfur and calcium analyses, the flow of
sulfur was estimated as follows:
SULFUR RATE, LBS/HR
Test Time< hours
1
2
3
4
5
Flue gas output
0.00
0.39
0.85
1.15
10.4
Fly-ash output
0.27
0.25
0.35
0.27
0.3
Bed retention
1.75
1.47
0.70
0.50
-8.8
Total output
2.02
2.11
1.90
1.92
1.90
Input
1.95
1.95
1.95
1.95
1.95
Input less
output
- .07
-0.16
0 .05
0.03
0.05
This and sulfur balance data for other tests are
included in Appendix C.
Before leaving Test 114 it should be noted that the
initial weight of limestone in the bed was less than
used previously so as to reduce the sorption time and
to study the effect of bed mass (or depth). The re-
sults in Table VIII show that decreasing bed depth
significantly decreases the time for 20% breakthrough
of sulfur dioxide.
Subsequent tests, with simultaneous reduction of
oxygen content and bed temperature increase, showed
that sulfur could be desorbed from the bed more rapidly
than with simple change in bed temperature. A concen-
tration of 8.1% S02 was observed in regeneration during
Test 119 when the oxygen level was reduced from 3% to
1% as the temperature was increased to 2000°F. This
variation is shown in Figure 34.
The figure also shows the trend of emissions when re-
ducing conditions were effected in the bed for a short
period, i.e., with ^80% of the combustion air passing
through the bed and the remainder of the air supplied
above the bed to hold constant the 3% oxygen content.
POPE, EVANS AND ROBBINS
-------
94
The test showed that the bed could be desorbed with
reducing conditions alone and without an increase in
bed temperature. The notable effects on nitric oxide
and hydrocarbons emissions tare discussed in the next
ceotion.
The reactivity of the limestone bed throughout a
number of sorption regeneratic-i cycles could not be
studied in detail, but it was apparent from a second
oycle operation that the 20% bieaicthrough time is
shortened on the second cycle. It was also apparent
that the rate- o.C iiiL-Ledfau: in sulfur dioxid.s content
in the flun yas, once some appears, is approximately
the same as the first cycle rate.
Attrition loss of the bed material during regeneration
appears to vary in the range of 2% to 4% (calcium) of
the original charge per hour.
6.9.4 Other Emissions
The data summary in Table VIII shows the variation in
nitric oxide and hydrocarbons emissions during the
limestone bed test series. The nitric oxide emission
at low bed temperature varied approximately in the
same range observed with the sintered ash bed opera-
tion, i.e., .20 to .30 lb/MBtu but appeared to be
more responsive to change in bed temperature. During
the desorption phase of Test 11£, the NO emission
increased from 0.21 to 0.52 lb/MBtu with temperature
increase from 1530°F to 1920°F at constant (3%) O2
in the flue gas. The NO emission did not increase
with temperature durinq Test 117. apparently heranw
the O2 content was reduced to 0.2%. The character-
istic reduction in NO emission with lower O2 content
was discussed in Section 6.6.
When reducing conditions were created in the limestone
bed during the low temperature sorption phase of
Test 119, the NO emission showed a decrease from 0.30
to 0.16 lb/MBtu despite the 3% O2 in the flue gas
(supplied by overbed air). This result indicates that
NO emission can be limited by a simple form of two-
stage combustion. Unfortunately, this particular
mode of operation did not favor sulfur sorption in
the bed as indicated in Section 6.9.3.
The reducing conditions phase of Test 119 also pointed
up the fact that hydrocarbons can be consumed with
overbed air at the 3% oxygen level. At this value
the hydrocarbon emission remained constant at
0.02 lb/MBtu in the change from oxidizing to reduc-
ing conditions. _The emission varied in the range of
0.02 to 0.05 lb/MBtu at low temperature operations. The
POPE. EVANS AND ROBBINS
95
0.12 value observed during Test 114 is thought to be
an instrument error. When the temperature is increased
during docorption, the hydrocarbons disappear except
when tho oxygon content is lovcrcd simultaneously. AL
the lower oxygen levels, the hydrocarbons emission is
sharply increased at any temperature.
Sulfur trioxide emission during the limestone bed opera-
tion vac ^ero.
Particulate cmiccion during the sorption pha3e of the
limestone bed tests and the energy lost in unburneu
carbon are summarized as follows:
FBC Fly ash Carbon Discharge Coal Energy Loss in
Test Collected Content to Atmos. Input Unburned Carbon
No.
lbs/hr
%
lbs/hr
lbs/hr
% of input
113
14.0
46
1.6
64
11.7
114
15.0
43
1.4
65
12 .0
115
15.6
42
1.5
65
11.6
116
14 .9
42
1.2
64
11.2
117
16.6
39
1.7
64
11.5
118
14.4
43
1.2
62
12.4
119
14.8
38
1.5
65
10.7
120
12.8
47
1.3
61
11.9
The particulate emission was about the same as
observed with the sintered ach bed operation and
was locc than that obcorvod with a sintered ash
bed and fine sorbent injection. The latter was
discussed in Section 6.7. The energy lost in
unburned carbon is about the same loss observed
with the sintered ash bed operation.
-------
96
7. RESULTS OF BOILER MODULE (.FBM) TESTS
Procedures employed m the FBM tests involving both
coarse and fine limestone injection are discussed in
Section 5.2. In general, the test conditions selected
were those observed to favor sulfur emission control
during the FBC tests. Tests for sulfur trioxide emis-
sion from the FBM are not discussed separately in this
section since results are comparable to low values
observed in the FBC tests (Section 6.4). The method
of gas sampling is discussed in 5.4 and the sampling
system is shown in Figure 20. Variations in emissions
during the course of the tests are shown in Appendix A.
Complete data summaries are presented in Appendix B
and sulfur balances m Appendix C.
7.1 SULFUR DIOXIDE EMISSION
Emission of sulfur dioxide from the FB?1 without sorbent
addition indicates that 90% to 95% of the sulfur in the
coal appears as sulfur dioxide in the flue gas. The
remainder of the sulfur is held in the fly ash. This
distribution of sulfur is shown in sulfur balances in
Appendix C.
When coarse, raw 1337 dolomite was injected into the
FBM while burning the 4.5% sulfur coal, the most favor-
able calcium utilization observed was 31.2%. The S02
reduction was 54.5%, the bed temperature 1600°F, the
oxygen content 3.5%, the stoichiometric feed ratio
1.75, and the sorbent particle size -7 +14 mesh. These
data and others pertaining to the coarse dolomite addi-
tion are summarized in Table IX. The results are com-
parable to values reported in Table I for the FBC
under similar test conditions. The coarse 1359 lime-
stone was not tested in the FBM because of its poor
performance in the FBC, as indicated in Figure 22.
When the finely divided sorbents (-325 mesh) were added
to the combustion of the 4.5% sulfur coal in the FBM,
the sulfur dioxide reductions and calcium utilizations
were found to equal.those observed in the FBC. The re-
sults of the tests are summarized in Table X, and a com-
parison with FBC data trends is shown in Figure 37.
The FBC Trend lines were reproduced from Figure 25.
x
The FBM results indicate a reduction of 74% at a Ca/S M
ratio of ].7 with the 1337 raw dolomite. This reduc- w
tion is exactly comparable to the FBC results as indi-
cated in Figure 37. A reduction of 74% observed in the ^
K
cn
u
X
H
+
r-
<
w
o
CO
u
<
<*
o
u
u*
h!
