SEPA
United States
Environmental Protection
Agency
Industrial Environmental Research EPA-600/7-78-088
Laboratory June 1978
Research Triangle Park NC 27711
Fuel Gas
Environmental
Impact:
Final Report
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Off ice of Research and Development, U.S. Environmental Protec-
tion Agency, have been grouped into nine series. These nine broad categories were
established to facilitate further development and application of environmental tech-
nology. Elimination of traditional grouping was consciously planned to foster technology
transfer and a maximum interlace in related fields. The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECHNOLOGY
series. This series describes research performed to develop and demonstrate instrumen-
tation, equipment, and methodology to repair or prevent environmental degradation from
point and non-point sources of pollution. This work provides the new or improved tech-
nology required for the control and treatment of pollution sources to meet environmental
quality standards.
REVIEW NOTICE
This report has been reviewed by the U.S. Environmental
Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the
views and policy of the Agency, nor does mention of trade
names or commercial products constitute endorsement or
recommendation for use.
This document is available to the public through the National Technical" Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-78-088
June 1978
Fuel Gas
Environmental Impact
Final Report
by
F.L. Robson (UTRC), W.A. Blecher (UTRC), and V.B. May (Hittman Associates)
United Technologies Research Center
Silver Lane
East Hartford, Connecticut 06108
Contract No. 68-02-2179
Program Element No. EHE623A
EPA Project Officer: Thomas W. Petrie
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ABSTRACT
United Technologies Research Center continues to demonstrate the poten-
tial environmental and economic benefits of integrated coal gasification/gas
cleanup/combined gas and steam cycle power plants. Several technical problem
areas requiring investigation or further definition were left over from UTRC's
work on EPA Contract No. 68-02-1099, reported in EPA-600/2-76-153 (June 19T6)..
Refinements in plant operational characteristics lower heat rates and reduce
emissions from previous values. An expanded study of plant environmental intru-
sions includes a look at potentially hazardous trace elements. Comparisons
made of integrated plants using air-blown and oxygen-blown gasifiers favor
air-blown operation. Careful theoretical design of plants with low tempera-
ture sulfur cleanup reduces to marginal levels the performance and cost advan-
tages of plants with high temperature cleanup. If gasifier steam feed rates
are kept low in all but fixed bed types, choice of gasifier among other major
generic types is not critical to achieving attractive systems using low tem-
perature cleanup. Excessive thermal NOX emissions may be avoided by departing
from conventional combustor designs. Fuel NOX and participates remain as un-
solved problems with use of high temperature cleanup. Sulfur removal to very
low levels is possible with integrated systems, but cost rises rapidly as it
becomes necessary to remove most of the COS as well as the HoS.
-------
Fuel Gas Environmental Impact
TABLE OF CONTENTS
Abstract ii
Figures v
Tables ix
Conversion Factors xiv
Acknowledgment xv
1. Summary ..... 1
Introduction 1
Power Plant Components 3
Overall Power Plant 13
References 23
2. Introduction 24
References 27
3. Conclusions 28
4. Recommendations for Further Work 30
5. Overview of Gasification and Cleanup Processes 31
for Use with Combined-Cycle Power Plants
Introduction 31
Gasification 33
Cleanup Systems 40
References 54
6. Detailed Description of Selected Gasification 56
and Cleanup Processes
Introduction 56
Selection of Processes 56
Gasification Process Description 63
Desulfurization Process Descriptions 78
References 89
7. Combined Cycle Power Generation Systems 90
Introduction 90
General Parametric Analyses 92
111
-------
TABLE OF CONTENTS (Cont'd)
8. Integration Studies 100
Introduction 100
Thermal Integration '. 101
Selexol Operation 106
Steam to Coal Feed Ratio 114
Air-Blown BCR Gasifiers 116
Parametric Studies of the U-Gas/Conoco System 127
Gasifier Modeling 127
References 134
9. Power System Emissions 135
Introduction 135
.Definition for Goals of Cleanup Systems 137
Sulfur Emissions 140
Nitrogen Oxides 144
Particulates 177
References 180
10. Other Environmental Intrusions 183
Introduction 183
Wastes from Coal Preparation and Handling 186
Wastes from Gasification, Cleanup and Gas Utilization . 194
Residuals 209
Trace Elements 210
References 229
11. Performance and Cost of Integrated Systems 231
Introduction 231
Performance 231
System Cost Estimates 284
References 291
Appendices
A. Water Treatment and Reuse 292
B. Fuel Processing Cost Basis 331
C. Powerplant Cost Analysis 338
IV
-------
LIST OF FIGURES
Fig. Title Page No.
1-1 Typical Present-Day Waste-Heat Combined Gas and Steam
Turbine System 2
1-2 BCR Entrained-Flow Gasifier 5
1-3 Selexol Low-Temperature Desulfurization 10
1-4 Conoco High-Temperatute Desulfurization 11
1-5 BCR/Selexol System 15
1-6 Potential Pre-Mix Combustor Layout 20
5-1 Basic Gasification Coal Combined Cycle Power System 32
5-2 General Fuel Processing Schematic 34
5-3 Typical Low-Temperature Acid Gas Removal Unit 43
6-1 Integrated Low-Temperature Cleanup System 60
6-2 Ash-Agglomerating Gasifier 65
6-3 BCR Entrained-Flow Gasifier 68
6-4 Low-Btu Molten Salt Coal Gasification Process 71
6-5 Gasification and Gas Cleanup 75
6-6 Ash Removal and Salt Recovery Section 77
6-7 Typical Flow Diagram-Selexol Acid Gas Removal Process .... 79
6-8 Conoco Process Block Diagram 81
6-9 Conoco High-Temperature Desulfurization 82
6-10 Claus Sulfur Recovery Process 86
6-11 Beavon Tailgas Cleanup Process 88
7-1 Waste-Heat Combined Gas and Steam Turbine System 91
7-2 Open Cycle Gas-Turbine Engine Performance Trade-off
Curves
93
v
-------
LIST OF FIGURES
Fig. Title Page No.
7-3 Trends for Combined-Cycle Systems 95
7-4 HTTTP Gas Turbine Parametric Study 96
7-5 Effect of Changes in Pressure Ratio 98
7-6 Results of Steam Cycle Parametric Study 99
8-1 Typical T-Q Diagram for Waste Heat Boiler 103
8-2 Effect of Fuel Temperature 105
8-3 Capital Equivalent of Heat Rate Improvement 107
8-4 Bumines/Selexol System with Fuel Gas Resaturation 118
8-5 Effect of Water Vapor in Fuel Gas Bumines-Selexol
System 119
8-6 Equilibrium Constant for H2S Absorption by Half -
Calcined Dolomite 123
8-7 Equilibrium Constant for COS Absorption by Half -
Calcined Dolomite 124
8-8 Dissociation Pressure for Calcium Carbonate 125
8-9 Sulfur Absorption by Half Calcined Dolomite 126
8-10 U-Gas Gasifier Equilibrium Composition at 2010R 128
8-11' High-Temperature Cleanup System Performance 129
9-1 Effect of Particle Size on Engine Lifetime 141
9-2 NOX Production from Combustors Burning Low-Btu
and Medium-Btu Gas 147
9-3 Nitric Oxide Formation in Gas Turbine Burner 149
9-4 Concentration-Time Profiles for Premixed H2-CO-Air
Mixture 151
9-5 NO Rate Parameter as a Function of Temperature 153
9-6 Equilibrium Combustion Products Low Btu Gas
144 Btu/SCF 154
VI
-------
LIST OF FIGURES
Fi-g- Title Page No.
9-7 Equilibrium NO Concentration Molten Salt Gasifier
Fuel 155
9-8 Adiabatic Flame Temperature Molten Salt Gasifier Fuel .... 156
9-9 Comparison of Conventional Combustor NOX - Equivalence
Ratio Relationship with that of a Well-Stirred Reactor . . . 157
9-10 Reduction in NOX Emission by Reduction in Residence
Time - Information Based on Experimental Data 159
9-11 FT4 Premix Rig with External Mixing 161
9-12 Comparison of Rapid Quench and Premix NOX with Baseline . . . 162
9-13 NOX Estimation for Premixed Combustion Based on
Extended Zeldovich Mechanism 164
9-14 Schematic of Premix Concept Applied to a Combustor
Dome 165
9-15 Diagrammatic Representation of Premix Combustor 167
9-16 Autoignition Characteristics-Residence Time in the
Pressure Tube Due to Local Flow Recirculations is not
Likely to Cause a Problem 168
9-17 Estimated Flashback Loop-Estimated Flashback Characteristics
for the Premix Tube Combustor and Montebello Test are
Shown 170
9-18 Calculated Stability Loop for Premixed Concept as Applied
to an Annular Combustor with 2600°F GET 172
9-19 Residual Nitrogen Species as a Function of Equivalence
Ratio and Residence Time 174
9-20 Optimization of the Initial Design Concept 175
9-21 Rich Burner Staging Characteristics 176
9-22 Extrapolated Fractional Efficiency of Particulate Removal
Device 179
10-1 Coal Preparation and Handling 187
VI1
-------
LIST OF FIGURES
Fig- Title Page No.
10-2 Particulates Emissions from Coal Preparation and Handling . . 189
10-3 Water Handling for Coal Preparation 193
10-4 Sources of Wastewater in a COGAS Power System 196
10-5 Schematic Representation of Waste Water Streams and
Anticipated Treatment 197
10-6 Solid Wastes Exiting a Generalized Integrated Plant 205
10-7 Possible Distribution of Trace Elements 219
11-1 System Flow Diagram-U-Gas/Selexol 235
11-2 System Flow Diagram U-Gas/Low Steam/Selexol 240
11-3 System Flow Diagram BCR/Air-Blown/Selexol .245
11-4 System Flow Diagram BCR/Oxygen-Blown/Selexol 250
11-5 Molten Salt System Flow Diagram 261
11-6 System Flow Diagram U-Gas/Low Steam/Conoco 267
11-7 System Flow Diagram-BCR/Air-Blown/Conoco 271
11-8 System Flow Diagram-BCR/Oxygen-Blown/Conoco 276
A-l Combined Cycle-Power Plant-Water Treatment and Reuse 293
A-2 Conventional Sour Water Stripping Process 303
A-3 Two Stage All-Distillation Process 304
A-4 Phosam-W Process for Ammonia Separation 306
A-5 Sour Water Generation and Treatment 308
A-6 Conventional Two-Bed Ion Exchange Process 315
A-7 Boiler Feed Water Treatment Schemes 316
A-8 Conventional Demineralizer and RO/Demineralizer Systems
for Boiler Feedwater Treatment 322
A-9 Sidestream Softening of Cooling Water 329
Vlll
-------
TABLES
Number Title Page
1-1 Gasifier Types Considered 4
1-2 Low Temperature Cleanup Processes 8
1-3 High Temperature Cleanup Processes 9
1-4 Power System Characteristics(D 14
1-5 Summary of Power Plant Characteristics 16
1-6 Potential Fate of Trace Elements* 22
5-1 Comparison of Air- vs Oxygen-Blown Gasification 37
5-2 Low Temperature Cleanup Processes 42
5-3 High Temperature Cleanup Processes 49
6-1 Summary of Low Temperature Integrated Systems 61
6-2 Feed Coal Compost ion 64
6-3 U-Gas Gasifier Effluent 66
6-4 Molten Carbonate Reactions 72
6-5 Molten Salt Gasifier Coal and Raw Gas Composition 74
8-1 BCR-Selexol Fuel Gas Regenerator Costs 106
8-2 Basic Selexol Designs 109
8-3 Selexol Utilities and Cost Summary Refrigerated Operation . . 110
8-4 Mol Balance for Ambient Temperature Selexol Design Ill
8-5 Selexol Cost and Utilities Summary 112
8-6 Refrigerated vs Ambient Temperature Selexol Operation . . . .113
8-7 Refrigerated vs Ambient Temperature Selexol Operation . . . .113
8-8 Summary of Performance and Cost Refrigerated vs Ambient
Temperature Selexol Operation 115
IX
-------
TABLES
Number Title Page
8-9 Effect of Steam Addition on Fuel Gas Chemical Heating Value . 117
8-10 Comparison of BCR Data 121
8-11 Comparison of High vs Low Steam Performance 130
8-12 Gasifier Equilibrium Compositions 132
8-13 Molten Salt Gasifier Equilibrium Composition 133
9-1 Emission Summary 136
9-2 Comparison of Emission Standards 138
9-3 Gas Turbine Fuel Specifications 139
9-4 Suggested Low-Btu Fuel Gas Cleanup System Goals . 142
9-5 Comparison of Selexol Designs 145
9-6 Theoretical Flame Temperatures 148
9-7 Characteristic Times for a Natural Gas Combustor 160
9-8 Particulate Loading in Gasifier Product Gases 178
10-1 Typical Composition of Illinois No. 6 Coal 184
10-2 Typical Form in Which Trace and Minor Elements Occur in Coal. 185
10-3 Coal-Pile Runoff Analysis at Selected Plants 192
10-4 By-Product Water Analysis from Synthane Gasification
of Various Coals 198
10-5 Gasification Wastewater - Estimated Composition 199
10-6 Cooling Tower Slowdown Characteristics 201
10-7 Summary of Solids Exiting the BCR and IGT Combined Cycles . . 206
10-8 Description of Slag and Salt Slurry Exiting the Molten
Salt Gasifier 207
10-9 Constituents of Coal Ash 208
-------
TABLES
Number Title Page
10-10 Minimum Acute Toxicity Effluent and Estimated Permissible
Concentration Values for Air and Water 212
10-11 Environmental Methodology for Ranking Trace Elements
in Air and Water 214
10-12 Environmental Ranking of Trace Elements in Air and
Water-Land 215
10-13 Trace Element Distribution Illinois No. 6 Coal 217
10-14 Modified Environmental Ranking of Trace Elements in
Air and Water-Land 218
10-15 Formation of Trace Element Compounds 221
10-16 Predicted Volatilities of Trace Elements 222
10-17 Retention of Trace Elements in Gasifier Ash 223
10-18 Content of Leachate from Gasifier Ash (Montana Rosebud Coal). 225
10-19 Trace Elements in Condensate From an Illinois No. 6 Coal
Gasification Test 226
10-20 Distribution of Trace Elements for Lurgi Gasification .... 227
10-21 Potential Rate of Trace Elements 227
11-1 System Performance Summary 232
11-2 Overall Power Generation Cost Summary 233
11-3 Material Balance for U-Gas/Air-Blown/Selexol 236
11-4 Material Balance for U-Gas/Air-Blown/Selexol - Low Steam . . 241
11-5 Material Balance for BCR/Air-Blown/Selexol 246
11-6 Material Balance for BCR/Oxygen-Blown/Selexol 251
11-7 Utilities - U-Gas/Selexol/Air-Blown 255
11-8 Utilities - U-Gas/Selexol/Air-Blown - Low Steam 256
XI
-------
TABLES
Number Title Page
11-9 Utilities BCR/Selexol - Air-Blown 257
11-10 Utilities BCR/Selexol - Oxygen-Blown 258
11-11 Molten Salt System 262
11-12 Utility Summary - Molten Salt System 266
11-13 Material Balance for U-Gas/Low Steam/Conoco 268
11-14 Material Balance for BCR/Air-Blown/Conoco 272
11-15 Material Balance for BCR/Oxygen-Blown/Conoco 277
11-16 Utility Summary - BCR/Conoco - Air-Blown 281
11-17 Utility Summary - BCR/Conoco - 02 - Blown 282
11-18 Utility Summary - BCR/Conoco - Air-Blown 283
11-19 Fuel Processing Cost Summary Low-Temperature Cleanup .... 286
11-20 Fuel Processing Cost Summary High-Temperature Cleanup .... 287
11-21 Power Systems Cost Summary 288
A-l Water Balances for the BCR/Air-Blown/Selexol Process Overall
Water Balance 294
A-2 Water Balances for the BCR/Oxygen-Blown/Selexol Process
Overall Water Balance 295
A-3 Water Balances for the IGT/Air-Blown/Selexol Process
Overall Water Balance 296
A-4 Cooling Water Requirements for BCR/Air-Blown/Conoco Process . 297
A-5 Cooling Water Requirements for BCR/Oxygen-Blown/Conoco
Process 298
A-6 Cooling Water Requirements for IGT/Air-Blown/Conoco Process . 299
A-7 Flow Characterization for BCR/Air-Blown/Selexol Sour Water
Treatment 310
xix
-------
TABLES
Number Title
A-8 Flow Characterization for BCR/Oxygen-Blown/Selexol Sour
Water Treatment 311
A-9 Flow Characterization for IGT/Air-Blown/Selexol Sour Water
Treatment 312
A-10 Cost Estimation of Sour Water Treatment 313
A-ll Summary of Pertinent Data for Conventional Ion Exchange
Resins 314
A-12 Raw Water Analysis 318
A-13 Performance Characteristics for Scheme 2 Boiler Feed Water
Demineralization* 319
A-14 Quality Requirements for Boiler Feedwater^D 320
A-15 Cost Summary for Demineralization of Boiler Feed Water . . . 321
A-16 Cost Comparison of Conventional Demineralizer and R/0
Demineralizer Systems . 323
A-17 Control Limits for Cooling Tower Circulating Water
Composition 325
A-18 Impact of Softening Cooling Tower Make-Up Water 326
A-19 Cooling Tower Water Treatment Cost 327
B-l Equipment List for Air-Blown BCR Process 332
C-l Power System Cost Breakdown 340
X1L1
-------
CONVERSION FACTORS
ft x 0.30W = m
ft3 x .02832 = m3
gal x .003785 = m3
GPM x U.ltOS = m3/s
atm x 101,325 = N/m2 (Pa)
lbf/in2 x 6895 = N/m2 (Pa)
lbf x U.UU8 = N
lbm x .U536 = kg
ton x 907.2 = kg
hp x 7^6 = W
Btu x 1056 = J
*Btu/ft3 x 37,288 = j/m3
Btu/rb raol x 2328 = J/kg mol
lbm/106 Btu x .U295 = kg/109 J
F subtract 32 x 5/9 = C
Differences between various "standard"
conditions (e.g., SCF vs. normal m3) may
require slight modifications to this con-
version factor.
xiv
-------
ACKNOWLEDGMENT
The work described herein was performed by the United Technologies Re-
search Center (UTRC) for the Synthetic Fuels Program, Fuel Process Branch,
Industrial Environmental Research Laboratory, Research Triangle Park, under
EPA Contract 68-0.2-2179 during the period September 1976 to December 1977
Included among those who assisted in performing this work were Messrs. R. L.
Sadala, S. J. Lehman, W. R. Davison and Dr. E. B. Smith of UTRC. Mr. V. B.
May headed the efforts of.Hittraan Associates, Inc. who provided expert assis-
tance in the area of effluent emissions and water use. The subcontract effort
of Fluor Engineers and Contractors, Inc. on gasification processes was ably
headed by Mr. J. Moe. The UTRC Program Manager was Dr. F. L. Robson and the
Deputy Program Manager was Mr. ¥. A. Blecher,
Technical Project Officers for the EPA were Mr, W. J. Rhodes until July
1977 and, subsequently, Dr. T. W. Petrie. Valuable guidance and comments re-
ceived from these gentlemen are gratefully acknowledged.
Process information was made available from current studies sponsored
by the Electric Power Research Institute (EPRl), The cooperation and comments
of Dr. M. J. Gluckman of EPRI are also gratefully acknowledged.
xv
-------
SECTION 1
SUMMARY
INTRODUCTION
The electric utility industry is the largest single consumer of basic
energy in the United States. As such, it is under massive pressure to uti-
lize indigenous fuels, especially coal, the most plentiful fossil fuel.
However, the industry realizes that environmental protection is a primary
concern and, thus, there is a great interest in methods of generating elec-
tricity with minimum environmental intrusion. One of the first studies to
identify advanced methods of generating electric power at minimum emissions
and cost was reported on in December 1970 (Reference 1-1) by the United
Technologies Research Center (UTRC) for the National Air Pollution Control
Administration, an EPA predecessor organization. In this study, the combined
gas turbine-steam turbine power cycle, or simply combined-cycle system, used
in conjunction with coal gasification and fuel gas cleanup (Figure 1-1) was
identified as the most attractive of the various advanced power systems in
terms of emissions and cost of power. Subsequent studies by a wide variety of
organizations (e. g., References 1-2, 1-3, and 1-4) have reached the same con-
clusion.
Since 1973, UTRC has been carrying out further studies for the EPA
Fuel Process Branch, Industrial Environmental Research Laboratory/Research
Triangle Park with the objective of providing more detailed definitions of the
technological status, the potential environmental intrustion, the performance,
and economics of coal gasification with both low- and high-temperature sulfur
cleanup processes integrated with combined-cycle power systems. The model
integrated power plants are nominal 1000-MW, base-load plants.
In this and previous studies (References 1-5 and 1-6) fixed-bed, fluid-
bed, entrained-flow and molten salt gasifiers have been considered as have
various types of physical and chemical sorbent cleanup systems. Power
systems technology has been both first-generation (~1980-82) systems oper-
ating at 2200 F* turbine inlet and second-generation (post 1985) systems
*EPA's policy is to use metric units. For the convenience of the reader
making comparisons to results of previous studies, non-metric units are
used herein. A list of conversion factors to metric units is given on
page xiv.
-------
TYPICAL PRESENT-DAY WASTE-HEAT COMBINED GAS AND STEAM TURBINE SYSTEM
AIR
COMPRESSOR
TURBINE
BURNER
FUEL
T-2000F
p~14 ATM
T~875 F
STEAM
BOILER
T~300 F
TO STACK
T-775 F
P~60 ATM
POWER TURBINE
STEAM
TURBINE
L
PUMP
CONDENSER
ELECTRIC
GENERATOR
70 MW
ELECTRIC
GENERATOR
50 MW
O
T
-------
operating at 2600 F. Emphasis has been placed on the latter since it appears
that it will be the last half of the 1980 decade before gasification processes
become available for utility use.
In the performance of these studies, UTRC has had the expert assis-
tance of Foster Wheeler Energy Corporation, Fluor Engineers and Constructors,
Inc., Hittman Associates, Inc. and the Allied Chemical Company. In addition,
the Electric Power Research Institute (EPRI) has provided considerable data on
several gasification processes which were the subject of current EPRI-
sponsored research at Fluor Engineers and Constructors. The Conoco Coal
Development Co. has provided data relating to their half-calcined dolomite
desulfurization process.
The major conclusion of these studies is that the integrated coal
gasification/sulfur cleanup/combined-cycle power system offers the potential
of lower environmental intrusion and lower cost electricity than conventional
power plants with flue gas desulfurization. During the course of the studies,
the gasifier operating conditions were updated and modified where possible to
reduce utility consumption and thereby increase the gasifier efficiency. This
is reflected throughout the entire power plant by lower capital and fuel costs
and lower emissions. Thus, each subsequent study phase has indicated an
increased attractiveness for the integrated power plants.
For the systems studied, sulfur removal costs showed little increase
as emissions were reduced to one-third or less of the EPA standard for
coal-fired power plants. While fuel gas cleanup equipment for medium-Btu gas
is smaller and less costly than for low-Btu gas, an integrated plant with an
oxygen-blown gasifier showed no advantage over its air-blown counterpart; in
fact, for the gasifier type considered, it had a higher heat rate, higher
power costs, and potentially higher NOX emissions.
During the course of the overall program, continual refinements were
made in the integration process which were more advantageous to the systems
using low-temperature cleanup. Thus, early estimates for the performance
advantages of the integrated power plant with high-temperature sulfur removal
have been decreased to marginal levels. Since the performance and cost of the
high-temperature cleanup system are based upon extrapolations of bench-scale
tests, whereas those of the low-temperature system are based upon adaptations
of commercial-scale equipment, it would not be unfair to presume that the
proven low-temperature systems should offer an attractive choice.
POWER PLANT COMPONENTS
The power plant consists of three basic elements: gasification, clean-
up and power generation. One of the most important factors in the gasified
coal/combined-cycle power system is the integration of the components to
achieve proper utilization of waste heat both in the steam cycle and in the
thermal regeneration of the fuel gas streams.
-------
Gasification
The major generic types of gasifiers have been examined including
fixed, entrained, and fluidized bed designs and the molten salt gasifier
which combines gasification and sulfur removal in a sodium carbonate melt.
Table 1-1 lists the gasifiers considered. Emphasis has been placed on those
gasifier operating characteristics, especially utility consumption, which are
a key factor in performance of the cleanup system and the combined-cycle
generating plant. For example, one change which has a widespread effect,
particularly on the air-blown, entrained-bed gasifier (Figure 1-2) is the
reduction in steam feed rate.
Generic Type
Fixed Bed
Entrained Bed
Fluid Bed
Molten Salt
TABLE 1-1 GASIFIER TYPES CONSIDERED
Specific Gasifiers Considered
Morgantown Energy Research Center
Status
Bituminous Coal Research (BCR)
Two Stage
Ton/hr test unit;
other fixed bed
designs are commer-
cially available
Oxygen-blown unit for
syn gas production
undergoing tests;
air-blown version
being designed by
Foster Wheeler
Energy Corporation
Koppers Totzek Atmospheric Pressure Commerical
Texaco Partial Oxidation
Commercial on oil
Institute of Gas Technology U-Gas Bench scale test
Pullman Kellogg and Rockwell
International
Ton/hr test unit
being built by
Rockwell
While no practical minimum steam feed rate has been established, on
a theoretical basis a value of 0.144 Ib steam per Ib coal was determined.
Previous studies had used values as high as 0.57. This reduction, coupled
with one in the coal transport gas flow rate, produced an increase from 78.5
percent to 83 percent in predicted cold gas efficiency (fuel gas chemical
heating value/coal chemical heating value). This improved efficiency is
accompanied by a reduction in both water vapor and carbon dioxide in the raw
fuel gas. By reducing the amount of steam diluent, it is theoretically
possible to come close to a 1:1 ratio between oxygen and carbon molecules in
the feed, thereby improving the product gas heating value due to the increased
production of CO instead of C02.
-------
FIG. 1-2
BCR ENTRAINED - FLOW GASIFIER
COAL
GAS
STEAM
TRANSPORT
GAS
SLAG
HOPPERS
SLAG
-------
Aside from an improved coal to gas conversion efficiency there are
several other beneficial effects. Reduced water vapor concentration means
less heat is lost at low temperature due to condensation of water in the fuel
gas stream. The increased chemical energy can be converted at combined-cycle
efficiency whereas sensible energy associated with C02 production in the
gasifier for the most part can only be converted at steam cycle efficiency.
Also, equipment capital savings are realized due to reductions in water
purification requirements, condensate processing and fuel gas volume.
The modifications of gasifier operating conditions can also produce
changes in the formation of potential pollutants in the gasifier. The sulfur
in the feed coal appears mainly as H2S and COS after gasification. Other
sulfur compounds are present but have not been satisfactorily identified.
When operating with excess steam, chemical equilibrium calculations for the
entrained bed gasifier indicate that approximately 96 percent of the sulfur
would appear as H2S and nearly 4 percent as COS. At the low steam flow
rates, the same calculations indicate approximately 6 percent COS. This could
affect the size and performance of the sulfur cleanup system as will be
described later.
The gasifier operating conditions also could affect the formation
of NOX. The two major sources of NOX emissions are fuel-bound nitrogen
compounds which are converted to NOX during combustion and thermal NOX
which results from locally high combustion temperatures. Equilibrium cal-
culations for the entrained bed gasifier indicate a 50 percent reduction
in fuel-bound nitrogen compounds (taken as ammonia) when the gasifier is
operated at the low steam conditions. While essentially all the fuel-bound
nitrogen compounds can be removed by water scrubbing as part of the low-
temperature gas cleanup system, there has been no method determined for
removal at high-temperature.
The previous discussion has focused on a particular type of gasifier,
the air-blown entrained bed type. In many respects the other gasifiers
studied are similar; however, some significant differences should be noted.
Both the fluid-bed (U-Gas) and molten salt types can expect a high degree of
catalytic activity as a result of their basic design concept. This should be
quite desirable since a major concern in other designs is that COS and NH3
levels will exceed equilibrium values.
Oxygen, rather than air, has long been used as the oxidant in gasifiers
when the desired off gas consists mainly of H2 and CO with a small amount of
C02- Usually called 'syn-gas,1 this gas with a heating value of approxi-
mately 300 Btu/SCF is typically used in the production of t^. Because of
the larger technology base of oxygen-blown gasification and also because of
the similarity of its combustion to that of natural gas, there has been
considerable interest in the use of this medium-Btu gas in the utility
industry as well as in industrial boilers.
In order to ascertain the benefits, if any, that could be obtained from
the use of this gas in integrated power plants, a comparison of power plants
-------
with oxygen- and air-blown entrained flow gasifiers with both high- and
low-temperature cleanup was made. The results indicated that the power plant
with the oxygen-blown gasifier had higher heat rates, higher costs and the
potential for very high NOX emissions as compared to its air-blown counterpart.
Fuel Gas Cleanup Systems
The key advantage of the integrated gasified-coal, combined-cycle
system is the opportunity to remove sulfur compounds prior to combustion
while the gas is at pressure. The process whereby the fuel gas is desul-
furized is the focal point of the environmental impact of the system. Both
high- and low-temperature cleanup systems have been investigated. Summaries
of the various cleanup systems reviewed are given in Table 1-2 (low-temperature)
and Table 1-3 (high-temperature). The Selexol* process (Figure 1-3) of Allied
Chemical was selected as typical of the low-temperature system while the
half-calcined dolomite process (Figure 1-4) of Conoco was selected for the
high-temperature system.
Similar to other sulfur removal systems, the Selexol process is sensi-
tive to the type of sulfur compound that must be removed, i.e., whether sulfur
in the form of COS must be removed in addition to H2S. While H2S is quite
soluble compared to other gases in the fuel, COS is only about 30 percent as
soluble as H^S. As a result, if the system must be sized for appreciable COS
removal to meet very low levels of sulfur emissions, its cost and utility
requirements increase dramatically.
In discussing sulfur emission levels, it is important to note that
all sulfur has been assumed to appear as either H2S or COS in the fuel gas.
Thus, while estimates may show emission levels of less than 0.1 lb/10° Btu,
these do not account for other compounds that may not be removed or processed
by the particular sulfur removal or recovery system. For example, data for a
Synthane gasifier (oxygen-blown) show a total concentration of up to 100 ppm
of other sulfur bearing compounds. If none of these were removed, the result-
ant emission from these compounds alone would be on the order of 0.05 to 0.1
lb/10° Btu. While data for this type of gasifier are not directly appli-
cable, they do provide an indication of the limitations of this study.
In the design of the Selexol sulfur removal system, operating temper-
ature is rather critical. Gas sweetening plants generally run with subambient
absorber temperatures. However, in the interests of simplifying the process
for application in the utility industry, a study was made of the effect of
Selexol operating temperature on the overall power plant performance and
cost. The results of this study indicated a decided reduction in the cost of
electricity due to lower solvent flow rate associated with low-temperature
operation. It should be noted that this study was based upon a Selexol system
designed for virtually complete H2S removal from the fuel gas (down to
*A registered trademark of Allied Chemical Corp.
-------
TABLE 1-2
LOW TEMPERATURE CLEANUP PROCESSES
Basis: 8400 tons/day Illinois No. 6 Coal Fed to BCH Gasilier. or 670(1 ppm of Influent lljS
Process
Chemical
solvent type
1. MEA
1. DEA
3. TEA
4. Alkuid
5. Benfield
6. Caracarb
Phyjical
solvent type
7. Sulfinol
8. Selexol
9. Reel not
Direct
conversion
10. Siret-
ford
11. Town-
send
Orybed type
12. Iron
sponge
Absorbent
Monoetha-
noliiirune
/
Ditlhanol
amine
Tri«tha-
nolamine
Potassium
dimethyl
arnino
acetate
Activated
potassium
carborijte
solution
Activated
potassium
carbonate
solution
Sul'olane
t
Dilsupro-
punoamina
Polyethyl-
ene glycol
ether
Methanol
Na,CO, +
anthrjquin
one iul-
fonic acid
Triethylene
glycol
Hydrated
Type of
Alisurhent
Aqueous
Solution
A(|ueous
Aqueous
solution
Aqueous
solution
Aqueous
solution
*rr
Organic
solvent
Organic
solvent
Organic
solvent
Alkaline
solution
Aqueous
solution
Fixed
Temp.
" f
80 to
120
100 to
130
100 to
150
70to
170
150to
200
150 to
250
•
80 to
120
3010
80
<0
ISO to
ISO
70 to
100
Pressure
Insensitive
to variation
in pressure
Insensitive
to variation
in pressure
Insensitive
to variation
in pressure
Insensitive
to variation
in pressure
1 • 80 Jim
Insensitive
to variation
in pressure
geimrally
> 300 usi
High
pressure
preferred
Efficiency uf S Removal
%H,S In-
fluent
99
99
99
99
99
99
99
99
99
99.9
99.9
99
Effluent
II, S
-100
-100
-100
-100
H,S
+ COS
-100
H,S
+ COS
-100
H,S
tCOS
-100
H,S
+ COS
-100
-100
-to
-10
H,S
tCOS
-ion
Absorbent
Cbaijcteiistics
Life
Unlim-
ited
No
deO'B-
jjnon
Rrtjenera-
tiu it
Thermal
Thermal
Thermal
With
steam
With
sttam
With
steam
Low
pressure
heuting
or with
steam
Selectivity
tlUV.IIll
Forms non-
iusorbs
COS. CS,
H,S, and
also ubsorbi
COK. CS,
and mer-
tJiil.ini
absorbs
COS
H,S
H,S
H,S
H,S ond
also towards
COS.CS,
jncl mer-
captdfis
Make up
rate
GO to
100%
< 5%
< 5%
< 5%
'
bOto
100%
Form of
Sullur
Heccvery
AiH.S
gas
AsH,S
gas
AsH.S
<)as
As H,S
gas
AsHaS
gas
As H,S
gas
A»H,S
gas
AsH.S
(JdS
Elemen-
l.'il
sulfur
Elemen-
tal
sulfur
Elemen-
tal
sulfur
Status
Commercial
Commercial
Commercial
Commercial
Commercial
Commercial
Commercial
Commercial
Commercial
Commercial
Commercial
8
-------
HIGH TEMPERATURE CLEANUP PROCESSES
Basis: 8400 tons/day Illinois No. 6 Coal Fed to BCR Gasificr, or G700 ppm of Influent H,S
. .
i 1 «
Process Absorbent
i
i |
1. Bureau
of Mines
2. Hnbcock
and
Wilcox
3. CONOCO
4. AirprcJ-
ucts
5. Battelle
Morth-
west
6. IGT-
t
i
i Type of
Bed
Temp.
°F
Sintered | Fixed
pellets of
Fe:03 !25%)
and fly ash
Fe-Oj
Half caic;ncd
dolomite
bed
Fixed
bed
Flmdized
bed
Calcined Fixed
dolomite
Molten
caroonates
(15%CcC03!
bed
Solution
Molten moral Spbshing
Meis- j (proprietary) | contact
sner [
i
1 COO to
1500
800 to
1200
1500 to
Pressure
Insensitive
to variation
in pressure
Insensitive
to variation
in pressure
~200psia
1800 H,S removal
1600 to
2000
1100 to
1700
900
is high at
low pressure
Insensitive
to variation
in pressure
Atmospheric
Hj S removal
is high
at low
pressure.
5-6 pslg
Efficiency of S
Removal
%Hj Sin-
fluent
-95
-99
-95
-95
-98
Effluent
H,S
ppm
-350
-75
-350
-350
-150
Absorbent
Characteristics
Life
>174
cycles
Wt loss
<5%
mini-
mum
5-6
cycles
Regener-
ation
With air
10-13%
with
steam
and CO,
80-00%
with
steam
and CO,
With
steam
and CO,
Elec-
troly-
tic
Selec-
tivity
toward
H7S.
COS
H,S,
COS
H,S.
COS
H,S.
COS.
fly ash
H,S.
COS
Make up
rate
<5%
1%of
circula-
tion
rate
Form of
Sulfur
Recovery
AsS05
gas
As
12-15%
SO, gas
AsH2S
gas to
Claus
process
AsH,S
gas to
Claus
process
AsH,S
gas to
Claus
process
Energy
Required
Elec.
kw
96.360
3330
Oth-
er
stu
Status
Pilot
Experi-
mental
Pilot
Aban-
doned
Pilot
Concep'
tiral
-------
SELEXOL LOW -TEMPERATURE DESULFURIZATION
CLEAN
FUEL GAS
FUEL GAS
FROM WATER WASH
SOLVENT COOLER
ABSORBER
POWER
RECOVERY
TURBINE
SOLVENT-SOLVENT
EXCHANGER
SOLVENT PUMP
OVERHEAD
CONDENSER
ACID GAS
CONDENSATE
TANK
CONDENSATE
I
ro
ID
(O
to
-------
CONOCO I-HGH — TEMPERATURE DESULFURIZATION
BOI LER
HOT FUEL GAS
MAKE-UP
C02
DESULFURIZER
SPENT ACCEPTOR
LOCK HOPPER
CO2
WATER
o
CO
I
M
ID
ID
SPENT ACCEPTOR
CONVERTERS
ACID GAS
DOLOMITE
SLURRY
ACID-GAS
STREAM
LIQUID-PHASE
CLAUSPLANT
SULFUR
STACK GAS
CO2+H2O TO
REGENERATOR
-------
approximately 35 ppm) and that total power plant sulfur emissions were on the
order of 0.2 to 0.4 lb/10" Btu. If less ambitious sulfur removal goals were
considered (Reference 1-7), it appears that there would be a smaller differ-
ence between ambient and refrigerated operation.
Performance was estimated for power plants using each of the gasifier
types followed by the Conoco half-calcined dolomite high-temperature
desulfurization system. Reactions within the desulfurizer are reported
to be close to equilibrium and are limited by the amount of water vapor
and carbon dioxide in the fuel gas. Low concentrations of these molecules in
the raw fuel gas improve desulfurization. The latest data for the air-blown
entrained bed gasifier with minimum steam feed were especially attractive for
use in a combined-cycle application with the high-temperature desulfurizer.
To maintain gasifier temperature within design limits under minimum steam
feed, oxidant feed must also be low; hence, the product gas has low concen-
trations of both water vapor and carbon dioxide. Desulfurizer performance
under these conditions resulted in an estimated emission after combustion of
0.09 Ib 802/10^ Btu. Similarly, use of the gasifier equilibrium model to
simulate the fluid bed/high-temperature cleanup system at reduced steam feed
rates indicated emissions of approximately 0.15 Ib 802/10^ Btu.
Removal of particulates and nitrogen compounds also must be considered
when evaluating fuel cleanup processes. For example, the level of particulates
needed to avoid excessive high-temperature turbine erosion is lower than the
environmental regulations. Thus, the water wash associated with low-temper-
ature systems which clean the fuel gas stream to gas turbine tolerance will
meet the EPA particulate standard. No such system is available for high-
temperature processes.
Fuel cleanup system characteristics also affect the quantity of nitrogen
compounds in the fuel which could form NOX during combustion. For systems
using low-temperature sulfur removal, a water scrub is incorporated which
effectively removes both particulates and ammonia, the major nitrogen compound
after gasification. Removing ammonia minimizes NOX formed during combustion
in conventional burners but thermally generated NOX is still a problem.
However, United Technologies sponsored tests carried out in conjunction with
the Texaco Development Company at their Montebello California Research Lab-
oratory have led to the identification of a potentially attractive burner
concept. The test results (References 1-8 and 1-9) indicate that a premix
burner should meet both current and proposed NOX standards. Thorough
premixing of fuel and air permits control of combustion temperature and
associated NO production. To avoid possible preignition, fuel temperature to
the gas turbine has been limited to 1000 F with both high- and low-temperature
cleanup systems. While this in effect penalizes the high-temperature cleanup
system by not allowing it to utilize the fuel gas sensible heat over 1000 F in
the combined cycle, this penalty is necessary if excessive NOX emissions are
to be avoided.
12
-------
Power System
Throughout the course of the ongoing studies, the power system has
been updated to reflect advances in gas turbine technology. The latest update
includes results from the recently completed DOE-sponsored High-Temperature
Turbine Technology Program (HTTTP) Phase I. The advanced turbine defined in
this program has a dual-spool gas generator and a free turbine driving the
generator. The overall pressure ratio is 18:1 at a turbine inlet temperature
of 2600 F. Compressor discharge air is precooled to 400 F prior to use in
cooling the rotating blades while water cooling is used in the static
sections. Advanced ceramic coatings were projected for a number of hot
section static parts. The characteristics of the power system are given in
Table 1-4.
The present study has adapted the flow path of the HTTTP engine but
has assumed ceramic vanes, which require no cooling, rather than the water
cooled concept. The resulting performance is essentially the same.
A gas turbine pressure ratio of 18:1 results in an exhaust temperature
that is sufficiently high to provide both superheat and reheating for a 2400
psi, 950 F reheat steam cycle. The use of such a high-performance steam system
would not be practical in a conventional distillate-fueled combined cycle.
This is one of the advantages of the fully integrated system. Little, if any,
evaporation is done in the heat recovery boiler in the gas turbine exhaust
stream. High-pressure saturated steam is raised in the gasifier and cleanup
process and can be extracted from the steam cycle for process use after having
produced some useful output. The net effect is a range of overall coal to
power^conversion efficiencies from 42 percent to 46 percent. These high
efficiencies are important not only because they minimize the fuel dependent
cost of power generation but also because increased plant efficiency implies
increased electrical output for a given size coal processing system. Thus,
the capital cost of fuel processing equipment per unit output is reduced.
This,^in turn, affects the degree to which the fuel can be cleaned within
existing economic constraints.
OVERALL POWER PLANT
The performance, cost, and environmental intrusion of the power plant
cannot be defined until the various components have been integrated. The
integrated plant is very complex involving the exchange of air and steam
between the fuel processing and power systems. A simplified schematic of a
typical integrated power plant with low-temperature desulfurization is shown
in Figure 1-5.
Performance
A summary of the performance, costs and major air emissions of eight
integrated power plants is given in Table 1-5. Integrated power plants
using fixed-bed gasification and the first-generation of advanced gas turbines
13
-------
TABLE 1-4
POWER SYSTEM CHARACTERISTICS^
Gas Turbine
Inlet Temperature, F 2600
Pressure Ratio 18
Inlet Flow Rate, Ib/sec 983
Unit Power, MW 181
Steam Turbine
Throttle/Reheat Temperature, F 950/950
'Throttle/Reheat Pressure, Psig 2400/585
Throttle/Reheat Flow Rate, Ib/sec 764/764
Condenser Pressure, in Hg 4
Unit Power, MW 453
Overall Power Plant
Gross Output, MW 1178
Net Output, MW 1098
Heat Rate,(2) Btu/kWh 7775
(1) For integrated plants with air-blown entrained bed gasifiers
and low-temperature fuel gas cleanup. Other types of gasi-
fiers and cleanup systems will affect power system character-
istics .
(2) Coal pile to busbar.
