EPA-600/2-77-235
November 1977
Environmental Protection Technology Series
LOW NOX COMBUSTION CONCEPTS
FOR ADVANCED POWER GENERATION
SYSTEMS FIRING LOW-BTU GAS
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
rep0urts of the Office of Research and Development, U.S. Environmental Protection
fn*6H ee,T 9rPuped Tto fjve series- These f ive broad categories were established to
• development and application of environmental technology. Elimination of
n9rf •"?' WrWSf C0"sc'9us|y planned to foster technology transfer and a maximum
in related fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5 Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION TECHNOLOGY
fione«o^nm«nrteaS ^^ ®2 r?*eB[ch Performed to develop and demonstrate instrumenta-
«?H ' Sn E •e?t| d meth,odo ??/ to r®Pa'r or prevent environmental degradation from point
SJlESTF01?! Sourc;es1of pollution. This work provides the new or improved technology
standa d treatment of pollution sources to meet environmental quality
EPA REVIEW NOTICE
reviewed by the U.S. Environmental Protection Agency, and aoDroved
>val does not signify that the contents necessarily reflect the views and
nor does mention of trade names or rnmmorHai r\r<-.Hn<%«» MM__«:*. .*._
trade names or commercli
Spring«eTvTrglnaaa22ai61.'° '^ PUb"C 'hr°U9h the NatiOr"" Technical 'n'ormation Service.
-------
EPA-600/2-77-235
November 1977
LOW NOX COMBUSTION CONCEPTS
FOR ADVANCED POWER GENERATION SYSTEMS
FIRING LOW-BTU GAS
by
T.J. Tyson. M.P. Heap, C.J. Kau,
B.A. Folsom, and N.D. Brown
Energy and Environmental Research Corp.
8001 Irvine Boulevard
Santa Ana, California 92705
Contract No. 68-02-1361
ROAP 21ADO-BA
Program Element No. 1AB013
EPA Project Officer: G. Blair Martin
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, N.C 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, D.C. 20460
-------
TABLE OF CONTENTS
Section Page
SUMMARY xi
I INTRODUCTION 1
2 FUEL CHARACTERISTICS 5
2.1 Gasification Processes 5
2.1.1 Gasifier Types 5
2.1.2 Specific Gasification Processes 7
2.2 Properties of LBG 13
2.3 LBG Product Gas Cleanup 16
2.3.1 Particulate Removal Systems 16
2.3.2 Sulfur Removal Systems 16
2.3.3 Bound Nitrogen Species Removal Systems 21
2.4 Product Gas Combustion Characteristics 21
2.4.1 Flame Stability 21
2.4.2 Furnace Heat Transfer 22
2.5 Low Btu Gas Combustion - Pollutant Emissions 25
3 ADVANCED POWER GENERATING SYSTEMS ANALYSIS .... 27
3.1 Separate Gasifier/Power Plant Comparative Analysis .... 28
3.1.1 Preliminary Screening • 29
3.1.2 System Design 32
3.1.3 Plant Optimization and Performance 41
3.1.4 Capital and Operating Cost Estimates 44
3.1.5 Heat Transfer Surface Requirements 52
3.1.6 Burner Designs 54
3.1.7 System NOX Emission Assessment 56
3.2 Integrated Gasifier/Power Plant Analysis 59
3.2.1 An Overview of COGAS System Efficiency 60
3.2.2 Gasifier Losses 63
3.2.3 Supercharged Versus Fired Exhaust Boilers 67
3.2.4 Generalized Approach to Cycle Analysis 72
3.3 Conclusions and Combustor Definitions for NOX
Emission Studies 80
ill
-------
Section
TABLE OF CONTENTS (Continued)
ESTIMATES OF NOX EMISSIONS FROM LBG COMBUSTORS 83
4.1 Methodology for Estimating NOX Emissions 83
4.2 General Characteristics of Combustors and
Selected Fuel 86
4.3 Adiabatic Gas Turbine Generator Results 92
4.3.1 General Flame Characteristics 92
4.3.2 Premixed Lean Combustion 93
4.3.3 Staged Combustion 100
4.4 Supercharged Boiler Results 138
4.4.1 General Flame Characteristics 140
4.4.2 Premixed Lean Combustion 140
4.4.3 Staged Combustion 144
5 CONCLUSIONS 171
REFERENCES 173
APPENDIX A ALTERNATIVE CONVENTIONAL AND COMBINED CYCLE SYSTEMS 175
APPENDIX B THE PREDICTION OF FURNACE PERFORMANCE FOR TANGENTIALLY-
FIRED UTILITY BOILERS 211
APPENDIX C KINETIC MECHANISM OF NOX FORMATION IN LOW BTU GAS
COMBUSTION 217
REFERENCES 230
IV
-------
LIST OF ILLUSTRATIONS
Figure
2-1 Schematic Diagrams of LBG Coal Gasi tiers ............ 6
2-2 Lurgi Gasifier Schematic Diagram (from Refs. 2 and 3) ..... 8
2-3 Schematic Diagram of Koppers Totzek (K-T) Entrained
Flow Gasifier ......................... 1U
2-4 Schematic Diagram of BCR Entrained Flow Gasifier (from Ref. 9) . 12
2-5 Schematic Diagram of Westinghouse Fluidized Bed Gasifier
(from Ref. 10) ......................... L*
3-1 Conventional Steam Plant, System Al, Lurgi LEG, Low
Temperature ..........................
3-2 Conventional Steam Plant, System Al, Koppers Totzek MBG,
Low Temperature ........................
3-3 Conventional Steam Plant with Gas Precooler, System Bl,
Lurgi LBG ... ........................ JD
3-4 Conventional Steam Plant with High Temperature Burners,
System Cl, Lurgi MBG, High Temperature ............. J/
3-5 Supercharged Boiler Combined Cycle, System HI, Lurgi LBG .... 38
3-6 02 Blown Steam Plant, System 15, Koppers Totzek MBG ...... 39
3-7 Side Elevation Showing Detail of Supercharged Boiler ^
and Burner ...........................
3-8 Plant Heat Rate as a Function of Furnace Pressure and Turbine ^
Inlet Temperature .......................
3-9 NOX Emissions Predicted by Combustion Engineering's Program
for Tangential Firing .....................
3-10a Unfired Steam Generator Design .................
3-10b Typical Gas and Steam Temperatures Unfired Steam Generator ... 70
3-11 Supercharged Boiler Combined Cycle Efficiency as a Function ^
of Excess Air .........................
3-12a Gas Turbine Plus Unfired Steam Generator .... ........ 75
3-12b Gas Turbine Plus Furnace-Fired Steam Generator ......... 75
3-12c Gas Turbine Plus Supplementary- Fired Steam Generator ...... 76
3-12d Supercharged Furnace-Fired Steam Generator Plus Gas Turbine . . 76
3-13 Brown Boveri Industrial Gas Turbine with External Combustor . . 77
3-14 Generalized Thermodynami c Cycle Analysis Flow Diagram ..... 79
4-1 Examples of Basic Element Coupling for Limit-Case
Investigations .........................
-------
LIST OF ILLUSTRATIONS (Continued)
Figure Page
4-1 Examples of Basic Element Coupling for Limit-Case
Investigations (Continued) 87
4-2 COGAS Power Plant with Adiabatic Gas Turbine Combustor 89
4-3 COGAS Power Plant with Supercharged Boiler 90
4-4 Premixed Adiabatic Gas Turbine Combustor - MKAP Analog
Schematic 94
4-5 NO from Lean Premixed Combustion —Adiabatic Gas Turbine
Combustor — No Fuel Ammonia — = 0.45 95
4-6 NO from Lean Premixed Combustion — Adiabatic Gas Turbine
Generator - No Fuel Ammonia - = 0.53 98
4-7 Superequilibrium 0 Concentration, Lean Premixed Combustion —
Adiabatic Gas Turbine Combustor 99
4-8 NO from Lean Premixed Combustion - Adiabatic Gas Turbine
Combustor — With Fuel Ammonia 101
4-9 Staged Adiabatic Gas Turbine Combustor — MKAP Analog Schematic . 102
4-10 NO Formation and Destruction in Well-Stirred Cfty -Air
Reactor with 1300 ppm NO Addition 104
4-11 NO Formation and Destruction in Well-Stirred Reactor with
1300 ppm NH3 Addition 105
4-12 Nitrogen Species from Premixed Rich Primary Zone — Adiabatic
Gas Turbine Combustor - $ = 2.00 110
4-13 Nitrogen Species from Premixed Rich Primary Zone - Adiabatic
Gas Turbine Combustor - = 1.67 Ill
4-14 Nitrogen Species from Premixed Rich Primary Zone — Adiabatic
Gas Turbine Combustor - = 1.33 112
4-15 Nitrogen Species from Rich Primary Zone —Adiabatic Gas
Turbine Combustor — = 1.10 113
4-16 Nitrogen Species from Premixed Lean Primary Zone —Adiabatic
Gas Turbine Combustor - = 0.90 114
4-17 Sum of Residual Nitrogen Species from Premixed Plug Flow
Primary - Adiabatic Gas Turbine Combustor 117
4-18 Sum of Residual Nitrogen Species from Premixed Plug Flow
Primary - Adiabatic Gas Turbine Combustor - N2 in Air
Replaced by Argon 119
4-19 Effective (NHa) Conversion Ratio for Premixed Rich Primary
Reactor - Adiabatic Gas Turbine Combustor 120
4-20 Nitrogen Species from Rich Primary Zone with Early Heat
Removal -Adiabatic Gas Turbine Combustor - = 1.67 122
vi
-------
LIST OF ILLUSTRATIONS (Continued)
Figure Page
4-21 Nitrogen Species from Rich Primary Zone with Late Heat
Removal -Adiabatic Gas Turbine Combustor — = 1.57 123
4-22 Nitrogen Species from Rich Primary Zone with Early Heat
Removal —Adiabatic Gas Turbine Combustor — $ = 1.33 124
4-23 Nitrogen Species from Rich Primary Zone with Late Heat
Removal —Adiabatic Gas Turbine Combustor — = 1.33 126
4-24 Nitrogen Species from Rich Primary Zone, Stirred/Plug Flow
Comparison — Adiabatic Gas Turbine Combustor 127
4-25 Nitrogen Species from Lean Primary Zone with Varying Stirred
Reactor Residence Time — Adiabatic Gas Turbine Combustor .... 128
4-26 Nitrogen Species from Premixed Rich Primary Zone —Adiabatic
Gas Turbine Combustor - Effect of Pressure 129
4-27 Nitrogen Species from Premixed Lean Primary Zone — Adiabatic
Gas Turbine Combustor — <{> = 0.90 130
4-28 Nitrogen Species from Premixed Rich Combustion -Adiabatic
Gas Turbine Combustor - Effect of Fuel Methane Content 132
4-29 Nitrogen Species from Premixed Lean Combustion - Adiabatic
Gas Turbine Combustor - Effect of Fuel Methane Content 133
4-30 Staged Adiabatic Gas Turbine Combustor - MKAP Analog Schematics
for Alternative Secondary Stage Arrangements 134
4-31 Staged Adiabatic Gas Turbine Combustor - MKAP Analog Schematic
for Simulated Diffusion Flame 136
4-32 NO Concentration and Temperature for Alternative Secondary
Stage Configurations —Adiabatic Gas Turbine Combustor 137
4-33 NO Concentration for Optimum Staged Adiabatic Gas Turbine
Combustor 139
4-34 Adiabatic Flame Temperature for LBG and Combustion Air for
Supercharged Boiler 141
4-35 Equilibrium NO Concentrations —Supercharged Boiler 142
4-36 NO Concentration and Temperature for Premixed Lean Combustion
- Supercharged Boiler 143
4-37 Staged Supercharged Boiler - MKAP Analog Schematic 145
4-38 Nitrogen Species and Temperature in Rich Primary Zone -
Supercharged Boiler - = 1.50 146
4-39 Nitrogen Species and Temperature in Rich Primary Zone -
Supercharged Boiler - = 1.40 147
4-40 Nitrogen Species and Temperature in Rich Primary Zone -
Supercharged Boiler - = 1.33 148
vii
-------
LIST OF ILLUSTRATIONS (Continued)
Figure Page
4-41 Nitrogen Species and Temperature in Rich Primary Zone —
Supercharged Boiler — = 1.15 149
4-42 Sum of Residual Nitrogen Species from Premixed Plug Flow
Primary -Supercharged Boiler 151
4-43 Nitrogen Species and Temperature in Rich Primary Zone -
Supercharged Boiler (NH3)Q =0 152
4-44 Nitrogen Species and Temperature in Rich Primary Zone -
Supercharged Boiler (NH3)Q = 250 ppm in Fuel 153
4-45 Nitrogen Species and Temperature in Rich Primary Zone -
Supercharged Boiler (NH3)Q = 1000 ppm in Fuel 154
4-46 Nitrogen Species and Temperature in Rich Primary Zone -
Supercharged Boiler - Reduced Pressure, P = 1.0 atm 156
4-47 Nitrogen Species and Temperature in Rich Primary Zone -
Supercharged Boiler -Reduced Pressure, P = 0.1 atm 157
4-48 Nitrogen Species and Temperature in Rich Primary Zone -
Supercharged Boiler- Ignition TSR = 10 msec 159
4-49 Nitrogen Species and Temperature in Rich Primary Zone -
Supercharged Boiler - Ignition TSR = 100 msec 160
4-50 Nitrogen Species and Temperature in Rich Primary Zone -
Supercharged Boiler - = 1.15 161
4-51 Nitrogen Species and Temperature in Rich Primary Zone -
Supercharged Boiler-No Fuel Methane 162
4-52 Nitrogen Species and Temperature in Secondary Stage -
Supercharged Boiler-Air Added Over 125 msec 164
4-53 Nitrogen Species and Temperature in Secondary Stage -
Supercharged Boiler-Air Added Over 500 msec 165
4-54 Nitrogen Species and Temperature in Secondary Stage —
Supercharged Boiler — Stirred Reactor Primary Stage 167
4-55 Nitrogen Species and Temperature in Primary and Secondary
Stage - Supercharged Boiler 168
4-56 Nitrogen Species and Temperature in Secondary Stage —
Supercharged Boiler (NH~)0 =0 169
B-l Furnace Schematic Showing Volume Modeled by Lower
Furnace Program 212
B-2 Lower Furnace Program Recirculating Flow Model 213
C-l Comparison of Measured(6) and Predicted NOX Concentration
at the Exit of a Stirred Reactor 227
C-2 Measured(S) and Predicted NO Retention in a Stirred Reactor ... 228
C-3 Measured(6) and Predicted Conversion of NH3 to NOX in a
Stirred Reactor 229
viii
-------
LIST OF TABLES
Table
2-1 Characteristics of LBG ..................... 15
2-2 Characteristics. of Low Temperature Cleanup Systems (after
Colton et al(14)) ....................... I/
2-3 Characteristics of High Temperature Cleanup Processes
(after Colton et aid4)) .................... ^
OQ
3-1 Fuel Gas Compositions .............. ....... to
3-2 Summary of Preliminary System Concept Analysis ......... 30
3-3 Summary of Systems Studied During Critical Analysis ...... 33
40
3-4 Major Components ........................
45
3-5 Plant Performance .......................
3-6 Plant Capital Costs ......................
4Q
3-7 Summary of Operating Cost Estimates .............. ^
3-8 Energy Cost Computations . . . ................. 50
3-9 Summary of Heat Transfer Surface Requirements ......... 53
3-10 Furnace Design for Burning Offgas from Coal Gasifiers -
MCR Burner Design Parameters ..................
91
4-1 Combustor Parameters ......................
4-2 LBG Composition Assumed for NOX Study ............. 92
221
C-l Kinetic Rate Constants .....................
ix
-------
SUMMARY
Several advanced power generating concepts firing low Btu gasified coal
were analyzed and the combined gas and steam cycle power plant with integrated
gasifier was identified as the most promising from fuel utilization and eco-
nomic points of view. Two representative combined cycle systems were chosen
for detailed nitrogen oxide emission analysis: (1) an advanced-technology
high-temperature gas turbine with a waste heat boiler; and (2) a supercharged
boiler with a current-technology gas turbine. Nitrogen oxide emissions were
investigated using a kinetic model which included over 100 reactions. The
model was validated by comparison with the best available experimental data
and then applied to idealized combustor configurations.
Staged combustion schemes involving rich primary zones and controlled
mixing secondary zones were found to minimize thermal nitrogen oxides and
nitrogen oxides produced from ammonia in the fuel gas. The minimum calcu-
lated nitrogen oxide levels were 150 ppm for the high temperature turbine
case with an equivalence ratio of 0.45 and 4000 ppm of fuel ammonia, and v
125 ppm for the supercharged boiler with five percent excess air and 500 ppm
of fuel ammonia. Both these calculations refer to uncorrected concentrations
measured at the stated equivalence ratio. While these results need to be
verified experimentally, they show the potential for nitrogen oxide emissions
within the Federal New Source Performance Standards without requiring ammonia
removal from the fuel gas.
XI
-------
1.0 INTRODUCTION
Although the United States has an abundant supply of coal, the
utilization of this resource has been hindered by the low competitive cost
of petroleum, its availability from foreign sources and its proven technology.
However, the recent rise in petroleum prices and the vulnerability of foreign
supply to international politics encourage the development of domestic coal
resources as a petroleum alternative.
Unfortunately, coal causes more environmental problems than petroleum.
One of the most severe problems is the high sulfur content of coal and the
corresponding sulfur oxide emissions. Much of the recent coal-related
research has focused on this problem and several approaches have been
explored: direct coal combustion with stack gas scrubbers to remove sulfur
oxides, use of sorbants for sulfur oxide control in fluidized bed combustors,
and the conversion of coal into low sulfur gaseous or liquid fuels. While
each of these options can potentially decrease emissions of sulfur oxides to
acceptable levels, the total environmental impact of each option from coal
extraction to final end use must be considered. This report examines one
aspect of this problem, the nitrogen oxide emissions from advanced power
systems fired with low Btu gasified coal (LBG).
The primary justification for using low Btu gas (LBG) in advanced power
systems is the potential for reduced sulfur emissions with improved thermo-
dynamic performance over conventional coal-fired power plants. Technological
and economic obstacles limit conventional plants without sulfur recovery to
overall efficiencies (coal pile to buss bar) in the range of 37 to 40 percent.
However, essentially all of the sulfur entering such plants in the form of
coal exits in the stack gas in the form of S02- Federal New Source Perfor-
mance Standards limit S02 emissions from coal-fired steam boilers to
1.2 pounds of S02 per 106 Btu heat release. This corresponds to 0.6 weight
percent sulfur in a coal with a heating value of 10,000 Btu/lb. Compliance
with this regulation requires either low sulfur coal or sulfur removal from
the coal or stack gas. However, low sulfur coal is in short supply and
sulfur removal or stack gas cleaning can lower overall efficiency to the
-------
low 30 percent range. These factors have motivated a search for alternate
solutions which are more efficient and environmentally acceptable methods of
coal utilization.
Low Btu coal gasification offers such an alternative solution. Since
the gasification process operates fuel-rich, the sulfur in the coal is con-
verted mainly to hLS in the fuel gas which can be removed more easily than
SO-. The HpS concentration is high and the total volume of gas to be treated
is small in comparison with S02 scrubbing.
The production of LEG has the disadvantage that the energy losses in
the gasification and sulfur cleanup processes may amount to more than
25 percent of the coal's heating value. These losses must be countered with
improvements in integrated gasification and power system design if the overall
energy conversion is to be economically competitive with conventional direct-
fired power plants with stack gas sulfur removal. One major energy loss is
the sensible heat of the hot LEG leaving the gasifier. This heat, amounting
to 10 to 15 percent of the coal's heat of combustion, is lost or used
inefficiently if the sulfur cleanup must be carried out at low temperatures.
The use of a high temperature cleanup process offers reduced losses since the
sensible heat adds to the heat released in final combustion. Modifying the
plant's thermodynamic cycle is another possible improvement. Although direct
coal conversion is limited by present technology to furnace firing, LBG can
be combusted in internal combustion engines such as gas turbines. The hot
turbine exhaust gases can also be used as a heat source for a steam cycle
giving rise to a combined gas and steam (COGAS) cycle. With turbine inlet
temperatures representative of current technology (2000°F) overall COGAS
plant efficiencies are comparable to conventional direct coal-fired plants
without sulfur recovery. With the promise of increased turbine inlet
temperatures the COGAS system can yield efficiencies significantly greater
than conventional power plants.
The environmental consequences of burning LBG in advanced power genera-
tion systems are not well-known. The precise composition of LBG depends upon
the coal characteristics and the details of the gasification process. The
-------
major fuel gases contain small amounts of hydrogen sulfide, carbonyl sulfide
and ammonia. Although methods of sulfur removal are well-documented the
fate of the ammonia is less certain. Robson, et al(1) estimate up to 600 ppm
of ammonia in a low Btu gas with low temperature cleanup and 3,800 ppm from
high temperature cleanup processes currently under development.
A substantial portion of the ammonia could be converted to nitric oxide
in the combustion process. Previous investigations of ammonia conversion in
combustion systems have shown that under premixed conditions the amount con-
verted is dependent upon mixture stoichiometry and can be as high as 80 per-
cent. In diffusion flames up to 50 percent of the ammonia can be converted
depending on the method of ammonia addition and the rate of fuel/air mixing.
Consequently, if LBG is combusted without ammonia removal the potential for
nitric oxide production may be greater than in the present coal.
This report investigates the combustion of LBG in advanced power systems
focusing on the rate of NOX formation from fuel nitrogen-bearing species such
as ammonia. Section 2 examines the characteristics of LBG gas. The alterna-
tive gasificiation processes, species distributions, heating value and other
fuel properties are summarized. Gas cleanup systems for removing sulfur
species, particulates and nitrogen-bearing species are also reviewed. The
design and performance of advanced power generating systems are documented
in Section 3. A comparative analysis of several alternative systems is pre-
sented and the combustion parameters for the most attractive systems are
determined.
In Section 4 the rates of NOX formation from these LBG combustion systems
are estimated. A kinetic model is applied to a simplified LBG fuel containing
carbon monoxide, hydrogen and methane with trace quantities of ammonia. A
series of "limit-cases" are evaluated based on operating constraints imposed
by practical systems. Although macro- and micro-scale mixing phenomena
dictate the NO emission levels in any practical system, this model esta-
blishes upper and lower emission limits and thus gives an indication of the
feasibility of direct use of high ammonia fuel gases.
3/4
-------
2.0 FUEL CHARACTERISTICS
Although identified by the single term low Btu gas, the characteristics
of the fuel are as varied as the air blown gasification schemes proposed for
its production. This section discusses the composition of coal-derived fuel
gases which could be used as fuels for power generation, their combustion
characteristics and pollutant production potential. These parameters of the
fuel gas depend to varying extent upon:
the feedstock;
the gasification process;
the thermodynamic state of the product gas; and
the method of product gas cleanup.
2.1 Gasification Processes
2.1.1 Gasifier Types
Low Btu Gas (LB6) is produced from coal through a partial pyrolysis
reaction where coal, steam and air combine in a gasifier pressure vessel to
make a hot combustible fuel gas with a heating value of approximately 150 Btu/
ft3. Three types of gasifiers have been used or proposed to be used to
generate LBG:
t Fixed bed
t Fluidized bed
t Entrained Flow Gasifier
Figure 2-1 compares the characteristics of each gasifier type. The fixed bed
is the oldest design and is basically a counterflow device. Air and steam are
blown upward through a packed coal bed in which the reaction occurs. As the
coal reacts, additional coal is continuously added at the top so that coal
particles move continuously downward through four reaction zones: drying,
devolatilization, gasification and char combustion. Ash is removed from the
bottom. The coal bed is continuously stirred to avoid coking and temperatures
are maintained low to prevent ash fusion. Typical LBG output temperatures are
in the range of 800 to 1200°F. The counterflow process and the low temperature
result in substantial carry-over of tars, oils and char.
-------
Coal
Counterf1ow
Coal Moves
Slowly
Downward,
Gases Move
Upward
Air
Coal
Steam
1500-2000°F
LBG
2500-3000°F
^
1
1 1
l\
y
/
Counterflow
Coal Mixes
Rapidly,
Gases Move
Upward
f •>
a
(y^
vV
\ /
Steam
Ash
Fluidized Bed
Parallel Flow
Gas and
Coal Move
Upward
Air
Steam
Coal
Entrained Flow
Figure 2-1. Schematic Diagrams of LBG Coal Gasifiers
-------
In fluidized bed designs the relative gas velocity is substantially
higher than in fixed bed designs, and the coal bed is turbulent with substan-
tial bottom-to-top mixing. As a first approximation the solid phase character-
istics may be assumed constant from bottom to top and the four reaction zones
are merged. This and the higher LBG output temperature (1500-2000°F) reduces
tar and higher hydrocarbon carry-over.
Entrained flow gasifiers have the highest gas phase velocity. All com-
ponents including the coal particles are introduced at the bottom and move
upward. Unlike the fixed and fluidized bed gasifiers, all constituents move
in essentially parallel flow. The high temperatures employed in entrained
flow gasifiers (2500-3000°F) essentially eliminate higher hydrocarbons.
2.1.2 Specific Gasification Processes
Gasification processes suitable for integration into a COGAS power plant
can be divided into two classes: current state-of-the-art processes which
have demonstrated commercial success, and second generation processes now in
the pilot plant stage. Three systems can be considered current state-of-the-
art and capable of restricted present day utilization:
t Lurgi
• Koppers Totzek
• Winkler
Lurgi
The Lurgi gasifier is a pressurized, stirred, fixed bed gasifier originally
developed in 1931. Figure 2-2 shows the essential design elements. Gasifica-
tion takes place in a countercurrent moving bed of coal at pressures above
300 psig^. Coal is fed intermittently through a pressurized coal lock, passes
downward through the reacting bed and exits as ash through a lock hopper on the
bottom.
Major operating problems center around the use of coking coals. A coal
distributor rotates beneath the bed surface to prevent the coal from bonding
together when it reaches the plastic state, and a rotating grate at the
bottom serves the same function for ash. A water jacket surrounding the
-------
Feed Coal
Coal Lock Hopper
Bed Stirring Drive
Quench
Water
Scrubbing
Cooler
Coal Distributer
or
Bed Stirrer
Raw Product Gas
Rotating
Grate
Grate Drivi
Water and Tar
Water
Jacket
Air and & U ^s
Ash
Lock
Hopper
Ash Outlet
Figure 2-2. Lurgi Gasifier Schematic Diagram (from Refs. 2 and 3)
8
-------
gasifier vessel provides some cooling and approximately 10 percent of the
steam necessary for gasification. The steam/air mixture entering the
gasifier also cools the ash preventing fusion. The crude product gas leaves
the gasifier between 600 and 1200°F (the temperature is a function of coal
type) and contains coal dust, oil, naptha, phenol, ammonia, tar oil, ash and
char, as well as C02, H2, CO, CH4> N2, H2S, COS and small quantities of
higher hydrocarbons. The product passes through a scrubber and cooling tower
to remove the tar before being cooled to about 370°F in a waste heat boiler.
The Lurgi process is complicated because of the large number of moving
parts which also limits its size. Although vessels of 16 feet in diameter
(4)
are planned, present day capabilities are limited to 12 feet diameterv '. The
12 feet units can produce up to 230 million Btu/hr of fuel gas and are mar-
keted in the United States by the American Lurgi Corporation in New York.
Winkler
The Winkler generator is a fluidized bed gasifier originally developed
in the 1920s for the production of synthesis gas from small friable fuels
(up to 8 mm in size) which were unsuitable for gasification in fixed beds.
The gasification takes place in a fluidized bed supported by steam/oxygen
or steam/air - as with fixed bed gasifiers, the feed stock must be noncoking.
The product gases leave the gasifier at between 1500 and 1900 F and are
relatively free from condensible hydrocarbons' . The original gasifier
design operated at a pressure of 1.5 atmospheres absolute and recently the
design has been reevaluated to allow high pressure operation.
The Winkler system is currently marketed by Davy Power Gas, Inc., Lakewood
Florida and sizes to 18 feet diameter providing up to 500 million Btu/hr of
fuel gas have been operated^ '.
Koppers Totzek (K-T)
The K-T process is an entrained flow gasifier as shown in Figure 2-3.
Pulverized coal, oxygen and steam are fed through a "burner" into the gasi-
fier vessel where they react at a slight positive pressure and at approxi-
mately 3300°F^6'. The ash slags and between 50-70 percent of it is removed
from the water quench tank. The product gas may be water quenched before
-------
To Low Pressure
Steam Drum
Pulverized Coal,
Steam and Oxygen
Bo Her Feed Water
Steam
cv *.<. i <.
Product
Gas Outlet
Upptr
-Water
Jacket
Boiler Feed Water
In
_ Burner
Cooling Water
Out
Ash Falls Into
Water Quench Tank
Figure 2-3. Schematic Diagram of Koppers Totzek (K-T)
Entrained Flow Gasifier
10
-------
passing through a waste heat boiler and further scrubbing systems. Tests
have been conducted with air instead of oxygen, but high preheat temperatures
of at least 982°C must be used to maintain slagging conditions. Because of
its high temperature, the K-T system is insensitive to the coal coking pro-
perties and the product gas is free of tars and other condensible hydro-
carbons. However, trace amounts of NH3 and HCN are produced.*
This system is marketed in the United States by the Koppers Co., Inc.,
Pittsburgh, Pennsylvania. The largests units produce up to 234 million Btu/
(4)
hr and larger units up to 560 million Btu/hr are under construction' '.
The three systems described above were operable forty years ago and are
not necessarily representative of future integrated coal gasifier electric
power generation plants. The limitations of fixed bed gasifiers (coking and
small size) will probably preclude their use in large scale COGAS plants.
Similarly, unless the coking problem in fluidized bed gasifiers can be
resolved, they will be limited to use with selected coal types.
Entrained flow and fluidized bed gasifiers have received increased
attention in recent years since they have been recognized as suitable for
large-scale LBG coal gasification^7'8^. The BCR entrained flow gasifier
and the Westinghouse fluidized bed gasifier are two examples of second
generation gasifier systems.
Bituminous Coal Research (BCR) Entrained Flow Gasifier
Figure 2-4 is a schematic diagram of the BCR entrained flow gasifier.
Gasification occurs in two stages. In the top stage crushed dried coal con-
tacts hot gas from the lower stage and partially gasifies leaving small char
and ash particles. The flow velocity is high enough to entrain the particles
so that all materials exit the top of the gasifier as a gas-solid suspension.
The char and slag are separated from the product gas in a cyclone separator and
returned to the lower stage gasifier vessel pneumatically suspended in steam.
The char reacts with the air in the lower stage and the slag drops out the
bottom. Both stages operate at a nominal pressure of 500 psia and the pro-
duct gas exit temperature is 1800°P .
NH3 =0.17 percent and HCN 288 ppm by volume
11
-------
COAL
RAW
FUEL
GAS
GASIFIER
STAGE II
LBG
GAS
CYCLONE
SEPARATOR
TRANSPORT GAS
CHAR
HOPPER
STEAM
AIR
SLAG HOPPERS
SLAG
Figure 2-4. Schematic Diagram of BCR Entrained Flow Gasifier
{from Ref. 9)
12
-------
Westinghouse Fluidized Bed Gasifier
i
The Westinghouse gasifier is a two-stage process combining sulfur
removal with the gasification process as shown in Figure 2-5. The upper
stage is a recirculating fluidized bed where hot coal gas and char react
to form LBG, and dolomite particles absorb sulfur. Coal is introduced at
the bottom, entrained at 20 to 40 ft/sec in a "draft tube" and transported
to the top of the bed. Dolomite particles injected at the top move down-
ward absorbing sulfur from the gases moving counterflow. Since char and
dolomite have different densities they tend to segregate in the fluidized
bed and are drawn off and rejected. Parti cul ate -laden LBG is drawn off
the top of the reactor and cleaned in a cyclone separator before exiting
at 1600°F. The char is recycled to the combustor gasifier where it reacts
in a fluidized bed supported by steam and air producing a hot fuel gas as a
feed to the upper gasifier vessel. The Westinghouse gasifier operates at
10 to 15 atmospheres and produces LBG at
2.2 Properties of LBG
Table 2-1 lists the characteristics of several LBGs produced with
several combinations of gasifiers and cleanup systems. The primary combus-
tible species are CO and H2 comprising 26 to 40 percent with the majority of
the remainder composed of hL, C02 and N«. Approximately half of the LBG is
nitrogen introduced into the process as air and this is responsible (in part)
for the low gas heating value.
Medium Btu gas with about 400 Btu/ft3 Is produced by replacing the air
with oxygen, eliminating the nitrogen.
The ratio CO/H2 is a function of gasifier design and the steam/air ratio.
Typical values range from 0.5 to 2.0.
Table 2-1 also shows that the gasification process converts most of the
sulfur to H2S with the remainder converted to COS. Trace amounts of S02, SOg
and S2 can also be produced but are not reported in the table. Ammonia, NH3,
is also produced in varying quantities and other nitrogen bearing compounds
such as HCN may also be produced but are not reported in the table.
13
-------
DEVOLATIZER
DESULFURIZER
Dolomite Feed
Recycle Gas
Dolomite « '
Draw Off Pot
J
Sulfided
Dolomite
Coal Feed
Hot Clean Fuel Gas
/^ Char
Draw Off
Pot
Recycle
Gas
Hot Gas
L_
Steam
Air
Char
Cyclone
Collector
Devolatizer Fines
COMBUSTOR
GASIFIER
Figure 2-5. Schematic Diagram of Westinghouse Fluidized Bed
Gasifier (from Ref. 10)
14
-------
Table 2-1. Characteristics of LBG
Name
Type of Bed
Cleanup System
Temperature
Temperature, °F
Pressure, atm
Higher Heating
Value, Btu/ft3
CO/H2 Mole Ratio
H2S/COS Mole Ratio
Mole Fractions
H20
H2
CO
C02
CH4
H2S
COS
N2
NHa
HCN
Reference
Lurgi
fixed
None
850
28.6
180
0.68
6.0
10.1
19.6
13.3
13.3
5.5
0.6
0.1
37.5
NR
' NR
3,11
Bureau
of
Mines
fixed
None
1000
34.0
139
1.49
6.0
8.2
13.8
20.6
5.9
2.8
0.6
0.1
47.6
0.15
NR
8
Bureau
of
Mines
fixed
Selexol
low
265
15.8
NR
1.49
NR
0.01
16.0
23.9
5.0
3.1
NR
0.01
52.0
0.03
NR
12
Bureau
of
Mines
fixed
Iron oxide
high
1070
15.8
NR
0.84
NR
3.9
18.2
15.3
11.1
2.8
0.01
NR
47.7
0.03
NR
12
MERC
fixed
None
NR*
10.9
104
1.43
NR
NR
10.7
15.3
12.7
2.1
0.2
NR
59.3
NR
NR
13
Winkler
fluidized
None
2000
6.8
118
1.62
NR
11.5
11.7
19.0
6.2
0.5
0.1
NR
51.1
•NR
NR
11
UGAS
fluidized
None
1900
18.0
150
1.47
NR
12.0
11.6
17.0
8.8
4.1
0.6
NR
45.4
NR
NR
11
BCR
entrained
flow
None
1800
34.0
125
1.39
4.6
10.5
12.0
16.7
8.8
3.1
0.46
0.1
47.7
0.38
NR
9
BCR
entrained
flow
Selexol
low
1000
23.8
125
1.42
NR
0.01
14.9
21.2
6.5
4.2
NR
0.01
53.2
NR
NR
12
BCR
entrained
flow
Conson
high
1070
23.8
125
1.29
NR
9.9
13.5
17.4
9.3
3.6
NR
0.01
45.9
0.43
NR
12
Koppers
Totzek
entrained
flow
None
1000
34
139
1.49
6.0
8.24
13.8
20.6
5.9
2.8
0.6
0.1
47.6
0.17
228 ppm
6
NR = Not reported
-------
2.3 LBG Product Gas Cleanup
Low Btu gas leaving the gasifier is termed "dirty" since it contains
several materials which can cause environmental problems. A cleanup system
is usually installed between the gasifier and combustor to remove these
materials. Three types of cleanup systems may be employed:
• particulate removal;
• sulfur removal; and
• nitrogen species removal.