U->
£3
o
W
z
dP
O
M
«
H
M
a
<
a
<
o
z
X
H
H
2
H
OS
s
D
£Q
to
2 S
O
CO
M
Cm
U
U
O
T.
Q
H
S
O
Eh
w
a
W
H
Eh
X
H
o
JS
c
s
o
05
p
Zi
Cm
m
D
m
in
rH
EVANS .AMD R03B1NS
97
c
o
cm-H
O -M
m o
O) o
3
m
df> 73
m t
0)
Pi
CM O O O
m rH r» co
i—1 (N CO *3"
r—i
fO
o
O
o
©
o
o
c
in
m
o
o
o
o
r|
m
©
ro
o
o
Cm
rH
(N
m
m
Oi
i—i
3
f0
u
K
H
rH
r—i
i-H
tn
nj <*>
O
in oo
3 oj
rH O
Cm
T3 Cu
oet.
CQ Q) o
Eh
in in u*
13 J-> •
Q»
¦H +>
E
0 -P
rH tO
ro
OS
cn ro u">
-------
13
0
'0
Pi
m
><
z
0)
>
z
0
3J
0
ffl
ID
z
a)
z
o
H
E-i
'J
3
Q
W
X
K
D
Cu
3
SORBENT INDICATED BY SYMBOL:
SORBENT MESH SIZE
TEST CONDITIONS:
REACTOR: FBM
SUPERFICIAL VELOCITY: 12-14 FPS
BED TEMPERATURE: 1500°?-1600°F
COAL: OHIO #8 PITTSBURGH SEAM,
UNWASHED, 4.5% S
BED: SINTERED ASH, 20"-24"
STATIC DEPTH
FBC DATA TREND LINE
FOR 1359 SORBENT-
~o -
10 2.0
Ca/S STOICHIOMETRIC RATIO
3.0
FIGURE 37. SULFUR DIOXIDE REDUCTION WITH SORBENT ADDITION TO THE FBM
BURNING A 4.5% SULFUR COAL
TABLE X. DATA SUMMARY FOR INJECTION OF -325 MESH SORBENTS INTO THE FBM
BURNING A 4.5% SULFUR COAL
FBM
Test
Sorbent
Bed
Depth
Bed
Temp.
Flue Gas
Ca/S
SOj Cone, ppm
% S02
% Sorbent
No.
Type
in.
°F
02 %
Ratio
Initial
Final
Reduction
Utilization
25
1337R
24
1550
3.0
1.70
3750
1100
71.5
42.0
"
"
1520
3.0
1.70
It
950
74.2
43.8
26
133^
20
1660
3.0
1.70
3750
1350
64.2
37.8
"
II
II
n
1.90
II
HOC
70.9
37.3
27
1359R
20
1570
3.0
2.00
3700
950
74.0
37.0
29
1359R
. 20
1600
3.0
1.70
3730
1500
73.5
35.2
"
"
II
"
2.00
"
1000
73.5
36.6
Z
10
-------
100
FBM with the 1359 lxmestone at a ratio of 2.0 is indi-
cated to be somewhat more favorable than the 70% reduc-
tion indicated in the FBC at this ratio.
Reductions in sulfur dioxide emission observed with the
medium sulfur coal are summarized in Table XI and are
compared with the FBC data tre/.ds in Figure 38. The
comparison indicates that the -325 mesh 1359 raw lime-
stone is as reactive in the FBM as in the FBC. The
hydrated forms of both sorbents indicated a reactivity
comparable to the raw stone.
,2 HYDROCARBONS EMISSION
Emission of hydrocarbons from the FBM was observed to
vary sharply with flue gas oxygen content in the same
manner as noted in the FBC tests, but the general level
of emission was somewhat higher. Concentrations varied
as shown in Figure 39 from ^4600 ppm at 0.5% O2 to 50 p
at 4.0% O2• At the 3% O2 level maintained during the F
tests, the concentration varied from 210 to 260 ppm.
During the FBC tests, the concentration varied from 50
100 ppm (Section 6.5). An average concentration of 230
ppm for the FBM test operation corresponds to 0.10 lbs
CH4/MBtu input.
These results indicate that a 4% oxygen content in the
flue gas would be necessary to limit hydrocarbon concen-
trations to 50 ppm {.02 lb CH /MBtu emission). The
excess air requicement would be approximately 24% unles
improvements in coal feeding methods are made.
Injection of sulfur control sorbents did not affect
hydrocarbons emission.
Carbon monoxide emission, determined by Orsat analysis,
was found to be nil at oxygen concentrations of 2% and
higher. CO concentrations of 0.4% appeared in the flue
gas when the oxygen content was reduced to 1%.
3 NITRIC OXIDE EMISSION
The concentration of nitric oxide in the flue gas from
the FBM was observed to increase from 280 ppm to 340
ppm with increase in oxygen content from 1% to 4%.
This variation, shown in Figure 40, is characteristic
of the variation observed in the FBC but the concentra-
tions are somewhat less {cf Figure 30). An average
value of 275 ppm for the FBM tests compares favorably
with an approximate average of 380 ppm for the FBC
operation. These concentrations correspond to emis-
sion values of 0.22 and 0.30 lbs NO/MBtu respectively.
EVANS .AKJD ROBB1NS
SNlSSOy QNV SNVAH 3cOd
ujcoujcococococogjcococolo
0000000000000
co 00 oi CO CO
fo fO W N) W M
M ^ N) NJ ^ ^
fO O to CO CO
tO M? Ul O
00 to M O
xj (Tl (Jl (J\ (Jl
ro ro nj ro
•£» co to to
CO Ul CD
CD (J\ a\ (J;
Ul VD Ul CO
to u> co co
to CO CO
COUItOfOCoCOCOfO
h3 ^
OJ
CO
to
ro
ro
to
ro
X ft
c.
CO
Co
Co
Co
< h
Ul
= en
: cn =
= CO
Co
CO
co
¦a cr
V£>
CO
-0
-J
(t> (t>
EC
s
&
rc
tc
p
rf
to
C
to
to
to
to
I—"
H' (D CD
0
s 0
s 0 s
: to
CO
0
kO
3TJ O
• rt a
sr
t-1
H"
1-*
(-¦
t—'
t—<
CT\
cn
a\
cn
ON
o>
0 n> to
to
= to
s O s
= 0
Ul
Ul
to
3 ro
O
0
O
0
0
0
Ul
13 a
a
0)
* n
&>
f+\
p- to
o
to
- o
I o
-¦ o
3
o
J •
• -a
1 3
w
CD <#>
a
e cn
o o
r+
H-
o
a
rt «jp
' to
• 0
tr
•
• 3
TOT
-------
% 'HIV SS30X3
E-»
Z
H
Eh
2
O
o
2
W
o
><
o
M
<
o
w
D
J
Cm
Z 3
O CQ
H t*
Eh
rf'W
C5 33
E-i £H
S
W Z
U H
T
O f-«
u ^
w
W H
a is
o o
cq a
< 12
u w
o o
05 >*
a x
:* o
K
W
2 >
o
Wdd 'SVG Sflli NI N0I-LVH1N30N0D ONVHiatM) KOffaVOOHCUH
SORBENT INDICATED BY SYMBOL:
10
20
. 30
40
50
60
70
'80
90
100
TEST CONDITIONS:
SYMBOL SORBEFT
MESH
~
1337 H
-325
~
1337 R
-325
O
1359 H
-325
o
1359 S
-325
H =
HYDRATE
R =
RAW
REACTOR:FBM
SUPERFICIAL VELOCITY: ^13 FPS
BED TEMPERATURE: 1550-1630°F
BED DEPTH: 19'-22'
COAL: OHIO #3 PITTSBURGH SEAM,
WASHED, 2.6% S
FBC DATA
TREND FOR
1337 S0SBENT-
v.
s
Ql
vO
-tt
~
vr
5sr-
O
Cr
"•S.@
~
Q
FBC DATA.