14
-------
BCR/SELEXOL SYSTEM
Tl
01
-------
TABLE 1-5
SUMMARY OF POWER PLANT CHARACTERISTICS
Air-Blown Air-Blown Air-Blown Air-Blown
BCR/ U-Gas/ U-Gas/ BCR/
Air-Blown Air-Blown 02-Blown 02-Blown
U-Gas/ Molten BCR/ BCR/
Po\
rerplant Type Selexol Selexol^' Selexol1^' Conoco
Net Power^ - MW 1098 1074 1109 1136
Net Efficiency - % 43.9 42.9 44.3 45.4
Capital Cost - $/kW 657 663 641 597
Power Cost^ - Mills/kWh 31.76 32.24 31.08 29.25
Ib
Ib
Ib
S02/106 Btu(5) 0.40 0.19 0.23 0.13
NO /106 Btu^ — <7) --^7) —(7) 2 52W
X
particulate/106 Btu^6^ <0.01 <0.03 <0.03 <0.10
Conoco ^) Salt Selexol Conoco
1145 955 1045 1098
45.7 43.3 41.8 43.9
543 683 725 709
27.08 32.99 34.87 33.88
0.19 0.21 0.39 0.73
0.165(8) — <7> ~<7) 2.60<8>
<0.10 <0.03 <0.03 <0.10
(1) Based on EPRI-furnished data
(2) Based on UTRC/IGT-developed low steam process
(3) Based on 700,000 Ib/hr coal flow
(4) Based on $1.00/106 Btu coal
(5) Emissions from power plant
(6) Emissions from power system
(7) Function of burner design <0.35 lb/10 Btu
(8) Fuel bound nitrogen only - burner design will add thermal NO
-------
(turbine inlet temperature of 2200 F) have not been included because their
performance and costs are significantly less attractive than the more advanced
second-generation gasifiers and gas turbines. A detailed discussion of the
less-advanced power plants is given in Reference 1-5.
Performance of the various advanced air-blown systems with low-tempera-
ture cleanup showed little difference between gasifier types. Generating
efficiency from coal pile to busbar varied between 43 and 45 percent. The
same is true of the air-blown fluid bed and entrained flow-type gasifiers
when coupled with the Conoco high-temperature cleanup system. For these, the
performance differences were so small that both can be considered to have an
efficiency of 46 percent. Thus, on the basis of performance, there is little
difference between integrated plants with either high- or low-temperature
cleanup especially when the low-temperature cleanup components are carefully
tailored to the power generating system.
It is of interest to note the comparison of integrated power plants with
air- vs. oxygen-blown operation of the entrained bed-type gasifier. With
low-temperature sulfur removal, a performance difference of approximately
two points was estimated (air-blown, n = 0.439 and oxygen-blown, n = 0.418).
The factors that cause this difference are associated with the loss of mass
in the fuel gas due to the oxygen separation process and the relatively high
steam feed needed in the oxygen-blown gasifier. In the case of the power
plant with the low-temperature cleanup system, the heat of vaporization of the
steam is completely lost when the water vapor in the fuel gas is condensed
before entering the cleanup system. In the case of high-temperature cleanup,
the water vapor remains in the fuel gas and acts in a manner similar to nitro-
gen as a diluent. While not enhancing performance, the fact that the water
vapor is available to do work in the gas turbine, and to provide sensible heat
in the waste heat boiler, results in a lesser difference between the oxygen-
and air-blown power plants when used with a high-temperature cleanup system.
Here the performance is separated by just over one point (air-blown, TI = 0.458
and oxygen blown, n • 0.444).
Costs
For the air-blown power plants with the Selexol cleanup process, capital
costs were quite close for the entrained bed, fluid bed and low-steam version
of the fluid bed gasifier, varying from $630/kW to $660/kW. The estimated
cost of electricity was also virtually the same for each, varying between 31
and 32 mills/kWh. Capital cost of the power plant with the oxygen-blown
entrained bed-type system with Selexol cleanup was approximately $720/kW.
While performance of this power plant was estimated to be some two points
poorer than that of the air-blown system, the cost of electricity (based on
coal at $1/106 Btu) was approximately 35 mills/kWh. This small difference
of approximately 10 percent, while likely to be real, is well within the
estimated accuracy limits of the costs available for use in the study and
probably would not represent a real deterrent to the use of an oxygen-blown
gasifier should the technology be preferred to that for an air-blown gasifier.
17
-------
However, considering the use of fuel gas for power generation only, there
appears to be no reason to opt for the oxygen-blown gasifier as it shows no
advantage in terms of emissions, performance or cost and would clearly be
somewhat more complex than the air-blown device.
The molten salt system can be considered to be a low-temperature system
since it incorporates a water wash (albeit at 175 F). It was estimated to be
comparable in cost to the others although the available cost estimates are
less definitive. Capital cost is estimated to be $720/kW while electricity is
estimated to cost 33 mills/kWh.
The integrated power plants using high-temperature desulfurization
processes show some advantage in terms of reduced cost of electricity.
However, they do not include provision for nitrogen removal from the fuel gas
and, thus, would not be environmentally acceptable. A reduction of 2 to 3
mills/kW over air-blown Selexol-based systems is indicated, but this advantage
is based on costs that include a larger number of unknown factors than do
those for the Selexol-based systems. Both cleanup and particulate removal
costs are based on extrapolation of small-scale lab test data as opposed to
low-temperature equipment which, while not having handled gasifier effluents,
has been built and run on a commercial scale. For the power plants using
air-blown gasifiers with Conoco cleanup, estimated power costs range from 27
to 29 mills/kWh. The high-temperature, entrained bed, oxygen-blown system has
only a slightly lower cost of electricity than that for its low-temperature
counterpart (34 vs 35 mills/kWh). Thus, for a power plant with an oxygen-
blown gasifier, there appears to be little incentive for the use of a high
temperature sulfur removal device. Those plants using air-blown gasifiers
show a 2 to 3 mill/kWh improvement and also have the potential for a sig-
nificant reduction in sulfur emissions. It must be reiterated, however, that
the high-temperature systems do not have identifiable methods of removing
fuel-bound nitrogen nor have methods of particulate removal been satis-
factorily defined.
During the course of this study improvements in integration as well
as improvements in both gasifier and power system performance have resulted in
an increase in the absolute value of overall power generating efficiency of
from 3 to 6 percentage points. This not only means a lower fuel cost, but
also permits the capital cost of gasification and cleanup to be spread over a
larger power output thus reducing the specific cost of these parts of the
system.
Environmental Intrusion
All of the power plants considered easily were able to meet the current
EPA standards for stationary power plant sulfur emissions of 1.2 Ib
802/10^ Btu. The power plant with the oxygen-blown gasifier shows an
advantage in terms of sulfur removal system size. Because there is no
nitrogen diluent, the quantity of gas to be processed is reduced by more than
one-third. With low-temperature cleanup, this results in an overall reduction
in system cost and utilities without a large effect on sulfur emissions. For
18
-------
the high-temperature case, equipment size, other than that of the absorber, is
more dependent on sulfur loading than gas flow rate and there is little
difference in size and cost. However, as noted earlier, due to the much lower
water vapor and C(>2 concentrations, the sulfur emissions estimated for the
air-blown system are significantly less than for the oxygen-blown system (0.1
vs. 0.7 lb/106 Btu.
Only the low-temperature systems such as Selexol incorporate a water
scrubber for ammonia removal. Thus, in the high-temperature systems, pro-
duction of NOX in the gas turbine could be quite high, as much as 2.5 Ib
NOX/10° Btu if all the nitrogen went to the oxide form. (Currently
proposed NOX standards for gas turbines are between 0.35 and 0.9 lb/106
Btu depending on the level of nitrogen in the fuel. The power plant standard
is 0.7 Ib NOX/1Q6 Btu). Clearly, if the high-temperature cleanup system
is to be viable, the combustor design must be such as to convert ammonia and
other fuel bound nitrogen compounds to N2 rather than NOX. This must be
done while limiting the formation of thermal NOX. Unfortunately, from the
viewpoint of thermal NOX formation, the fuel gas from the oxygen-blown
systems has a much higher flame temperature than even methane. Design of a
N0x-free combustor for that fuel could be a problem and certainly will be
more difficult than for the low-Btu gas.
For those power plants with the low-temperature cleanup, ammonia has
been removed and only thermally generated NOX is a problem. For thermal
NOX control, several promising combustion techniques have been identified.
After regeneration, the hot fuel gas and compressor discharge air combine to
produce flame temperatures that would result in excessive NOX emissions in
a conventionally designed combustor. Premixed, fuel-lean mixtures effectively
reduce flame temperature and consequently limit thermal production of nitrogen
oxides. In the premix concept, shown in Figure 1-6, air and fuel are well
mixed prior to combustion thereby significantly reducing the combustion time
and temperature and, thus, the production of NOX. Fuel-lean conditions,
however, have been shown to result in the conversion of virtually all ammonia
to NOX, which is unacceptable from an emission standpoint. Research has
shown that fuel-rich combustion can convert virtually all of the fuel-bound
nitrogen to N2 while minimizing the production of thermal NOX. Work
currently underway for the EPA at the Government Products Division, Pratt &
Whitney Aircraft has demonstrated NOX emissions as low as 50 ppm (-0.23
lb/10" Btu) while burning a fuel doped with 0.5 percent nitrogen (Reference
1-10).
In addition to sulfur and nitrogen oxides, sources of emissions exist
throughout the coal and fuel gas processing train. Many of these are similar
to those in coal-fired steam plants, particularly in regard to coal prep-
aration and heat rejection. Residuals from the cleanup process will include
waste water which must be treated prior to disposal. In general, treatment
will be highly dependent on individual plant design and will be affected by
plant location.
19
-------
POTENTIAL PRE-MIX COMBUSTOR LAYOUT
IsJ
O
MIXING ZONE OF FUEL AND AIR
COMPRESSOR DISCHARGE
GASIFIER FEED DUCTS
COMBUSTOR CASE
P
O
-------
The overall environmental intrusion must also consider the many trace
elements found in coal. Consideration of trace elements became a significant
part of the program effort. Because of the many trace elements present in
coal and the difficulty of discussing all of them, a ranking method was applied
to limit the number considered. The ranking procedure was based on the ratios
of concentration in typical Illinois No. 6 coal to allowable concentration in
air, water or solid waste as given by Minimum Acute Toxicity Effluent (MATE)
values and Estimated Permissible Concentrations (EPC) for each of the ele-
ments. The elements having the highest values of these ratios could have the
highest potential for producing harmful effects. This selection process was
tempered by consideration of the potential form that the element or its
compounds might take, as well as experience as to which have proven to be
troublesome in other applications. The elements As, B, Be, Cd, Cr, Hg, Ni,
Pb and V were selected as potentially troublesome.
An attempt to discuss the fate of trace elements during the gasifi-
cation of typical Illinois No. 6 coal and subsequent use of the gas is dif-
ficult. From the time that coal enters the plant boundary, there is the
potential for it to cause some type of environmental intrusion. Except for
some chemicals introduced for water treatment, the coal feed is the source of
all trace element emissions. The gasifier itself, a closed vessel, discharges
no pollutants directly to the environment. However, coal preparation and feed
equipment as well as gasifier exit streams all must be considered.
Those dust and particulate emissions that occur prior to gasification
will likely contain trace elements in a distribution of concentration similar
to that of the parent coal. The fate of trace elements in the gasifier,
whether they leave in the ash/slag or off-gas, is a function of factors such
as gasifier configuration, bed type and operating conditions. Trace elements
by themselves are not very volatile. Yet slag or ash particles from gasifiers
seem to be significantly depleted in most of the trace elements in coal. In
the high-temperature reducing atmospheres typical of gasifiers, trace elements
thus, would seem to be volatilized in the form of compounds such as carbonyls,
sulfides and hydrides. Those compounds condense in quench water.
While there are few data available on the fate of trace elements, a
number of estimates have been based on data from Reference 1-11. Table 1-6
presents the results of a calculation based on such an estimate to determine
the distribution of the elements between the slag and sour water. It is based
on the assumption that the amount of each element not appearing in the ash
will be found in the sour water. It shows that mercury, arsenic and cadmium
go predominantly to the sour water and may require special care. It may be
possible to utilize the sour water in the ash quenching process. If all the
elements were to appear in the slag or ash, disposal might be simplified if
guards against leaching are taken. As an example of the potential magnitude
of the problem, should all of the nickel estimated to appear in the sour water
be allowed to escape, it would require 3.6 million gallons per minute of water
to dilute that element to the level of the EPC.
21
-------
TABLE 1-6 POTENTIAL FATE OF TRACE ELEMENTS*
Flowrate (Ib/hr)
Element Feed Final Residue Sour Water
Hg 0.084 0.008 0.076
As 16.8 4.4 12.4
Pb 7.7 4.9 2.8
Cd 0.62 0.23 0.39
V 11.9 8.3 3.6
Ni 10.5 8.0 2.5
Be 0.70 0.57 0.13
B 140 126 14
Cr 10.5 10.5 0
% in Residue/
% in Sour Water
10/90
26/74
64/36
27/63
70/30
76/24
81/19
90/10
100/0
*Based on a coal feed rate of 700,000 Ib/hr
CONCLUDING REMARKS
The results of this study have updated prior work and have further
strengthened the conclusion that the integrated gasification/sulfur cleanup/
combined-cycle power system will be one of the more environmentally and
economically attractive medthods of generating electric power in the future.
Further work in the areas of the costs and performance penalities
associated with deep sulfur removal (removal of significant amounts of COS)
appears necessary. Also, it is important to recognize that an evaluation must
be made of methods to remove fuel-bound nitrogen prior to combustion and of
possible combustor modifications which might convert these compounds to
N2.
22
-------
REFERENCES
1-1. Robson, F. L., et al.: Technological and Economic Feasibility of
Advanced Power Cycles and Methods of Producing Nonpolluting Fuels
for Utility Power Stations. PB198-392, December 1970.
1-2. Harris, L. P., and R. P. Shah: General Electric Phase II Final
Report - Open Cycle Gas Turbines and Open Cycle MHD. NASA CR-134949,
Vol. II, Part 3, December 1976.
1-3. Beecher, D. T., et al.: Westinghouse Phase II, Final Report: Summary
MV Combined Gas-Steam Turbine Plant with or Integrated Low-Btu
Gasifiers. NASA CR-134942, Vol. I, November 1976.
1-4. Crozier, R., editor: Assessment of Technology for Advanced Power
Cycles. Report of Ad Hoc Panel on Advanced Power Cycles, National
Academy of Science, May 1978.
1-5. Robson, F. L., A. J. Giramonti, W. A. Blecher, and G. Mazzella: Fuel
Gas Environmental Impact-Phase Report. EPA-600/2-75-078, (NTIS No.
PB249-424), November 1975.
1-6. Robson, F. L., W. A. Blecher, and C. Colton: Fuel Gas Environmental
Impact. EPA-600/2-76-153, (NTIS No. PB257-134), June 1976.
1-7. Chandra, K. B. McElmurry, E. W. Neben, and G. E. Pack: Economic
Studies of Coal Gasification Combined Cycle Systems for Electric
Power Generation. EPRI AF-642, January 1978.
1-8. Crouch, W. B., et al.: Recent Experimental Results on Gasification
and Combustion of Low Btu Gas for Gas Turbine. ASME Turbine Conf.,
April 1974.
1-9. Crouch, W. B., and R. D. Klapatch: Solids Gasification for Gas
Turbine Fuel, 100 and 300 Btu Gas. llth Intersociety Energy
Conversion Engineering Conference, September 1976.
1-10 Hosier, S. A., and R. M. Pierce: Advanced Combustion System for
Stationary Gas Turbine Engines. Second Symposium on Stationary
Source Combustion, New Orleans, August 1977.
1-11. Attari, A. J. Pau, and M. Mensinger: Fate of Trace and Minor
Contaminants of Coal During Gasification. EPA-600/2-76-258,
(NTIS No. PB270-913), September 1976.
23
-------
SECTION 2
INTRODUCTION
Previous EPA-sponsored studies (References 2-1 and 2-2) at UTRC
demonstrated the potential environmental and economic benefits of integrated
coal gasification/gas cleanup/combined-cycle power plants. However, there
remained several technical problem areas requiring investigation or further
definition. One such area involved an expanded study of plant effluents
including an examination of the possible fate of potentially hazardous trace
elements. Another problem area was the comparison of integrated power plants
using air-blown versus oxygen-blown gasifiers supplying fuel to the advanced
power systems. To date, neither have been compared on a consistent basis
using the latest data on gasifier performance and effluent discharges. A
third area involved the use of advanced technology gasifiers such as the
bed and molten salt types which have the potential for higher integrated pbwei*
plant performance, but which have not been well defined in the integrated
power system mode from the technical, economic and environmental viewpoint.
Also, to provide a better understanding of the basic capabilities and limita-
tions of these advanced gasifiers, the modeling work done in the previous
program phases needed to be extended to consider them. The current program
addresses these areas in a manner consistent with the earlier work to provide
a meaningful comparison of these gasifiers on both an environmental and eco-
nomic basis when operating in an integrated power generating system using coal
gasification and advanced combined gas and steam turbine cycles with fuel gas
cleanup before combustion.
Prior investigations carried out under EPA Contract No. 68-02-1099
resulted in the major conclusion that integrated power plants using low-Btu
coal gasification, fuel gas sulfur removal, and combined gas and steam cycle
power generation offer the potential for reduced environmental intrusion while
generating electric power at costs competitive with or less than conventional
coal-fired stations with flue gas desulfurization. Coal processing systems
studied in UTRC's prior work for the EPA included entrained-flow (BCR-type)
and fixed bed (Morgantown Energy Reseach Center design, referred to herein as
the BuMines gasifier) medium pressure (25-atm) gasifiers with both high- and
low-temperature cleanup systems.
24
-------
Each of the coal conversion processes was mated with a combined-cycle
generating system for producing electric power. Additionally, systems utiliz-
ing existing technology such as the Koppers-Totzek atmospheric-pressure,
oxygen blown coal gasifier, and the Shell or Texaco-type partial oxidation oil
gasifier were studied. These studies are reported in References 2-1 and 2-2.
For continuity, the results are summarized in the early sections of this
report.
A further study consideration was the utilization of the latest data
developed under the sponsorship of other agencies or by other contractors
that could affect the conclusions of these previous studies. For example,
gasifier modeling performed during the earlier EPA-sponsored studies had shown
the advantages of low steam feed rates. Heat and mass balances available at
that time were not optimized for integration with combined-cycle power gener-
ating systems. As yet unpublished data, made available by the Electric Power
Research Institute (EPRI), substantiated that a marked performance improvement
can be achieved by that relatively simple change in approach. Other programs,
such as the High Temperature Turbine Technology Program sponsored by DOE have
provided technological data regarding the potential for improved combined-
cycle performance. Results from that and other ongoing programs have also
served as a basis for the selection of the most economically attractive
combined-cycle configuration consistent with current and predicted future coal
prices.
Because of the importance of these issues to the electric power gener-
ating industry, EPRI has participated in this program by allowing the use of
information generated under contracts sponsored directly by them. In sub-
stance, EPRI has provided process data on air- and oxygen-blown two-stage
gasifiers and on advanced fluid-bed gasifiers. Many of these data were
developed by Fluor Engineers and Constructors as part of an EPRI-sponsored
study. That study, subsequently reported on in Reference 2-3, sought to
identify whether significant economic and/or environmental incentives exist
for using various current and advanced gasification processes coupled with
combined-cycle power plants to produce electricity. The study did not explore
in depth the environmental intrusion of these integrated power plants. How-
ever, it provided the latest available revisions of the respective gasifier
operating conditions.
Fluor has also acted as a direct subcontractor to UTRC to provide process
engineering in the areas of fuel gas cleanup, system integration, and effluent
definition and control. Hittman Associates, who participated in the earlier
study, has continued as a subcontractor to provide overall evaluation of the
various discharges of air and water pollutants and solid wastes from the inte-
grated plant. Allied Chemical Co. has provided performance and cost estimates
for the Selexol desulfurization system and the Institute of Gas Technology has
cooperated in studies aimed at reduced steam feed to the U-Gas gasifier. UTRC
has provided power system characteristics and analysis, overall systems
integration, and environmental intrusions.
25
-------
In order to achieve the goal of comparing the environmental, performance
and cost characteristics of integrated power plants having both low- and
high-temperature sulfur cleanup processes, five program objectives were
defined:
1. To provide a more detailed evaluation than was done in previous
studies of the effluents from the integrated power plant and the control
methods that could be applied.
2. To compare the operational, economic, and environmental character-
istics of integrated power plants using oxygen vs. air-blown gasification
in an entrained flow (BCR-type) gasifier with both high- and low-temperature
cleanup.
3. To consider the merits of advanced gasifiers such as the fluid bed
and the molten salt gasifiers when integrated with a combined-cycle power
system.
4. 'To update previous work so that a meaningful comparison of perfor-
mance, emissions, and cost can be made between the present and previous
systems.
5. To utilize modeling techniques developed under previous contracts to
identify the effect of gasifier operating parameters on overall system
emissions and performance characteristics.
In carrying out these objectives, several other areas were uncovered that
warranted investigation. Among these were: 1) the effect of operating
temperature on the Selexol system performance and cost; 2) the perfor-
mance of the fluid-bed gasifier at reduced steam feed rates; and 3) the
use of nonselective, low-temperature desulfurization systems.
The conclusions from the study and recommendations for further work
follow this introduction. Sections 5, 6 and 7 deal with the individual
processes which make up the integrated gasification/sulfur cleanup/com-
bined-cycle power plant. Subsequent sections describe the overall power
plant integration, power plant air emissions and other environmental intrusion.
A final section presents the performance and resultant cost of electricity for
the integrated power plants.
26
-------
REFERENCES
2-1 Robson, F. L., A. S. Giramonti, W. A. Blecher, and G. Mezzella: Fuel Gas
Environmental Impact: Phase Report. EPA-600/2-75-078, November 1975.
2-2 Robson, F. L., W. A. Blecher, and C. B. Colton: Fuel Gas Environmental
Impact. EPA-600/2-76-153, June 1976.
2-3 Chaubra, K., B. McElmurry, E. W. Neben and G. E. Pack: Economic Studies
of Coal Gasification Combined-Cycle Systems for Power Generation: Final
Report. EPRI AF-642, January 1978.
27
-------
SECTION 3
CONCLUSIONS
1. Revisions in gasifier and power system operational characteristics have
resulted in lower integrated power plant heat rate and reduced emissions com-
pared to the estimates made for similar integrated power plants in the prior
UTRC studies carried out for the EPA.
2. The revised integrated power plants have become an even more attractive
method for providing electricity at a lower cost with less environmental
intrusion than a conventional steam plant with flue gas desulfurization.
3. Integrated power plants having oxygen-blown gasifiers with either low-
or high-temperature sulfur cleanup appear to be less attractive than their
air-blown counterparts. They exhibit higher heat rates, higher costs and
the potential for higher emissions of NOX.
4. For the integrated power plants studied having various advanced, air-blown
gasification systems, little difference between power plants was noted in
emissions, performance' or generating cost when operating with low-temperature
desulfurization and at low gasifier steam feed rates. Thus, gasifier selection
may be based on other considerations.
5. Power plants with a high-temperature cleanup process, for which bench
scale data only are available, have estimated performance and cost advantages
over plants with low-temperature sulfur cleanup, for which process data are
available. Careful design of plants with low-temperature cleanup can reduce
these estimated advantages to marginal levels.
6. In high-temperature cleanup processes, the problems associated with
efficient removal of particulates less than 10 u and the removal or harmless
conversion of nitrogen compounds in the fuel gas remain to be resolved. Until
such systems can be better identified, associated cost and performance penaltid
cannot be accurately defined.
28
-------
7. At the elevated fuel gas temperatures needed to attain highly efficient,
integrated power plant operation, conventional gas turbine combustor designs
would result in excessive NOX formation even with low-Btu gas. A premix
burner offers the potential for off-stoichiometric combustion and greatly
reduced NOX.
8. For the selected low-temperature desulfurization system (Selexol),
removal of sulfur to very low levels can double equipment cost if it is neces-
sary to design for COS removal.
9. Any advantages in power plant performance or cost resulting from the use
of an ambient temperature Selexol process rather than a refrigerated Selexol
process are lost when sulfur removal below the level of 1.2 Ib S02/10" Btu
is desired.
10. Data needed to predict the fate of the numerous trace elements in the coal
feed are scarce. While not definitive, it appears that water scrubbing of
the fuel gas would be desirable in controlling the discharge of the compounds
of these elements.
29
-------
SECTION 4
RECOMMENDATIONS FOR FURTHER WORK
1, Further investigations are required to identify the cost and effect on
overall power plant performance of sulfur removal to levels of 95 percent
and beyond.
2. An investigation should be conducted to compare the effects of removing
fuel-bound nitrogen prior to combustion with the effects of combustion
modifications. Of interest are the amounts of nitrogen removed, the cost,
effect on overall system performance and the reduction in nitrogen oxide
emissions.
30
-------
SECTION 5
OVERVIEW OF GASIFICATION AND CLEANUP PROCESSES
FOR USE WITH COMBINED-CYCLE POWER PLANTS
INTRODUCTION
The basic technology for production of low-Btu gas is available.
Gasification has been in common use since the 1800's. A number of desulfuriza-
tion processes have been developed for gas cleanup. These are designed for use
in sweetening natural gas as well as in the cleanup of the various types of
low-Btu or synthesis gas.
In addition to these existing processes, a number of advanced types are in
various stages of development. In general, they are intended to provide for
more efficient gasification of widely varying coal types while having increased
throughput and consequently lower cost. This section presents a brief descrip-
tion of the basic types of gasifiers and cleanup processes that are potentially
available and evaluates their applicability for use in combined cycle power
generation.
j>ystem Arrangement
An integrated system is shown schematically in Figure 5-1. It shows the
primary interconnections between the gas turbine, steam bottoming cycle and
coal processing or gasification and cleanup system. The gasification process
requires air which can be supplied from the compressor discharge of the gas
turbine. Using this high pressure air permits pressurized operation of the
entire fuel processing system. This is generally beneficial since it reduces
component size and improves the performance of many desulfurization systems.
Also, many gasification processes need steam which can be supplied by extraction
from the steam cycle after having done some useful work. Low pressure steam
can also be extracted for use in various parts of the cleanup process.
Because the gasification process takes place at elevated temperature, the
vessel is generally cooled. Heat rejected there can be used to raise high-
pressure steam for the steam cycle. Additional steam may be raised when
cooling the hot fuel gas leaving the gasifier. This would be the case if a
low-temperature cleanup process were to be used.
31
-------
BASIC GASIFICATION COAL COMBINED CYCLE POWER SYSTEM
HIGH PRESSURE STEAM
FOR STEAM CYCLE
COAL
oo
S3
COM-
PRESSOR
GASIFICATION
& CLEANUP
GAS TURBINE
BURNER
TURBINE
GAS TURBINE
EXHAUST
o
u
01
t>
LOW PRESSURE STEAM
STEAM CYCLE
WASTE HEAT
RECOVERY
GENERATOR
STACK
GAS
C
01
-------
The remainder of the plant is a conventional combined-cycle. Gas turbine
exhaust is used to raise some steam for the steam turbine and to superheat that
steam as well as that raised in the fuel processing system. The steam condi-
tions are dependent on the available temperature levels and vary with gas
turbine and fuel processing system parameters.
Gasification and Cleanup
The fuel processing part of the power plant is shown in simplified form
in Figure 5-2. It assumes the use of a low-temperature cleanup system. In the
coal handling section, coal is processed as required by the gasifier. It is
fed to the gasifier at ambient condition. The gasifier must include lock-
hoppers or other devices to feed the coal to the gasifier at pressure. Steam
and air are fed to the gasifier as required and jacket heat is used to generate
high-pressure steam for the steam cycle. Ash removed from the gasifier may
require cooling or other handling facilities not shown here.
Gas leaving the gasifier must be cooled to permit removal of ammonia,
hydrogen sulfide and particulates, at least in the case of low-temperature
systems. In the cooling process heat may be used to raise steam and/or to
reheat the clean fuel gas. This is discussed in Section 8 under thermal
integration.
The water scrub consists of a particulate removal section and a section
designed for ammonia removal. The particulate slurry produced, in most cases,
can be recycled through the gasifier. The sour water, containing ammonia,
carbon dioxide and hydrogen sulfide, is processed to remove ammonia. The
remaining gases are sent to the sulfur recovery unit.
Sulfur in the fuel gas is removed in the H2S absorber. The degree of
sulfur removal and the amount of other gases absorbed differs among the various
processes. These absorbed gases are stripped and sent to the sulfur recovery
unit for recovery of elemental sulfur. While the degree of sulfur removal from
the fuel gas is of prime importance, the amount of steam or power needed in the
stripping operation and the concentration of I^S in the feed to the sulfur
recovery unit are important in measuring performance of the desulfurization
process.
GASIFICATION
A high pressure, air-blown gasifier appears to be a logical choice for
use with a combined cycle generating system. However, other types can be
adapted and one goal of the current study is to compare the merits of air-vs.
oxygen-blown operation.
Chemical Reactions
In the production of low-Btu gas, the highest yield would be achieved if
all of the carbon were converted to carbon monoxide and all hydrogen were
released in the molecular form. In practice many competing reactions exist and
33
-------
GENERAL FUEL PROCESSING SCHEMATIC
HIGH PRESSURE
STEAM
HIGH PRESSURE
STEAM
STEAM
COAL
FEED
COAL
HANDLING
GASIFICATION
GAS COOLING
WATER SCRUB
AIR ASH
\ f
" BFW
BFW
SOUR
WATER
H2S CLEAN
ABSORBER
PARTICULATE
SLURRY
PROCESS
STEAM
CONDENSATE
SOUR WATER
STRIPPER
t/-\IVI .
i I
AMMONIA
RECOVERY
ACID
GAS
BFW STEAM
it
SULFUR
RECOVERY
ACID
GAS
H2S STRIPPER
T I
VI
CD
I
O
CO
I
01
PROCESS CONDENSATE
STEAM
AMMONIA
ELEMENTAL
SULFUR
n
PROCESS CONDENSATE
STEAM
ro
-------
both C02 and H20 are formed. The process is quite complex and can be viewed as
a number of sequential reactions (Reference 5-1). First, some of the fuel is
oxidized to form C02 and
2H2 + 02 * 2H20 (5~2)
These reactions release a significant amount of heat which enables the
highly endothermic gasification reactions to take place.
C + H20 * CO + H2 (5~3)
C + C02 + 2CO (5"4)
These are commonly known as the steam-carbon and Boudouard reactions.
Boudouard equilibrium is generally used as a test for possible carbon formation.
It determines the minimum C02 concentration.
In addition, methane can be formed by the reaction of hydrogen with carbon
or carbon monoxide. Heating of the coal also results in devolatilization which
produces carbon, methane and other hydrocarbons. At low temperatures, these
hydrocarbons do not react and appear as heavy tars and oils in the gasifier
product.
When these gases coexist at high temperature, equilibrium is generally
determined by the water gas shift reaction,
H2 + C02 > H20 + CO (5~5)
which reaction is generally close to equilibrium.
Equilibrium concentration of methane can be calculated from the gas phase
reaction:
CH^ + H20 * CO + 3H2 (5~6)
However, measured concentrations are generally greater than would be predicted
by equilibrium.
Formation of methane in the gasifier is of major importance when the
objective of the gasification plant is to produce pipeline quality gas. When
the production of fuel gas is the objective, the major consideration is the
efficient conversion of the energy in the coal to chemical energy (heating
value) of the product gas. This can be achieved with gas containing hydrogen
and carbon monoxide as the major combustible components rather than methane.
35
-------
Sulfur and Nitrogen Compounds —
Reactions involving sulfur and nitrogen are important from an
environmental standpoint. Sulfur content of American coal varies widely.
Western coals are generally low in sulfur while Eastern coals have relatively
high sulfur content. Because of the reducing atmosphere in the gasifier, most
of the sulfur is converted to hydrogen sulfide and carbonyl sulfide. The
proportions of each are related by the reaction:
COS + H20 > C02 + H2S (5~7)
At equilibrium, H2S generally accounts for approximatly 95 percent of the
total sulfur.
The conversion of coal nitrogen during gasification is more complex.
Where oil and tar are produced, some of the nitrogen will appear in pyridine
compounds. The remainder of the nitrogen will be converted to ammonia or molec-
ular nitrogen. At high temperatures, some hydrogen cyanide may be formed.
Chemical Equilibrium—
Gasification processes vary greatly in the degree to which the reactions
approach equilibrium. High-temperature and catalytic action within the gasi-
ifier result in a close approach to equilibrium. Sophisticated computational
techniques are available to calculate the equilibrium concentration of virtu-
ally all potential products. The results of such a calculation are presented
in Section 8 for the gasifiers studied under this contract.
Oxygen vs Air-Blown Gasification
The basic feed streams needed for gasification are coal, steam and a
source of oxygen. Most processes can be designed to run with either pure
oxygen or air as the source of oxygen. Processes used in producing pipeline
gas are oxygen-blown and use steam to generate hydrogen for methanation. When
oxygen is used, reactions are sufficiently exothermic as to make the addition
of steam necessary to keep the temperature within practical limits . With air,
the nitrogen diluent provides a quenching effect and complete gasification can
theoretically be achieved with minimal steam used to reduce reaction tempera-
ture. The role of steam is critical in combined cycle,power generation and is
addressed in Section 8.
The choice between oxygen and air is usually determined by the end use of
the gas. Oxygen is required when pipeline or high-Btu gas is desired. There,
the presence of nitrogen would be unacceptable. Where transportation and
interchangeability with natural gas are not a requirement, then medium- and
low-Btu gas can be considered. Oxygen-blown operation will produce a medium-
Btu gas with a heating value in the range of 300 Btu/SCF. An example of the
difference in composition is given in Table 5-1 for a BCR-type gasifier. For
air-blown operation, a value of 150 Btu/SCF is typical since the nitrogen dilu-
ent accounts for approximately 50 percent of the product gas. The primary
fuel constituents of both are hydrogen and carbon monoxide which each have a
heating value of approximately 325 Btu/SCF.
36
-------
TABLE 5-1
COMPARISON OF AIR- VS OXYGEN-BLOWN GASIFICATION
BCR-Type Gasifier
Coal - Illinois No. 6 - 700,000 Ib/hr
AIR-BLOWN
OXYGENS-BLOWN
CH4
H2
CO
C02
H2S
COS
N2
NH3
H20
HHV-Btu/SCF
Oxidant/Coal Ratio
Steam/Coal Ratio
Transport Gas/Coal
Cold Gas Efficiency
Mols/Hr
3,775
15,315
32,190
3,396
751
76
53,753
479
2,213
111,948
Ratio
Mol %
3.37
13.68
28.75
3.03
0.67
0.07
48.02
0.43
1.98
171.2
2.78
0.144
0.088
83%
Mols/Hr
4,522
22,098
26,459
9,369
770
77
336
479
10,787
74,897
Mol %
6.04
29.50
35.33
12.51
1.03
0.10
0.45
0.64
14.40
270.8
0.594
0.597
0.088
85.7%
37
-------
When used as a fuel, the primary difference between low- and medium-Btu
gas is the flame temperature. Because of the nitrogen diluent, the storchio-
metric flame temperature of the low-Btu gas can be as much as 800 F lower
than that of the medium Btu gas. Of interest is the fact that the stoichio-
metric flame temperature of methane or high-Btu gas is lower than that of
medium Btu gas. Methane requires four times as much-air as does an equivalent
volume of CO or H2 (the major constituents of medium-Btu gas) while having a
heating value of just over three times that of those fuels.
Flame temperature can be very important when retrofitting existing indus
trial processes. For that purpose medium-Btu gas would be selected. Also, i^
has the advantage of being more economically transported. However, in the gas
turbine, flame temperature is not important, in fact the high flame tempera-
ture associated with medium-Btu gas may result in a NOX problem discussed
in Section 9, and selection can be based on other factors.
Air-blown operation is generally favored because oxidant for the gasifi-
cation process can be bled from the gas turbine compressor. The nitrogen is
returned to the gas turbine with the fuel gas and produces useful work as it
is expanded through the turbine. Oxygen-blown operation requires a separate
oxygen compressor in addition to an air separation plant. It does, however,
result in smaller cleanup equipment size since the fuel gas volume is approxi
mately half that of the air-blown system. This, plus the possible synergism
of industrial utilization of the fuel gas, could make the oxygen-blown gasifi*
a viable alternative.
Types of Gasifiers
While the reactions are common to all gasifiers, i.e., partial combustion
followed by reducing type reactions, the means of achieving them differ wideljj
Gasifiers can be classified in four main categories:
(a) Moving Bed
(b) Fluidized Bed
(c) Entrained Bed
(d) Molten Bath
Within each category, gasifier characteristics can vary due to operating
conditions (temperature and pressure), method of coal feeding, method of ash
removal and other important aspects. However, it is a convenient categoriza-
tion and represents a primary design difference between the processes.
Moving Bed Gasifier—
This category is typified by the Lurgi gasifier although, in this report
data for the Bureau of Mines stirred bed gasifier have been used. It is some
times referred to as a "Fixed Bed" as it involves a bed of coal supported by
a grate. Oxidant and steam are introduced below the grate and move rather
slowly up through the bed. Product gas is removed from the top while ash is
discharged at the bottom.
38
-------
Combustion temperature in "dry bottom" operation is controlled to below
the ash fusion point. Temperature of the gas leaving the top of the unit is
approximately 1000 F. Because of the relatively low temperatures, the efflu-
ent gas contains significant amounts of condensible oil and tar vapors. As a
result, the gas is usually quenched with water to condense and remove these
constituents. This is a major disadvantage in combined-cycle operation since
most of the sensible heat in the off-gas is lost.
The process is sensitive to type and size of the coal feed. Caking
coals tend to form clinkers and a stirring device is provided to break up any
lumps that may form. Fines are undesirable and must be briquetted or used
elsewhere. Coal is fed by lock-hopper. The high coal inventory in the gasi-
fier provides stability against variations in coal flow and composition.
A recent advance in moving bed gasifiers, which is now in the development
stage, is the "slagging" mode of operation. In this operation, the gasifier
is oxygen-blown and the combustion zone is controlled at a temperature which
is above the melting point of the ash. Ash is removed from the gasifier in
molten form. This type of operation reduces the steam requirement and in-
creases the gasification rate relative to "dry bottom" gasifiers.
Fluidized Bed Gasifier—
As the name implies, solids in the gasifier are suspended in a fluid
state by an upflow of gas through the gasifier. As is characteristic of a
fluidized bed, temperature is quite uniform and there are no hot spots. Thus,
temperatures can be maintained below the ash fusion point while being high
enough to minimize tars and oils in the off-gas (generally 1600-1800 F).
However, this is a problem in many fluid bed concepts. Also, since there must
be a considerable carbon inventory, ash disposal is a problem. The fluid bed
is well mixed and ash removal is a matter of extracting a part of the well-
mixed bed. As a result, the ash removal stream contains a substantial amount
of unreacted carbon.
A problem common to gasifiers of this type is the inability to use
strongly caking coal without pretreatment. However, they are tolerant to var-
iations in coal size and require less coal processing equipment.
Many of the problems associated with fluidized bed gasification are
claimed to be eliminated in the ash-agglomerating gasifier under development
at IGT. Relatively low amounts of carbon are contained in the ash and no tar
is expected in the off-gas. Operation of that gasifier is described in Sec-
tion 6.
Entrained Flow Gasifiers—
Gasification by means of an entrained flow process involves suspension of
relatively small coal or coal char particles in a high velocity gaseous medium
Residence time of the coal particles in the gasifier is relatively low and high
temperatures are used to maximize gasification rate. As a result, entrained
flow gasifiers operate under ash slagging conditions.
39
-------
The atmospheric pressure Koppers-Totzek (K-T) gasifier is probably the
best working example of a single-stage entrained flow process. It is oxygen-
blown and reaction temperature is in excess of 2600 F. The high temperature
results in high reaction rates and the off-gas is reported to be very close to
equilibrium. The coal feed is pulverized and the gasifier is not affected by
the use of caking coals. Application of the K-T process to combined cycle
power generation is discussed in Section 6.
In a two-stage entrained flow gasifier, coal is injected into an upper
stage where it is partially gasified by hot gases produced in the lower stage.
Gas and char leave the upper stage at a temperature near 1800 F. Char is
separated from the gas and injected into the lower stage together with steam
and an oxygen containing gas. The char is partially combusted at temperatures
of 2800 to 3000 F. Molten ash slag is separated from the gases formed and is
removed from the bottom of the lower stage.
As with the single-stage gasifier, any grade of coal can be used since
the pulverized particles are in the reaction zone. The carbon content in the
ash is quite low because of the high reaction temperature. The capacity of
entrained flow gasifiers is much higher than that of moving or fluidized bed
gasifiers since the flow is not restricted by bed characteristics.
Molten Bath Gasifiers—
In this type of process, the reactants are contacted in a high tempera-
ture molten fluid. The melt serves to disperse the reactants and provides
heat transfer and thermal storage for the process. It can also have a cata-
lytic effect on gasifier reactions and/or combine chemically to remove sulfur
from the product gas. As in the entrained flow process, the gasifier can
handle all types of coal. It has the additional advantage of being relativel}
insensitive to coal size, thus reducing the processing requirements. Ash is
withdrawn with the melt (the ash forms the molten bath in some cases) and con'
tains on the order of 1 percent unreacted carbon.
Gas leaving the process is free of tars but may contain vapors from the
melt. In the case of a molten salt type gasifier, the vapors could be harmful
and provision must be made to insure their removal.
CLEANUP SYSTEMS
Fuel gas cleanup systems consist of processes for sulfur removal, nitroge'
compound (ammonia) removal, and particulate removal. Sulfur removal processes
are generally divided into two categories. These are:
1) Low-temperature desulfurization
2) High-temperature desulfurization
Low-temperature desulfurization processes require cooling of the dirty gas and
operate at temperatures below 250 F. At these temperatures, it is possible to
use water as a means of particulate and ammonia removal. High-temperature
40
-------
iesulfurization processes operate at or near gasifier exit conditions and
their use will require other means of nitrogen and particulate control.
Control of nitrogen and particulates in systems with high-temperature desul-
furization is discussed in Section 8.
Low-Temperature Desulfurization Processes
Low-temperature processes for desulfurizing raw producer gas are commer-
cially available and have been widely used for natural gas sweetening and
treating synthesis gas in the chemical process industry; e.g., the manufacture
of ammonia, methanol and oxo-alcohols. These systems normally operate below
250 F and are commonly classified into the following categories:
Chemical Solvent Processes
Physical Solvent
Direct Conversion Processes
Dry Bed Processes
A literature survey resulted in a list of 38 low-temperature processes.
A summary of the important characteristics of some representative processes is
given in Table 5-2. Figure 5-3 illustrates the schematic for a tyical absorp-
tion - stripping process for low-temperature acid gas removal.
Chemical Solvent Processes —
Chemical solvent processes employ aqueous solutions of organic and/or
inorganic agents which are capable of forming "complexes" with the acid gas
components, notably H2S and CC-2, present in the raw gas stream. The
absorption solution is regenerated by decomposing the "complex" at elevated
temperature thereby releasing the acid gases for subsequent recovery. The
solution is recycled for further absorption. These processes are essentially
insensitive to the partial pressure of acid gases in the feed and generally
exhibit little or no selective absorption of t^S over carbon dioxide. The
chemical processes may be sub-divided into those using amine scrubbing solu-
tions and those based on alkali scrubbing solutions.