2.3.1 Particulate Removal Systems
Particulate matter in LBG consists of small char and ash particles and its
removal is desirable because of the potential for turbine blade erosion down-
stream of the combustor and atmospheric particulate emission. Since solid and
liquid particles are several orders of magnitude more dense than the gaseous
constituents, they are usually removed with mechanical equipment. Cyclone
separators, electrostatic precipitators and various filters are suitable for
this application. Since the gas phase constituents are not affected, parti-
culate removal devices will not be discussed further. Reference 14 summarizes
the pertinent aspects of particulate removal systems as applied to LBG.
2.3.2 Sulfur Removal Systems
A variety of sulfur removal systems have been developed or proposed for
LBG. All are primarily designed to remove H2S but may also remove other
gases as well. The temperature of the cleanup process is an important con-
sideration and sulfur removal systems may be divided into two categories: low
temperature processes requiring gas temperatures less than 250°F, and high
temperature processes occurring at or near gasifier outlet temperatures.
Low temperature processes can be subdivided into four groups according
to their principle of operation:
• chemical solvent processes,
• physical solvent processes,
• direct conversion processes, and
t dry bed processes.
Table 2-2 compares the characteristics of each type.
16
-------
Table 2-2. Characteristics of Low Temperature Cleanup Systems (after Colton et al(14))
Basis: 8400 tons/day Illinois No. 6 Coal Fed to BCR Gasifier, or 6700 ppm of Influent H2S
Process
Absorbent
Type of
Absorbent
Temp.
°F
Efficiency of
S Removal
H2S Effluent
Pressure Influent H2S ppm Life
Absorbent Characteristics
Selectivity
Regeneration Towards
CHEMICAL SOLVENT TYPE
1. MEA
2. DEA
3. TEA
4. Alkazid
5. Benefield
6. Catacarb
Monoethanol
amine
Diethanol
amine
Triethanol
amine
Potassium
dimethyl
ami no
acetate
Activated
potassium
carbonate
solution
Activated
potassium
carbonate
solution
Aqueous
solution
Aqueous
solution
Aqueous
solution
Aqueous
solution
Aqueous
solution
Aqueous
80 to
120
100 to
130
100 to
150
70 to
120
150 to
250
150 to
250
Insensitive 99 ~ioo
to variation
in pressure
Insensitive 99 ~100
to variation
in pressure
Insensitive 99 ~100
to variation
in pressure
Insensitive 99 ~100
to variation
in pressure
1-80 atm
99 H2S + COS Unlimited
~100 No degra-
dation
Insensitive 99 H2S + COS
to variation ~100
in pressure
generally
300 psi
Thermal Forms non-
regen, comp.
with COS,
CS2
Thermal Absorbs C02,
does not
absorb COS,
CS2
Thermal H2S
With Steam H2S is high
With Steam H2S - partial
also absorbs
COS, CS2
With Steam
Make up Sulfur
Rate Recovery
50 to As H2S
100% gas
5% As H2S
gas
5% As H2S
gas
As H2S
gas
5% As H2S
gas
As H2S
gas
Status
Commerc i a 1
Commercial
Commercial
Commercial
Commercial
Commercial
-------
Table 2-2. (Continued)
CO
Process Absorbent
Efficiency of
S Removal
Absorbent Characteristics r ,
Type of Temp. H2S Effluent Selectivity
Absorbent Of Pressure Influent H2S ppm Life Regeneration Towards
Make up Sulfur
Rate Recovery Status
PHYSICAL SOLVENT TYPE
7.
8.
9.
Sulfinol Sulfolane
plus Dilso-
propanol
amine
Selexol Polyethylene
glycol ether
Pectisol Methanol
Organic 80 to High pres- 99 H2S + COS
solvent 120 sure pre- r»100
ferred
Organic 20 to 99 H2S + COS
solvent 80 ~100
Organic 0 99 ~100
solvent
Low pressure H2S, and also
heating or absorbs COS,
with steam CS2 and
mercaptans
H2S also
absorbs COS
H2S
As H2S Commercial
gas
As H2S Conmercial
gas
Commercial
DIRECT CONVERSION
10.
Stretford Na2C03 P^us
anthracuin
one sulfonic
acid
Alkaline 99.9 -vlO
solution
H2S
50 to Elemental Commercial
100 sulfur
11. Townsend Triethylene Aqueous 150 to
glycol solution 250
99.9
-10
H2S
Elemental Commercial
sulfur
DRY BED TYPE
12. Iron Hydrated
sponge
Fixed
bed
70 to
100
99 HoS + COS
H2S and also
towards COS,
CS2 and
mercaptans
Elemental Commercial
sulfur
-------
Chemical solvent processes employ aqueous solutions to scrub the dirty
gas forming complexes with H2S, C02 and other components. The solution is
then regenerated at elevated temperatures and recycled. The affinity for
C02 is undesirable since most LBG contains much more C02 than H2S and C0?
removal adds unnecessarily to the cleanup system operational cost.
Physical solvent processes utilize absorption rather than chemical
reaction to remove certain gas phase species. The absorption is proportional
to the partial pressure of the gas to be removed favoring high pressures for
good efficiency. These processes selectively remove H2S over C02, and remove
other sulfur-bearing species as well.
In direct conversion processes sulfur-bearing species are absorbed in
a solvent and reduced directly to elemental sulfur by subsequent reaction.
In dry bed processes LBG is forced through a dry bed where sulfur-bearing
species are absorbed as solids. These processes are best suited to low
concentrations.
In contrast to the low temperature cleanup systems, which can be con-
sidered commercial, high temperature desulfurization processes can only be
termed "under development". The characteristics of some of the high tem-
perature processes are listed in Table 2-3. None of the processes have been
commercialized to date, although the Bureau of Mines Sintered Iron Oxide
Process and Consolidated Coal's Half Calcined Dolomite Process will probably
be the earliest to claim commercial status. To date, no evidence exists to
show that these processes are capable of removing either ammonia or other
nitrogen compounds.
Both low and high temperature cleanup processes discussed above can
easily remove sulfur compounds from the fuel gas to ensure that total sulfur
oxide emissions (from the combustion unit and the sulfur recovery process) are
below the allowable S02 emissions for large coal-fired power plants
(1.2 lb/166 Btu).
19
-------
Table 2-3. Characteristics of High Temperature Cleanup Processes (after Colton, et
Basis: 8400 tons/day Illinois No. 6 Coal Fed to BCR Gasifier,
or 6700 ppm of Influent h^S
Efficiency of S
Removal Absorbent Characteristics c- „,
Process
1. Bureau of
Mines
2. Babcock
& Uilcox
3. Consolid.
Coal
4. Air
INS
o
5. Batten e
Northwest
6. 1ST -
Meissner
Absorbent
Sintered
pellets of
Fe203 (25%)
and fly ash
FezOa
Half cal-
cined
dolomite
Calcined
dolomite
Molten
Carbonates
. (15% CaCOa)
Molten
Metal
(proprie-
tary)
Type of
Bed
Fixed Bed
Fixed Bed
Fluidized
Bed
Fixed Bed
Solution
Splashing
contact
Temp.
OF
1000 to
1500
800 to
1200
1500 to
1800
1600 to
2000
1100 to
1700
900
Pressure
Insensitive
to variation
in pressure
Insensitive
to variation
in pressure
-200 psia
H2S removal
is high at
low pressure
Insensitive
to variation
in pressure
Atmospheric
HgS removal
is high at
low pressure,
5-6 psig
* H£S Effluent
Influent H2S ppm Life Regeneration
-95 -350 >174 With air
cycl es
wt. loss
<5%
~99 -75
-95 -350 10-132
with steam
and CO?
minimum 80-90% with
5-6 steam and
cycl es C02
-95 -350 With steam
and COz
-98 -150 Electro-
lytic
Selectivity Make up Sulfur
Towards Rate Recovery
H2S, COS <5» As S02 gas
As 12-15%
S02 gas
H2S, COS 1% of As H?S gas
circu- to Claus
lation process
rate
H2$, COS As HoS gas
to Claus
process
HgS, COS, As H2S gas
fly ash to Claus
process
H2S, COS
Energy
Regui red
Elec. Other
kW Btu Status
Pilot
Experimental
96,360 Pilot
Abandoned
Pilot
9,830 Conceptual
-------
2.3.3 Bound Nitrogen Species Removal Systems
The removal of nitrogen-bound species such as NH3 and HCN from LEG is a
difficult problem. Some low temperature sulfur removal systems also remove
NH3, but there are at present no satisfactory high temperature methods.
2.4 Product Gas Combustion Characteristics
As the name implies, the volumetric energy content of LBG is small com-
pared to other common fuel gases such as natural gas. Thus, if LBG is to be
substituted for natural gas the pipe supplying the burners must either be
larger or operated at higher pressure. Other minor changes include a smaller
air supply and smaller wet products of combustion per unit of heat released.
These changes are minor and pose no significant problems in fuel con-
version. The major factors determining the suitability of firing LBG in
specific furnace designs are flame stability and furnace heat transfer.
2.4.1 Flame Stability
Ball, Smithson, Engdahl and Putnam(15) have discussed the substitution
of low Btu gas for natural gas in combustion processes using the critical
velocity gradient at flashback as an indication of flame stability*. A flame
is stable if the stream velocity and burning velocity are equal at some point
along the flame front. If the burning velocity exceeds the stream velocity
the flame moves upstream and "flashes back" into the burner. Conversely, if
the stream velocity exceeds the burning velocity at every point the flame
moves downstream and eventually blows off. Both of these critical gradients
(at flashback and blow off) are proportional to the square of the burning
velocity (the mass burning rate per unit area of flame surface).
These considerations apply directly to premixed flames but must be
modified for the diffusion flames commonly used in utility boilers. Here
flame stability results from the recirculation of heat to the base of the
f»
The boundary velocity gradient theory of blow off and flashback was developed
by Lewis and von Elbedo) to correlate stability criterion for premixed
flames burning on circular or rectangular ports.
21
-------
flame providing a continual source of reignition. Since maximum flame
temperatures are lower with LBG than with natural gas, more recirculation
should be necessary to maintain stability. However, burners specifically
designed for lean fuel gases have operated satisfactorily for many years and
that suggests that flame stability is not a critical problem for LBG.
2.4.2 Furnace Heat Transfer
When LBG is substituted for natural gas in a utility boiler it is
important that the heat removed from the furnace be the same for both fuels.
Since a substantial portion of the heat transfer in a furnace is due to radia-
tion, which is proportional to the fourth power of the radiant gas temperature
the lower flame temperature of LBG could lower furnace heat transfer
substantially.
As a first order approximation, Hottel and SarofirrP ' assumed the fur-
nace to be a well-stirred tank and calculated the radiative heat transfer as
a function of combustion characteristics. Their model further assumed that:
t the properties of the gases in the furnace are uniform;
• the gas is gray;
• the heat sink has a constant temperature;
• heat losses through the wall are negligible;
• the gases leave the radiant section at a temperature AT
below the effective radiating temperature of the gases in
the furnace T ; and
t convective heat transfer is negligible compared to radiative
loss (although the assumption is later relaxed).
Under these conditions the net heat flux from the gases to the walls
n is given by:
vg net 3
Vet ' R ' (T94 - T14> (1)
22
-------
AT
(2)
1 R r- + T~- - l
eg CS el •
where
Tj is the sink temperature
e is the gas emissivity
E! is the emissivity of the sink surface
Cs is the fraction of the total surface area AT covered by the sink
If H is the total heat input, m is the product mass flow rate, and c
is the average specific heat,
H = A cp (TAF - To)
where TA(- is a pseudo adiabatic flame temperature which ignores dissociation.
Similarly:
H - Qg net = * CP
Elimination of Tg between equations (1) and (5) gives
Q'D1 + T* = (1 + A' - Q')* (6)
where
Q' = g net AF " o (the dimensionless furnace
H TAF efficiency)
HF
}—ji—Tf— T \ (the dimensionless firing density)
n Mr Mr 0
23
-------
T = ^i/TAF ^e rat10 °f s^n'c temperature to
pseudoadiabatic flame temperature)
A1 = T/Tnr (the ratio of gas temperature drop
to pseudoadiabatic flame
temperature)
Equation 6 specified the furnace efficiency as a function of firing density,
heat sink temperature and the temperature drop. If the furnace is well-stirred,
AT -*• 0 and equation (6) simplifies to:
Q'D1 + T* = (1 - Q1)* (7)
The results of this analysis may be used to illustrate the difference
between firing a furnace with LBG and natural gas. As a simple example con-
sider a utility boiler with a radiant chamber 30' x 60' x 80' high with a
total heat input of 300 x 10 Btu/hr fired with natural gas with 15 percent
excess air at 600°F. T.p is approximately 4140°R and T, will be approximately
1160°R if T is 3000°F, e = 0.38 and e, the emissivity of the walls is 0.8.
*7 J
Thus:
(GS
l'R 2.63 + 1.25 - 1
= 6,250 ft2
_
4140
= 0.28
D, 300 x 106
6,250 (0.1712 x 10"8) (4140)3 3620
Solving equation (7) iteratively gives:
Q1 = 0.5045
and thus the total heat transfer is:
6
Qnet = 0.5045 x - x 300 x 10° Btu/hr
= 174 x 106 Btu/hr
24
-------
If the same furnace were to be fired with LB6 produced from a Lurgi gasifier,
600 F preheated air and 10 percent excess air, the pseudoadiabatic flame
temperature would be 3,560°R. Therefore,
D1 = 0.204
Q' = 0.4427
Qnet = 153 x 106 Btu/hr
These approximate calculations give some indication of the radiatability of
LBG compared to natural gas.
As expected, the furnace heat transfer with LBG is lower. However, the
difference is not large, particularly considering the approximations involved.
If the furnace was to be designed specifically for LBG, the radiative section
should be slightly enlarged over natural gas designs.
2.5 Low Btu Gas Combustion - Pollutant Emissions
Carbon monoxide and nitrogen oxides are the atmospheric pollutants most
likely to be formed during the combustion of coal-derived LBG. Nitrogen
oxide formation is a natural consequence of efficient combustion, whereas the
emission of significant quantities of carbon monoxide is an indication of poor
combustion conditions. Carbon monoxide emissions can be prevented by providing
complete fuel/air mixing and avoiding rapid quenching of the combustion
products.
Nitrogen oxides can be produced from two sources:
/[
• the oxidation of molecular nitrogen (thermal NO); and
• the oxidation of nitrogen-bearing compounds (such as ammonia in
LBG)
The reactions forming NO from molecular nitrogen are of two types: the
reactions originally proposed by Zeldovich, and those in which the nitrogen
molecules are attacked by hydrocarbon fuel fragments producing nitrogen inter-
mediaries which subsequently form NO. The production of NO from nitrogen-
bearing compounds is controlled mainly by the availability of oxygen, although
temperature also has some effects. The adiabatic fl'ame temperatures of low
25
-------
Btu fuel gases are relatively low, and thus their potential to form thermal
NO is also low. However, fuel NO formation is almost independent of flame
temperature and relatively high conversions of NH3 to NO are expected,
particularly with vigorous fuel/air mixing.
(18)
Rawdon et alv ' have recently burned both medium and low Btu fuel gas
in an experimental tunnel furnace using a typical boiler burner and measured
NO emissions. Thermal NO levels between 80 and 150 ppm were measured and
thermal NO levels between 50 and 60 ppm were reported for a turbo furnace
burning 90 Btu/ft3 blast furnace gas. Conversions of NH3 to NO from 60 to
20 percent were reported for NH3 concentrations of 0.04 and 0.2 percent.
Thus the presence of ammonia, in coal-derived LBG may cause significant NO
emissions.
26
-------
3.0 ADVANCED POWER GENERATING SYSTEMS ANALYSIS
Power plants fired with coal-derived LEG will only be competitive with
conventional direct-fired power plants if the capital costs and energy losses
in the gasification process are countered by plant improvements such that the
overall efficiency and cost of electricity are comparable. Furthermore, since
these improvements may drastically alter the combustion parameters, it is
important to identify those advanced power generating systems with the best
performance and those most likely to be constructed based on economic
considerations.
This section discusses a comparative analysis of several advanced power
generating systems based on separate sets of assumptions:
• Separate gasifier and power plant - fuel transfer from gasifier
to power plant is the only coupling. All components are representa-
tive of existing technology with no advances in the state-of-the-art.
• Integrated gasifier/power plant combined cycle system with heat,
and mass and fuel transferred between the components, and each
component optimized for optimum overall system performance.
An analysis based on the first set of assumptions was conducted by Com-
bustion Engineering under subcontract to Ultrasystems. This approach is
reasonable as a first approximation and applies directly to situations where
the gasification complex and power plant must be separated by some distance
or cannot be directly integrated into the power plant for other reasons. The
next section discusses the results of this study.
Integrating the gasifier into the power plant and optimizing each com-
ponent can improve overall systems performance substantially, but would
require some development work extending the state-of-the-art in gasifier and
power plant design and a careful analysis of the entire power plant to identify
optimums. Consequently, the results of the second analysis apply to more
advanced future power plants to be constructed some time in the mid to late
1980s. The results are presented in Section 3.2.
27
-------
3.1 Separate Gasifier/Power Plant Comparative Analysis
This section presents the results of a comparative analysis of LBG-fired
power plants conducted by Combustion Engineering under subcontract to Ultra-
systems. Two basic restrictions were imposed upon the power generating
systems:
• An integrated gasifier-power plant system was not to be con-
sidered. The fuel was to be purchased "over the fence".
• All components in the system must represent existing technology
and require no advances in the state-of-the-art.
Systems meeting these requirements were then assumed to be fired with two
alternative fuels representative of existing coal gasification technology:
t A fuel gas typical of that produced by a large air-blown Lurgi
gasifier; and
• A medium heating value fuel typical of that produced by an oxygen-
blown Koppers Totzek (K-T) system.
The compositions assumed for these fuels are given in Table 3-1.
Table 3-1. Fuel Gas Compositions
Component
CO
H2
CH4
C02
H20
N2
LHV (Btu/scf)
60°F and 1 atmosphere
Low Btu Lurgi
14.1 vol percent
20.9
5.8
12.5
6.6
40.1
172
Medium
53 vol
36.4
9.25
0.3
1.05
272
Btu (K-T)
percent
Both fuels were assumed to be delivered to the power plant at up to 1500°F and
10 atmospheres.
28
-------
The study began with a preliminary screening of 35 conceptual power
plants and six were chosen for further detailed analysis. These plants were
then analyzed thermodynamically to determine heat and material balances and
to calculate overall plant performance at design point. The capital and opera-
ting costs were also estimated to identify the most cost-effective designs.
Finally, the combustion conditions, heat transfer surface areas and special
materials for each type of combustion were determined.
3.1.1 Preliminary Screening
Thirty-five conceptual power plant systems were considered in the pre-
liminary screening. These concepts can be conveniently classified into the
following groups:
• Conventional steam power plant with or without fuel gas preceding
or high temperature burners.
t Combined cycle systems with waste heat or supplementary-fired
steam boilers
• Supercharged boiler combined cycle systems
• Combustors fired by oxygen for those systems with medium Btu
gas (since oxygen is required for the gasificiation process)
A complete description of each system is given in Appendix A and the major
findings are summarized in Table 3-2. The twelve most promising systems
were provisionally selected for more detailed consideration and are listed
below:
Al Conventional plant baseline case
Bl High temperature fuel gas and precooler
G3 Supplementary-fired combined cycle
G5 Supplementary-fired combined cycle with fuel gas cooler
Gl Supplementary-fired combined cycle with fuel gas cooler
G2 Supplementary-fired combined cycle with fuel gas cooler
HI Supercharged boiler combined cycle
E5 Exhaust-fired combined cycle
Cl Conventional power plant with high temperature burners
H2 Supercharged boiler with process steam fuel gas precooler
II Oxygen-fired supercharged boiler
14 Oxygen-fired supercharged boiler fuel gas precooler
29
-------
Table 3-2. Summary of Preliminary System Concept Analysis
CJ
o
Class Description
A Conventional Utility Steam
Power Plan
B Conventional Power Plant
with Gas Precooler
C Conventional Power Plan
with High Temperature
Burners and Partial
Precooling
D Industrial Boiler
Systems
No.
1
2
1
2
3
1
2
3
1
2
3
4
Fuel
Press.
Low
High
Low
Low
High
Low
Low
Low
Low
Low
High
High
Gas
Temp.
Low
Low
High
High
High
High
High
High
Low
High
High
Low
Remarks
Baseline system, all components of current design,
low NOX potential, retrofit possible, but high
capital cost.
High pressure fuel more suited to combined cycle,
otherwise fuel throttling loss.
Components of current design, produces process
steam in quantities dependent upon load.
Uses hot fuel as air heater, explosive potential
of gas precooler, retrofit unlikely.
Fuel throttling loss.
Burner development could be required, possible NOX
problems
Partial precooler producing process steam burner
development not warranted.
Partial gas precool to heat air.
All have high plant heat rates, complex system
design, marginal delivery time and cost advantage
may require new steam turbine design.
-------
Table 3-2 (Continued)
Class Description
E Waste Heat Boiler
Combined Cycle Systems
F Exhaust-Fired Boiler
Combined Cycle Plant
G Supplementary-Fired
Boiler Combined Cycle
Plant
H Supercharged Boiler
I Supercharged Boiler
Combined Cycle Oxygen
No.
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
•3
o
4
1
4
3
2
Fuel
Press.
Low
Low
High
High
High
Low
Low
High
High
High
Low
Low
High
High
High
High
High
I nui
L.UVY
Low
High
High
Low
Low
Gas
Temp.
Remarks
Low
High
Low
Higher plant heat rates than corresponding G systems,
good load demand flexibility
High High NOX potential
High Low NOX because of process steam heater
Low No advantage to offset higher plant heat rates than
High G systems
Low
High
High
Low
High
Low
High
High
Low
High
I nw
L.UW
High
Low plant heat rate, components are of current design,
low NOX potential
High NOX potential because of hot fuel gas
Compact design,;low heat rate plant, lower capital cost,
existing technology but not widely used in the U.S.,
NOX unknown.
System design not warranted for low pressure fuel.
Low As for systems H, but cost of oxygen may affect any
High advantage.
Low
Low
-------
Robson, et al^7'8'9^ has made extensive studies on the utilization of
coal-derived fuel gas in waste heat and supplementary-fired boiler combined
cycle systems. Consequently, to prevent repetition it was decided to
eliminate systems G and E from further consideration, and to concentrate
upon four variations from conventional utility power plant technology:
• The use of a gas/water heat exchanger to cool the fuel gas and
produce process steam.
• The use of high temperature burners in a conventional boiler.
• The use of oxygen (instead of air) with a medium heating value
gas.
• A supercharged boiler combined cycle steam plan.
The six systems listed in Table 3-3 were selected for detailed analysis.
The first two are conventional power plants fired with the two fuel gases
and the other four represent the four variations listed above.
3.1.2 System Design
Each of the six systems selected for detailed analysis was developed and
optimized for a 500 MW plant with an acceptably low overall plant heat rate
in keeping with usual engineering practice and the limitations of state-of-
the-art components.
This required an increase in plant complexity and capital cost because
of certain special characteristics of LEG such as the low air/fuel ratios
which necessitated stack gas coolers. A simplified schematic diagram and
heat and material flows for each system are shown in Figure 3-1 through 3-6,
and Table 3-4 summarizes the major components.
All plants except system Hl-L (supercharged boiler combined cycle)
utilize a single, conventional balanced draft, controlled circulation boiler
producing a maximum of 3,310,000 Ibs/hr main steam flow with superheat and
reheat steam conditions of 1011°F/2685 psig and 1101°F/501 psig respectively.
Superheat steam temperature control is via burner tilt or gas recirculation.
The furnace is single cell with tangential or alternate opposed wall burner
designs. Extremely high gas recirculation (205 percent) was included for
32
-------
Table 3-3. Summary of Systems Studied During Critical Analysis
GO
OJ
Fuel Properties (as Purchased)
Description
Conventional Plan
Conventional Plan
Conventional Plan
with Precooler
Conventional Plant
with High Tem-
perature Burners
Supercharged
Boiler Combined
Cycl e
Conventional
Boiler with Q£
System Type
Al-L Low Btu
(Lurgi)
Al-K Medium
Btu
(Koppers)
Bl-L Low Btu
(Lurgi)
Cl-L Low Btu
(Lurgi )
Hl-L Low Btu
(Lurgi)
I5-K Medium
Btu
(Koppers)
Pressure
Psig
10
10
10
10
132
10
Temp.
OF
250
250
1500
1500
654
250
Remarks
Baseline
Baseline
Fuel enters the plant at 1500°F, is cooled
to 750°F in a waste heat boiler and fired
through conventional burners.
Fuel enters the plant at 1500°F and is
fired through high temperature burners.
Furnace pressure 10 atmsopheres. Turbine
inlet temperature 2000°F producing 33 per-
cent of the power output.
Air for combustion replaced by oxygen.
-------
SYSTEM A1~L
CO
-p.
I 1
I Q3
Q4 L
............jy.-
' f
Q17
STEAM
WATER
FUEL
AIR
GAS
I
)
SH
^^^^
RH
ECON
Q16
ICRl
1
|
1
1
fr7
1
1
|Q8
1
HIGH
PRESSURE
HTR
STEAM
TURBINE
GEN
500MW
LOW
PRESSURE
HTR
QZ A
FAN
FD FAN
I *
1-1/2" Hg
CONDENSER
ID.FAN
STACK
Figure 3-1. Conventional Steam Plant, System Al, Lurgi LBG, Low Temperature
-------
SYSTEM Al-K
CO
f-
V
>
Q15
013
FD FAN x—
Q3
1
•
L».
Q2
Q
J
mmm m
\
1
1
• MM
•
i
SH
^
GR
FAN
I
1
•
I
AIR
OTER
|
1
•
1
wmmm
•
•
1
;«
Q:
04
RH
ECON
14
1
no
vp
LO
• •
/
WATER
FUEL
1 AIR
L GAS
• • • • *^ ~mmmm*mmmmmm*m^ »—>•»
' Qs i
1 «.2. y......
|Q7 ..Q6 ? •
? : w
t-^h i — i
^. .... [STEAM GEN
j SJJ^INt 500 MW
^A A ' ^^
f
~~t m^^
| : HIGH ; LOU
|Q8 y PRESSURE HTR- PRESSURE HTRl
I] OH —..I.. X*"^*N^-^- __
1 *<*•*• I yl£ /^ ^^^^ r—
•-- > --0T- > -e-(>r r
• * ,n , ^^>^-, i
III A /
1 j L_ 1 I
/
ID FAN f
STACK
^S
Figure 3-2. Conventional Steam Plant, System Al, Koppers Totzek MBG, Low Temperature
-------
oo
r~
i
•
?i.7....Jy
i ^V
A A A Q16
vvv""*" "~
013 014
»••• ^MM ••• ^B *• «H^^*
WASTE HEAT BOILER
Q18
FD FAN f—
no
...Si..
J
It
1 I—..
* <
-, 11
s
» M
>"<
i
i
12
• •
YSTEM
SH
V
pq
GB
FAN
^
j
AIR
HEATEf
i
(
j
B:
•^•^
» *
»
•
i
Q
\
L-L
» ««B^
^ *
Rl
EC
15
Q]
• MM
.Q4.
^
DN
nq
10
» VI
•
-1
1
1
-I-
i
Im
"Q7.
I
w
j
L
p
* ^
IQS
1
i
i
HP
HTR.
rp-^
WATER
FUEL
AIR
— — — r.A^
\ -Q5
s.Q6, i
? \ Jf
^"^"^CTCflM /vr»i
1 ^ ' tAM GtN
L^UmBlNE 50MW
• • «
« • * •
• • •
Qll --I--- Q12 X ^-^* • — i
"tT-T- <> 7pT-l ^JCOHDEHSER- -
1 " __J HTD V^-,-^1 — 1 1
i 1 i IITR- 1
. • • • II
u_^ u ,
STACK 1 1
V
Figure 3-3. Conventional Steam Plant with Gas Precooler, System Bl, Lurgi LBG
-------
SYSTEM Cl-L
CO
Q4 I ;
• • ~( \
I
V Q5
f» «• * • v * • ••**•«•• ••••
3-1/2" Hg
CONDENSER
FD FAN
STACK
Figure 3-4. Conventional Steam Plant with High Temperature Burners,
System Cl, Lurgi MBG, High Temperature
-------
f -
Q21
SYSTEM Hl-L
Q3 04
1-1M r<- •
:
7 tf
CO
oo
_£-
IQ20
SH
RH
AIR ? Q2
COMPRESS |
Q5
Q6
STEAM
TURBINE
Q13
^
Q15
r^
Q7
GAS TURBINE
QU
I HP
1 HTR
ST;
I
ICK|
Q18
L-Z-.
^^J
^
Ql
Q19
— "^ GEN
1 ^ 217MW
Qft
Q1E
^ L
—j — 1 i
t > kQ9 r 5>
1 r j
] r JQle
i
I
i
I
|
ijn
|Q2
Q10 ^-
1
u
5
i
:>
T
Qll
k
?
1
1
IQ24
1
i
1
W
STEAM
WATER
FUEL
AIR
GAS
GEN
308MW
-1/2" Hg
CONDENSER
Figure 3-5. Supercharged Boiler Combined Cycle, System HI, Lurgi LBG
-------
SYSTEM I5-K
CO
I Q4. Q5 I
• •--••«•-, ............ ^ ......
Vs i . r
U -- - 1
^^^ ^^^^^.M I ^••
... ......
Q13
SH
I i.._™
STEAM
WATER
FUEL
3-1/2" Hg
CONDENSER
STACK
Figure 3-6. O Blown Steam Plant, System 15, Kopper Totzek MBG
-------
Table 3-4. Major Components
System
Boilers
Steam Turbines
Gas Turbines
Fans:
FD
ID
GR
-fa.
o Pumps :
Boiler Feed
Circ.
Al-L
1
IMP, IIP, 2LP
0
2
2
1
2
3
Bl-L
1
IMP, IIP, 2LP
0
2
2
1
2
3
Cl-L
1
IMP, IIP, 2LP
0
2
2
1
2
3
Hl-L
3
1HP, IIP, 2LP
3
0
0
0
2
0
Al-K
1
1HP, IIP, 2LP
0
2
2
1
2
3
I5-K
1
1HP, IIP, 2LP
0
0
2
2
2
3
Special Heaters
Stack
2 RECUP Fuel/Gas
2 REGEN Air/Gas
2 RECUP Fuel/Steam
2 REGEN Air/Gas
2 REGEN Air/Gas
3 RECUP Steam/Gas
3 RECUP Water/Gas
2 RECUP Water/Gas
2
2 REGEN Air/GAS
2 RECUP FUel/Gas
2 RECUP 02/Gas
1
-------
the 02 Blown Steam Plant (System I5-K), in order to maintain peak gas (and
hence, metal) temperatures within acceptable limits. The major differences
between these boilers and conventional LBG boilers are the fuel piping, the
burners, stack gas coolers and the introduction of flue gas recirculation
through the windbox instead of through the hopper bottom.
The Supercharged Boiler Combined Cycle Plant (System Hl-L) includes
three 10 atmosphere once-through boilers. A side elevation through one
possible design is presented in Figure 3-7. The LBG fuel would be fired
through a single fixed vane (SV) burner in the bottom of each furnace. Each
of the boilers has a maximum continuous rating of 612,000 Ibs/hr main steam
flow with superheat and reheat steam conditions the same as the other plants.
Superheat and reheat steam temperature control is achieved via desuperheat
spray. Flue gas leaves the boilers at 2000°F and is delivered to three gas
turbines; heat losses from the system are minimized by routing the air from
the compressors to the boilers in an annulus surrounding the flue gas pipe.
Upon leaving the gas turbines the flue gas passes through a waste heat
recovery chain consisting of an evaporator-superheater, economizer and a
feedwater heater. The flue gas at 300°F then passes to the stack. The three
gas turbine generator sets deliver a total net output of 191 MW; a single
steam turbine generator delivers 308 MW net.
3.1.3 Plant Optimization and Performance
The optimization of each of the six systems was relatively straight-
forward with the possible exception of the combined cycle system number Hl-L.
The desire for high plant efficiency, low stack temperature and the limita-
tions of existing materials and components dictated the plant arrangements.
The optimization of the combined cycle system number Hl-L was somewhat
more complex. The relationships between gas turbine inlet temperature,
pressure ratio (or equivalently furnace pressure) and plant net heat rate
holding stack temperature and steam turbine output constant are shown in
Figure 3-8*. The plant heat rate is seen to improve with increasing turbine
Heat rate is the number of Btus required to provide 1.0 kWh of electriritv
and is inversely proportional to overall efficiency.
41
-------
AIR
to
<
O
AIR
AIR FUEL
IGNITER
JWT
SKETCH OF
PROPOSED BURNER DESIGN
n
GAS TURBINE
V. X
Figure 3-7. Side Elevation Showing Detail of Supercharged Boiler and Burner
42
-------
10,000 r
9,000
4->
CQ
at
a:
-------
inlet temperature and furnace pressure (within the range indicated), a direct
result of the improvement in gas turbine performance. A turbine inlet tem-
perature of 2000°F and furnace pressure of 10 atmospheres were selected for
System Hl-L since they present no insurmountable problems for the boiler
design and represent conservative estimates of industrial gas turbine capa-
bilities in the near future. For example, G.E. is currently offering a 70 MW
oil-fired industrial gas turbine with a compressor pressure ratio of 9.7 and
a 1950°F turbine inlet temperature; United Aircraft has built several slightly
larger machines which operate at these levels. The 2000°F turbine inlet tem-
perature selection may be conservative since the characteristic temperature
"spikes" found for conventional machines are not present for this combustor/
boiler arrangement. Thus the turbines should be able to withstand higher
inlet temperatures.
Table 3-5 summarizes the performance of each system. The overall plant
efficiencies and heat rates for all plants are nearly the same with the excep-
tion of the combined cycle plant number Hl-L which is substantially better.
Since this analysis assumed no integration of gasifiers and power plant, the
efficiencies, heat rates, etc., are based on the heat content of the fuel gas
rather than the parent coal. The overall gasifier and plant performance from
coal pile to busbar must also consider gasifier losses. Systems Bl-L and
Cl-L which utilize the sensible heat of the fuel gas should have 10 to 15 per-
cent improved gasifier efficiency over the other systems.