TREND FOR
,1359 SORBENT
1.0 2.0
STOICHIOMETRIC RATIO, Ca/S
3.0
FIGURE 38. SULFUR DIOXIDE REDUCTION WITH SORBENT ADDITION TO THE FBM
BURNING A 2.6% SULFUR COAL
-------
104
TEST CONDITIONS:
REACTOR r FBM 20" x 72"
SUPERFICIAL VELOCITY: 12-14 FPS
BED TEHPERATURE: 1770-1840°F
BED: SINTERED ASH, STATIC DEPTH 13"
COAL: OHIO S8 PITTSBURGH SEAM,
WASHED, 2.6% S
FIGURE 40. FBM - VARIATION OF NITRIC OXIDE CONCENTRATION
WITH FLUE GAS OXYGEN CONTENT
POPE, EVAIMS AND ROBBINS
105
Two tests conducted at a bed temperature of 1800°F
indicate NO emissions as high as 0.28 lbs/MBtu but
values observed at 1600°F varied from 0.18 to .24 lbs
NO/MBtu. These results suggest a correlation with bed
temperature, but scatter of additional data points at
the higher temperature might prove otherwise.
Injection of sulfur control sorbents reduced the nitric
oxide emission by about 30% in one test, but the results
were not reproduced in subsequent tests with tne same
sorbents. In general, the nitric oxide emission was not
affected by sorbent injection.
A small increase in NO emissions (5% - 10%) was noted
during the transition from a low nitrogen coal (1.6% N2)
to one of higher nitrogen content (2.5% N2). The in-
crease was less than the data scatter, however, and the
test was felt to be inconclusive. Nitrogen oxides con-
centrations, determined by the phenoldisulfonic acid
procedure generally agreed with values of nitric oxide
determined by infrared absorption. These are shown on
the emission curves in Appendix A.
.4 EFFECTS OF FLY ASH RECIRCULATION AND STEAM INJECTION
Recirculation of fly ash from the collector hopper to
the base of the fluidized bed was tested as a means of
improving sulfur capture. The procedure was discussed
in Section 5.2 and the system shown in Figure 10. Re-
circulation was initiated during a steady-state reduc-
tion of sulfur dioxide with sorbent injection, and con-
tinued for one hour. The results are indicated as
follows:
1359R
Fly
FBM
Coal
Sulfur
Sorbent
Ash
S02
S02
Test
Feed
Cont.
Feed
Recir.
Emission
Reduc.
, Sorbe:
No.
lbs/hr
%
lbs/hr
Rate
lbs/MBtu
%
Util
29
720
4.5
220
0.0
1.7
73.5
36.6
"
"
80%
1.7
73.5
36.6
32
720
2.6
108
0.0
1.8
65.0
36.0
"
"
"
it
80%
1.8
70.5
39.1
These results indicate little or no improvement in sul-
fur capture with 80% ash reinjection during a one-hour
period. However, a one-hour period is not sufficient to
achieve steady state in a recirculating mode and some
improvement in utilization might have been found at
steady state. Although 25 to 30 hours would be required
to approach steady state, the marginal improvement in
POPE EVANS AND ROBB1NS
-------
106
the first hour, which would show the largest increment
of improvement, suggests that the once-through material
is essentially inert.
Continuous recirculation, with limestone injection, would
cause infeasible dust loadings and would require sending
some of the collected dust to waste so as to avoid "chock-
ing" the system.
Nitric oxide and hydrocarbons emission were not affected
by recirculation.
Injection of steam into the inlet air during sorbent
addition improved the sulfur capture but the improve-
ment was probably due to a simultaneous decrease in
bed temperature. The observations are summarized as
follows:
Coal Inlet Air
FBM
Sulfur
Water
Bed
S02
S02
Test
Content
Content
Temp.
Ca/S
Emission
Reduc.
Sorb.
No.
%
% Vol.
°F
Ratio
lbs/MBtu
%
Util.
20
2.6
0.5
1780
1.46
2.20
40
27
H
"
8.8
1700
1.46
1.50
58
40
21
2.6
0.5
1680
1.37
1.50
56
41
n
n
8.8
1600
1.37
1.25
62
44
The reduction in Test 20 appears to show a significant
effect from water injection except for the fact that
the bed temperature was also reduced. Test 21 shows
that virtually the same reduction can be produced at the
lower temperature without water injection. Nitric oxide
and hydrocarbons emissions were unaffected by the water
injection.
Since bed temperature can be adjusted readily with bed
depth, there appears to be no advantage to water injec-
tion. A disadvantage would be a slight reduction in
the boiler thermal efficiency.
7.5 PARTICULATE EMISSION
Isokinetic samples of particulate matter were drawn from
the long duct above the FBM at a point just upstream oi
the induced draft fan as shown in Figure 10. The sample
point is downstream of the FBM cyclones and the samples
taken were proportional to the rate of particulate dis-
charge to atmosphere, when fine sorbents were injected,
the particulate emission increased as expected. Typi-
cal data are summarized as follows:
POPE. EVANS AND ROBBINS
107
FBM
Test
No.
Type Rate
lbs/hr
Additive
Type Rate
lbs/hr
Fly Ash
Collected
lbs/hr
Fly-ash
Emission
lbs/hr
Collec-
tor
Effi-
ciency
27
Unwashed
760
No Additive
156
10.5
34
Unwashed
760
1359R 220
332
16.5
94
28
Washed
745
No Additive
102
8.9
91
Washed
745
1359R 150
230
12.4
94
29
Unwashed
720
No Additive
135
12.1
92
Unwashed
720
1359R 175
295
14.7
9:
31
Washed
800
No Additive
108
7.7
93
Washed
800
1359H 65
200
11.4
94
32
Washed
720
No Additive
115
10.9
91
Washed
720
1359R 97
180
13.7
93
One fly-ash sample taken from the cyclone discharge
during the addition of sorbent in FBM Test 24 was
analyzed for particle size distribution by micro-
scopic count. The size distribution, shown in Figure 41,
indicates that 90% (by number) of the material was
smaller than 5 microns. Assuming spherical particles of
equal depsity, only about 52% (by weight) of the particles
were smaller than 5 microns. The sorbent was 1337R, -325
mesh fed at a rate of 260 lbs/hr with the washed coal at
800 lbs/hr. The particle size distribution of fly ash
collected in the cyclone was not determined.