The principal reactions involved in gas sweetening with amine solutions
(10-30 percent weight) may be represented as:
RNH2 + H2S + RNH3 HS (5~8)
RNH2 + C02 + H20 - RNH3 HC03 (5~9>
Monoethanol amine (MEA) will easily reduce the l^S content below 4 ppm;
however, it is not considered selective, even though the rate for C02 absorp-
tion is less than for H2S. The principal disadvantage of MEA is that it will
react with COS and CS2 forming nonregenerable compounds. Diethanolamine
(DBA) will not react with these contaminants and is favored for service where
41
-------
TABLE 5-2
LOW TEMPERATURE CLEANUP PROCESSES
Basis: 8400 tons/day Illinois No. 6 Coal Fed to BCR Gasilier, or 6700 ppm of Influent H2S
Process
Chemical
solvent type
1. MEA
Z DEA
3. TEA
4. Alkuid
6. Benfield
6. Catacarb
Physical
solvent type
7. Sulfinol
8. Selexol
9. Ractiiol
Direct
conversion
10. Strat-
ford
11. Town-
send
Drybed type
12. Iron
sponga
Absorbent
Monoetha-
nolamina
Oiethanol
amine
Trietha-
nolamina
Potassium
dimethyl
amino
acetate
Activated
potassium
carbonate
solution
Activated
potassium
carbonate
solution
Sulfolane
+
Dilsopro-
panoamina
Polyethyl-
ene glycol
ether
Methanol
Na,CO,+
anthraquin-
onesul-
fonic acid
Triethylene
glycol
Hydrated
Fe,0,
Type of
Absorbent
Aqueous
solution
Aqueous
solution
Aqueous
solution
Aqueous
solution
Aqueous
solution
Aqueous
solution
Organic
solvent
Organic
solvent
Organic
solvent
Alkaline
solution
Aqueous
solution
Fixed
bed
Temp.
° F
80 to
120
100 to
130
100 to
160
70 to
120
ISO to
250
ISO to
250
80 to
120
20 to
80
<0
150 to
250
70to
100
Pressure
Insensitive
to variation
in pressure
Insensitive
to variation
in pressure
Insensitive
to variation
in pressure
Insensitive
to variation
in pressure
1 - 80 atm
Insensitive
to variation
in pressure
generally
> 300 psi
High
pressure
preferred
Efficiency of S Removal
%H, Sin-
fluent
99
99
99
99
99
99
99
99
99
99.9
99.9
99
Effluent
H,S
ppm
~100
~100
-100
-100
H.S.
+ COS
~10O
H,S
+ COS
~100
H.S
+ COS
-100
H,S
+ COS
-100
~100
-10
-10
H,S
+ COS
-100
Absorbent
Characteristics
Life
Unlim-
ited
No
degra.
dation
Regenera-
tion
Thermal
Thermal
Thermal
With
steam
With
steam
With
steam
Low
pressure
heating
or with
steam
Selectivity
toward
Forms non-
regen. comp.
with COS,
CS,
Absorbs CO, ,
does not
absorb
COS, CS,
H,S
H,S
H,S
is high
H.S-par-
tial also
absorbs
COS, CS,
H,S.and
also absorbs
COS, CS,
and mer-
captans
H, S also
absorbs
COS
H,S
H,S.
H,S
H, S and
also towards
COS. CS,
and mer-
captant
Make up
rate
50 to
100%
<5%
<5%
<5%
SOW
100%
Form of
Sulfur
Recovery
AsH.S
gas
AsH,S
gas
AsH.S
gas
AsH,S
gas
AlH.S
gas
AsH,S
gas
AsH.S
gas
AsH.S
gas
Elemen-
tal
sulfur
Elemen-
tal
sul f u r
Elemen-
tal
sulfur
Status
Commercial
Commercial
Commercial
Commercial
Commercial
Commercial
Commercial
Commercial
Commercial
Commercial
Commercial
42
-------
TYPICAL LOW-TEMPERATURE ACID GAS REMOVAL UNIT
TREATED GAS «*
3>
Ji
I
^
tt>
to
RAW GAS
FEED
RICH
SOLVENT
*- ACID GAS
STEAM
p
Ol
I
OL'
-------
COS and CS2 are likely to be present. Like MEA, DBA solutions are not
selective for H2S and will seldom reduce the H2S content below 100 ppm. Ter-
tiary amines, such as triethanolamine and methyl-diethanolamine, while not as
reactive as the other amines, have the advantage of being selective towards
H2S removal. The tertiary amines are two to four times more costly and find
little application in industrial gas sweetening.
The alkali scrubbing system may be represented by the following chemical
react ions:
M2C03 + H2S * MRS + MHC03 (5-10)
M2C03 + C02 + H20 * 2 MHC03 (5-11)
A number of processes have been based on these reactions, the Benfield and
Catacarb representing the most advanced versions.
The earlier processes, such as Seaboard and Vacuum Carbonate, were based
on dilute solutions of sodium carbonate (3-4 percent weight) and were capable
of removing 80-90 percent of the H2S. Regeneration in the Seaboard process
was by air resulting in a dilute acid gas stream while the Vacuum Carbonate
system used vacuum regeneration with steam. These processes were superseded
by the hot potassium carbonate system. In the "hot pot" processes, an aqueous
solution of 25-35 weight percent I0j C03 is used to absorb acid gases at
temperatures in the range of 200-250 F. With low H2S/C02 ratios, the process
is capable of sweetening the gas to 5 ppm. Some selective H2S absorption
can be achieved by taking advantage of the relatively slow rate for C02
absorption. In addition to removing H2S and C02, the process can remove
COS and CS2 by hydrolysis of these components to C02 and H2S. The
Catacarb and Benfield processes are improved versions of the Bureau of Mines
"hot pot" systems. They employ activators to increase the rate of absorption
thereby decreasing the required solution circulation rate. Disadvantages of
the hot potassium carbonate systems are the relatively high steam consumption
for regeneration, sensitivity to operating pressure and possible inability to
handle mercaptans and thiophene.
Not listed in Table 5-2, the tripotassium phosphate process was developed
by the Shell Oil Company specifically for H2S removal via the reaction:
K3P04 + H2S + K2HP04 + KHS (5-12)
The nonvolatility of the agent, its nonreactivity with COS and CS2, and
partial selectivity toward H2S in the presence of C02 gives it certain
advantages over the araine systems. However, when operated for high H2S
selectivity, the process only gives about 90 percent removal efficiencies.
Conversely, with high H2S removal the steam consumption becomes excessive
due to CC>2 absorption.
44
-------
Physical Solvent Processes—
Physical solvent processes all use organic solvents to remove acid gases.
The physical absorption is directly proportional to the partial pressure of
the acid gas components. These processes are most applicable to high-pressure
gas treating where appreciable quantities of sour gases are present. After
absorption, the "loaded" solvent is regenerated by heat and/or pressure
reduction giving a concentrated stream of H-2^ plus CC>2 and a recyclable
lean solvent. Due to the higher solubility of I^S in these organic sol-
vents, selective absorption of H2S over CC>2 can be achieved. In general,
these processes have two major disadvantages: the solvents have a great
affinity for absorbing heavy hydrocarbons (€5+) which contaminate the gas
stream fed to sulfur recovery units; and the solvents are quite expensive so
that large solvent losses cannot be tolerated.
As a group, these processes were developed for bulk removal of acid
gases but, for low H2S concentration, they are capable of giving a sweetened
gas having less than 5 ppm l^S. In order to maximize the solubility of
acid gases and minimize solvent loss through vaporization, the processes are
generally operated at or below ambient temperature. In addition to removing
^28 and CC>2, the solvent processes are all capable of removing COS, CS2
and mercaptans without solvent degradation. They also dehydrate the gas to a
low dew point. The low heats of solution for acid gases result in appreciably
lower steam requirements for solvent regeneration compared to steam require-
ments for the chemical solvents.
The Sulfinol process is unique in that it combines the characteristics
°f a solvent process and an amine process. The physical absorbent, Sulfolane,
gives high acid gas loadings at high acid gas partial pressures, giving it
bulk removal capacity and the chemical absorbent, DIPA, reduces residual
acid gases to very low values. However, the presence of the chemical solvent
reduces the H2S selective H2S absorption capability for this system
compared to the straight solvent processes.
Direct Conversion Processes—
Two types of processes fall into this category:
a. those based on oxidation reduction reactions, and
b. those based on the stoichiometric reaction of H2S with SC>2
in the presence of a solvent.
The first type involves the absorption of H2S in alkaline solutions con-
taining oxygen carriers. The H2S is subsequently oxidized to elemental
sulfur by air fed to the regenerator. There the sulfur product is flotated
and collected as a froth at the regenerated solution interface. Processes of
this type are in common use in Europe for removal of H^S and sulfur recovery
from manufactured gases and coke-oven gas. The Ferrox and Manchester processes
di-isopropanolamine
45
-------
employ a suspension of iron oxide in an aqueous solution of sodium carbonate
to absorb H2S. With multistage absorption, essentially complete removal of
H2S is obtainable; however, the product is of low quality due to salt contam-
ination. Chemical replacement costs are high. Both the Thylox and Giammarco
Vetrocoke (References 5-2, 5-3) processes use alkaline solutions of arsenates
and are capable of reducing the H2S to less than 1 ppm. Partial removal of
COS, CS2, and mercaptans is also possible. Again the sulfur product is con-
taminated and the use of arsenates makes these processes potentially hazardous
The Stretford and Takahax processes are similar in that alkaline solutions of
quinone sulfonic acids are employed. The addition of vanadium salts increase
the rate of oxidation of hydrosulfide to sulfur resulting in higher solution
loadings. Close to 100 percent H2S removal is claimed for these processes as
is a high purity (99%) sulfur product. However, substantial amounts of thio-
sulfates are formed resulting in sludge deposition and corresponding chemical
makeup .
Generally speaking, the low solution loadings exhibited by this group of
processes make them uneconomical for treating large volumes or very sour gas
streams. They are best suited for sour gases containing less than 1.0 percent
H2S with sulfur production under 20 tons/day. These processes, as well as
those in the following group, are almost totally selective for H^S removal.
In the second group of direct conversion processes are those in which H2
is absorbed in a solvent and converted to elemental sulfur by the Glaus type
reaction with S02 .
(5-13)
2 H S + SO + 3 S + H_0
The Townsend process uses an aqueous solution of an organic solvent, such as
triethylene glycol, to sweeten the gas, dehydrate the gas and convert H2S to
elemental sulfur. A portion of the product sulfur is burned to S02 which is
absorbed by fresh solvent and the S02 rich solvent is used to contact the sour
gas. The IFF, Nalco, and Deal processes operate in a similar manner employing
other solvents. While a high purity (99.7%) sulfur product is claimed, none
of these processes have been commercialized.
Dry Bed Processes —
These sweetening processes are based on absorption of acid gases by a
fixed bed of solid absorbent. Due to their low absorbent loading, they are
applicable to gases containing low concentration of H2S and mercaptans, per-
haps less than 500 ppm. These processes can be subdivided into the iron oxide
processes and the various molecular sieve processes.
The iron oxide or dry box process is one of the oldest processes known
for removing sulfur compounds from industrial gases. In the iron sponge
systems, wood shavings impregnated with hydrated ferric oxide are used to
absorb H^S :
(5-14)
2 Fe203 + 6 H2S * 2 Fe^ + 6 H20
46
-------
Regeneration of the absorbent is carried out with air:
2 Fe2S3 + 302 - 2 Fe^ + 6S (5-15)
This process is best suited for small to medium gas volumes with low sulfur
contents; otherwise the sponge bed life would be too short to be economical.
The process is selective towards H2S and mercaptans and will partially
remove COS and CS2. Sweetened gas of less than 5 ppm H2S is easily obtained
However, sulfur recovery would not be economical when using the iron sponge
system.
Molecular sieves can be made to have pore sizes which will permit
selective absorption of H2S over C02- These processes are characterized
by the various regeneration schemes employed; i.e., via hot combustion gases,
not S(>2 gas as in the Haines process, or hot air. In the latter two modes,
elemental sulfur is produced via the oxidation of the absorbed H2S. The
sieve processes also appear to be economically attractive for small to medium
gas volumes having low H2S content. Additionally, for efficient H2S removal,
the raw sour gas should have a water content below 20 Ib/MMSCF since water
will also be absorbed by the molecular sieve structure.
Selection Considerations—
Low-temperature desulfurization systems for application in low-Btu fuel
gas plants will have to treat large volumes of sour gas, 500-1000 MMSCFD,
having total sulfur content in the range of 0.4 to 1.0 percent by weight. In
addition to H2S, the raw gas will contain CC>2, COS, CS2, probably mercaptans,
cyanides and heavy hydrocarbons. Of the types of processes described above,
it is evident that the liquid scrubbing processes using physical solvents and
some chemical solvents are the best suited. These processes are currently
available and can easily reduce the sulfur content of the gas to 100 ppm. Such
a gas, when combusted, would result in S02 emissions well below present EPA
regulations for conventional steam stations. As such, these processes are cap-
able for serving both first-generation and second-generation coal gasification
Plants and should meet any future regulations for S02 emissions.
High Temperature Cleanup Systems
High-temperature systems for sulfur removal are not presently available
ln commercial scale although there are several in various stages of develop-
ment. Active work involves use of limestone and dolomites which have potential
ln the range of 1500-2000 F. Other systems receiving attention employ iron
oxides, molten salts, and liquid metals. These systems operate by chemical
reaction of the absorbent with sulfur compounds in the gas, forming the cor-
responding metal sulfides. The degree of desulfurization attainable depends
ln part on the chemical equilibria for the particular system at the operating
conditions. As with low-temperature processes, economics dictate that the
sulfided absorbent be regenerated for reuse.
-------
The only commercial experience with high-temperature desulfurization
reported in the literature is that of the Frodingham Desulfurization Process
(Refs. 5-4, 5-5). This process employed fluidized beds of pulverized iron
oxide operating at 650 F to 800 F. In the early 1960's, a commercial plant
treating 32 MMCFD of crude coke oven gas containing 1.0 percent H2S was
operated at the Exeter Works of the South Western Gas Board. Essentially
complete (99.9%) removal of H2S was achieved with 90 percent removal of all
organic sulfur compounds other than thiophene. The sulfided absorbent was
regenerated with air at 1000-1100 F resulting in a S02 stream which was sub-
sequently converted to sulfuric acid. Major difficulties were experienced in
the solids handling system which produced fine oxide dust resulting in exces-
sive losses of the absorbent. In addition, operation of the sulfuric acid
plant was erratic due to low S02 concentrations in the regenerator off-gas.
Several processes currently under development which may prove commercially
viable for use with second generation gasification systems are listed in
Table 5-3.
Bureau of Mines Sintered Iron Oxide Process — (References 5-6 5-7 5-8
5-9.) '
This process, under development at the Morgantown Energy Research Center,
is based on a sintered absorbent consisting of a mixture of iron oxide (Fe203)
and fly ash. This sorbent satisfied the primary requirements for high-temper-
ature sulfur removal in that it is readily available and inexpensive, it has
reasonable absorption capacity for sulfur, can be regenerated for repeated use
and is resistant to fusion and disintegration over the operating temperature
range of 1000-1500 F. The absorbent is prepared by mixing iron oxide with "as
received" fly ash to a total iron oxide content of about 35 percent. Iron
oxide contents above 40 percent were unsatisfactory because the fusion temper-
ature was lowered within the operating range. The mixture is extruded into
1/4" x 3/8" pellets and then sintered to develop the required hardness.
Reaction chemistry is reported in Reference 5-7 as follows. During
absorption, two iron sulfides are produced, FeS and FeS2, with the empir-
ical composition approaching FeS]_ 5.
(5-16)
During regeneration, air oxidizes the FeS/FeS2 to Fe203 and S02 .
6FeS1 5 + 1302 * 2Fe304 + 9SC>2 (5-17)
04 + 1/2 02
Excess oxygen in the air from the reaction that produces the Fe304 converts
it. to Fe203.
48
-------
TABLE 5-3
HIGH TEMPERATURE CLEANUP PROCESSES
Basis: 8400 tons/day Illinois No. 6 Coal Fed to BCR Gasificr, or 6700 ppm of Influent H2S
Process
1 . Bureau
of Mines
2. Babcock
and
Wilcox
3. CONOCO
4. Air prod-
ucts
5. Battelle
North-
west
6. IGT-
Meis-
sner
Absorbent
Sintered
pellets of
Fe,O, (25%)
and fly ash
Fe,03
Half calcined
dolomite
Calcined
dolomite
Molten
carbonates
<15%CaC03)
Molten metal
(proprietary)
Type of
Bed
Fixed
bed
Fixed
bed
Fluidized
bed
Fixed
bed
Solution
Splashing
contact
•
Temp.
°F
1000 to
1500
800 to
1200
1500 to
1800
1 600 to
2000
1 1 00 to
1700
900
Pressure
Insensitive
to variation
in pressure
Insensitive
to variation
in pressure
~200psia
H, S removal
is high at
low pressure
Insensitive
to variation
in pressure
Atmospheric
HjS removal
is high
allow
pressure.
5-6 psig
Efficiency of S
Removal
%H2S In-
fluent
-95
-99
-95
-95
-98
Effluent
H2S
ppm
-350
-75
-350
-350
-150
Absorbent
Characteristics
Life
>174
cycles
Wt loss
<5%
mini-
mum
5-6
cycles
Regener-
ation
With air
10-13%
with
steam
and CO,
80-90%
with
steam
and CO,
With
steam
and CO,
Elec-
troly-
tic
Selec-
tivity
toward
H,S,
COS
H,S.
COS
H,S.
COS
H,S.
COS,
fly ash
H,S.
COS
Make up
rate
<5%
1%of
circula-
tion
rate
Form of
Sulfur
Recovery
AsSOj
gas
As
12-15%
SO, gas
AsH,S
gas to
Claus
process
AsHjS
gas to
Claus
process
AsHjS
gas to
Claus
process
Energy
Required
Elec.
kw
96.360
9830
Oth-
er
stu
Status
Pilot
Experi-
mental
Pilot
Aban-
doned
Pilot
Concep-
tual
-------
A practical problem during regeneration is the burnup of accumulated
carbon deposits within the bed. This produces locally high temperatures that
tend to harm the absorbent. Recent reports indicate that the fly ash can be
replaced by silica to produce a more rugged material.
Concern over the possible absorption of hydrogen or the conversion of
hydrogen and carbon monoxide to water and carbon dioxide by reaction with the
ferric oxide is expressed in Reference 5-10. Equilibrium calculations indi-
cate the potential for high sulfur removal levels, especially at low steam
feed rates. Carbonyl sulfide absorption also appears to be practical.
An area that has not been investigated is the potential effect of the
iron oxide bed on ammonia in the fuel gas. It is possible that it will cata-
lyze the decomposition of ammonia which is generally assumed to be present in
concentrations well in excess of those at equilibrium. This could be a very
attractive feature since the presence of ammonia and consequent NOX produc-
tion is a major drawback in high-temperature desulfurization.
Conoco Half Calcined Dolomite Process—
This process (Reference 5-11) evolved as an adaption of the C02 Accep-
tor Process (Reference 5-12) for producing low-Btu fuel gas from coal and
incorporates the use of a half-calcined dolomite acceptor for sulfur capture
as studied by Squires and coworkers (References 5-13 to 5-16). Basically the
process chemistry involves the following reaction:
[CaC03 • MgO] + HS2 •* [CaS • MgO] + H02 + C02 (5-19)
A maximum operating temperature for this process is imposed by the par-
tial pressure of carbon dioxide in the gas phase; i.e., the temperature should
not exceed that at which the C02 partial pressure is equal to the decomposi-
tion pressure for CaCC>3 via the following endothermic reaction:
CaC03 + CaO + C02 (5-20)
Should that temperature be exceeded, C02 would be released into the clean
fuel gas and the regeneration of the unreacted CaO will make the process more
complex. The effect of this temperature limitation on desulfurization perfor-
mance is discussed in Section 8 where the various factors are quantified.
Although no data have been reported for COS adsorption by half-calcined
dolomite, high COS removal efficiencies are predicted thermodynamically
according to the reaction:
[CaC03 - MgO] + COS + [CaS • MgO] + 2 C02 (5-21)
50
-------
The process as described by Conoco involves desulfurizing the raw gas
in a fluidized bed of half-calcined dolomite acceptor at 1600-1700 F, accord-
ing to reaction (5-19). The sulfided acceptor is regenerated by the addition
of steam and C02 at reduced temperature, thereby reversing the absorption
reaction. Regeneration is conducted in a fluidized bed at around 1300 F giv-
ing a dilute H2S off-gas,.less than 10 percent (volume). Because the low I^S
off-gas content prohibits the direct use of a vapor-phase Glaus unit for sulfur
recovery, Conoco is proposing the use of a liquid phase sulfur recovery system
based on the Wachenroeder reaction
2H2S + H2S03 + 3S + 31^0 (5-22)
All important features of the process have been confirmed (Ref. 5-11) in
bench-scale studies. There remains to be determined the effect of scale-up
as well as the effect of ammonia and the ability to remove particulates and
alkali fumes from the effluent. The process is described in detail in Section
° and the effect of gasifier operating conditions on sulfur removal examined
in Section 8.
Air Products Fully Calcined Dolomite Process—
As with the Conoco process, this system employs dolomite as the sulfur
acceptor. However, the acceptor is in the fully calcined form; i.e.,
[CaO • MgO], and the process therefore consists of three steps: absorption,
regeneration and calcination.
Absorption of hydrogen sulfide takes place at around 1600-1700 F via the
reaction with calcined dolomite:
[CaO -. MgO] + H£ * [CaS • MgO] (5-24)
For this system, the H2S removal efficiency is independent of pressure but,
for a given gas composition, it decreases with increasing temperature. From
a practical viewpoint, there is a minimum temperature at which this process
should operate; that being the temperature at which the decomposition pres-
sure of CaC03 equals the partial pressure of C02 in the gas. Below this
temperature carbon dioxide will also be absorbed according to the reaction.
CaO + CC-2 * CaC03
While the carbonate can also react with H2 in accordance with reaction
v5-l9)} this is detrimental for two reasons:
a. Additional acceptor calciner capacity is required with an associated
increase in heat input, and
b. The large exothermic heat of C02 absorption, 75,000 Btu/mol, will
necessitate some means for heat removal from the absorption bed.
51
-------
Like the Bureau of Mines and Conoco processes, COS removal appears to
be thermodynamically attractive. Residual COS content may be estimated from
the chemical equilibrium associated with the absorption reaction:
[CaO • MgO] + COS * [CaS • MgO] + C02 (5-25)
Regeneration of the sulfided acceptor is conducted similarly to the
Conoco process. Steam and carbon dioxide are reacted with the sulfided
dolomite at 1100-1200 F resulting in a dilute I^S off-gas and half-calcined
dolomite.
[CaS • MgO] + H20 + C02 -* [CaC03 • MgO] + t^S (5-26)
Hydrogen sulfide in the regenerated off-gas may be converted to elemental
sulfur via the liquid-phase Wachenroeder process or first concentrated and
then fed to vapor-phase Claus units.
The regenerated half-calcined dolomite must be calcined before
recycle to the subsequent absorption cycle. Calcination is effected at 1900
F with air to drive off carbon dioxide:
[CaC03 • MgO] > [CaO • MgO] + C02 (5-27)
This reaction is endothermic and requires the use of fuel to preheat
the air. Excessive temperature, above 2000 F, during calcination can result
in deactivation of the acceptor.
There is currently no development activity in progress.
IGT-Meissner Process —
This process is being developed by the Institute of Gas Technology in
conjunction with its U-Gas Process. The process, still in the conceptual
stage, utilizes a splashing molten metal-gas contact to remove H2S from the
gas. The contact results in the formation of a metal sulfide which is then
decomposed electrolytically to release H2S and regenerate the molten metal
for recycle. The operating temperature is 900 F and a high sulfur removal
efficiency (98 percent) is projected. The molten metal absorbent is propri-
etary. The estimated characteristics given in Table 5-3 are preliminary
(Reference 5-17). Further development is being directed toward establishing
mass transfer rates.
Battelle Northwest Process —
The Battelle process (Reference 5-18) utilizes calcium carbonate,
CaC03, dissolved in a tertiary mixture of alkali metal carbonates to
remove H2S at high temperature. The tertiary carbonate system, consisting
of potassium carbonate, lithium carbonate, and sodium carbonate, has a
52
-------
eutectic melting point around 750 F. Under operating conditions it contains
about 15 mol percent CaC03. Besides removing sulfur compounds from the gas
stream, this solvent will also scrub out the fly ash constituents from the raw
gas.
The system under study at Battelle contacts the molten salt and raw gas
in a co-current flow venturi scrubber at temperatures from 1100 to 1700 F.
Hydrogen sulfide is removed from the gas by chemical reaction with the car-
bonate solvent:
("i
CaC03 + H2S - CaS + C02 + H20
Unfortunately, due to the chemical complexity of the molten salt system, the
equilibrium constant cannot be accurately predicted. Observed K values have
been a factor of ten below the calculated values. Qualitatively, the I^S
removal efficiency improves with temperature and is inversely proportional to
pressure. Experimental data at atmospheric pressure have indicated high H2S
removal, >94 percent, at salt loadings under 50 percent of capacity. Regen-
eration of the salt is conducted with steam and C02 at 1000-1100 F giving
an off-gas having H2S concentrations suitable as feed to a Glaus unit for
recovery of elemental sulfur.
This process is presently in the pilot plant stage where it will be
demonstrated on an actual gasifier raw gas at low pressure, 0-10 psig. It is
doubtful that this system can achieve high H2S removal efficiencies at high
pressure and its ability to handle other sulfur contaminants is yet to be
demonstrated. Materials of construction for a commercial unit will also
present a formidable problem.
Babcock and Wilcox Process—
This process is chemically similar to the Bureau of Mines process in
that it utilizes iron oxide to remove H2S from the gas at high temperatures.
The difference lies in the material used by the two processes. While the
Bureau of Mines' process uses a sintered material made from iron oxide and
fly ash, the Babcock and Wilcox process starts out with carbon steel and gen-
erates an iron oxide scale on the steel surface which is then used as the
desulfurization agent.
While the reactions are not pressure sensitive, the equipment lends
itself to low (ambient) pressure operation. The desulfurizer uses a modified
regenerative type air heater and is referred to as a "regenerative desulfur-
izer11. Sulfur removal efficiency greater than 90 percent is projected. Re-
generant gas is expected to contain 10 to 13 percent by volume S(>2. The
process concept has been demonstrated on bench scale equipment and a hardware
design has been developed.
53
-------
REFERENCES
5-1 Archer, D. H., et al.: Coal Gasification for Clean Power Production.
Clean Fuels From Coal Symposium II, IGT, Chicago, Illinois, June
1975.
5-2 Maddox, R. N.: Gas and Liquid Sweetening. Second Edition, J. M.
Campbell, CPS (1974).
5-3 Jennett, E.: Assessment of the Giammarco Vetrocoke Process. Proc.
Gas Conditioning Conf., University of Oklahoma (1964).
5-4 Reeve, L.: Desulfurizatipn of Coke Oven Gas at Appleby-Frodingham.
Journal of the Institute of Fuel, p. 319, July 1958.
5-5 Bureau, A. C., and M. J. F. Olden: Operation of the Frodingham
Desulfurization Plant at Exeter. The Chemical Engineer (206), CE55-62,
March 1967.
5-6 Abel, VJ. T., F. G. Shultz and P. F. Langdon, Removal of Hydrogen
Sulfide from Hot Producer Gas, Report No. RI 7947 (1974).
5-7 Oldaker, E. C., A. M. Poston, and W. L. Farrior, Removal of Hydrogen
Sulfide from Hot Low Btu Gas with Iron Oxide - Fly Ash Sorbents,
Report No. MERC/TPR-75/1 (1975)
5-8 Oldaker, E. C., A. M. Poston, and W. L. Farrior, Hydrogen Sulfide
Removal from Hot Producer Gas eith a Solid Fly Ash Iron Oxide ,
Absorbent, Report No. MERC/TPR-75/2 (1975)
5-9 Farrior, W. L., A. M. Poston, and E. C. Oldaker, Reggenerable Iron
Oxide Silica Sorbent for the the Removal of H2S from Hot Producer
Gas, Paper presented at Fourth Energy Resources Conference Univer-
sity of Kentucky (January, 1976).
5-10 Jones, C. H. and J. M. Donohue: Comparative Evaluation of High
and Low Temperature Gas Cleaning for Coal Gasification - Combined
Cycle Power Systems. EPRI AF-416, April 1977.
54
-------
REFERENCES (Continued)
5-11 Curran, G. P., B. J. Koch, B. Pasek, M. Pell, and E. Gorin: High
Temperature Desulfurization of Low-Btu Gas. EPA-600/7-77-031,
(NTIS No. PB271-008), April 1977.
5-12 Curran, G. P., et al.: C02 Acceptor Gasification Process. Adv.
Chem. Series No. 69 "Fuel Gasification", pp. 141-146 (1967).
5-13 Squires, A. M., et al.: Desulfurization of Fuels with Calcined
Dolomite. AIChE Symposium Series 115, Vol. 67 (1971).
5-14 Ruth, L. A., et al.: Desulfurization of Fuels with Half Calcined
Dolomite. Environmental Science and Technology, Vol. 6. No. 12:1009.
November 1972.
5-15 Squires, A. M.: Process for Desulfurizing Fuels. U.S. Patent 3,481,
834, December 2, 1969.
5-16 Squires, A. M.: Cyclic Use of Calcined Dolomite to Desulfurize
Fuels Undergoing Gasification. Adv. Chem. Series No. 69, "Fuel
Gasification", p.p. 205-229 (1967).
5-17 Letter Communication November 15, 1974. Dennis Duncan, Institute
of Gas Technology, to M. S. Dandavati, Hittman Associates, Inc.
5-18 Moore, R. H.: Removal of Sulfur Compounds and Fly Ash from Low Btu
Gas. Battelle Pacific Northwest Laboratories. BN-SA-210 (1973).
55
-------
SECTION 6
DETAILED DESCRIPTION OF
SELECTED GASIFICATION AND CLEANUP PROCESSES
INTRODUCTION
This section presents both a description of and the rationale
Leading to the selection of particular gasification and desulfurization
processes for integration with a combined-cycle power system. The general
approach was to select gasifiers that were representative of the various
general types. These were combined with selected high- and low-temperature
cleanup systems for comparison of the resulting integrated power plants.
SELECTION OF PROCESSES
Gas_if_j.er_g_
During this and earlier phases, at least one representative of each
of the major gasifier types was selected for study. The coal gasifiers
that have been used in defining the integrated power plants are:
1. Bureau of Mines Stirred Bed Gasifier
2. U-Gas Ash Agglomerating Gasifier
3. Koppers-Totzek Entrained Flow Gasifier
4. BCR-type, Two-Stage Entrained Flow Gasifier
5. Molten Salt Gasifier
Of these, all but the Koppers-Totzek process operate above atmo-
spheric pressure. While pressurized operation presents a number of
operational problems, it is a definite advantage when attempting integration
with a combined-cycle power plant. This fact showed up in both the per-
formance and cost of power from a plant using an oxygen-blown K-T gasifier.
While the process is very desirable in all other respects, the need to
compress the product gas resulted in a heat rate approximately 10 percent
worse than the closest competitor. Also, the combined effect of performance
and cost of air-separation and other additional equipment resulted in a
cost of electricity nearly 30 percent greater than the competition. It was
concluded that the process does not lend itself to use with combined-cycle
power generation.
-------
The Bu Mines Gasifier, which was selected as typical of the "Fixed
Bed" type, does not permit recovery of the sensible heat in the gasifier
effluent. This is due to the need to quench that stream and condense out
the tars and oils that would otherwise foul the downstream equipment. In
the current phase, emphasis was placed on overall performance improvement
by the use of high performance power system technology. This includes a
high temperature gas turbine with a 2400 psi reheat steam bottoming cycle.
Efforts at integration of the Bu Mines gasifier with such a power plant
showed them to be incompatible due to the lack of heat recovery from the
raw fuel gas and the large quantity of steam required by the gasifier. It
was concluded that the performance estimates contained in the previous
report (Reference 6-1) are representative of the capabilities of systems
using this particular gasifier. The estimated heat rates were 9200 Btu/kWh
for low- and 8600 Btu/kWh for high-temperature cleanup. These can be
compared with current estimates of 7900 and 7500 Btu/kWh with low- and
high-temperature cleanup respectively for the advanced type gasifiers.
Therefore, no further work was done on the power plants having a Bu Mines-
type gasifier.
While the oxygen-blown, slagging Lurgi gasifier promises much better perfor-
mance than the Bu Mines process, the only oxygen-blown gasifier that was used was
the BCR-type two stage process. It was studied for comparison with its air-blown
counterpart. Gasifier operating characteristics did not differ as widely and the
effect of air separation and medium- vs low-Btu gas could be better appreciated.
For integration with a combined-cycle power plant, the important
gasifier characteristics are:
1. No condensible hydrocarbons in effluent
2. Low gasifier exit temperature
3. Low steam feed rate.
The remaining gasifiers, U-Gas, BCR-type two-stage and the molten salt-type
appear to be good selections on those bases. In terms of power plant heat
rate, cost of electricity and emissions, the estimated performance of all
three is the same within the accuracy of the study data.
Low-Temperature Desulfurization Processes
The majority of low-temperature desulfurization processes that were
identified are commercially operative and could be used with both current
and advanced gasification processes. In selecting those most applicable in
treating coal derived fuel gas, the following factors were considered:
a. Sulfur removal capabilities, not only with respect to H2S, but
also other sulfur compounds such as COS and CS2-
57
-------
b. Selective absorption of sulfur compounds over carbon dioxide.
Since CC>2 need not be removed from fuel gas intended for use in
advanced power cycles, absorption of C02 represents an increased
operating load on the system.
c. Type of absorbent insofar as the treated fuel gas may be contamin-
ated by entrained or volatilized solvent which could be detrimental
to downstream system components.
d. The system's tolerance to other contaminants present in the raw
fuel gas such as ammonia, cyanides, phenols, and tars.
e. Overall energy requirements and operating cost.
A preliminary screening, reported in Reference 6-2, led to the selec-
tion of three processes for detailed comparison in an integrated system.
These are the Benfield chemical solvent process and the Selexol and Rectisol
physical solvent processes. This comparison involved a preliminary per-
formance estimate of an overall integrated power system using each of the
three processes.
For the purpose of this comparison, the entrained-flow BCR type
gasifier was used under the following operating conditions:
Coal Type
Illinois No. 6
Feed, Ib/hr 2000
Gasifier Operation
Temperature, F 1800
Pressure, psia 500
Air, Ib/lb, coal 3.422
Steam, Ib/lb coal . 0.567
Gasifier Production
Net Gas, SCF/lb coal 74.53
Slag, Ib/lb coal 0.087
Raw Gas Analysis, Volume %
N ' CO C02 H2 CH4 H2S COS NH3 1^0
47.70 16.74 8.84 11.98 3.14 0.46 0.10 0.38 10.66
58
-------
There is a net heat rejection from the gasifier of 0.384 MM Btu/hr
-"vis" Telenetl%c>°treaf^TT^^IT ,'TellTT' """ d"ul^"tion
The acid gas from the regenerator was designed for a^igh^'lO*
nrJHoS^nnrior»f-*-?i*-t/-**-i^«.«ui_' ,_ i _ °
the recoverv of ^1 ^m^n*-ai o,,^fur
ri,m* p the>^8rated cleanup systems is shown in
Figure 6-1 Raw producer gas from the BCR gasification system at 1750 F is
first cooled in the heat recovery section to about 300 F. The heat extracted
i- available for regenerate heating of the clean fuel gas, boiler feed
water preheating and steam generation. ThlS sectton vaS optimized to give
he highest overall plant efficiency for each case. The gas is further
cooled below the dew point to about 120 F via direct water scrubbing which
also removes most of the ammonia present in the raw gas. Sour water from
this scrubbing operation is first steam stripped and the stripped gases are
ent to an ammonia recovery section. The cooled, ammonia-free producer gas
desulfurized ln the low-temperature acid gas removal section producing a
clean fuel gas and a regenerated acid gas stream containing 13-22 percent
hydrogen sulfide. Sulfur is recovered from the latter stream and from
the ammonia recovery off-gas stream in a vapor phase Claus unit A
tailgas treating section is included and recovers a minimum of 99 percent
Complete material balances for the Selexol Benfield and Rectisol
processes in this application are given in Reference 6-1. Performance of
the cleanup processes are summarized in Table 6-1 together with the utility
requirements.
As expected for a chemical solvent system, the Benfield. process
requires 2.5 to 3 times more low-pressure steam for solvent regeneration
than do the two physical solvent systems. This is partially offset by the
higher power consumption required by the Selexol and Rectisol processes for
mechanical refrigeration to obtain subambient operating tempratures.
Using these data, a preliminary analysis of the integrated power
stations gave the following relative performance characteristics:
Cleanup
System
Selexol
Benfield
Rectisol
Clean Fuel
HHV
Btu/SCF
142.6
136.5
144.0
Fuel Gas
Gasifier
Cold Gas
Efficiency
% Coal HHV
73.4
74.9
73.4
Overall
Station
Efficiency
% Coal HHV
31.2
30.5
31.4
59
-------
INTEGRATED LOW-TEMPERATURE CLEANUP SYSTEM
BCR
GASIFIER
HEAT
RECOVERY
02
-•- 01
T
SLAG
COAL TRANSPORT GAS
~"1 REM
ACID GAS
REMOVAL
7\ CONDENSATE
GAS
SCRUBBER
CONDENSATE
*_ AMMONIA
CLEAN
FUEL GAS
STACK
P
'O5
I
-------
H20
H2S
M. W.
TABLE 6-1
SUMMARY OF LOW- TEMPERATURE INTEGRATED SYSTEMS
Process Selexol Benfield Rectisol
Feed Streams C1)
BCR Gas
Flow, mph ^26.U5 ^26.1*5 U26.1*5
T p 1750 1750 1750
P psia ^50 ^50 U50
Product Stream
Sulfur
Flow, rb/hr 7^.9 7l+. 9 76.2
T F 300 300 300
P psia " ~
Transport Gas
Flow, mols/hr
T p
P psia
Product Gas
Flow, mols/hr
T P
P psia
N
CO
C02
H
3^.20
300
550
332. 8U
100
U30
5^-70
19-20
8.66
13-73
3-59
0.11
100 ppm
25-^7
1^2.6
3U.20
300
550
3^9-^7
250
U30
52-32
18.36
7.02
13-1^
3.M*
5.66
0.06
100 ppm
2^.81
136.5
3^.21
300
550
329.U6
90
1*30
55-20
19-37
7.88
13-86
3.63
0.06
10 ppm
25-31
lUU.O
HffV Btu/scf
Utilities.
Cooling Duty, MMBtu/hr 2.858 3-287 3-008
Steam @ 1300 psia, Ib/hr 106.5 10°'5 lo6'5
@. 65 psia, Ib/hr 1020.0 250^ 7 779-6
Electric Power, kw 60.8 25-5 Ul.8
Boiler Feed Water, Ib/hr 219.2 222.8 223-1
Steam Condensate, Ib/hr 1233-3 2718.3 99^-8
Peed Gas Cooling, MMBtu/hr^ U.862 5-156 U.6l6
Based on 2000 Ib/hr Coal Feed to Gasifier
Available for STM Generation and/or Gas Reheat
61
-------
The efficiencies above do not necessarily represent the optimum power
cycle configuration and fuel gas regenerative heating that can be achieved.
However, in terms of their relative magnitudes for a particular system
configuration, it can be concluded that the three low-temperature desul-
furization processes can give comparable performance.
For a variety of reasons, including the relatively harmless solvent
and moderate absorber temperature, the Selexol process was selected as the
typical low-temperature process. Sulfur recovery is accomplished in a
Glaus plant with a Beavon unit for tail gascleanup.
High-Temperature Desulfurization Processes
Four high-temperature processes, the Conoco Half-Calcined Dolomite, Air
Products Fully-Calcined Dolomite, Bureau of Mines Sintered Iron Oxide, and
Battelle Molten Salt processes were compared on an integrated power plant basis in
a manner similar to the low-temperature comparison. The comparison, complete witl
process flow sheets is reported in Reference 6-1. Results of the comparison are
shown below:
Gasifier
and Cleanup Overall
Clean Fuel Cold Gas Station
Cleanup HHV Temp. Efficiency Efficiency
System Btu/SCF _F - % Coal HHV % Coal HHV
Conoco 125.2 1610 76.1 36.0
Air Products
Case 1 143.1 1550 53.8 29.1
Case 2 126.5 1630 73.7 35.5
Bureau of Mines 124.8 1000 72.5 31.6
Battelle 125.3 1610 76.1 34.9
For the Air Products fully calcined dolomite process, the large
variation in performance is caused by absorption of C02 at the lower
desulfurizer temperature. The reaction is exothermic. Thus, the bed must
be cooled and re-calcination uses a like amount of energy.
The Bureau of Mines process suffers from the need to cool the raw gas
down to 1000 F for desulfurization. Also, since the sulfur comes off the
regeneration step as SC>2, the recovery of elemental sulfur is more
difficult. However, it is ideally suited for the moving bed or Bu Mines-
type gas ifier for which it is being developed.
Of the two remaining systems, despite their relatively comparable
performance, the Conoco process is preferred. The consequences of potential
carryover of alkali metals are too severe to consider the use of a high-
temperature molten carbonate system without a downstream water wash.
62
-------
GASIFICATION PROCESS DESCRIPTION
For each process considered in this report, flow sheets and mass
balances are given in Section 11. The coal composition given in Table 6-2
applies to all but the molten salt gasifier. While data for that process
are based on an Illinois No.6 coal, the composition varies to some extent.
U-Gas Gasification Process
The U-Gas process is being developed by the Institute of Gas Technology.
*t is a fluidized bed system that can produce either low- or raedium-Btu gas
with either air- or oxygen-blown operation.
The use of a fluidized bed has many inherent advantages. In particular
the bed acts as a catalyst for the gasification reactions and is expected
to permit operation at relatively low temperature while completely gasifying
the feed. The U-gas process is distinguished from other fluid bed processes
ln that it utilizes an "ash agglomeration" technique to concentrate the ash
and remove it with minimum carbon content while operating the bed with a
relatively high carbon content.
The gasifier is shown schematically in Figure 6-2 from Reference
°~3. The key to operation of the gasifier is the agglomeration and
separation of the low carbon content ash from the bed. The U-gas gasifier
accomplishes this and maintains a bed of approximately 70 percent carbon and
30 percent ash by proper design and operation of the grid and the fines
return system in the bottom of the gasifier. The grid is sloped toward
one or more inverted cones contained in the grid. Part of the fluidizing
steam and air flow through the grid while the remaining fluidizing gas
flows upward at high velocity through the throat at the cone apex. The
ratio of steam to air in the fluidizing gas fed to the cone is chosen so
that the resulting submerged jet in the cone is hotter than its surroundings.
The temperature of the jet is maintained near the softening point of the
ash^particles for the specific coal being gasified. Ash particles prefer-
entially stick together, and the agglomerates grow until they are heavy
enough to move downward counter to the force of the gas stream in the apex
°f the cone. Thus, they fall out of the fluidized bed.
Fines elutriated from the fluid bed are separated from the product gas
by two cyclones in series: the first inside the gasifier and the second
outside. Fines removed by the first cyclone are returned to the bed by a
standleg. Fines removed by the external cyclone are entrained in the inlet
air/steam to the gasifier grid cone where they are instantly gasified
because of the high temperature in that region.