3.1.4 Capital and Operating Cost Estimates
Capital cost estimates were prepared for each of the systems by assuming
an average U.S. site. Plant capital costs including field erection, are sum-
marized in 1974 dollars in Table 3-6 using the standard Federal Power Commis-
sion accounting system. Twenty-four percent has been added for engineering
and construction management (13 percent) and for contingency (11 percent).
Interest during construction is calculated at an annual rate of 8 percent
using an S-curve expenditure schedule for four years of construction, giving
a total of 17 percent. The total estimated cost is also shown in dollars per
net kilowatt. These cost estimates are within plus or minus 10 percent.
44
-------
Table 3-5. Plant Performance
en
System
Net Power Plant Eff. %
Net Power Plant Heat
Rate Btu, 1 kW/hr
HHV Heat Energy Input
106 Btu/hr
New Power Plant Output
MW
Steam Turbine Output
Al-L
36.2
9440
4720
500
500
Bl-L
36.3
9420
4710
500
500
Cl-L
36.3
9420
4710
500
500
Hl-L
43.7
7814
3899
499
308
Al-K
37.8
9040
4520
500
500
I5-K
38.3
8920
4460
500
500
MW
Steam Turbine
Throttle Conditions
psia/°F/°F
Steam Turbine Back
Press., Inch Hg.
Gas Turbine Power
Output, KW
Gas Turbine Inlet
Temp., °F
Gas Turbine Back
Press., psia
Power Plant Aux.
Power, MW
Stack Temp., °F
Steam Turbine
Output, %
2535/1000/1000
3.5
2535/1000/1000 2535/1000/1000 2535/1000/1000 2535/1000/1000 2535/1000/1000
3.5
3.5
3.5
191
3.5
3.5
---
_—
17.4
250
100
—
—
17.4
300
100
2000
15.2
17.4 5.2 15.4 16.4
300 300 250 350
100 61.7 100 100
-------
Table 3-6. Plant Capital Costs
Plant Number
Land and Land Rights
Structure & Improvement
Boiler Plant Equip.
Turbine Gen. Equip.
Accessory Elec. Equip.
Misc. Power Plant Equip.
Station Equip.
Subtotal
Engineering, Design
Const. Management, Temp.
Const. Facilities &
Contingency
Subtotal (1974)
Escalation (1974)
Interest During Const.
Total Estimated Cost
Estimated Cost /Net KW
Al-L
572
16,900
38,900
26,500
1,100
1,450
1,300
86,812
20,835
107 ,647
-0-
18,300
125,947
252
Bl-L
571
16,900
38,970
26,400
1,100
1,450
1,300
86,691
20,806
107,497
-0-
18,274
125,771
252
Cl-L
571
16,900
36,440
26,400
1,100
1,450
1,300
84,161
20,199
104,359
-0-
17,741
122,101
244
Hl-L
557
15,900
27 ,400
32,300
1,150
1,830
1,360
80,497
19,319
99,817
-0-
16,969
116,786
234
Al-K
570
16,900
36,020
26,300
1,100
1,450
1,300
83,640
20,074
103,713
-0-
17,631
121,344
243
I5-K
571
16,900
36,020
26,400
1,100
1,450
1,300
83,741
20,098
103,839
-0-
17,653
121,491
243
-------
Capital costs for equipment in Combustion Engineering's scope of supply
were estimated by C.E. This includes boilers, fans, economizers, air heaters,
steam and feedwater piping, stacks, etc. These items are all included in
boiler plant equipment. Land is included at typical costs for average U.S.
sites. The balance of the equipment for the plants was cost estimated by
scaling previous consulting engineering firm estimates for similar plants.
The combined cycle plant number Hl-L has steel stacks supported on the steel
work of the waste heat recovery sections and the remainder of the plants each
have a single concrete stack. Stack dimensions are summarized below:
_ ,. Stack Stack
system Height (ft) Diameter (ft)
Al-L 260 23.8
Bl-L 210 22.1
Cl-L 210 22.1
Al-K 260 20.8
I5-K 180 15.0
Structures and improvements include site improvements, buildings, and
foundations. Boiler plant equipment costs are significantly less for the
combined cycle System Hl-L due to reduced steam.
Turbine generator equipment includes steam turbines and generators,
condensers and auxiliaries, circulating water system, cooling towers, make up
water system, steam turbine auxiliaries, gas turbines and generators and gas
turbine auxiliaries. The gas turbines in the combined cycle plant, Hl-L
result in higher costs in this area, but these are more than offset by the
lower boiler plant equipment costs.
Accessory Electrical Equipment includes auxiliary power transformers,
switchgear and buses, electrical control board, relay boards, motor control
boards, and miscellaneous substations. These cost totals are essentially
the same for all plants.
Miscellaneous power plant equipment includes laboratory and sampling
equipment, tools, lockers, emergency equipment, portable cranes and hoists,
47
-------
and communication equipment. This equipment costs more for the combined
cycle plant Hl-L due to the addition of gas turbines.
Station equipment includes the main power transformers, buses, and
insulators. The cost for the combined cycle plant Hl-L is slightly larger
than the other plants due to the increased number of generators.
Performance and operating costs for 1974 operation are summarized in
Table 3-7 and the details of the calculations are listed in Table 3-8. The
gross electrical output is approximately 520 MW for each plant after cor-
recting for typical auxiliary power requirements including condenser cooling
pumps, boiler circulating pumps, FD, ID and GR fans, and transformer losses
(0.5 percent of the net output). The boiler feed pumps are steam turbine
driven and are included in the steam cycle. The net plant heat rate is the
gross (higher heating value plus sensible) fuel heat input divided by the
net output. Annual net generations are the net full load output times 5000
and 7500 hours per year. These are equivalent to load factors of 57 and
86 percent. Capital costs are from Table 3-8. A fixed charge rate of 20 per-
cent of capital cost per year is assumed as typical for an average U.S. site.
Annual fuel costs for 1974 are based on $1.25/10 Btu as the delivered
price for 10 psig Lurgi LBG and $1.40/106 Btu for 147 psig Lurgi LBG. The
delivered price for Koppers Totzek MBG is $1.50/10 Btu. Plant 15 has an
additional cost of $0.55/10 Btu for oxygen for the boiler, resulting in a
total of $2.05/10 Btu. These costs are based upon estimates of coal and
gasification costs. For example, Koppers Totzek estimates the cost of
300 Btu/ft3 fuel gas at $1.00/106 Btu based on coal available at $0.80/106
Btu. However, recent coal costs are as high as $1.60/10 Btu. The best
price for a long-term, large volume contract for 3 percent sulfur bituminous
coal is currently $12/ton at the mine mouth and unit train transportation
adds $6/ton for a typical 350 mile distance. The total $18/ton is equivalent
to $0.80/10^ Btu for a coal with a heating value of 11,300 Btu/lb. The cost
of Koppers Totzek 300 Btu gas using this coal would be $1.50/10 Btu.
Economic evaluations included in Combustion Engineer's coal gasifica-
tion program indicate LBG adds $0.45/10 Btu to coal cost while MBG adds
48
-------
Table 3-7. Summary of Operating Cost Estimates
PLANT NUMBER
Nominal Plant Output MW
Net Plant Heat Rate
Number of Gas Turbines
Steam Throttle PSIG
SH DEGF
Peneat DEGF
Waste Heat FW Heater
Gross Plant Output MW
Auxiliary Power MW
Transformer Losses MW
Net Output MW
Annual Net Generation
Million KUHR
At 5000 Hr/Yr
At 7500 Hr/Yr
Capital Costs - $1000
Fuel Cost - S/M11 BTU
Annual Fixed Charges at
ZOVYr - $1000
Annual Operating and Ha Int.
Labor - $1000
Annual Maintenance
Mat, i Supplies - $1000
At 5000 Hr/Yr
At 7500 Hr/Yr
Annual Fuel Cost - SI 000
At 5000 Hr/Yr
At 7500 Hr/Yr
Total Annual Operating
Costs - $1000
At 5000 Hr/Yr
At 7500 Hr/Yr
Energy Cost MIlls/KWHR
At 5000. Hr/Yr
At 7500 Hr/Yr
Al-L
500
9,440
-0-
2.520
1,000
1,000
-0-
520
17.4
2.6
500
2.500
3,750
125,947
1.2S
25,189
1.320
1.040
1.560
29,500
44.250
57.049
72.319
22.820
19.285
Bl-L
500
9.420
-0-
2.520
1,000
1,000
1
519
16.4
2.6
500
2,500
3.750
125.771
1.25
25.154
1.320
1.038
1.557
59,437
44.156
56,950
72,187
22,780
19.250
Cl-L
500
9.420
-0-
2,520
1,000
1.000
-0-
519
16.4
2.6
500
2,500
3.750 '
122.101
1.25
24,420
1.320
1.038
1,557
29.437
44.156
56,216
71,453
22,486
19,054
Hl-L
499
7.814
3
2.520
1,000
1,000
3
506.8
5.2
2.6
499
2,495
3,742
116.786
1.40
23,357
1,320
1.394
2,091
27,294
40,941
53,365
67,709
21 ,389
18,092
• Hf^^^^t—m
Al-K
500
9.040
-0-
2,520
1.000
1.000
-0-
518
15.4
2.6
500
2.500
3,750
121.344
1.50
24.269
1.320
1,036
1,554
33.900
50.850
60.525
77,993
24,210
20,748
^— •— »^— n*™^*,
15-K
500
8.920
-0-
2.520
1.000
1.000
-0-
519
16.4
2.6
500
2.500
3,750
121.491
2.05
24.298
1,320
1.038
1,557
45.719
68,572
72.371
95.748
28,948
25,533
49
-------
TABLE 3-8
ENERGY COST COMPUTATIONS
1. Net Plant Heat Rate
Total Fuel Fired
Gross MW - Auxiliary Power
Total Fuel Fired
Net Plant MU ~
2. Annual Net Generation = Net Plant MW x Hours
Maintenance Material
and Supply Cost
= Gross KWHRS/yr x Mills/KWHR Cost
4. Annual Fuel Cost
Net Plant v Fuel ($/10b BTU)
Annual Net .,__ .
Generation Heat Rate Cost
5. Total Annual
Operating Cost
Annual Annual Annual Annual
Capital + Fuel + Labor + Maintenance
Cost Cost Cost Materials
6. Unit Energy Cost
Total Annual Operating Costs
Annual Net Generation
50
-------
$0.70/10 Btu (the same as the above estimates for MBG). Therefore, the
total price used for Lurgi LBG is $1.25/106 Btu; for Koppers Totzek MBG,
$1.50/10 Btu. This prices are for fuel delivered at low pressure (10 psig)
and have been used for all systems except HI. Plant 15 needs 138.5 Ibs
oxygen per 10 Btu of fuel for combustion. At the 1974 U.S. average oxygen
price of $8/ton, this adds $0.55/106 Btu to the MBG for a total price of
$2.05/106 Btu.
It should be noted that the above prices are for fuel delivered at low
pressure (10 psig). In the Lurgi gasification process, the fuel will be
expanded from 132 psig through a turbine to 10 psig and this will yield
more than enough power to drive the gasifier air compressors. However, if
the gas is delivered at gasifier pressure, additional energy will be required
to compress the air for the gasifier and the cost of the fuel will be corres-
pondingly higher, $1.40/106 Btu for Lurgi LBG at 654°F and 132 psig compared
to $1.25/10 Btu for the low pressure Lurgi systems. The annual fuel costs
are the product of these rates, the net heat rates, and the annual hours of
power generations (Item 4 of Table 3-8).
Annual operating and maintenance labor costs represent plant staffs of
78 people for each plant and are based on 1974 staff salary levels plus
35 percent administrative cost. Maintenance materials and supplies costs are
based on 0.40 mills/kWh (gross) for the steam plans and 0.55 mills/kWh (gross)
for the combined cycle plant. These represent composite total plant systems.
Maintenance rates for the combined cycle plants are prorated according to the
percentages of steam turbine and gas turbine outputs. The annual maintenance
materials and supplies costs are the products of the composite plant rates,
the gross plant outputs, and the hours per year (Item 3 of Table 3-8).
The total annual operating costs are the sums of the fixed charges and
the fuel, labor and maintenance costs (Item 5 of Table 3-8). Unit energy
costs are the total annual operating costs divided by the annual new power
generation (Item 6 of Table 3-8). The resulting energy costs are the overall
evaluation of plant cost and performance economics for comparison purposes,
assuming the same plant location and power transmission costs.
51
-------
The supercharged boiler combined cycle plant (Hl-L) has the lowest
overall energy costs, capital costs per kw, net heat rate and the lowest
fuel costs.
The three other Lurgi LEG plants (Al-L, Bl-L and Cl-1) have nearly
identical overall energy costs approximately 6 percent higher than Plant
Hl-L. Plant Al-L fired with Koppers Totzek fuel has overall energy costs
approximately 16 percent higher than Plant Hl-L; primarily due to the higher
fuel cost and the use of oxygen in the boilers.
3.1.5 Heat Transfer Surface Requirements
The combustion chambers for each of the six systems were analyzed with
Combustion Engineer's Lower Furnace Program described in Appendix B. This
program has demonstrated good agreement with test data from tangentially-
fired natural gas utility boilers, but has not been calibrated for other
fuel gases. Although the precise numerical results obtained here cannot be
verified with test data, they should be mutually consistent, allowing
accurate comparison of the six systems.
The results for tangential firing are listed in Table 3-9. Although
the Lower Furnace Model could not be directly applied to opposed wall firing,
the results should be similar. A comparison of tangential and horizontal
firing with natural gas based on Combustion Engineer's Engineering Standards
showed that for the same heat inputs, furnace size, and burner location, the
furnace outlet temperatures (and hence, waterwall absorptions) were nearly
identical for the two firing modes. Unfortunately the very high gas recircula-
tion for System 15 was well-beyond the range of C.E.'s Standards and compari-
son was impossible.
The results of the Lower Furnace Program (waterwall absorption, furnace
outlet temperature, and net radiation flux from the furnace) provide the
necessary inputs for determining the heat transfer surface requirements. A
modified version of a C.E. surfacing program was prepared to handle LBG
products. As with the Lower Furnace Program, a high degree of confidence in
the results listed in Table 3-9 should be assumed for mutual comparison only.
52
-------
Table 3-9. Summary of Heat Transfer Surface Requirements
System
Al-L
Bl-L
Cl-L
Hl-L
Al-K
I5-K
tn
co
Superheater (FTr)
Reheater (FT2)
Economizer (FT2)
Furnace & Backpass
(FT2)
64,676
74,735
439,833
34,354
64,552
92,220
247,690
30,452
54,367
67,891
31,871
2,514 TOT 52,693 48,585
5,169 TOT
19,227 TOT
72,412 64,827
317,742 243,573 TOT 243,690
77,290
30,452 29,492
Evaporator
Special Heaters
Fuel Heaters (2)
02 Heaters (2)
Evap - SH (3)
Par. FW (2)
252,800 TOT 97,500 TOT
45,200 TOT
517,242 TOT
233,108 TOT
169,430 TOT
* In cases where more than one unit was included in a classification such as
superheater, TOT refers to the total surface area for all such units.
-------
Material requirements for the furnace and boiler are generally similar
to those of conventional steam power plants. Low temperature components
such as the economizers coils are carbon steal with chrome-moly alloy steels
for high temperature superheat and reheat tube assemblies. Burner nozzles are
generally cast stainless steel. The boiler-combustors for system Hl-L will
contain a larger percentage of ferritic and stainless alloy steels due to
the expected higher gas temperatures.
Each system was examined for additional components which may require
special materials due to high temperature service. These components were
found to be:
System Component Service
Bl-L Fuel Piping Carries 1500°F gas @ 10 psig to Precooler
Cl-L Fuel Piping Carries 150QOF gas @ 10 psig to Burners
Cl-L Burner Nozzles Firing 1500°F gas
Hl-L Gas Pipe to Carries 2000<>F gas
Gas Turbine
Material selections were made consistent with these requirements for the
capital cost comparisons. To allow for uncertainty in the corrosive potential
of the fuel, overly expensive materials were selected in some cases but the
incurred increment in capital cost was not sufficient to decrease the accuracy.
of the comparisons.
3.1.6 Burner Designs
Full-load burner design parameters are presented in Table 3-10 for each
of the systems. Tangential burner preliminary designs have been prepared for
all systems except Hl-L. These burners are of the gas spud type. Systems
Al-L Bl-L Al-K and I5-K utilize tilting burners for reheat steam temperature
control and system Cl-L has a fixed nozzle design, relying on gas recirculation
54
-------
Table 3-10. Furnace Design for Burning Offgas from Coal Gasifiers - MCR Burner Design Parameters
CJI
in
FUEL
TO
BURNER
OXIDANT
TO
BURNER
GR TO
BURNER
MIXTURE
TO
BURNER
FURNACE
SYSTEM
OXIDANT
GR THRU WBOX
OVERFIRE AIR
EXCESS AIR
FIRED
GROSS HEAT INPUT
FLOW
TEMP.
PRESS.
DENSITY
FLOW
TEMP
FLOW
TEMP.
FLOW
TEMP.
DENSITY '
HEAT INPUT
NHP/PA
WIDTH
DEPTH
%
I
*
106BTU/HR
1068TU/HR
LBH/HR
°F
PSIG
LBM/FT3
LBM/HR
°F
LBM/HR
°F
LBM/HR
°F
LBM/FT3
106BTU/HR
106BTU/HR-FT2
FT
FT
Al-L
AIR
10
0
15
4640
4720
1,650,000
451
10
.0591
3.340,000
364
499,000
511
3,839.000
400
.0461
304
1892.50
43.33
43.17
Bl-L
AIR
10
15
15
3990
4310
1,420,000
750
10
.0445
2,870,000
487
429,000
543
3,299,000
493
.0417
335
2.64
40
40
Cl-L
AIR
10
15
15
3990
4710
1,420,000
1500
10
.0275
2,870.000
487
429.000
543
3,299,000
493
.0417
335
2.75
41
41
Hl-L
AIR
0
0
15
3650
3899
1,300.000
654
162
.5405
2.626.000
671
0
2.626.000
671
.3533
381
12.5 (Ei. of 3)
10.25
10.25
Al-K
AIR
15
0
15
4470
4520
820,000
250
10
.0650
2.970.00 0
460
570.000
523
3.540.000
467
.0432
337
2.86
40
40
I5-K
OXYGEN
205
0
5
4380
4598
807,000
710
10
.0395
611
392
2.910,000
784
3.521,000
731
.0393
631
3.08
40
40
-------
for steam temperature control. (Thermal expansion for this burner ruled out
the possibility of tilt).
Preliminary opposed wall burner designs were also prepared for each
system based on scaled-up Combustion Engineering Type R burners. Front and
rear wall firing with three rows of four or five burners each was utilized
for these systems. A single fixed vane (SV) burner mounted in the bottom of
the furnace was employed in each of the System HI boilers.
3.1.7 System NOX Emission Assessment
Although the Combustion Engineering study was primarily directed toward
the design and optimization of these systems from fuel utilization and economic;
viewpoints, NO emissions were also calculated.
/\
Combustion Engineering's Lower Furnace Program described in Appendix B
and reference 21 utilizes the axial temperature/time distribution of the
combustor products generated by the basic program in combination with an NO
formation rate based upon the Zeldovich mechanism. This program has been
calibrated for both oil and natural gas firing and calculates NOX emissions
within ±15 percent for oil firing and ±10 percent for natural gas firing.
The burner designs described in the previous section were analyzed with
this program and the results are presented in Figure 3-9 in relation to the
maximum temperature in the furnace. Emissions were assumed to contain 5 per-
cent of the total NOX as N02. As expected, since the Zeldovich mechanism is
very temperature sensitive, emissions rise dramatically with maximum
temperature.
Although the Combustion Engineering program cannot be directly applied
to the supercharged boiler*, it was applied to limit cases. The emissions
for the supercharged boiler system shown in Figure 3-7 cover a 28-fold range.
The minimum value is based on an atmospheric pressure boiler with the same
*The NO subroutine is limited to atmosphere pressure calculations.
A
56
-------
«°°
o
t—t
>•
(/)
00
£
o
I—I
-------
physical size and firing rate as the supercharged boiler. The maximum value
was extrapolated from the minimum value by correcting for the effect of
increased pressure in the Zeldovich mechanism. The basic lower furnace heat
transfer program does have a high pressure capability and it was found that
a change in pressure from one to ten atmospheres had very little effect on
peak temperatures, although the exit gas temperature was lower at high pres-
sure because of the increased overall absorption. The predicted temperature
profiles were relatively flat at approximately 2900°F and consequently no
temperature correction was necessary.
The increased NO emissions with increased pressure were estimated by
/\
evaluting the following two effects:
• The reaction rate is higher because of the increased partial
pressure of the reactants.
• The furnace residence time is longer at high pressures due to
radiant heat transfer requirements.
Habelt and Selker^21^ assumed the ratio of the NO concentration produced in a
cloud of combustion products of uniform composition to the NO concentration at
equilibrium to be given by:
[NO] -Cqt
= 1 - e i
[N0]eq
where:
t = residence time
Cj = 2K [N2J [0]/[NO]eq
K = 1.36 x 1014 e"37943/T cc/mol-sec
(reaction rate for N2 + 0 -»• NO + N)
T = temperature (°K)
58
-------
•=
They also assumed that for excess oxygen conditions [NO] is independent
of pressure. . q
Applying these relationships to the pressurized boiler with a 1.0 second
residence time at 10 atmsopheres and assuming equilibration of oxygen atoms,
the pressure correction factor was calculated to be 28.
The results of this simplified analysis can be considered at best a
rough approximation. Major questions regarding this procedure include:
• The lower furnace model is an empirical tool requiring "calibra-
tion" for different fuels and has not been used with LBG. It is
also specific for corner-fired boilers.
t The Zeldovich mechanism assumption is difficult to justify,
particularly if the majority of the N0x is produced in the heat
release zone because it neglects superequilibrium oxygen atom
concentrations and Fenimore-type reactions.
• There is no mechanism to account for NO formation from fuel
ammonia.
These limitations cast serious doubt on the accuracy of the C.E. com-
puter model applied to pressurized combustors firing LBG. Section 4 dis-
cusses the application of a modular kinetic model to calculate limit-case
emissions based on the best available kinetic mechanisms.
3.2 Integrated Gasifier/Power Plant Analysis
The comparative analysis in Section 3.1 demonstrated that subject to the
limitations of existing technology and a separate gasifier, the best alterna-
tive (of those examined) was a combined gas and steam (COGAS) cycle power
plant. In particular, the COGAS system had:
• Lowest heat rate (highest efficiency)
• Lowest capital cost
• Lowest annual fuel cost
t Lowest net energy conversion cost (mils/kWh)
59
-------
This section extends the analysis of COGAS systems to integrated gasifier/
power plants where heat and mass transfer in addition to fuel transfer
couple the two units. The purpose is to provide some general background
comments pertaining to thermodynanric cycle studies of gasifier-combined.
cycle systems and to suggest a general methodology for further systems
studies. An overview of COGAS systems is first presented summarizing some of
the important design variables and how they affect overall performance. The
discussion then focuses on gasifier design and losses and the trade-offs
between a supplementary-fired waste heat recovery boiler and a supercharged
boiler, two factors which strongly influence combustor requirements. Finally,
descriptions of six general classes of combustion devices which are most
likely to evolve in the course of gasifier-combined cycle plant development
are given.
3.2.1 An Overview of COGAS System Efficiency
The principal losses contributing to inefficiency in a coal gassification-
combined cycle power plant are: the condenser heat rejection; the combustion
product latent and sensible heat emitted through the exhaust stack; and any
heat loss associated with the gassification product cooling. Other less
significant losses are: heat transfer between various hot elements of the
system and the surroundings; exhaust stream energy emissions from auxiliary
equipment; and by-product streams containing combustibles, e.g., the tar-like
residue from the coal gasification process which may be emitted as a waste
stream or returned to the gasifier.
The efficiency of a coal gasification-combined cycle power plant can
be defined by:
Fffir-ionrv - Net Electrical Energy Output
trnciency - Higher Heating Value of Input Coal
This in turn can be expressed in terms of the system losses as:
Losses
Efficiency = 1 -
HHV of Coal
where losses include all energy outflow fluxes from the system except electri-
cal energy and the heat content of the coal is the only energy input to the
60
-------
system . The basic concept of combined cycle systems is to cut these losses
by effectively utilizing the high temperature turbine exhaust gases in a
Rankine cycle to raise steam. The high turbine exhaust loss is thus replaced
by condenser and boiler stack losses which combine to yield a lower total
loss. Variations on this theme come about when the turbine exhaust tempera-
ture is too low (due to low turbine inlet temperature) to permit an efficient
Rankine cycle. This low steam temperature and excess of available low grade
energy from the exhaust which cannot be utilized in an economizer leads to
high stack temperatures. Although the latter situation may be improved some-
what by incorporating a second low pressure steam cycle in the steam generator,
a more effective approach is to improve the "quality" of the turbine exhaust
by supplementary firing the turbine exhaust boiler, thereby raising its tem-
perature. Once the efficiency of the steam cycle has been sufficiently
improved, any further supplementary firing serves only to bias the combined
cycle towards a pure Rankine cycle, and hence, to lower the overall efficiency.
Provided the exhaust energy can be used effectively, the key to enhanced
efficiency is to increase the fraction of fuel energy extracted by the gas
turbine. This decreases the turbine exhaust energy, a large portion of which
is necessarily lost through the condenser and stack. Higher gas turbine
efficiency is accomplished through increased turbine inlet temperature. In
the idealized Brayton cycle, efficiency is a function of compressor pressure
ratio only and is independent of turbine inlet temperature. However, due to
inefficiencies in both the compression and expansion processes the relation-
ship between the temperature increase due to combustion and the temperature
rise associated with compression becomes an important factor. High compres-
sion ratios are permissible only with a high degree of energy addition in the
combustor which, in turn, leads to high turbine inlet temperature. High inlet
temperatures also lead to high exhaust temperatures, and hence, to more effective
Rankine cycle waste energy recovery systems. For low turbine inlet temperature,
optimum compression ratios for COGAS systems are somewhat lower than for stand-
alone gas turbines due to the trade-off between gas turbines and waste heat
Any electrical input to the plant is combined with the generator cutout tn
the net and any auxiliary heating is combined with the HHV of the coaK
61
-------
recovery efficiencies. Similarly, establishing optimum gas turbine reheat
and recuperation requires consideration of the complete COGAS system. This
is discussed further in later sections.
An alternative to supplementary firing the exhaust boiler and a more
effective approach to improving the utilization of turbine waste heat is
the integration of a supercharged boiler with the gas turbine combustor.
Here the supplementary fuel addition is upstream of the turbine rather than
downstream in the waste heat recovery boiler. Again sufficient steam raising
(boiling and superheating) is accomplished through additional firing of the
gas turbine combustor (supercharged boiler) to allow full utilization of the
turbine exhaust for feedwater heating. The greater effectiveness of the
supercharged boiler is a result of the greater mass and energy flow through
the gas turbine, and hence, greater turbine power extraction than for a fired
waste heat boiler.
Attempts to optimize net plant heat rate and define combustion devices
on the basis of separate studies of the gasifier and combined cycle plant
(treating the fuel as "coming over the fence") have inherent difficulties
and the results must be examined with considerable caution. This applies
equally to pollution potential, plant efficiency and cost benefit trade-off
studies. The results of these optimization attempts do not consider the
interaction of gasifier and power plan so that optimizing one unit could
degrade the performance of the other. The gasifier and combined cycle sys-
tems can be intimately coupled through several heat and mass transfers and
trade-offs:
• Heat exchange between the gasifier's raw gas (prior to H2S
removal) and the boiler feedwater
• Gasifier and gas turbine compressor requirements
• Gasifier and gas turbine combustor pressure losses
• Fuel gas temperature and composition
The next section describes the effects of some of these variables on
overall system performance and demonstrates that the effectiveness of the
62
-------
gasifier depends greatly on the manner in which it is matched to the combined
cycle. Thus, it is impossible to define a meaningful gasifier efficiency
without examining the coupled system. The following section examines the
trade-offs between a supplementary-fired waste heat recovery boiler and super-
charged boiler as system design parameters vary.
3.2.2 Gasifier Losses
The coal gasification process yields a product which-is lower in both
total energy and available energy than the total energy input. It is con-
venient to classify the energy losses as either "irretrievable" or "potential".
Irretrievable losses are those gasifier energy losses which cannot be utilized
by the power plant. Examples include:
• Water or steam added to the gasifier
• Carbon loss in ash
• Residual tars not returned for combustion or gasification
t Gasifier heat losses to the environment
Potential losses are those gasifier energy losses which could be utilized
by the power plant depending upon power plant design. Two important examples
include:
• Sensible heat in the fuel gas
t Work of air compression
, It is difficult to speak of gasifier efficiency or losses independently
since the gasifier and combined cycle plant are so closely coupled. However,
a useful figure-of-merit or efficiency for the gasifier portion of such a
system can be defined as the ratio of the net power plant efficiencies for
a gasifier-combined cycle system to the corresponding efficiency of a similar
combined cycle system burning coal directly (ignoring the design problems
and the energy penalties associated with corrosion suppression and particulate
and sulfur cleanup). For example, if a gasifier could be operated with losses
equal to those incurred burning coal directly, then it would have 100 percent
efficiency under this definition. Any degradation in plant performance from
this ideal situation is assigned as gasifier inefficiency.
63
-------
Using this definition, a typical one-atmosphere gasifier feeding a
conventional boiler (with or without preceding and feedwater heat exchange)
will have an efficiency of approximately 83 percent which corresponds to the
irretrievable losses cited above. A system mismatch, such as a high pressure
gasifier in conjunction with a combined cycle employing a low temperature gas
turbine feeding a fired exhaust boiler, will yield efficiencies on the order
of 73 percent. The additional 10 percent loss is due to throttling the major
portion of the fuel gas prior to entering the fired exhaust boiler. Another
example of mismatch is employing gasifier intercooling for feedwater heating
in an unfired turbine exhaust waste heat boiler. This not only diminishes
the turbine power but is especially ineffective because of the overabundance
of low temperature heat already available. Here the gasifier efficiency may
be on the order of 78 percent. On the other hand, a well-optimized gasifier-
COGAS system should be able to operate with 90 percent gasifier efficiency
corresponding to the minimum irretrievable losses.
3.2.2.1 Water Content - Irretrievable Loss
LBG and its products of combustion contain more water than the products
of combustion of coal. This is due to the introduction of steam in the
gasification process which provides hydrogen donation via the water/gas
reaction. In the air-blown Lurgi process the steam addition rate is approxi-
mately one pound of steam per pound of coal. Roughly 50 percent of this water
appears as water vapor in the LBG (which is knocked out if the gas is quenched)
and eventually all of it appears as additional water vapor in the final products
of combustion leaving the power plant stack. In either case its latent heat
is lost and the lower heating value (LHV) of the fuel is thus diminished. In
a conventional boiler burning a typical Lurgi gas the efficiency is degraded
by approximately 10 percent which corresponds to the full reduction of the
lower heating value due to the presence of additional water vapor. This con-
clusion is independent of gasifier intercooling. However, if the LBG is fired
in a gas turbine the additional water vapor adds to the working fluid passing
through the gas turbine much as in a Rankine cycle. That is, the expenditure
of the latent heat in the process of creating the vapor provides the turbine
with a high pressure gaseous working fluid without the penalty of compressor
64
-------
energy. Although the vapor is raised to high temperature in the gas turbine
it is not expanded over the very large pressure ratio of a Rankine cycle, and
hence, the incremental efficiency associated with the water vapor working
fluid is not up to typical steam cycle efficiency. In a typical COGAS system
employing an unfired waste heat boiler the incremental efficiency associated
with the additional water vapor is approximately 30 percent*. This is less
than the overall cycle efficiency, and hence, represents a degradation in
performance. The effective gasifier efficiency is down by 5 percent, or
approximately one-half of the penalty associated with burning the LBG in a
conventional boiler. In the COGAS system, as gasifier water injection increases,
the gas turbine power fraction and the fuel/air ratio both increase while the
turbine exhaust sensible heat, the steam-side power, and the condenser heat loss
all decrease more than with the conventional boiler, thus verifying the net
advantage of the COGAS system.
The supercharged boiler COGAS system has the same advantage indicated
above for the waste heat boiler. Supplementary-fired waste heat boilers and
fully-fired exhaust boilers suffer the same losses as conventional boilers,
but to a lesser degree.
3.2.2.2 Other Irretrievable Losses
Interceding the gasifier product knocks out approximately 50 percent
of the injected water and eliminates this portion of the water vapor as an
effective gas turbine working fluid. In this instance the effective gasi-
fier efficiency is reduced approximately 7% percent due to water injection
when coupled with supercharged or waste heat boiler COGAS systems. Thus
the effect of intercooling just due to the water vapor effect alone is roughly
one point in overall system efficiency.
InJ« III WhS- -?dd1t'°nal fuel 1s bu™ed to raise steam which is injected
into the gasifier, 30 percent of that fuel energy is converted into work.
65
-------
Losses associated with heat loss to the environment, carbon loss in the
ash, and residual tarts which are not returned for combustion as part of the
gasification process amount to typically 7-8 percent. This loss is probably
reducible to 5 percent with an optimum design.
3.2.2.3 Sensible Heat - Potential Loss
Typical LBG raw product gases leave the gasifier with sensible heat
equivalent to 10-15 percent of the heat of combustion of the coal input. For
pressurized systems, typically one-third of this can be associated with the
increase in energy resulting from combustion air compression. Some or all of
the sensible heat can be effectively used by the COGAS system depending upon:
the system integration of gasifier and power plant; the need for cold H2S
cleanup; and the combined cycle characteristics. In one extreme the sensible
heat may be dumped overboard in a quench process or it may be used for indus-
trial processes which require low quality heat. In the other extreme, the
raw gas may pass through a hot cleanup process (if available) and into a
combined cycle system with undiminished temperature in such a manner as to
utilize all of the sensible heat effectively. Alternately, if an unpres-
surized gasifier is used to feed a conventional boiler, all of the sensible
heat can be effectively utilized through a heat exchange with the feedwater
prior to cold cleanup and no losses are incurred. In between these extremes
are situations in which partial recovery is accomplished through a somewhat
ineffectual exchange of heat between raw gasifier products (prior to cold
cleanup) and various portions of the COGAS cycle.
A rough rule describing the cumulative effects of intercooling is that
it becomes more significant as turbine inlet temperature increases. There
are three potential effects of gasifier intercooling on performance:
• Loss of high pressure gaseous working fluid (water vapor)
through condensation.
• Reduction of gas turbine power fraction through decrease in
turbine energy throughput.
• Inability to effectively utilize low grade heat from inter-
cooler in feedwater heating or steam raising.
66
-------
The first effect was discussed in the last section and represents
typically one point in performance for supercharged and waste heat boiler
systems. The second effect is of little importance in low excess air super-
charged and low excess air exhaust-fired systems because the turbine energy
throughput is essentially unaltered. In these cases the reduction of fuel
enthalpy is manifested in a reduction of heat input to the supercharged or
exhaust-fired boilers. This loss is retrieved by feedwater heating and boil-
ing from the intercooler. However, for high excess air systems this effect
can be appreciable because turbine energy throughout is substantially affected
by fuel interceding. The effect can be several points in system efficiency.