POPE. EVANS AIMO ROBBINS
-------
801
109
DISCUSSION OF FBC AND FBM TEST RESULTS
On the basis of performance observed during the test pro-
gram, the fluidized-bed boiler appears to offer pollution
control advantages with respect to all three of the chem-
ical pollutants studied, i.e., sulfur oxides, nitrogen
oxides and hydrocarbons. On the other hand, control of
particulate emission may be somewhat more difficult with
injection of fine sorbents for sulfur emission control.
Factors which relate to possible advantages in boiler
maintenance are apparent. These considerations and the
effects of dominant variables are discussed in the follow-
ing paragraphs.
Emission of sulfur dioxide from combustion of coal in a
fluidized bed contains 90% to 95% of the input sulfur.
The balance is retained in the fly ash probably as a
pyrite form. A very small amount of sulfur appears as
sulfur trioxide in the flue gas.
In the control of sulfur dioxide emission, effectiveness
of sorbent materials was seen to depend primarily on
sorbent type, feed rate, particle size, bed operating
temperature, oxygen content in thr flue gas and, to a
lesser extent, on bed depth. The effect of sorbent was
shown in the comparison of reductions with the 1337 dolo-
mite and the 1359 limestone. The dolomite proved to be
superior on a Ca/S basis, i.e., when the magnesium frac-
tion was discounted as a sorbent. On a weight basis,
however, the 1359 limestone was more effective particu-
larly when ground to a -325 mesh particle size.
The improvement in desulfurization, observed with in-
creased stoichiometric feed ratio of the limestone, is
accompanied by a decline in the sorbent utilization.
Utilization of the finely divided, raw 1359 limestone,
under the most favorable conditions was found to vary
from 40% at a Ca/S ratio of 1.0 to 33% at a ratio of
2.0 and 28% at a ratio of 3.0. This result is consistent
with decline in the driving force in the reaction, i.e.,
the sulfur dioxide concentration in the system. In
terms of SO2 emission reduction, the performance indi-
cates that 80% of the sulfur emitted from a 4.5% sulfur
coal could be captured with a Ca/S ratio of 2.7.
Grinding the sorbents to a fine particle size (-325 mesh)
markedly improved sulfur capture (and the sorbent-utili-
zation) despite the expectation that the residence time
of fine particles in the fluidized bed would limit desul-
furization. The improved utilization is apparently the
result of greater reactive surface per unit mass of
sorbent, and the ease of calcining the small particle
POPE. EVANS AND ROBBINS
-------
110
as the initial step in the desulfurization reaction.
The attempt to find a particle size which would pro-
vide an optimum between residence time and reactive
surface failed to show such an optimum. The small
particle size (325 mesh) was more reactive than any
larger size at constant bed depth.
Increase in hed residence time by increasing the bed
depth from 10 to 13 incnes indicated a small imprcvs-
mpnt In sulfur capture at the low bed tempsrature.
This result and the failure to observe an optimum sug-
gests that product shell diffusion i"? controlling even
with small sorbent particles. This conclusion is fur-
ther supported by the fact that increasing particle
residence time by reducing superficial gas velocity
did not show an improvement in sulfur capture.
'rne rapid improvement in deaulfurisafcion with reduction
in particle size suggests that fine grinding may be
necessary for effective utilization of the 1359 lime-
stone. The corresponding lime hydrate, which occurs
naturally in the fine state, was equally as effective
as the fine raw stone but is considerably more expen-
sive. Other, less durable limestones, may tend to decrep
tate in the bed and initiyaLe Lhe grinding requirement.
The reactivity of sorbents in the fluidized bed was
tound to be greater in eveij iusLance at a bed tempera-
ture of 1550°F than at 1800°F. This behavior is con-
sistent with thermodynamic predictions for the reaction
but equilibrium in the bed is improbable. It is incon-
sistent with kinetic considerations. A possible explana*-
tion is that the lower bed temperature producer a soft,
highly porous calcine with minimum crystal growth. At
temperatures DSlow lSOO'F Uie ieacLiviLy may be reduced
by failure of the sorbent to calcine.
Operation of a fluidized-bed boiler at 1550°F instead of
1800°F does not mean that less heat is transferred out
of the bed. The bed temperature is reduced from 1800°F
to 1550°F by increasing the bed depth and hence the bed
cooling contact surface. The fact that the gases leave
the bed at a lower temperature means a lower heat loss
in the gas and hence an even greater heat removal from
the bed. The input energy is fixed by the superficial
velocity range.
The low bed operating temperature should reduce boiler
tube slagging.
POPE. EVAJSIS AND HOBBINS
111
Heat required to calcine the sorbent does not create a
demand on the system since it is supplied, for the most
part, by heat release from the desulfurication reaction.
Standard reaction energies indicate that one pound of
CaC03 would absorb 775 Btu in calcination but would re-
lease 1300 Btu if fully converted to sulfate. The ener-
gies balance, if the utilization is 37%--roughly the
utilization observed in the test program. The sensible
heat loss with sorbent feed will be small by comparison.
Ilowcvci-, the use of a sorbent must be considered in the
design of the boiler since heating of the sorbent and
calcination both take energy and honcc tend to reduce bod
temperature. This energy, removed from the bed in the
form of a hot solid and hot CO2, is recovered, in part,
in the convection zones.
The fact that sulfur capture is favored by increase in
excess air is readily apparent from the limestone bed
investigation. This study clearly demonstrated that
sulfur oan bo capturod offoctively for a period of timo
in a bed of limestone and then discharged from the bed
by reducing excess air and increasing bed temperature.
The sulfur release apparently follows the reaction:
C (or H2) + CaSOn ¦+ CaO + S02 + C02 (or H20)
It was shown that culfur release may occur with oxygon in
the flue gas probably because of local roducing conditionc
in the bed. Mildly reducing conditions in the bed accel-
erate the sulfur release and offcct highor oulfur concen-
trations in the flue gas.
Most significant is the fact that concentrations of sul-
fur dioxide in the off-gas from the bod during the regen-
eration period may be thirty times the untreated flue gac
concentration. Concentration as high as 8.1% observed
during regeneration markedly increases the feasibility of
sulfur recovery. Concentrations in excess of 8.1% might
be achieved by designing the regeneration region 00 ac to
minimize heat loss. This, in turn, would reduce the fuel
and air requirement and so reduce dilution of S02 by
CO2 and N2.
Recycle of the limestone through absorption and regener-
ation phase might provide the means for improving the ef-
fective limestone utilization beyond the present limit.
This will depend on how well the reactivity is retained
and the long-term attrition rates. Additional work in
this area is indicated. Utilization per cycle might be
increased by larger percentages of excess air.
POPE EVANS AND ROBBINS
-------
112
The method of sorbent feed into the bed appears to be
optional in the FBC with no clear advantage for any of
the methods under study. The test of optimum sorbent
distribution by injection on all four sides of the FBC
unit showed the same sulfur control as single side in-
jection. These results speak well of the mass transfer
within the FBC bed. The two-point feed system used in
the FLM appears to be as effective as any of the systems
used in the FBC.
Failure to observe a consistent, beneficial effect from
recirculation of spent sorbent in the fly ash suagests
again the product shall limitation. Wetting the fly a^h
before recycle may improve the sorbent utilization by
breaking down the particle as the core becomes hydrated.
This procedure has not been tested.