With Illinois No.6 coal feed, the base case composition out of the
gasifier is given in Table 6-3. All of the sulfur in the coal is assumed
to appear as l^S or COS and the two compounds are assumed to be at
equilibrium with 94 percent of the total being I^S. Ammonia is also
63
-------
TABLE 6-2
FEED COAL COMPOSITION
Constituent
Carbon
Hydrogen
Sulfur
Nitrogen
Ash
Water
Weight %
67.4
5 i
2 Q
9 . 6
-^ 2
8>7
4 2
Higher Heating Value - 12,200 Btu/lb
-------
FIG. 6-2
ASH-AGGLOMERATING GASIFIER
CRUSHED
COAL
FEED-LOCK
HOPPER
PRETREATMENT
(IF NECESSARY)
STEAM
GENERATION
GASIFIER
AIR (OR OXYGEN)
AND STEAM
AIR (OR OXYGEN)
AND STEAM
RAW GAS TO
PURIFICATION
SECOND-STAGE
DUST REMOVAL
ASH-LOCK
HOPPER
WATER
ASH/WATER
65
-------
TABLE 6-3
U-GAS GASIFIER EFFLUENT
Mol Percent
CO 18.0
C02 8.1*7
H20 10.62
CH^ 3.03
N2 1*3.82
NH3 0.03
H2S 0.59
COS 0.02
Mol. Wt 23.98
HHV - Btu/SCF 138.8
Cold gas efficiency (2) 80.1 percent
Steam/coal ratio 0.557
Air/coal ratio 3.01
66
-------
assumed to appear in equilibrium concentration due to the catalytic effect
°* the fluid bed. The relatively low level would permit the use of high
emPerature cleanup without ammonia removal.
Raw coal is crushed to 1/4 in. size. The feed may contain up to
Percent < 200 mesh material as generated in the crushing step. Noncaking,
^'bituminous coalg Qr Agnate can be fed directly to the gasifier from
he crusher. Caking coals (Eastern bituminous, for example) must at
Present be pretreated by contact with air in a fluidized bed operating at
8asifier pressure and 700 to 800 F. An oxidized outer layer forms on the
°al Particles, and this prevents agglomeration and possible blockage in
the
, Heat evolved during pretreatment is removed by generating steam in
Coti^Fansfer coils which are immersed in the fluidized bed pretreater.
ai that has been pretreated is fed to the gasifier. Off-gases are fed to
the b°ttom of the gasifier to destroy all tar and oils that evolve during
Pretreating process.
is Tne gasifier is a refractory lined, hot-metal-wall vessel. Steam
generated to provide cooling for the pressure vessel while the fluid bed
ceaction takes place at temperatures as high as 2000 F. System pressure
^ be as low as 100 psia (minimum level is determined by economics) but in
i8 aPPlication is 400 psia, as determined by gas turbine pressure ratio
a Pressure drop in the fuel processing system and burner fuel distribution
The operating conditions within the gasifier result in a
°f tar8 and °ils' ThUS n° Special cleanup Procedures are
Entrined Flow Process
R The two-stage entrained flow gasifier developed by Bituminous Coal
Search, inc. (BCR) was selected as the second-generation gasification
^tem for this study. An oxygen-blown, 120 ton/day pilot plant to produce
j * 10& SCFD of SNG ig under development at Homer City, Pennsylvania. De
"
8
c"8n
<>f an air-blown version is currently underway at the Foster Wheeler
in * schematic diagram of the BCR gasifier with its auxiliaries is shown
pul 8Ure 6-3 (Reference 6-4). Run-of-the-mine coal is crushed, dried, and
ve;Yerized until 70 percent passes through a 200 mesh screen. The pul-
fr^ed coal " metered from the feed hopper into hot transport gas recycled
ga°Vhe gas purification section and then fed into the upper stage of the
W ler' In this stage, the coal is entrained by the hot gases from the
te! Sta8e> is rapidly heated and devolatilized. At the high (1800 F)
h JPerature, the products are expected to be methane, carbon monoxide,
*>at!?8en and unreacted char. The gases leave the upper stage at approxi-
Cel
l7 1800 F.
67
-------
BCR ENTRAINED- FLOW GASIFIER
FIG. 6-3
COAL
TRANSPORT
GAS
GAS
SLAG
HOPPERS
V
/
Y
STEAM
SLAG
R1-19-3
68
-------
residual char is removed from the gas by cyclone separators
p recycled via superheated steam to the lower stage of the gasifier.
of the char then reacts with steam and air at 2800 F to form synthesis
and molten slag. The hot synthesis gas, containing unreacted char,
s to the upper stage for reaction with coal as described above. Molten
g collects and drains from the bottom of the lower stage into the slag
where it is water quenched.
Overall, the gasifier reactions are endothermic and the process heat
r Cement is supplied by combustion of char with air. The air rate is
th to maintain tne operating temperature in the upper stage while
in «-k°Wer 8^a8e temperature is controlled by steam addition. Temperature
da lower stage is fairly critical since too high a temperature will
fr 8^e tne refractory and too low a temperature will cause the ash slag to
eeze and accumulate,
Ot the current study, estimated performance of the gasifier was
ba airved from work done by Fluor under contract to EPRI. The estimate is
^nt °n minimum steam feed and is designed to improve performance of the
ai ^rated power plant. This is discussed in Section 8. Estimates for both
t*h'i anc* oxygen~blown operation have been given in Table 5-1. Note that
is 6 C^e air-blown case has a relatively low steam/coal ratio, that ratio
th a^>?rox:'-mately 0.6 for oxygen-blown operation. This is due to the lack of
Illtrogen diluent which helps to quench the reaction in the first stage.
j.,
PUte oxygen» a significant amount of steam is needed to control the
low e oxygen» a sgn
er stage temperature.
fo le c°ld gas efficiency is 83 percent for air- and 85.7 percent
n °xygen-blown operation. While data defining this gas composition are
te .available, it is expected that at the high temperature and with sufficient
te **ence time, formation of tars, oils and phenols will be avoided. The
th ant gas is expected to be quite clean relative to that encountered in
e Coving bed type of process.
Gasification
cat. he molten salt gasification process is well suited to this appli-
to 1°n*. Ifc uses no steam and the catalytic effect of the melt, in addition
ca , llllmizing reactor volume, should tend to produce levels of ammonia and
the °ny* Sulfide that are close to equilibrium. Because a large part of
are Su*fur is retained in the melt, total sulfur compounds in the raw gas
of _at: relatively low levels and at equilibrium COS would be on the order
ca, . PPm. Other attributes of the process include the ability to handle
Co.}11^ coals and the ability to operate with either coarse or finely ground
of • Because both sulfur and ash are retained in the melt, a large part
reg e system is devoted to processing the melt to recover sulfur and
fuel erate tne sodium carbonate. This process also incorporates a final
tiQ ^as desulfurization step so that the use of a conventional desulfuriza-
n Process is not required. Thus the process consists of several sections:
69
-------
1. Gasification
2. Gas cleanup
3. Ash removal
4. Salt recovery
Gasification and Gas Cleanup—
In the gasifier, as indicated in Figure 6-4, coal is suspended in a
bed of molten sodium carbonate and is gasified with air to produce low-Btu
fuel gas. Coal mixed with recycled sodium carbonate is continuously
charged to the gasifier, and the vigorous turbulence in the molten salt bed
results in uniform distribution of coal throughout the bed. Air is injected
into the molten bed and the partial oxidation of coal produces a low-Btu
fuel gas free of coal ash. The ash is retained in the melt. Furthermore,
it is also free of ammonia, tars, and heavy hydrocarbons. These constituents
are destroyed in the molten sodium carbonate. Under conditions of low-Btu
gasification, a high degree of sulfur retention as Na2§ results in the
melt such that the raw fuel gas is low in l^S. Trace amounts of alkali
metals must be removed from the gas prior to firing so that a low-temperature
scrub is necessary.
A number of complex chemical reactions, many of which involve cata-
lytic interaction with the molten sodium carbonate, occur in the bed.
Table 6-4 indicates, in simplified form, the most imporant of these.
Partial oxidation, reaction (6-1), is the main route for gasification of
carbon. If the air to coal ratio is too high, formation of C0£ by
reaction (6-8) becomes important. This is undesirable because the fuel gas
quality suffers and the reactor tends to run hotter owing to the highly
exothermic nature of reaction (6-8). In accordance with reactions (6-2)
and (6-3) the volatile matter in the coal is broken down by the catalytic
action of the molten salt and as a result the low-Btu fuel gas is free of
tars and heavy hydrocarbons. It is reported that ammonia is also destroyed
by the molten salt (Reference 6-5). Equilibrium calculations performed
during this study did show some ammonia, however, it was at a relatively
low concentration (180 ppm) and if not removed would not exceed allowable
emission levels. The water-gas shift reaction, reaction (6-4), is maintain6*1
at equilibrium, and CC>2 and 1^0 concentrations in the low-Btu gas
product are low, being driven there by the reducing environment. Reaction
(6-5) indicates an important feature of the process. Sulfur in the coal
reacts with the melt to produce Na2$. An equilibrium does exist between
Na2§ in the melt and t^S in the gas (reaction (6-6)). However, because
CC-2 and 1^0 concentrations are low, most of the sulfur is retained in
the melt. Since I^S levels are low resulting COS levels are very low
(reaction (6-7)). Reactions (6-9), (6-10), and (6-11) are of minor signi-
ficance in low-Btu gasification.
To remove ash and sulfur from the gasifier, a purge stream of melt is
withdrawn and treated. Coal ash is separated, sulfur is removed as H2S,
and sodium carbonate is recovered for recycling to the gasifier. The ash
level in the melt can be allowed to accumulate to as much as 30 percent,
thereby minimizing the purge treating requirements.
70
-------
FIG. 6-4
LOW-BTU MOLTEN SALT COAL GASIFICATION PROCESS
AIR
COAL + CARBONATE
I
LOW-BTU
FUEL GAS
(CO, H2, N2)
MELT PURGE
(ASH, SODIUM CARBONATE, SULFUR)
71
-------
TABLE 6-k
MOLTEN CARBONATE REACTIONS
C + 1/2 02 •* CO (6-1)
CmHn (COAL) ->- VOLATILES + C (6-2)
VOLATILES CO + H2 + CH^ (6-3)
CO + H20 J C02 + H2 (6-1+)
S (IN COAL) + MELT •* Na2S (6-5)
Na2S + C02 + H20 ? Na2C03 + H2S (6-6)
H2S + CO 2 COS + H2 (6-7)
C + 02 -*- C02 (6-8)
C + HO -*• H2 + CO (6-9)
C + 2H2 J City (6-10)
Na2C03 + H20 ^ 2 NaOH + C02 (6-11)
72
-------
tK °Perating conditions of the molten salt gasifier must be such that
k bed is above the fusion temperature of sodium carbonate (1570 F), yet
m w the temperature where excessive vaporization of sodium carbonate and
],_ rial corrosion problems arise. Normal operation is maintained between
ant* 1800 F. The temperature can be conveniently controlled by
p . tln8 the air to coal ratio. Since the process is basically one of
J-al oxidation, increasing the air to coal ratio increases the gasifier
Perature at the expense of fuel gas heating value. The operating
pr Ure is determined primarily by fuel gas pressure requirements.
tj_ 8ure does not have a significant effect on the fuel gas composi-
(J ' Table 6-5 shows the reference coal composition and the fuel gas
resulting from gasification at 411 psia and 1800 F with 1000 F
_. control alkali metals, two stages of gas cleaning are employed. In
ent *rst stage, a fluid bed cooler is used to quench the gas so the
lned and
and vaporized sodium salts condense and fuse on the surface of
in the fluid bed. By providing condensation sites while
Or *n8 the gas, the fluid bed prevents formation of a sodium carbonate
s0j: lum °xide fume. In addition to removing vaporized and entrained
lay, m Carbonate, the fluid bed cooler also serves to prevent carbon
gas Wtl> The reaction 2CO -*• C02 +C is avoided by rapidly quenching the
heat • tlle same t"ne a hi8h level of recovery of the fuel gas sensible
to r ls achieved. In the second stage of cleaning, a water wash is employed
m°ve finai traces of alkali metals.
; l§ure 6-5 shows a flow schematic of the gasification and gas clean-up
n fns *n the plant. Ground coal mixed with makeup sodium carbonate is
erred to lock hoppers which alternately feed the gasifier. To remove
d sulfur from the system, a purge stream of melt is withdrawn and
° the ash removal and salt recovery section.
Th
d K raw fuel 8as f^om the gasifier is quenched in the riser leg of the
in tk cooler to 1200 F or lower. Contact with the large mass of solids
bed rapidly quenches the gas. Entrained or vaporized sodium
cotu*ensed and solidified on the surface of sand particles in the
The fluid bed rises through a boiler where heat is removed and
8°Ud °WS ^nto the disengager where the cooled fuel gas separates from the
U and flows out through a system of cyclones. To remove the trapped
g Carbonate from the fluid bed cooler, an intermittent purge is taken.
if Particles, now coated with sodium carbonate, are recycled to the
-where the sodium carbonate is recovered. The sand particles are
ln the gasifier purge. To replenish the fluid bed an intermittent
°* sand is required.
th-e fluid bed cooler the clean fuel gas is further cooled by heat
with the clean gas in the recuperator. It then passes through an
73
-------
TABLE 6-5
MOLTEN SALT GASIFIER
COAL AND RAW GAS COMPOSITION
Illinois No. 6
Carbon (wt. %} 59-62
Hydrogen (wt. %) ^.U6
Oxygen (wt. %) 8.H5
Nitrogen (wt. Jf) 1.00
Sulfur (wt. JS) 3.10
Ash (wt. %) 10.37
Moisture (wt. %} ^ 13.00
Higher Heating Value (Btu/lb) 10755
Raw Fuel Gas Composition
CO (vol. %} 28.33
H2 (vol. %) 13.79
CHi^ (vol. %) 1.50
C02 (vol. %) 3.08
H20 (vol. 55) 2.35
N2 (vol. %} 50.85
H2S (vol. %} 0.10
Higher Heating Value (Btu/Scf) 152
74
-------
GASIFICATION AND GAS CLEANUP
SAND
COAL +
CARBONATE
J
FLUIDIZED BED
COOLER AND DISENGAGER
X*
\
BFW
WATER WASH
STEAM
MELT PURGE TO
ASH REMOVAL
CLEAN
FUELGAS
RECOVERED
RECUPERATOR
H20
Na2CO3
SOLUTION
SOLUTION
MAKE UP CO2 ABSORBER
( SALT RECOVERY SECTION)
H2OMAKE UP TO
ASH REMOVAL
SECTION
O
O)
!
tn
-------
absorber where a portion of the C02 in the gas is absorbed to Provjd®.
makeup C02 for the salt recovery section. The sodium carbonate soiuci
in the absorber reacts with both C02 and H2S thereby reducing the H2S ^
content of the gas stream from 1000 down to less than 200 ppm. Partic ^
carried out of the fluid bed cooler will be removed in the absorber.
the makeup C02 absorber, the fuel gas goes to the water wash which ^
removes final traces of alkali metals. The clean fuel gas is then rene
in the recuperator and sent to the gas turbine combustor.
For the system studied here, the detailed stream compositions are
presented in Section 11.
Ash-Removal and Salt Recovery —
The ash removal and salt recovery section treats the purge stream
which removes ash and sulfur from the gasifier. By employing technology^
similar to that in practice in the pulp and paper industry, ash and su
are removed and the sodium carbonate is recovered. Figure 6-6 is a si
plified flow diagram of the ash removal and salt recovery section.
The melt purge withdrawn from the gasifier at 1800 F and 400 psia 1
first quenched to 430 F by a circulating stream of sodium bicarbonate
the quench tank. The outlet stream from the quench tank flows to a hy
clone where solids are separated from the solution which is recycled
the quench tank. The quenched slurry is then flashed to produce 2o p
steam at 250 F.
From the flash tank, the slurry is sent to the dissolving tank
it is mixed with additional sodium bicarbonate solution and makeup
which includes all the process condensate collected during the proces •
The sodium salts are thus allowed to dissolve. The sodium sulfide an
sodium hydroxide originally present in the melt purge are neutrali-26
sodium bicarbonate, according to the following reactions:
Na2S + NaHC03 -* Na2C03 + NaHs
NaOH + NaHC03 -* Na2C03 + H20
The outlet stream from the dissolving tank is treated by a comb 1 nf^
of clarification, filtration and washing of the solids to minimize so
loss in the ash. Ash free filtrate solution is pumped to the H2S st^P
which uses the steam previously produced from the flash operation. atore
H2S stripper is operated at low pressure (2.15 psia) and at low temper
(130 F) in order to minimize steam consumption. Along with the H2S s
C02 is also removed according to the following reactions:
NaHC03 + NaHS -* Na2C03 + H2S ,5)
2NaHC03 _* Na2C03 + C02 + H20 (~
76
-------
ASH REMOVAL AND SALT RECOVERY SECTION
MELT FROM
GASIFIER
QUENCH
TANK
STEAM TO
H2S STRIPPER
FLASH
TANK
X
HYDRO-
CLONE
x
Y
NaHCO3 SOLUTION
CALCINER
SODIUM CARBONATE TO
GASIFIER-*
DISSOLVING
TANK
ASH FILTER AND WASH
ASH TO DISPOSAL
BICARBONATE
CENTRIFUGE
RECYCLE C02
H2S TO CLAUS PLANT
H2S
STRIPPER
FUEL GAS TO
WATER WASH
STEAM
CARBONATOR
MAKE UP CO2
ABSORBER
FUEL GAS
FROM
RECUPERATOR
Tl
P
O)
-------
The overhead leaving the H2S stripper at 130 F, containing water
vapor, H2S and C02i is cooled to condense and remove water vapor. The
remaining gas stream (33% H2S) is sent to conventional Glaus plant for
converting I^S to elemental sulfur.
The stripped solution is reheated to 174 F and pumped to an absorber
where makeup C02 and some H2S are absorbed from the fuel gas according
to the following reactions:
Na2C03 + C02 + H20 2NaHC03 (6-16)
+ H2S NaHO>3 + NaHS (6-1/J
This makeup C02 £s required to compensate for C02 losses such as
the C02 lost in the l^S stripper and C02 lost in the gasifier by
reaction of sodium carbonate with sulfur to form Na2S. The makeup C02
absorber operates at the fuel gas pressure and at high temperature (!'-'
in order to minimize the amount of water condensed from the fuel gas.
The solution is further carbonated with recycle C02 to precipitate
solid sodium bicarbonate in the carbonation tower. The solution is coo
from 196 F to 140 F in the carbonation tower. At 140 F the sodium bicar
bonate exceeds its solubility limit, and precipitates out of the soluti-0
The resulting slurry with 12% solids is separated by a combination
hydroclones and centrifuges. Sodium bicarbonate with 12% moisture is
decomposed by indirect steam heating at 350 F in a fluidized bed calcine
to produce sodium carbonate for recycling to the gasifier melt. The ^ 2
produced during decomposition is recycled to the carbonation tower. Tne ^g
filtrate solution, along with condensate from other parts of the process
recycled to the slag quench and dissolving tanks.
DESULFURIZATION PROCESS DESCRIPTIONS
Selexol Process Description
The Selexol process, developed by Allied Chemical Corp., is a physi^
absorption process. It uses the Selexol solvent (dimethyl ether of P°W &
ethylene glycol) (Reference 6-6) to selectively remove H2S in the presen
of C02. Commercial applications currently include both selective ab-
sorption and bulk removal of acid gas.
For this application, a typical schematic is shown in Figure 6-7.
gas enters the absorber after having been cooled by heat exchange with
product gas. Where a high degree of COS removal is required, gas fr°m t0
flash tank also enters at the bottom of the absorber. This is necessary
improve the process selectivity for COS with respect to C02- For the
78
-------
TYPICAL FLOW DIAGRAM-SELEXOL ACID GAS REMOVAL PROCESS
PRODUCT GAS
ABSORBER
RAW GAS
CLEAN
GAS
1
\ /
\/
A
REFRIGERATION
FLASH VESSEL
C-v
cw
ACID GASES
SEPARATOR
STRIPPER
• STEAM
SOLVENT
BLOWDOWN
oo
I
o
ro
CO
Tl
-------
basic systems currently under study, COS levels were not sufficiently
to warrant sizing of the system for any gases other than l^S and the
flash tank and recycle compressor were not required. To achieve very
emissions, systems designed for COS removal were also investigated.
Lean solvent, having been cooled by heat exchange with the rich ^
solvent from the bottom of the absorber, is further cooled by refrigera
and introduced at the top of the absorber. Almost all of the ^2S is ^
absorbed as the gas flows up through the tower. COS is less soluble a^ ^
only 30 percent of that component is removed. Approximately 15 percen
the C02 is removed along with minor amounts of the other gases. The ^
solvent vapor pressure is quite low (0.0002 mmHg at 60 F) (Reference
so that it does not contaminate the product gas.
The rich solvent is let down in pressure through a power recoveryure
turbine and the absorbed gases are recovered by a combination of pre
letdown and stripping. Heat of absorption is low and most of the ste
needed to maintain the temperature differential between absorber and ^
stripper. Acid gas to the Glaus plant contains between 24 and 39 perc
H2S.
Other sulfur compounds, such as methyl mercaptan, carbon disul l
thiophene are more soluble than H2S and will be removed if present. wj.th
solvent is not degraded by impurities in the fuel gas. This, combine
the low vapor pressure, results in very low solvent makeup requireme
Conoco Process Description
The process is shown in block diagram form in Figure 6-8. Tne e
passes through the fluidized bed desulfurizer where both t^S and CO
removed. Dolomite is regenerated with a mixture of carbon dioxide an
steam and the off gas sent to the sulfur recovery unit. Because the
dolomite gradually becomes non-reactive, approximately 2 percent of
recirculating dolomite acceptor must be withdrawn and replaced
processed prior to discharge to convert the CaS to
In addition to make-up dolomite, both steam and carbon dioxide &..&
needed in the regenerator and spent dolomite converter. Steam is av
from the bottoming cycle and the carbon dioxide must be separated tr
either the fuel gas or the gas turbine exhaust gas stream.
A schematic of the equipment arrangement and flow paths is shown
Figure 6-9. The process is described fully in Reference 6-7 and the
systems used herein are based on the basic design presented in that
Gasifier product gas flows to the bottom of a fluidized gas desu
furizer. Most of the l^S in the gasifier gas reacts with the CaCOs
component of the dolomite as follows:
80
-------
COA1OCO PROCESS BLOCK DIAGRAM
CLEAN GAS
MAKE-UP
DOLOMITE
00
DESULFUWIZATION
RAW GAS
MAKE-UP
REGENERATIONS
SULFUR RECOVERY
SULFUR
SPENT DOLOMITE
WASTE DOLOMITER
CONVERSION
VI
DC
O
co
O
O
CJI
Tl
P
oo
-------
CONOCO HIGH -TEMPERATURE DESULFURIZATION
CO
00
o
co
MAKE-UP
DOLOMITE
1660F CLEAN GAS
WASTE WATER
LOCK
HOPPER
RAW FUEL'
GAS
MAKE-UP
CO-
ABSORBER
ACID COOLER
LIQUID PHASE
CLAUSREACTOR
REGENERATOR
SEPARATOR
STORAGE SULFUR BURNER
LOCK
HOPPER
— WASTE HEAT
RECOVERY
MAKE-UP
STEAM
EXPORT
SULFUR
SPENT
DOLOMITE
HYDROCLONE
DOLOMITE
CONVERTERS
COOLER (~}
COOLING
WATER
WASTE
SLURRY
WATER
T]
P
CO
-------
H2S + MgO-CaC03 + MgO-CaS + C02 + H20 (6-18)
Although no data have been reported for the COS absorption high removal
efficiencies are predicted according to the reaction:
COS + MgO-CaC03 + MgO-CaS + 2C02 (6-19)
The desulfurized gas must then be cleaned of particulates and cooled
prior to use in the combined cycle. This integration process is discussed
in other sections of this report. The effect of gasifier operating condi-
tions on the degree of desulfurization is discussed in Section 8.
The fluidized acceptor regenerator is maintained at 1300 F. The sulfided
acceptor is recarbonated at these conditions by the reverse reaction,
MgO-CaS + C02 + H20 + MgO-CaC03 + H2S (6-20)
The carbonated magnesium component of the make-up acceptor is also calcined
at these conditions,
MgC03-CaC03 + MgO-CaC03 + C02 (6-21)
Spent acceptor (2 percent of the circulating flow) is withdrawn from
the regenerators. This spent acceptor must be treated before disposal.
Over 75 percent of the calcium component of this stream is in the form of
CaS. If this were disposed of directly to the station ash pit, H2S gas
would slowly evolve as the CaS was hydrolyzed. To avoid this condition, the
spent acceptor is directly contacted with C02 and water in three stages of
stirred reactors to convert the CaS to CaC03. The overall reaction
is,
MgO-CaS + 2C02 + H20 •»• MgC03'CaC03 + H2S (6-22)
Acid gas resulting from the acceptor stripping operation is passed through
two stages of cyclones to remove entrained acceptor. The sensible heat
content of this stream is then exchanged with recycle gas and with
boiler feed water. Electrostatic precipitators are provided to remove
entrained dust. The gas then flows to the bottom of the liquid-phase
Claus reactor.
The concentration of H2S in the gas from the acceptor regenerators
is only about 3.6 volume percent. The liquid-phase Claus reaction developed
for this process is uniquely suited to processing this gas. This was
demonstrated during the experimental work described in Reference 6-8.
Liquid sulfur is produced by the reaction,
83
-------
2H2S + H2S03 -»• 3 S + 3 H20 (6-23)
Liquid sulfur and liquid water flow from the reactor to a decanter-type
separator. Unreacted gas, saturated with water vapor at 310 F, is
compressed and returned to the acceptor regenerator reactor.
The sensible heat content of the liquid water from the separator is exchanged
with the feed acid and charged to the S02 absorption tower. A slip stream of
water is rejected to maintain water balance.
Approximately one-third of the sulfur produced is burned with stoichio-
metric air to produce S02 in the pressurized combustor. Excess heat is
removed via cooling tubes in the walls. Water flowing down through the
packed tower absorbs the S02 in the gas by,
S02 + H20 * H2S03 aqueous. (6-24)
Most of the exothermic heat of reaction is removed by side stream
coolers. The vent gas from the absorption tower is at 90 F and 205 psia.
It probably would contain some residual S02 (assumed in this case as 0.3
volume percent).
Sulfur Recovery
In the case of the Conoco process, sulfur recovery is an integral part
of the process. For low-temperature processes, where acid gas is produced
with greater than 20 percent H2S, the Glaus process is used extensively.
While it is a direct means of recovering elemental sulfur, recovery efficiency
is generally in the 90 to 98 percent range and the tail gas can contain a
significant amount of sulfur as H2S, S02 or COS. Simple incineration of
the tail gas is one possibility. In this study, the use of a Beavon unit
for tail gas sulfur removal was elected. Both the Claus and Beavon units
are described below.
Claus Sulfur Recovery
The basic Claus process was developed in the late nineteenth century.
Since that time, many modifications have been made in the process by various
engineering firms. The overall reaction that is the basis of the process
is:
3H2S + 1.502 -> 3S + 3H20 + 286,000 Btu (6-25)
84
-------
used in th* ""ited
Direct oxidation
Split flow
Straight through
Sulfur recycle
rare-baSed °n the same Principle but find use depending on the acid gas con-
centration in the feed.
In the original Glaus process, called "direct oxidation", H2S was
si i oxidized with air over a bauxite or iron ore catalyst in a
ingle reactor. The reaction (6-25), which is highly exothermic, resulted
excessive temperatures and poor yields. This led to recycle of the tail
• tor temperature control. It is now used for dilute acid feeds.
are ?"ol,imVrovements were introduced in the 1930 's by I.G. Farben. These
In th ii Spflt: flow" and "straight through" versions shown in Figure 6-10.
bur * Split fl°w" Process' whicn is used here, one third of the H2S is
med completely to sulfur dioxide in a waste heat boiler:
H20 + 1.502 — »- S02 + H20 + 236,000 Btu. (6-26)
°2 is then reacted with the remainder of the H2S over the catalyst:
2H2S + S02 — »~3S + 2H20 + 50,000 Btu, (6-27)
and less heat is involved.
In the "straight through" version, which is used at high (50 to 100
d^rcent) H2S concentrations, the acid gas is partially oxidized as in the
ect oxidation process but in a waste heat boiler rather than over a
c aiyst. A large part of the conversion is achieved in the boiler and the
e° gas is sent to one or more catalytic stages for the desired conversion.
H Tne acid gas streams feeding Glaus plants sometimes contain so little
he reaction temperatures cannot be maintained without supplemental
a, ' . This gives rise to additional Glaus plant variations, including
Pr h 10tl °^ hydrocarbons to acid gas ahead of the burner and/or indirect
eating of the acid gas and/or air ahead of the burner or catalytic
eactor .
ha ' 6 <^aus Plant variation, particularly useful for dilute acid gas
V^ng low sulfur oxide content, is called "sulfur recycle." In this, some
stuct sulfur is burned with air to produce S02 and heat. The resultant
eam is mixed with the acid gas in the appropriate ratio and sent to a
85
-------
FIG. 6-10
GLAUS SULFUR RECOVERY PROCESS
WASTE HEAT
BOILER
CATALYTIC
REACTORS
CONDENSERS
NOTES:
1. DASHED LINE IS ADDED FOR THE SPLIT-FLOW VERSION OF THE PROCESS
2. ADDITIONAL SECTIONS CONSISTING OF CATALYTIC REACTOR AND CONDENSERS
CAN IMPROVE CONVERSION EFFICIENCY
86
-------
catalytic reactor. This variation is applicable to H2S concentrations of
iess than 15 percent.
Complete conversion to sulfur is prevented by equilibrium limits in
e reactor. Improved conversion is achieved by adding reactor stages.
°e products of conversion are sulfur vapor and water vapor. They are
condensed and removed between stages. Conversion efficiencies are generally
ln the mid-ninety percent range but up to 98 percent is attainable. The
sulfur product is 99.9 percent pure and can be marketed.
Tail Gas Treating - This process consists of a hydrogenation step
ollowed by a Stretford section for sulfur conversion. Named for D. K.
eavon, the process was developed by the Ralph M. Parsons Company. Final
tail gas concentrations in the range of 40 to 80 ppm sulfur (as S02) have
been reported for the process.
In the process, entering tail gas is mixed with hot flue gas as shown
ln Figure 6-11. Sufficient hydrogen is supplied, either with the tail gas
°r by fuel-rich combustion in the burner, to convert all the sulfur to
2s- This is done over a cobalt-molybdate catalyst at temperatures
between 550 and 750 F.
In the Stretford section, the H2S is reacted with oxygen to form
elemental sulfur and water. The process uses the following reaction
sequence:
Absorber -
H2S + Na2C03 - — - > NaHS + NaHC03 (6-28)
faction hold tank -
4NaV03 + 2NaHS + 2H20 - > Na^Og + 4NaOH + 2S (6-29)
Na2V409 + 2NaOH + H20 + ADA - > 4NaV03 + 2ADA (reduced state)
°xidation tank -
2ADA (reduced state) + 02 - — > 2ADA + H20 (6-30)
In the absorber, H2S reacts with sodium carbonate solution to form
s°dium bicarbonate and sodium hydrosulf ide. The catalytic sodium vanadate
olution reacts with the NaHS in the hold tank to form elemental sulfur.
^1J blowing with sodium anthraquinone disulfonate (ADA) regenerates the
°aium vanadate catalyst. The ADA is regenerated by oxidation.
The overall effectiveness of the Beavon process is dependent on
onversion efficiency of sulfur compounds to H2S and subsequent efficiency
the Stretford unit. High conversion efficiencies are predicted for S,
COS and CS2 (Reference 6-10). The Stretford is capable of reducing
H2S content to less than 1
ppm.
87
-------
BEAVON TAILGAS CLEANUP PROCESS
TREATED TAI LGAS
FUEL GAS i
AIR i
00
00
LINE BURNER
CATALYTIC
REACTOR
ABSORBER
7|
s. ^
A.
^ **.
£
[REACTION
I HOLD
I TANK
COOLER
fc
S*~^\
v
^ \
J
A
MAKE-UP
WATER
STRETFORD UNIT
OXIDIZER VENT
MAKE-UP
CHEMICALS
SURGE
TANK
SETTLING^
TANK J
OXIDIZER
ELEMENTAL
SULFUR
AIR
SORBENT SLOWDOWN
CONDENSATE TO
SOUR WATER
STRIPPER
T]
P
cn
-------
REFERENCES
Robson, p. L., A. J. Giraraonti, W. A. Blecher and G. Mazzella: Fuel
24Q r-"7'ronmental Impact, Phase Report. EPA-600/2-75-078 (NTIS No. PB
^-, November 1975.
6-2 R h
oson, F. L., W. A. Blecher and C. B. Colton: Fuel Gas Environment
impact. EPA-600/2-76-153,(NTIS No. PB257-134), June 1976.
' J* G' : Clean Fuel from c°al is Goal of U-Gas Process. Oil
Gas Journal, Vol. 75, No. 31, August 1, 1977.
6-4 He
egarty, W. P. and B. E. Moody: Evaluating the Bi-Gas SNG Processes.
nem. Eng. Progress, Vol. 69, No. 3:37, March 1973.
olzberger, T. W. : Combined Cycle Power Generation Using Molten
^ait Coal Gasification. The Fourth Annual International Conference
on Coal Gasification, Liquifaction and Conversion to Elecricity,
university of Pittsburgh, August 2-4, 1977.
e, J. p. : Gas Purification with Selexol Solvent in the
ew Clean Energy Proceses. American Chemical society, 167th National
"eeting, Los Angeles, California, April 1974.
6-7 c
urran, G. p., B. J. Koch, B. Pasek, M. Pell, and E. Gorin: High
emperature Desulfurization of Low-Btu Gas. EPA-600/7-77-031
^NTIS No. PB271-008), April 1977.
6~8 c,
g "an> G. P., j. x. Clancey, B. Pasek, M. Pell, G. D. Rutledge,
n E. Gorin: Production of Clean Fuel Gas from Bituminous Coals.
pep°rt to Office of Research & Devlopment, U.S. Environmental
rotection Agency, Under Contract EHSD 71-15, Period - March 1972 to
une 1973, EPA Report No. 650/2-73-049, (NTIS No. PB 232-695/AS)
"ecember 1973.
6-9 B
fNrr8' W' D>: Characterization of Glaus Plant Emissions. EPA-R2-73-188
^ris NO. PB220-376), April 1973.
6-io c
CoV^nau8h, E. C., et al.: Technology Status Report: Low/Medium Btu
(NTT Gasif ication and Related Environmental Controls, EPA600/7-77-125B
VINIIS No. PB274-843), June 1977.
89
-------
SECTION 7
COMBINED-CYCLE POWER GENERATION SYSTEMS
INTRODUCTION
The term "combined cycle" can be used for a variety of systems where
two cycles operating at different temperatures are interconnected. The
rejected by the higher-temperature cycle is recovered and used in the loW-on
temperature cycle to produce additional power and improve overall conver
efficiency. The higher-temperature cycle is referred to as a topping eye ^
while the lower-temperature cycle is called a bottoming cycle. The com 1 ^
tion used in this study is a gas turbine topping cycle with a steam bot
cycle. It is currently the only combination of commercial significance.
A simplified schematic of the combined gas- and steam-cycle plant is ^
shown in Figure 7-1. Exhaust gases leaving the gas turbine are typical y^
a temperature near 1000 F. This places an upper limit on the steam temp ^
ture generated in the waste heat boiler. As higher steam temperatures at
generally associated with a higher steam cycle thermal efficiency, it iS^ ^a
desirable to raise the gas turbine exhaust temperature. To do this wou t&
a decrease in the work extracted from the gas turbine cycle. Clearly. Off
is a tradeoff between gas turbine and steam cycle parameters. This tra
is made even more complex by the addition of the gasification and clean P^
processes. These present a source of heat to raise steam for the bottom^^
cycle. They also use steam that can be extracted from the steam cycle
doing some useful work.
Y&
The current study uses gas turbine and steam-cycle parameters tnft"e
developed as part of a DOE-sponsored study, the High Temperature Turbin^ ^e
Technology Program (HTTTP). As part of that program, parametric studie ^^
conducted to determine the optimum combination of gas turbine inlet te P
ture, pressure ratio, metal cooling methods, steam cycle conditions an ^
various other factors affecting overall power generating efficiency.
these were evaluated in the context of a system comprised of a molten s ^g
type gasification system integrated with a combined cycle power Platlt'on of *
section presents a review of the process that led to the HTTTP selecti
combined cycle power plant having the following characteristics:
90
-------
WASTE-HEAT COMBINED GAS AND STEAM TURBINE SYSTEM
AIR
COMPRESSOR
TURBINE
BURNER
FUEL
T-2000 F
P~14 ATM
T-875 F
STEAM
BOILER
T-300 F
TO STACK
T~775 F
P~60 ATM
POWER TURBINE
L
PUMP
CONDENSER
ELECTRIC
GENERATOR
70 MW
ELECTRIC
GENERATOR
50 MW
P
••j
I
-------
Gas turbine pressure ratio 18:1
Gas turbine firing temperature 2600 F
Static parts cooling fluid Water
Rotating parts cooling fluid Pre-cooled air
Steam cycle pressure 2400 psi
Steam temperature 950 F
Reheat temperature 950 F
These same characteristics have been used with the exception of the
cooling method for the static parts. The use of ceramic vanes which requi
no cooling has been assumed. This results in a very minor change in gas
bine performance. Previous phases of this EPA study had used a 24:1 gas
turbine pressure ratio and non-reheat steam cycle. The reduced gas turbm
pressure ratio results in exhaust temperatures which in concert with waste
heat from the gasifier are able to support the high performance steam eye •
The result is a significant improvement in overall power generating efficl
GENERAL PARAMETRIC ANALYSES
Analyses performed in earlier phases of this program and in the HTT
program were first directed towards defining the operating conditions ot
the gas-turbine engine. This was done by conducting trade-off studies tha^
relate the interaction of engine pressure and temperature levels to perfor
mance levels and design requirements. Initially, these conditions were
defined with respect to the engine alone, then modified to account for the
interactions with the remainder of the system, thereby identifying a gas- ^
turbine engine that provides an overall optimum system. For example, typ1
cal curves relating to the interactions among compressor pressure ratio a
turbine cooling techniques with open-cycle gas turbine engine power outpu
and efficiency are shown in Figure 7-2. This figure indicates that, while
the relationship of performance and specific power is a strong function o
cooling system type, there is a range of compressor pressure ratios Sener* ff
between 12:1 and 24:1 which demonstrates a compromise between specific P°
which is a measure of capital cost, and efficiency.
However, when the steam bottoming cycle is added to the design, the P
lem of optimizing the gas turbine becomes more complex. Two of the more
important factors considered were the effects of the gas turbine operating
point on steam system efficiency and the relative costs of the two system • ^
Increasing the compressor pressure ratio of the gas turbine while maintai ^
the defined turbine temperature has the effect of increasing the work °u ?g
of the engine and decreasing the engine exhaust temperature. Although t 1
augments the performance of the gas turbine cycle, it leaves less energy . ^
the steam bottoming cycle. As a result, because the exhaust temperature ^
lower, the steam temperature is lower and the efficiency of the steam cy
then lower. Thus, the goal is to select an operating condition that P*°
the highest overall performance.
92
-------
FIG. 7-2
0pEN CYCLE GAS-TURBINE ENGINE PERFORMANCE TRADE-OFF CURVES
as
o
ti
Uj h-
IU <
45
40
35
30
25
20 L
40
CERAMIC VANES
WATER COOLED VANES AND
PRECOOLED AIR FOR BLADES
40
.•in
PRECOOLED AIR
:
16
CONVENTIONAL
COOLING
THEORETICAL
LIMIT
8
NOTE: NUMBERS 8 THROUGH 40 ALONG CURVES
INDICATE COMPRESSOR PRESSURE RATIO
1
I
1
1
100 120 140 160 180 200 220
NET POWER PER UNIT OF AIRFLOW — KW-SEC/LB
240
78-03-160-2
93
-------
Definition of the steam bottoming cycle is dependent upon the waste hea
available from the gasifier and cleanup system as well as that available fro
the gas turbine exhaust stream.
Analyses were performed to obtain the best match among these components
for a definition of the best overall powerplant performance and cost. The
key items are the steam turbine and the waste heat boiler. The selection o
operating conditions for these components seeks to ensure a sufficiently nig
rate of heat transfer by maintaining sufficient temperature difference betw
the gas turbine engine exhaust and the steam being heated in the waste heat
boiler. The smaller this temperature difference, the larger and more costly
the heat exchanger must be. In contrast, the larger the difference, the tno
overall efficiency decreases. These effects are discussed under thermal
integration in Section 8. Selection of a "standard" combined-cycle power
plant results in a slight penalty for some of the systems studied, but did n
materially affect the overall conclusions.
On the basis of results acquired from general parametric studies, it lS
possible to identify the areas of interest concerning gas turbine and steaw^
turbine operating characteristics. Figure 7-3 shows the generalized perfor" ^
mance of simple-cycle and combined-cycle systems operating on liquid fuelj
demonstrates the expected effect with the integration of a coal gasificatio
system. The range of interest defined for subsequent analyses is indicate
the shaded areas in Figure 7-3.
Operating Conditions
Gas turbine pressure ratio and firing temperature are the primary facc
affecting overall system performance. However, the techniques used for coo
ing the high-temperature parts of the turbine and the allowable metal temp^
ture have a direct bearing on performance at a given firing temperature.
part of the HTTT program for DOE, a metal temperature of 1600 F was selecte
as the maximum practical limit. Also, a cooling scheme that would make fu ^
utilization of existing technology was selected. It consists of water coo
for the static parts of the engine and the use of pre-cooled air for
the rotating turbine parts.
Using the molten salt gasifier, a parametric analysis was conducted
results of this analysis indicated that an overall pressure ratio of 18:1 1
nearly optimum for turbine stator inlet temperatures of 2600 F and 3000 F
while a pressure ratio of 14:1 is nearly optimum for an operating temperatu
of 2400 F. Figure 7-4 presents the overall system efficiency versus net ,fl
specific power as a function of the thermodynamic parameters of Pressure.rfet
and turbine operating temperature. As indicated, as the turbine stator 1°
temperature increases, the efficiency increases, but at a decreasing rate. ^
The difference in maximum efficiencies between 2400 F and 2600 F is appr°x
mately one percent, whle the difference between 2600 F and 3000 F is 0.5
percent. In addition, the effect of pressure ratio on efficiency is *esSttfeef
pronounced at the higher operating temperature levels. The difference be ^
the optimum and lowest efficiency of the systems investigated is 1.5 perce
at 3000 F, 2.8 percent at 2600 F, and 4.3 percent at 2400 F.
94
-------
FIG. 7-3
TRENDS FOR COMBINED - CYCLE SYSTEMS
•
O
Uj
O
ii~
u
i
RANGE OF
INTEREST
!8
; '
" COMBINED DISTILLATE - FIRED
CYCLE
Pr = 40
!6
COMBINED CYCLE/GASIFICATION/ SULFUR C
..16...