Finally, the third effect can be significant for COGAS systems with
moderate turbine inlet temperature (2000°F) incorporating high excess air
boilers (supercharged or exhaust). In these systems there is an overabundance
of low quality exhaust heat for feedwater heating and boiling which makes it
impossible to effectively utilize the intercooler heat. The effect becomes
less significant for high turbine inlet temperature waste heat systems and
is nonexistent for low inlet temperature systems which employ low excess air
combustors. The latter have a sufficiently large quantity of high quality
heat input that they can effectively utilize the lower temperature inter-
cooler heat for feedwater heating.
3-2.2.4 Work of Air Compression - Potential Loss
Compression is manifested as an increase in enthalpy of the raw gas
product as well as an increase in its available energy. It is recovered in
part in any thermodynamic cycle which converts heat to work. However, if a
pressurized gasifier is utilized for a one-atmosphere conventional boiler
only about 35 percent of this energy will be recovered. The net effect is
a loss of mechanical energy equal to approximately 3 percent of the heat of
combustion of coal which is equivalent to a 9 percent reduction in gasifier
efficiency.
3-2.3 Supercharged Versus Fired Exhaust Boilers
If a conventional steam plant is fired with LBG from a one-atmosphere
gasifier there is approximately a 17 percent degradation of net heat rate
67
-------
over direct coal firing. This leads to an overall gasifier-steam plant
efficiency of roughly 32 percent using state-of-the-art technology. This
conclusion is independent of whether the raw gasification product is uncooled
or fired in the boiler at a high temperature because only low quality (i.e.,
low temperature) heat is required in a conventional boiler; there is no need
for heat input at the high peak temperature associated with combustion. The
usual boiler heat transfer process is over a large temperature differential
giving rise to high entropy production (irreversibility).
The utilization of the high temperature of combustion in a gas turbine
topping cycle is the key reason for increased performance in a COGAS cycle.
The waste heat.from the gas turbine exhaust is used as the heat input to a
conventional steam turbine bottoming cycle.
Depending upon overall cycle design, the heat in the gas turbine exhaust
may or may not have sufficient temperature for an optimum steam bottoming cycle.
If the temperature is too low, it can be raised either by burning more fuel in
the gas turbine combustor and transferring the heat to the steam cycle (super-
charged boiler) or by bringing more fuel directly in the turbine exhaust
(supplementary firing). The supercharged boiler is the preferred COGAS system
for use with LBG for turbine inlet temperatures up to levels well in excess
of 2000 F. The precise inlet temperature at which the unfired waste heat
boiler system becomes superior has not yet been determined, but it may be as
high as 2400°F. In this high temperature regime the supercharged boiler would
be fired with high excess air.
The principal reasons for the thermodynamic superiority of a supercharged
boiler over a fired exhaust boiler include:
• It allows the system to take advantage of additional water vapor
in the combustion products due to gasifier steam injection.
• It allows higher gas turbine energy throughput, and hence, higher
turbine fractional power due to increased mass flow through the
turbine.
68
-------
• Due to its high operating temperature, it more effectively
serves the basic purpose of providing high quality energy
input for steam raising which, in turn, allows the effective
utilization of the low quality turbine exhaust energy.
The cumulative effects lead to advantages which decrease as turbine
inlet temperature increases. At low turbine inlet temperatures the improve-
ment can amount to several percentage points in overall system efficiency.
The utilization of water vapor dominates the other advantages, and hence,
the attractiveness of supercharged boilers over fired exhaust boilers is
predominantly a consequence of firing LBG - a factor generally ignored in
most studies.
3-2'3>1 Low Turbine Inlet Temperature Combined Cycle Systems
Figure 3-10a shows a typical design for a counterflow unfired waste heat
boiler where the turbine exhaust gases first encounter the superheat section,
then the evaporator sections and finally the economizer. Figure 3-10b shows
the gas and steam side temperature distribution for a typical low turbine
inlet temperature case and illustrates two important and related points.
First the quality of the steam generated is poor with an attendant low steam-
side efficiency. ^Secondly, the "pinch point" constraint forces a high gas
stack temperature . There is, of course, a trade-off between attainable steam
quality and stack losses, and both of these factors'are related to turbine
exhaust temperature. When the exhaust temperature becomes sufficiently high
that an efficient steam cycle can be operated off the waste heat without
excessive stack losses, then there is no need for supplementary firing. For
temperatures below this critical value supplementary firing is required to
bring up the steam-side efficiency and minimize stack losses. This supple-
mentary firing can occur either in the gas turbine combustor (i.e., a super-
charged boiler) or in the waste heat boiler.
For turbine inlet temperatures of 2000°F and below there is a clear
advantage to the supercharged boiler fired by a pressurized gasifier. The
There is an excess of low quality heat which is unusable.
69
-------
—-Economizer
—• Evaporator
-— Evaporator
[-——Superheater
Gas Turbine Exhaust
Figure 3-10a. Unfired Steam Generator Design
40 60
HEAT TRANSFER-X
Figure 3-10b. Typical Gas and Steam Temperatures
Unfired Steam Generator
70
-------
advantage lies primarily in the higher fraction of system power delivered
by the gas turbine while the steam-side efficiency remains the same for both
systems. The low excess air supercharged boiler, typical of low turbine
inlet temperature systems, fully utilizes the gas turbine exhaust heat for
feedwater heating, and hence, the boiler stack losses are low. Further, the
condenser losses on the steam-side are lower for the supercharged boiler case
because of the lower fractional steam-side power. The higher fractional gas
turbine power for the supercharged boiler is due to the higher energy flow
through the turbine under these conditions.
Assuming the fired exhaust boiler is fed by a separate one-atmosphere
gasifier, then the supercharged system has roughly a three percentage point
advantage in overall system efficiency over the exhaust-fired boiler. The
advantage becomes substantially greater if the same high pressure gasifier
feeds both the gas turbine and exhaust boiler as is sometimes assumed in
making the comparison. Overall gasifier-COGAS efficiency for these low
turbine inlet temperature systems can reach approximately 37 percent. If
gasifier interceding is employed, then 36 percent efficiency can be achieved
at 2000°F.
3-2.3.2 Moderate Turbine Inlet Temperature Combined Cycle Systems
Most attention to date has been given to supercharged boilers and furnace-
fired exhaust boilers under the low excess air conditions which are optimum for
low turbine inlet temperatures. However, as turbine temperature increases, the
amount of supplementary firing required to bring the system to optimum per-
formance decreases with an attendant increase in excess air. Excessive supple-
mentary firing does little to enhance the steam-side efficiency while it
increases the steam-side fractional power, and hence, reduces overall efficiency.
The incremental efficiency associated with excessive supplementary firing (that
is, the efficiency of conversion of the incremental addition of fuel) at 2200°F
turbine inlet temperature is on the order of 37 percent for the exhaust boiler
and 38% percent for the supercharged boiler. These figures are high in com-
parison to a conventional Rankine cycle considering the gasifier losses. This
is a consequence of the low incremental change in stack losses due to the
decreasing excess air. The overall cycle efficiency is greater than 40 per-
cent for the optimum supercharged boiler with 2200°F turbine temperature,
and hence, excessive supplementary firing degrades the system performance.
71
-------
Figure 3-11 illustrates qualitatively how the supercharged boiler COGAS
system efficiency depends on excess air and turbine inlet temperature. For
low turbine inlet temperatures the performance increases with decreasing excess
air until the minimum excess air line for satisfactory combustion is reached.
As inlet temperatures increase the optimum excess air level becomes greater
than the minimum line. To the left of this optimum point (that is, at lower
excess air) the system performance increases with increasing excess air. The
curve marked (*2000°F) shows equal system performance for the minimum excess
air supercharged boiler and the gas turbine-waste heat boiler. This result
is often misinterpreted as meaning that for turbine temperature above 2000°F
it is no longer desirable to use a supercharged boiler. As can be seen from
the figure, this is not the case. Under optimum excess air conditions the
supercharged boiler appears to be attractive for turbine temperatures well
above 2000°F. The actual turbine temperature for which the supercharged
boiler is no longer attractive, and for which performance increases mono-
tomically with increasing excess air, has not been established at this time.
3.2.3.3 High Turbine Inlet Temperature Combined Cycle Systems
Again referring to Figure 3-11, as turbine inlet temperature climbs, a
point is eventually reached where the steam cycle receiving heat from an
unfired waste heat boiler is sufficiently efficient that any supplementary
firing will not improve steam cycle performance but will degrade overall
performance by reducing the gas turbine power fraction. The optimum waste
heat boiler COGAS systems have efficiencies which can reach 50 percent with
turbine inlet temperatures of 2800°F. Gasifier interceding would degrade
this efficiency by several percentage points.
3.2.4 Generalized Approach to Cycle Analysis
The preceding discussion demonstrated that the influence of system
design and operational variables on combustor requirements is subtle and
easily defies intuition. A particular danger in such an analysis is the
omission of sufficient generality of concepts and attention to detail in the
approach. For example, if the analysis above had been restricted to unfired
waste heat boilers or low excess air supercharged boilers, an entire class
72
-------
0)
•r—
O
-------
of concepts would have been omitted. These are the "high excess air"
supercharged boiler and the "supplementary-fired" waste heat boiler which
may exist singly or together in the optimum system. Similarly, one might
conclude that several stages of gas turbine reheat are desirable on the
basis of no pressure drop in the reheat combustors, but this, conclusion
might prove wrong upon inclusion of a more detailed treatment.
A generalized and systematic analytical approach is required to properly
define optimum combustors under various combined-cycle constraints. Six basic
types of combustors should be considered. The four most common combustors are
shown schematically in combined-cycle application in Figure 3-12 and are des-
cribed below.
• Adiabatic Gas Turbine Combustor (Figure 3-12a). Figure 3-13
shows a Brown Boveri design for an industrial gas turbine firing
LBG.
t Furnace-Fired Turbine Exhaust Boiler (Figure 3-12b). This is
essentially a conventional LBG boiler which receives vitiated
combustion air from the turbine exhaust.
• Supplementary-Fired Turbine Exhaust Boiler (Figure 3-12c). In
order to increase the effectiveness of waste heat boilers it
may be cost-effective for systems with moderate turbine inlet
temperatures to supplementary fire a combustor located just
upstream of the waste heat boiler. This combustor uses only a
portion of the available oxygen contained in the gas turbine
exhaust and is designed to limit the maximum gas temperature
entering the steam generator to approximately 1200 to 1300°F.
This allows the steam generator to maintain its relatively
simple arrangement shown in Figure 3-10. The combustor must
be designed to handle high excess air and high gas velocities
without flame instability. With conventional fuels the burners
are located directly in the duct work ahead of the boiler.
0 Supercharged Boiler (Figure 3-12d). This is the system which
has received considerable attention in the course of the present
investigation.
74
-------
FUEL
COND.
- EXHAUST
Figure 3-12a. Gas Turbine Plus Unfired Steam Generator
FUEL
COND.
EXHAUST
Figure 3-12b. Gas Turbine Plus Furnace-Fired Steam Generator
75
-------
FUEL
COND.
EXHAUST
Figure 3-12c. Gas Turbine Plus Supplementary-Fired Steam Generator
EXHAUST
AIR
FUEL
COND.
Figure 3-12d. Supercharged Furnace-Fired Steam Generator
Plus Gas Turbine
76
-------
External Combustor
Exhaust ^
Multi Stage
Turbine
Axial Flow
Compressor
Air Inlet
Figure 3-13. Brown Boveri Industrial Gas Turbine with External Combustor
-------
In addition to the four combustor types indicated above there are two
more which deserve attention. These are shown schematically in combined-
cycle application in Figure 3-14 and are described below.
• Gas Turbine Reheat Combustors (Figure 3-14). There is a cycle
advantage to interstage reheating in the gas turbine portion of
the cycle. The increase in turbine power fraction can 'be very
substantial, while at the same time increasing the exhaust tempera-
ture, and hence, the effectiveness of the waste heat recovery. The
excess air level is continuously decreased as more stages are added.
Particular attention must be given to excessive reheat combustor
pressure drop in optimizing such a system. Use of reheat combustors
is referred to as "carnotizing" the gas turbine cycle and their use
in combined cycle application should prove effective.
• MHD Combustors (Figure 3-14). MHD combined cycles fired with LBG
offer the potential for a high efficiency power plant, and hence,
the associated combustors deserve attention. Very high combustion
temperatures (5000°F) are required which implies a high degree of
recouperation as shown in Figure 3-14. An "interstage" recouperator
followed by a conventional low pressure turbine stage is shown.
This allows high temperature recouperation which could not be
obtained otherwise. NO control via post-flame reduction and
A
hold-up in the recouperation and waste heat boilers are possi-
bilities to be investigated.
Figure 3-14 also shows the flow diagram for a suggested generalized
cycle analysis computer program where all six types of combustors are
potential elements in the system. Optimization would include the selection
of the appropriate combustor type for a given set of constraints (e.g., tur-
bine inlet temperature). Trade-off studies of pollutant potential versus
systems efficiency could be conducted with parametric variations of combustor
excess air and temperature around the optimum design point. Both high pres-
sure and one-atmosphere gasifiers are included. This allows the possibility
of feeding a furnace-fired or supplementary-fired turbine exhaust boiler with
low pressure LBG.
78
-------
BYPASS
UD
("I
HIGH P
FUEL GAS
FEEOWATER HEATER
OR
WASTE HEAT BOILER
POWER IN
•MHO DUCT-
..
POWER OUT
. LJ
• orn—
POWER OUT
BYPASS
STEAM
COAL
ONE-
ATMOSPHERE
GASIFIER
(PROVISIONS FOR
INTERSTAGE COOLING)
1-ATM
FUEL GAS
TO GENERAL
STEAM CYCLE
CONDENSER
WASTE HEAT
EXH
FEEDWATER HEATER
FEEDWATER
AIR
FUEL.GAS GENERATOR
Q) ADIABATIC GAS TURBINE
© SUPERCHARGED BOILER WITH
VARIABLE EXCESS AIR
CD SUPPLEMENTARY COMBUSTOR
FOR WASTE HEAT BOILER
COMBINED CYCLE
REHEAT COMBUSTOR
FURNACE-FIRED BOILER USING
TURBINE EXHAUST PRODUCTS
MHD COMBUSTOR
GAS TURBINE
POWER OUT
<>REC
RECOUPERATION
Figure 3-14. Generalized Thermodynamic Cycle Analysis Flow Diagram
-------
The flowchart labels the "fuel in", "gasifier steam in", "compressor
power in", "turbine power out", "condenser heat ejection", and "exhaust
stack heat ejection" points in the cycle. This information, appropriately
manipulated, provides all of the gross cycle efficiency characteristics.
Much significant detail has been omitted from the flow diagram in order
to emphasize the principle features. Feedwater heating through regeneration,
economizers, and gasifier precooling are shown only implicity. Similarly,
steam side reheat, gas turbine recouperation, and compressor interstage
cooling are not shown in detail. Such refinement as desuperheat control
are not shown at all, but should be included in the analysis.
3.3 Conclusions and Combustor Definitions for NOX Emission Studies
Based on the results of this study, several conclusions can be dr"awn.
• The COGAS cycle firing LBG from an integrated gasifier has high
potential efficiency when compared with other advanced power
generating systems.
• Interceding the fuel gas between the gasifier and combustor is
undesirable.
• Supercharged boilers offer distinct advantages over fired exhaust
boilers.
A properly designed and coupled gasifier-combined cycle system with sulfur
removal can potentially achieve fuel utilization 90 percent as effective as the
hypothetical direct firing of pulverized coal in a comparable combined cycle
system without sulfur recovery. The integrated gasifier-combustor can be
thought of as a multistage combustion system with a rich gasifying primary
incorporating water injection followed by hot H2S cleanup and a final burnout
stage. Further, if one speculates that turbine inlet temperatures under
direct coal firing will necessarily have to be lower than with LBG firing
due to difficulties in particulate removal, then the two systems may have
fully comparable overall efficiencies . Overall conversion efficiencies of
A 200°F reduction in turbine inlet temperature has roughly the same impact as
the gasifier losses. This comparison does not include any energy penalties
associated with sulfur and particulate cleanup in a direct fired system, but
includes cleanup in the LBG system.
80
-------
gasifier-combined cycle systems of 40 percent or greater can potentially
be achieved with turbine inlet temperature of 2200°F and potential efficiencies
approaching 50 percent are attainable as turbine inlet temperatures climb to
2800°F.
In general, interceding of the gasifier product degrades system performance
and the magnitude of the degradation increases with turbine inlet temperature.
Gasifier interceding has essentially no effect for the limiting COGAS system
in which an atmospheric-pressure furnace-fired exhaust boiler receives hot
combustion air from a low temperature gas turbine whose power is a small
fraction of the total system power output. On the other hand, there is
substantial degradation in a COGAS system where all of the LBG is burned in
a high temperature gas turbine combustor and steam is raised only in an unfired
waste heat boiler. For high excess air supplementary firing to raise steam
either in the gas turbine combustor (supercharged boiler) or the waste heat
boiler, interceding effects are also significant. For the low excess air
supercharged boiler there is roughly a 2-1/2 percent degradation in gasifier
efficiency due to knocking out the LBG water vapor (roughly 50 percent of the
injected steam remains as water vapor in the raw LBG), an effect not present
with the atmospheric furnace-fired exhaust boiler. This degradation is not
observed if the LBG gas composition is held fixed during intercooling. All
of the above conclusions assume utilization of the intercooling heat for
feedwater heating and boiling to the extent allowable under the constraints
of the particular system.
Supercharged boilers are preferable to fired exhaust boilers and the
advantage increases as excess air decreases. Optimum excess air increases
with increasing turbine inlet temperature. In making such a comparison it
is assumed that the exhaust boiler is fired with an atmospheric gasifier.
However, it may be unrealistic to have a separate one-atmosphere gasifier in
the high excess air case, and thus advantage of the supercharged boiler may
be even more substantial.
At low excess air the supercharged boiler advantage is largely a conse-
quence of gasifier water injection. The additional water vapor in the products
of combustion is most effectively utilized by expanding it through the tur-
bine. Thus the advantage of the supercharged boiler becomes considerably
81
-------
more pronounced with LBG than with conventional fuel. Robson^ ', for example,
shows only a one percentage point advantage for the conventional fuel low
excess air supercharged boiler over the fired exhaust boiler. With LBG the
advantage can be three percentage points in efficiency.
The supercharged boiler COGAS system with low excess air is less effective
than unfired waste heat boiler systems for turbine inlet temperature in excess
of 2000°F (low excess air is the optimum firing condition only for inlet tem-
peratures well under 2000°F). However, if the supercharged boiler is optimized
with regard to excess air, it then maintains an advantage over other COGAS
systems until turbine inlet temperatures are considerably higher, possibly in
excess of 2400°F, at which point the unfired waste heat boiler has the
advantage.
82
-------
4.0 ESTIMATES OF NOX EMISSIONS FROM LBG COMBUSTORS
The analyses in the previous chapter identified the combined gas and
steam turbine power plant fired with LBG from an integrated coal gasifier
as the most desirable of the advanced power generating systems from a fuel
utilization point-of-view. This chapter discusses the NO emissions from
combustors suitable for this system focusing on NO produced by reduction
of nitrogen-bearing fuel species such as NH3 and HCN.
The purpose of this study is to develop an understanding of the impor-
tant N0x formation mechanisms and to adjust the combustion parameters of
two specific LBG combustors to minimize NO . A kinetic model, which includes
/\
a reaction set capable of describing NOY formation and destruction, has been
A
applied to a series of "limit-cases" based on practical operating constraints
of combustors fired with ammonia-containing LBG. Although macro-and micro-
scale mixing will dictate the NOX levels in any practical system, this model
establishes upper and lower emission limits, and thus, gives an indication of
the feasibility of direct use of high ammonia fuel gases.
The next section outlines the methodology for estimating NO emissions
and describes the kinetic model in more detail. A following section (4.2)
describes the fuel and combustor characteristics examined and the last two
sections (4.3 and 4.4) give the results for the two specific combustors
analyzed.
4.1 Methodology for Estimating NQV Emissions
Having identified two specific combustors of interest, the adiabatic
gas turbine and the supercharged boiler, "limit-case" estimates for NO
emissions were made over a range of system constraints on inlet conditions,
fuel composition, and exit conditions.
Limit-case situations are those in which some single aspect of the
physics or chemistry dominates and the remaining phenomena can either be
ignored or modeled by simple idealized processes. We use the term here to
refer to the methodology whereby we search for the lower bounds on NOX emis-
sions which are set by the chemistry. Implicit in this approach is the
assumption that all physical processes associated with fuel-air-product
83
-------
contacting and heat transfer can be accomplished in an ideal and optimum
manner dictated by the chemistry. Having established this chemical-lower-
bound, the question of how close one can achieve the optimum physical trans-
port behavior is then open to scrutiny.
Fundamental to the success of this approach is, of course, an authori-
tative description of the finite-rate chemistry pertinent to fuel nitrogen
(in this instance NH3) conversion and thermal fixation of Ng. Of particular
importance is the inclusion of the proper chemistry to describe the genera-
tion of stable intermediate nitrogen compounds (e.g., HCN, NH^) and their
subsequent equilibration path under rich conditions. The role of fuel break-
down and the effect of hydrocarbon radicals on the production and destruction
of nitrogen compounds must adequately be described as must the influence of
XN compounds on NO production and distribution. Appendix C sets forth a
A
proposed kinetic mechanism for NO formation in low Btu gas combustion. This
X
set of approximately 100 reactions was employed in all of the NOX estimates
of this investigation. The rationale for selecting this set of reactions and
rate constants is described in detail in the appendix.
In searching for the lower bound on NO emissions set by the chemistry
/\
it is assumed that the optimum chemical reactor system which achieves this
lower bound can be described in terms of an interacting set of stirred and
plug flow reactors which exchange matter and heat with one another and with
the surroundings. The optimum coupling between these basic elements which
yields minimum NO emissions is unknown at the outset of the search and may
A
indeed be very complex involving several parallel and feedback paths as
illustrated in Figure 4-1. The simplest (computationally) of these systems
is shown in Figure 4-la which illustrates a "series" or sequentially-staged
set of coupled reactors. SR refers to stirred reactors and PFR to plug flow
reactors. Q represents heat loss. The Ultrasystems Modular Kinetics Analysis
Program (MKAP) performs the nonequilibrium chemistry and element-to-element
bookkeeping on such a coupled system and provides for an arbitrary distribu-
tion of mass and energy transfer along the PFR elements. Figure 4-lb illus-
trates hypothetical systems with feedback loops. In one case the feedback
84
-------
(A)
AIR
(F)
FUEL--
F,
SERIES CONNECTED
4. la
F-
A
F,A
FEEDBACK LOOPS
4.1b
PFR
PFR
T
SR
Figure 4-1. Examples of Basic Element Coupling for Limit-Case
Investigations
85
-------
interaction is shown of two well-stirred zones with different residence
times and heat loss factors. The other illustration shows a plug flow zone
with distributed feedback from both exhaust products and from a well-stirred
zone. Such a system might be used for example, to model a free shear layer
combustion zone with distributed fuel addition which is stabilized through
coupling to a recirculation zone, and which is further complicated by the
presence of external exhaust gas recirculation. Figure 4-lc shows examples
of "parallel" couplings in which no influence of downstream behavior is felt
upstream. However, continuous exchange between reactor elements is allowed.
The upper sketch indicates how parallel couplings of this sort can be used
to simulate diffusion flame behavior wherein the flame zone is modeled by a
distribution of stirred reactors at various equivalence ratios. The products
from this flame zone are then transported back to the lean and rich sides of
the flame.
A definitive methodology for determining the optimum interacting set of
reactors is not known at present, and is the subject of current theoretical
and experimental research. For the purposes of this study, we have adopted
a simpler notion. It has been assumed that a close estimate for the lower
NO emission bound set by the chemistry can be obtained by examining a simple
series coupling of a rich primary plug flow reactor followed by a secondary
reactor which completes the combustion and achieves the correct combustor
exit conditions. The combined system of reactors is subject to certain
design constraint bounds on overall heat transfer rates and residence
times (size).
The principle variables over which an optimum was sought are primary
zone equivalence ratio, temperature, and residence times and secondary zone
combustion air entrapment rate and heat transfer rate. The SR and PRF
subelements of the MKAP program are standard chemical kinetic calculation
routines using fully implicit integration. They are documented in Reference 18
4.2 General Characteristics of Combustors and Selected Fuel
The analyses in Section 3 point to two types of combustors for use in
advanced power generating systems designed for optimum fuel utilization. The
following combustors were selected for the NOX emission studies.
86
-------
FUEL
AIR
H
i
^^•••^•^M
1
V~*"" PFRi
J
1
PFR2
1
l»- PFR.
J J
1
(
(
{
4.1c
Parallel Connected Elements
Figure 4-1. Examples of Basic Element Coupling for Limit-Case
Investigations (Continued)
87
-------
• Adiabatic gas turbine combustor
t Supercharged boiler
The adiatatic gas turbine combustor will be optimum for future COGAS
systems when turbine inlet temperatures reach 2400°F and above. For maximum
performance it will require no interceding of the gasifier product. It is
implicitly assumed that hot gas H2$ removal techniques will be developed on
a time scale consistent with the development of high temperature turbines.
Figure 4-2 is a partial schematic diagram of a COGAS power plant with
integrated gasifier and an adiabatic gas turbine combustor. Only the major
heat, work and mass flows important to combustor design have been shown. The
compressor feeds both the pressurized gasifier and the combustor with com-
pressed air at 10 atmospheres and temperature up to 1000 F depending upon the
degree of recouperation (not shown). The combustor operates at high excess
air to maintain combustor outlet temperature (turbine inlet temperature) at
2800°F.
A low excess air supercharged boiler firing low temperature LBG from an
interceded gasifier with a turbine inlet temperature of 2000 F was selected
as the second combustor for this study for three reasons. First, as distinct
from the adiabatic high temperature gas turbine combustor, this combustor
represents the optimum choice for a COGAS system when restricted to present
state-of-the-art technology in regard to H2S clean up and turbine inlet tem-
perature limitations. Second, although the choice of low excess air is not
necessarily optimum for 2000°F turbine inlet temperature it is representative
of designs suitable for use with low turbine inlet temperatures and it serves
as a vehicle to illustrate the distinctive features of low excess air firing
as distinct from the pure gas turbine combustor. And third, the supercharged
boiler appears to be the optimum choice for the near future operating at
higher excess air levels with turbine inlet temperatures approaching 2400 F.
Hence, its study at this time is pertinent not only to immediate COGAS system
development, but also to developments over the next decade.
Figure 4-3 is a partial schematic diagram of a COGAS power plant with a
supercharged boiler. A heat exchanger and low temperature cleanup system
have replaced the high temperature cleanup system in the previous example
88
-------
COMPRESSOR
AIR
INLET
CD
a-
COMPRESSED AIR
COAL
HOT
LBG
o
on •-•
=>U-
(O i-.
to to
UJ *^t
fV {J
a.
I
STEAM
FROM
RANKINE
CYCLE
HIGH
TEMP
CLEANUP
SYSTEM
HOT
LBG
SULFUR
COMPOUNDS
PARTICULATES
a--
ADIABATIC
GAS TURBINE
COMBUSTOR
STEAM
TO
RANKINE
CYCLE
WASTE
HEAT
BOILER
EXHAUST
TO
STACK
Figure 4-2. COGAS Power Plant with Adiabatic Gas Turbine Combustor
-------
SUPERCHARGED
BOILER
TURBINE
AIR
INLET
VO
o
1 , a
J_L _
Q
INI OC
M LU
QC •— •
Z3 U.
to •-•
to to
K S
Q.
I
iAT
(CHANGER
^W5
STEAM
TO
RANKINE
CYCLE
1 LOW TEMP.
SYSTEM
1 1
SULFUR PARTI
COMPOUNDS
STEAM
FROM
RANKINE
CYCLE
STEAM
TO
RANKINE
CYCLE
STEAM
TO
RANKINE
CYCLE
WASTE
HEAT
BOILER
\
EXHAUST
TO
STACK
Figure 4-3. COGAS Power Plant with Supercharged Boiler
-------
and the LBG is assumed to enter the combustor at 150°F. The supercharged
boiler operates with low excess air (nominally 5 percent) and outlet tem-
perature is controlled by heat removal to the Rankine cycle.
Table 4-1 summarizes the characteristics of the two combustors selected
for analysis.
Table 4-1. Combustor Parameters
Parameter
Air temperature
Air pressure
Fuel gas temperature
Fuel gas pressure
Outlet temperature
Excess air
Heat removal
Adiabatic
Gas Turbine
Combustor
1000
10
1500
10
2800
High
No
Supercharged
Boiler
Combustor
600
10
150
10.
2800
5
Yes
Units
°F
atm
°F
atm
°F
%
—
Unfortunately, the selection of a fuel gas for analysis is not as straight-
forward. Section 2 demonstrated that the properties of LBG vary considerably
depending upon gasifier design and operation and the characteristics of the
coal feed and other inputs. Since the purpose of this analysis is to examine
the formation of NOX, and in particular, the contribution of nitrogen-bearing
fuel species and the role of hydrocarbons, the fuel described in Table 4-2 was
selected. This LBG has a species distribution similar to that produced by an
air-blown (LBG) gasifier as listed in Table 2-1. Since the LBG is assumed to
enter the combustor after a cleanup process, the H2S and COS components were
eliminated. The fuel was then doped with 4000 ppm of ammonia, an amount esti-
mated by Robson( , as a potential level associated with high temperature
gas cleanup.
91
-------
Table 4-2. LEG Composition Assumed for NO Study
A
Specie Mole %
H20 10.1
H2 19.6
CO 13.3
C02 13.3
CH4 5.5
N2 37.6
NH3 Up to 0.4
4.3 Adiabatic Gas Turbine Generator Results
In order to establish the chemical constraints on the control of NOV
/\
formation in adiabatic gas turbine combustors burning LBG, a set of numerical
computations was performed using the MKAP methodology outlined in Section 4.1.
The intent of these computations was to expose the dominant chemical mechanisms
leading to NO formation and destruction and to establish their dependence on
the gross design features of gas turbine combustors. An additional goal was
to set estimates on the lower achievable limits of NOX emissions as dictated
by chemical considerations.
4.3.1 General Flame Characteristics
The first group of calculations was done to establish general combustion
characteristics of the selected LBG fuel gas such as flame temperature varia-
tion with stoichiometry and autoignition temperature for premixed fuel and
air at the inlet conditions. The peak adiabatic flame temperature is 3660°F
*
and occurs at an equivalence ratio of 1.05. This high value is due to the
compensating effects of:
§ high air.preheat (due to air compression in both the gasification
and final combustion processes)
^Approximately 75°F less than adiabatic flame temperature for coal/air with
room temperature air and 100°F higher than that for methane/air.
92
-------
• losses associated with the Lurgi process due to water addition
(dilution and latent heat effect), heat loss and tar losses.
An overall equivalence ratio of 0.45 corresponds to the desired turbine inlet
temperature of 2800°F.
Using the MKAP program it was found that the autoignition temperature for
this fuel is approximately 1400°F over a broad range of fuel/air ratio. This
is on the order of 100°F above the unignited mixture temperatures for all but
very rich mixtures, and hence, for the premixed conditions of interest here
an external energy source or backmixing of hot products is required for igni-
tion and flame stability. For the numerical experiments performed in this
study a 1.0 msec residence time adiabatic-stirred reactor was used to provide
ignition. In the limit of very rapid initial mixing this is a reasonable
representation of the ignition process, although residence times may be con-
siderably longer. Slower mixing of the reactant (even in the absence of back-
mixing) may give rise to diffusion zones which can attain autoignition condi-
tions in locally-rich regions without the presence of an ignition source.
The effect of ignition zone residence time is examined later in the
study. In general, it is found that if the stirred reactor residence time is
increased while maintaining total (plug flow and stirred reactor) residence
time fixed, then there is little effect for fuel-lean systems. Under rich
conditions, systems with larger stirred reactors appear to be richer in their
general characteristics with higher concentrations of hydrocarbon fragment and
lower concentrations of oxidative radicals.
4.3.2 Premixed Lean Combustion
Having determined these flame characteristics, the second task was to
establish the NOX emission levels under premixed conditions at the overall
equivalence ratio of 0.45. This situation then becomes a point of departure
to examine other staged combustor concepts. It serves as a useful yardstick
with which to measure NOX control effectiveness. Figure 4-4 shows a schematic
of the simple MKAP analogy to this situation where a stirred reactor ignition
region is followed by an adiabatic plug flow reactor. Figure 4-5 shows the
NO growth with time at pressures of 1 and 10 atmospheres for the case of zero
93
-------
LBG Fuel Gas, 1500°F, 4000 ppm NH3
P=10 ATM
Outlet Conditions
0=0.45,1 = 2800°F
.Ignition
T = 1.0 ms
Adiabatic Plug Flow
Air 1000°F
Figure 4-4. Premixed Adiabatic Gas Turbine Combustor - MKAP Analog Schematic
-------
100
90
80 -
70
60
o
50-
o
o
40
10
100
\
P-10 atm.
Corresponds to Simple
Zeldovich (/F Dependence)
200
TIME (MSEC)
300
400
Figure 4-5. NO from Lean Premixed Combustion - Adiabatic
Gas Turbine Combustor - No Fuel Ammonia
NOTE: = 0.45; Toutiet = 2800°F; (NH3)0 = 0.0
95
-------
NH- content in the fuel. Prompt NO of 6 ppm and 23 ppm is indicated for the
1 and 10 atmosphere cases, respectively. The NO growth rate follows the simple
Zeldovich mechanism for lean systems as expected. Making the usual Zeldovich
assumptions (0/02 equilibration, NO far below equilibrium, and N radicals at
steady state concentration) and using first Zeldovich (N2 + 0 •* NO + N) reaction
rate constant from the Baulch{19) (as is used in MKAP, see Appendix C) then It
follows that the rate of NO formation is given by
[NO] - 0.21 exo [.37.4 (^ - 1)] | [NZ]
where f"N9l and fo?l are mole fractions, P is in atmospheres, T is in K, and
[NO] is in ppm/msec. This gives at P =1, T = 1800, and = 0.45**:
[NO] = 0.044 ppm/msec
which matches the slope of Figure 4-5. Furthermore, the 10 atmosphere curve
shows a production rate which is (10)*5 higher than the one atmosphere slope
which is again consistent with the simple Zeldovich mechanism.
A similar and more convenient expression for the temperature dependence
of NO production rate is obtained by logrithmetically differentiating the
Prompt NO is here defined as the extrapolation of the linear portion of the
curves to zero time. It includes the NO produced in the 1.0 msec stirred
reactor which is primarily a consequence of high superequilibnum 0-atm
concentrations.
For the assumed LBG composition the N2 and excess 02 mole fractions are
(with C02 and H20 products of combustion):
**
0
.45
.53
Tf(°F)
2800
3000
N2
.68
.67
°2
.085
.069
96
-------
above Zeldovich expression and approximating the resulting expression for
temperatures near 3000°F. Thus:
..