Sulfur trioxide elimination with sorbent use is consist-
ent with the active nature of the compound. Its absence
could make electrostatic precipitation of fly ash more
difficult unless the design of the system exploits the
high carbon content of the primary fly ash.* On the
other hand, boiler tube corrosion should be reduced.
Emission of hydrocarbons from the fluidized-bed boiler
clearly precludes its operation at very low values of
excess air (5%) but the advantage is noted that hydro-
carbons can be eliminated with only moderate rates of
excess air. The test results suggest that a 4% oxygen
content in the flue gas will be necessary to prevent
hydrocarbons emissions from the FBM operation. This
oxygen content corresponds to an excess air rate of 24%--
a value which compares favorably with values of 40 - 50%
commonly used in coal fired industrial boilers.
The loss in energy from hydrocarbon emission would
probably be as great as the heat saved by lower excess
air operation, as estimated in Section 6.5
The lower excess air requirement for the FBC operation
(17%) suggests a better distribution of volatile matter
in the smaller bed. The potential seems to exist for
decreasing the excess air requirement about 10% while
still burning essentially all hydrocarbons and CO if
the fuel distribution system is substantially improved.
A fuel saving on the order of 1/2% would then be
realized.
See Appendix A, Enclosure 45.
POPE. EVANS AND ROBBINS
113
Nitrogen oxides emission from the FBM were found
to be less than the_ emission from the FBC
(0.22 vs 0.30 lbs/MBtu). This result may be related
to the higher hydrocarbons emission from the FBM,
possibly by the reaction:
(CHx) + 2N0 - N2 + C02 + (H20)
The flue gas oxygen concentrations were 3.0% for
both the FBM and FEC tests. This suggests that b
hydrocarbon gas properly dispersed at the grid might
reduce the NO emission without affecting the sulfur
control functions.
The moderate sensitivity of nitric oxide emission to
flue gas oxygen content suggests that the level can be
reduced by lowering the average oxygen concentration in
the bed, i.e., by reducing conditions. A NO reduction
of 50% was observed in Test No. 119 with reducing con-
ditions in the bed and overbed air to make up the 3%
oxygen content (cf. Figure 36). Unfortunately, this
mode of operation is not conducive to sulfur capture
in the sulfate form. Hydrocarbons from the bed were
effectively consumed by overbed air.
These results would indicate that nitric oxide emission
can be reduced in a limestone bed without aggravating
the hydrocarbons emission by two-stage combustion, i.e.
by reducing conditions in the bed and an oxidizing en-
vironment above. It may be possible to capture sulfur
as the sulfide in a cyclic operation under these condi-
tions .
The nitric oxide emission from the FBM is favorable in
comparison with emissions from other combustion units
of equal size. The average value of 0.22 pounds/MBtu i
less than reported values for most conventional boilers
A full scale boiler made up of modules according to the
present concept may not be subject to the increase in
NOx emissions generally observed with increase in unit
capacity.
Most of the fine sorbent added to the bed is collected
in a single stage mechanical cyclone operating at 95%
efficiency. Controlling emission of the remaining 5%
may present a problem if subsequent tests show that 90%
of the particulate is smaller than 5 microns when fine
sorbents are used. The microscopic count showing this
distribution applied to one sample. Additional data
are needed for a firm conclusion regarding particulate
emission control.
POPE EVANS AND ROBBINS
-------
114
9. ECONOMIC ANALYSIS
9.1 GENERAL
9.1.1 Economic Evaluations in Industry. Investment deci-
sions in the selection of steam and power generating
equipment are made in a number of ways. The factor^
which are utilized vary from industry to industry
and from company to company within an industry.
Typically, however, a central steam supply is viewed
as a long-term investment not subject to the same
rapid pay-out demands as a process investment might
be.
Whatever factors are applied, a rational technique of
structuring the decision-making process is required.
It is possible, for example, to apply the present-
worth method. By this technique all capital and
operating expenses are reduced to a single dollar
figure, the "present worth" of all present and future
expenses. A number of other investment appraisal
techniques exist but present worth appears to be the
most popular.
Making application of the present-worth method in a
sophisticated manner requires that predictions be
made as to the future cost of labor, the future cost
of fuel, etc.; and, when certain investments may be
deferred, the future cost of money. Fortunately, in
the field of steam power generation an extensive
statistical base exists on which reasonable projec-
tions of future costs may be made. The various alter
natives are then evaluated on the present-worth basis
The best apparent choice is that alternative which
has the lowest present worth. Computerized evalua-
tions make possible sensitivity checks, i.e., the
effect of an incremental change in each cost ingre-
dient may be evaluated so as to determine which are
the most significant.
Unfortunately, when air pollution control is added to
the list of plant requirements and this requirement
also includes control of gaseous emissions, the sta-
tistical base becomes very limited. In addition,
even current capital and operating costs involved in
pollution control techniques, other than the selec-
tion of a low sulfur fuel, are based on a limited
number of "paper" evaluations.
POPE, EVANS AMD ROBBING
115
This section of the report is not a complete invest-
ment appraisal; instead it is intended to indicate
how limestone- addition to a fluidized-bed boiler may
affect costs. These data might then be utilized in
a complete investment appraisal.
9.1.2 Treatment of Incremental Costs. The costs which
are included in the evaluation are those which are
directly attributable to limestone injection. It
must be assumed that in th3 selection of a fluidized-
bed boiler, limestone injection is not treated as an
afterthought. The boilers, the plant, the auxiliaries
and the pollution control systems are designed with
maximum degree of integration. Examples of such
integration include: a single receiving point for
coal and limestone; a single bulk conveyor system;
a single, but properly, partitioned storage silo;
the use of the preheated combustion air or possibly
flue gas to dry the limestone before pulverizing;
and the use of the boiler's induced draft fan to
provide any suction required on the limestone system.
The boiler itself receives all dust vented from the
limestone handling and storage system, etc.
When a single system serves two functions, it is
reasonable to attribute only an incremental cost to
the function being evaluated. Therefore, limestone
addition to a new, properly designed fluidized-bed
boiler plant is far less costly than limestone addi-
tion to an existing plant or to a new plant in which
pollution control is an afterthought.
The cost estimates were based on the assumption that
if a new plant were being built it would include two
boilers. Costs were therefore estimated for the
500,000 lb/hr plant and then divided by two to indi-
cate costs attributable to a single boiler. This
approach was taken in order that this report be
consistent with earlier analyses. For readers who
wish to determine capital costs for a single boiler
installation or for more than two boilers, the well-
known six-tenths factor has been found to apply to
equipment and construction of this type.
9.2 BASIS OF PERFORMANCE ESTIMATES
The analysis of limestone use has been based on the
experimental performance data obtained with the single
full-scale fluidized-bed module. This module has many
features in common, especially dimensions, with a
250,000 lb/hr shop-assembled boiler. The key dimensions
which are similar are bed height and cell width.
POPE EVANS AND ROBBINS
-------
116
2.1 Limestone vs Dolomite. Performance data were obtained
for both, dolomite and high calcium limestone. These
data indicated that dolomite (53% CaCC>3 ) was superior
in controlling emissions when the measure of superi-
ority was stoichiometric addition rate based on the
calcium content only. However, limestone and dolo-
mite are both sold on a weight basis with little re-
gard to chemical composition. So, although the cal-
cium in limestone is less effective than that in dolo-
mite, a much lesser total weight of limestone is re-
quired for a given S02 reduction. For this reason
the economic analysis is based on the data obtained
with the high ofvlcium limestone (97% CslCO^ ) .