SIMPLE CYCLE/
DISTILLATE - FIRED
36
40
SPECIFIC POWER
78-03-160-1
95
-------
£
3
-
DC
LLJ
44
43
42
40 —
HTTTP GAS TURBINE PARAMETRIC STUDY
Wagt = 815.6 LB/SEC
Iso CONDITIONS
TT5 = 2600 F
TT5 = 2400F
TJ5 = 3000°F
38
A 80
200
22O
240
260
280 300 320
OF /\\Rf \_O\N -KXN- SEC I \_B
340
360
-------
c 81n8 the data of Figure 7-4, a comparison of capital cost and heat rate
bet 6 mac*e as snown in Figure 7-5. Taking cost and performance differences
W®en a pressure ratio of 22 and 18 shows a capital cost increase of approxi-
•jan-i ^ 57.7 per kW as performance improves fron a heat rate of 8224 to one of
tu/kWh. The capital expenditure to get such improvements in heat rate
Be" e Justified as coal costs exceed approximately $1.00 per million Btu.
K« Use °f the rapid trend in this direction, the higher efficiency appears to
De desirable.
Qe ne costs for Figure 7-5 were developed using data from previous phases
$488 pr°8ram- Power system costs were taken as $122 per gas turbine kW and
t£v per steam turbine kW. Gasifier and cleanup costs are generally insensi-
Pres to.pressure» at least over the range being considered here. For the
wei uUrizec* e
-------
FIG.
EFFECT OF CHANGES IN PRESSURE RATIO
.60
r8300
-8200
1
Dfl
I
LU
<
ce
<
UJ
'8100
-8000
1-7900
280
.50
-260
Q
O
11.
-240 u.
O
U
1
Q
ii
-220
40
L200
14
16
18 20
PRESSURE RATIO
22
98
-------
>-
U
U]
Q
I
o
FIG. 7-6
RESULTS OF STEAM CYCLE PARAMETRIC STUDY
44
43
TURBINE TEMPERATURE °F = 2600°F
PCOND = 3-5IN-HG
ISO CONDITIONS
42
41
40 -
39
270
OPR = 18
OPR = 12
PSI/F/F
3500/1000/1000
D 2400/1000/1000
O 2400/950/950
A 1800/950/950
0 1450/950/950
1250/950 NO REHEAT
280
NET POWER PER UNIT OF AIRFLOW-KW- SEC/LB
78-03-160-3
99
-------
SECTION 8
INTEGRATION STUDIES
INTRODUCTION
Thermal integration can have a significant effect on overall system
performance. Effective thermal integration simply means utilizing available
thermal energy with minimum temperature difference between the source and
the substance it heats. Thermal energy is available from the gasifier jacket,
hot raw fuel gas, turbine cooling air, gasifier bleed air as well as from
some of the process waste heat. Much of this heat is available at a tempera-
ture level suitable for raising relatively high-pressure steam. Moreover,
process heat requirements generally involve the use of low-pressure steam
which can be withdrawn from the steam cycle after having done some useful
work.
When considering the complete system, the highest efficiency in the use
of thermal energy is achieved when it is converted at combined-cycle effi-
ciency. This means using high-temperature cleanup or reheating the fuel gas
after a low-temperature cleanup. A number of thermal integration studies,
including the effect of fuel gas reheat, were conducted during the earlier
phases and are summarized in this section. In general, the results of those
studies have been applied in defining the current integrated system schema-
tics used for determing performance and environmental intrusion. In the case
of clean fuel gas temperature, a maximum limit of 1000 F was established to
allow the use of a pre-mix combustor for NOX control. While this effec-
tively limits some of the advantages of high-temperature cleanup, it appears
to be essential if NOX emission levels are to be within allowable limits.
Besides thermal integration, other integration studies were conducted to
determine the effects, especially environmental effects, of operating tempera-
ture on the Selexol system and of steam feed rate on the U-Gas gasifier. Also
examined was the effect of achieving equilibrium in each of the gasifier efflu
ents. This last study was intended to identify potential species and concen-
tration levels that could be expected, at least on the basis of equilibrium
predictions.
The Selexol solvent is quite sensitive to operating temperature level.
In natural gas sweetening plants, absorbers generally run at subambient
temperatures. Because of the utility industry's desire to avoid the
100
-------
of a refrigeration system as part of the process plant, a study
operating temperature was made. For the degree of sulfur removal consi-
red in this study, refrigerated operation showed a decided advantage. Two
r P erent levels of sulfur removal were considered and the advantage of
frigerated operation was seen to increase with increased sulfur removal.
. Earlier phases of this study had shown the effect of reduced steam feed
0 the gasifier. This improved cold gas efficiency and reduced the heat lost
y condensation of water vapor in the raw fuel gas when it is cooled prior to
lfurization. Data used in this study for the BCR-type gasifier showed the
r steam feed rate (0.15 Ib steam per Ib coal vs. 0.57 Ib steam per Ib coal)
ted in a change in cold gas efficiency from 78.5 percent to 83 percent.
reduced steam feed rate also results in a lower air/coal ratio and a lower
Ot»centration of both CC>2 and H20 in the raw gas. These low concentrations
end to improve performance of the high-temperature Conoco desulfurizer.
For the U-Gas process, the effect of steam feed rate on desulfurizer
?*rformance was determined and the point of minimum practical steam feed
ed on discussions with IGT) was used in the subsequent system evalua-
. Performance improvements (approximately 3%) for the U-Gas system with
steam feed were also identified for low-temperature sulfur removal.
t, Gasifier modeling efforts of previous phases were continued. Considering
tje goals of this study, the most important part of the modeling effort was
ne adaption of the chemical equilibrium calculation procedure to model the
Ja^fication process. Use of this tool allowed equilibrium limits to be
eflned for the formation or reduction of potential pollutants.
INTEGRATION
8, The gas turbine operating parameters and steam cycle conditions have a
£*«if leant effect on the overall thermal integration. Previous studies, for
!*amPle, considered the use of a supplementary fired waste heat boiler as a
t?ans of improving overall system performance. The idea was simply to raise
ConVas temperature in the boiler. This, in turn, allows improved steam
onions which increase steam cycle efficiency and improves the efficiency
•• utilization of waste heat. Overall efficiency is given by:
(8-1)
n = • • ••• • : : ••-••
ncc w
. . s
101
-------
where
ncc = combined cycle efficiency
n»t = §as turbine efficiency
ns = steam cycle efficiency
Wf = primary fuel flow
Ws = supplementary fuel flow
TPX = gas turbine exhaust temperature
Ta = ambient temperature
Tst = stack temperature
A high turbine firing temperature combined with utilization of gasifier
and raw fuel gas thermal energy make it possible to utilize a high-performance
steam cycle without resorting to supplemental firing.
When there is no supplemental firing of the waste heat boiler, the
equation simplifies to:
T — T
•'•ex La
(8.2)
and the importance of steam cycle efficiency is quite apparent, all other
things remaining equal.
In each of the systems studied, virtually all steam is raised in the
gasifier jacket and the hot fuel gas boiler. The gas turbine waste heat boilfi*
is used almost exclusively for superheat, reheat and economizing. A typical
Temperature Vs. Heat Flow (T-Q) curve for the waste heat boiler is shown in
Figure 8-1. It shows the problems involved in utilizing the thermal energy °*
the turbine exhaust while maintaining minimum temperature differentials consi3
tent with high-cycle efficiency. Because of the varying amounts of heat
available from the gasifier, fuel gas stream, and other sources, waste heat
boiler designs vary considerably among the various systems.
In the case of the oxygen-blown BCR type system with Selexol cleanup,
the waste heat boiler is quite complex. In that system, it would be neces-
sary to raise some steam at reheat pressure in the waste heat boiler to
achieve a 300 F stack temperature which was set as a lower limit for all
systems. Modification of the steam conditions could eliminate some of the
102
-------
TYPICAL T-Q DIAGRAM FOR WASTE HEAT BOILER
FIG. 8-1
1200
1000
u_ 800
o
n
)
< 600
QC
III
400
200
•70 F PINCH
GAS SIDE PROFILE
40 F PINCH
REHEATER SUPERHEATER
100 200
STEAM SIDE PROFILE
DEAERATION
600 700 800
Q-MMBTU/HR
78-02-177-1
103
-------
complexity; however, further iteration of the design was judged unwarranted
as it would not materially change the study results. All heat recovery
systems were able to be designed with a minimum superheater pinch of 70 F and
boiler pinch of 40 F.
An important factor that does not appear in Equation 8-2 for combined-
cycle efficiency is the effect of fuel gas temperature. The sensible heat
associated with the fuel gas can be converted at full combined-cycle effi-
ciency if the gas were reheated after cleanup and prior to being burned. The
alternative to reheating would be to use that same heat in the steam cycle.
Efficiency of the steam cycle is significantly lower (lstm/ncc ~ 0<6^
than the overall combined-cycle efficiency.
The effect of reheating the clean fuel gas by heat exchange with the hot
raw gas in a BCR system with Selexol cleanup is shown in Figure 8-2. As
clean fuel temperature to the burner is increased, overall plant efficiency
is improved at the cost of an increase in heat transfer surface area. The
performance improvement depends heavily on the temperature level below which
sensible heat in the hot fuel gas stream is not recovered (i.e., minimum
temperature at exit of boiler or hot side of fuel regenerator). Recovery of
this low-temperature heat is made more difficult by the presence of water
vapor, sulfur compounds and ammonia in the dirty gas. These substances cause
the mass flow rate of the dirty gas to be higher than that for the clean gas
on the cold side of the regenerator. As a result, the temperature drop on
the hot side of the regenerator is significantly less than on the cold
side.
One of the undesirable aspects of a low-temperature cleanup system is
that the low-temperature causes most of the water vapor in the dirty gas
stream to condense. This will result in the need for costly heat exchange
equipment to withstand the weak acids which will be present. However, since
it is necessary to cool the gas, it appears desireable to utilize as much of
the heat in the dirty gas stream as is possible to improve system performance.
In addition to showing system efficiency, Figure 8-2 shows the effect of
regeneration temperature on heat exchange area. While the relationship is
also a function of system configuration and subject to further optimization,
in general higher temperatures will require larger exchangers made out of
more expensive materials. As an example, using the area requirements of
Figure 8-2, a cost comparison was made for a regenerator having a cold-side
outlet temperature of 750 F and one with a 1000 F outlet temperature. The
results in Table 8-1 show that fabricated equipment costs were estimated to
be $17.50/ft2 for the low-temperature unit and $27.50/ft2 for the 1000 F
unit. In the high-temperature case, 18-8 stainless steel and chrome alloy
steel are employed for the tube and shell materials. While low alloy carbon
steel can be used for the shell in the low-temperature case, stainless steel
would still be required for the tube service due to the H2~H2S environment
at the operating temperature.
104
-------
FIG. 8-2
EFFECT OF FUEL TEMPERATURE
BCR/SELOXOL SYSTEM
37
O
Uj
o
35L
HEAT EXCHANGE AREA
30
600
700 80°
FUEL TEMPERATURE, °F
900
1000
R04 -35-1
105
-------
TABLE 8-1
BCR-SELEXOL FUEL GAS REGENERATOR COSTS
Fuel Supply Temp - F 1000 . 750
Required Area - ft2 88,500 48,500
Exchanger Cost - $ 3.04 x 106 1.06 x 106
Total D&E Cost - $ 5.5 x 106 1.9 x 106
Cost with interest and escalation - $ 7 x 106 2.5 x 106
The resultant delivered and erected cost differential is still quite
substantial due to the significantly larger area required for the high-
temperature regeneration. The incremental capital investment is approxi-
mately $4.5/kW for a performance improvement of approximately 1.5%. This is
equivalent to a decrease in heat rate of more than 120 Btu/kWhr.
Figure 8-3 presents the capital cost equivalent of heat rate improvements'
As coal costs increase to more than $1/10^ Btu, an increment in capital
investment of approximately $5/kW is warranted. In the current study, a
somewhat lower performance exchanger has been assumed, with the cold fuel
being reheated to approximately 900-950 F.
SELEXOL OPERATION
A general description of the Selexol process was given in Section 6.
The basic flow diagram for the process was shown in Figure 6-7. It is a
physical solvent system with absorption taking place at high pressure and
stripping at ambient pressure. Refrigeration may be added to enhance solvent
capacity and thereby decrease solvent flow rate, steam requirements and
equipment size. A solvent flash tank with gas recycle compressor may be
needed to improve selectivity. The desirability of this additional equipment
is dependent upon the particular application.
One of the goals of the current study was to compare operation of the
ambient-temperature Selexol process to that of the refrigerated process.
Other goals initially made were the identification of the sensitivity of the
ambient-temperature system to sulfur removal level and a reassessment of the
advantages of catalytic conversion of COS to t^S upstream of the Selexol
absorber. The latter goal was later discarded mainly on the basis of a la'*
of a well defined need for such conversion. Latest estimates of COS level ltl
the raw gas show that a system designed for l^S removal only can still
produce emissions on the order of 0.4 Ib S02/106 Btu, well below current
power plant standards. Nonetheless, technical problems are associated with
CDS conversion because it is catalyzed and operates at approximately 600 to
700 F.
106
-------
FIG. 8-3
CAPITAL EQUIVALENT OF HEAT RATE IMPROVEMENT
LOAD FACTOR = 0.7
ANNUAL CAPITAL CHARGE = 17%
I
<
rT
3
u
i -
m
i
i
3
a
UJ
FUELCOST-$/106BTU
107
-------
At that point in the process there is still a significant amount of particu-
late matter in the stream. The problem of plugging the catalyst bed possibly
could be solved by the use of a fluid bed. However, applicable experience
with the required type of catalyst in this service is not available.
Refrigerated Operation
Initial Sel,exol sizing studies (Reference 8-6) had been based on
refrigerated operation and required two basic designs, the choice of which
depended upon the feed gas composition. These compositions are typified by a
BuMines-type feed that has little COS and a BCR-type feed having a high level
of COS. To achieve a level of 100 ppm sulfur in the clean gas requires a
design based on H2S removal in the case of the BuMines-type feed and one
based on COS removal for the BCR-type feed. The feed, product, and Glaus
feed gas compositions are presented in Table 8-2.
For the H2S-based design, solvent flow rate is adjusted to remove the
major portion of the H2S and is therefore capable of removing only some 30
percent of the COS. In the COS-based design, solvent flow rate is increased
to remove most of the COS as well as the H2S. Because of the relatively
low solubility of COS in the solvent, a system sized for COS removal must
have a proportionately higher solvent flow rate and utility consumption will
increase accordingly. Moreover, the solubility of COS is only about twice
that of C02- Simply increasing solvent flow rate would result in removal
of a significant amount of the C02 from the fuel gas, thereby lowering its
mass flow and decreasing the I^S fraction in the Glaus gas. To avoid this,
the flash tank and recycle compressor shown in Figure 6-7 are included.
These improve selectivity to a point where only 30 percent of the C02 is
removed while virtually all the COS is absorbed.
A comparison of the utilities and costs associated with the two designs
is given in Table 8-3. While not shown in Table 8-3, the approximate cost
and utilities were estimated (Reference 8-1) for the H2S-based design
should the required sulfur removal level be relaxed to allow 1000 ppm sulfur
in the clean gas. Solvent flow rate and corresponding utilities and cost are
reduced by only 12 percent over the 100 ppm design. Thus, for refrigerated
operation, the sensitivity to sulfur removal level appears to be quite low,
at least between 100 and 1000 ppm total sulfur (actually 32 ppm and 900 ppm
H2S). Additional studies showed that for refrigerated operation, solvent
flow rate and associated utilities and cost were quite insensitive to the
level of H2S in the feed within the limits studied. That is, the control-
ling factor in the design is that area of the absorber where the lowest
sulfur levels are achieved. Thus, the Selexol system utilities and cost are
largely a function of only the molar flow rate, the total pressure and, of
course, the type of sulfur compound that must be removed.
Ambient Temperature Operation
Selexol designs using an ambient temperature (100 F) absorber were
developed (Reference 8-2) for the same feed compositions as previously
described for the regrigerated systems. Mol balances for the resulting
108
-------
TABLE 8-2
BASIC SELEXOL DESIGNS
Refrigerated Operation
Removal to 100 ppm Sulfur
BuMines Type Feed (H2S Based Design)
MOL BALANCE MOLS/HR.
Feed Product Glaus Gas
2 57,792.7 57J6U. 5 128.2
0 26,616.0 26,527.6 88.1*
°2 6,717.9 5,61*7.1 1,070.8
2 17,791*.3 17,767.3 27.0
3,1*86.8 3,1*55.6 31.2
639.2 3.6 (32PPm) 635.6 (31*)
Os 11.6 7.6 (68Ppm) 1*.0
3 93.0 30.5 62.5
otal 113,251.5 111,203.8 2,Ql*7.7
BCR Type Feed (COS Based Design)
Feed Product Glaus Gas
C2 65,678 65,676 2
Cn 26,217 26,215 2
3 2 11,663 8,015 3,61*8
cjj. 18,369 18,369
Ul 5,202 5,198 *
If 616.6 1.2 (lOppm) 615.1*
UtT 11*1.3 10.8 (90Ppm) 130.5
3 61*.2 1.1 62V7
rn
°tal 127,951.1 123,U86.5 U,l*61*.6
109
-------
TABLE 8-3
SELEXOL UTILITIES AND COST SUMMARY
REFRIGERATED OPERATION
Bu Mines Feed BCR Feed
Total sulfur 100 100
in product - ppm
Controlling Species H2S COS
Steam (50 psi) Ib/hv 105,000 305,000
Net Power - bhp 21,000 47,000
Net Power - kW 17,400 38,950
Capital Cost - $106 21.3 49.2
systems are presented in Table 8-4. The table also gives the performance
estimates for a BuMines-type feed with removal down to 500 ppm and a modified
BCR-oxygen-blown feed with removal down to 60 ppm total sulfur.
Of interest is the degree of COS removal achieved at the various overall
removal levels. Comparing the two BuMines cases, at 100 ppm total sulfur 46
percent of the COS is removed while at 500 ppm sulfur only 35 percent of the
COS is removed. This indicates that solvent flow rate was significantly
increased and that the ambient temperature design is quite sensitive to
removal to levels of less than 500 ppm in the product gas.
The sensitivity of the ambient-temperature systems to sulfur removal
level is shown in Table 8-5 which gives a summary of utilities and costs for
each of the four designs. Comparing the two BuMines cases, decreases of 42
percent in steam consumption, 22 percent in power and 23 percent in cost are
noted as the sulfur removal requirements are relaxed from 100 to 500 ppm.
The COS-based design requires a flash tank and recycle compressor in addition
to increased solvent flow to achieve an 18 percent H2S level in the Glaus
feed gas.
Selection of Operating Temperature
It had been anticipated that the change to ambient temperature operation
of the Selexol absorber would prove to be desirable both from a performance ^0
cost standpoint. However, a review of the results shows that this is apparently
not true. Tables 8-6 and 8-7 present a summary of the effect of temperature
change for a BuMines-type feed where H^S removal determines the design and a
BCR-type feed where COS is the controlling species. In general, the refrig"
erated system requires less steam, more power, and is less costly than its
ambient-temperature counterpart. The BuMines design is of particular interest
110
-------
MOL BALANCE FOR AMBIENT TEMPERATURE SELEXOL DESIGN
Feed
Bu Mines Type Feed
100 ppm Sulfur Out
Product
N2
CO
co2
H2
CHI;
H2S
COS
NH3
H20
57,892.7
26,616.0
6,717.9
17, 791*. 3
3,1*86.8
639.2
11.6
93.0
239.1*
57,603.2
26,1*17.3
5,318.1
17,732.6
3,1*22.1*
3.9 (35ppm)
6.3 (57ppm)
12.6
139.5
TOTAL 113,1*90.9
110,655-9
Claus Gas
2,835.0
Feed
BCR Type Feed
100 ppm Sulfur Out
Product
289.5
198.7
1,399.8
61.7
61*. 1*
63.3 (22*)
5.3
80.1*
99.9
N2
CO
C02
H2
H2S
COS
H20
65,678
26,217
11,663
18,369
5,202
616.2
11*1.3
61*. 2
283
65,671
26,211
9,300
18,368
5,191*
f ?•*-*'*
1.2
11.2
159
(lOppm)
(90ppm)
128,231*.1 121*,915.1*
Claus Gas
7
6
2363
1
1*
615.1*
130.1
1.5
306
3,l*3l*.0
(18*)
Feed
TOTAL 113,ll90.9
Bu Mines Type Feed
500 ppm Sulfur Out
Product
N2
CO
co2
H2
Cfy
H2S
COS
NH3
H20
57,892.7
26,6l6.0
6,717.9
17,79^.3
3,1*86.8
639.2
11.6
93.0
239.1*
57,666.5
26,1*60.8
5, 621*. 3
17,71*6.1
3,1*36.5
1*6.6 (l*19ppm)
7.5 (67ppm)
30.2
135.3
111,153.8
Claus Gas
226.2
155.2
1,093.6
1*8.2
50.3
592.6 (25*)
62.8
2,337.1
CO
C02
H,/
COS
H20
Feed
BCR-Oxygen Blown Feed
60 ppm Sulfur Out
Product
381.6
30,018.9
10,630.5
25,070.7
5,130.0
951*.0
7.2
257.U
379.5
29,766.9
8,113.2
2l*,97l+.l
5,020.8
0.7 (lOppm)
3.6 (53ppm)
118.2
72,1*50.3 68,376.9
Claus Gas
2.1
252.0
2,517.3
96.6
109.2
953.1*
3.6
381*. 3
1*,318.5
(22*)
-------
TABLE 8-5
SELEXOL COST AND UTILITIES SUMMARY
Ambient Temperature Absorber
Total S in effluent - ppm
Controlling Species
Steam (50 psi) - Ib/hr
Net Power - bhp
Net Power - kW
Capital Cost - $106
Bu Mines
Feed
92
H2S
252,000
5,151*
k,21Q
3U.6
BCR
Feed
100
COS
760,000
30,815
25,532
95.8
Bu Mines
Feed
1*86
H2S
1^7,000
**,029
3,338
26.6
02 -Blown
BCR Feed
63
H2S
82,200
1,230
1,019
27.1
-------
TABLE 8-6
REFRIGERATED VS. AMBIENT-TEMPERATURE SELEXOL OPERATION
BuMines Type Feed (H2S Controls)
100 ppm Sulfur in Product
Ambient-Temperature Refrigerated A
252,000 106,200 lU6,000
- kW
_ 13,000
Cost-$106 3U.6 21.3 13.3
TABLE 8-7
REFRIGERATED VS. AMBIENT-TEMPERATURE SELEXOL OPERATION '
BCR Type Feed (COS Controls)
100 ppm Sulfur In Product
Ambient-Temperature Refrigerated /\
t
- ib/hr 760,000 305,000 ^55,000
- kW 25,530 38,9^0 _ 13>Uoo
C°st - $106 95.8 119.2 U6.6
113
-------
in this study since that will be the basis for the integrated system designs
described in Section 11. For fuel gas from the BuMines-type gasifier,
refrigerated operation shows less advantage than it does in the case of the
BCR-type gasifier.
The effects of changing from refrigerated to ambient-temperature opera-
tion are summarized in Table 8-8. Low-pressure steam use is debited at the
rate of 20 Ib/kWh while the incremental costs of the turbogenerator and
condenser were $500/kW for the low-pressure steam (exclusive of boiler). The
net effect of ambient-temperature operation is a $10/kW increase in cost and
a 50 Btu/kWh decrease in heat rate. It can be seen from Figure 8-3 that this
would pay off only when coal reaches a cost of approximately $6 per 10"
Btu. Thus, while the overall effect of a change to ambient-temperature
operation would not be very significant, it is definitely not economical and
would result in operation at conditions where the Selexol design is quite
sensitive to sulfur removal level. Under those conditions, any unexpected
factors that might harm Selexol performance would have a much greater effect
on the ambient-temperature system. Therefore, refrigerated operation was
chosen as the design basis for this study.
Effect of Sulfur Removal Level
Based on the data presented in Tables 4-3 through 4-5, it is clear that
the major factor determining Selexol system utility consumption and cost is
whether the system design is based on H2S or COS removal. For the H2S-
based design with refrigerated operation, only a 12% decrease in solvent
flow rate (and cost) was noted when the product gas was allowed to increase
from 100 to 1000 ppm total sulfur (32 to 900 ppm H2S). Thus, it appears
that at refrigerated operating conditions, there is only a slight penalty
involved with removal to low H2S levels. However, in the case of ambient
operation with the same feed composition, a 30 percent increase in cost and
60 percent increase in steam consumption is noted as removal level is reduced
from 500 to 100 ppm total sulfur (420 to 35 ppm I^S). Thus, it appears
that at ambient temperature the system is somewhat more sensitive to the
degree of H2S removal.
STEAM TO COAL FEED RATIO
Earlier studies (e.g., Reference 8-3) have shown the benefits to be
achieved by lower steam/carbon feed ratios. With coal as the feed, this is
usually referred to as the steam/coal ratio. For combined-cycle power
generation, methane and hydrogen yields are not important so long as the
resultant gas will permit stable combustion. This allows a great degree of
latitude in operating conditions and generally would permit gasification with
no hydrogen other than that contained in the coal.
To test the effect of steam/carbon feed ratios, it is of interest to
review the somewhat easier case of an oil gasifier using the equilibrium
calculation procedure for the gasifier. Typical ranges for steam/oil ratios
would be 0.0 to 0.4 (Ib/lb) while the air/oil ratios would be 6.0 to 6.5.
The minimum air/oil ratio that would provide 1 atom of oxygen per atom of
114
-------
TABLE 8-0
SUMMARY OF PERFORMANCE AND COST
REFRIGERATED VS. AMBIENT TEMPERATURE SELEXOL OPERATION
Performance Effect of Change from Refrigerated to
Ambient Temperature
Change
Stream Consumption
P°Ver Consumption
Effect on Output
Effect on Heat Rate
+lU6,000 Ib/hr
- 13,000 kW
Performance Effect
- 7,300 kW
+13,000 kW
+ 5,700 kW
- **5 Btu/kWh
Cost Effect of Change from Refrigerated
To Ambient Temperature
Gh
Ch
a
-------
carbon is 5.0. However, to achieve a reasonable reactor size, it is neces-
sary to increase this to 6.0 thereby providing sufficient air to increase
combustion temperature and, thereby, reaction rates.
Table 8-9 shows that change in volumetric heating value of the fuel gas
is very small over the range of operational steam/oil ratios. When viewed
in terms of chemical heating value per pound of oil consumed, the output of
the gasifier is constant over the range of steam/oil ratios considered.
Also shown in Table 8-9, the effects of steam addition on composition are an
increase in hydrogen and carbon dioxide production coupled with a decrease
in CO. Stoichiometrically, each additional molecule of hydrogen in steam
brings with it ah oxygen atom which will react with one CO molecule to form
CC>2. Each added H2 molecule means one less CO molecule. Moreover, the
higher heating value of H2 and CO are almost the same so their effects on
gas heating value cancel. Further examination of the product gas shows that
about 25 percent of the input steam shows up as hydrogen and the remainder
leaves as water vapor in the product gas. Thus, the net effect of steam
addition on the product gas is minimal and the heat needed to raise the
steam is mostly lost since the latent heat cannot practically be recovered
from the water vapor in the fuel gas.
To maintain reactor temperature, air/oil ratio would have to be increased
thereby reducing chemical heating value. Even if it were possible to keep
the water vapor in the fuel gas stream (as with hot cleanup) it would be used
at a relatively low efficiency since gas turbine expansion ratios are signifi"
cantly less than those of steam turbines.
In the case of low-temperature cleanup systems, virtually all water
vapor is condensed during the gas cooling process. One method of recovering
some of this heat is by using the latent heat to heat water from a fuel gas
resaturator (Figure 8-4). This increases the mass of the fuel gas stream.
The effect on performance is illustrated by studies of resaturation of the
clean gas in the BuMines/Selexol system.
The potential performance improvement to be achieved with the addition
of water vapor to the fuel gas was estimated by varying the cleanup system
output composition. The results are shown in Figure 8-5. In determining
performance, no penalty was associated with the addition of the vapor so the
trends shown in that curve are the maximum that can be achieved. In practice,
heat must be provided for humidification and could detract from performance.
In terms of water used per megawatt of power, the incremental power produced
requires about 14,000 Ib/hr for each additional megawatt of electrical power.
For this system a 300 F dew point will produce a water vapor mol fraction of
.248. This was taken as a practical maximum since further resaturation would
require the use of heat at temperatures over 300 F while producing power at &
relatively high heat rate of 15000 Btu/kWh.
AIR-BLOWN BCR GASIFIERS
The earlier phases of this study had used available data for the BCR-tyPe
gasifier that had a high steam feed rate and was more suited to synthesis gas
116
-------
113 J
139^0
117.78
139^5
TABLE 8-9
EFFECT OF STEAM ADDITION ON FUEL GAS CHEMICAL HEATING VALUE
Fuel - Venezuelan Residual Oil
Air/Oil Ratio =6.0
Steam/Oil Ratio
Fuel Gas Characteristics 0, 0.2 O.U
Mole Fraction Hp
Mole Fraction H20
Mole Fraction CO
Mole Fraction C02
H2 SCF/lb Oil
CO SCF/lb Oil
HHV Btu/SCF
Gas Produced SCF/lb Oil
Output Gas HHV - Btu/lb Oil
.0335
.2335
.015U
16.66
26.U9
122.9
.0612
.215
.021*8
17.83
25.32
118. U
.0858
.1985
.0328
18.91
2U.2U
11U.2
122.13
139^7
117
-------
BUMINES/SELEXOL SYSTEM WITH FUEL GAS RESATURATION
oo
CLFAM GAS
TO SULFUH
RECOveRV
-------
EFFECT OF WA TER VAPOFt IN FUEL GAS BUMINES—SELEXOL. SYSTEM
8OO
3.340
790
780 —
-
-
i
i
:
h
LL
770 —
760 —
750 —
:
-
—
-
•
740
0.08 0.12 0.16
MOL FRACTION H2O IN FUEL GAS
0.20
0.24
0.28
.316
.
.-
-------
or methane production. Work subsequently performed under EPRI-sponsorship
(Reference 8-4) has resulted in descriptions of gasifier operation at lower
steam feed rates that are more suited to combined-cycle power generation. A
comparison of the two estimates is given in Table 8-10. The improved cold gas
efficiency is a good indication of relative performance but does not account
for all of the factors that will affect system efficiency.
A careful inspection of the data in Table 8-10 shows that the remarkable
increase in cold gas efficiency of nearly six percentage points is due only in
part to the reduced steam feed rate. The reduction in transport gas require-
ments is also quite significant. Transport gas is compressed to higher than
gasifier pressure and must be cooled prior to compression. A considerable
amount of heat must be supplied via combustion in the gasifier to increase both
transport gas and steam temperatures. Surprisingly, while the absolute hydro*
gen yield decreased with reduced steam, the mol fraction of hydrogen increased-
Thus, there is certainly no reason to expect a combustion problem as a result
of the change.
Low Temperature Desulfurization
In terms of sulfur removal, a more desirable ratio of H2S to COS is
evident. While total sulfur molar flow rate remains the same, the higher
partial pressure resulting from lower total molar flow rate will mean a cor-
responding reduction in Selexol system size, cost, and utilities.
The reduced C02 concentration could permit the use of a nonselective
desulfurization process. The effect of substituting a nonselective cleanup
process for the Selexol system was evaluated using utilities typical of a
hot-potassium system. The effect on overall performance was estimated to
be negligible. However, the Selexol process was retained as the baseline
system for the study. It gives reasonable performance representative of
low-temperature processes and will probably result in cost estimates that
are slightly conservative.
High-Temperature Desulfurization
The Conoco half-calcined dolomite absorber is quite sensitive to both
temperature and gas composition. Reactions in the absorber closely approach
equilibrium (Reference 8-5) and desulfurizer performance can be estimated usi°?
data for H2S and COS reactions with the dolomite. H2S absorption involves
the following reaction:
[CaC03 . MgO] + H2S [CaS . MgO] + H20 + C02 (8-3)
for which the equilibrium H2S concentration in the gas is given by
[H2S] = [H20] [C02] P/K (8-4)
120
-------
COMPARISON OF BCR DATA
Coal - Illinois No. 6 - 700, 000 Ib/hr.
Air-Blovn Gasifier
CHU
H2
CO
C02
H2S
COS
N2
NH3
H20
HHV-Btu/SCF
Air/Coal Ratio
Steam/Coal Ratio
Transport Gas/
Coal Ratio
Cold Gas Eff/2'
1975 EPA
Mols/Hr (l)
5,188
18,319
26,151
11,751
687
lit 2
65,^99
573
13,953
11+2,263
138.0
3.09
.567
.1+26
7 O cot
| O. p/»
Study
Mol %
3.65
12.88
18.38
8.26
O.U8
0.10
1+6.01+
0.1+
9.81
EPRI Data
Mols/Hr (l)
3,775
15,315
32,190
3,396
751
76
53,753
1+79
2,213
111,91+8
171.2
2.78
.088
Q3%
Mol %
3.37
13.68
28.75
3.03
0.67
0.07
1+8.02
0.1+3
1.98
(l) Includes transport gas
(2) Based on net gas make - excluding transport gas
-------
where [ ] = mol fraction of component
P = total pressure, atm
K = equilibrium constant
From Figure 8-6 and Equation 8-4, it is clear that the degree of H2S
removal increases with temperature for a given gas composition. Also, the
degree of removal at a given temperature is inversely proportional to tempera-
ture and concentration of the reaction products, C(>2 and H20.
Although no data have been reported for COS adsorption by half-calcined
dolomite, high COS removal efficiencies are predicted thermodynamically
according to the reaction:
[CaC03 . MgO] + COS [CaS . MgO] + C02 (8-5)
where the residual COS concentration is
[COS] = [C02]2 P/K (8-6)
The equilibrium constant for this reaction is shown in Figure 8-7. The
degree of COS removal like that for H2S increases with increasing temperature
and decreasing C02 concentration.
From the above, it would appear that reduced steam feed to the gasifier
with its associated reduction in H20 and C02 in the off-gas should
improve desulfurization performance. However, with the low H20 and C02
content, a third reaction becomes important, that of decomposition of the
acceptor. It imposes a maximum operating temperature for this process
depending on the partial pressure of carbon dioxide in the gas phase; i.e.,
the temperature should not exceed that at which the C02 partial pressure is
equal to the decomposition pressure for CaC03 via the following endothertnic
reaction:
Ca C03 CaO + C02 (8-7)
The equilibrium decomposition pressures for calcium carbonate are given in
Figure 8-8.
An example of the characteristics of the Conoco absorption process is
given in Figure 8-9 using a BCR-type feed having a high water vapor and C02
composition; the partial pressure of C02 in the product gas is shown as a
solid line for each of two different operating pressures (the variation with
temperature is due to the water gas shift reaction). The limiting or mini«nuin
partial pressure shown by the dashed line is the equilibrium partial pressu*6
of C02 above CaC03. The partial pressure of C02 must be greater than
this to prevent decomposition of the CaC03 to CaO. The intersection (Point
A) indicates that at the absorber operating pressure of 450 psia, the maxii°oin
operating temperature could be 1780 F and the residual sulfur would be 325 ""'
122
-------
FIG. 8-6
EQUILIBRIUM CONSTANT FOR H2S ABSORPTION BY HALF - CALCINED DOLOMITE
1000[
I
I
8
100 —
10 —
K =
-------
FIG. 8-
EQUILIBRIUM CONSTANT FOR COS ABSORPTION BY HALF-CALCINED DOLOMITE
104
103
.'
I
f>
-
<
a
!
102
10
K =
-------
FIG. 8-8
DISSOCIATION PRESSURE FOR CALCIUM CARBONATE
1200 1400 1600 1800
TEMPERATURE, F
''I III!!
2200
R1-19 19
125
-------
FIG. 8-9
SULFUR ABSORPTION BY HALF CALCINED DOLOMITE
0.
LU
.::
I
111
CC,
0.
1
t
Q-
(N
O
40
/
-
30
/
X
10
,
MINIMUM PCO2 TO PREVENT
DECOMPOSITION
MAX TEMPERATURE BASED ON THERMAL LIMITATION
600
a.
0
Q
i
11
i
i
1
•
in
ij
Q
500 -
400-
300
MAX TEMPERATURE
TO AVOID
DECOMPOSITION
1600
1700
1750
1800
TEMPERATURE, °F
126
-------
°mt A"). However, the product gas enters the absorber at only 1750 F and
th ^lance constraints limit the maximum temperature to about 1650 F since
f e combined chemical reactions are endothermic to the extent of 112 Btu/mol of
/ §as. The residual sulfur content at this temperature is around 600 ppm
vpoint B in Figure 8-9).
In the case of the air-blown BCR gasifier with low steam feed, desulfurizer
,.essure is 340 psia and partial pressure of carbon dioxide in the feed is only
Psia. Thus the temperature in the absorber must be limited to 1600 F to
l7n decomposition. For that reason, the raw gas which leaves the gasifier at
^ u F must be cooled prior to desulfurization. The lower temperature reduces
lee ecluilbrium constants in Equations (8-4) and (8-6), but that reduction is
j Ss than 50 percent while C02 and t^O concentrations are reduced by a
tor of 2.5 and 5.0 respectively. The resultant sulfur in the gas (based on
brium) is only 106 ppm as opposed to approximately 600 ppm for the
steam feed case.
STUDIES OF THE U-GAS/CONOCO SYSTEM
Because of the potential benefits of reduced steam feed rates, the effect
at tlle U-Gas system was investigated. The gasifier equilibrium model developed
H_. ^C was used to predict the effect of variations in steam feed rates while
w- |Jtaining a fixed reaction temperature and constant heat leak. Discussions
Of —* supported the validity of the study and led to their recommendation
•25 lb H20/lb coal as the minimum practical ratio for the U-Gas process.
ift p°r an Illinois No. 6 coal feed, the concentration of ^0 and C02
ji^ "e raw gas as a function of steam/coal ratio to the gasifier is given in
CQjJi*6 8-10. The effect on sulfur removal is shown in Figure 8-11. A steam/
CQ feed ratio of 0.1 is limiting from a theoretical standpoint in that the
Ale part-ial pressure is too low to prevent decomposition of the carbonate.
£0 °' as C02 concentration is reduced, there is the potential for carbon
the at*0n v^a the Boudouard reaction. A steam/coal ratio of 0.2 coupled with
Thisresidual moisture in the coal gives a total H20/ coal ratio of 0.24.
ten, Va*ue was selected as the basis for operation of a U-Gas/Conoco high-
rature desulfurization system. The estimated product gas characteristics
l§h-temperature desulfurizer performance for this system are given in
8-11.
hav *- should be noted that in both the BCR and U-Gas cases, the desulfurizers
Of t, en assumed to run at a temperature approximately 100 F lower than that
evai e 9°n°co design upon which performance and cost is based. Quantitative
avaj,ati0n of the effect on approach to equilibrium of that difference was not
a la able- A longer residence time will most likely be required resulting in
8gr desulfurization system.
MODELING
V6re Modeling techniques developed in earlier phases (References 8-3 and 8-6)
Used to verify data obtained during this program. In addition to use of
127
-------
FIG. 8-1
U-GASGASIFIER EQUILIBRIUM COMPOSITION AT 2010R
3.1
3.0
I
u
<
'
2.9
•
a
ii
I
o
CvJ
I
D
<
O
O
0.10r
0.09 -
0.08 -
0.07 -
0.06 -
2.8
0.05 -
STEAM/COAL FEED RATIO
128
-------
FIG.8-11
HIGH-TEMPERATURE CLEANUP SYSTEM PERFORMANCE
0.8-1
0.7-
0.6-
0.5-
,
CD
S
CQ
-J
§
CD 0.4
2
<:~'
.:
Q
,
U
3
',
0.3-
0.2-
0.1-
o-1
i
a
I
I
•i
o
'
I
00
I
CO
CN
I
700 -
600 -
STEAM/COAL FEED RATIO
77-09-87-1
129
-------
Comp.
TABLE 8-11
COMPARISON OF HIGH VS. LOW STEAM PERFORMANCE
U-Gas Gasification/Conoco Desulfurization
Coal Feed - Illinois #6 - 700,000 Ib/hr
High Steam Feed
Gasifier Out Desulfurizer Out
Mol/Hr
Mol/Hr
Low Steam Feed
Gasifier Out Desulfurizer Out
Mol/Hr Mo/Hr
CHj^ U,002
H2 '20,335
CO 23,738
co2 11,178
HgS 779
COS 32
N2 57,801
NH3 U2
H20 lU,013
Total 131,920
HHV-Btu/SCF 138.8
Air/Coal Ratio 3.01
Steam/Coal Ratio 0.557
Transport Gas/ 0.053
U,002
21.U98
22,57^
13,071
90
12
57,801
U2
13,539
132,629
138.1
3,72U
16,926
29,126
5,832
763
ill
55,053
31
3,606
115,102
162
3,12k
17,003
29,0^9
6,732
17
3
55,053
31
i*,275
115,887
160.9
2.86
0.2
0.053
Coal Ratio
Cold Gas
Efficiency - %
80.2
81.5
-------
e equilibrium model in evaluating the effects of steam/coal ratio, it was
,ed to identify the possible pollutant compounds and potential levels that
l§ht be expected for each of the gasifiers studied. These are presented in
Cables 8-12 and 8-13. While the results given in Tables 8-12 and o-13 are
ubject to the limitations of an equilibrium calculation not absolute in char-
, r> they are useful to get a feel for the degree of optimism or pessimism
°ut pollution control for the fuel gas from each type of gasifier.
131
-------
TABLE 8-12
GASIFIER EQUILIBRIUM COMPOSITIONS
•
H20
CO
H2
co2
E
1975 Data
j Adiabatic
Predicted' Eauilib.
.0981
.1838
.1288
.0826
N2 .U6ol*
OH,
NH3
COS
H2S
COS/H2S
T-°F
.0365
! .OQl*
.001
j . OOU8
.021
1750
.0816
.1920
.1693
.0801*
.1*1*83
.0222
CR/Air-Blown
Current Data (Low Steam Feed)
Predicted
.0198
.2875
.1368
.0303
.1*802
.0337
Adiabatic
Equilib.
.0177
.2908
.1511
.0276
.1+762
.0291
.0001*6 .001*3 .00023
.0002 .00067 .0001*2
.0051*3
.037
1595
.0067 .0069
0.1 .061
1700 , 1636
Isothermal
Equilib.
BCR/0^ - Blown
Predicted
c.
Adiabatic
Equilib.
.0160 .ll*l*0 . .1339
.2997 .3533 j .3527
.161*8
.0211*
.1*693
.0215
.00022
.2950
.1251
*
.001*5
.o6oU
.006U
.OOOU .0010
j .0068 .0103
.059 0.1
! 1700
.3183
.1237
Isothermal
Equilib.
i
Predicted
II- Gas
Adiabatic
Equilib.
.1272 .1062 .0891*
.3760
.31*62
.1012
.180 ; .1851
.151*1 .1883
.081*7
; j
.0075
.0528
.00009
.00038
.0107
.035
I
1700 1619
.0072
.0315
.00008
.1*382
.0303
.00032
I
.00036 i .00021*
.0103
.035
i 1700
.0059
.01*1
1660
.0852
.1*278
.0178
Isothermal
Equilib.