[NO]
_ [37.4(1800) lldT
L T - 2J T
which for T = 3000°F (1922°K) gives
L or dT
NO
or
[NO] =0.53). The NO
production rate at 3000°F is approximately six times that at 2800°F which
reflects the temperature increase as well as the slight reduction in N? and
excess 02 concentrations . On the other hand, note that the prompt NO (inclu-
ding that formed in the 1.0 msec stirred reactor) has increased only by a
factor of two which implies that 0-02 equilibration is more rapid at higher
temperatures with an attendant lower fractional overshoot.
In an attempt to rationalize the prompt NO level in Figure 4-5 the oxygen
atom superequilibrium overshoot has been plotted in Figure 4-7. The very
high levels in the 1.0 msec stirred reactor and during the first two milli-
seconds of the plug flow reactor are responsible for the prompt NO production.
This is somewhat less of an increase than predicted by the simple Zeldovich
mechanism. r w»n.n
97
-------
500
200
TIME (MSEC)
300
NO from Lean Premixed Combustion — Adiabatic
Gas Turbine Combustor - No Fuel Ammonia
Figure 4-6.
NOTE: 4> = 0.53; Toutlet = 3000°F: (NH3)0 = 0.0
98
-------
500
400
300
200
20
TIME (MSEC)
30
Figure 4-7. Superequilibrium 0 Concentration, Lean Premixed
Combustion - Adiabatic Gas Turbine Combustor
NOTE: = 0.45; Toutlet = 2800°F; P = 1.0 atm; (NH3)Q = 0.0
99
-------
Without bound nitrogen in the fuel a 10 atmosphere premixed system at
2800°F flame temperature and reasonable residence times (on the order of
100 ms), has an NO level (corrected to stoichiometric) of approximately
85 ppm. The attainment of such a low level, however, in the absence of
premixing is a formidable task. Fuel/air contacting must be achieved in
such a manner as to minimize the residence times in zones of near-unity
stoichiometric ratio through rapid dilution in order to prevent substantial
NO production under locally high temperature conditions.
In Figure 4-8 the same premixed calculation at 2800°F is repeated with
the addition of 4000 ppm of NH3 in the fuel. Here we see an immediate con-
version of 86 percent of the NH3 to NO. Following this "prompt" conversion
the rate of increase of NO is again given by the simple Zeldovich mechanism.
Clearly the resulting 2000 ppm NO level (corrected to stoichiometric) is an
indication that more effective control measures must be sought. The next
series of calculations examine other control alternatives.
4.3.3 Staged Combustion
Figure 4-9 presents a MKAP analog schematic of a large-scale gas turbine
combustor concept based on staging. A long residence time rich primary
"hold-up" zone is provided in anticipation that it is possible to "cook" the
fuel NH3 to N2 in an oxygen-depleted environment given sufficient time. The
primary zone is followed by secondary burnout and dilution to the desired
turbine inlet temperature. In this lean secondary process any residual nitro-
gen compounds will convert to NO. The design challenge for the secondary and
dilution zones is to avoid diffusion flame production of thermal NO in locally
hot, near-stoichiometric regions. One possibility would be to reduce the
primary zone exit temperature by heat exchange with the dilution air as indi-
cated in the schematic.
4.3.3.1 Fuel Nitrogen Conversion Under Rich Conditions
The nonequilibrium chemistry associated with the transfer of fuel nitro-
gen to N?, NO, or HCN in a rich primary is complex and not well-understood.
However, two significant results can be derived from an examination of the
100
-------
Q.
D.
1200
non
1000
900-
win.
700.
ROD.
ROO,
400
300.
?on-
TOO.
0.
NH3 IH
TYPICJ
m., = 1015 PPM (-4
L LURGI LBG
~~ 86% CONVE
300 PPM IN FUEL)
_ —
:RSION OF (NH3)o
83% CONVERSION OF (NHjo
1.0 MS STIRRED REAC
FOLLOWED BY PLUG FL
100 2C
OR IGNITION ZONE
W REACTOR
0 3C
P=10
ATM
ATM
0
o
o
TIME(MSEC)
Figure 4-8. NO from Lean Premixed Combustion - Adiabatic
Gas Turbine Combustor - With Fuel Ammonia
NOTE: * = 0.45; Tflame = 2800°F
101
-------
o
ro
SECONDARY AIR (1000°F)
LOW BTU FUEL GAS, 1500°F, 4000 PPM
P = 10 ATM
AIR, 1000°F
STIRRED
REACTOR
IGNITION
STAGE
T = 1 MSEC
PRIMARY STAGE
VARIABLES: 0, T,
HEAT TRANSFER
HEAT TRANSFER
1 II II 1 1
SECONDARY STAGE
VARIABLES: SECONDARY AND
TERTIARY AIR STAGING
DISTRIBUTIONS
0 =• 0.45
T = 2800°F
TERTIARY AIR
Figure 4-9. Staged Adiabatic Gas Turbine Combustor - MKAP Analog Schematic
-------
numerical results presented in this section. First, in the moderately rich
primary combustion process a "prompt" transfer of fuel N to N2 occurs within
the first millisecond which can account for two-thirds of the effectiveness
of the primary zone. This prompt reduction of fuel N is accompanied by a
significant generation of superequilibrium HCN and NO. Ideally then the
function of the subsequent long primary hold-up period is to drive all of
the remaining N-containing constituents to their low equilibrium values.
Under very rich conditions this goal is thwarted by further synthesis of
superequilibrium HCN followed by a freezing-out due to the slow HCN destruc-
tion process under rich conditions.
The second result is associated with the long primary hold-up zone.
After a period of 100-200 msec the composition becomes independent of the
initial level of NH3 for a broad range of primary zone stoichiometry ( = 0.8
- 1.7). On the lean side this is merely a statement that equilibrium has been
reached. At intermediate equivalence ratios (=1.3) it is indicative of reach-
ing a universal approach-to-equilibrium condition. Under richer conditions
($=1.67) it represents the achievement of a common approach-to-equilibrium for
NO and NH3< Further, under these rich conditions (=1.67) the HCN nonequili-
brium synthesis and freezing processes are dominated by Fenimore-type reactions,
and hence, are independent of the initial NH3 levels.
Before proceeding with a presentation of the primary zone numerical
calculations it is appropriate to suggest possible chemical mechanisms which
dominate the early NO formation period. Figures 4-10 and 4-11 present nitro-
gen balance diagrams from 2-msec residence time stirred reactor calculations
for methane-air combustion at an equivalence ratio of 1.33. They are perti-
nent to this discussion because they are doped in one case with a high initial
NO level (1300 ppm) and in the other with a high initial NH3 level (1300 ppm),
and the overall results of these calculations'compare well with stirred reactor
data obtained at Exxon (see Appendix C). The circles represent the various
significant nitrogen species which exchange N with one another via the reaction
paths indicated by connecting lines. The molecular partner in the forward
direction is indicated on the line and the direction is signified by the arrow-
head. The numbers within the circles represent the net production rate of the
N-containing molecule. All rates have the units of ppm per msec. Finally,
103
-------
Figure 4-10. NO Formation and Destruction in Well-Stirred CH. -Air Reactor
with 1300 ppm NO Addition
NOTE: Equivalence Ratio = 1.33; Residence Time = 2 msec
-------
Figure 4-11. NO Formation and Destruction in Well-Stirred Reactor with
1300 ppm NH3 Addition
NOTE: Equivalence Ratio = 1.33; Residence Time = 2 msec
-------
the concentrations of the various nitrogen molecules are given by a mass
balance across the reactor.
where [cj is the concentration of the ith species in (ppm), [Cijinit is
the concentration entering the stirred reactor, T is the residence time in
msec, and Fc'^l is the production rate of the ith species per unit mass in
the reactor. For example, in Figure 4-10 [NO] = -325, T = 2, [N0]-jnit = 1300.
Hence, [NO] = 650 ppm which corresponds to 50 percent retention.
Figure 4-10 is for the case of initial NO addition. It can be seen that
N atom production and destruction plays a central role in the prompt decay of
NO. The Myerson reaction
NO + CH -> CHO + N (1)
appears to be responsible for direct reduction of NO as well as producing N
atoms. These NO atoms in turn enter into both: production of NO, primarily
via the extended Zeldovich reaction
N + OH -> NO + H, (2)
and reduction of NO via the reverse Zeldovich reaction
NO + N -»• N2 + 0 (3)
The destruction of NO by N outweighs production of NO by N roughly in the
ratio 4:3. N atoms are also produced by the Fenimore reaction
CH + N2 -> HCN + N (4)
106
-------
Although this promotes reduction of NO it is also the primary producer of
HCN. How effectively HCN is subsequently removed in the.hold-up zone (if
at all) is unclear. The other Fenimore reaction
CH2 + N2 -v HCN + NH
is substantially less effective in this instance as a producer of HCN. The
NH so produced has little overall effect on NO production or destruction as
can be seen from the balance diagram. Sufficient N atoms are formed from
NH + H •*• N + H2 (6)
to provide an NO sink which compensates for the small NO production through
NH + OH + NO + H2 (7)
Approximately 25 percent of the HCN production is through a direct conversion
of NO via the Myerson reaction
NO + CH -»• HCN + 0.
(8)
This, however, accounts for less than 10 percent of the NO destruction. Finally,
an active loop exists between NO and HNO. Although this loop has only a small
(less than 10 percent) direct effect on NO depletion, it does effect the 0, OH
and H radical population, and hence, exerts an indirect effect.
With this background on prompt behavior with large initial NO levels we
now examine Figure 4-11 which is more representative of the situation at hand.
In this instance the stirred reactor is doped with 1300 ppm of ammonia. The
process is essentially a cascading down from NH3 to N as shown. This large
N atom source in the absence of high initial NO concentrations provides a net
source of NO primarily from the extended Zeldovich reaction (2). The pro-
duction of NO by N outweighs the destruction of NO by N in the ratio 1.6:1.
NO is depleted by the Myerson reaction (1) and the reverse Zeldovich reaction
(3) which leads to N2 and is the final link in the transfer of nitrogen from
107
-------
NH? to N2. The Fenimore reactions (4) and (5) act to provide a net
production of NO through the generation of N atoms. These reactions are
again a source of HCN. With the reaction scheme used in this study the
linkage between NH3 and HCN is through NO generation from NH3 followed by
the Myerson reaction (8). In the case at hand this linkage is weak with
only a small portion of the HCN so generated. It is possible, however, that
under different equivalence ratios the NH3-HCN linkage could be stronger.
The NO-HNO loop again appears primarily as a factor in the 0, OH, and H
radical balance. The small time constant of the initial or "prompt" NH3
conversion is clearly associated with the rapid methane combustion process
producing hydrocarbon fragments as intermediaries and with the rapid decom-
position of NH3 yielding N atoms under rich conditions.
The effect of residence time and degree of backmixing on "prompt"
behavior has not been thoroughly explored. In general it was found that
under relatively lean conditions there is no discernable difference between
a longer residence time stirred reactor and a stirred reactor followed by a
plug flow section with the same total residence time. However, under richer
conditions (
-------
• Initial fuel ammonia content, (NH-L
• Air nitrogen replacement by Argon
The discussion is necessarily speculative due to considerable uncertainties
in the basic chemical kinetic set and rate constants employed in the
calculations.
Equivalence Ratio and Plug Flow Residence Time
Figures 4-12 to 4-15 show the nitrogen species history for an adiabatic
plug flow rich primary at various equivalence ratios (2.0, 1.67, 1.33, 1.1).
Essentially all of the "prompt" behavior has been completed within the stirred
reactor ignitor and the subsequent behavior is slowly varying as seen in these
figures. The stirred reactor output is plotted along the ordinate. The solid
curves are for the fuel NH3 concentration set equal to zero. Figure 4-16 is a
similar calculation under lean conditions (=0.9) included here for comparison.
In examining the primary zone behavior for various equivalence ratios
there are several aspects to scrutinize. First, how effective was the prompt
reduction of nitrogen compounds (that is, NO precursors which will convert to
NO in the secondary stage}? What is the distribution of these species? How
effective is the rich hold-up zone in promoting equilibration of these species?
To what degree does nonequilibrium generation of HCN take place? Under what
conditions does the HCN level freeze? Does the system reach a condition which
is independent of the initial fuel nitrogen level?
Figure 4-12 presents the results for the richest case (4>=2.0) examined.
More than 80 percent of the fuel N is still present as NO precursors following
the stirred reactor. The system is too rich and most of the NH3 does not
decompose in the stirred reactor. The NH3 which does decompose, converts
primarily to N2 with approximately 150 ppm of NO also appearing. With time
the NH3 and NO slowly decay towards their equilibrium values. Although very
little HCN appears initially, nonequilibrium synthesis of HCN proceeds through-
out the entire period apparently due to nonequilibrium hydrocarbon fragments
participating in the Fenimore-type reactions. The indirect conversion of NH
to HCN through the generation of NO and the subsequent conversion of NO to
109
-------
10,000
1000-
O.
Q.
100
10
HCN
10
TIME (MSEC)
100
1000
Figure 4-12. Nitrogen Species from Premixed Rich Primary Zone — Adiabatic
Gas Turbine Combustor
NOTE: 4> = 2.0; Toutiet = 2955°F; Coordinate values from stirred reactor
zone TSR = 1 msec; LOG/LOG coordinates; NHo = 4000 ppm in fuel
-------
10,000
1000
Q.
CL
= 2230 ppm (-4000 ppm in fuel)
(NH3)0 = 0.0
1000
TIME (MSEC)
Figure 4-13. Nitrogen Species from Premixed Rich Primary Zone — Adiabatic Gas
Turbine Combustor
NOTE: = 1.67; Toutlet = 3167°F: Ordinate values from reactor ignition zone
T = 1.0 msec; LOG/LOG coordinates
-------
10,000
ro
•v EQUILIBRIUM VALUE
1000
TIME (MSEC)
Figure 4-14. Nitrogen Species from Premixed Rich Primary Zone - Adiabatic
Gas Turbine Combustor
NOTE: 4> = 1.33; Tout-|et = 3475°F; Ordinate values from stirred reactor
ignition zone TSR = 1.0 msec; (NHaJo = 2005 ppm (4000 ppm in fuel);
LOG/LOG coordinates
-------
CO
10,000
1000
Q.
O.
o 100
i
«_>
o
10
NO
s HCN
N
X
NO
10
100
1000
TIME (MSEC)
Figure 4-15. Nitrogen Species from Rich Primary Zone - Adiabatic Gas Turbine Combustor
NOTE: $ = 1.1; Toutlet = 3650°F; Ordinate values from stirred reactor ignition zone
= 1.0 msec; LOG/LOG coordinates; NH3 = 4000 ppm in fuel
-------
10,000
1,000
u
o
100
10
(NH3)Q = 1706 ppm (-4000 ppm in fuel)
(NH3)Q = 0
10
TIME (MSEC)
100
1000
Figure 4-16. Nitrogen Species from Premixed Lean Primary Zone - Adiabatic Gas
Turbine Combustor
NOTE
: = 0.9; Toytiet = 3600°F; Ordinate values from stirred reactor ignition
zone TSR =1.6 msec; LOG/LOG coordinates
-------
HCN through the Myerson reaction is also a possibility although the NO levels,
and hence, the reaction rates are low. Again, these comments must be taken as
speculative since the HCN production and destruction reactions are poorly
understood under rich conditions.
Figure 4-13 shows the results for the*=1.67 case. Approximately 50 per-
cent of the fuel NH3 has converted to N2 within the stirred reactor. Appre-
ciable "prompt" NO and HCN are generated and NO decays slowly toward equili-
brium. Superequilibrium HCN synthesis continues during the first 100 ms of
the primary zone. After 100 msec the HCN concentration becomes frozen, indi-
cating that the decomposition reactions are either very slow or improperly
modeled.
Initial Fuel Ammonia Content
Figure 4-13 also shows the behavior for the 4>=1.67 case in the absence
of any fuel ammonia. All of the NO precursors, HCN, NH3> and NO itself
experience prompt synthesis in the stirred reactor which is continued during
the first 30 msec of the plug flow. The HCN generation is identical to that
with fuel nitrogen present indicating that the mechanism is entirely Fenimore.
The NO generation is apparently also due to the Fenimore mechanism of N atom
generation followed by the modified Zeldovich reactions. When the N atom
production ceases due to a decay in hydrocarbon radicals, the HCN freezes
and the NO commences its decay to equilibrium. The NH3 synthesis again
apparently rests on the N atom generation. It is speculated that the dominant
generation mechanism is
N + OH -»• NH + 0
NH + OH •»• NH2 + 0
NH2 + OH -v NH3 + 0
Again, when the N atom production ceases the NH3 starts to decay towards equili-
brium. Surprisingly this final reduction of NO and NH3 towards equilibrium
becomes identical with that for the case with fuel nitrogen present.
The results for 4>=1.33 are shown in Figure 4-14. The fuel NH3 essentially
vanishes completely within the stirred reactor with two-thirds of the N appear-
ing as N2. Of the remaining portion most is manifested as NO with about 100 ppm
115
-------
of HCN. The NO and HCN decay very slowly towards their equilibrium states
with the NO reaching that value (90 ppm) in approximately 200 msec. At this
point 60 ppm of HCN still remains yielding a total amount of NO precursors
of 8 percent of the initial level of NH3- Such a result certainly must be
considered encouraging provided a practical combustor design can be evolved
with primary zone residence times in excess of 100 msec.
In Figure 4-15 the results for 4>=1.1 are presented. This case is clearly
too lean with only about 50 percent of the fuel nitrogen promptly converted to
H9. The remaining portion is primarily converted to NO within the stirred
reactor at a level of 900 ppm which is approximately the equilibrium value.
50 ppm of prompt HCN decays to zero in approximately 100 msec probably through
reactions with OH to form CN. In comparison with the undoped case the HCN
decay is quite slow. This may be a consequence of a HCN production contri-
bution through the Myerson reaction due to the high initial NO levels in the
case with fuel nitrogen present. The influence of initial level of NH3 is
felt only for the first 100 msec after which the system is equilibrated. The
results for 4>=0.9 are included in Figure 4-16 for completeness. They are
similar in nature to the for various residence times. On the lean
side, thermal NO production dominates with peak values occurring at approxi-
mately 4>=0.9. On the rich side the residual nitrogen fraction has a minimum
at an equivalence ratio around 1.4 and the magnitude of this fraction decreases
with time, reaching a value of 0.10 at t=200 msec.
Air Nitrogen Replaced by Argon
In order to assess the importance of N2 content, a set of calculations
was conducted with the nitrogen content of the air replaced by an equal mole
fraction of argon. The mixture (02-Ar) temperature was adjusted to yield the
116
-------
Increasing Residence
Time (msec)
t = 0 msec
plug flow
residence
time
°-8 1.0 1.2
Equivalence Ratio,
Figure 4-17.
N°TE:
Sum of Residual Nitrogen Species from Prefixed Plug
Flow Primary - Adi aba tic Gas Turbine Combustor
.:- .«vl =.1500°F; P = 10 atm; TSR = 1 msec
; NH3 = 4000 ppm in fuel K
117
-------
same flame temperature as the corresponding low Btu fuel gas/air mixture.
Considering the N2 content of the low Btu fuel gas, at <|>=1 the (02-Ar) system
replaces about 70 percent of the total N2 with Ar and at =0.5 it replaces
about 80 percent.
Figure 4-18 shows the results plotted in the same format as Figure 4-17.
On the lean side there is a reduction of thermal NO generation in proportion
to the N2 concentrations as expected. On the rich side the Fenimore reaction
rates are reduced in proportion to the N2 reduction, and hence, HCN, NH3 and
NO generations are all reduced. For example, at =1.4 and t=50 msec, N2 is
reduced by approximately 60 percent and the residual nitrogen fraction is cut
by more than 50 percent, showing the significance of the N2 content on the
rich side behavior. At larger residence times this effect is less pronounced.
Summary of Results
Figure 19 presents the Premixed Primary Reactor data for LBG fuel gas and
air in the form of an "effective NH3 conversion ratio" which is defined by:
j[NO] + [NH3] +[HCH][ - | [NO] + [NH3] + [HCN]f
effective NH. conversion ratio s —
(MHO
(HHJ
3'0
3'0
That is, it is the excess nitrogen-containing molecules, over and above those
existing in the absence of NH3 doping, normalized by the initial (NH3)Q. This
ratio is a measure of additional N-containing molecules due to the presence of
initial NH3. If it equals zero it signifies that there is no effect due to
the initial doping, i.e., the system behavior has become independent of the
NH3 initial conditions. Figure 4-19 shows that these conditions occur for
t=s200 msec and 0.8 < * < 1.7. For * < 1.1 this is due to the reactor achiev-
ing a near-equilibrium state. At =1.33 it is due to the independence of HCN
production on the presence of NH3 and to the equilibration of NO. At =1.67
Figure 4-13 previously demonstrated that the concentrations for the ammonia-
doped and undoped cases merge at t=200 msec. Again the HCN production is
independent of (NH3)Q. In the absence of initial ammonia, NO is produced via
N generation by the Fenimore reactions. NH3 is synthesized in a similar manner
in the absence of initial doping by a chain which is initiated by N and NH
118
-------
Increasing Residence
Time (msec)
t = 0 ms
plug flow
residence
time
t = 50 ms
t = 200 ms
t = 500 ms
0.0
0.4
1.0 1.2 1.4
Equivalence Ratio,
Figure 4-18. Sum of Residual Nitrogen Species from Premixed
Plug Flow Primary - Adiabatic Gas Turbine
Combustor - N£ in Air Replaced by Argon
NOTE:
such that flame temperature is
as for Air; Tfuei - 150QOF; P = 10 atm; TSR = 1 ms-
(NH3)fuel - 4000 ppm; [AR]/[02] • 3.79 (same as for air
119
-------
1.0
0.8
0.6
I/I
Ol
o
QJ
> 0.4
-t->
o
O>
0.2
0.5
Increasing
Residence
Time (msec)
1.0 1-5
Equivalence Ratio, 4>
= 0 msec
plug flow
residence
time
50
Figure 4-19. Effective (NHs) Conversion Ratio for Premixed
Rich Primary Reactor - Adiabatic Gas Turbine
Combustor
NOTE: P = 10 atm; T$R = 1 msec (ignition); NHs = 4000 ppm
in fuel; Tair = 1000°F; Tfuel = 150QOF
120
-------
production through the Fenimore reactions. Both NO and NH- so generated
eventually reach the levels associated with NHg doping, and therefore, follow
the same decay behavior.
In summary, a baseline rich primary reactor burning fuel containing
4000 ppm NH3 with 1.33 < 4> < 1.45 designed for a residence time on the order
of 200 msec will have a residual nitrogen species concentration of less than
200 ppm. This appears to be near the optimum design point without requiring
excessive residence times.
4.3.3.3 Deviations from Baseline Primary Reactor Specifications
Now that the factors influencing N0x production in the baseline primary
reactor have been discussed, the following additional variables will be
addressed:
• Heat transfer from plug flow reactor
• Stirred reactor residence time
• Pressure
t Fuel composition - methane content
Heat Transfer Effects
Heat transfer effects were examined by comparing the nitrogen-bearing
species time histories for the adiabatic case with those for an instantaneous
heat removal equivalent to a temperature drop of 267°F. The effect of heat
extraction from the rich primary zone depends on when the heat transfer takes
place. Figure 4-20 shows the results for the c|>=i.67 case with early heat
removal immediately following the stirred reactor zone (1.0 ms). The effect
is to reduce all of the reaction rates which results initially in a retardation
of the decay of the cumulative NO precursors. After long periods, however, the
effect of cooling appears to be small.. Figure 4-21 indicates that taking the
same amount of heat out late in the process (t=500 ms) has essentially no effect,
Figure 4-22 shows the effect of early cooling (after 1.0 ms) for 4>=1.33. The
heat transfer here was increased to produce a 475°F temperature reduction to
further emphasize the effects. In this case the reduction in temperature is
sufficiently great to freeze the HCN at its initial level and to retard the
121
-------
10,000
ro
Q.
Q.
(_)
o
1000
Adiabatic, Toutlet = 3167°F
Heat Loss, Toutlet = 2900°F
Heat Removed at \ = 1 msec
— HCN
10
100
TIME (MSEC)
1000
Figure 4-20. Nitrogen Species from Rich Primary Zone with Early Heat Removal —
Adiabatic Gas Turbine Combustor
NOTE: = 1.67; Ordinate values from stirred reactor ignition zone TCD = 1 msec;
[HCN]eq < 5 ppm; (NH3)Q = 2230 ppm (-4000 ppm in fuel); LOG/LOG coordinates
-------
ro
200
Q.
Q_
100
Adiabatic Toutlet =. 31670F
Heat Loss, Toutiet = 290QOF
Heat removed after 500 msec
Heat
Removed
Here
HCN
HCN
10 100
TIME AFTER HEAT REMOVAL (MSEC)
Figure 4-21. Nitrogen Species from Rich Primary Zone with Late Heat Removal
— Adiabatic Gas Turbine Combustor
NOTE: = 1.67; Ordinate values from uncooled plug flow zone, TSR = 500 ms;
(NH3)0 = 2230 ppm (-4000 ppm in fuel); (HCN)eq = 5 ppm; LOG/LOG coordinates
1000
-------
10,000
ro
-P.
1,000
Q.
a.
o
o
o
Adiabatic, Tout]et = 34750F
Heat Loss, Toutiet = 3000°F
Heat removed at 1 msec
1000
TIME (MSEC)
Figure 4-22. Nitrogen Species from Rich Primary Zone with Early Heat
Removal — Adiabatic Gas Turbine Combustor
NOTE:
-------
decay of NO. The combined NO and HCN remain high until t=500 ms when the
trend is reversed due to the NO seeking a lower equilibrium value. Fig-
ure 4-23 indicates that cooling late in the process (t=500 ms) for <|>=l.33
yields a lower combined NO and HCN due to the NO seeking a lower equilibrium
value. Such late cooling is, of course, unrealistic and is included here
only to provide insight into primary zone behavior.
Stirred Reactor Residence Time Effects
Figure 4-24 presents an extreme case in which the entire =0.9) is examined. Stirred reactors with residence times of 0.5,
1.0, and 5.0 msec were followed by plug flow reactors. The concentrations,
when plotted again total stirred and plug flow reactor residence time, show
no effect of this variation.
Pressure Effects
Figure 4-26 illustrates the effect of pressure reduction for the 4>=1.33
case. An extreme example was chosen (P=l atm) in order to emphasize the
effect. In comparison with the 10 atmosphere case (also plotted), the NO
level increases and the HCN level decreases as pressure is reduced. The
increase in NO level is consistent with the increase of equilibrium NO con-
centrations with reduction in pressure under rich conditions due to the
increase in equilibrium oxygen concentration as the pressure drops. The
Fenimore HCN production rate is expected to be roughly proportional to
pressure.
Figure 4-27 shows similar results for reduced pressure at 4>=0.9. For
lean conditions there is little pressure effect since equilibrium NO con-
centrations are pressure-independent on the lean side.
125
-------
ro
120
100
80
0.
§
I—t
I
z
UJ
i
60
40
20
Adiabatlc, Toutlet = 3475°r
Heat removed AT = 475°F
teat removed after 500 msec
NO
HCN
NO
10 100
TIME AFTER HEAT REMOVAL (MSEC)
1000
Figure 4-23. Nitrogen Species from Rich Primary Zone with Late Heat Removal -
Adiabatic Gas Turbine Combustor
NOTE: = 1.33; Ordinate values from uncooled plug flow zone, T$R = 500 ms;
= 2005 ppm ( 4000 ppm in fuel); Log time coordinate; (HCN)eq<5 ppm
-------
10,000
ro
1,000
Q.
Q.
§
*—t
i
o
g
Line data are for plug flow
Point data are for 500 msec stirred reactor
TIME (MSEC)
1000
Figure 4-24. Nitrogen Species from Rich Primary Zone, Stirred/Plug Flow Comparison -
Adiabatic Gas Turbine Combustor
NOTE: = 1.33; Toutlet " 3475°F; Ordinate values from stirred reactor ignition zone
TSR = 1 msec; (NH3)0 = 2005 ppm (4000 ppm in fuel); No heat transfer; LOG/LOG coordinates
-------
INS
00
10,000
1000
a.
a.
LU
<_>
1
100
10
(NH3)Q = 1706 ppm (4000 ppm In fuel)
(NH3)Q - 0
a~T<.D = 1/2
OK
A ~TSR= l
O ~ TSB = 5
Stirred reactor resldenct time
followed by plug flow
10
100
1000
TIME (MSEC)
Figure 4-25. Nitrogen Species from Lean Primary Zone with Varying Stirred Reactor
Residence Time - Adiabatic Gas Turbine Combustor
NOTE: 4> = 0.9; Toutlet = 3600°F; LOG/LOG Coordinates; (N0)eq = 3680 ppm
-------
r\>
to
10,000
CL.
O.
O
§
o
1,000
100
10
P =10.0 atm
P = 1.0 atm
NO
HCN
10
100
1000
TIME (MSEC)
Figure 4-26. Nitrogen Species from Prenrixed Rich Primary Zone - Adiabatlc
Gas Turbine Combustor — Effect of Pressure
NOTE: * » 1.33; Toutlet = 3475°; Ordinate values from stirred reactor
ignition zone TSR = 1 msec; (NHaJo = 2005 ppm (4000 ppm in fuel);
LOG/LOG coordinates
-------
CO
o
10,000
.J!!L ——
1,000
Q.
Q.
100
10
HCN
P = 10.0 atm
^ p = 1.0 atm
10
100
TIME (MSEC)
Figure 4-27. Nitrogen Species from Premixed Lean Primary Zone — Adiabatic
Gas Turbine Combustor
NOTE: 0 = 0.9; Tout]et = 3600°F; Ordinate values from stirred reactor
ignition zone TSR = 1 msec; (Nf^)* = 1706 ppm (4000 ppm in fuel);
LOG/LOG coordinates
1000
-------
Fuel Composition Effect
The effect of fuel methane content was simulated by replacing the
methane with CO and \\2 such that the overall C/H ratio, equivalence ratio,
and enthalpy remained unchanged. This is equivalent to allowing the hypo-
thetical reaction
CH4 + 1/2 02 -*• CO + 2H2
to proceed instantaneously. Figure 4-28 compares the results for $=1.33
with and without methane. The absence of hydrocarbon fragments eliminates
the Myerson and Fenimore-type reactions, and hence, HCN is absent in Fig-
ure 4-28. The NO level is only slightly affected showing a somewhat higher
level in the absence of the Myerson decay mechanism and the Fenimore N atom
generation.
Figure 4-29 shows the results for $=0.9. Again, there is essentially
no effect on NO as expected.
4.3.3.4 Secondary Stage Burnout and Dilution
The function of the second stage combustor is to complete the energy
release and dilute the combustion products to a suitable turbine inlet tem-
perature. The turbine inlet temperature for the adiabatic gas turbine com-
bustor is 2800°F.
The challenge for the second stage design is to minimize the production
of thermal NO associated with finite rate mixing coincident with the minimiza-
tion of the conversion of those nitrogen specie leaving the first stage to NO.
A secondary stage following a nonoptimum primary operating at =2.0 with high
HCN, NH3, and NO is first examined to illustrate the essential features of the
secondary combustor design problem. The optimum configuration with a primary
zone operating at $=1.33 is then addressed.
Figure 4-30 shows the MKAP analog schematic for a $=2.0 nonoptimum primary
stage with two possible secondary stage configurations. The upper schematic
shows a combustor with secondary air addition distributed axially along the
combustor. The distribution of equivalence ratio is shown below the figure
under the assumption of rapid fuel/air mixing upon air admission. Note that
131
-------
10,000
CO
ro
1000
Q.
Q_
No fuel methane (equivalent to CH. + % Q -*• CO + 2H?)
Base fuel (from Figure 4-12)
1000
TIME (MSEC)
Figure 4-28. Nitrogen Species from Premixed Rich Combustion -Adiabatic Gas Turbine
Combustor — Effect of Fuel Methane Content
NOTE:
= 1.33; Toutjet = 3475°F; Ordinate values from stirred reactor ignition zone
R = 1.0 msec; p = 10 atm; (^)Q = 2005 ppm (4000 ppm in fuel); LOG/LOG Coordinates
-------
CO
CO
10,000
1,000
D.
Q.
100
10
~ — - _ _ _HCN
NO
NO
No fuel methane (equivalent to CH4 + -'s 02 -»• CO + 2H2)
Base fuel (from Figure 4-14)
10
TIME (MSEC)
100
1000
Figure 4-29.
Nitrogen Species from Premixed Lean Combustion -Adiabatic
Gas Turbine Combustor - Effect of Fuel Methane Content
NOTE: 4> = 0.9; Toutiet = 3600 F; Ordinate values from stirred reactor
ignition zone T$R = 1 msec; (NHaJo = 1706 ppm (4000 ppm in fuel);
p = 10 atm; LOG/LOG coordinates
-------
AIR INTO FUEL
u>
LBG FUEL GAS, 1500°F,
4000 PPM NH
STIRRED
REACTOR
IGNITION
ZONE
P = 10 ATM
AIR, 1000°F
RICH PRIMARY
: 2.0, T = 500 ms
AIR, 1000°F
0.45
T = 28000F
Figure 4-30.
Staged Adiaba tic Gas Turbine Combustor - MKAP Analog Schematics for
Alternative Secondary Stage Arrangements
-------
it passes from an initial rich value of 2.0 through stoichiometric conditions
to a final lean condition of 0.45. If the time constant of this delayed
mixing process is substantial, considerable thermal NO will be produced during
the temperature excursion near =1. To emphasize this behavior a mixing time
constant of 50 msec was chosen for the computations. The lower schematic
represents the opposite mixing limit in which the fuel is added to the secondary
combustion air in a distributed fashion. Again, under the assumption of rapid
mixing upon fuel admission the distribution of equivalence ratio goes from zero
to 0.45 without experiencing a stoichiometric state. Clearly under these
idealized conditions the thermal NO production will be small.
Two additional cases also deserve attention. Instantaneous mixing in
which all of the primary products and air mix rapidly at the inlet to the
secondary zone, and a simulated diffusion flame. Figure 4-31 shows the MKAP
model of the diffusion flame. The fuel is introduced to the secondary air
at three discrete axial stations and at each station the diffusion flame
behavior is simulated by first bringing the fuel and air into contact in a
parallel cluster of three stirred reactors (indicated schematically as a
single stirred reactor at each entry station). The stirred reactors have a
3.0 msec residence time and a distribution of equivalence ratios of 0.9, 1.0
and 1.1. The effluent from these flame simulations is then returned to the
main flow. The choice of residence time is arbitrary and implicitly reflects
-fuel nozzle scale, spacing, and turbulence intensity. The 3 msec value used
here is simply for illustrative purposes.
Figure 4-32 shows the secondary combustor NO and temperature distribution
for the four cases. The long dash curves represent the instantaneously mixed.
situation. In that case NO rises rapidly in the first few msec as the primary
zone NO precursors are converted to NO. The NO level is lower than the pri-
mary exit only due to distribution. Following this "prompt" secondary conver-
sion the NO increases very slowly due to thermal fixation since the temperature
does not exceed 2800°F. The solid curves in Figure 4-32 correspond to the upper
schematic of Figure 4-30. A temperature excursion to 3700PF occurs and the NO
level goes over 1000 ppm before decreasing back to 850 ppm by fi.nal dilution
Following this there is a gradual rise in NO due to thermal fixation.