2.2 Raw Stone vs Hydrate. Hydrates of limestone and dolo-
mite were also evaluated as an additive and these were
found to be slightly more effective than the raw stones.
However, as in the case of dolomite vs limestone noted
above, the cost of a ton of calcium delivered as the
hydrate is higher than the calcium delivered in the raw
stone. Since the slightly higher utilization of the
hydrate does not compensate for its much greater cost,
only the raw stone has been considered in the evalua-
tion .
2.3 Particle Size. Increased utilization of the raw stone
is found with decreasing particle size. This is illus-
trated in Figure 27 where S02 reduction is plotted
against particle size for constant additive rate. Since
the smallest particle size used, -325 mesh, gave the
best results, this size has been assumed for the economic
evaluation.
3 PERFORMANCE DATA
The reduction of sulfur dioxide emission from the full
scale module using 1359 limestone at a bed temperature
of 1600°F was noted earlier to be about the same as
that achieved in the pilot scale unit. It was also
noted that if percent reduction is plotted against stoi-
chiometric ratio similar values are found for both the
2.6% and 4.5% sulfur coals. This plot is given as
Figure 42, and is an average of the 1359 lines in
Figures 25 and 26.
When the stoichiometric ratio is converted to a weight
basis, pounds of limestone per 100 pounds of coal,
separate curves are generated for each coal. These
are given in Figures 43 and 44. For additional clarity,
the ordinate in these figures was converted to the
ratio--sulfur in emissions/sulfur input.
POPE EVANS AND ROBB1NS
SNISBOU ONV SNVA3
TTcJOrf
SULFUR DIOXIDE REDUCTION, %
Ou>oo-j
-------
10 20
LIMESTONE FEED RATE, LBS/100 LBS COAL
FIGURE 44. RATIO OF SULFUR EMISSION TO SULFUR INPUT VS LIMESTONE FEED RATE
FOR THE 2.6% S COAL
10 20 30
LIMESTONE FEED RATE, LBS/100 LBS COAL
FIGURE 43.
RATIO OF SULFUR EMISSION TO SULFUR INPUT VS LIMESTONE FLOW RATE
FOR THE 4.5% S COAL
-------
120
It was found that even without additive not all the
sulfur would appear as S02 in the flue gas. Typically,
10% of the input sulfur was found in the ash. There-
fore, Figures 43 and 44 show the ordinate at 90% with
a zero additive rate. These curves then form the basis
for the operating cost analysis in that they relate
sulfur emissions to the required weight of additive.
9.4 CAPITAL REQUIREMENTS FOR EQUIPMENT
9.4.1 Description of Raw Stone Feed System. Limestone must
be received, stored, preparea and injected, captured
and disposed of. As noted earlier, the major portion
of the system has been integrated with the coal han-
dling system, and therefore the size of the system is
relatively independent of the sulfur content of the
fuel. The limestone injection system is charged with
a storage silo increment, dust collector increment,
etc. (See Table XII). In some instances, pneumatic
ash-conveyors for example, the smallest system commer-
cially available would be used with or without lime-
stone addition.
A block outline of the combined coal/limestone/ash
handling system is given as Figure 45. The system
shown and the costs tabulated below are assumed
constant regardless of the sulfur content of the coal
and the degree of emission control required. Although
some capital cost reduction would be achieved for a
precisely sized system, it would be poor judgment for
the plant designers not to provide for use of high
sulfur coal even though use of a lower sulfur coal is
planned,' and for maximum emission control since doing
so would not affect the capital significantly.
9.4.2 Description of Dust Collector System. The reacted
limestone is carried out of the boiler, along with the
carbon rich fly ash in the flue gas. Collected in a
cyclone, the spent stone and fly ash are pneumatically
injected into the Carbon-Burnup Cell*. Here the car-
bon content of the fly ash is burned in an oxygen rich,
high temperature environment. Carried out once again
by flue gas, the spent limestone and fly ash are col-
lected in the secondary mechanical collector for dis-
posal. Depending on local regulations regarding
The Carbon Burnup Cell is an integral component of a
fluidized-bed boiler in which the relatively unreactive
carbon remaining in fly ash can be burned so as to improve
the boiler's efficiency. It is fully described in U.S.
Patent 3,508,506.
POPE, EVANS AND ROBBINS
121
TABLE XII. SUMMARY OF CAPITAL COST COMPONENTS FOR LIMESTONE
ADDITION PER BOILER. 500,000 LB STEAM/HR PLANT
CONSISTING OF TWO 250,000 LB/HR COAL-FIRED,
FLUIDIZED-BED BOILERS
Line
1 Incremental site improvements $ 1,000
2 Incremental unloading hopper,
storage silos, transfer belt
and bucket elevator 7,000
3 Surge hopper 2,000
4 Dryer, pulverizer and classifier 38,000
5 Storage hopper 2,000
6 Incremental mechanical handling and
injection systems 10,000
7 Incremental dust collector costs 8,000
8 Incremental ash handling and storage 5,000
9 Controls and instruments 10,000
10 Miscellaneous steel 5,000
11 Incremental electrical, mechanical,
utilities, etc. 10,000
Subtotal (Lines 1 through 11) $ 98,000
13 Contingency @ 10% of Line 12 9,800
14 Total (Line 12 + Line 13) $107,800
POPE, EVANS AND HOBBINS
-------
123
particulate emissions, an electrostatic precipi-
tator* or wet scrubber may also be required. It
should be assumed that near urban areas regulations
will require particulate emissions on the order of
0.2 lbs per 106 Btu input.
The ash content of the coal is assumed to be 10%
and 7% for the 4-5% S and 2.6% S coals respectively.
The total particulate matter emanating from the com-
bustion of each coal is snown in Figures 46 and 47
as a function of additive feed ratio. Curves which
show the probable variation in precipitator load
were added. The curves assume all ash goes over-
head, 40% utilization of CaO, 85% efficiency on
the mechanical collector and 10% carbon in the fly
ash. Omitted is bed material attrition which may
add to the particulate load.
Discussions with precipitator manufacturers failed
to provide a basis on which to estimate the costs
of additional capacity requirements due to limestone
addition. It appears that the resistivity of fly
ash increases when S03 is not present. However,
carbon in the fly ash may compensate so that pre-
cipitator efficiency will not be seriously impaired.*
The preferred method of defining precipitator require-
ments is to use one of the portable or pilot precip-
itators owned by precipitator manufacturers .
Cyclone collector costs are relatively independent
of dust loading except that an increment has been
provided for heavy duty construction, increased
hopper capacities and increased unloading capacities.
The size distribution and composition of the fly
ash emanating from a fluidized-bec boiler is now
under study. Some preliminary work has indicated
that about 95% of the particles leaving the com-
bustor are collectable in a low efficiency mechanical
collector. Of the particles bypassing the collector,
99.9% were under 20 microns.
Ash is moved to the ash section of the common silo
via a pneumatic conveyor. Except for the increased
silo capacity requirement due to the added limestone
essentially no capital cost increase is required for
ash disposal.
*See Appendix A, Enclosure 45 for proposed design
arrangement for minimizing electrostatic precipitator
costs.