.0893
.2058
.2052
.0698
.^179
.0058
.00035 .00028
j
.00018 j .00018
.0058
.031
.0057
.031
1533 ( 1660
-------
TABLE 8-13
MOLTEN SALT GASIFIER EQUILIBRIUM COMPOSITION
J:jole. Fractions Kellogg Data Equilibrium
CO
H2
CO,
C(s)
NaOH(g)
cos
NH3
HCN
NaCN(g)
HCNO
HCO
0.1+969191
0.2768238
0.131+7269
0.0301001
0.022914U8
0.011+6700
0.0113919
0.00521+93
0.001+1051+
0.00201+29
0.00096ol+
0.0000651+
0.1+8962
0.29089
0.15312
0.022750
0.018882
0.0068163
0.011713
0.0056839
0.000021+0
0.000lt7»tl
0.0000292
Equilibrium
0.1+8957
0.29085
0.15290
0.022810
0.018908
0.0067769
0.011699
0.0056828
0.0000239
O.OOQl+760
0.0000293
0.0001790
0.0000560
0.0000283
0.0000070
O.OOOOOll*
0.0000013
0.0000009
0.0000001+
0.0000003
0.0000001
0.0000001
Totals
1.0000000
1.0000025
1.0051+637
133
-------
REFERENCES
8-1 Private Communication. Allied Chemical Co., May 1975.
8-2 Private Communication. Allied Chemical Co., January 1977-
8-3 Ro"bson, F. L. , A. J. Giramonti, W. A. Blecher, and G. Mazzella: Fuel ,\
Gas Environment Impact - Phase Report. EPA-600/2-75-078, (NTIS Mo. PB2^9-1*2
November 1975.
8-U Chandra, K., B. McElmurry, E. W. Weben, and G. E. Pack: Economic Studies °
Coal Gasification Combined Cycle Systems For Electric Power Generation.
EPRI AF-6U2, January 1978.
8-5 Curran, G. P., B. J. Koch, B. Pasek, M. Pell and E. Gorin: High-Temperate6
Desulfurization of Low-Btu Gas. EPA-600/7-77-031, (NTIS No. PB271-008),
April 1977.
8-6 Robson, F. L., W. A. Blecher, and C. Colton: Fuel Gas Environment Impact-
EPA-600/2-76-153, (NTIS No. PB257-1310, June 1976.
134
-------
SECTION 9
POWER SYSTEM EMISSIONS
INTRODUCTION
c A significant advantage of integrated coal gasification/gas cleanup/
Ca lnec* cycle power plants is the reduced amount of air pollutants they
be designed to emit as compared to conventional coal fired plants with
6 S ^esul^urizati°n- This section discusses those emissions for which
dp • contr°l equipment has been incorporated or where special component
^ X8ns are required; i.e., sulfur dioxide, nitrogen oxides and particulates.
th em*ssi°ns of these primary pollutants are summarized in Table 9-1 for
lo Seven power plant configurations under consideration. herein. In all
""temperature cases, emission levels will be equal to or better than
v-j lcable standards. However, the systems with high-temperature cleanup
te, require additional fuel processing or combustion modification to
j Ce nitrogen oxides resulting from ammonia or other nitrogen compounds
the fuel.
t . Sulfur levels given in Table 9-1 include the effect of Glaus plant
th ~^>as cleanup and are nearly all due to the presence of H2S and COS in
in ^- Low-temperature designs are quite sensitive to COS and increases
Or the COS level could have a significant effect on desulfurizer performance
a c°st. High-temperature desulfurizer performance estimates are based on
as ec*u*-libriuin calculation and, to avoid decomposition of calcium carbornate
OB a result of the low C02 concentration in the fuel gas, are based upon
rating temperatures some 100 F lower than those at which Conoco did
sting.
NQ NOX levels in Table 9-1 assume that all ammonia is converted to
for" ^e use °^ a Pre~mixe^ combustor promises to reduce thermal NOX
de mation by 80 percent over conventional combustors. Test data were
Te ®l°ped by Turbo Power and Marine Systems, a subsidiary of United
nologies Corporation, in cooperation with the Texaco Development
135
-------
TABLE 9-1
EMISSION SUMMARY
u>
High-Temperature Cleanup
Emission
Th/1 n Btu
Sulfur
Gas Turbine Exhaust
Sulfur Recovery
Nitrogen Oxides
Thermal
U-Gas
High-Steam
Selexol
0.18
0.002
< 0.35
liow— Teiupei
U-Gas
Low-Steam
Selexol
0.23
0.003
< 0.35
BCR
Air-Blown
Selexol
0.39
0.006
< 0.35
-••p
BCR
Op- Blown
Selexol
0.37
0.006
> 0.35
Molten
Salt
0.21
o.ooU
< 0.35
U-Gas
Low-Steam
Conoco
0.15
O.OU
< 0.35, .
0.1T(2)
BCR
Air-Blown
Conoco
0.09
0.0k
< 0.35
2.52
BCR
02- Blown
Conoco
0.69
o.oU
> 0.35
2.60
Fuel Bound
Particulates
.01
.01
.01
.01
.01
.03
.03
.03
(l) Based on 100$ conversion to NO
(2) NH3 in fuel gas "based on equilibrium calculation - actual likely higher
-------
p
mPany. These data demonstrate this reduction when various low-, medium-
c high-Btu gases are burned in a pre-mixed burner. Extrapolation from
6U5rent data shows that levels of 75 ppm (corrected to 15% 02 in the
aust) could be achieved with the pre-mix concept in all but the oxygen-
°yn systems. For those medium-Btu gases, the fuel temperature may be
ited to some value lower than 1000 F in order to limit NOX formation.
There are problems associated with a pre-mix design, but ongoing
search and development should solve them. Unfortunately, the pre-mix
ncePt relies on fuel lean combustion to achieve the lower thermal NO
( ~*
_,3) is converted to NO. Thus, where there is NH3 in the fuel, an
, • — — *. j, »_, ^j VSLA ^UVvil ^VvULl W \SlllbS W tj W -*r V ** w w w v> .—•—»•— —... — _ _ ., v _ _.,. v _ ...» _ . . X
' lssion level. Under these conditions virtually all fuel-bound nitrogen
-•i ** — w t,ULLVe L I t?U CU ll\J • i.UUO , WIIGI.C UHCJ.«i J-o nnj 4-11 i~ut^ j.uv^J., all
ernate approach will be needed. Tests have been run with staged combus-
r techniques using a fuel rich burn followed by a rapid quench. The
t.Sults show effective reduction of NO production. An evaluation of combus-
°n modification techniques as opposed to fuel processing is needed to
ect the best approach.
Erosion of gas turbine blading limits the quantity and size of parti-
iate matter that may be in the fuel. Low-temperature particulate removal
, °cesses are available to meet these limits. No high-temperature process
t s been demonstrated to be effective enough. Such a process is essential
, the use of high-temperature desulfurization, and its feasibility has
etl assumed herein when postulating high-temperature desulfurization
Since limited data are available, a definitive particulate
requirement cannot be stated. It is not clear whether or not
t - 7temperature systems will be able to meet the needs of advanced gas
c rbines in this important area. It is believed that current estimates are
w°nservative and could be met by a water wash. Of course, a water wash
uld obviate the basic advantage of a high-temperature desulfurizer.
PINITION OF GOALS FOR CLEANUP SYSTEMS
Absolute guidelines are lacking for the degree of cleanup needed to
adverse health and ecological effects. The capabilities of the
; 1lable or potentially available equipment here played an important part
t, arriving at the goals for the systems studied herein. Additionally,
g 6re are basic guidelines, such as current or proposed EPA standards for
v turbine and other power generating systems. They must be equalled or
tered for the systems to be attractive. These emission standards are
esented in Table 9-2.
Because of the high metal temperatures, corrosion and erosion can be a
6st- m *"n Sas turbines. As a result, gas turbine manufacturers have
g ablished stringent specifications for fuels to be burned in industrial
T, Plications. A summary of pertinent specifications is given in Table 9-3.
^ajor problem in current use of gas turbines by utilities is hot end
tosion. This results from alkali metal salts and sulfur in the fuel.
corrosion agent is the alkali metal sulfate which attacks the oxidation-
137
-------
TABLE 9-2
COMPARISON OF EMISSION STANDARDS
Standards For Proposed for Gas Turbines'^
Conventional Plants (Ref. 9-1)
Coal-Fired Oil-Fired Distillate-Fired
Sulfur as S02
Ib/MMBtu 1.2 0.8 1.0
Nitrogen Oxides
as N02-lb/MMBtu 0.7 0.3 0.35 to 0.6
Particulates
Ib/MMBtu 0.1 0.1 None
(1) Standards are given in ppm in stack gas with 15% oxygen.
(2) Variation is due to allowance for fuel-bound nitrogen -
additional allowance made based on heat rate.
resistant coatings. Once these coatings have been penetrated, rapid
oxidation of the base alloy occurs. Thus, although the turbine could
withstand fuels having sulfur contents well above those allowed by environ-
mental constraints, the presence of alkali metals reduces the allowable
content to a much greater degree. The presence of particulates also contri-
butes to corrosion since any spall ing of the coatings due to impingement
could result in subsequent oxidation of the base alloy.
Particulate Loading
One of the major concerns in setting fuel specifications is the
particulate content. As can be seen in Table 9-3, the particulate loadings
are low, varying from less than 1 ppm to about 30 ppm. The reason for the
low allowable loadings is the potential erosion of the turbine blades.
A number of attempts have been made to utilize coal directly in a
turbine (References 9-2, 9-3, 9-4). Due to excessive erosion, all have
failed to obtain reasonable machine lifetime. Yet, the actual limitations of
particulate content or size distribution have not been adequately defined.
A previous UTRC study (Reference 9-5) on methods of cleaning emissions from
jet engine test cells showed that the mean diameter of the particles
emitted from liquid-fuel burning engines was 0.1 y with over 99 percent
of the particles being less than 1 p. Since erosion is not a problem with
these engines, it can be concluded that particles of 1 u or smaller in the
fuel or combustion products are not harmful to engines. In fact, it was
pointed out in this study that the use of combustion additions to eliminate
visible emissions resulted in agglomerated particles of greater than 1 u
138
-------
Constituent
Sulfur
Particulates
Metals
P&WA Spec. 527
1.8 Mol % H2S
(1)
0.08 It/106 ft3
(0.00056 gr/ft3)
9-3
GAS TURBINE FUEL SPECIFICATIONS
Suggested
<1 Mol > K2S or Less Than
Amount Required to Form
5 ppm Alkali Metal Sulfates
0.01 gr/ft3
HO y Maximum
Westinghouse
2% by Weight
(3)
Limits "by Material
General Electric
(1)
Less Than Amount
Required to form
3 ppm Alkali Metal
Sulfates
30 ppm (Weight)
(0.01 g/ft3)
Vanadium
Sodium and
Potassium
Calcium
Lead
Copper
<0.2 ppm (Weight)
<0.6 ppm (Weight)
0.1 ppm (Weight)
0.1 ppm (Weight)
0.2 ppm (Weight)
see Sulfur Spec,
0.5 ppm (Weight)
0.5 ppm (Weight)
10 ppm (Weight)
2 ppm (Weight)
(l) For aircraft-derived turbine using gaseous fuels
(2) For industrial turbines; subject to revision
(3) Liquid fuel specifications
(U) Given in Ref. 5-9 as U x 10~ gr/ft of 2y to lOy particulate in gaseous fuels
See Sulfur Spec.
-------
(above the size range which results in refraction). While deposition did
occur, there was no erosion which indicates that the soft compounds which
left the airstream and impinged on the blades did not damage them.
Tests in the United Kingdom on fluid-bed combustion (Reference 9-6) also
indicate that soft particles with quite large diameters of about 50 u do
not appear to cause erosion. However, hard particles may result from coal
combustion above the ash fusion point Tests carried out in Australia
(Reference 9-7) indicate that such particles cause erosion at sizes above
6u. Perhaps the most useful guide is work done in the petro-chemical industry
on turbine erosion (Reference 9-8). Results of this work are shown in
Figure 9-1, where engine life is shown as a function of particle size and
loading. The,shape of Figure 9-1 indicates that the relationship between
lifetimes, particulate loading and particulate size would be of the form
Constant (9-1)
L w . c
where
L = lifetime
p = particle size
C = loading
Thus, for a given lifetime, a reduction in particle size would be
accompanied by higher allowable loadings. For example, assuming linear
particle size distribution, capture of particles of 2 u and above would
allow twice the particulate concentration compared to what would be allowed
by capture of 4 y and larger particles.
Suggested Low-Btu Fuel Gas Specifications
The values for loadings in Table 9-3 are based upon methane fuel. For
low-Btu fuel, both the heating value and the density will affect the allowabl6
loadings. For fuel gases with heating values of 150 Btu/ft-^ (about 2500
Btu/lb), the value in Table 9-3 would be reduced by approximately a factor of
eight if the solids loading to the turbine were to be kept constant. Using
the foregoing approach, Table 9-4 has been prepared to serve as a guideline
for low-Btu fuel gas.
In addition to the usual fuel property limitations, Table 9-4 includes
a value for fuel nitrogen compounds expresed in terms of NH3. The limitation
is based upon 90 % conversion of fuel nitrogen to NOX.
SULFUR EMISSIONS
The variation in sulfur emission shown in the summary table is somewhat
arbitrary. In most cases, changes in equipment design could be effected to
140
-------
EFFECT OF PARTICLE SIZE ON ENGINE LIFETIME
FIG. 9-1
PARTICULATE LOADINGS.gr/ft3
102
2 5 104
LIFETIME, hrs
RO2 -192--1
-------
TABLE 9-U
SUGGESTED LOW-BTU FUEL GAS CLEANUP SYSTEM GOALS
Property
Sulfur
Particulates
Specification
0.05 Mol % or Less Than Amount
to Form 0.6 ppm Alkali Metal
Sulfates
U ppm (weight)
(0.0012 gr/ft3)
Resulting,
O.UH Ib
<0.01
Metals
Vanadium
Sodium and Potassium
Calcium
Lead
Copper
Nitrogen Compounds
<0.03 ppm (weight)
See Sulfur Spec
<0.012
<0.012
<0.0025
500 ppm as NHo
0.3 Ib NO,/***
142
-------
the levels. Also, estimated raw gas compositions have a decided
on cleanup performance. Such factors have been examined in an
, - to identify the sensitivity of each design to sulfur emission
level.
An important factor in all systems, especially as removal efficiency
ases, is the form in which the sulfur appears in the raw fuel gas. In
study it has been assumed that all sulfur would be in the form of COS
cr ^28; however, potentially significant quantities of other sulfur
0lnPounds could be present (as compared to a 100 ppm level in the clean
?as) Since other compounds have not been definitively or quantitatively
* entified, their effect on the system has not been evaluated in any
etail. While it is anticipated that the cleanup processes will be able to
°Pe with such gases, removal or recovery performance will likely suffer.
j Us» where sulfur emissions are shown to be on the order of 0.1 lb/10^
J-ui it is quite possible that they could be higher by a factor of two or
at the associated equipment may be more complex than anticipated.
Systems
With the exception of the molten salt system, all the systems
herein incorporate a Selexol desulfurizer with sulfur recovery via a
t aus plant with Beavon tail-gas cleanup. Glaus plant emissions are vir-
ually 2ero an(J the emission level is ultimately determined by the Selexol
Qystem design. In the case of the molten salt gasifier, sulfur removal
nC°Urs both in the gasifier and in the CC>2 recovery tower which is a
tj"Cessary part of the salt recovery process. The l^S is stripped from
e solution and sent to a Glaus plant with tail gas cleanup as in the
designs. Thus, in both cases, sulfur emissions depend only on the
of removal of sulfur compounds from the main fuel gas stream.
Sei
ie*ol Design Basis—
(, Each of the Selexol designs is based on removal of virtually all H^S
own to approximately 30 ppm). Under these conditions, only 30% of the
t,S is removed. Thus, nearly all of the sulfur emissions are due to COS in
6 fuel gas. The level of COS in the raw gas would be on the order of 5%
a total sulfur at equilibrium (see Tables 8-12 and 8-13). As hard data
t e generally unavailable, estimates vary between equilibrium and two times
enf e
-------
When solvent flow rate is sized for H2S removal, approximately one
third of the COS is removed. To remove COS to the 100 ppm level requires
that solvent flow rate be significantly increased. To achieve a high
degree of removal of a particular component requires a solvent flow-rate
inversely proportional to its solubility. The effect of this change on the
performance and cost of two air-blown BCR-type gasifiers designs is com-
pared in Table 9-5. Utilities increase by a factor greater than two, while
cost is approximately doubled. Power output was penalized at the rate of
20 Ibs steam/kWh giving a total decrease in plant output of some 38 MW in
going from I^S to COS based design. Of the $48/kW increase in fuel
system cost, some $10/kW can be attributed to the decreased output.
The Selexol desulfurizer systems for the other gasifiers considered
were sized on the basis of H2S removal. Therefore, a similar improvement
in COS removal can be expected for each of the other systems with a com-
parable increase in both cost and utility consumption. The conclusion to
be drawn is that desulfurizer cost can vary by a factor of two depending on
the composition of the raw fuel gas.and the degree of sulfur removal which
is required.
Molten Salt System—
Because 85% of the sulfur is retained in the melt, total sulfur in the
raw gas is quite low. The COS level is reduced correspondingly and, even
with no COS removal, the emission level is quite low. After leaving the
gasifier, the raw gas is scrubbed with a sodium carbonate solution to
recover C02- During this process, most of remaining H2S is removed.
In this study, no COS removal is assumed (Reference 9-9). This may be a
conservative estimate as COS is reported to be hydrolyzed (Reference 9-10) i°
the hot potassium carbonate solution of the Benfield system. Under appro-
priate conditions complete COS removal is claimed for that process.
NITROGEN OXIDES
Conventional gas turbine burners firing distillate fuels are concerned
only with thermally generated NOX. Its formation is both temperature and
time dependent. Water injection is the current method used to control
thermal NOX emissions, but they could eventually be controlled by burner
design. Fuel gas from coal can be expected to contain significant amounts
of nitrogen bearing compounds (usually lumped together as ammonia). At the
levels expected and under the expected burner design conditions, virtually
all of the nitrogen in the ammonia will form nitric oxide (Ref. 9-11). It
so happens that measures most appropriate for reduction of thermal NOX
tend to enhance the production of NO from ammonia. Because of this, both
problems will be treated separately. Low-temperature systems which include
a water wash need be concerned only with thermal NOX. In the high-
temperature systems, a choice must be made between further fuel processing
and combustion modification.
144
-------
TABLE 9-5
COMPARISON OF SELEXOL DESIGNS
BCR Air-Blown Gasifier
H2S Based COS Based
Design Design
' P Steam Required - Ib/hr 131,500 317,000
6r ^quired - kW 20,1*00 U8,800
Cost - $106 26.6 51.1
,
ln Clean Gas - ppm 32 10
4
ln Clean Gas - ppm ^50 90
i
^ Emission - Ib S02/MMBtu 0.39 0.08
Output - MW 1,098 1,060
(jj Astern Cost - $106 277- 5 319-1
W uding Contingency, Escalation and
During Construction)
Cost - $/kW 253 301
Efficiency - 0.1439 O.k2k
145
-------
Thermally Produced Nitrogen Oxides
Data for various types of fuel gas in conventional burners show that
low-Btu fuel gas will produce little thermal NOX. The range of data for
low-, medium-, and high-Btu gas with fuel at ambient temperatures is given
in Figure 9-2. However, reheating of the clean fuel gas, while improving
efficiency, results in a higher flame temperature and, thus, more thermal
NOX. As Figure 9-2 shows, burning at off-stoichiometric conditions,
i.e., at lower than stoichiometric flame temperatures, offers a means of
controlling production of thermal NOX. The data in this figure were
obtained by Turbo Power and Marine Systems a subsidiary of United Technologic8
Corporation. They have been carrying out a series of tests on low- and
medium-Btu fuel gases produced by an experimental gasifier at the Texaco
Development Company's Montebello, California research facility.
Of particular interest are results from testing of premixed (fuel and
air mixed prior to introduction into the combustor) burners. Figure 9-2
(Reference 9-12) shows the NOX emissions as a function of source temperature
rise for several values of fuel chemical heating value. In the test series,
the fuel gas was delivered at essentially ambient temperature. The theore-
tical stoichiometric temperatures are given. In conventional burners,
indicated by the areas between the solid lines, stoichiometric temperatures
are generally achieved. Also shown in Figure 9-2 are the approximate emissio0
for burners in which the fuel and air have been premixed. These are shown
as off-stoichiometric conditions since complete mixing will eliminate the
possibility of stoichiometric combustion which occurs in conventional
burners. While no significant reduction was noted for low-Btu gas (the
emissions were already low), use with the medium-Btu gas indicated a large
reduction.
Stoichiometric flame temperatures given in Figure 9-2 apply to the point
of maximum burner AT. Reduced burner AT corresponds to part throttle
conditions for a gas turbine. At part throttle conditions, combustion air
temperature is decreased with a corresponding decrease in stoichiometric
temperature and NOX production. As flame temperature is increased above
3600 R, the increase in thermally produced NOX is exponential. For the
systems studied here, fuel gas temperatures ranged from 800 to 1000 F and
the resultant stoichiometric flame temperatures, given in Table 9-6, are
in the high thermal NOX zone. Figure 9-3 (Reference 9-13) gives an estimate
of the concentration as a function of combustor exit temperature and theore-
tical flame temperature. It differs from the data of Figure 9-2 in that it
shows a stronger temperature dependence. However, it is clear from both
figures that conventional combustors will result in the production of
excessive thermal NOX. (Note that the data of Figure 9-2 is at atmospheric
pressure and must be corrected for turbine pressure ratio.).
146
-------
IMOX PRODUCTION FROM COMBUSTORS BURNING
LOW-BTU AND MEDIUM-BTU GAS
FIG. 9-2
60
X - OFF STOICHIOMETRIC MEDIUM BTU FUEL
O - OFF STOICHIOMETRIC LOW-BTU FUEL
50
40
MEDIUM-BTU DATA BAND
(300 BTU/SCF)
\ (PPM) 30
20
NATURAL GAS DATA BAND
(1000 BTU/SCF)
900
1000
LOW-BTU DATA BAND (100 BTU/SCF)
1
1
1
1100 1200
BURNER AT
1300
'FLAME
= 4450 R
1 FLAME
= 4320 R
'FLAME
= 3650 R
1400
76-02-91-4
147
-------
TABLE 9-6
THEORETICAL FLAME TEMPERATURES
Air at 793 F
Equivalence Ratio = 1.0
System
U-Gas Selexol
U-Gas Low-Steam Selexol
BCR Air-Blown Selexol
BCR 02-Blown Selexol
Molten Salt
U-Gas Low-Steam Conoco
BCR Air-Blown Conoco
BCR 02-Blown Conoco
Fuel
HHV
Btu/SCF
158.6
169.7
177-9
332. U
151*. o
160.9
170.0
268.0
Gas
Temp
F
oLf.il,
QLj.il
926
926
805
1,000
1,000
1,000
Fuel/Air
lb/rb
.681
.639
.608
.262
.72U
.675
.639
.335
Flame
Temperature
F/R .
STS-AW*
3870/14330
39U6/UH06
U14914A95*
3800/14260
3805A265
3900/14360
142140A700
148
-------
FIG. 9-3
NITRIC OXIDE FORMATION IN GAS TURBINE BURNER
CONSTANT BULK GAS FLOW RATE
FIXED BURNER VOLUME
STOICHIOMETRIC FUEL/AIR RATIO IN RECIRCULATION ZONE
BURNER PRESSURE = 12.5 ATM
_^. HIGH-BTU ^.
FUELS
LOW-BTU FUELS
1.0 -
0.11
3400
3600 3800 4000 4200
MAXIMUM COMBUSTION TEMPERATURE
4400
- R
4600
76-02-91-1
149
-------
Thermal Formation Mechanism —
Complex computer simulations have been developed by United Technologies
and others that model the combustor internal flow-field, the combustion
reactions, and the thermal NOX kinetics (Reference 9-14 through 9-19). The
simplified NOX kinetic predictive techniques are generally based on the
fact that the NO formation rate is very slow relative to the hydrocarbon
combustion reaction rate so that the two can be decoupled in predicting NO
formation; i.e., the combustion reactions can be assumed to be at equili-
brium in estimating NO formation rates. This is illustrated in Fig. 9-4
which is the result of a kinetic model considering both combustion and
NO kinetics (Reference 9-20). For that system, the combustion reaction is
essentially complete in 60 microseconds at which time the NO concentration
is several orders of magnitude less than its final or equilibrium value.
The thermal mechanism for formation of NO from nitrogen and oxygen was
originally proposed by Zeldovich (Reference 9-21). It consists of a chain
of two reactions for the production of nitric oxide:
0 + N2 * NO + N
N + 0+ NO + 0
While there is evidence of the formation of "prompt" NO (Reference
9-22), it is believed that the bulk of the NO is formed by the thermal
mechanism. A third reaction,
OH + N + NO + H (9
can become important when the oxygen concentration becomes low under
fuel-rich conditions. Several equivalent solutions to the above equations
are given in References 9-23 through 9-25. They all assume that the
nitrogen atoms are in equilibrium with the products of combustion and that
the flame temperature is constant.
By making several approximations, a simplified solution (Reference
9-25) can be obtained that shows good agreement with more complex models
and correlates with test data:
, = 3* (9
-4)
-5)
150
-------
FIG. 9-4
CONCENTRATION-TIME PROFILES FOR PREMIXED H2-CO-AIR MIXTURE
j
2
INLET TEMPERATURE 1800 R
PRESSURE = 10 ATM
COMBUSTION TEMPERATURE = 4460 R
EQUIVALENCE RATIO = 1.0
TIME, SEC
151
10
-------
where p _ XNQ _ _ mole fraction NO _
(Xpj())e mole fraction NO at equilibrium
t = apparent residence time, sec
and 6 = 4.24 x IfllS Pl/2 (x vl/2 -114.572 (9-6)
T 2 RT
where 9 = NO formation parameter, sec~l
T = temperature, K
P = pressure, atm
X = mole fraction of N~
R = gas constant, 1.987 cal
mole - °K
As suggested in Reference 9-23, if XN is assumed to be constant,
the curve shown in Figure 9-5 would result. The mole fraction of N2 is
in the range of .71 to .73 for distillate and low-Btu gas. It can be as
low as 0.6 for medium-Btu gas but this only introduces a 10% error. The
above are a close approximation for values of p on the order of 0 to 0.2
and equivalence ratios near unity. Equivalence ratio is defined as the
ratio of actual fuel/air ratio and the stoichiometric fuel/air ratio.
Equilibrium composition for a low-Btu fuel is shown in Figure 9-6. In
estimating NOX formation, the parameters of interest in Equation 9-5 are
equilibrium NO concentration and adiabatic flame temperature. These are
presented in Figures 9-7 and 9-8 as a function of equivalence ratio and
fuel temperature. While fuel temperature does have a significant effect on
flame temperature and rate constant, it is also apparent that adjustment of
the equivalence ratio (off-stoichiometric combustion) could be valuable in
offsetting their bad effects.
Reduction of Thermal NOX—
The NOX reduction potential for off-stoichiometric combustion is
shown in Figure 9-9 for two modern aircraft engines burning liquid fuel.
Data for methane with air in a well stirred reactor are also shown. All
data have been corrected to similar conditions to permit comparison. It
appears that the practical limit that can be achieved as well as the
ability of the combustor to achieve that limit tend to favor lean conditions
(0 < 1.0). -
152
-------
FIG. 9-5
NO RATE PARAMETER AS A FUNCTION OF TEMPERATURE
1000
5-05-223-3
153
-------
FIG.
EQUILIBRIUM COMBUSTION PRODUCTS LOW BTU GAS
144BTU/SCF
1.10°
1.10-2
1.10-3
C/3
I
'
I
Q
1.10
_4
uj 1.1Q-5
1.10-
1.10
/
-8
1.10
1.10-9
N2
H2O 0.010
COS 0.007
144 BTU GAS
TF = 696K
TA= 696 K
F/ASTOIC=.73
FUEL COMPOSITION,
MOLE FRACTIONS
02
0.0 0.4000E+00 0.8000E+00 0.1200E+01 0.1600E+01 0.2000E+01
EQUIVALENCE RATIO
154
-------
FIG. 9-7
EQUILIBRIUM NO CONCENTRATION MOLTEN SALT GASIFIER FUEL
1Q-3
k 1Q-4
I
I
I
1Q-S
10-6
_l_
1.4
0.4
0.6
0.8
1.0
1.2
1.6
78-02-157-5
155
-------
FIG.9-8
ADIABATIC FLAME TEMPERATURE MOLTEN SALT GASIFIER FUEL
2400
2200 -
2000 -
I
LU
EC
D
DC
LU
1800 -
IfiOO
1400 -
1200 -
1000
78-02'
156
-------
Ln
:.
-
__
-
-
COMPARISON OF CONVENTIONAL COMBUSTOR NOX—EQUIVALENCE RATIO
RELATIONSHIP WITH THAT OF A WELL-STIRRED REACTOR
120
100
80
60
X
y
i
-
_
z
y
C
.
g
LU .-
— 40
X
o
20
D
ENGINE A CORRECTED TO
ENGINE B STD. DAY TAKEOFF
ENGINE B STD. DAY TAKEOFF
METHANE/AIR WELL STIRRED REACTOR
CORRECTED TO ENGINE B STD. DAY
TAKEOFF CONDITIONS
'RESENT GENERATION
OF COMBUSTORS
PRACTICAL LIMIT
(WELL STIRRED REACTOR!
0.4
0.6
0.8
1.0 1.2 1.4 1.6
"PRIMARY" ZONE EQUIVALENCE RATIO
1.8
2.0
2.2
-.
CD
CD
-------
Comparing the engine results with those of the well-stirred reactor
reveal the very important effects on NOX formation of fuel preparation
and mixing. With gaseous fuel, the well-stirred reactor probably represents
the best level of mixing that can be achieved in practice without premixing
the fuel and air prior to combustion. For the benefits of mixing to be
worhwhile, the conclusion is inescapable: if either a rich or a lean
combustion alone is selected to reduce NOX, considerably better mixing
will be required than is presently obtainable with conventional liquid-fueled
combustors.
The reduction in NOX that can be expected from reducing residence
time in the flame zone for lean combustion systems is shown in Figure 9-10.
The data shown represents relatively unsophisticated "early quench" methods
where large amounts of relatively cool dilution air are introduced very
early, at or before the end of the primary zone. This is done to freeze
chemical reactions. As can be seen, to achieve significant reductions in
NOX with this method alone, very large reductions in residence time are
required. In general, the required reductions cannot be achieved without
disturbing combustor performance.
Mixing on a molecular level is necessary for the combustion of a
mixture to take place at a flame temperature equivalent to the bulk fuel-air
ratio. Typical characteristic times, shown in Table 9-7 for combustion of
natural gas in air in the FT4 combustor, indicate that mixing time is the
largest characteristic time of those for the processes involved. Therefore,
if the full benefit of low flame temperatures were to be realized by
burning off-stoichiometric, premixing of the reactants prior to introductio°
into the combustor would be necessary. If this were not done, combustion
would be controlled by mixing and diffusion flames would exist with their
associated high temperatures.
When good premixing is achieved, large reductions in exhaust NOX can
be obtained. Figure 9-11 shows the arrangement of a combustor rig in which
premixing to an equivalence ratio of 0.75 was achieved external to the
combustor in a high-pressure drop mixing chamber. The fuel was natural
gas and the test pressure was 1.3 atmospheres. The residence time after
mixing was limited to 8 msec to avoid autoignition. The fuel and air
mixture was introduced into a modified FT4 combustor through a multitude of
individual tubes. Residence time in the primary zone was estimated at 3
msec.
The premixed results are compared in Figure 9-12 with the results from
a conventional FT4 combustor when burning No. 2 liquid fuel at the condi-
tions for the rig tests. The results of an early quench to about the same
bulk equivalence ratio are also shown. The low NOX benefits of premixing
the fuel and air prior to combustion are clearly revealed. The limited
mixing available in the conventional combustor is demonstrated by the early
quench results which, although showing a significant reduction in NOX, do
not show anything close to the potential of the premix rig.
158
-------
FIG. 9-10
REDUCTION IN NOV EMISSION BY REDUCTION IN RESIDENCE TIME-INFORMATION
A
BASED ON EXPERIMENTAL DATA
:
i
s
It
I
•'
~2Q 40 60
ESTIMATED PERCENT REDUCTION IN PRIMARY ZONE RESIDENCE TIME
100
78-02-105-3
159
-------
TABLE 9-7
CHARACTERISTIC TIMES FOR A NATURAL GAS COMBUSTOR
PRIMARY
EQUIVALENCE
RATIO
0.7
1.0
1.2
VAPORIZATION
TIME, ms
0
0
0
TURBULENT
MIXING
TIME, ms
10
10
10
COMBUSTION
TIME, ms
0.20
0.13
0.2
NOX FORMATION TIME
..... TO 300 PpmV, ms
19
0.7
106
RESIDENCE
TIME, ms
5
5
5
-------
FT4 PREM/X RIG WITH EXTERNAL MIXING
NAT. GAS/AIR INJECTION
oo
I
o
10
o
01
REMOTE CONTROL
INLET VALVE
35% OF TOTAL AIR AT 725" F
FLOW METERING
ORIFICES
MIXING CHAMBERS
Jj
P
(O
I
-------
FIG. 9
|2
COMPARISON OF RAPID QUENCH AND PREMIX NOV WITH BASELINE
140
120
o
•.
CNI
O
X
O
100
80
60
40
20
NO. 2HH.
BASELINE
0
0.012
EARLY QUENCH
7
CH4~AIR PREMIX
J
J L
0.013 0.014
0.015 0.016 0.017
FUEL/AIR RATIO
0.018
162
-------
The effects of both residence time and equivalence ratio are shown in
Figure 9-13 based on the Zeldovich mechanism. As stated earlier, it is a
'lose approximation only near * = 1.0 (0.8 to 1.2). However, for preliminary
r*si§n purposes it is considered to be adequate. While it would be feasible
;° dgsign a conventional diffusion flame burner, studies at United Technologies
^Reference 9-25) have shown that NOX control would severely constrain
*si8n parameters and compromise other desirable combustion performance
characteristics.
P
reillix Concept Design Considerations—
If the gaseous fuel can be thoroughly mixed with the combustion air on
* molecular level prior to combustion, the flame temperature will correspond
0 that appropriate to the mixture bulk equivalence ratio. In addition,
Slnce the mixing is achieved external to the combustor, mixture residence
lme in the flame zone can be reduced. These two factors enable such a
PJ-emixed system to take full advantage of the emissions reduction capability
Shown in Figure 9-13. This is done by operating the primary zone lean with
9 mean equivalence ratio about 0.8, and reducing the residence time in the'
"tery zone to a minimum.
A preliminary design for a low-Btu gas combustor using this concept
!as Prepared as part of the Department of Energy-sponsored High Temperature
^rbine Technology Program (HTTTP) conducted by the Power Systems Division
°f united Technologies and UTRC (Reference 9-26). this preliminary design
*as based on the results of the joint P&WA-Texaco-Montebello test program
°r the combustion of gasified residual fuel oil in an FT4 combustor.
t The concept is illustrated in Figure 9-14. Gaseous fuel is supplied
0 a manifold formed in the head of the combustor. Passing through and
dealed in the manifold is an array of premix tubes about one inch in
d*atneter which are open at their forward end to admit air entering from the
lffuser section. Fuel from the manifold is admitted to individual tubes
n this array through a number of metering ports arranged at the forward
Dnd °f each tube in order to maximize the mixing length available. In the
|;ren>ix tube, the entering fuel jets mix with the air flowing through the
tubes. This mixing is enhanced by turbulence grids attached to the premix
fUbe inlets. The pressure drop across these grids also serves to inhibit
tjel from flowing forwards out of the tube inlets into the air-casings in
e event of a disturbance in the smooth supply of air to the premix
The premixed fuel and air are discharged from the premix tubes into
Primary zone where combustion takes place. The downstream portion of
dome is formed into an air manifold for cooling purposes. Cooling air
- — *»*^ i O J- U i. IIIC \~M. J.LJLL.U ULl MI J- *, «»»M. m.^-m- -~ — ___ *~, I 1 (_»
^admitted into the air manifold through ports on its sides. The premix
"-lib • . - - - - . .
manifold is discharged as a concentric annular jet from small annular
-•»*. t-^Cxl .L LI L. U L- LI C- CI.LJ- UK* iti.x.***.w •»•••——,_o--j---- t
'es pass completely through the cooling air manifold. Cooling air from
1 manifold is discharged as a concentric annular jet from small annular
's in the faceplate around the end of each premix tube. This air curtain
formed also serves as a flashback inhibitor, as will be described below.
163
-------
NOX ESTIMATION FOR PREMIXED COMBUSTION BASED ON EXTENDED
ZELDOVICH MECHANISM
FIG.9-1
05
CO
o
O
z
Ml
1.0
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.10
0.08
0.06
0.05
0.04
0.03
0.02
0.01
0.008
0.006
0.005
0.004
0.003
0.002
0.001
PINLET = '8 ATM
TFUEL = 1460°R
T = RESIDENCE TIME IN
MILLISECONDS
136 BTU/SCF FUEL
COMPONENT
H20
CO?.
CO
CH4
H2
COS
H2S
I
167
117
83
50
33
I/
10
< i
o
r
•
Q
Q_
-
O
0.2 0.4 0.6 0.8
EQUIVALENCE RATIO
1.0
1.2
1.4
164
-------
SCHEMA TIC OF PREMIX CONCEPT APPLIED TO A COMBUSTOR DOME
Ul
COOLING AIR INLET
I.D.
FUEL METERING
PORTS
GRID OVER
EACH INLET
AIR
VIEW
> A
COOLING AIR INLETS
O.D.
GAP
FOR COOLING
FUEL '
AND
AIR MIXTURE
FUEL
VIEW A
VIEW OF FACE-PLATE
Tj
P
co
-------
Further cooling air for the faceplate itself is discharged into the primary
zone from the manifold through many "transpiration" cooling holes in the
faceplate.
Figure 9-15 shows a diagrammatic representation of the flame stabiliza-
tion in the low velocity regions of the jets of the premix concept. Again,
this is a simplified diagram and does not attempt to show the additional
flameholding which takes place in the recirculation zones on the faceplate
between individual jets. The path length for fuel consumption is indicated.
It will depend upon mixing tube diameter and rate of spread of the jet.
Autoignition—All premixed combustion devices are susceptible to a
degree of risk from the dual hazards of autoignition and flashback.
Autoignition, as strictly defined, is the temperature at which a fuel and
oxidant mixture will spontaneously ignite when kept for an infinite residence
time. When applied to flowing systems, it occurs when the residence time
exceeds the ignition delay of the mixture. Flashback occurs when the flow
velocity of the arriving combustible fuel and air mixture at some point in
a flowing system becomes smaller than the burning velocity, and an external
combustion wave is then able to propagate against the mixture stream into
the supply duct. Both hazards, either occurring separately or together,
can result in destructive burning in the supply duct, which might lead to
secondary failures in the combustor and downstream components.
Using the collected data shown in Figure 9-16, the possibility of
autoignition in the HTTTP-type design was evaluated. Provided there is not
significant heat transfer to the mixture in the premix tubes, it was
concluded that autoignition should not be a problem. Such heat transfer is
not likely with careful design. However, Figure 9-16 shows the definite
limitations imposed on heat transfer and fuel temperature by autoignition.
Flashback—Flashback is usually encountered in boundary layer and wake
regions of a flow where local low velocities exist. It is usual to relate
flashback to the boundary layer through a critical boundary velocity
gradient. For turbulent pipe flow, the velocity gradient is given
(Reference 9-27) by:
dU = 0.023 p°-8 U1-8 (9~7
dy (2R)°-2 y°-8
where:
R = tube radius
U - mean velocity of mixture flow through tube
166
-------
FIG.9-15
DIAGRAMMATIC REPRESENTATION OF PREMIX COMBUSTOR
DOME
FACE
167
-------
FIG. 9'
II
AUTOIGNITION CHARACTERISTICS-RESIDENCE TIME IN THE PRESSURE TUBE DUE
TO LOCAL FLOW RECIRCULATIONS IS NOT LIKELY TO CAUSE A PROBLEM
1000
100
•I
! !!
tj,
I
I
1 10
I
Q
1.0
0.1
H2~AIR MIXTURES:
= 1.0
&
3
§
fl
SYM
O
A
O
D
SOURCE
AFAP
AFAPL
VEZIROGU
JOST Si
CROFT
PAT MS
14.21
16.51
1.0
1.0
CONDITION
SHOCK TUBE
WITH FLOW
SHOCK
TUBE
WITH DOWN
STREAM
FLAME
DESIGN SAFETY LINE
FOR HYDROCARBONS SPRAYS
SYMBOL
INITIAL DESIGN POINT
ADDITIONAL DATA
FOR HYDROCARBONS
SYM
+
^^
^3
S?
1Z&
SOURCE
NASA
SPADACCINI
MESTRE &
DUCOURNEAU
TABACK
MULLINS
0
0.57
0.08
3.0 5.9
0.070.11
0.075
0.834
PATMS
5.5
1.0
5,11
17.28
1.0
FUEL
JET A
JP4
KEROSENE
JP4
KEROSENE
_ — •
PRE-
VAPORlZe
60/jSPRA
COARSE
96pSPRA
-^
i
i
1
1
1
0.0002
0.0004
0.0006
0.0008
0.0010
0.0012
0.0014
168
-------
value of dU/dy, at which flashback takes place is termed gcr£t and is
Wh' t6(* a8ainst mixture bulk equivalence ratio to form a closed loop within
te flashback takes place. Since burning velocity is a function of
a Perature and pressure, a series of such loops exists for a given fuel
cording to the test conditions.
Figure 9-17 presents estimated flashback loops for the preliminary
burner and for the premix Montebello testing (Reference 9-28). Also
n is the operating range of critical velocity gradient for the Montebello
8' For convenience they are shown at a single bulk value of hydrogen
tne fuel, although in reality the bulk value varied. The operating
8e is to the left of and outside the flashback loops.
g. The Montebello testing could have been subject to incipient flashback
v 1°e the operating range overlaps the estimated flashback loops in critical
e °Clty gradient. Poor mixing in the premix tubes could result in hydrogen
Ursions from the bulk values across into the loop areas. Visual inspection
^ the Montebello test hardware suggests that such incipient flashback may
i e taken place and provides a degree of confidence in the validity of the
re°Ps, Also, the method of estimating flashback loops apears to be a
s°nable one for preliminary design purposes.
shown in Figure 9-17 is the operating point of the preliminary
. burner. As indicated, it falls well outside the area of concern and
Co 6S a reasonable degree of confidence in the feasibility of the premix
n
tan -Staging, — Premixed combustion systems have inherent limitations on the
gn 8e °f fuel-air ratios over which they operate. If the premixing is
and near coraPlete, the mixture will burn at the bulk fuel-air ratio
tjj n°t over a range of fuel-air ratios as with other systems. This means
H- combustion is not possible outside the bulk flammability limits of the
xt
combustor is also different from most gas-turbine combustors
H K- .Inanner in which the flame is stabilized. Most conventional combustors
Op xlize the flame front aerodynamically by means of swirling air and
itit S6(* a*~r Jets strategically placed to create rotating toroidal vortices
th ° w^icn the separate fuel supply is introduced. In the present case,
as Wetl-mixed fuel and total combustion air are introduced in the combustor
a large number of individual, unswirled jets containing all the reactants.
Hg To hold the flame stable in such a system, it is necessary to use a
fj^ anical f landholder which creates the necessary aerodynamic blockage.