135
-------
SCHEMATIC REPRESENTATION OF FUEL NOZZLE ARRAY
AIR 1000°F
o£= Or=f
EQUIVALENT MODULAR ARRAY
= 2.0
PARTIAL PRODUCTS OF
COMBUSTION FROM
PRIMARY
T = 3000°F
' /f
I
i i
i
i
•X
AIR 1000°F
STIRRED REACTORS
3.0 MSEC RESIDENCE TIME
PLUG FLOW REACTOR
2800°F
0 = 0.45
Figure 4-31.
Staged Adiabatic Gas Turbine Combustor
MKAP Analog Schematic for Simulated
Diffusion Flame
136
-------
1000-
800-
D-
O-
o 600-
o
o
Air addition over
50 msec
Simulated diffusion
flame
Instantaneous Mixing
Products added over
50 msec
•4000
-3000
•2000
-1000
Time After Initiation of
Secondary A1r Injection (msec)
Figure 4-32. NO Concentration and Temperature for
Configurations
Gas Turbine Corabustor
NOTE: Primary zone exit conditions from Figure 4-12
See Figure 4-30 for schematic.
137
-------
The dot-dash curves in Figure 4-32 correspond to the lower schematic
of Figure 4-30. Since fuel is added to the secondary air over a 50 msec
period the mixture increases in temperature until autoignition of the
remaining fuel occurs at approximately 40 msec. There is no overshoot in
temperature and the conversion of the primary NO precursors is followed by
a slow thermal fixation as in the instantaneously mixed case.
The short dashed curves in Figure 4-32 correspond to the finite rate
diffusion flame shown in Figure 4-31.
The temperature curve represents the mean axial temperature and does
not reflect the local temperature excursion within the diffusion flames.
These temperature and stoichiometry distributions about the mean give rise
to the increased NO as indicated. The peak flame temperature predicted by
this was 3700°F and produced an NO level twice that of the idealized rapid
mixing case.
Finally, Figure 4-33 provides an indication of the lower bound NO emis-
sion attainable if an optimum =1.33 primary feeds a rapid mixing secondary.
The results imply that NO levels under 100 ppm may be feasible. It is
probably prudent, however, to assume that even under optimum design condi-
tions finite rate mixing will contribute an additional 80 ppm of thermal NO
(50 percent of the diffusion flame NO calculated in the previous example).
Thus NO levels of 150 ppm can reasonably be expected from a well-designed
LBG-COGAS system firing a fuel with high ammonia content in an adiabatic gas
turbine combustor at an overall equivalence ratio 4>=0.45.
4.4 Supercharged Boiler Results
In order to establish the chemical constraints on the control of N0x
formation in supercharged boilers a set of numerical computations was under-
taken using the MKAP methodology following generally the same lines as the
previous adiabatic gas turbine combustor analysis. Again, the objective was
to derive estimates of the lower achievable limits for supercharged boiler
NO emission as dictated by chemical considerations by exploring the dependence
of NO emissions on specific design features.
A
138
-------
£
Q.
IX.
110
100
90
80
70
60
50-
40
30
20
10
•
Primary: 1.0 msec st
Plug from E
t = 1 33
Toutlet = '
,,_ Secondary Instantnr
4> final =
Toutlet =
+ ,
0.1 1
1rred reactor Ignition zone
>00 msec
1475 F
eous mixing
0.45,,
2800°F
Follows Simple
Zeldovich
TIME (MSEC) 1(
S
^^
j
-*/
•
3 101
AFTER INSTANTANEOUS SECONDARY AIR INJECTION
Figure 4-33.
NO Concentration for Optimum Staged Adiabatic
Gas Turbine Combustor
139
-------
As outlined in Section 4.2, the baseline supercharged boiler system is
a low excess air (5 percent) supercharged boiler firing a low temperature
(150°F) fuel gas typical of the product from a coal-fired LEG gasifier with
interceding and low temperature gas cleanup. The gas composition is identi-
cal to that used in the adiabatic gas turbine combustor study except that the
ammonia content is taken as 500 ppm to reflect a partial ammonia cleanup in
the low temperature sulfur removal process. The boiler operates at 10 atmos-
pheres pressure and feeds a gas turbine at 2000°F inlet temperature. Combus-
tion air is supplied at 600 F.
4.4.1 General Flame Characteristics
Much of the general information regarding LBG flame chemistry presented
in Section 4.3 is directly applicable to combustion in supercharged boilers
and will not be repeated in this section. The principal difference in the
two systems lies in the lower fuel nitrogen and lower flame temperature for
the supercharged boiler fuel and in the nonadiabatic low excess air secondary
stage combustor associated with the boiler.
Figures 4-34 and 4-35 shows the adiabatic flame temperature and equili-
brium NO as functions of equivalence ratio and fuel temperature. The maximum
adiabatic flame temperature 3160°F is 500°F below the gas turbine case. At
5 percent excess air the adiabatic equilibrium NO level is approximately
1000 ppm. Note how sharply the equilibrium values fall off as 4>=1 is
approached. The peak values are located on the lean side at approximately
=0.8.
4.4.2 Premixed Lean Combustion
Figure 4-36 presents the results for a premixed lean combustor operating
at 5 percent excess air. It provides an "uncontrolled" estimate of NO emis-
sion levels against which to measure the success of the staged combustion sys-
tem discussed in the next section.
A 1.0 msec stirred reactor provides ignition and "prompt" conversion of
essentially all of the fuel nitrogen to NO. The stirred reactor is followed
by a plug flow heat exchange section. The heat transfer distribution is shown
140
-------
4000
3500
3000
3500
UJ
O.
2000
1500-
600°F Fuel Temp.
150°F Fuel Temp.
NOTE: 1. LBG Composition from
Table 4-2
2. Air Inlet T = 600°F
1000
0.0
Figure 4-34.
0.5 1.0
EQUIVALENCE RATIO
for LBG and Coition Air for
141
-------
3000
2500
2000
OL
CO
1500
o
o
o 1000
500
0
Fuel Temp = 600°F
150°F
NOTE: 1. LBG Composition
from Table 4-2
2. Air Inlet T = 600°F
I i i i 1 1 1 L
EQUIVALENCE RATIO
Figure 4-35. Equilibrium NO Concentrations - Supercharged Boiler
142
-------
3500
CO
Plug Flow 1500 msec
150
1000
1000
1500
TIME (MSEC)
Figure 4-36. NO Concentration and Temperature for Premixed Lean Combustion
— Supercharged Boiler
NOTE: 5% excess air; (NH3)Q = 500 ppm in fuel; Heat removal as shown above
-------
in Figure 4-36 and has a 2:1 peak-to-minimum ratio with the peak occurring
at the midpoint of the boiler. For the purposes of this computation an
arbitrarily long residence time was assumed (in excess of 1 sec). For these
conditions the 2800°F temperature cutoff for NOV kinetics occurred at one-
A
third the length down the boiler and 345 ppm of NO resulted. If the boiler
residence time had been reduced by a factor of two the NO level would have
been 330 indicating the insensitivity of the calculation to an increase in
overall residence time once the gas temperature has dropped below 2800°F.
4.4.3 Staged Combustion
The objective of staging the supercharged boiler is to reduce the fuel
nitrogen conversion to well below 200 ppm (based on 5 percent excess air) and
to reduce the thermal NO in the heat exchange section to well below 150 ppm
using the same control strategy as the adiabatic gas turbine combustor in
Section 4.3. Figure 4-37 shows the MKAP schematic for the staged super-
charged boiler. The additional degree of freedom present here is the ability
to simultaneously stage the secondary stage heat release and heat transfer
to minimize long periods at high temperature.
4.4.3.1 Fuel-Rich Primary Stage - Baseline Case
The following discussion addresses the NO kinetics for this baseline
A
reactor as the following parameters are varied:
• Equivalence ratio, 4>
• Adiabatic plug flow residence time, T
t Initial fuel ammonia content, (NHgJg
Equivalence Ratio and Plug Flow Residence Time
Figures 4-38 through 4-41 show the nitrogen species and temperature
histories for an adiabatic plug flow rich primary operating at various
equivalence ratios (1.5, 1.4, 1.33, 1.15). The ignition in all cases was
provided by a 2 msec stirred reactor and all of the prompt behavior is con-
fined to this Ignition device. The fuel and air temperatures are the base-
line values of 150°F and 600°F, respectively. The fuel NH3 level is the
baseline value of 500 ppm.
144
-------
LOW BTU GAS |VARIABLE COMPOSITION
I1500F, 500 PPM NH3
4s.
tn
HEAT TRANSFER
TTTTTT
t
'STIRRED^
P = 10 ATM f REACTOR
IGNITION
AIR, 600°F
RICH PRIMARY
*,T
n n m
HEAT LOSS
SECONDARY
5% EXCESS AIR
tfiiU
• ' AIR
Figure 4-37. Staged Supercharged Boiler - MKAP Analog Schematic
-------
O.
a.
o
cj
300
250
200
150
100
50
0
NH
NO
3SOO
3000
2500
2000
1500
1000
100
200 300
TIME (MSEC)
400
Figure 4-38. Nitrogen Species and Temperature in Rich Primary Zone
— Supercharged Boiler
NOTE: = 1.5; Ordinate values from stirred reactor ignition zone
TSR = 1.0 msec; NH3 = 500 ppm in fuel; No heat transfer
QC
=>
Q.
s
146
-------
300
250
200
150
o 100
50
NO
100
200 300
TIME (MSEC)
3500
3000
2500
2000
S:
UJ
1500
1000
500
400
500
Figure 4-39. Nitrogen Species and Temperature In Rich.Primary Zone
- Supercharged Boiler
NOTE: + = 1.4; Ordlnate values from stirred reactor ignition zone
TSR = 1.0 msec; NH, = 500 ppm 1n fuel; No heat transfer
147
-------
150
100
Q-
Q.
.
cc
t_J
§
50
1 1
Additional Concentrations at
Time =1.0 sec
0 = 9.5 ppm
OH = 1.09 ppm
H = 1.61 ppm
3500
3000
2500
2000
CXL
•=c
o:
a.
•Si
LU
1500
1000
100
200 300
TIME (MSEC)
400
500
500
Figure 4-40. Nitrogen Species and Temperature 1n Rich
Primary Zone - Supercharged Boiler
NOTE: 4» = 1.33; Ordinate values from stirred reactor
ignition zone TSR = 1.0 msec; (NHaJo = 263 (500
ppm in fuel); No heat transfer
148
-------
3500
3000
2500
CL.
Q_
O
o
2000
OL
IjJ
O-
Figure 4-41.
NOTE:
TIME (MSEC)
Nitrogen Species and Temperature in Rich
Primary Zone -Supercharged Boiler
* = 1.15; Ordinate values from stirred reactor
ignition zone TSR = 1.0 msec; (NH3)0 = 243 (500
ppm in fuel); No heat transfer
149
-------
For 4>=1.5 (Figure 4-38) there is practically no prompt fuel-N reduction
and the situation does not improve as residence time is increased. This is
distinguished from the gas turbine rich primary zone situation for which the
temperature and fuel ammonia concentration were considerably higher. For
=1.4 the situation improves slightly with 15 percent prompt reduction of the
NO precursor species increasing to 30 percent with a 200 msec plug flow
residence time. HCN generation increases as equivalence ratio decreases
becoming significant for only equivalence ratios of 1.33 (Figure 4-40) and
lower (Figure 4-41). For richer conditions the hydrocarbon radical concen-
trations are apparently not sufficiently large to promote HCN generation at
the low prevailing temperatures.
Under the leaner conditions in Figures 4-41 (
-------
Plug Flow
Residence
Time
1-2 1.4
EQUIVALENCE RATIO,
Figure 4-42. Sum of Residual Nitrogen Species from
Premixed Plug Flow Primary - Supercharged
Boiler
NOTE: 5J,1,rTed reactor ignition zone TSR = 1.0 msec);
= 500 ppm in fuel
151
-------
150
100
D_
D-
CtL
o
2
O
50
NO
HCN
3500
3000
2500
2000
1500
1000
100
200 300
TIME (MSEC)
400
500
500
Figure 4-43. Nitrogen Species and Temperature in Rich
Primary Zone -Supercharged Boiler
NOTE: = 1.33; Ordinate values from stirred reactor
ignition zone T$R = 1.0 msec; (NH3)o = 0; No
heat transfer
152
-------
o
z
o
100 200 300
TIME (MSEC)
400
3500
3000
2500
2000
UJ
a:
o.
UJ
1500
1000
500
500
Figure 4-44. Nitrogen Species and Temperature in Rich
Primary Zone - Supercharged Boiler
Values from
reactor
153
-------
3500
D.
C_
o
I—
OL
h-
_
o
o
500
100
200 300
TIME (MSEC)
400
Figure 4-45. Nitrogen Species and Temperature in Rich Primary Zone
- Supercharged Boiler
NOTE: <}> = 1.33; Ordinate values from stirred reactor ignition zone
TSR = 1.0 msec; (NHsJo = 526 ppm (1000 ppm in fuel); No heat
transfer
154
-------
the baseline case with fuel NH3 = 500 ppm, the results shown on these
figures are as expected with one exception. For the case of no NH3 in the
fuel (Figure 4-43) there is a synthesis of a small amount of NH3 (10 ppm)
over 500 msec. One can speculate that this is due to N and NH atom genera-
tion through the Fenimore reactions. These in turn would combine with OH
radicals to eventually form NH-.
3
4-4'3'2 Deviations from Baseline Primary Reactor Specifications
Now that the factors influencing N0x production in the baseline primary
reactor have been discussed, the following additional variables will be
addressed:
• Pressure
• Stirred reactor residence time
• Heat transfer from plug flow reactor
• Fuel composition - methane content
Pressure Effects
Figures 4-46 and 4-47 show the effects of reduced pressure on the *-l.33
baseline case. The pressure levels are 1.0 and 0.1 atmospheres, respectively.
The kinetically-limited NO levels increase with reduced pressure similar to
the equilibrium NO level. The NH3 levels are essentially zero at these lower
pressures with equilibrium strongly favoring NO. The HCN concentrations also
diminish as pressure is dropped.
One of the reasons for selecting reduced pressures is to provide computa-
tional data to compare with experimental HCN data from one atmosphere and
lower pressure flames. In general, the experimental results have not exhibited
the relatively high predicted levels of HCN for high pressure conditions. It
now appears that HCN is generated only under restricted conditions of tempera-
ture, pressure, equivalence ratio, fuel composition, and degree of backmixing
(e.g., stirred reactor residence time). Conditions favoring HCN generation
are high temperature, high pressure, high CH4 content/and a high degree of
backmixing. Under these conditions HCN synthesis increases with increasing
equivalence ratio. For example, in comparing the supercharged boiler with the
155
-------
150
3500
(NO)
3000
100
2500
Q-
D-
2000
o
o
50
1500
1000
HCN
500
100
200 300
TIME (MSEC)
400
500
Figure 4-46. Nitrogen Species and Temperature in Rich Primary Zone
— Supercharged Boiler
NOTE:
-------
3500
100
200 300
TIME (MSEC)
500
400
500
Figure 4-47. Nitrogen Species and Temperature in Rich
Primary Zone -Supercharged Boiler
NOTE: <*> =1.33; Ordinate values from stirred reactor
ignition zone TSR = 1.0 msec; (NH3)0 = 263 (500
in fuel); No heat transfer; Pressure = 0.1 atm
157
-------
gas turbine results of Section 4.3 there is a considerable difference in the
equivalence ratio range over which HCN is generated in appreciable quantities.
With the higher gas turbine primary reactor temperature, HCN generation
increases as equivalence ratio increases. This is consistent with the notion
that backmixing or richer conditions lead to higher hydrocarbon radicals pro-
vided the temperature is sufficiently high. On the other hand, supercharged
boiler primary zone temperatures are apparently too low to promote HCN genera-
tion for equivalence ratios greater than 1.33. This situation may be altered
for stirred reactor residence times greater than 1.0 msec.
Effect of Stirred Reactor Residence Time
Figures 4-48 and 4-49 show the effect of increased stirred reactor resi-
dence time. Large increases of HCN are indicated which are apparently due
to a correspondingly large increase in hydrocarbon fragments associated with
increased backmixing. NO levels also increase somewhat which is probably due
to N atom generation via the Fenimore reaction followed by the extended
Zeldovich reactions (N + OH). The overall effect of increased stirred reactor
residence time is to make the systems appear richer with an attendant decrease
in nitrogen species reduction effectiveness.
Heat Transfer Effects
Figure 4-50 shows the effect of a distributed heat loss corresponding
to 200°F for the =1.15 case (Figure 4-41). Under these conditions the heat
loss has very little effect which is consistent with the results obtained
in Section 4.3 for the gas turbine analysis. A small amount of NH3 is
generated under these conditions indicating that lower temperatures favor
these synthesis reactions.
Fuel Methane Content
Figure 4-51 presents the =1.33 case with the methane content allowed
to go instantaneously to CO and H,,. Comparison should be made with Fig-
ure 4-40. In the absence of methane the 0, OH and H radical concentrations
are increased by large factors as indicated on the two figures and as a
consequence the prompt NH3 reduction is much larger and the prompt reduction
of NO precursors is over 70 percent. The synthesis of NH3 could be initiated
by the presence of the large H content combining with N2 to yield N and NH.
158
-------
3500
200 300
TIME (MSEC)
3000
2500
2000
o:
=3
<
LU
CL
LU
h-
400
509
NOTE:
msec,
,„
3 - 00 ppm 1n fuel; No heat transfer
159
-------
3500
3000
2500
2000 <
ex
1500
1000
500
100
200 300
TIME (MSEC)
400
Figure 4-49. Nitrogen Species and Temperature in Rich Primary
Zone - Supercharged Boiler
NOTE: 4> = 1.33; Ordinate values from stirred reactor ignition
zone TSR = 100 msec; NH3 = 500 ppm in fuel; No heat transfer
160
-------
IV)
100
Q.
D-
o
o
50
100
HCN
200 300
TIME (MSEC)
3500
3000
2500
2000
1500
1000
500
400
500
Figure 4-50. Nitrogen Species and Temperature in Rich Primary
Zone - Supercharged Boiler
N°TE: tnn1>15; °i:dlnate va]ues from stirred reactor ignition
zone TSR =1.0 msec; (NH3)0 . 243 ppm (500 ppm in fuel)-
Heat transfer = -71 Btu/lb (equivalent to -20QOF AT) '
161
-------
150
100
Q.
Q_
C_)
z
o
50
Additional Species Concentrations at
1.0 msec Plug Flow:
0 = 33 ppm
OH = 608 ppm
H = 1200 ppm
NH.
3500
3000
2500
2000
o:
UJ
Q.
1500
1000
500
100
200 300
TIME (MSEC)
400
500
Figure 4-51. Nitrogen Species and Temperature in Rich
Primary Zone — Supercharged Boiler
NOTE: = 1.33; Ordinate values from stirred reactor
ignition zone T$R = 1.0 msec; (NH3)o = 263 (500 ppm
in fuel); No heat transfer; No fuel methane (equiva-
lent to CH4 + 1/2 02-*-CO + 2H20 proceeding
instantaneously)
162
-------
4'4'3-3 Secondary Stage - Distributed Heat Exchange and Energy Release
The rich primary stage discussed in the previous section serves two
functions. First, it provides a means to partially control the fuel nitro-
gen conversion to NOX, and second, it allows the subsequent staging of energy
release (combustion air addition) and heat transfer to control thermal NO.
This section presents the results of secondary stage calculations using the
primary reactor data from the previous section as input. In all cases a
primary reactor equivalence ratio of 1.15 was used since this was most effec-
tive in minimizing the effect of fuel nitrogen.
Figures 4-52 and 4-53 present species and temperature distributions for
a 1.0 second residence time secondary stage. They differ only in the distri-
bution of secondary air. In Figure 4-52 the air is added over 125 msec and
in Figure 4-53 it is added over 500 msec. In both cases the primary residence
time was taken as 500 msec and the primary zone heat loss corresponded to a
200 F drop in primary zone temperatures.
The heat transfer distribution is shown in the sketches on Figures 4-52
and 4-53 and has a peak-to-minimum heat flux ratio of 3:1 with the peak
located at a point 25 percent of the length down the secondary combustor.
This distribution was based on preliminary engineering estimates of radiative
heat transfer. Perturbations on this distribution and on the required
secondary residence time (as dictated by heat transfer limitations) were
found to have little effect on the final NO level. Hence, refined heat
transfer calculations were not performed.
With the air distribution in Figure 4-52, the NO level is relatively
uneffected until the system leans out and the equivalence reaches 1.0 at
approximately 100 msec into the system. At this point the remaining HCN
rapdily converts to NO over a period of 50 msec and thermal NO production
commences and continues until the temperature drops below the thermal thres-
hold of 2700-2800°F which occurs 350 msec downstream. Further downstream
the NO level remains constant at 110 ppm.
In Figure 4-53 the secondary air addition is delayed relative to Fig-
ure 4-52 and occurs uniformly over the first half of the combustor, thus
reducing peak temperatures and eliminating thermal NO entirely. The NO level
163
-------
Stirred reactor TCP = 1.0 msec
followed by plug flow (T =
(NH3)Q in fuel = 500 ppm
Secondary:
0.95
Air distribution over 125 msec
Heat removed as shown in sketch
3000
2500
2000,
Q.
1500
1000
1000
TIME (MSEC)
1400
1 500
1500
Figure 4-52. Nitrogen Species and Temperature in Secondary Stage - Supercharged Boiler -
Air Added Over 125 msec
-------
tn
Stirred reactor T$R » 1.0 msec
followed by plug flow (500 msec)
Secondary:
= 0.95
Air distribution over 500 msec
Heat removed as shown in sketch
600 700
900 1000 1100
TIME (MSEC)
3000
2500
2000
S
LU
a.
1500
1000
1200 1300 1400 1500
500
Figure 4-53. Nitrogen Species and Temperature in Secondary Stage -Supercharged Boiler -
Air Added Over 500 msec
-------
at first decreases due to dilution and then increases to its final value of
90 ppm as the HCN is converted.
Figure 4-54 shows the effect of entirely eliminating the long residence
time primary reactor. The output of a 1.0 msec adiabatic stirred reactor at
an equivalence ratio of 1.15 (see Figure 4-41) was used as input to the
secondary stage and the heat transfer and air addition distributions were
maintained the same as in Figure 4-53. Again, thermal NO is eliminated and
the final NO level is 150 ppm. Thus the benefit of a long primary cooking
period is on the order of a 60 ppm reduction in NO emission. Probably a com-
promise design is best in which a 200 msec primary zone participates actively
in the heat transfer process. Such a combustor would have emission levels on
the order of 125 ppm.
Figure 4-55 combines a single plot, the primary and secondary distribu-
tions from Figures 4-50 and 4-52 respectively.
Finally, Figure 4-56 provides a calculation identical to that of Fig-
ure 4-54 except that the fuel NHL has been eliminated. The difference in
NO levels following the stirred reactor persists throughout the system and
indicates that the net fuel nitrogen contribution to the exhaust emissions
in Figure 4-54 is approximately 100 ppm.
166
-------
cn
I
UJ
o
Primary:
• 1.15
Stirred reactor T
(NH3)0 1n fuel -
Adiabatlc
Secondary:
* - 0.95 at exit
Air distribution over 500 msec
Heat removed as shown in sketch
1500
200
300
400 500 600
TIME (MSEC)
3000
2500
2000
I
1000
700
800
900
1000
Figure 4-54. Nitrogen Species and Temperature in Secondary Stage -Supercharged Boiler
— Stirred Reactor Primary Stage
-------
CO
isn
130
110
•^
t.90
f
3
I 70
E
J
J
f
3 50
30
10
^
^
"\
_______!
^NO
^-^
HCN
NH3
+
^
'
«1.0
"\
_ \
^^ -
^ ~_
Primary:
= 1 1 5
Stirred reac
followed by
(NH3)Q in fu
Secondary:
* = 0.95 at
Air distribu
Heat removed
Figure 4-52
tor TSR = 1.0 m
plug flow 500 m
el - 500 ppm
exit
tion over 125 m
as shown in sk
0 . 100 300 SCO 700 900 1100 13
' .
sec
sec
sec
stch
00 15
3100
2700
c
2300
1900
1500
1100
00
TIME (MSEC)
Figure 4-55. Nitrogen Species and Temperature in Primary and Secondary Stage - Supercharged Boiler
-------
vo
600 800
TIME (MSEC)
1000
1200
3300
2900
•• 0.95 at exit
Air distribution over 500 msec
Heat removed as shown in sketch
1400
500
Figure 4-56. Nitrogen Species and Temperature in Secondary Stage — Supercharged Boiler
-------
5.0 CONCLUSIONS
The major conclusions of this study were:
• A chemical kinetic reaction set is available which may be suitable
for engineering calculations; however, it needs further validation,
modification and reduction.
• Gasifier-COGAS systems appear to be potentially efficient energy
conversion devices if developed as integrated systems.
• Low N0x emission combustors can probably be developed without the
need for NH3 removal from the gasifier raw gas.
For both concepts studied it was found that careful combustor design can
contribute significantly to reduction of N0x emissions from low Btu gas sys-
tems. Staging the combustion process was found to be desirable in both cases.
The optimum design for an adiabatic gas turbine combustor was found to
be a rich (1.33 <> < 1.45) primary reactor section with a residence time of
at least 200 msec followed by a gradual mixing of the primary products into
the dilution air stream. N0x emissions for such a system could potentially
be lower than those for an unstaged system by factors of five or more.
For the supercharged boiler case the optimum configuration consisted of
a rich primary zone (*=1.15) and a secondary zone with gradual air introduction
The potential NOX reduction compared with the unstaged configuration is a
factor of three in this case.
Although this study considered only idealized combustor configurations
and utilized a kinetic model that has not been completely validated the
results can be taken to indicate that there is an excellent likelihood that
NOX emissions from low Btu gas power generation systems can be held within
acceptable limits. If validated, this conclusion will influence the need
for high temperature ammonia cleanup.
It must be stressed that the calculations described in this report were'
made with the most appropriate reaction set available at that time and with
full recognition of some of its deficiencies. Experimental evidence exists
indicating that under certain conditions in premixed flames a large quantity
of the NH3 is converted to HCN prior to or coincident with the formation of
171
-------
NO or N~. The current reaction set does not allow for this type of exchange
nor is the oxidation of HCN adequately described.
One interesting observation is that for certain conditions (stoichiometry,
time and temperature) final NO levels appear to be independent of the initial
fuel nitrogen concentration. This observation is perhaps more relevant to
other coal-derived fuels (e.g., liquids) with high nitrogen contents, and there-
fore, high NO emission potential. Suitable combustion chamber design can possi-
bly overcome NO emission problems by maintaining rich products at relatively
A
high temperatures to allow nitrogen fragments to equilibrate.
172
-------
4-
REFERENCES
?' F-L:' and Giramonti, A'°" An Advanced-Cycle Power System Burnina
led and Desulfurized Coal. Proceedinas nf the Fi»«cf <:am-inr>M «~
, , . wvrni. i i W.CCUiiiya ui trie nrst seminar on
November,P597uZ and Combustion Gases- Geneva, Switzerland,
2' atdthI'4thFSvnthlHrLp^li-rOCrSS ^ The ^Oute to SNG frorn Coal• Presented
at the 4th Synthetic Pipeline Gas Symposium, Chicago, October, 1972.
3. Shaw, H., and Magee, E.M., Evaluation of Pollution Control in Fossil
Fuel Conversion Processes, Lurgi Process, EPA 650/2-74-009-C (1974).
Banchik, I.N., "The Winkler Process - A Route to Clean Fuel from Coal "
S*»^ -'
6' wFiatrhTThprocess ^ISip!;":' anrdD|fanady' J'F-' "dean Environment
Aspects
?' Gasr^ednd'oes^UuSH1?' ?-J'S A" Advanced- Cycle Power System Burning
the DisulXurlSSnn I c C?a1', Proceedln9s of the First Seminar on
f Fuel and Combustion Gases. Geneva, Switzerland,
November, 1970
8.
s:
^ — — "G--:-
Colorado, November 1975, hHA-bUU/^/b-212 hRDA 47, August 1976, Page 359.
10. Salvador, L.A., Vidt, E.J., and Holmgren, J.D., "The Westinqhouse
1Z C° 1ne
.
J-i1Z?«n^ C°f 1neS Cy^le PrOCess: Status °f Technologan Environ-
mental Considerations". Paper presented at the EPA Symposium on thl
65^'5 °f FUel C°nVerSi
"' "' C1ea" ^s from
173
-------
12. Smith, E.B., Mador, R.S., "Coal Gas Combustion in Industrial Gas
Turbines".
13. Gill more, D.W. and Liberature, A.J., "Pressurized, Stirred, Fixed
Bed Gasification", in Symposium Proceedings: Environmental Aspects
of Fuel Conversion Technology II. EPA-600/2-76-149, June 1976.
14. Colton, C.B., Dandavati, M.S., and Bruce, May V., "Low and Intermediate
Btu Fuel Gas Clean-UP", presented at the EPA Symposium on the Environ-
mental Aspects of Fuel Conversion Technology, Hollywood, Florida,
December 1975.
15. Ball, D., Smithson, G., Engdahl, R., and Putnam, A., Study of Potential
Problems and Optimum Opportunities in Retrofitting Industrial Processes
to Low and Intermediate Energy Gas from Coa, EPA 650/2-74-052, May 1974.
16. B. Lewis and G. Von Elbe, Combustion. Flames and Explosions of Gases,
Academic Press, New York, third edition, 1959.
17. Hottel, H.C. and Sarofim, A.F., "Radiative Transfer, first edition,
McGraw Hill, 1967.
18. IKAP — Industrial Kinetics Analysis Program, A Computer Program for the
Analysis of Chemically Reacting Gas Mixtures, Ultrasystems, 1971.
19. Baulch, D.L., et al, "High Temperature Reaction Rate Data No.l",
Dept. of Physical Chemistry, University of Leeds, England, May 1968.
20. Bueters, K.A., et al, "Performance Prediction of Transentially Fired
Utility Furnaces by Computer Model", 15th International Symposium on
Combustion, Tokyo, Japan, August 25 - 31, 1974.
21. Habelt, W.W. and Selker, A.P., Operating Procedures and Prediction for
NO Control in Steam Power Plants Paper presented at the Central States
Sefition of the Combustion Institute - March 26, 27, 1974.
174
-------
APPENDIX A
ALTERNATIVE CONVENTIONAL AND COMBINED CYCLE SYSTEMS
This appendix gives the details of the conventional and combined
cycle power generating systems investigated by Combustion Engineering as
part of the preliminary screening process. See Section 2 for cycle
comparisons.
175
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT 68-02-1361
CE CONTRACT 9*73
PROJECT 901001
SYSTEM CONCEIT NO: Al
SYSTEM DESIGNATION: Conventional Utility Steam Power Plant
PRIMARY FUEL TYPE: Clean, low temperature, low pressure, low BTU gas
STEAM TURBINE CYCLE:
BOILER: 1 - Controlled circ; 2400 psia, 100S/1005°F, 3.5-106lbm/hr main ste-m flow
AIR PREHEATER: 2 - Ljungstrom
BURNERS: C-E Type T (Tangential Firing)
TURBINE-GENERATOR: Current designs; 500 HW (lOOt)
GAS TURBINE CYCLE:
COMBUSTOR:
TURBINE-GENERATOR
NOxCONTROL: Tangential Firing with Overfire Air, Flue Gas Recirc.
SYSTEM SKETCH:
Primary
Sec. Fuel
Ign. rue
Boiler
Stack
FD Fan
ADVANTAGES: . LOW plant net heat rate
- All system components of current design
- Low NOx potential
- Low operating cost
- High turndown ratio
- Conventional controls
- Good system flexibility in following gasifier thruput (via primary fuel
storage or secondary fuel capability)
- Retrofit of existing boiler practical
- Good system flexibility in followinj1 load demand (via primary fuel
storage or secondary fuel capability)
_ Potentially good system reliability
DISADVANTAGES: . High capital cost
176
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT 6842.1361
CE CONTRACT «67)
PROJECT 901001
SYSTEM CONCEPT NO: &
SYSTEM DESIGNATION: Conv.ntion.1 Utility St.a. Pow.r Pl.nt
PRIMARY FUELTYPE: „..„, low t-p.ratur. f hlgh
STEAM TURBINE CYCLE:
'F, 3.5xlO« Ibm/hr ..in .t.aa flov
T (Tang.ntlal Firing)
" Current d..lgna; 500 MU (100X)
CAS TURBINE CYCLE:
COMBUSTOR:
TURBINE-GENERATOR:
NO.CONTROL: T«g.ntt.l Firing with Ov.rfir. Air. Flu. 0.. R.circ.
SYSTEM SKETCH:
Prlaary Pimi t
. Sac. Fual .
Ign. Iruai
Stack
ID Fan
ID Fan
ADVANTAGES: . ^ plant n.t h..t rat.
- All syittn conponcnts of currant daalcn.
- Low MOx potential .
- Low optra ting co«
- High turndown ratio
- Convantlal controls
- Retrofit of axiatlng boilar practical
"or...
DISADVANTAGES:*
- High capital coat
- High pr...ur. f».l .or. id.ally .uit.d to cortinad cycl. application..
177
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL CASIFIERS
TASKC
EPA CONTRACT 68-OM36I
CE CONTRACT 967J
PROJECT 901001
SYSTEM CONCEPT NO: Bl
SYSTEM DESIGNATION: Conventional Utility Stcar, Plant with ^as frecoolcr
PRIMARY FUEL TYPE: Clean, hij-h tenperature, low pressure, low BTU gas
STEAM TURBINE CYCLE:
BOILER: 1 - Controlled circulation; 2400 PSIA,' lOOS/lOOS"17; 3.5 x 100 Ihn/hr nain stc.ir
AIRPREHEATER: 2 - [,jun.CStron
BURNERS: C-l:. type T (Tangential Firinp)
TURBINE-GENERATOR: Current designs; 500 »W (100%)
GAS TURBINE CYCLE:
COMBUSTOR:
TURBINE-GENERATOR:
NO* CONTROL: Fuel "Tas Precooling, Tangential ririnp wAVei-fire Air, rlue (\ns ".ecirc.
SYSTEM SKETCH:
Pflpary Fue
Process Ste
Stack
Air
' t**1 Piant net heat rate
_ A11 system components Of current desiRn
- Low NOx potential
- Low operating cost
- High turndown ratio
- Conventional controls
- Good system flexibility in following gasifier thruput (via primary fuel
storage or secondary' fuel capability)
- Retrofit of existing boiler practical
- Process steam produced for plant heating, auxiliaries, or power
- Rood system flexibility in following load demand (via primary fuel storage
or secondary fuel capability)
- Potentially good system reliability
DISADVANTAGES: - High capital cost
- Heavily dependent on gas cooling for system availability
- Process steam flow dependent on plant load
178
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT M4M36I
CE CONTRACT 967 J
PROJECT 901001
SYSTEM CONCEPT NO: B2
SYSTEM DESIGNATION: Conventional Utility Steam Plant w/fias Precooler
PRIMARY FUEL TYPE: Clean, high temperature, low pr...ur., low BTU «...