POPE. E\^IMS AND FSOBBllsIS
-------
9000
02
X
w
ta
w
Eh
X
.J
Ck
TOTAL FLY ASH
CYCLONE COLLECTION
PRECIPITATOR
-LOAD
10 20
LIMESTONE FEED RATE, LBS/100 LBS COAL
FIGURE 47.
TOTAL FLY ASH RATES FOR FULL LOAD OPERATION OF A 250,000 LB PER HR.
FLUIDIZED-BED BOILER WITH LIMESTONE ADDITION (2.6% S COAL)
TOTAL FLY ASH
CYCLONE COLLECTION
LIMESTONE FEED
PRECIPITATOR LOAD
10 20
LIMESTONE FEED RATE, LBS/100 LBS COAL
FIGURE 46. TOTAL FLY ASH RATES FOR FULL, LOAD OPERATION OF A 250,000 LB PER HR.
FLUIDIZED-BED BOILER WITfl LIMESTONE ADDITION (4.5% S COAL)
-------
126
9.4.3 Summary of Capital Costs. The cost data which ap-
pears in Table XII was based on manufacturer's in-
formation where applicable and on published estimates
for components and systems. Outdated information was
adjusted using the well-known Marshall and Stevens
Equipment Cost Index. As noted earlier, these costs
represent one-half the incremental cost of including
limestone addition in the design and construction of
a new plant containing two 250,000 lb/hr fluidized-
bed boilers.
9.5 ANNUAL OPERATING COSTS
The major element of operating cost is the delivered
cost of the raw limestone. Other components of oper-
ating costs are incremental labor costs, incremental
maintenance costs, increased disposal costs, power
costs for pulverizing, recovery of capital, taxes and
insurance and a small cost for the thermal effect of
limestone additions.
9.5.1 Delivered Cost of Limestone. Delivered costs of
limestone are variable, as are the costs of coal,
and dependent on plant location, rate of consump-
tion, mode of transportation, and market conditions.
The most definitive evaluation of limestone econom-
ics, by TVA, assumed a cost of $2.05 per ton for
crushed limestone. Studies by Esso Research and
Engineering and A. M. Kinney, Inc. also used this
limestone cost. This cost, as in the TVA study, is
a $1.35 per ton vendor's cost and a $0.70 per ton
shipping cost.
The same value, based on $2.05 per ton, will be
assumed in this evaluation, although costs above or
below this value may be found to be more appropri-
ate in an actual investment analysis.
9.5.2 Incremental Labor Cost (Plant Handling). Two men
are employed in the 500,000 lb/hr steam plant as
coal and ash handlers. They both work during the
day shift, five days per week. During other periods,
materials are drawn from live storage. No increase
in staffing requirements is anticipated as a result
of the decision to use limestone injection. To ac-
count for occasional overtime, however, a cost of
$0.15 per ton of limestone has been allocated for
plant handling.
POPE. EVANS AND HOBBINS
127
A watch supervisor and a watch fireman monitor the
operation of the plant's two boilers and auxiliaries.
No extra watch positions are required due to the addi-
tion of limestone.
9.5.3 Incremental Power Costs (Pulverizing). The only sig-
nificant power requirement because of limestone addi-
tion is that due to the pulverizer. An evaluation of
limestone grinding by A.M. Kinney, Inc. indicated less
than 35 kwn per ton of stone, while TVA's evaluation
indicates that ^43 kwh per ton of limestone is re-
quired to grind to 99%, -325 mesh. At a conservative
$0.009/kwh the power cost would be on the order of
$0.32 to $0.39 per ton of stone. A cost of $0.40 per
ton will be used in this analysis.
9-5.4 Thermal Effect. When limestone utilization approaches
about 40%, it is possible to realize a net thermal
gain from limestone injection. Depending on the cost
of coal, the method of drying, the exit gas temperature
and the precise degree of utilization, costs of from
1$ to 54 per ton of raw stone might be used for this
factor. This analysis will use 5C per ton of raw
limestone for thermal effect.
9.5.5 Incremental Maintenance Costs. In many economic
analyses, annual maintenance is simply assumed at
2 - 5% of capital. In this study, the incremental
maintenance costs are assumed to be made up of a
fixed portion, and a value dependent upon throughput.
For the fixed portion, 2»s% of the incremental invest-
ment will be used. For the tonnage dependent portion,
a value of $0.20 per ton will be used to account for
pulverizer wear. The sum of these two factors will
exceed 5% of capital for several of the cases analyzed
below.
9.5.6 Disposal Costs. Fly-ash disposal costs are the most
variable ingredient in any industrial coal-fired
boiler cost analysis. Costs may vary between $0.00
per ton to $1.00 per ton depending on local market
conditions for fly ash or the distance to a landfill.
Ash disposal is often by sluice to a fill area. In
this case, ash disposal costs are more properly ex-
pressed as a capital cost (sei TVA's treatment, for
example). For this evaluation, a cost of $0.25 per
ton of raw stone will be assumed to be borne by the
steam plant.
POPE. EVANS AND ROBBINS
-------
128
9.5.7 Summary of Annual Operating Cost Ingredients. Shown
below is a summary of the cost ingredients for the
annual operating cost analysis. The costs are divided
into two categories—fixed costs, independent of the
degree of emission control, and variable costs which
are proportional to the degree of control.
TABLE OPERATING COST INGREDIENTS SUMMARY
FOR FINE LIMESTONE INJECTION IN A
500,000 LB/HR FLUIDIZED-BED BOILER PLANT*
A. Fixed Costs
1. Interest, depreciation,
taxes and insurance
@ 14% of $107,800
2. Maintenance
@ 2-1/2% of $107,800 =
Total
$15,100/annum
2,700
$17,800/annum
Fixed cost per ton of coal,
13 tph x 6,000 hrs/yr. - $17,800/(13 x 6,000)
$0.23/ton of coal
B. Variable Costs
1. Limestone, 1/4" x 0,
vender's price
2. Shipping
3. Power for pulverizing
4. Thermal effect
5. Incremental maintenance
6. Disposal
Total
$/Ton of Limestone
1.35
.70
.40
.05
.20
.25
2.95
Made up of two 250,000 lb/hr boilers
POPE EVANS A.ND ROBEZNS
129
9.5.8 Annual Operating Costs. Applying the cost data of
the previous section and the limestone requirements
for the two coals (4.5% S and 2.6% S) from Figures 43
and 44, the annual operating costs are shown below.
The two measures of performance (emission equivalent,
% S, and S02 removed, %) are two ways of expressing
the same thing. Emission equivalent, % S, is related
to % S02 removed by the equation:
Emission equivalent, % S
(100 - SO2 removed, %) x (Actual % S in coal)
= _ :
TABLE XIV. ANNUAL OPERATING COST DATA FOR FINE LIMESTONE
INJECTION
Case
1. For the 4.5% Sulfur Coal
1.
Emission equivalent, % S
3.5
2.5
1.5
1.0
0.6
2.
SO2 removed, %
2?
45
67
78
87
3.
Additive rate, Tons of
limestone/ton to coal
.037
.12
.21
.28
.37
4.
Fixed cost, $/ton of coal
.23
.23
.23
.23
.23
5.
Variable cost, $/ton
of coal
.11
.35
.62
.83
1.09
6.