C0 , re
-------
FIG. 9-
ESTIMATED FLASHBACK LOOP-ESTIMATED FLASHBACK CHARACTERISTICS
FOR THE PREMIX TUBE COMBUSTOR AND MONTEBELLO TEST ARE SHOWN
10,000,000
1,000,000
111
n
CD
100,000
10,000
<
i
0
I
1000
100
HTTTP DESIGN POINT
MONTBELLO TESTS:
1 ATMOSSPHERE
FUEL TEMP. 80°F
20 40 60
%H2 BY VOLUME
80
100
170
-------
j The type of dome design being considered provides aerodynamic blockage
r lame stability through the faceplate material between individual
s "llx tubes. Thus, the dome acts rather like a perforated plate flameholder,
on aS that described "by Jamieson (Reference 9-29). The initial reactions
2o Such flameholders take place in the multitude of little recirculation
s created between the individual mixture jets, and only later spread
°Ss the jets. These zones were not shown in Figure 9-15 for simplicity.
On stability characteristic must be defined for such flameholder when
Co rat^n8 with low-Btu gas. The stability characteristic consists of a
relation of aerodynamic and geometric parameters of the flameholder,
w- ^ed against the mixture equivalence ratio so as to form a closed loop
"in which stable combustion is possible. United Technologies Corporation
demonstrated good success with premixed hydrocarbon/air flameholders
Us ^ fc^e Mar1uardt stability parameter (Reference 9-30), and it has been
ed again here. It is defined as:
StabiUty Parameter = JL . JL . Jk i°°° ^x lO'3 (9-8)
d d0 P Tn
re> V = velocity of flow at flameholder
jde_ = effective dimension ratio, (= 1.6 for disks)
d
d = baffle dimension
PO = reference pressure (= 1 atmos.)
P = static pressure in flow
TO = mixture total temperature
. stable combustor operation it is necessary to always operate
n th-e loops shown in Figure 9-18. These loops are much narrower than
those for more conventional gas turbine combustors which do not use
ixing. The loops for premixed systems are usually not wide enough to
ta«- • y most operating requirements. The design point mean equivalence
re X? was selected as 0.78 on the basis of satisfying NOX emissions
^Ulrements. This point is shown on Figure 9-18 inside the loop.
to Once this point was decided, a totally fixed geometry system is forced
Cf Work along a smooth, continuous operating line which will eventually
Ca ? t'le lean limit of the stability loop at some part-power condition,
ne Sln8 a lean blowout of the combustor. To prevent lean blowout, it is
fu ^Ssary to use a variable geometry system. By making this part of the
*lo System> ifc is possible to have the moving parts external to the engine
wPath. Variability of the fuel system is to be achieved by fueling the
.lx dome in discrete zones such that the functioning zone equivalence
10 is always kept inside the stability loop.
171
-------
FIG. 9"'
CALCULATED STABILITY LOOP FOR PREMIXED CONCEPT AS APPLIED TO AN
ANNULAR COMBUSTOR WITH 2600°F CET
100
f!
c
D
',
Q
a
o
i i
to
CO
1
o
Oi
Q- I
'I-S
10
1.0
0.1
0.01
P&WA PERFORATED
PLATES-PROPANE
LEAN
LIMITS
\
UNSTABLE REGION
120 BTU/SCF
(14% H2)
STABLE
REGIONS \
DATA POINT \
FROM PREVIOUS '
MONTEBELLO FT-4A
^-*-START
ZONE 1 I
HTTTP
SCHEME 1
PATH
BO BTU/SCF
(10'
DESIGN POINT
RICH LIMITS
0.2 0.6 1.0 1.4 1.8
0 - EQUIVALENCE RATIO
2.2
2.6
3.0
172
-------
"h" , F*"*>ure 9-18, by way of illustration, displays the operating line for
ar . a three-zone premix was done. At the design point all three zones
in operation. As power is reduced by reducing the fuel flow, the lean
t^ lt is approached until Zone III is shut down; and the total fuel flow at
r Point goes through Zones I and II only. This increases the equivalence
lo of these zones and moves the operating characteristic away from the
g n limit. Further reduction in power causes the lean limit to be approached
Sn ln Until Zone II is shut down. This occurs just below synchronous idle
le Start-up of the gas turbine is accomplished on Zone I at an equiva-
Ce ratio of unity for ease of starting. Zoning is achieved through
°f internal baffling in the fuel manifold surrounding the premix
s in the dome of the combustor.
re While low-temperature cleanup systems achieve virtually complete
hi K^* °^ nitrogen compounds, possibly the most serious shortcoming of
jj-^ "temperature cleanup systems is their inability to remove fuel-bound
t, r°gen. in conventional combustors with stoichiometric air or greater
ha U st°*chioraetric air, almost all ammonia (at low, < 1% concentrations)
been shown to be converted to NO (Reference 9-11). Unless reliable and
re0ri0mical means can be developed to modify fuel composition in the gasifier.
co £Ve ammonia at high temperature or decompose it to nitrogen in the
ustor, the viability of high-temperature cleanup is questionable.
-------
FIG. 9
19
I
LJ
' :
'
•!
to
Ml
LJJ
a
LLJ
(3
O
Q
I
3
( !
RESIDUAL NITROGEN SPECIES ASA FUNCTION OF EQUIVALENCE
RATIO AND RESIDENCE TIME
NH-
^4000 PPM IN FUEL
TFUEL= 18 1088K
TAIR
PRESSURE = 10 ATM
RESIDUAL NITROGEN SPECIES =
(NO, NH3 AND HCN)
1.0
0.8
0.6
0.4
0.2
\
1\
t = 1 IDS
t = 50 ms
/j
0.5
1.0
EQUIVALENCE RATIO,
1.5
2.0
174
-------
OPTIMIZATION OF THE INITIAL DESIGN CONCEPT
TA|R = 810K
AIR
SECONDARY AIR TO ACHIEVE
0 OVERALL OF 0.45
PRIMARY0 = 1.33
T{= 1088K
FUEL
\v •
i^sr
TI NH3 ' *
PRIMARY ZONE
RESIDENCE
TIME = 500msec
T= 2190K
1 1
NOm = 43 ppm
MIXING
ZONE
1
~~1
[y TT1=1813K
0 = 0.45
A
[) NO = 67 ppm
1
SECONDARY ZONE
RESIDENCE
TIME = 100msec
INSTANTANEOUS
NOTE: N0m AND HCNm INCLUDE
SECONDARY DILUTION EFFECT
O
CO
10
O
-------
FIG.**
1000
800
600
400
RICH BURNER STAGING CHARACTERISTICS
RICH BURN/QUICK QUENCH CANCEPT
10% PRIMARY AIR
o
Q.
Q
• :
O
z
Q
in
U
in
'i
DC
O
o
200
100
80
60
40
20
10
.20% PRIMARY AIR
i
!\
I
50PSIA
600°F
0.5% FUEL
NITROGEN
0.1 0.2 0.3
EQUIVALENCE RATIO, OVERALL
0.4
176
-------
with' t^e °kservec' reduction to 50 ppm is significant. Also, for the case
p . ®% primary air, CO emission levels are quite low. Apparently it is
rei l. e to move the position of minimum NO formation while maintaining a
atively constant CO characteristic.
PARTICULATES
en - e effluent from each gasifier will contain varying amounts of
m ,rained particulate matter. To date, no quantitative assessment has been
g ? °f particulate loadings in gases produced by the specified coal
ch lcati°n processes. The Koppers-Totzek gasifier effluent has been
p.racterized as having an entrained particulate concentration of 11.57
na Scf (Reference 9-33). Westinghouse has prepared a quantitative approxi-
bedl°n °f raw gas particulate loading and size distribution for fluidized
a Sasif ication. The projection is based on fluidized bed combustion
9.., ® theoretical model developed by Kunii and Levenspiel (References 9-33,
th ^e ^est inghouse projections and the particulate loads estimated for
^ 8asif iers studied here are shown in Table 9-8. The molten salt bed
t^. °§8 gasifier removes particulates before they exit the gasifier. For
• s reason a lower particulate loading, nearly all entrained sodium salts
8 ^pecte
(DP summary of removal system performance developed by Westinghouse
toeference 9-34) is shown in Figure 9-22. Since the gas turbine is believed
ty 6 tolerant to particles of less than 2 micron diameter, the venturi
th Scr"bber appears ideally suited for this duty. It is also apparent
r , those concepts amenable to high-temperature operation have significantly
Uced removal capabilities in the size range of interest.
fil F°r the ni8h-temperature systems studies here, use of a granular bed
est-er was assumed. Particulate removal down to .01 to .02 grain/SCF is
^ imated (Reference 9-35). Other candidates include a woven metal cloth
be-Ufactured by the Brunswick Corporation and the panel bed sand filter
devel°Ped at City College, New York (References 9-36, 9-37, and
Performance of the panel bed filter is reported to be as high as
with no particles larger than 5 micron penetrating the filter. For
pa purposes of this study, it is important that a representative cost for
9^3 lculate removal be used. Data obtained from References 9-35 , 9-38 and
Of aH show installed costs for the various concepts to be in the range
to $100 per actual CFM for units of the size used here.
177
-------
Process Configuration
Westinghouse (Estimate
for Fluidized-Bed
Gasification)
BCR-Air (Entrained Bed)
BCR-Oxygen (Entrained Bed)
IGT-Air (Fluidized Bed)
Kellogg-Air (Molten Bed)
TABLE 9-8
PARTICULATE LOADING IN GASIFIER
PRODUCT GASES
Projected Dust Load
10 to 30 gr/SCF
15 to U5 gr/SCF
15 to U5 gr/SCF
5 to 25 gr/SCF
0.1 to 1 gr/SCF
Particle
Size Distribution
10 to 25$ < 10 y
5 to 15? < 5 y
15 to 35$ < 10 y
10 to 20% < 5 M
15 to 35$ < 10 y
10 to 20% < 5 y
10 to 25$ < 10 y
5 to 15$ < 5 y
10 to 30$ < 10 y
5 to 20$ < 5 y
178
-------
EXTRAPOLATED FRACTIONAL EFFICIENCY OF
PARTICULATE REMOVAL DEVICE
FIG. 9-22
0.01
o
111
G
G
u
ui
O
u
UJ
...
f '
- 10.0
95.0
0.01
- 99.9
0.1
PARTICLE DIAMETER, MICRONS
78-O2-157-6
179
-------
REFERENCES
9-1 Federal Register, Monday, October 3, 1977, Part III. EPA-Stationary
Gas Turbines - Standards of Performance for New Stationary Sources.
9-2 Hagard, H. R. and F. D. Buckley. Experimental Combustion of Pulverized
Coal at Atmospheric and Elevated Pressure. Trans. ASME, Vol. 70, p. 729
(19^8).
9-3 Smith, J., et al. Bureau of Mines Progress in Developing Open and Closed-
Cycle Coal Burning Gas Turbine Power Plants. J. Eng. Power, Vol. 38, No.
k, October 1966.
9-^ McGee, J., Coal-Fired Gas Turbines, Mech. Eng., Vol. 8l, May 1959.
9-5 Robson, F. L., et al. Analysis of Jet Engine Test Cell Pollution Abater^0
Methods. Technical Report No. USAF AFWL-TR-71-18, May 1973.
9-6 Hoy, H. R. and H. G. Roberts. Fluidized Combustion of Coal at High Pres-
sure AIChE Symposium Series, Air Pollution and Its Control, 68, No. 12,
1972.
9-7 The Coal Burning Gas Turbine Project, Report of Interdepartmental
Committee, Department of Minerals and Energy, Department of Supply.
(Commonwealth of Australia), (1973).
9-8 Dust Erosion Parameters for Gas Turbines. Petro/Chemical Engineering.
December 1962.
9-9 Talbert, W. M., et al. High Temperature Turbine Technology Program
Plant Study and Design. Pullman Kellogg Report No. RED-77-1333, Feb.
9-10 Field, J. H., et al. Pilot Plant Studies of the Hot-Carbonate Process
Removing Carbon Dioxide and Hydrogen Sulfide. Bureau of Mines Bulletin ''
1962.
9-11 Wendt, J. I. L. and C. V. Sternling. Effect of Ammonia in Gaseous 0
Nitrogen Oxide Emissions. Journal of the Air Pollution Control
Vol. 2U, No. 11, November 197^, pp. 1055-1058.
180
-------
REFERENCES (Continued)
Crouch, W. B. and R. D. Klapatch. Solids Gasification for Gas Turbine
Fuels, 100 and 300 Btu Gas. Intersociety Energy Conversion Engineering
Conference. Paper 70903^, September 1976.
C-iramonti, A. J. Advanced Power Cycles for Connecticut Electric Utility
Stations. UTRC Report L-971091-2, January 1972.
Roberts, R., L. D. Aceto, R. Kollrach, D. P. Teixeria, and J. M. Bonnell:
Analytical Model for Nitric Oxide Formation in a Gas Turbine Combustor.
Journal, Vol. 10, No. 6, June 1972.
Mador, R. J. and R. Roberts: A Pollutant Emission Prediction Model for
Gas Turbine Combustor. Paper presented at 10th Annual AIAA/SAE Propulsion
Conference, San Diego, October 21-23,
, S. A. and R. Roberts. Low-Power Turbo Propulsion Combustor Exhaust
Missions. Vol. 1, Theoretical Formulation Design and Assessment. AFAPL-
TR-73-36, June 1973.
, S. A. and R. Roberts: Low-Power Turbo Propulsion Combustor Exhaust
issions, Vol. 2, Demonstration and Total Emission Analysis and Prediction.
AFAPL-TR-73-36, April 1971*.
M°iser, S. A. and R. Roberts. Low-Power Turbo Propulsion Combustor Exhaust
, Vol. 3, Analysis. USAF AFAPL-TR-73-36, July 1971*.
, S. A., R. Roberts and R. Henderson: Development and Verification
°f an Analytical Model for Predicting Emissions from Gas Turbine Engine
Corabustors During Low-Power Operation, NATO AGARD-CP-125, April 1973.
Marteney, P. J.: Analytical Study of the Kinetics of Formation of Nitrogen
°xide in Hydrocarbon-Air Combustion. Combustion Science and Technology,
V°l. 1, pp. Ii6l-l*69, 1970.
zeldovich, J.: The Oxidation of Nitrogen in Combustion and Explosives.
Physiochmica, U.S.S.R., Vol. 21,
'^a
penimore, C. P.: Formation of Nitrix Oxide in Premixed Hydrocarbon Flames.
13th International Symposium on Combustion, Salt Lake City, 1970.
**3
shaw, H.: Fuel Modification for Abatement of Aircraft Turbine Engine Oxides
°f Nitrogen Emissions. NTIS Report No. AD752581, October 1972.
^
j, A. A.: Kinetics of NO and CO in Lean Premixed Hydrocarbon-Air
Blames. Combustion Science and Technology, Vol. U, 1971, pp. 59-6U.
181
-------
REFERENCES (Continued)
9-25 Sarli, V. J.: Variation of NO Formation with Time and Humidity in the Com-
bustion of JP-Fuels. United Aircraft Report No. UAR-L52, 1972.
9-26 Carlson, N. G., et al. Development of High-Temperature Subsystem Technology
to a Technology Readiness State, Phase 1 - Final Report, ERDA FE-2292-19
April 1977.
9-27 Lewis, B. and G. VonElbe: Combustion, Flames and Explosion of Gases, Academic
Press, 2nd. Edit., 196l, p. 250.
9-28 Madden, T. J., R. H. James and H. R. Schwartz: Evaluation of Gasified
Residual Fuel Oil and Coal in a Low Pressure Single Segment Gas Turbine
Burner Rig - Phase III Final Report, Pratt & Whitney Aircraft Report No.
PWA-52lt9, March 1975.
9-29 Jamieson, J. B.: Premixed Primary Zone Studies Using a Multiple-Port Baffle,
Proc. Cranfield Intl. Symposium Series, Vol. II - Combustion & Heat Transfer
in Gas Turbine Systems, Edit. E. R. Norster, Pergamon Press, 1971, pp. 123-lW.
9-30 The Marquardt Company, Technical Report AFAPL-TR-70-Sl, January 1971.
9-31 Martin, G. B. : NOX. Consideration in Alternate Fuel Combustion. EPA-600/2-
76-11*9, (NTIS No. PB257-182), Symposium Proceedings: Environmental Aspects of P-
Conversion Technology II, December 1975, Hollywood, Florida.
9-32 Mosier, S. A. and R. M. Pierce: Advanced Combustion System for Stationary-
Gas Turbine Engines. Second Symposium on Stationary Source Combustion, Nev
Orleans, August 1977-
9-33 Advanced Coal Gasification System for Electric Power Generation, Annual
Technical Report for August 9, 1972-June 30, 1973. Submitted to Office of
Coal Research by Westinghouse Electric Corporating, Lester, PA, 1973.
9-3^ Clean Power Generation from Coal. Final Report by Westinghouse Electric Corp.
Research and Development Center, April 197U, PB 2'31il88.
9-35 Private Communication, Combustion Power Co., May, 1977-
9-36 Squires, A. M. and R. Pfeffer: Panel Bed Filters for Simultaneous Removal of
Fly Ash and S02. Journal of Air Pollution Control Assn. (20), No. 8,
August 1970.
9-37 Lee, K. C., et al.: The Panel Bed Filter. EPRI AF-560, May 1977-
9-38 Jones, C. H. and J. M. Donohue: Comparative Evaluation of High and Low
Temperature Gas Cleaning For Coal Gasification - Combined Cycle Power Systems,
EPRI AF-U16, April 1977.
9-39 Beecher, D. T., et al.: Energy Conversion Alternatives Study Westinghouse
Phase II, Final Report. NASA CR-13^9^2, November 1976.
182
-------
SECTION 10
OTHER ENVIRONMENTAL INTRUSIONS
INTRODUCTION
j_ he previous section has dealt with the major air emissions from the
8rated power plant. With the exception of thermal NOX, all the poten-
Pollutants arose because of the nature of the coal feedstock. From the
ejc- th-at the coal enters the plant boundary to the time that solid wastes
Ex ' there is the potential to cause some type of environmental intrusion.
fe *~pt for some chemicals introduced for water treatment, the original coal
tr is the source for all the major air, water, solid waste and minor and
element
emissions.
St *he organic portion of coal is largely composed of polycyclic aromatic
t£ Vctures and, in addition to carbon and hydrogen, contains significant quan-
aiujles °f organic oxygen, nitrogen, and sulfur, residual biological compounds,
ca ,°tller organic compounds. Functional groupings include methoxyl, hydroxyl,
H °n^1-' and carboxyl. The form into which organic sulfur and nitrogen com-
su with other elements or compounds in the coal is not known, but it is pre-
Hg e<* to be largely or entirely in heterocyclic aromatic rings. Coal mineral
Ueer contains trace to appreciable quantities of practically all the ele-
ts in the periodic table.
fr ?e Proximate and trace-elements analysis of coals may vary considerably
an * mine to mine, or even from seam to seam (Reference 10-1). The proximate
An ysfs ^or coal 1ists only moisture, volatile matter, fixed carbon, and ash.
an j fcimate analysis is only slightly more definitive; a complete ultimate
for tvPical Illinois No. 6 coal is listed in Table 10-1 from Refer-
e }°~2. Typical modes of occurrence of trace and minor elements in coal
given in Table 10-2 (Reference 10-3).
j A general procedure that considers concentration and toxicity of trace
Q ments is described for choosing those most likely to be of environmental
adnf rn' Usin? that Procedure> tne trace elements arsenic, beryllium, boron,
mium, chromium, lead, mercury, nickel and vanadium have been selected for'
183
-------
TABLE 10-1
TYPICAL COMPOSITION OF ILLINOIS NO. 6 COAL
Proximate Analysis, As Received
Moisture
Volatile Matter
Ash
Fixed Carbon
Total
Ultimate Analysis, Dry Basis
Carbon
Hydrogen
Sulfur
Nitrogen
Oxygen (by difference)
Ash
Minor Eleracnta
Wt-%
100.0
70.1
4.88
3.74
1.07
10.11
10.10
Silicon
Iron
Calc ium
Chlorine
Potassium
Sodium
Trace Elements
Titanium
Magnesium
Boron
Fluorine
Zinc
Manganese
Strontium
Zirconium
Lithium
Barium
Arsenic
Copper
Vanadium
Chromium
Nickel
Selenium
Lead
Tellurium
Molybdenum
Germanium
Cobalt
Tin
Antimony
Bismuth
Beryllium
Cadmium
Samarium
Ytterbium
Mercury
Silver
2.0
1.4
0.35
0.23
0.17
0.14
ppm
700
570
200
61
49
48
37
35
33
31
24
19
17
15
1
13
11
8.1
7.0
4.3
3.6
2.0
1.1
1.1
.0
0.89
0.74
0.56
0.12
0.10
184
-------
TYPICAL FORM IN WHICH TRACE AND MINOR ELEMENTS OCCUR IN COAL
00
Elemen t Form
Sb Sulfide
As Oxide, sulfide
Ba Carbonate, sulfate
with Ca
Be ORO*
Bi Sulfide
B ORO, borate
Cd Sulfide
Ca Oxide, carbonate, sulfate
Cl PORO+ sodium chloride
Cr PORO, oxide
Co PORO, sulfide
Cu CuFeS2, sulfide
F CaF2
Ge PORO, carbonate
Fe Carbonate, sulfide, oxide
Pb Sulfide
**
Li SQ
Mg PORO, carbonate, SQ
Mn Carbonate in CaCO,, SQ
Hg PORO, elemental, sulfide
* ORO - Organic occurrence
+ PORO - Partial organic occurrence .
** SQ - Silicates, clay, quartz
Element
Mo
Ni
N
K
Sm
Sc
Se
Si
Ag
Na
Sr
S
Te
Th
Sn
Ti
V
Yb
Zn
Zr
Form
Sulfide
Sulfide
ORO
KC1, carboi
SQ
Oxide
PORO, sulf:
Oxide , SQ
Element, s
PORO, carb
PORO, with
PORO, sulf
Iron tellt
SQ
Carbonate ,
PORO, SQ
PORO
SQ
Sulfide
Oxide, SQ
-------
discussion of their fate in the integrated plant processes. All these ele-
ments except boron and vanadium are included on EPA's list of criteria pol"
lutants. Because of the potentially corrosive activity of boron and vanadium
salts in the hot section of the gas turbine, these elements have also been
selected for further study.
The following presents a general discussion of the sources of pollution
in an integrated plant and describes, wherever possible, methods of control-
ling the air, water or solid waste emissions. The trace elements, in parti-
cular those chosen for examples, are also discussed.
WASTES FROM COAL PREPARATION AND HANDLING
Figure 10-1 is a schematic block diagram of coal handling and preparation
facilities. Run-of-mine (ROM) coal will be received from the mine by trucks.
The amount of coal stored at a commercial-sized, combined-cycle power plant
of 1,000-MW capacity will depend upon the availability of coal from the mine>
the location of the plant, and storage space available at the plant. Coal
will be received on a continuous basis and stored in an active coal pile. ^
is a common practice to stockpile enough coal for a three to fifteen weeks
supply. This coal will be stored in an inactive pile and will be utilized i°
case of unusual circumstances, such as unavailability of coal on a continuous
basis due to bad weather or other uncontrollable conditions.
Size RedjactJLon
Coal size reduction is done in stages. Primary crushers, either roll-
type or rotary screen-type, are employed to reduce the raw coal to a top si&e
of 3 inches. Coal is reduced further with secondary crushers and screening
crushers. Secondary crushers will reduce coal to a top size of from one-and'
one-half to one-and-three-fourths inches, screening crushers will reduce coal
to a top size of from 3/8 to 1 inch. Different types of crushers are used
for coal cleaning; from an environmental point of view, crushers producing
lesser amounts of fines are desirable. The commonly used crushers are
hammer-mill and ring crushers, single-roll crusher, double-roll crusher,
rotary breaker, and pick breaker.
Screening
Screening is used for sizing coal for further processing. The crushed
coal is sized into various fractions: 3" x 1/4", which is further processed
in heavy medium separators; and 1/4" x 0 which is further sized by desliming
screens into two different sizes, 1/4" x 1/2 mm and 1/2 mm x 0. Coal parti"
cles may be screened wet or dry. The various types of screening equipment
used include gravity-bar screens (or grizzlies), revolving screens, shaking
screens, vibrating screens, and sieve bends. Wet screening produces large
quantities of contaminated water whereas dust is generated by dry screening-
186
-------
COAL PREPARATION AND HANDLING
ROM COAL
FIG.110-1
FLOTATION
1/2 mm x 0
VACUUM
FILTER
THICKENER
SETTLING
POND
ROTARY BREAKER
ROCK TO TRUCKS
TOP SIZE 3 IN.
RAW COAL
STORAGE
RAW COAL
SIZING AND
SCREENING
1/4 IN. x 0
3 IN. x 1/4 IN.
SETTLING
TANKS
DESLIMING
SCREENS
1/4 IN. x 1/2 mm
PRIMARY
HYDROCYCLONES
CENTRIFUGES
THERMAL
DRYER/PULVERIZER
STORAGE
SILOS
HEAVY MEDIUM
SEPARATORS
CENTRIFUGE
TO THERMAL DRYER
OR STORAGE SI LOS
REFUSE
SECONDARY
HYDROCYCLONES
TO FINAL
• •
REFUSE BIN
•*• TO GAS PRODUCTION
78-02-138-1
187
-------
Thermal Drying
Moisture content in the feed coal acceptable to different gasifiers will
vary, depending upon the gasifier configuration. During coal sizing and
screening, the water content of coal will increase to a level as high as 10 to
15 percent. Generally, the moisture content of coal is reduced to 2 to 6
percent by drying with hot gases in thermal dryers. The hot gases are usually
the gaseous combustion effluent from a burner or air heater. After being used
for drying the wet coal, this gas is vented to the atmosphere from an elevated
vent. Temperature control is essential in drying to prevent spontaneous com'
bustion when the drying medium contains high levels of oxygen. The types
of thermal dryers available are rotary dryers, screen-type dryers, fluidized"
bed dryers, cascade dryers, and multilouver dryers.
Air Pollution and Control
Atmospheric discharges from coal preparation units appear to be the
largest single source of particulate emissions from the combined-cycle plant
if coal is prepared on site. The major sources of particulate emissions, in~
eluding fugitive dust, are shown in Figure 10-2. Dust is composed of coal
particles, typically 1 to 100 microns in size, similar to the parent coal in
composition. Fugitive dust is generated from coal unloading, crushing,
screening, and conveying operations. More coal dust will be generated for
gasifiers requiring smaller particle sizes. Pneumatic conveying of coal also
will generate more dust and better control will be necessary. Inactive coal
piles do not cause a major dust problem but active coal piles could have
higher than usual dust emissions under high wind conditions. Coal stored in
silos will produce no dust emission.
Coal dryers in the combined-cycle power system will utilize the stack
gas from the heat recovery boiler. This gas is available at a temperature of
about 300F. Major components present in the stack gas are carbon dioxide,
nitrogen, oxygen and water with small amounts of NOX and SC>2 that are
generated during the combustion of the low-Btu gas in the turbine burner.
exhaust gas from the dryer will include, in addition to the components above
coal particulates, water and trace amounts of volatile organics that may be
present in the coal. The exhaust gas leaves the dryer at a temperature of
about 220F. The coal is heated to a temperature of about 200F. Both BCR
and oxygen-blown processes use part of the stack gas for coal drying. For a
1,000-MW plant, a flow rate of about 80,400 moles/hr is required. The U-Gas
gasifier does not require coal drying. In general, vent gases from the
dryer will have a higher moisture content than the stack gas.
Particulate Control Methods
The particulate control methods applicable to different sources of
particulate dust generation also are shown in Figure 10-2. Crushing, convey"
ing, pulverizing, loading and unloading equipment can be operated with enclo"
sures or hoods. Water sprays can be used to prevent dispersion of dust
188
-------
FIG. 10-2
PARTICULATES EMISSIONS FROM COAL PREPARATION AND HANDLING
UNCONTROLLED
EMISSION
CONTROL
METHODS
CONTROLLED
EMISSION
10TON/HR
FROM
8400 TON/DAY
COAL HANDLING
FACILITY
_ 40 TON/HR
ENCLOSURES OR HOODS
DUST COLLECTORS
-VACUUM CLEANING
- LOUVER TYPE COLLECTOR
- CYCLONES
-WET SCRUBBERS
- FABRIC FILTERS
- ELECTROSTATIC PRECIPITATOflS
WATER SPRAYS
MINIMIZE UNCOVERED STORAGE
SPRAYING OF WATER SOLUABLE
ACRYLIC POLYMER
PROPER HANDLING AND DISPOSAL
» CYCLONES _
_• WET SCRUBBERS
LESS THAN
225 LB/HR
(ASSUMING DUST LOADING
OF 100 GRAINS PER ACTUAL
CUBIC FEET AND GAS VOLUME
OF 100,000 ACTUAL CUBIC
FEET)
BAG FILTERS
LESS THAN
30 LB/HR
78-02-138-2
189
-------
particles at loading and unloading points, at transfer and discharge points
from crushing and screening, and from the coal storage area. Outdoor coal
storage piles should be covered with coatings, such as acrylic polymer,
to provide protection from wind and rain.
Application of suitable dust collection devices will also help reduce
the particulate emissions. Vacuum-cleaning systems can effectively minimize
the particulate dispersion around the plant. The louver-type collectors can
operate at 99 percent efficiency when collecting plus 100-mesh dust and are
suitable for collecting heavy dust. Primary cyclones can effectively collect
particles larger than 200 mesh. Secondary cyclones in series with primary
cyclones can collect all the plus 10-micron particles. Wet scrubbers can re""
move submicron particles, but consume more energy. Fabric filters can operat
at a 99 percent efficiency, but their application is limited to temperatures
below 500F. Electrostatic precipitators can remove submicron particles.
The particulate loading rate from thermal dryers, if uncontrolled, can
be as high as 100 to 300 grains per actual cubic foot. Such a high emission
rate can be controlled by operating primary and secondary cyclones in the rflW
gas transfer line. A wet scrubber can also be used in place of secondary
cyclones.
Water Pollution and Control
Two major sources of wastewater from the coal preparation and handling
area are the water generated from coal processing units such as wet screeni°8
and the runoff water from raw coal storage and coal waste deposits. Water
from the processing units contains fine and colloidal particles, dissolved
mineral matter, and salts, all of which were present in coal. Chemical
agents added to this water for solids removal by flotation and flocculation
are also present. Approximately 2600 gpm of wastewater will be generated
from processing 350 TPH (8400 TPD) of raw coal. However, most of it will be
treated and recycled for reuse.
The second source of wastewater, runoff water from coal storage and wast
deposits, is contaminated with high concentrations of sulfate and metal ions-
The amount of runoff water will depend on the rate of precipitation at the
site.
Coal Pile Runoff Water Characteristics—
Coal pile runoff is the drainage from the coal storage area usually
resulting from rain. Runoff may cause pollution if allowed to enter waterway5
or to seep into aquifers. The nature of coal pile runoff depends on the tyPe
of coal used. Generally, there are two types of coal pile runoff. The first
is neutral or slightly alkaline, contains ferrous ion, and originates from
coal containing large amounts of alkaline materials or small amounts of
pyrite. The second kind of runoff is highly acidic, contains large amounts
of dissolved iron and aluminum, and'is produced from pyrite-rich coal.
190
-------
Pyrite is oxidized by atmospheric oxygen and hydrolyzed to form ferrous
Sulfate and sulfuric acid according to the reaction
2FeS2 + 702 + 2H20 - 2FeS04 + 2H2S04. (10-1)
Additional sulfuric acid may be formed if the ferrous ion is further
°xidized to the ferric state. When rain falls on coal piles, the acid is
Washed out and eventually becomes part of the coal pile drainage. At low pH,
*etals such as aluminum, copper, manganese, zinc, and others are also dis-
s°lved, further degrading the water.
fu Coal pile runoff water characteristics can vary widely, depending on
Jhe coal characteristics, weather conditions, and the length of contact
bet*een coal and water. The amount of rainfall that percolates through coal
Stor*ge piles is a function of the area of the pile and the rainfall The
!rea of the coal storage pile for a 1000-MW electric plant may range from 12
1° 60 acres, depending on how high the coal is piled and the number of
°Perating days the plant has in storage.
Characteristics of coal pile runoff at selected steam-electric power
s are given in Table 10-3 (Reference 10-4). The coal types are unknown.
water is treated along with other waste streams from the integrated plant.
ater Handling and Treatment —
„ Water with less than 5 percent solid matter neutral pH, low conductivity,
nd low bicarbonate can be used in coal preparation plants. The waatewater
r°» the coal preparation units can be handled and treated to this quality by
ed water circuit consisting of thickeners, cyclones, filters, and/or
bowl centrifuges. Solids are separated by sedimentation. Flocculation
s the size of particles, thereby aiding the sedimentation process.
f peculation agents are employed for treatment, among which starches
Probably those most commonly used.
, Thickeners are used to remove solids from contaminated water containing
ess than 10 percent solids. The overflow from the thickener may contain
i*ss than 1 percent solids of very fine size that cannot settle. The under-
l°w is 60 percent solids. Cyclones are also used to remove suspended
S°U
-------
TABLE 10-3
COAL-PILE RUNOFF ANALYSIS AT SELECTED PLANTS
(mg/l)
Plant A B C D E P <>
lU.32 36.U1 - „
6,000
28,970 - - 5,800
100 - - 200
1.35
2.77 6.13
21,700 10.25 8.8H
6,837 19,000 - - 861
1,200
15.
1.8
Alkalinity
Total solids
TDS
TSS
Ammonia
Nitrate
Phosphorous
Turbidity-
Acidity
Total hardness
Sulfate
Chloride
Aluminum
Chromium
Copper
Iron
Magnesium
Zinc
Sodium
PH
6
1,330
720
610
' 0
0.3
-
505
-
130
525
3.6
-
0
1.6
0.168
0
1.6
1,260
2.8
0
9,999
7,7^3
22
1.77
1.9
1.2
-
-
1,109
5,231
U8l
-
0.37
-
-
89
2.U3
160
3
- - e
15.7 - - 0.05
- - ,
0.368 i+,700 1.05 0.9 0<0f
12.5
2.7 2.1 6.6 6.6
192
-------
FIG. 10-3
WATER HANDLING FOR COAL PREPARATION
WASTEWATER
(10% SOLIDS)
. HYDROCYCLONES
I
RECYCLE WATER
FOR REUSE
(4% SO LIDS)
40% SOLIDS
THICKENERS
RECYCLE WATER
FOR REUSE
(<1% SOLIDS)
60% SOLIDS
SETTLING
PONDS
_ SOLIDS
TO DISPOSAL
CENTRIFUGE
VACUUM
FILTER
RECYCLE WATER
FOR REUSE
(<0.5% SOLIDS)
RECYCLE WATER
FOR REUSE
(«=0.5%SOLIDS)
SOLIDS
TO DISPOSAL
SOLIDS
TO DISPOSAL
KEY:
ALTERNATE WATER
-HANDLING METHODS
78-02-138-3
193
-------
Wastewater resulting from raw coal piles or coal waste piles can be
treated by neutralization. Acid or alkali is added to neutralize alkaline of
acidic wastewater. Common alkalies used are lime, limestone, or soda ash.
Sulfuric acid is used to neutralize alkaline water.
Solid Waste and Disposal
It is estimated that about 25 percent of the raw coal mined is disposed
of as waste. For a coal conversion plant processing 8400 TPD of prepared coal»
the coal refuse will amount to 2100 TPD. The coal refuse consists of waste
coal, slate, carbonaceous and pyritic shales, and clay asociated with a coal
seam. There are two basic types of coal refuse, coarse and fine. The coarse
refuse is separated mainly by crushing the lump coal to a smaller size; the
fine refuse is mainly separated from the thickeners as underflow and impounded
into the settling ponds. The coarse refuse is normally disposed of in an
embankment. Further discussion of the disposal of these wastes is presented
in the discussion for the integrated plant.
WASTES FROM GASIFICATION, CLEANUP AND GAS UTILIZATION
The major air emissions resulting from the preparation and combustion oi
the fuel gas have been identified in Section 9. Thus, the emphasis here will
be upon potential water borne and solid wastes.
Waste Water Sources
Water requirements for a coal conversion plant are high. Variation of
water requirements results from a combination of process requirements, plus
cooling and wastewater streams subjected to varying degrees of conservation
and reuse. Low-Btu coal gasification requires about 1 to 3 pounds of water
per pound of feed coal. Low-Btu gasification combined with cleanup and a
COGAS power cycle will need an additional 2 to 4 pounds of water per pound 01
feed coal. This water is used in these processes for different objectives.
Steam is required for gasification, water is used for scrubbing the raw gases
from the gasifier, a large amount of steam is generated in the boilers for
use in steam turbines, and circulating cooling water removes excess heat
generated in the process.
Part of the water entering the process comes out as wastewater. A
portion of the steam fed to the gasifier remains unreacted and exits with raw
gas. It is condensed in knock-out drums and combined with the sour water
stream. Raw gas from the gasifier is scrubbed with water to absorb ammonia
and to lower further the raw gas temperature. The scrubbing water also dis-
solves hydrogen sulfide and carbon dioxide, and phenols, cyanides, and organ"
ics that may be present. This constitutes the sour water stream and is the
dirtiest water generated in the process.
Regeneration wastes are produced during resins regeneration of the
demineralizers used to deionize boiler feedwater. Cooling tower blowdown is
another source of wastewater. Raw water treatment wastes and boiler cleaning
194
-------
Wastes are also generated from the power system. These sources are shown in
Flgure 10-4. Coal pile drainage and yard runoff wastes will also need treat-
*ent before they are discharged. Figure 10-5 shows the sources of wastewater
fr°™ the integrated plant and the anticipated treatment. The characteristics
Of these wastewaters, the applicable treatment methods, and possible reuse of
the treated water are described in the following sections.
-5£tewater Characteristics
The characteristics of process sour water, coal pile drainage, boiler
and cooling tower blowdown, water treatment wastes, equipment cleaning
Wastes, and floor and yard drains have been identified. These categories
lnclude all types of wastewaters generated by the facility.
°ur Water Characteristics— ,
Sour water is the dirtiest wastewater generated in the process. A
Portion of it originates from the use of steam within the gasifiers. The
^sequent condensation of the unused steam usually occurs simultaneously with
Jhe condensation of hydrocarbon liquids in the offgas from fixed and fluidized
bed gasifiers. The vapor phase contains hydrogen sulfide, ammonia, carbon di-
?xide, and volatile cyanides as major impurities. The water condensed in the
rno<*-0ut drum contains these gases. The gaseous mixture is further cooled
7 ^ter scrubbing Part of the hydrogen sulfide, carbon dioxide, and ammonia
dissolve in the scrubbing water. Small amounts of phenols and other organics
> also be present in the case of the fluidized bed gasifier. However the
Stitute of Gas Technology claims that phenols and organics are not produced
n their fluidized bed gasifier.
Sour water characteristics will depend on the composition of the raw
*as. the scrubber design, and its operating conditions. Sour water character-
lstics are not available for the processes considered in this work; however,
formation from other processes has been used to determine the sour water
CharacteristicS. A by-product water analysis from Synthane gasification of
Various coals is given in Table 10-4 from Reference 10-5. Approximately 60
Pefcent of the nitrogen in coal is converted to ammonia. The concentration
?f Cyanide is notably small (0.6 mg/1 or lower). Thiocyanate levels are also
low» compared to coke plant weak ammonia liquor. The variation in phenol
!°ncentration for different coal feeds is wide, from 1,700 to 6,000 mg/1.
?e entrained bed BCR gasifier does not produce phenols; thus they will be
absent in the wastewater generated in the scrubbing operation of this configu-
tation. Sour water may contain as much as 2 to 5 percent by weight of
ailtt*onia and 2 to 6 percent by weight of hydrogen sulfide. Suspended char
Jarticles also may be present in the wastewater from the scrubbing operation.
fable 10_5 shows the composition of gasification wastewater as estimated by
he Bechtel Corporation (Reference 10-6). This wastewater composition was
CotllPiled from a number of sources including process design data and analyses
°f other gasification liquors. The sour water characteristics of Table 10-4
d 10-5 are not used in the present study; but these characteristics in-
cate the types of compounds that may be present in the wastewater.
195
-------
SOURCES OF WASTEWATER IN A COGAS POWER SYSTEM
CHEMICALS
• WASTEWATERT*
i
BOILER
TUBE
CLEANING
CHEMICALS
CLEAN GAS TO
BURNER
COMBUS AIR
vo
SLOWDOWN
RAW WATER
WATER
TREATMENT
CHEMICALS
FLOOR AND
YARD DRAINS
WASTE-
WATER
POWER SYSTEM
GAS TURBINE
STEAM GENERATING
BOILERS
STEAM TURBINE
I
I
1
I
1
LEGEND
LIQUID FLOW
GASANDSTEAMF LOW
/ / f- CHEMICALS
ff # ff OPTIONAL FLOWS
A WASTEWATER SOURCE
EXHAUST TO
STACK CHEMICALS
*# ff ff ff
CONDENSER
RECIRCULATION
- I l
I * 1 ~*r~SH DEMINERALIZER l«*
JWASTEWATERJ ' . t
WASTEWATER
DISCHARGE
TO WATER
BODY
MAKE-UP WATER
ONCE THROUGH
COOLING WATER
-------
FIG. 10-5
SCHEMATIC REPRESENTATION OF WASTE WATER STREAMS AND
ANTICIPATED TREATMENT
SOURCES
TREATMENT
REUSE OR DISPOSAL
BOILER
BL°WDOWN
*AW WATER
TMENT
WASTES
RAW GAS -
WW ATPR
SCRUBBER
COAL STORAGE
PLANT YARD AREA
COOLING TOWER .
BOILERS
DEMINERALIZERS
' SUSPENDED SOLIDS
REMOVAL STRIPPING
SUSPENDED SOLIDS
REMOVAL OIL
REMOVAL
NEUTRALIZATION
SOFTENING
SETTLING
SOFTENING
SUSPENDED SOLIDS
REMOVAL
NEUTRALIZATION
1 '
RECYCLE TO
SCRUBBER AND
COOLING TOWER
COAL PREPARATION
OR DISCHARGE
TO RIVER
RECYCLE TO
COOLING TOWER
RECYCLE TO
COOLING TOWER
COAL PREPARATION ,
OR <;i Ar;
HANDLING
BOILERS
' PRECIPITATION OF
Cu AND Fe
COAL PREPARATION
OR INCINERATION
DISCHARGE TO RIVER
PLANT
NEUTRALIZATION
SETTLING
DISCHARGE
TO RIVER
197
-------
TABLE 10-k
BY-PRODUCT WATER ANALYSIS FROM SYNTHANE GASIFICATION OF VARIOUS COALS
mg/1 (except pH)
00
PH
Suspended Solids
Phenol
COD
Thiocyanate
Cyanide
NH3
Chloride
Carbonate
Bicarbonate
Total sulfur
Coke
Plant
9
50
2,000
7,000
1,000
100
5,000
-
-
-
(l) 85 percent free NH_
(2) Not from same
analysis
Illinois
No. 6
Coal
8.6
600
^2,600
15,000
152
0.6
8,000 (i)
500
6,000 (2)
11,000 (2)
1,^00(31
(3)
Wyoming
Subbitu-
minous Illinois
Coal Char
8-7 7-9
lUO 2k
6,000 200
^3,000 1,700
23 21
0.2 0.1
9,520 2,500
31
— _
"
a" . toO
S0= = 300
3
SO^ = 1,1*00
s^o% = i ,000
North
Dakota
Lignite
9-2
6k
6,000
38,000
22
0.1
7,200
—
Western
Kentucky
Coal
8.9
55
3,700
19,000
200
0.5
10,000
-
Pittsburgh
Seam
Coal
9.3
*r * —J
23
1,700
19,000
188
0.6
11,000
-
-------
TABLE 10-5
GASIFICATION WASTEWATER - ESTIMATED COMPOSITION
(pH = 8.^)(D
Concentration
Constituent mg/1
Phenols 1^20
Other Organics 3,000
Thiocyanates 2
Hydrogen Sulfide 12
Ammonia 5Q
Hydrogen Cyanide 5
Total Dissolved Solids 930
Chloride 300
Calcium 2
Ferrous or Ferric Ions 1
!l) The gasification system is not specified
199
-------
Cooling Tower Slowdown Characteristics—
In the operation of a closed cooling system, with wet cooling towers,
the warm circulating water returning to the cooling system is cooled by
evaporating a small fraction of it. The amount of water evaporated is a
function of the temperature change of the water between the inlet and outlet
of the cooling system; approximately one percent of the circulating water is
evaporated for each 10F drop, assuming a latent heat for water of 1000 Btu/lb-
Additional water is lost to the atmosphere when water is entrained in
the air draft (drift loss). Drift losses in modern mechanical-draft towers
average 0.005 percent of the cooling-water circulating rate (0.002 percent in
modern natural-draft towers), while in older and less efficient towers it can
be as much as 0.2 percent of the circulating rate (Reference 10-4).