STEAM TURBINE CYCLE:
: 24°° P?U; 100S/100S'F: " x 106
main steam flow
S: C-E Type. T (Tangential Firing)
TURBINE-GENERATOR: Current designs; 500 w (100%)
CAS TURBINE CYCLE:
COMBUSTOR:
TURBINE-CENERATOR:
NO,CONTROL: Fuel Gas Precooling,Tangential Firing wAVerfire Air. Flue Q« Recirculation
SYSTEM SKETCH:
n r
Primary Fuel
Power
Stack
ID Faa
ADVANTAGES: - Low plant net heat rate
- All system conixments of current design
- Low operating cost
- Very high turndown ratio
- Conventional controls
f?r lant
«uxlllari«s or power
• High NOjj potential
- High capital cost
" 1?"™^* of listing boiler may not he practical
- Explosive potential of gas precooler
179
-------
Primary Fuel.
Process SteaSr
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT 68-02-136!
CE CONTRACT »673
PROJECT 901001
SYSTEM CONCEPT NO: B3
SYSTEM DESIGNATION: Conventional Utility Steam Plant with Gas Precooler
PRIMARY FUEL TYPE: Clean, high temperature, high pressure, low Bill gas
STEAM TURBINE CYCLE:
BOILER: 1-Controlled clrc; 2400 psia, 1005/1005'F; 3.5xl06 Ibm/hr main steam flow
AIR PREHEATER: 2-LJungscrom
BURNERS: CE-Type T (Tangential Firing)
TURBINE-GENERATOR: Current designs} 500 MW (100%)
GAS TURBINE CYCLE:
COMBUSTOR:
TURBINE-GENERATOR:
NO*CONTROL: Fuel Gas Preceding, Tangential Firing with Overflre Air, Flue Gas Reclrc.
SYSTEM SKETCH:
Steam Tur
Stack
Air
ADVANTAGES: _ Low piant net heat rate
- All system components of current design
- Low NOx potential
» Low operating cost
- High turndown ratio
- Conventional controls
- Good system flexibility in following gasifier thruput (via primary fuel storage or secondary
fuel capability).
- Retrofit of existing boiler practical :
- Good system flexibility in following load demand (via primary fuel storage or secondary fuel j
- Potentially good system reliability. capabilitj
DISADVANTAGES:
- High capital cost
- Heavily dependent on gas cooling for system availability
- Process steam flow dependent on plant load
- High pressure fuel more ideally suited to combined cycle applications.
180
-------
FURNACE DESIGN FOR BURNING OFFCAS
FROM COAL CASIFIEKS
TASKC
EPA CONTRACT 68-02-1J6I
CE CONTRACT «673
PROJECT 901001
SYSTEM CONCEPT NO: Cl
SYSTEM DESIGNATION: Conventional Utility Steam Plant w/High Tenperature Burners
PRIMARY FUEL TYPE: Clean, high tenperature, low prenurt. low BTU KM
STEAM TURBINE CYCLE:
BOILER: 1 • Controlled circulation; 2400 PSIA, 100S/100S°F; 3.5 x 10^
BURNERS: High teit|>erature Sulzer or new technology
TURBINE-GENERATOR: Current designs, 500 M* (100?)
GAS TURBINE CYCLE:
COMBUSTOR:
TURBINE-GENERATOR:
NOx CONTROL: Overfire Air, FlueGas Reclrc.
SYSTEM SKETCH:
Primary Fuel
S«c. Fuel
Tg" »..-\
Bo-Mo,
Stean
Power
Stack.
Air
ADVANTAGES: ' Low plant net heat rate
,- Low operating cost
- High turndown ratio
- Conventional controls
108d deffland Cvia secondary fuel capahility)
DISADVANTAGES: • Burner development may be required
- Very high NOx potential
- High capital cost
181
-------
FURNACE DESIGN FOR BURNING OFFCAS
FROM COAL CASIFIERS
TASKC
EPA CONTRACT 68-02-1J6I
CE CONTRACT 9673
PROJECT 901001
SYSTEM CONCEPT NO: c-
SYSTEM DESIGNATION: Conventional Utility Steam Plant w/High Temperature Burners
PRIMARY FUEL TYPE: Clean, high temperature, low pressure, low BTU gas
STEAM TURBINE CYCLE.
BOILER: I - Controlled circulation; 2400 PSIA. 1005/100S°F; 3.5 x 106 lbm/hr main steam flow
AIR PREHEATER: 2 - Ljungstrom
BURNERS: High temperature Sulzer or New technology
TURBINE-GENERATOR: Current designs; 500 NW (1004)
GAS TURBINE CYCLE:
COMBUSTOR:
TURBINE-GENERATOR:
NOx CONTROL: Overfire air; Flue Gas Recirc.
SYSTEM SKETCH:
Primary
Process.Steam
Power
Stack
ADVANTAGES: - Low plant net heat rate
- Low operating cost
- High turndown ratio
- Conventional controls
- food system flexibility in following gasifier thruput (via prijnary fuel storace or
secondary fuel capability) s
- Retrofit of existing boiler practical
- Process steam produced for plant heatine, auxiliaries, or power
- food system flexibility in following load demnd (via primary fuel storage
or secondary fuel capability)
- Potentially good system reliability
DISADVANTAGES: .
development may be required
- High NOX potential
- High capital cost
182
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT AM7-IM1
CE CONTRACT W J
rROJECT 901001
SYSTEM CONCEPT NO: C3
SYSTEM DESIGNATION: Conventional Utility Steam Plant w/High Temperature Burners
PRIMARY FUEt TYPE: Clean, high temperature , low Pr.,.ur.. low BTU ga.
STEAM TURBINE CYCLE:
A?R^ATERCrfro«ieSvr10n! 24°° •*»• ^S/iOOS'F; 3.S x 106
iURNERS: High temperature Sulzer or New technoloev
TURBINE-GENERATOR: Current designs; 500 »« (100?)
CAS TURBINE CYCLE:
COMBUSTOR:
TURBINE-GENERATOR:
NOx CONTROL: Overfire Air; Flue Ras Recirc.
main steam flow
Stack
ADVANTAGES: - High turndown ratio
- Conventional controls
Plant h««"»«. auxiliarlea or pow.r
108d
- Potentially good system reliability
DISADVANTAGES: . High plant net heat rate
" J*™8* development nay be reouired
- High N0j( potential
- High capital cost
- High operating cost
" SSSJJr S01*?11?1 of *« Precooler
Retrofit of existing boiler may not be practical
storage
"orage or
183
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT 68-02-1361
CE CONTRACT 9673
PROJECT 901001
SYSTEM CONCEPT NO: M
SYSTEM DESIGNATION: industrial Boiler System
PRIMARY FUEL TYPE: Clean, low temperature, low pressure, low BTU gas
STEAM TURBINE CYCLE:
BOILER: n . Natural circulation, shop assembled; 1800 psia 9SO°F; 0.4 x 106 Ibm/hr steam flow
AIRPREHEATER:2 - Ljungstrom
BURNERS: Front wall
TURBINE-GENERATOR: Current HP turbo design; possible new LP design; SOO HW (100*)
GAS TURBINE CYCLE:
COMBUSTOR:
TURBINE-GENERATOR:
NO*CONTROL: Overfire Air; Low excess air operation
SYSTEM SKETCH:
Stack
ADVANTAGES: . ym system components of current design
- Very high, turndown ratio
- Good system flexibility in following gasifier thruput (via primary fuel
storage or secondary fuel capability)
- Good system flexibility in following load demand (via primary fuel storage
or secondard fuel capability)
- Shop assembled boilers reduce load time for delivery
DISADVANTAGES: . yery high plant net heat rate
- May require new steam turbine design
- High NOx potential
- Very high operating cost
- Control system complexity
- Potentially poor system reliability
- High capital cote
184
-------
FURNACE DESIGN FOR BURNING OFFCAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT M4M36I
CE CONTRACT »»7J
PROJECT 901001
SYSTEM CONCEPT NO: D2
SYSTEM DESIGNATION: industrial Boiler System
PRIMARY FUEL TYTE: cle.n( hlgh temp«ratur«, low pt«».ur., iw BTU gM
STEAM TURBINE CYCLE:
BOIUR: il_Natural circulation, shop aaseabled; 1800 paia, 950*p; 0.4x10 lbn/hr main it rum flow
AIR PREHEATER: 2-LjunRStroo
BURNERS: Frontwall
TUR1INE-CENERATOR: Curr«nt HP tort, d.ilgn; poidbU n«r LP da.lgn ; 500 MW (100Z)
CAS TURBINE CYCLE.
COMIVSTOR:
TURBINE-CENERATOR:
N0>CONTROL: Ov«rfir« Air; Low wceu* air op.r.tioo.
SYSTEM SKETCH:
Proc«» St«an
PrUurr fur
Stack
ADVANTAGES: -All syatan conpon«tvts' of currant dasl^n.
-Prot*»» Bteam produced for plant haatlng, auxlllarlta or power
-Vary high turndown ratio
-Good ay*tam flexibility in following Ka»ifi.r. thruput (via primary fual
atoraga or >econdary fual capability)
-Good ayitam flexibility in followinB load demand ( via prioary fuel
atoraga or secondary fuel capability )
-Shop aiaenblad boliars reduca laad tine for delivery
MS ADVANTAGES.
-Vary high plant net heat rata
-Hay require new steam turbine d«ilgn
-Vary high operating cost
-Control ayitera complexity
-Potentially poor system reliability
-High KOx potantlal
-High capital coat
185
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT 68-02-1361
CE CONTRACT 9673
PROJECT 901001
SYSTEM CONCEIT NO: 03
SYSTEM DESIGNATION: industrial Boiler System
PRIMARY FUEL TYPE: Clean, low temperature, high pressure, low BTO gas
STEAM TURBINE CYCLE:
BOILER: 11-Natural circulation shop assembled; 1800 psla 9SO*F; 0,4x10 Ibra/hr main stean flow
AIRPREHEATER: 2-LJungstrom
BURNERS: Front wall
TURBINE-GENERATOR: Current HP turb. design; possible n«w LP design; 500 MW (100*)
GAS TURBINE CYCLE:
COMBUSTOR.
TURBINE-GENERATOR:
NOx CONTROL: Overfire Air; low excess air operation
SYSTEM SKETCH:
Stack
ADVANTAGES: - All system components of current design
- Low NOx potential
- Very high turndown ratio
- Good system flexibility in following gaslfier thruput (via primary fuel storage
or secondary fuel capability)
- Good system flexibility in following load demand (via primary fuel storage or secondary fuel
capability
- Shop assembled boilers reduce lead time for delivery
DISADVANTAGES:- Very high plant net heat rate
- Hay require new steam turbine design
- Very high operating cost
- Control system complexity
- Potentially poor system reliability
- High pressure fuel more Ideally suited to combined cycle applications
- High capital cost
186
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT 684MMI
CE CONTRACT 9673
PROJECT 901001
SYSTEM CONCEPTNO: DA
SYSTEM DESIGNATION: Industrial Boiler System
PRIMARY FUEL TYPE: Clean, high temperature. high pressure. low BTU gas
STEAM TURBINE CYCLE:
0Vnbltd8 180° P-U- 950§F' O-
..In .t.M flow
BURNERS: Front wall
TURBINE-GENERATOR: Current HP turb. design; possible ».„ LP design; 500MW (100X)
GAS TURBINE CYCLE.
COMBUSTOR:
TURBINE-GENERATOR:
Np« CONTROL: Overflre Air; Low excess all operation
SYSTEM SKETCH:
Process Steam
Primary Fua
Stack
ADVANTAGES: . AH uyttm comooneatt of cutr,nt d..tgll
- Viry high turndown ratio
- Good ayatra flexibility in following gaalfleif thruput (via primary fual
•toraga or aacondary fuel capability). P"«ary tuai
- 'roeaaa ataam produced for plant heating auxiliaries or power
" lomd d"Mnd (vla priMry
- Shop assembled boiler* reduce lead Cine for delivery
DISADVANTAGES:
- Very high plant net heat rate
- May require new steam turbine design
. - Very high operating cost
- Control system complexity
- Potentially poor system reliability
- High pressure fuel more Ideally suited to coined cycle applications
- High capital cost
187
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT 684)2-1361
CE CONTRACT »673
PROJECT 901001
SYSTEM CONCEPT NO: El
SYSTEM DESIGNATION: Waste Heat Boiler Combined Cycle Plant
PRIMARY FUEL TYPE: Clean, low temperature, low pressure, low BID gal
STEAM TURBINE CYCLE:
BOILER: 5 - Natural circ.; 3 cycles to 1250 psia/1005*F ( max. )
AIR PREHEATER:
BURNERS:
TURBINE-GENERATOR: Current designs; 150 MW ( 30* )
GAS TURBINE CYCLE:
COMBUSTOR: Current design
TURBINE-GENERATOR: 5 gets; Current design ; 350 MW (701)
NOxCONTROL: w,t,r or atum injection do combustor.
SYSTEM SKETCH:
Power
Primary Fuel
ADVANTAGES:
Boil
Stack
- Low plant net heat rate
- All system components of current design
- Very high turndown ratio
- Good system flexibility in following gasifler thruput ( via primary fuel
storage- )
- Good system flexibility in following load demand ( via primary fuel
storage )
- Potentially good system reliability
- Low operating cost
DISADVANTAGES:
- High NOx potential
- Control system complexity
- High capital cost
188
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL CASIFIERS
TASKC
EPA CONTRACT 6*02-1361
CE CONTRACT 9673
PROJECT 901001
SYSTEM CONCEPT NO:
SYSTEM DESIGNATION:
PRIMARYFUELTYPE:
£2
w.at. H..t Boil.r Combined Cycle Plant
Clean. htgh temperature. low pr...ur., low ,TO ...
t0 125°
STEAM TURBINE CYCLE:
CHEATER: '
BURNERS:
TURBINE-GENERATOR: Current dealgns; ISO MW (30Z)
(
GAS TURBINE CYCU:
COMBUSTOR: Curr.nt d«.lgn
TURBINE-GENERATOR: 5 ..t. ; Curr.nt d..tgn ; 350 MW (70X)
NOx CONTROL: -tmmm
St««m or water inj action to combustor.
SYSTEM SKETCH:
Proceae Steam
Primary Fue
APVANTACES:
DISADVANTAGES:
AH ay*t*m component! of current design
Low plent net heat rate
Low operating cost
Very high turndown ratio
Stack
«-lfl« thruput (via pri-ary fuel
• High NOx potential
• Control ayatem complexity
• High capital coat
189
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL CASIFIERS
TASKC
ETA CONTRACT 6N-02-UM
CE CONTRACT M73
PROJECT 901001
SYSTEM CONCEPT NO: E3
SYSTEM DESIGNATION: Vaste Heat Boiler Combined Cycle Plant
PRIMARY FUEL TYPE: Clean, low tanp«rature, high pressure, lov BID gas
STEAM TURBINE CYCLE:
BOILER: 5 ~ Natural circ.; 3 cycles to 1250 peia/1005'F (max.)
AIR PREHEATER:
BURNERS:
TURBINE-GENERATOR: Currant designs; ISO MW (30Z)
GAS TURBINE CYCLE:
COMBUSTOR: Current design
TURBINE-GENERATOR: 5 sets; Current design ; 350 MW (70Z)
NO* CONTROL: Steam or water Injection to combust or.
SYSTEM SKETCH:
Boil<
Primary Fuel
Mr
Stack
ADVANTAGES: - Low plant net heat rate
- All system components of current design
- Lov operating cost
- Very high turndown ratio
- Good system flexibility in following gaslfier thruput (via primary fuel storage)
- Good system flexibility in following load demand (via primary fuel storage)
- Potentially good system reliability
DISADVANTAGES: - High NOx potential
- Control system complexity
- High capital cost
190
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT 6842-1361
CE CONTRACT W7J
PROJECT MINI
SYSTEM CONCEPT NO: EA
SYSTEM DESIGNATION: ,Waste Heat Boiler Combined Cycle Plmnt
PRIMARY FUEL TYPE: Clean, high t«mp«r»tur«, high pressure, low BTU gas
STEAM TURBINE CYCLE:
BOILER: 5-Natural clrc.; 3 cycles to 1250 pela/1003*F (MX.)
AIRPREHEATER:
BURNERS:
TURBINE-GENERATOR: 5 sets; Current design; -150 MV (30Z)
GAS TURBINE CYCLE: '
COMBUSTOR: Current design
TURBINE-GENERATOR: 5-i*ts, Current deilgn; 330 MU (70Z)
NO» CONTROL: steam or water Injection to coabustor.
SYSTEM SKETCH:
Boil*i-
Power
Cond.
Stack
ADVANTAGES:
- Low plane net heat rate
- Low operating cost
- Very high turndown ratio
- Good system flexibility in following gasifier thruput (via primary fuel storage)
- Process steam produced for plant heating auxiliaries or power
- Good system flexibility in following load denand (via primary fuel storage)
- Potentially good system reliability
DISADVANTAGES:
Combustor development may be required
1 Very high NOx potential
1 Control system complexity
High capital coat
191
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT 684)2-1361
CE CONTRACT 967J
PROJECT 901001
SYSTEM CONCEPT NO: E5
SYSTEM DESIGNATION: Waste H«at Boiler Combined Cycle Plant
PRIMARY FUEL TYPE: Clean, high temperature, high pressure, low BTU gas
STEAM TURBINE CYCLE: ,,-...- , s
BOILER: 5- Natural clrc.; 3 cycles to 1250 psia/1005'F (max.)
AIR PREHEATER:
BftplUCDC*
TURBINE-GENERATOR: Current designs; 150 MW C30X)
GAS TURBINE CYCLE:
COMBUSTOR: Current design
TURBINE-GENERATOR: 5 sets; Current design; 350 MM C/OZ)
NOx CONTROL: Steam or water Injection to conbustor
SYSTEM SKETCH:
Process Steam
Primary Fuel
Stack
ADVANTAGES: - Low plant net heat rate
- All system components of current design
- Very high turndown ratio
- Low operating cost
- Good system flexibility in following gaslfler thruput (via primary fuel storage)
- Process steam produced for plant heating auxiliaries or power
- Good system flexibility in following load demand (via primary fuel storage)
- Potentially good system reliability
DISADVANTAGES:
High NOx potential
Control system complexity
High ctpital cost
192
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EM CONTRACT 684)2-1161
CE CONTRACT 9*73
PROJECT 901001
SYSTEM CONCEPT NO: Fl
SYSTEM DESIGNATION: Exhaust Fired Boiler Combined Cycle Plant
PRIMARY FUEL TYPE: Cl..n, low temperature, low pr...ur., low BTU gat
STEAM TURBINE CYCLE:
BOILER: 1-Controll.d circ.; 2400 p.ta. 1005/1005'F; 2.8x10* lbm/hr main .tea. flow
AIRPREHEATER:
BURNERS: CE - Type T (Tangential Firing)
TURBINE-GENERATOR: Currant design; 400MH (BOX)
GAS TURBINE CYCLE:
OOMBUSTOR: Currant design
TURBINE-GENERATOR: 2 sets; Currant daalgn; 100 MW (20J)
NO* CONTROL: Tangential Firing w/0verflr« Air , Flue Ga. Recirc.
SYSTEM SKETCH:
Primary.
Power
Boilerf
1,
t>
J~
%_
Steam
ff
Turb.
^
H.X.
r—
Puir.p
_, Power
Power
Cond.
Process Steam
Stack
ADVANTAGES: - Very low plant net heat rate
- All system components of current design
- Low NOx potential
- Low operating cost
- High turndown ratio
ir "txl"U'y ln following g».i£i.r thruput (via primary fuel .tor...)
8tean Produ«d for plant heating auxiliaries or powir
lomd d€man
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
ETA CONTRACT 68-02-1J61
CE CONTRACT 9673
PROJECT WlOOt
SYSTEM CONCEPT NO: «
SYSTEM DESIGNATION: Exhaust Fir«d Boiler Combined Cycle Pl«nt
PRIMARY FUEL TYPE: Clean, high temperature, low pressure, low BTU gas
: 1-Controlled cite; 2AOO p.ia. 1005/1005'F; 2.8xl06 Ibm/hg main sti
AIR PREHEATER:
BURNERS: CE - Type T (Tangential Firing)
TURBINE-GENERATOR: Currant designs; 400MW (BOX)
flow
GAS TURBINE CYCLE:
COMBUSTOR: Current design
TURBINE-GENERATOR: 2-sets; Current design; 100 MW (ZOI>
NOx CONTROL:
SYSTEM SKETCH:
Tangential Firing with Overfire Air, Flue Gai Recite.
Primary Fuel
•oeess Steam
Air
Boiler. L
Steam
•
^-<^
^>
kr^l
t
^^
b«_
. L_
Chaarr Tui-Hj.
^
/ > H.X.
\Jj J~
Ml.-. A
Gas Turb.'
—* Power
Power
ir.d.
Process St
Stack
ADVANTAGES: - Very low plant net heat rate
- All system components of current design
- Low HQx potential
- low operating cost
- High turndown ratio
- Good system flexibility in following gasifier thruput (via primary fuel storage)
- Process steam produced for plant heating auxiliaries or power
- Cood system flexibility in following load demand (via primary fuel storage)
- Potentially good system reliability
DISADVANTAGES:
High capital cost
Control system complexity
• High capital cost
194
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GAS1FIERS
TASKC
ETA CONTRACT M4MMI
CE CONTRACT «&7J
PROJECT MIOOI
SYSTEM CONCEPT NO: 73
SYSTEM DESIGNATION: Exhaust Fired Boiler Combined Cycle Plant
Clean, low temperature, high pressure, lew BTU gas
PRIMARY FUEL TYPE:
STEAM TURBINE CYCLE:
BOILER: 1-Controlled clrc; 2*00 psla, 1005/1005'F; 2.8xl06 Ibm/br main
AJRPREHEATER:
BURNERS: CE-Type T (Tangential Firing)
TURBINE-GENERATOR: Current deslgnt; AOO MW (80%)
•teas flow
GAS TURBINE CYCLE:
COMBUSTOR: Current design
TURBINE-GENERATOR: 2-sets; Current design; 100MH (20X)
NOx CONTROL:
Tangential Firing with Overfire Air, Flue Gae Reclrc.
SYSTEM SKETCH:
Primary.
Process Steam
Turb.
Boiler
Steam
"1 0
y owv
J ,
Coabu
^^~
S
^f
Cc.._
/"
(
>
T..|.Ug^^'^
. r^
1
H.X^ 1
l^.rS. <
Air
L^^pT
1
Gas.
—» Power
Cond.
.X.
Power
Proctss
team
Stack
ADVANTAGES:
Very low plant net heat rate
All system components of current design
Low NOx potential
Low operating cost
Ulgh turndown ratio
Good system flexibility in following gaslfler thruput (via primary fuel storage)
Process ittarn produced for plant heating auxiliaries orpower
Good system flexibility In following load demand (via primary fuel storage)
Potentially good system reliability
DISADVANTAGES:
High capital cost
Control system complexity
195
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EFA CONTRACT MMH-I36I
CE CONTRACT 9673
PROJECT 901001
SYSTEM CONCEPT NO: F4
SYSTEM DESIGNATION: Exhauat Fired Boiler Combined Cycle Plant
PRIMARY FUEL TYPE: Clean, high temperature, high preaaure. low BTU gee
'i-Controllad circ; 2400 paia, 1005/1005'F; 2.8xl06 Ibra/hr main ataaa flow
STEAM TURBINE CYCLE:
BOILER:
A1RPREHEATER:
BURNERS: Hi Temp. Sulzer or Hew Technology
TURBINE-GENERATOR: Current dealgna; 400MW (802)
GAS TURBINE CYCLE:
COMBUSTOR: Current deaign
TURBINE-GENERATOR: 2 aeta; Current dealgn; 100 HW (20X)
NOxCONTROL: Ovarfira Air, Flue Gaa Recirc.
SYSTEM SKETCH:
•Primary Fuel
Turb
Stack
Cooip
ADVANTAGES: - Very low plant net heat rate
- Low operating cost
- High turndown ratio
- Good ayatem flexibility in following gasifler thruput (via primary fuel storage)
- Froceaa steam produced for plant heating auxiliaries or power
- Good ayatem flexibility in following'load demand (via primary fuel storage)
- Potentially good system reliability
DISADVANTAGES: - Burner development may be required
- High NOx potential
- High capital coat
- Control ayaten complexity
196
-------
FURNACE DESIGN FOR BURNING OFFCAS
FROM COAL GASIFIERS
TASKC
ETA CONTRACT »MM J41
CB CONTRACT WJ
*oiooi
SYSTtMCONCErTNO: F3
SYSTEM DESIGNATION: Exhaust Fired Boiler Combined Cycle Plant
PRIMARY FUEL TYPE: Clean, high temperature, high pressure, low BTU gas
STEAM TURBINE CYCLE:
BOILER: 1-Contrclled cite; 2400 pila, 1005/1005'F; 2.8x10° Ibm/hr main ac«an (low
AIKMtEHEATEK:
BURNERS: CE-Typa T (Tangential Firing)
TURBINE-GENERATOR: Currant deilgni; 400HV (80Z)
GAS TURBINE CYCLE:
COMBUSTOR: Current daalga
TUtBINC-CENEIUTOR: 2-.«t«-, Cutr.nt dulgn; 100MU (20S)
NOxCONTROL: Tangfcntial Firing with Ovarflr* Air, Flu* Ga* Racltc.
SYSTEM SKETCH:
Primary Full
Stack
ADVANTAGES: . very low plant net heat rate
— All aystea component* of current design
- Low NOx potential
- Low operating cost
- High turndown ratio
- Good system flexibility in following gaslfier thruput (via primary fuel atoragt)
- Steaa produced for plant heating auxiliaries or power
- Good system flexibility in following load demand (via primary fuel storage)
- Potentially good syitem reliability
DISADVANTAGES:
High capital cost
Control system complexity
197
-------
FURNACE DESIGN FOR BURNING OFFCAS
FROM COAL CASIFIERS
TASKC
EM CONTRACT «M»MMl
CE CONTRACT 9673
PROJECT 901001
SYSTEM CONCEPT NO:
SYSTEMDES.CNAT.ON:
PRIMARY FUEL TYPE:
G1
Supplementary Fired Botl.r Combined Cycl. Pl.nt
C1«an- lotf <»?««"" , low pressure. low BTU gas
clrc5
' 1005/1005
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COALGASIFIER3
TASKC
EPA CONTRACT 6M1-I Ml
CE CONTRACT 9671
PROJECT MIOOI
SYSTEM CONCEPT NO:
SYSTEM DESIGNATION:
PRIMARY FUEL TYPE:
C2
Supplementary Firtd Boiler Combined Cycle Plant
Clean, high temperature, low preaaur*. low BTU gaa
STEAM TURBINE CYCLE:
BOILER: 1-Controlled clrc; 2400p«U, IOOS/1003'F; 2.1x10 Ibn/hr main ateam (low
AIR PREHEATER:
BURNERS: CE-Type T (Tangential Firing)
TURBINE-GENERATOR: Currant dealgna; 300 MW (601)
GAS TURBINE CYCLE:
COMBUSTOR: Currant dealgn
TURBINE-GENERATOR: 3-aata; Currant daalgn; 200 HH (40X)
NOxCONTROL: Tangential Firing with Ovarflra Air, Flua Gaa Raclrc.
SYSTEM SKETCH:
Bolla
Primary Fual
Procaaa Staan
Stack
ADVANTAGES:
Vary low plant nat haat rat*
All ayatam conponanti of currant daalgn
Low NOx potential
Low operating coat
High turndown ratio
Good ayaten flexibility In following gaalfler thruput (via primary fuel atorage)
Procaaa (team produced for plant heating auxiliaries or power
Good eyitem flexibility in following load deoand (via primary fuel atoragi)
Potentially good ayaten reliability
DISADVANTAGES: _
High capital coat
Control ayatem complexity
199
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL C.ASIFIERS
TASKC
EPA CONTRACT 684M 361
CE CONTRACT 9673
PROJECT 901001
SYSTEM CONCEPT NO: C3
SYSTEM DESIGNATION: Supplementary Fired Boiler Combined Cycle Plant
PRIMARY FUEL TYPE: Clean, low temperature, high pressure, low BTU gee.
*llA|SJ!ilISOTCVCLEi-Controll*l eirc; 2400 psia, 1005/1005'F; 2.1xl06 lb»/hr main .tea. flow
AIR PREHEATER:
BURNERS: CE-Typ« T (Tangential Firing)
TURBINE-GENERATOR: Currant designs; 300 MH (JOX)
GAS TURBINE CYCLE:
COMBUSTOR: Current design
TURBINE-GENERATOR: 3-sets; Current design; 200 MW (40X)
NOxCONTROL: Tangential Firing with Ovarfire Air, Flue Gas Reclrc.
SYSTEM SKETCH:
Process Steam
*_=-
Primary
Stack
ADVANTAGES: _ Very low plant net heat rate
- All system components of current design
- Low NOx potential
- Low operating cost
o lyiteTlflexibility in following gasif ler thruput (via primary fuel storage)
Process steam produced for plant heatini? auxiliaries or power
Good system flexibility in following load demand (via primary fuel storage)
Potentially good system reliability
DISADVANTAGES:
High capital cost
Control system complexity
200
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT MUM 361
Ot CONTRACT M7J
PROJECT 901001
SYSTEM CONCEPT NO: C4
SYSTEM DESIGNATION: Supplementary Fired Boiler Combined Cycle PUnt
PRIMARY FUEL TYPE: Clemn. high Ce.pei.ture. high pret.ure, low BID gas
STEAM TURBINE CYCLE:
BOILER: 1-Controlled clrc; 2400psla, 1005/1005'P; 2.1x10* lb«/hr main .tern, flow
AIR PREHEATER: '
BURNERS: HI Temp. Sulzer or New Technology
TURBINE-GENERATOR: Current designs; 300 HW (60ZX
GAS TURBINE CYCLE:
NOx CONTROL:
3-sets; Current design; 200 MU (40Z)
Flue Gas Reclrc.
SYSTEM SKETCH:
Stack
ADVANTAGES:
Very low plsnt net heat rate
Low operating cost
High turndown ratio
. .llffig" load d— ' (vl'
DISADVANTAGES:
- Burner development may be required
- Very Ugh NOx potential
- Control system complexity
- High cspltsl cost
201
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT M-02-U4I
CE CONTRACT 9673
PROJECT
SYSTEM CONCEPT NO: G5
SYSTEM DESIGNATION: Supplementary Fired Boiler Combined Cycle Plant
PRIMARY FUEL TYPE: Clean, high temperature, high pressure, low BTO gas
1-Controlled clrc; 2400 p«la, 1005/1005*Fj 2.1xl06 Ibm/hr nain iteaa flow
. AtRPREHEATER:
WTBNERS CE-Type T (Tangential Firing)
TURBINE^CENERATOR: Current design.; 300 MW (60X)
GAS TURBINE CYCLE:
COMBUSTOR: Current design
TURBINE-GENERATOR: 3-sets; Current design ; 200 MH (40X)
NOx CONTROL: Tangential Firing with Overflre Air. Flue CM Reclrc.
SYSTEM SKETCH:
Process Steao
Stack
ADVANTAGES: - Very low plant net heat rate
- All system components of current design
- Low NOx potential
- Low operating cost
- High turndown ratio
- Good system flexibility in following gaslfler thruput (via primary fuel storage)
- Process steam produced for plant heating, auxiliaries or power
- Good system flexibility in following load demand (via primary fuel storage)
- Potentially good system reliability
DISADVANTAGES:
High capital cost
Control system complexity
202
-------
FURNACE DESIGN FOR BURNING OFFCAS
FROM COAL CASIFICRS
TASKC
EPA CONTRACT M4MMI
d CONTRACT 9*73
HtOJBCT MIOOI
SYSTEM CONCEPT NO: HI
SYSTEM DESIGNATION: Supercharged Bollar Combined Cycle
PRIMARY FUEL TYPE: Clean, low temperature, high preaaure, low BTU gaa
STEAM TURBINE CYCLE: . '
BOILER: 1-Controllad circs 2400 pa la, 100S/100ST} 2.8xl06 Ibm/hr main steam flow
AIR PREHEATER:
BURNERS: CB-Type T (Tangential Firing)
TURBINE-GENERATOR: Current dealgna; 400 MW (BOX)
GAS TURBINE CYCLE:
NOx CONTROL:
SYSTEM SKETCH:
Curr«nt d«*lgni; 100 MW (20X)
Ttngtntlml Firing with Ov.rflr. Air, Flu* GM Rrclre.
Stack
ADVANTAGES:
DISADVANTAGES:
• V«ry low plant n«t haat rat*
• Low NOx potantlal
' Low operating coat
• High turndown ratio
• Good ayttan flaxiblllty In following gaalflar thruput (via primary fual atoraga)
• Addition of procaaa ataan bollar or Gaa turblna tall-and poaalbla
1 Low bollar alia and haatlng aurfaea raqulrananta
• Good ayatam flaxlbllity in following load daaand (via prUury fual atoraga or
aacondary fual capability
• Fotmtlally good ayatan reliability
> Bollar development required
• High capital coat
• Control ayaten complexity
203
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL CASIFtERS
TASKC
EPA CONTRACT 6842-lttl
CE CONTRACT M73
PROJECT 901001
SYSTEM CONCEPT NO: H2
SYSTEM DESIGNATION: Supercharged Boiler Combined Cycl«
PRIMARY FUEL TYPE: Clean, high temperature, high pressure, low BTU gas
STEAM TURBINE CYCLE:
BOILER: 1-Controlled clrc; 2400 psia, IOOJ/1005'F; 2.8x10* lb»/hr main .team flow
AIR PREHEATER:
BURNERS: CE-Type T (Tangential Firing)
TURBINE-GENERATOR: Currant daslgnii 400 MW .(801)
GAS TURBINE CYCLE:
COMBUSTOR: Current design
TURBINE-GENERATOR: 2-«eta; Current deilgns; 100 MW (20X)
NOx CONTROL: Tangential Firing with Overfire Air; Flue Ga« Raclrc.
Primary Fuel
SYSTEM SKETCH:
lailtr
Stack
Process Steam
ADVANTAGES: - Very low plant net heat rat*
Low NOx potential
> Low operating cost
• High turndown ratio
• Good system flexibility in following gaslfler thruput (via primary fuel storage)
> Process steam produced for plant heating auxiliaries or power
• Addition of process steam boiler or gas turbine tall-end possible
> Good system flexibility in following load demand (via primary fuel storage)
• Potentially good system reliability
DISADVANTAGES:
Boiler development required
Control system complexity
High capital cost
204
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT 684MMI
CE CONTRACT 967)
PROJECT 901001
SYSTEM CONCEPT NO: H3
SYSTEM DESIGNATION: Supercharged Boiler Combined Cycle
PRIMARY FUEL TYPE: Clean, low temperature, low praeeure, low BTU gee
STEAM TURBINE CYCLE:
BOILER: 1-Controlled ctrc; 2400 p«la, 1005/1005'F; 2.8xl06 Um/hr
AIR PREHCATER:
BURNERS: CB-Type T (Tangential Firing)
TURBINE-GENERATOR: Current designs; 400 MW (801)
••In (team flow
GAS TURBINE CYCLE:
OOMBUSTOR: Current design
TURBINE-GENERATOR: 2-«eta; Current deilgni; 100 MW (201)
NO* CONTROL:
SYSTEM SKETCH:
Tangent**! Firing; with Overfire Air,; Flue Gae Reclrc.