Total Cost*,
$/ton of coal (4 + 5)
.34
.58
.85
1.06
1.32
Case
2. For the 2.6% Sulfur Coal
1.
Emission equivalent, % S
2.0
1.5
1.0
0.6
2.
SO2 removed, %
23
42
62
77
3.
Additive rate, Tons of
limestone/ton to coal
.028
.065
.105
.155
4.
Fixed cost, $/ton
of coal
.23
. 23
.23
.23
5.
Variable cost, $/ton
of coal
.08
.19
. 31
.46
6.
Total Cost*,
$/ton of coal (4 + 5)
. 31
.42
.54
.69
^These results are plotted in Figures 48 and 49.
POPE EVANS AND ROBB1NS
-------
130
§
O
b
O
z
o
Eh
p$
W
04
0
Q
1
E-t
CO
O
o
1.40
1.20
1.00
.80
.60
AO
.20
\
\
<~AC
rUAL
PERFOI
*MANCI
2
UTI1
jIZAT"
PW0 T]
JAL PI
AT 1
ACTl
[MES
:rforj
1ANCE
TOTAL EMISSION
EQUIVALENT, % SULFUR
FIGURE 4E
T2TAL C0ST 0F CONVERTING A HIGH SULFUR (4.5%)
COAL TO A LOWE?. SULFUR COAL EQUIVALENT BY LIMESTONE ADDI-
TION TO A 250,000 LB/HR FLUIDIZED-BED BOILER
POPE, EVANS AND ROSBINS
131
1.00
.80
.60
.40
20
\
actu;
VL PEF
iform;
VNCE
UTILII
2 TIM3
2ATIOI
:s ACr
U1ANC1
* AT_
rUAL
J
PERFOI
I 2
TOTAL EMISSION EQUIVALENT, % SULFUR
FIGURE 49. ESTIMATED TOTAL COST OF CONVERTING A MEDIUM SULFUR (2.6%)
COAL TO A L0V7ER SULFUR COAL EQUIVALENT BY LIMESTONE ADDI-
TION TO A 250,000 LB/HR FLUIDIZED-BED BOILER
POPE. EA/ANS AND ROBBINS
-------
132
The results of one additional case is also plotted on
these figures. A lower curve shows the costs for a
utilization rate twice as good as that actually found
experimentally in the fluidized-bcd boiler. This indi-
cates a hypothetical lower limit to costs in a fluid-
ized-bed boiler if additional research reveals methods
of achieving a utilization on the order of 80%. Re-
sults on this order have been reported by British ex-
perimenters .
9.6 COMPARISON WI'iH COGTS F'OR ALTERNATIVE METHODS
9.6.1 Use of Low Sulfur Coal. The use of low sulfur coal
may be an economical alternative to limestone addi-
tion where low sulfur coal is locally available.
Where this coal is not available locally, it must be
shipped and this may markedly increase its cost.
This is the case for Chicago, as an example, where
the low sulfur coal might come from West Virginia.*
Table XV presents the costs for burning three "local"
coals in a fluidized-bed boiler with limestone addi-
tion and costs for burning an "imported" low sulfur
coal in the same boiler without limestone. The costs
for limestone addition are derived from the test pro-
gram performance curves.
It is clear from this comparison that, for the case
estimated, the cost of energy is less with limestone
injection than with the low sulfur coal.
9.6.2 Limestone Injection into Conventional Boilers. Dry
limestone injection into conventional boilers may be
somewhat less effective than injection into a fluidized-
bed boiler. Until test results from operation, full-
scale units of both designs are available, no economic
comparisons are meaningful.
9.6.3 Other Flue Gas Control Processes. A number of survey
articles have been published reviewing the costs of
the alternative stack gas cleaning processes.
Almost all of this work pertains to large utility
boilers, not industrial boilers, and is therefore not
truly comparable to the data presented above. In
every case, the capital costs are significantly higher
than for limestone injection and would be more unfavor-
able when reduced in scale. Some process developers
claim a profit on operations when markets exist for the
sulfur form produced and other factors are favorable.
*
Low sulfur coal from Wyoming is presently being brought
into Chicago by Commonwealth Edison and with shipping
costs alone exceeding ^»B.OO per ton.
POPE EVANS ANO ROBBINS
9ri:Eaow qnv snwvh a.dod
o (-3
3
o o
Di
& rt
{-> ffc
I-"
O
n o
o
I- Cn
W
H- rt
C
3
t—'
fD rt
M
CO 0
H-
rt
0 'O
&>
3 H
3
O
a
3 O
1—'
<-!• a>
CD
a
-4
rt 3
O
p. h-
O
0 CO
3 CO
H*
o
3
CO
U)
CD
tQ
C
H-
<
&>
M
H>
vj
3
rt
CO
-J
rt
O
ft
U)
3*
U)
o
U>
<30
CO
£
c
-J
l-t
o
o
fi>
XT
C
CO
(D
O
Hi
.fck
o
H1
o
«
CO
G
1—•
hh
u>
C
-J
o
u>
a
o
13
n
o
c
3
rt
H
co < as
OJ
rt pj
(D
•»
C H"
D>
CO
\ c
rt
o
H-
o
cr-
3
iQ
o w
o
C
o
3
h-'
Cfp rt
hh
•vl
(D
C
3
rt
H
¦o
h->
\
Q
rt
0
Cj
0
01
3
1—'
o
O
o
41
CO
ft
CO
o
ft
N)
c
33 W
w
0) (D
c
lQ & H
to
c c
Ml
0>
H- O
a
(A
H rt
n
0
(D H-
3 0
ro 3
3
rt
CO
dP
-o O
H f
CO
\ 0 3 H-
Si » ci g
p>
CD rt (D
(0
rt 0 co
(I)
C rt rt
H- 0
O 3
3
o *3
41
•
ts ai
O
rt i—'
C *
en
-------
13 4
This could never be the case for pulyerized limestone
injection. Continuing research on flue-gas cleaning
processes may provide a process applicable to indus-
trial boilers of either the conventional or fluidized-
bed design.
9.7 CONCLUSIONS
Review or the results outlined above lead to the follow-
ing observations:
1. Limestone injection to a fluidized-bea boiler couJd
be used at a reasonably low capital cost ( $3/kw)
when the limestone system is treated as an integral
part of the steam supply system.
2. Operating costs for limestone injection to a fluidized-
bed boiler will be a small multiple of the raw stone
cost (1*1.5) when the plant design is such that in-
creased labor requirements are avoided.
3. In those areas where coal enjoys a natural cost ad-
vantage over natural gas, a fluidized-bed boiler
with limestone injection may provide the plant owner
with an economically feasible method of providing
steam and complying with local air quality regula-
tions. Conventional boilers may not, in many cases,
be able to provide such a feasible alternative.
4. One final conclusion is warranted, in part by the
results discussed above and in part by information
recently published by the Federal Power Commission
on declining gas reserves: when the investment
appraisal techniques utilized by a potential boiler
plant owner provide for a sophisticated treatment
of cost trends, coal-fired, fluidized-bed boilers
utilizing limestone injection may appear favorable
even when coal does not currently enjoy a natural
cost advantage.
POPE, EVANS AND ROBBINS
APPENDIX A
ENCLOSURES
POPE. EVANS AND ROBBINS
------- |