Bailor
Stack
ADVANTAGES:
DISADVANTAGES:
• Very low plant net heat rate
• Low NOx potential
• Low operating coit
• High turndown ratio
flexlbllity ln Allowing 8aelf ler thruput Cvla primary fuel .tor.g.)
turbiM tall-nd po"lb:-
Cvta
load
• Boiler development required
• High capital coet
Control system complexity
205
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT 684)2-1361
CE CONTRACT 9673
PROJECT 901001
Primary tu»l
SYSTEM CONCEPT NO: H4
SYSTEM DESIGNATION: Supercharged Boiler Combined Cycle
PRIMARY FUEL TYPE: Clean, high temperature, low pressure, low BID gae
STEAM TURBINE CYCLE:
BOILER: 1-Controlled clrc; 2400 psla, 1005/1005*F; 2.8x10* Ibm/hr; main steam flow
AIRPREHEATER:
BURNERS: CE-Typ« T (Tangential Firing)
TURBINE-GENERATOR: Current designs; 400 MH (80*)
CAS TURBINE CYCLE:
COMBUSTOR: Current design
TURBINE-GENERATOR: 2-sets; Current designs; 100MW (20Z)
NOx CONTROL: Tangential Firing with Overflre Air; Flue Gas Reclrc.
SYSTEM SKETCH:
Boiler
Stack
Proeaee Steaa
ADVANTAGES:
DISADVANTAGES:
Very low plant net heat rate
Low NOx potential
Low operating cost
High turndown ratio
Good system flexibility in following gaslfier thruput (via primary fuel storage)
Process steam produced for plant heating auxiliaries or power
Addition of process steam boiler on boiler or gas turbine tall-end possible
Low beiler slzt and heating surface requirements
Good system flexibility In following load demand (via primary fuel storage)
Potentially good system reliability
Boiler development required
High capital coat
Control system complexity
206
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GAS1FIERS
TASKC
ETA CONTRACT 6M2-IMI
CE CONTRACT W3
PROJECT 901001
SYSTEM CONCEPT NO: H
SYSTEM DESIGNATION: Oj Blown Supercharged Coabined Cycle Plant
PRIMARY FUEL TYPE: Clw, low tMfmntm> hlgh pr.Mur.t lflw ^ g.§
STEAM TURBINE CYCtE:
1-Controll«<1
BURNERS: N«w Tvchnology
TWUINE-GENERATOR: Curr.nt d.itgn. 400 MW (80X)
GAS TURBINE CYCLE:
COMBUSTOR:
TURBINE-GENERATOR: 2...t.. Curtfnt
f,. IOOJ/IOOS-FJ 2.8x10* ib./hr
NOx CONTROL:
SYSTEM SKETCH:
blown
_Eoilei
ADVANTAGES:
m »—T1 n
.?"«!.
Sock
- Very low plant net heat rate
- Vary low NOx potential
- Low operating coat
- Good ayaten flexilllity in following g«»i.er snr
- Low boiler alee and heating surface requlrementa
- Good ayatea flexibility in following load demand (vie primary fuel atora«e)
thruput (via Prt«.ry fu«l ator.g.)
nSADVANTAGES:
- Burner development required
- Boiler development required
" High capital coat
- unknown turndown ratio
- Control ayaten complexity
- Applicability limited to 0. blown gaalfler
- Unknown eyatem reliability
- Hi coat of 0
- Applicability Halted to H2-free low BTO gaa
207
-------
FURNACE DESIGN FOR BURNING OPFGAS
FROM COAL CASIFIERS
TASKC
EPA CONTRACT M-02-IMI
CE CONTRACT »673
PROJECT 901001
SYSTEM CONCEPT NO:
SYSTEM DESIGNATION:
PRIMARY FUEL TYPE:
12
0 Blown Supercharged Steaa Power Plant
Clean, low temperature;, low pressure, low BTU gat
1-Controlled circ.; 2400 pala, '1005/1005'P; 2.8xl06 Ibm/hr main steam flow
STEAM TURBINE CYCLE:
BOILER:
AIRPREHEATER:
BURNERS: New Technology
TURBINE-GENERATOR: Current design; 400 MW C80X)
CAS TURBINE CYCLE:
COMBUSTOR:
TURBINE-GENERATOR:
2-ieta; Current designs; 100 Hw C20X)
NO* CONTROL:
SYSTEM SKETCH:
0. blown
Primary Fuel
U-*CompT
Boiler
[Stack
ADVANTAGES: - Very low plant net heat rate
- Very low NOx potential
- Low operating cost
- Good system flexibility In following gasffier thruput (via primary fuel storage)
- Low boiler sice and heating surface requirements
- Good system flexibility in following load demand (.via primary fuel storage)
DISADVANTAGES: . Burner development required
- Boiler development required
- High capital coat
- Unknown turndown ratio
- Control system complexity
- Poor system flexibility in following load demand (via primary fuel storage)
- Applicability limited to 0. blosn gasifler
- Unknown •ystem reliability
- Hi cost of 0.
- Applicability limited to N -free low BTU gas
208
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT M42-IMI
CC CONTRACT M73
PROJECT 901001
SYSTEM CONCEPT NO: 13
SYSTEM DESIGNATION: QJ Blown Sup.rch.rgrt St... Po«.r Pl.nt
PRIMARY FUELTYPE: Cl..n. high temp.r.tur., low pr...ur.. low BTU (M
STEAM TURBINE CYCLE:
BURNERS: New Technology
TURBINE-GENERATOR: Curr.nt d.eignj 400 HW (80%)
GAS TURBINE CYCLE:
COM1USTOR:
TURBINE-GENERATOR: 2-..t.; Curr.nt d..lgn.: 100 MH (20X)
NOx CONTROL: Oj Blown
SYSTEM SKETCH:
> "«/1003-F, 2.8,106 IWhr ..In .t.« flow
Prljury Fu.l
ProcMt St«u
Stack
ADVANTAGES:
- V«ry low plant net heat rat.
- Low operating cost
" flS"!"!,-""? Product<1 '« Pl«nt h.attng .uxlli.rl.. or g.alfl.r
- low boll.r six. and h.atlng .urfac. r.qulrm.nt8
WSADVANTACES; .
- Botl.r development required
- V.ry high NOx potential
- High capital co.t
- Unknown turndown ratio
- Control syitem complexity
- Poor iy.tem flexibility In following priury fuel .upnly
- Poor .y.tem flexibility In followlnl load Xmand P
- Unknown system reliability
- Applicability limited to 0 blown ga.ifler
- HI cost of 0. 2 *
- Applicability lialted to Nj-fr.e low BSU gas.
209
-------
FURNACE DESIGN FOR BURNING OFFGAS
FROM COAL GASIFIERS
TASKC
EPA CONTRACT WWM-I36I
CE CONTRACT 967J
PROJECT 901001
SYSTEM CONCEPT NO: 14
SYSTEM DESIGNATION: 0, Blown Supercharged Combined Cycle Plant
PRIMARY FUEL TYPE: Clean, high temperature, high pressure, low BTU gas
STEAM TURBINE CYCLE: 6
BOILER: l-Controlled circ.; 2400 psla, 1005/1005*F; 2.8x10 Ibm/hr main steam flow
AIRPREHEATER:
BURNERS: New Technology
TURBINE-GENERATOR: Current design; 400 MM (80Z)
GAS TURBINE CYCLE:
COMBUSTOR:
TURBINE-GENERATOR: 2-sets; Current designs; 100 HW (20Z)
NOx CONTROL: 0 Blown
SYSTEM SKETCH:
Boiler
Primary Fuel
Stack
Process Steam
ADVANTAGES:
- Very low plant net heat rate
** Low operating cost
- Process steam produced for plant heating auxiliaries or gaslfier
- Low boiler size and heating surface requirements
DISADVANTAGES: _ Burner development required
- Boiler development required
- Very high NOx potential
- High capital cost
- Unknown turndown ratio
- Control system complexity
- Poor system flexibility In following primary fuel supply
- Poor system flexibility In,following load demand
- Unknown system reliability'
- Applicability limited to 02 blown gaslfier
- Hi cost of 0,
- Applicability limited to N2-frse lew BTU gas
210
-------
APPENDIX B
THE PREDICTION OF FURNACE PERFORMANCE FOR
TANGENTIALLY-FIRED UTILITY BOILERS
Two requirements necessitated the development of an engineering model
for predicting furnace performance in water-wall boilers. These are:
• The selection of the proper grade of steel for the water-
wall requires a knowledge of axial heat absorption rate.
. The^furnace outlet temperature must be known within about
-50 F in order to correctly size superheaters, reheaters and
economizers.
The assessment of the performance of furnaces fired with low Btu gas described
in Section 3 utilized Combustion Engineering's "lower furnace model". The
development of this program was described by Bueters et al(20> and the region
modeled by this program is shown in Figure B-l.
The furnace is divided into a number of horizontal strips as shown in
Figure B-2. Heat absorption to the wall and the gas temperature in the exit
plane are given by the simultaneous solution of an energy balance on each
strip. This energy balance is given by
where
absorbed by Sl1ce 1 and emi'tted by all other
q^ = sensible heat entering the itn slice
Qj = heat released in the ith slice
q1o = axial emission of the itn slice
a
qio = sensible heat leaving the ith snce
qior = Cation from the ith slice to its bounding wall
211
-------
Upper Furnace Program
Furnace Outlet Plane
Lower Furnace Program
Steam Sen. Program
Figure B-l. Furnace Schematic Showing Volume Modeled by
Lower Furnace Program
212
-------
Furnace Outlet Plane
1 = n
ft
"^
Elevation Z = 0
Etc.
o
r-j
0)
10
. v
i
•o
M
U
U -
Lower Furnace Program Sample Gas
Circulation Pattern
Figure B-2-. Lower Furnace Program Reclrculatlng Flow Model
213
-------
The solution of this energy balance requires a description of:
the furnace flow pattern,
the heat release distribution,
the radiative properties of the furnace gases, and
the radiation exchange.
The Furnace Flow Model
The flow in a tangenti ally-fired utility boiler is characterized by a
"donut" shaped fireball and the model that has been adopted is shown in
Figure B-2. The furnace is divided into three zones: a recirculation zone,
a heat release zone, and a plug flow zone. The flow originates in the heat
release zone, circulates through the hopper region and peels off to rejoin
the exit stream. The products generated in each strip are assumed to be
directly proportional to the heat released in that strip.
The Heat Release Distribution
The lower furnace source condition is prescribed by two empirical
parameters :
t The release zone height, RZH, (heat release is assumed uniform
over this region), and
• The release zone centroid RZC which locates the heat release
zone with respect to the furnace bottom.
In each strip the heat is assumed to be liberated instantaneously with
the fuel forming C02 and H20. Thus each strip consists of fully-mixed, fully-
burned combustion products.
Radiation Properties
Heat absorption rates calculated on the basis of carbon dioxide and water
vapor emissivities are too low. Gas emissivities calculated from charts are
modified by the "FE Operator" as follows:
where e^ is the combined emissivity of C02 and H20 for the appropriate
214
-------
partial pressures and path length and the wall-to-gas and gas-to-gas
absorptivities are similarly modified:
a = (FE - 1 + ah)FE .
Values of FE have been determined for gas, oil and coal-firing and include
corrections for soot radiation and convection.
Radiation Exchange
The "lower furnace program" allows for axial exchange between strips,
but each strip only radiates to its own bounding surface wall.
Bueters, Cogoli and Habelt^20' have described the development of the
"lower furnace program" in detail and show comparisons between measured and
calculated absorption rates and furnace exit temperatures.
215/216
-------
APPENDIX C
KINETIC MECHANISM OF NOx FORMATION IN LOW BTU GAS COMBUSTION
C.I INTRODUCTION
As the systems study progressed, it became increasingly obvious that
accurate estimates of NO formation in supercharged boiler-combined cycle
systems burning low Btu gas would not be possible. Consequently, it was
decided to approach the problem by constructing limit-case models. These
models are based upon the conjecture that chemical kinetics set the limit
of NO production, but than in practice, emissions are dictated by mass
transfer. Consequently, for the limit-case idealized mixing is assumed and
constraints are imposed upon the system (inlet conditions, heat loss rate,
residence time, exit temperature) corresponding to practical conditions. An
essential feature of this model is the availability of a kinetic mechanism
for NO formation.
Two factors make it necessary to synthesize a new kinetic mechanism
capable of describing NO formation in low Btu gas combustion:
1. The presence of hydrocarbon species. Most coal-derived fuel
gases will probably contain hydrocarbons (mainly methane), in
addition to carbon monoxide and hydrogen.
2. The presence of ammonia. The yield of NO from ammonia in flames
is strongly dependent upon the mixture equivalence ratio. In
fuel-lean mixtures almost 100 percent of the ammonia is converted
to NO, however, the conversion rate decreases significantly as
the mixture become more fuel-rich.
C.2 NO FORMATION IN GASEOUS SYSTEMS
Nitric oxide can be formed from two nitrogen sources during the combus-
tion of coal-derived fuel gases:
Molecular nitrogen producing thermal NO
Reduced nitrogen compounds (NH3, HCN) producing fuel NO.
Experimental observations have characterized the most significant features
of nitric oxide formation from both sources.
217
-------
C.2.1 Thermal NO
Considerable effort has been expended in recent years to establish
the kinetic mechanism which describes thermal NO formation in flames. In
the post-flame gases, in regions far removed from the reaction zone,
observed NO production rates can be explained on the basis of the Zeldovich
mechanism(1). However, in regions close to the flame front, NO production
rates exceed those predicted by the Zeldovich mechanism if equilibration of
oxygen atoms is assumed. Two explanations have been proposed to account for
this rapid formation of NO in, or close to, the flame front: non-equilibrium
of oxygen atoms (0 atom overshoot), and the attack of hydrocarbon radicals
on nitrogen compounds, producing CN or NH type radicals. Both explanations
are valid, but neither is exclusive. Oxygen atom overshoot does occur, but
there is irrefutable evidence that hydrocarbon fuel fragments rupture nitro-
gen molecules producing nitrogen intermediates.
Fenimore^ used the term "prompt" NO to describe the rapid formation of
thermal NO in the flame front, and postulated two likely reactions involving
nitrogen and fuel fragments which would produce intermediates which could
then be converted to NO:
CH + N2 -* CHN + N
C2 + N2 + CN + CN
The existence of nitrogen intermediates in hydrocarbon flame fronts has been
confirmed by De Soete^, Eberius^3', and Haynes^ '. These investigators
report the presence of both HCN and "NH^1 in hydrocarbon flames.
C.2.2 Fuel NO
Several studies have been conducted in which simple nitrogen compounds
were added to both premixed flames^1'5', and well-stirred reactors( ', in
order to investigate the effect of temperature, pressure, fuel type, nitrogen,
additive, additive concentration and equivalence ratio on fuel NO formation.
Summarizing the most significant results of these investigations it appears
that:
Reaction zone equivalence ratio is the most important parameter
controlling fuel NO formation.
218
-------
Compounds containing NH bonds tend to give higher conversion
rates than compounds containing CN bonds in lean mixtures.
However, in rich mixtures the conversion appears to be almost
independent of the type of nitrogen compound.
;
Conversion decreases as the concentration of nitrogen additive
increases, and this effect is most apparent for fuel-rich
mixtures.
Flame temperature always has a relatively small influence on the
formation of fuel NO, but it is enhanced for rich mixtures and
additives with CN bonds.
Fuel NO formation appears to be higher in rich hydrogen flames
than in rich hydrocarbon flames.
Pressure does not appear to have a very significant effect upon
fuel NO formation.
It appears that the formation of fuel NO can be characterized by a rather
simplistic view. The initial fuel nitrogen compound breaks down in the
reaction zone producing an intermediate nitrogen compound XN. This XN inter-
mediate is then subject to two competitive reaction paths:
Path A XN + R'O •»• NO +....'.
Path B XN + R"N •»• Ng + . . . .
Path A produces NO and is the faster of the two paths in lean mixtures.
However, Path B dominates in fuel-rich mixtures, allowing the formation of
molecular nitrogen. It can be seen that some similarity exists between the
formation of fuel NO and the formation of prompt thermal NO by fuel fragments.
XN intermediaries are produced by fuel fragments which are free to react
through either Path A or Path B.
Path B allows the reduction of an intermediate compound XN to nitrogen
It is also known that NO itself behaves as a fuel nitrogen compound. Thus,*NO
can be reduced by R»N (Path B), or it can be converted to a nitrogen-containing
219
-------
intermediate. One possibility for the reduction of NO is the reverse of the
first step of the Zeldovich mechanism
NO + N + N2 + 0
Myerson' ' has indicated another route which will eventually allow the forma-
tion of N2 from NO. He observed that hydrocarbon addition .to simulated com-
bustion products containing NO and oxygen caused a significant reduction in
NO which he attributed to the occurrence of such reactions as
CH + NO ->• HCN + 0
CH + NO -»• HCO + N
C.3 SYNTHESIS OF A REACTION SET
The kinetic mechanism synthesized for use in the estimation of NO forma-
tion in combustors burning lowxBtu gas should include the folTowing features:
1. The ability to predict thermal NO formation and to take account
not only of the Zeldovich mechanism, but also the formation of
nitrogen intermediates produced by the reaction of fuel fragments.
2. The ability to be able to predict correctly the conversion of
ammonia to NO and the destruction of NO by fuel-rich mixtures.
3. The reaction set should not be restricted to carbon monoxide and
hydrogen since many coal-derived fuels contain hydrocarbons.
Reaction involving higher hydrocarbons than methane were not to
be included.
The majority of the reactions which were included had been listed by
(8)
Engleman^ . The reaction set contains 27 species and 100 reactions. The
total number of reactions are listed in Table C-l in four groups:
reactions describing the CH, oxidation mechanism
reactions involving N, NO, NpO or N02
reactions of ammonia and intermediates
reactions between carbon, hydrogen and nitrogen.
The reaction rate is expressed by
-E/RT
kf = AT
220
-------
Table C-l
kf = AT"N exp (-E/RT)
REACTIONS DESCRIBING CH4 OXIDATION MECHANISM
A
(cc, mole,
CH4 +
CH30 +
CH20 +
CHO +
co2 +
H2 +
H20 +
H +
H +
CH +
CH +
CH +
CH3 +
CH2 +
CH3 +
CH2 +
CH2 +
CH4 +
CH4 +
CH4 +
CH3 +
CH3 +
M
M
M
M
M
M
M
0
°2
CH4
CH3
HO
OH
H,
CH20
CH20
H
OH
H
0
0
°2
• CH3 +
= CH20 +
= CHO +
- CO +
= CO +
= H +
= HO +
+ M
+ M
- CH2 -f
= CH2 +
= CHO +
= CH2 +
+ CH, +
- CH4 +
- CH3 *
= CH +
' CH3 +
•• CH3 +
' CH3 +
- ,CH20 +
= CH30 +
H + M
H + M
H + M
H + M
0 + M
H + M
H + M
OH + M
H02 + M
CH3
CH2
H
H20
H
CHO
CHO
H
H20
H2
HO
H
0
2.00
4.00
8.00
2.50
1,00
2.
3.
8.
1.
1.
3.
5.
1.
3.
4.
2.
3.
3.
5.
1.
2.
2.
00
00
00
50
00
20
00
00
00
00
00
00
00
00
00
00
50
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
sec) N
17
40
33
20
15
14
15 '
15
15
12
12
11
12
12
10
11
11
13
10
10
12
9
0.0
7.5
4.5
1.5
0.0
0.0
0.0
0.0
0.0
0.0
0.0
-0.5
-0.7
0.0
0.0
0.0
-0.7
0.0
-1.0
-1.0
-0.5
-1.0
E
(kcal)
88.0
22
87
16
100
96
105
0
1
17
8
10
2
7
0
6
26
5
10
8
-0
28
.6
.0
.8
.0
.0
.0
.0
.0
.1
.0
.0
.0
.0
.0
.5
.0
.0
.0
.0
.3
.5
Ref,
8
10
10
10
8
8
11
8
8
10
10
8
8
10
8
10
11
8
11
11
11
9
221
-------
Table C-l. (Continued)
REACTIONS DESCRIBING CH4 OXIDATION MECHANISM (Cont.)
A
(cc, mole,
CH, •
3
CH, •
3
CH2 -
CHO -
CH20 H
CHO H
CH20 H
CHO H
CH
CH2 -
CO
CO
CO
H00
2
H00
2
HO
HO
H
HO
H
i- HO,
2
i- 00
2
h °2
i- 0
i- HO
I- HO
i- 0
H H
h °2
H HO
^ HO
f °2
i, LJO
nUo
* H
* HO
+ HO
f H2
f H02
f 0
f HO
= CH
4
= CH90
2
= CH20
= CO
= CHO
= CO
= CHO
= CO
= CHO
= CH
= co2
= co2
= CHO
= 09
2
= H90
£.
= H20
= H
= HO
= H
- H2
REACTIONS INVOLVING N,
N20 •
N
f M
f M
' N2
= N
+ 09
d
+ HO
+ 0
+ HO
+ H20
+ H20
+ HO
+ H2
+ 0
+ H20
+ H
+ 0
+ °2
+ H9
2
+ 0,
2
+ 0
+ H20
+ HO
+ °2
+ 0
NO, N20, OR N02
+ 0 + M
+ N + M
1.00
3.00
5.00
3.00
1.00
3.00
2.00
3.00
5.00
5.00
5.60
1.00
3.00
2.50
5.00
6.00
2.50
2.50
2.50
8.00
1.00
4.00
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
sec) N
11
13
11
11
14
10
11
10
11
11
11
13
12
13
13
12
13
14
13
9
14
21
-0.
0.
-0.
-1.
0.
-1.
-1.
-1.
-0.
-0.
o.
0.
0.
0.
0.
0.
0.
5
0
5
0
0
0
0
0
5
5
0
0
0
0
0
0
0
0.0
0.
-1.
0
0
0.0
1.6
E
(kcal)
6.
30.
7.
0.
0.
0.
4.
0.
6.
6.
0
0
0
5
0
0
4
0
0
0
1.08
60.
37.
0.
0.
1.
0
1
7
0
0
5.2
1.9
0.0
7.0
50.0
225.0
Ref.
8
8
8
10
10
10
10
10
8
8
10
8
10
8
8
8
8
8
8
8-
8
8
222
-------
Table C-l. (Continued)
REACTIONS INVOLVING N,
N02
°2
HO
N
N
NO
H
N20
N20
N
N
NO
N02
NO
NO
H
H
HNO
HNO
HNO
HNO
+ M
+ N
+ N
+ NO
+ °2
+ N02
+ N20
+ 0
+ 0
+ N02
+ N02
+ N20
+ 0
+ H02
+ HO
* NO
+ HNO
+ NO
+ HNO
+ 0
+ HO
= NO
+ M
= H
' N2
= NO
= N20
= HO
= NO
' N2
= NO
•• N2
= N02
= NO
= N02
= H02
+ M
= H2
= HO
" N2°
= HO
= H20
NO, N20 OR N02 (Cont.)
A
(cc, mole,
+ 0 + M
= N02 + M
+ NO
+ 0
+ 0
* °2
* N2
+ NO
+ °2
+ NO
+ °2
+ N2
+ °2
+ HO
+ N
= HNO + M
+ NO
+ N20
+ H20
+ NO
+ NO
1.00
1.00
6.00
3.10
6.00
1
8
1
1
4
1
2
1
5
5
2
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
1.00
2.00
1.00
5.00
1.00
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
sec) N
16
19
11
13
9
12
13
14
14
12
12
12
13
11
10
16
13
12
10
11
12
0.0
1.0
-0.5
0.0
-1.0
0.0
0.
0.
0.
0.
0.
0.
0.
-0.
-0.
0.
0.
0.
-0.
-0.
0
0
0
0
0
0
0
5
5
0
0
0
5
5
0.0
•—•— •— -^"^— i
(kcal)
65.0
0.0
8.0
0.334
6.3
60.0
15.0
28.0
28.0
0.0
0.0
40.0
1.0
3.52
96.5
0.0
2.5
26.0
41.55
0.0
1.0
^^MMMMB
Ref
8
9
10
8
8
8
8
8
8
8
8
9
8
9
9
9
8
8
9
8
10
223
-------
Table C-l. (Continued)
REACTIONS OF AMMONIA AND
NH3 4
NH3 4
NH3 4
NH3 4
NH2 4
NH2 4
NH2 4
NH2 4
NH2 4
NH2 4
NH 4
NH 4
NH 4
NH +
NH 4
NH 4
NH +
H +
NH 4
H 4
REACTIONS
CHN 4
CH2 4
°2 -
0
H
HO
°2 '
0
H
HO
NO
NH2 =
0
NO
H
NH
HO
0
N
N20 =
HO
HNO =
BETWEEN
N
N2 =
NH2 4
NH2 4
NH2 4
NH2 +
NH +
NH +
NH +
NH 4
N, 4
NH3 4
N +
N2 4
N 4
N2 4
NO 4
NO 4
H 4
H 4
H20 4
NH 4
CARBON,
CH 4
CHN 4
INTERMEDIATES
H02
HO
H2
H20
H02
HO
H,
H£0
H20
NH
HO
HO
H2
H,
H,
H
N,
NO
N
OH
HYDROGEN
N,
NH
(cc
5
8
1
4
1
9
1
3
1
1
1
2
1
3
5
5
6
1
5
2
A
, mole, sec) N
.00 E
.20 E
.90 E
.00 E
.00 E
.20 E
.40 E
.00 E
.20 E
.70 E
.70 E
.40 E
.00 E
.60 E
.00 E
.00 E
.00 E
.00 E
.00 E
.00 E
4 H
4 11
4 11
4 10
4 13
4 H
4 11
4 10
4 10
4 H
4 10
4 12
4 12
4 11
4 11
4 H
4 H
4 11
4 H
4 H
-0
-0
-0
-0
0
-0
-0
-0
0
-0
-0
0
-0
-0
-0
-0
-0
-0
-0
-0
.5
.5
.67
.68
.0
.5
.67
.68
.0
.63
.7
.0
.68
.55
.5
.5
.5
.5
.5
.5
E
(kcal)
56.
0.
3.
1.
50.
0.
4.
1.
0.
3.
0.
0.
1.
1.
2.
5.
0.
30.
2.
23.
0
0
4
1
5
0
3
3
0
6
1
0
9
9
0
0
0
0
0
0
Ref.
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
8
8
8
8
8
AND NITROGEN
5
1
.00 E
.00 E
4 H
4 14
0
0
.0
.0
16.
55.
0
0
10
10
224
-------
Table C-l. (Concluded)
REACTIONS BETWEEN CARBON, HYDROGEN AND NITROGEN (Cont.)
A
(cc, mole,
CHO
CHN
CN
CN
CN
CN
CHN •
co2 •
CH20 H
CH30 H
CO H
CH3 H
CHO H
+ N
+ 0
+ NH
* NO
f °2
*• 0
*• OH
H N
i- NO
i- NO
i- HNO
^ HNO
- NO
- CH
= CH
= CH
= CO
= CO
= CO
= CN
= CO
= HNO
= HNO
= co2
= CH4
= CO
+ NO
+ NO
+ N2
+ N?
+ NO
+ N
+ H20
+ NO
+ CHO
+ CH20
+ NH
+ NO
+ HNO
1
1
1
3
3
5
2
2
5
5
5
5
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
2.00
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
E +
sec) N
14
13
14
11
11
11
11
11
11
11
8
11
11
0
0
0
0
0
-0
-0
-0
-0
-0
-0
-0
.0
.0
.0
.0
.0
.5
.6
.5
.5
.5
.5
.5
-0.5
i^ — i ^Wl
E
(kcalj
48.6
72.0
40.0
0.0
0.0
9.63
5.0
25.0
27.0
4.23
7.0
0.0
2.0
i^M^HM
Ref
10
9
10
8
8
9
8
8
9
9
9
8
8
225
-------
where the units are note, cm3, sec, °K and kcal. The values of the constants
used originally, A, N, and E, were those listed by Engleman^ ' or estimated
by the method of Tunder, et an9'. A recent report by Benson, et ar ' lists
estimated Arrhenius parameters for several of the reactions included in the
synthesized set. All of the calculations carried out used the rates given in
Table C-l, all of which are within the estimates given by Benson, et ar '.
No attempt was made to screen the reaction set.
C.4 EVALUATION OF REACTION SET AGAINST STIRRED REACTOR DATA
Having assembled a reaction set which seemingly included all the key
reactions reported in the literature concerning NO formation and destruction,
the reaction set was used in the KAP program in the perfectly stirred reactor
mode, and an attempt was made to calculate the stirred reactor data reported
by Bartok, et ar . This data had been used previously by Waldman et ar /
for CH.-02-N2 mixtures. The experimental investigation also included the
addition of NH~ and NO to the stirred reactor. A comparison between the
measured and predicted NO concentrations is presented in Figures C-l, -2 and
-3. The stirred reactor calculations were carried out with a heat loss of
0.028/sec°K (see reference 11 for further discussion).
The 33 reaction set proposed by Waldman et ar ' was tested with the
methane/air stirred reactor data with the stated heat loss and found to be
unsatisfactory. Predicted NO concentrations were almost one order of magni-
tude below the experimental data. Also, for equivalence ratios greater than
1.33 the calculated NO concentrations were independent of equivalence ratio.
The mechanism was inadequate because it did not include reactions of the
Path B type which allow the formation of N2 from XN intermediates.
226
-------
ro
ro
-4
Predicted 2.0 msec
Nominal Residence Time
100 150
PERCENT STOICHIOMETRIC AIR
Figure C-l. Comparison of Measured' ' and Predicted NO Concentration
at the Exit of a Stirred Reactor
200
-------
ro
ro
oo
§ 100
H
0-
5
0
§
H
« 50
H
W
O
0
i
i i i i i i i
O '20° PPm ADDITION NO
Z\1300 ppm ADDITION NO
+ PREDICTED
I i i i i i
- £* A o A ;
-
- 4^
^^^/»
! A '.
1 1 1 1 1 1 1 1 1 1 1 1 1 1
JO 100- ISO 20
PERCENT STOICHIOMETRIC AIR
Figure C-2. Measured' ' and Predicted NO Retention in a Stirred Reactor
-------
ro
ro
vo
I
S
O
i— i
eo
s
I
O
w
I
50
50
1
1
O 1300 ppm ADDITION NH3
+ PREDICTED
i i i i i i
8
o * • o
0*
§* ;
j* -
i i i 1 i i i i 1 i r i i
100 150
PERCENT STOICHIOMETRIC AIR
200
Figure C-3. Measured^6' and Predicted Conversion of NH3 to NOX in
a Stirred Reactor
-------
REFERENCES
1. De Soete, G.G., "La Formation Des Oxydes D'Azote dans la Zone
D'Oxydation des Flammes d'Hydrocarbures". Compte rendu final des
travaux Contrat No. 73-56 avec le Ministere de la Protection de la
Nature et de 1'Envlronnement. Institut Francais de Petrole, June 1975.
2. Fenimore, C. P., Thirteenth Symposium (International) on Combustion,
The Combustion Institute, Pittsburgh, PA, 1971, p 373
3. Eberius, K. H., Comments, Fourteenth Symposium (International) on
Combustion, The Combustion Institute, Pittsburgh, PA, 1973, p 775.
4. Haynes, B.S., Combustion and Flame, Vol. 28, p 113 (1977).
5. Sarofim, A. F., Williams, G. C., Modell, M. and Slater, S.M., "Conver-
sion of Fuel Nitrogen to Nitric Oxide in Premixed and Diffusion Flames".
Paper presented at the AIChE 66th Annual Meeting, Philadelphia, 1973.
6. Bartok, W., Engleman, V.S. and del Valle, E.G., "Laboratory Studies
and Mathematical Modeling of N(L Formation in Combustion Processes".
Exxon Research and Engineering Company Report No. GRU-3GNOS-71, EPA
No. APTD1168, NTIS No. PB 211-480, 1972.
7. Myerson, A.L., Fifteenth Symposium (International) on Combustion,
p. 1085, The Combustion Institute, 1975.
8. Engleman, V.S., "Survey and Evaluation of Kinetic Data on Reactions
in Methane/Air Combustion", EPA-600/1-76-003, NTIS No. PB 248-139/AS,
January 1976.
9. Tunder, R., Mayer, S., Cooke, E. and Shieler, L., Aerospace Corp.
Report No. TR-001 (9210-02)-!, 1967.
10. Benson, S.W., Golden, D.M., Lawrence, R.W., Shaw, R., and Woolfolk, R.W.
Final Report EPA Grant No. R-800798, 1975.
11. Waldman, C.G., Wilson, Jr., R.P. and Maloney, K.L., "Kinetic Mechanism
of Methane/Air Combustion with Pollutant Formation", EPA-650/2-74-045,
1974.
230
-------
TECHNICAL REPORT DATA
(I'lcase rcqd Imunictknis on the n-ivrsc before completing)
1. Rl.PORT NO.
EPA-600/2-77-235
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Low NOx Combustion Concepts for Advanced Power
Generation Systems Firing Low-Btu Gas
5. REPORT DATE
November 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
T.J.Tyson, M.P.Heap, C.J.Kau, B.A.Folsom, and
*N. D. Brown
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Energy and Environmental Research Corp.
8001 Irvine Boulevard
Santa Ana, California 92705
10. PROGRAM ELEMENT NO.
1AB013; ROAP 21ADD-BA
11. CONTRACT/GRANT NO.
68-02-1361
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final: 6/73-3/77
14. SPONSORING AGENCY CODE
EPA/600/13
,5. SUPPLEMENTARY NOTES jERL-RTP project officer for this report is G. Blair Martin, Mail
Drop 65, 919/541-2235. Coauthor Brown is with Combustion Engineering, 1000 Pros-
pect Hill Road. Windsor. CT 06095.
6>A The report gives results of an analysis of several advanced power genera-
ting concepts firing low-Btu gasified coal. A combined gas-turbine/steam-cycle power
plant with integrated gasifier was the most promising from fuel utilization and econ-
omic viewpoints. Two representative combined cycle systems were chosen for detai-
led NOx emission and analysis: an advanced-technology high-temperature gas turbine
with a waste heat boiler; and a supercharged boiler with a cur rent-technology gas tur-
bine. NOx emissions were investigated using a kinetic model, which was validated by
comparison with best available experimental data and then applied to idealized corn-
bus tor configurations. Calculations indicate that staged combustion involving rich
primary zones and controlled mixing secondary zones minimizes thermal NOx and
NOx produced from ammonia in the fuel gas. Minimum calculated NOx levels were:
150 ppm for the high temperature turbine, with a 0.45 equivalence ratio and 4000 ppm
of fuel ammonia; and 125 ppm for the supercharged boiler, with 5% excess air and 500
ppm of fuel ammonia. These results need to be verified experimentally, but they
show the potential for achieving NOx emissions within the Federal NSPS without
requiring ammonia removal from the fuel gas.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air Pollution
Combustion
Nitrogen Oxides
Electric Power Generation
Coal Gas
Gas Turbines
Ammonia
Air Pollution Control
Stationary Sources
13B
21B
07B
10A
21D
13G
13. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
237
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-t (9-73)
231
------- |