Because of the water losses from evaporation, the remaining water
becomes more concentrated with dissolved solids. If the concentration level
of any of the soluble salts exceeds its solubility level, the salt will
precipitate. Some of the salts are characterized by reverse solubility, that
is, their solubility decreases with increasing temperature. When cooling
water saturated with such a salt is heated in the process condensers, the
salt will deposit as a scale on the condenser tube walls and reduce heat
transfer across the tubes.
Scale formation is controlled by discharging a blowdown stream from the
cooling system to limit the concentration of dissolved solids. The amount of
blowdown is a function of the number of concentration cycles, that is, the
ratio of the content of the critical component in the circulating water to
that in the makeup water. Since the drift loss is fixed for a given tower,
the blowdown flow is varied to achieve the desired concentration cycles.
Thus, for a tower with a 0.005 percent drift, blowdown may vary from 65
percent of makeup for a saltwater system to 20 percent of makeup for the
typical freshwater tower.
A variety of chemical additives may be used to treat water circulating
in the cooling system to control scaling, erosion, and fouling. These
additives will appear in the blowdown along with matter originally present in
the makeup stream. Biological growth in the circulating water is usually
inhibited by chlorinating the water. Sulfuric acid is often added to cooli^S
water to increase the solubility of the dissolved solids and to lower the
make-up requirements due to blowdown. Pentachlorophosphate is sometimes
added to inhibit fungi attack on wooden cooling towers.
There may be particular problems associated with leakage from the
high-pressure gas process train into the cooling system. Such leakages, if
they occur, will also be present in the cooling system blowdown.
Typical cooling tower blowdown characteristics, obtained from an
operating power plant in Sioux Falls, South Dakota, are given in Table 10-6-
200
-------
TABLE 10-6
COOLING TOWER SLOWDOWN CHARACTERISTICS
Characteristic^1' Cooling Tower Slowdown
pH 8.7
Temperature (Winter), °F 69
Temperature (Summer), F 8H
Alkalinity (as CaC03) 123
BOD 5-Day H.I
COD 2.7
Total Dissolved Solids 1,293
Total Suspended Solids 37
Ammonia (as N) 0.03
ncentrations in ppm, except pH.
201
-------
Boiler Slowdown Characteristics—
The formation of scale also is a major problem associated with the
operation of boilers or waste heat recovery systems. The primary cause of
scale formation is the reverse solubility of many of the scale forming salts.
The higher the temperature and pressure of boiler operation, the more insol-
uble become the salts which form scale. Calcium and magnesium salts are the
most common ingredients of boiler scales. Calcium deposition is primarily
due to the thermal decomposition of calcium bicarbonate according to the
following equation:
Ca(HC03)2 -»• CaCC-3 + C02 + H20 . (10-2)
Deposits of iron oxide, copper oxide, and other metallic oxides are
frequently found in boilers operating with feedwater having high dissolved
oxygen content. These deposits are left by corrosion resulting from dissolved
oxygen and carbon dioxide.
Boiler blowdown is the most widely used control method for scale
formation. The amount of blowdown required is a function of the allowable
concentration of scale forming or other undesirable components in the boiler
and the degree to which the make-up water is purified. High-pressure boilers
have quite stringent contaminant limits. For example5 the allowable concen-
tration of silica varies from 125 ppm at pressures under 300 psi down to 0.5
ppm at pressures in excess of 2000 psi. As a result, the allowable number of
concentrations can be quite low in a high pressure steam system. At pressures
above 600 psi, silica present in the boiler will vaporize along with other
contaminants and escape with the steam. To eliminate silica condensation and
resultant fouling of the turbine, it is necessary to maintain extremely low
silica concentrations in the boiler which can result in a high amount of
blowdown. Other methods, such as steam washing, can be used to reduce the
contaminant vapor content of the steam permitting higher boiler water concen*
trations and reducing the required blowdown quantity or makeup water quality*
In the steam washing method, high-pressure steam bypasses the turbine for a
short period.
Boiler blowdown contains all of the boiler feedwater additives as well
as the soluble matter originally present in the boiler feedwater. Scale
formation is usually inhibited by adding chemicals, such as phosphates, whic"
precipitate scale-forming salts to form sludge. Chelating agents are also
widely used. They form complex compounds with scale-forming metal ions, thus
increasing their solubility. Sodium sulfite or hydrazine is often added to
boiler feedwater to inhibit corrosion from dissolved oxygen.
Boiler blowdown is alkaline with a pH of 9.5 to 10 for hydrazine treated
water and a pH of 10 to 11 for phosphate treated water. Hydrazine treated
boilers produce blowdown containing up to 2 ppm ammonia, those treated with
phosphate may contain up to 50 mg/£ phosphate and up to 100 mg/j, hydroxide
alkalinity.
202
-------
Water Treatment Wastes —
Water treatment waste streams are described by pH, suspended solids
c°ncentration, and concentration parameters typical of the processes involved
Or toxic elements involved in the processes.
Clarification wastes consist of clarifier sludge and filter washes.
sludge could be either alum or other ionic based sludges from
c°agulant chemicals. If the clarifier is used for both, lime softening and
1-arif icat ion, the sludge will contain calcium carbonate and magnesium
ydroxide . Filter washes contain suspended solids occurring either as light
carryover floe from the clarifier or occurring naturally in unclarified raw
^ater. The underflow from clarifiers used in treating makeup water contains
trom 0.5 percent to 5 percent solids. This underflow will go to a sludge
^ckener, where the solids content increases to 10-15 percent. The super-
natant water from the thickener is returned to the clarifier inlet.
Ion exchanger wastes may be either acidic, alkaline or neutral. Usually,
wastes do not contain suspended matter. They may, however, contain
ium sulfate and calcium carbonate precipitates because of the common ion
effect.
t*
Cleaning Wastes —
6
A variety of cleaning formulations are used to clean scale and corrosion
Posits from boilers and condensers. The cleaning procedure is usually
e^Pendent on the composition of the surface-adhering materials. Cleaning
ons are usually grouped by composition into three principal categories.
6 first category includes the alkaline cleaning mixtures with an oxidizing
Th
^r -- and a copper chelating compound, usually ammonia, for copper removal.
6 oxidizing compound converts metallic copper deposits to a divalent copper
6 which then reacts with ammonia to form a soluble complex. The wastewater
t fluents contain an ammonium ion, oxidizing agents, high levels of dissolved
Pper and iron, and have high alkalinity.
The second category includes acidic cleaning mixtures. These mixtures
remove scale caused by water hardness. They contain a strong acid
3ftH
j^ a fluoride salt to remove silica. Waste streams from such mixtures are
^ ually acidic and may contain phosphates, fluorides and BOD, as well as
&e quantities of iron, copper, and hardness forming salts.
The last group of formulations includes solutions containing alkaline
^ating agents and anticorrosion additives. These cleaning mixtures may be
alone or after acid cleaning to neutralize residual acidity as well as
" remove additional amounts of scale-forming materials. Their use generates
stewater containing alkalinity, BOD, phosphate, and scale-forming components,
j>r In addition to the three categories of cleaning formulas above, several
s Oprietary formulations have been developed and are manufactured by companies
in cleaning chemicals. Most of these chemicals are similar to
described; the resulting wastes contain alkalinity, BOD, phosphate,
compounds, and scale-forming compounds such as iron, copper, and
ness-forming salts.
203
-------
Floor and Yard Drains—
The flood drains generally contain dust and fines and floor scrubbing
detergents. This stream also contains lubricating oil or other oils washed
away during equipment cleaning, oil leakage from pump seals, and oil collected
from spillage around the storage tank area of oil processing gasifiers.
Solid Wastes
Power generation by any of the seven integrated power systems considered
in this study will result in appreciable quantities of various solid wastes.
This discussion is confined to wastes exiting the coal preparation, gasifica-
tion, gas cleanup, power system and auxiliary facilities. Therefore,
construction debris, solids from sewage systems, garbage disposals, and
waste problems that are part of developmental efforts, are not addressed in
this report. Solids formed as a result of control technology selection such
as water treatment sludges, elemental sulfur and recovered entrained particu*
lates are discussed later in this section under the heading of Residuals.
Figure 10-6 shows the points at which solids exit a generalized integrate
system. Mine tailings such as pyrite and other mineral matter will be removed
during coal processing. Intermittent losses of coal solids will occur during
preparation for the gasifier. Solid wastes exiting the coal preparation unit
were discussed earlier in this section. The mineral residues of gasified coal-
will exit the gasifiers as slag. Small quantities of dissolved gases may be
present in the slag. Spent acceptors, such as dolomite in the Conoco gas
cleanup process, will exit the gas cleanup operations.
Tables 10-7 and 10-8 show the output rates for some solid wastes
generated in the specified integrated systems.
Based on current data it is impossible to give a detailed description
of the gasifier slags. The variety of possible ash characteristics is as
unlimited as the number of possible coal compositions. Furthermore, gasifier
operating conditions will influence the slagging characteristics of the ash
contained in any specified coal feed.
Table 10-9 cites three ash characterizations. The first two are general
descriptions of coal ash formed by coal combustion. The third column of Tab*
10-9 characterizes the ash exiting a Lurgi gasifier. The distribution of
mineral matter is similar to that of the general description although some
constituents are present in quantities that do not fall in the ranges given
for the general ash characteristics.
Solid Waste Disposal— ,
With the possible exception of spent dolomite, the solid waste generate
during integrated power generation will require no additional treatment
to ultimate disposal. Solids from air and water control technologies,
may require further treatment prior to disposal.
204
-------
FIG. 10-6
SOLID WASTES EXITING A GENERALIZED INTEGRATED PLANT
COAL-
AIR OR OXYGEN
COAL PREPARATION
MINE TAILINGS
COAL
SIZED COAL FEED
GASIFICATION
SLAG
RAW FUEL GAS
GAS CLEANUP
SPENT ACCEPTORS
CLEAN FUEL GAS
GAS TURBINE AND
STEAM GENERATION
-*• FLUE GAS
78-02-138-5
205
-------
TABLE 10-7
SUMMARY OF SOLIDS EXITING THE
BCR AND IGT COMBINED CYCLES
Configuration/Type of Solid Waste
BCR/Selexol/Air and Oxygen
Slag
IGT/Selexol/Air
Slag
BCR/Conoco/Air
Slag
Dolomite Slurry (Total)
Water
Dolomite
Inerts
Total Solid Wastes
BCR/Conoco/Oxygen
Slag
Dolomite Slurry (Total)
Water
Dolomite
Inerts
Total Solid Wastes
IGT/Conoco/Air
Slag
Dolomite (Total)
Water
Dolomite
Inerts
Total Solid Wastes
Quantity of Waste, Ib/hr
70,590
73,6^0
70,590
62,200
25,660
2,370
132,800*
70,590
57,360
31,510
23,6^0
2,180
127,950
73,61+0
61,520
33,800
25,380
2,3^0
136,700
*Totals are not exact due to rounding of all values to four significant digits •
206
-------
TABLE 10-8
DESCRIPTION OF SLAG AND SALT SLURRY
EXITING THE MOLTEN SALT GASIFIER
Slurry Constituents Quantity, Ib/hr
Aqueous Phase
Na C03 (aq) M^°
NaHC03 (aq) 9,873
NaHS (aq) 2,150
H20 (1) 81,386
Total aqueous phase 98,2^9
Na2C03 (s) . 20,J
NaHC03 (s) '
Sand (s)
Slag (s) T2?589
Total solid phase 98,2U8
Total solid waste 196,^97
207
-------
TABLE 10-9
CONSTITUENTS OF COAL ASH
Constituent
Si02
A12°
Fe203
Ti02
CaO
MgO
Na20
K20
so3
C and volatiles
P
B
U and Th
Cu
Mn
Ni
Pb
Zn
Sr
Ba
Zr
Percent (1)
30-50
20-30
10-30
0.4-1.3
1.5-4.7
0.5-1.1
0.4-1.5
1.0-3.0
0.2-3.2
0.1-4.0
0.1-0.3
0.1-0.6
0.0-0.1
trace
trace
trace
trace
trace
trace
trace
trace •
Percent d ]
20-60
10r35
5-35
0.5-2.5
1-20
0.3-4
1-4
1-4
0.1-12
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
I Percent (2)
59.0
23.7
4.7
0.9
3.7
0.9
1.4
0.8
NS
5.0
US
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NOTE: NS denotes value not specified.
(1)
Conventional coal-fired boiler
(2)
Lurgi gasifier
208
-------
Landfilling is the primary technique used for ultimate disposal of solid
Wastes. Ash disposal via ponds was considered, but is not discussed here
°ecause this method of ultimate solid waste disposal is highly site specific.
Unless large amounts of land are available adjacent to the other facilities,
^e costs associated with waste transport become prohibitive. Landfilling is
ln common use for both industrial and municipal solid waste disposal. If
Properly designed, landfills are also suitable for disposal of hazardous
Landfill designs are site specific; local soil, rainfall, and
table characteristics are among the major design parameters. The quan-
and physical and chemical properties of the solids to be landfilled
re important process specific variables.
Despite the fact that every landfill design is unique, certain
8eneralizations can be made. Generally, to be economical, landfills should be
°cated within ten miles of the point of waste generation. Landfills should
,°t interface local water tables. A minimum of not more than two feet of soil
etween the landfill and anticipated high groundwater level is required in
everal states. For this reason, landfill capacities should be determined by
Sa^le land volume rather than land area.
A regular program of local air and water monitoring around the landfill
"°uld be established. Materials from the combined cycle power systems
^Posited in a landfill are poorly characterized. Thus, the precise efi
these wastes are unknown. As a result, a regular monitoring program and
--v-v. in a landfill are poorly characterized. Thus, the precise effects
^ these wastes are unknown. As a result, a regular monitoring program and
Bailed records of all landfilling and monitoring activities are essential.
t. One alternative to landfilling of solid wastes is to develop ways to use
c em beneficially as raw materials. Slag could be used a constituent of
Oristruction materials such as concrete building blocks and asphalt. Another
°*et*tial application is to use slag as an abrasive. Fragmented slags from
„ er industries have been used in sand blasters to remove paint from marine
Residuals are generated as a result of applications of control and dis-
technologies. The following paragraphs briefly discuss possible adverse
al effects which could be caused by these wastes. Where necessary,
control and disposal practices are identified.
^ Gaseous, liquid, and solid residuals exit the systems controlling emis-
^Otls to air. Particulates removed from cyclones can be returned to the gasi-
tSr °r landfilled with solids from the slag removal system. Two residual
i,?ams exit the Glaus process. Spent Glaus catalyst may require chemical sta-
0 Cation to neutralize ammonia, organic sulfur compounds, and carbonaceous
^Pounds prior to disposal in the landfill to avoid leaching problems. Glaus
6i Ur may be sold as a by-product or landfilled. New uses for sulfur are
^Jg developed in order to increase demand for this commodity. Potential
Rations include sulfur-based fertilizers, construction materials and
ing foams.
709
-------
Additional sulfur, of high purity, is recovered by the Beavon process and
can be marketed or disposed of with Glaus sulfur. The Beavon oxidizer vent is
a clean gas stream, containing nitrogen, oxygen and water. Minor quantities
of ammonia may be present in the Beavon feed gas. If present, it will exit
with the oxidizer vent gas. Two water streams from the Beavon unit, the pro-
cess condensate and the Stretford sorbent blowdown, will require further treat'
ment. Stripping out ^S from the condensate stream should make that stream
suitable for reuse in the plant. The Stretford sorbent blowdown stream
should be sent to the wastewater treatment system to remove thiosulfates,
thiocyanates and sulfates.
Black water, a wastewater generated by the liquid-phase Glaus process, is
found in the Conoco configurations. This water can be used to quench gasifie*
slag and to remove spent dolomite in slurry form.
The water and wastewater treatment facilities will generate solid resi"
duals. Solid wastes from water and wastewater treatment processes are collec"
tively referred to as sludges. There are two types of sludge in water and
wastewater. The first is suspended solid materials which form sludges through
the mechanisms of coagulation and sedimentation. The second is dissolved
chemical salts. Precipitation of these salts forms sludges.
The physical and chemical properties of a sludge depend upon the charac-
teristics of the feed water and the treatment operation selected. Clarifier
sludges from raw water treatment are either alum or iron salt sludges, depend-
ing on the type of chemical coagulant selected. For example, alum sludges
consist of aluminum hydroxide, inorganics such as clay and sand particles, a°d
organic matter including plankton. The sludge from a lime soda softening clar
rifier is calcium carbonate, and also contains magnesium hydroxide, iron or
aluminum hydroxide, suspended mineral matter, and organics present in lesser
amounts. Treatment of coal pile runoff and floor and yard drainage also pro-
duces sludge. The primary chemical salts present in such a sludge are calciuin
carbonate and hydroxides of iron, aluminum, chromium, zinc and manganese.
Some coal fines, dust, and oil also may be present. Water demineralization
also may produce sludges, calcium sulfate being the primary constituent.
These sludges can be disposed of by landfilling. Mixing the sludge from the
softening unit with slag should further reduce the risk of leaching of undesi
rable materials to local groundwaters.
TRACE ELEMENTS
The distribution of trace elements in Illinois No. 6 coal, and the forms
in which they possibly occur in coal were described earlier in this section-
Trace elements are present in less than 1 percent concentrations. However,
they must be considered from the standpoint of their potential environmental
impact.
Even in trace amounts, very toxic substances could have adverse effects-
Because of the large number of elements present, it is important to identify
those most likely to be of environmental concern. Having done this, discus-
sion of the environmental impact can focus on them.
210
-------
^ Element Evaluation
If trace elements or compounds of them are discharged to the environment
*s a result of integrated power system operations, they may cause adverse
"ealth or ecological effects. To aid in evaluating and ranking the elements
r their compounds which may have adverse environmental impact, a convenient
°°1, suggested by the EPA Technical Project Officer, uses the methodology of
Wultimedia Environmental Goals (MEG's) (Reference 10-7). MEG's characterize
nvironmental pollutants by postulating emission level goals and ambient
evel goals for air, water, and land. These goals are the result of consis-
6Tlt extrapolations which consider the following data: existing standards and
CrUeria; threshold limit values; acceptable risk levels for human exposure to
uspected carcinogens or teratogens; contamination considered reasonable for
Protection of ecosystems; and cumulative potential in living organisms.
Both acute and chronic effects are of interest for assessing the
*Jvironmental impact of trace elements from integrated power systems. MEG's
dress acute and chronic effects in several ways including the following:
ltximum Acute Toxicity Effluents (MATE's) are emission level goals which give
timates for concentrations of pollutants in undiluted emission streams that
not adversely affect those persons (for health effects) or ecological
ms (for ecological effects) exposed for short periods of time; estimated
6rmissible Concentrations (EPC's) are ambient level goals based on standards
. criteria, toxicity and carcinogenic or teratogenic potential. Toxicity-
3sed EPC's represent concentrations of toxic substances in emission streams
a ^?a> after dispersion, will not cause the level of contamination in the
lent media to exceed a safe continuous concentration for exposure of
6rsons (for health effects) or ecological systems (for ecological effects).
In the absence of data from actual gasification plants it has been
med that all the trace elements present in coal are discharged to air,
j_ r and land in the same proportions as they are present in coal. Accord-
0.8 to the method suggested by the Project Officer, dividing the concentration
inj^race elements in coal by the corresponding concentration for these elements
lcative of adverse health or ecological effects will help in establishing
HA
The same could be done with ecological MATE's and EPC's.
r
Va^-ues for tne elements. Values of the two MEG's described above, the
s and EPC's for adverse health effects through air and water have been
Water EPC's are confused by the many standards set for water. In cases
a standard exists, the EPC is the standard and the MATE is exactly 5
t 6s the EPC. In other cases, where no standard exists, EPC and MATE values
not simply related. In present lists of MEG values, water and land
are proportional. Thus, rankings of a given trace element distribution
identical for land and water and are designated by water-land. The MATE
values for the trace elements in air and water are shown in Table
The method employed for ranking the trace elements in terms of health
fQ1?cts (or ecological effects, if desired) is a four step procedure as
211
-------
TABLE 10-10
MINIMUM ACUTE TOXICITY EFFLUENT AND ESTIMATED
PERMISSIBLE CONCENTRATION VALUES FOR AIR AND WATER
Element
Ag
As
B
Ba
Be
Cd
Co
Cr
Cu
Ge
Hg
Mn
Mo
Ni
Pb
Sb
Se
Sr
V
Zn
Mg
Ti
Coal
Cone . ppm
0.11
2U.O
200.0
31.0
1.0
0.89
3.6
15.0
19.0
U.3
0.12
U8.0
7.0
15.0
11.0
1.0
13.0
37.0
17-0
U9-0
570.0
700.0
Air
(yg/m3)
10
2
3,100
500
2
10
50
1
200
560
50
5,000
5,000
15
150
500
200
3,000
500
U,ooo
8,000
6,000
Water
(UK/1)
250
250
147,000
5,000
30
50
750
250
5,000
3,^00
10
250
75,000
230
250
7,500
50
U,ooo
2,500
25,000
90,000
90,000
Air
(yg/m3)
0.02U
0.005
7.»*
1.0
0.01*
0.12
0.12
0.12
0.5
1.3
0.1
12.0
12.0
0.2h
0.36
1.2
0.5
5.5
1.2
9-5
lli.O
ll*.0
Water
(yg/i)
50*
50*
U3
1,000*
ll
10*
0.7
50*
1,000*
8
2*
50*
70
l.k
50*
7
10*
27
7
5,000*
83
83
*Standard
**Toxicity Based
212
-------
Step 1 . Obtain the ratio values of coal concentration and MATE, and
C0al concentration and EPC for all the trace elements of interest as revealed
by analyses of the coal and corresponding MEG values. In terms of simple
ecluations Step 1 can be stated as:
(«M)i "
(RE)i =
Where: (RM)i: Ratio value for trace element i using MATE
(RE)i: Ratio value for trace element i using EPC
(TEC)i: Concentration of trace element i in coal
j>tep 2. Give ranking numbers (MRANK)i and (ERANK)i to the elements on
.e (RM)i and (RE)i lists, respectively, based on the values, with rank one
8iven to the smallest ratio value. Thereby (MRANK)i and (ERANK)i will be
°btained for all the trace elements. (MRANK)i denotes the rank of trace
element i when MATE is used and (ERANK)i denotes the rank of trace element i
When EPC value is used.
.Step 3. Add the (MRANK)i and (ERANK)i to obtain (TVAL)i for all trace
Cements. In effect, this weights equally the measures of acute and chronic
adverse health (or ecological) effects to obtain a single ranking:
(TVAL)i = (MRANK)i + (ERANK)i. (10-5)
gtep 4. Give a new rank (NRANK)i based on the (TVAL)i values, with
given to the element having highest value of TVAL. This corresponds to
16 highest overall ranking for adverse health (or ecological) effects.
Employing the four steps above on the coal considered in this study
u«ing MATE and EPC values such as in Table 10-10, (RM)i and (RE)i values
("r air and water are calculated. Some of these are shown in Table 10-11.
RANK)! or (ERANK)I values are shown in the left column. The elements are
t,sPlayed in descending order of the rank which resulted. Steps 3 and 4 are
en done to obtain (TVAL)i, and based on these values for the elements, the
(NRANK)i is given to each element. The (TVAL)i and (NRANK)i values for
r and water are shown in Table 10-12. From this table it is evident that
•j.SetUc, nickel, and titanium rank in the top five for both air and water-land,
's rank is due mainly to its high concentration in coal. It should
ent no problem if it remains in the slag. Chromium is second on the list
air. However, since chromium, a nonvolatile element, should also remain
slag, it may not escape, either with any air emission that may occur or
the sour water generated during scrubbing. Nonetheless, chromium, if
from the slag, will affect the ground water quality. Elements present
slag may be in the form of oxides, sulfides, or carbonates; these com-
tl(ls tend to generate leachate of the elements. Cadmium and lead are ranked
213
-------
TABLE 10-11
ENVIRONMENTAL METHODOLOGY FOR RANKING
TRACE ELEMENTS IN AIR AND WATER
Air
Water
or
( ERANK ) i
22
21
20
19
^~?
18
.j_ w
17
16
15
-1- >
1 ll
1 3
-»- .J
1 P
_l_c_
11
10
Q
7
8
W
7
i
6
14
3
-J
2
1
(RM)i
Cr
As
Ni
Be
Ti
Mg
Cu
Cd
Co
Se
B
Ba
V
Sr
Zn
AS
Mn
Ge
Hg
Sb
Mo
15.0 -
12.0
1.0
0.5
0.12
0.095
0.095
0.089
0.071*
0.072
0.065
0.061+5
0.062
0.03!+
0.0123
0.0122
0.011
0.0096
0.008
0.0021+
0.002
O.OOlU
(RE)i
As 1+
Cr
Be
Ni
Ti
Mg
Cu
Ba
Pb
Co
B
Se
V
Cd
Sr
Zn
Ag
Mn
Ge
Hg
Sb
Mo
,800.0
125.0
100.0
62.5
50.0
1+0.7
38.0
.31.0
30.56
30.0
27.03
26.0
11+.17
6.73
5.16
1+.58
l+.O
3.31
1.2
0.83
0.58
Se
Mn
As
Ni
Cr
Pb
Be
Cd
Hg
Ti
V
Mg
Ba
Co
B
Cu
Zn
Ge
Sr
Ag
Sb
Mo
(RM)i
0.26
0.192
0.096
0.0652
0.06
0.01+1+
0.0333
0.0178
0.012
0.0078
0.0069
0.0063
0.0062
0.001+8
0.001+3
0.0038
0.0019
0.0013
0.008
0.001+
0.0001
0.0009
(RE)j
Ni
Ti
Mg
Co
B
V
Sr
Se
Mn
Ge
As
Cr
Be
Pb
Sb
Mo
Cd
Hg
Zn
Ba
Ag
Cu
L— - -"
10. TI
6*87
5.1*
^ 'k?
i'.3T
A Qo
U . 7
°'Jg
0^
• ->
0.25
0.22
n 1^
0 .-1-
n 1
0 ••*•
O.o9
n 0°"
u *
n 01
u «*
o.o°3
Q.OO*
oo1'
214
-------
TABLE 10-12
ENVIRONMENTAL RANKING OF TRACE
ELEMENTS IN AIR AND WATER-LAND
Air
Water-Land
Element
(MRANK)i
39
36
31t
32
38
26
25
21*
23
23
19
16
11*
12
10
8
6
As
Cr
Be
Ni
Ti
Mg
Cu
Pb
Co
Ba
Cd
Se
B
V
Sr
Zn
Ag
Mn
Ge
Hg
Sb
Mo
1
2
3
U
5
6
7
8
9
10
11
12
13
lU
15
16
17
18
19
20
.21
22
Element (TVAI
Ni
Se
Mn
Ti
As
Mg
V
Cr
Co
Pb
Be
B
Cd
Sr
Hg
Ge
Ba
Sb
Zn
Cu
Mo
Ag
Ul
37
35
31*
32
31
29
29
28
26
26
26
21
20
19
18
13
10
10
8
8
5
215
-------
in the middle for both air and water-land. Mercury, near the bottom of the
list, is volatile and will likely appear in quench water. It possibly could
gain a higher ranking if a volatility factor is considered for water-land
ranking. Vanadium shows higher environmental rank through water-land than
through air. Boron appears in the middle of the list for both air and water.
These rankings are useful for a first cut at relative environmental
effects caused by various elements. The method gives a preliminary comparison
among different elements which includes both their concentration in coal and
their MEG values. The ranking could be misleading if all considerations are
not taken into account, such as with the case of chromium.
Considering the volatilities of trace elements is a valuable refinement
to the ranking procedure. The (RM)i and (RE)i values of trace elements desig"
nated in Table 10-11 can be multiplied by the corresponding volatility factors-
The percentage change values for Illinois No. 6 coal in Table 10-13 were
taken to be representative values for volatilities. These are taken from
Reference 10-3 and are a summary of analytical results of feed and residue
samples of coal hydrogasified in a bench-scale unit. The same procedure as
outlined previously is then followed to establish the ranks of trace elements
in air and water-land. These modified ranks are shown in Table 10-14. As
expected, mercury ranks significantly higher for water-land. Arsenic,
cadmium, lead, beryllium mercury, nickel, vanadium, and boron appear near the
top of either air rankings or water-land rankings. Chromium has a low effect
through air or water-land and ranks near the bottom of the lists. As stated
earlier, this reflects the nonvolatile nature of chromium which causes it to
remain in slag. This modified ranking further justifies the focus on the
trace elements which were selected and asterisked in Table 10-14. When
volatility is considered, it appears that selenium would be a logical choice
for addition to the list of those to be more critically examined.
Environmental Impact of Trace Elements
The gasifier, an enclosed vessel, discharges no pollutants directly to
the environment. However, the fate of the trace elements, whether they leav
in the gasifier raw gas or in the gasifier ash/slag is controlled by factors
such as reactor configuration and bed type, extent of pretreattnent, type of
coal feed system, and operating conditions. The trace elements that leave
gasifier in the raw fuel gas and hot ash/slag streams may find their way mt°
the atmosphere and groundwater through various vent gas, water, and solid
discharges from the downstream processing steps including the water treatmen
reuse system, the sulfur recovery unit, and the slag quench.
The possible discharge sources of trace elements are shown in Figure
10-7. Particulates discharged to air from coal storage and coal preparation
will probably have the same trace element distribution as the parent coal.
Water from the same sources will contain suspended coal solids; again, the
trace element distribution in these solids will be the same as in the paren ^
coal. The refuse from coal preparation will be mainly coal associated miner
matter, with a few trace elements associated with the mineral matter.
216
-------
TABLE 10-13
TRACE ELEMENT DISTRIBUTION ILLINOIS HO. 6 COAL
Element
Feed, ppm
(1)
% Change
(2)
Antimony
Arsenic
Barium
Beryllium
Bismuth
Boron
Cadmium
Calcium
Chlorine
Chromium
Cobalt
Copper
Fluorine
Germanium
Iron
Lead
Lithium
Magnesium
Manganese
Mercury
Molybdenum
Nickel
Nitrogen
Potassium
Samarium
Selenium
Silicon
Silver
Sodium
Strontium
Sulfur
Tellurium
Thorium
Tin
Titanium
Vanadium
Ytterbium
Zinc
Zirconium
1.0
2k
31
1.0
1.1
200
0.89
3500
2,300
15
3.6
19
61
It. 3
lit, 000
11
33
570
1*8
0.12
7.0
15
10,000
1,700
0.71*
13
20,000
0.11
1,1*00
37
38,000
8.1
1
2.0
700
17
0.56
1*9
35
-28
-33
0
-21*
-51
-10
-76
-3k
-Ik
0
0
0
-26
-9
-8
-1*7
0
+2
-19
-96
-3
-7
-76
0
0
-1*2
0
-67
+7
0
-79
-1*1
__
-1(5
-3
-18
-9
-27
0
Final
Residue, ppm
(3)
NOTES:
1.
2.
3.
0.72
16
31
0.76
0.5l*
180
0.21
2,300
590
15
3.6
19
k5
3.9
13,000
5.8
33
580
39
0.005
6.8
Ik
2,1(00
1,700
0.7lt
7.5
20,000
0.036
1,500
37
7,800
J*.8
1
1.1
679
lit
0.51
36
35
Feed composition is the same as in Table 10-1.
Average values of analytical results from two similar gasification runs in Hygas.
Minus sign indicates loss and plus sign indicates gain of any given element.
The percent losses shown for PDU (Hygas) runs are good to i 6 percent (i.e.,
losses less than 10-12/8 are not significant).
217
-------
TABLE 10-lU
MODIFIED ENVIRONMENTAL RANKING OF
TRACE ELEMENTS IN AIR AND WATER-LAND
Air
Element
'*As
*Be
*Pb
*Ni
*Cd
Se
Ag
*B
*V
Ti
Zn
*Hg
Mn
Sb
Mo
*Cr
Mg
Cu
Co
Ba
Sr
Ge
(NRANK)i
1
2
3
U
5
6
7
8
9
10
11
12
13
lit
15
16
17
18
19
20
21
22
Water-Land
Element
Se
Mn
*As
*Pb
*Cd
*B
*Hg
*V
Ti
*Be
*Ni
Zn
Sl3
Ag
Mo
Mg
Co
*Cr
Sr
Cu
Ba
Ge
^Elements selected for focus (See Table 1-6).
218
-------
^7-361
FIG. 10-7
POSSIBLE DISTRIBUTION OF TRACE ELEMENTS
AIR/PARTICULATES
FROM
COAL
STORAGE
WATER/SOLIDS
AIR/PARTICULATES
c
LAND/REFUSE
SULFER
SOLIDS/SPENT CATALYSTS
AMMONIA
AIR/PARTICULATES
LAND/SLAG DISPOSAL
SOUR WATER
TREATMENT
H 1
WATER/SOLIDS
J
. COAL
SLAG
FUELGAS
—# if- WATER STREAMS
LAND/SLUDGE DISPOSAL
78-01-92-4
219
-------
The coal conversion reaction takes place in a reducing atmosphere and
may form compounds of the trace elements such as hydrides, carbonyls or sul"
fides which may be more volatile than the element. Some trace elements in
the coal react with the organic matter in ash to form organometallic compounds-
The high operating pressures in the gasifier may cause carbon monoxide to
form carbonyls with iron and nickle. The formation of three typical compounds
and their properties are shown in Table 10-15 (Reference 10-8). Many of the
trace elements volatilize to varying extents during the high -temperature pro"
cessing. The forms in which these elements appear in the raw gas are deter-
mined by their chemical form in the feed coal and by processing conditions.
The predicted volatility of some elements is given in Table 10-16 from Refer-
ence 10-9. Elements having a high volatility and high toxicity can adversely
affect the environment if released to the atmosphere. Such elements include
mercury, selenium, arsenic, and lead. In the case of zinc, boron, and fluor-
ine, the degree to which they volatilize has not yet been determined, but may
be rather significant. If only 10 percent of these elements are volatilized,
large quantities will be present in the raw gas from the gasifier since they
are present in the coal in relatively higher concentratins than other trace
elements .
It should be noted that volatilities estimated by the various researcher6
are based mainly on work done for the EPA at IGT (References 10-3 and 10-10)'
This work centered on the analysis of feed and residues from the various
stages of the HYGAS process. It includes data from both bench scale and pil°
plant work. Because gasifier conditions may vary widely from those of the
HYGAS process, the form in which the trace elements appear will vary as wil*
their associated volatilities. Nevertheless, these extrapolations are of
interest and should be helpful in obtaining a proper perspective on the trace
element problem.
Trace elements which are nonvolatile or are only partially volatile wil^
leave the gasifier with the slag. Theoretically, if the trace element cofflP0
sit ion of the entering coal and the exiting slag is known, the trace elen»enc
leaving the gasifier with the raw gas can be estimated. However, trace
element analysis is a difficult laboratory procedure; thus, the estimated *&
gas composition from the analysis of Lurgi ash and feed coal does not corf6
late with the analytical trace element composition of the raw gas, as showfl
Table 10-17 from Reference 10-11.
The slag from the gasifier is quenched and some of the trace elements
the mineral matter may dissolve in the quench water. The slag form of the
mineral matter reduces the level of water contamination. A very low scha
^^
of particulates to atmosphere from slag handling will occur. Slag is Sen t6
ally transported to the disposal site; further distribution of trace eleinefl
from slag will depend on the type of disposal method used. The trace elefl6 ^
in slag material are not likely to respond independently to weathering, ^e&
ing, burying or other chemical or physical processes; their behavior more
likely will be tied directly to the behavior of the major mineral
220
-------
TABLE 10-15
FORMATION OF TRACE ELEMENT COMPOUNDS
co
Compound, Structure Synthesis Path M.P., C
Ni(roK Niflp 3 ^ Ni(nn)^ -PR
Nickel Carbonyl
T?pfnn^ TPp i sro - -- -^ fiv ( ro 1 * °i
Iron Carbonyl ^50 C
CO CO CO
/ \/ \ / \ H2' C°
/ * ^ ' r* £- l nt-^ \ Ki
-U UO UO " - UUV^U- — " ^ ^^^ \ ^w / o
\ / \ / \ / 3 28
CO CO CO 150°C
B.P. , C Comments
U3 oxidises slowly
in air
103 oxidizes readily
in air
52 unstable in air,
has catalytic
pf f pnts .
Octa(carbonyl)-di cobalt
-------
TABLE 10-16
PREDICTED VOLATILITIES OF TRACE ELEMENTS
Possible % of Content of Coal
Element Volatile
Cl 90+
Hg • 90+
Se 71*
As 65
Pb 63
Cd 62
Sb 33
V 30
Ni 2k
Be 18
Zn (10)
B (10)
F (10)
Ti (10)
Cr nil
Mainly based on data for Pittsburgh Seam Coal from Ref. 10-10, and indicated
10$ for Zn, B, F and Ti, in the absence of data.
222
-------
TABLE 10-17
RETENTION OF TRACE ELEMENTS IN GASIFIER ASH
Predicted *
Ash Retention, %
Present in Coal
36$
67
76
h
81
53
50
Actual
Amount in Lurgi
Gas, % of Element
Present in Coal
11*0.1$
1.11
86.0
13U.U
88.5
11U.8
120.6
3sumes 100% of coal ash goes to the gasifier ash.
223
-------
Slag from the gasifier will be disposed of either by burial in mine spoil
or by landfill. The trace elements present in the slag may leach to some
extent and enter the ground water. The leaching of trace elements from slag
will be lower than leaching from ash. Leaching experiments have been con-
conducted for power plant ash by Ecology Consultants, Inc., (Ft. Collins,
Colorado) and Peabody's Central Laboratory. Both reported that if leaching
occurs with water originally meeting drinking water standards, then the
leachate will still meet drinking water standards after reaching apparent
equilibrium leaching conditions. Specifically, both lead and mercury will
occur in this leachate only in the parts per billion range. Similar leaching
studies (Reference 10-11) were conducted by Peabody's Central Laboratory on
gasifier ash produced with Montana Rosebud Coal and Illinois No. 5 and 6
seam coals. Analyses of leachate from these experiments are presented in
Table 10-18.
The raw gas from the gasifier contains entrained char particles. The
sizes of some of these could range to micron and submicron levels. Particles
in the range of 0.5 to 10 microns in diameter cannot be removed even with
high-efficiency cyclones. Concentrations of lead, cadmium, selenium, arsenic,
nickel, and chromium increase markedly with a decrease in particle size.
Most of these elements are concentrated in particles between 0.5 and 10 P in
diameter. The char particles recovered in the cyclones are generally recycled
back to the gasifier.
The raw gas, when scrubbed with water, will have more of the particles
removed; a trace quantity of particulates may remain in the gas. Volatilized
trace elements present as a vapor will condense and enter the quench water.
Table 10-19 (Reference 10-5) shows trace elements in Synthane condensate from
an Illinois No. 6 coal gasification test and Table 10-20 (Reference 10-12)
shows a breakdown of trace elements found in streams from one commercial
Lurgi coal gasification facility. The raw gas from the Lurgi gasifier is
scrubbed with water. This mixture is then treated in an oil-water separator.
The bottoms from the separator are removed as tar. The top floating layer is
tarry oil and is drained out. In the remaining water, tarry gas liquor is
further treated to remove phenols and ammonia. Assuming that all the vola-
tilized trace elements end up in the sour water generated by scrubbing the
raw gas, and using the volatility data of Table 10-16, the quantities of
trace elements in slag and sour water can be estimated. An estimate of the
trace element distribution for a BCR gasifier processing coal for a 1,000-MW
power plant is shown in Table 10-21.
In the BCR/Air-Blown/Selexol, BCR/Oxygen/Selexol, and IGT/Air-Blown/
Selexol processes, trace elements that may be neither condensed during raw
gas cooling nor removed during the subsequent scrubbing operation, but remain
in the process gas stream may be removed in the acid gas removal unit. The
recovery of ammonia from sour water by stripping is done at a temperature of
about 230F. Most trace elements present in the sour water will not vola-
tilize at this temperature and, thus, will remain in the water. Part of
this water is recycled for scrubbing and a buildup of trace elements may
occur. The rest of the stripped water is used as makeup to cooling towers.
Trace elements entering with this makeup cooling water may leave by drift or
blowdown.
224
-------
TABLE 10-18
Element
Lithium
Lead
Vanadium
Antimony
Zinc
Chromium
Copper
Manganese
Silver
Nickel
Cadmium
Beryllium
Arsenic
Mercury
CONTENT OF LEACHATE FROM GASIFIER ASH
(MONTANA ROSEBUD COAL)
Concentration (pp"b)
1 Week _ _ 2 Weeks
750
100
80
57
39
lit
11
io
8
< 8
< 3
< 3
< l
0.1
820
100
80
\g
12
20
20
30
10
< 8
< o
< 3
< j_
0.5
225
-------
TABLE 10-19
TRACE ELEMENTS IN CONDENSATE FROM
AN ILLINOIS NO. 6 COAL GASIFICATION TEST
Element
PPM
Calcium I*
Iron 3
Magnesium 2
Aluminum 0 . 8
Selenium 360
Potassium 160
Barium 130
Phosphorus 90
Zinc 60
Manganese Uo
Germanium UO
Arsenic 30
Nickel • 30
Strontium 30
Tin 20
Copper 20
Columbium 6
Chromium 6
Vanadium 3
Cobalt 2
226
-------
TABLE 10-20
DISTRIBUTION OF TRACE ELEMENTS FOR LURGI GASIFICATION
'°>>iine
Vine
50.0
27-0
92.0
10.86
52.0
56.0
93.6
50.93
99.551*
100$
Tarry Gas Liquor^
66.5
6.5
89.0
1+5.0
1*3.917
2.05!+
140.30
O.Ul
3.75
2.25
1.1*96
.11+
• 90
.08
1+.33
8.12
.03
Tar 0113?
1.25
1*.25
.001*
2.1
.003
.016
.65
.006
TABLE 10-21
POTENTIAL RATE OF TRACE ELEMENTS
Element
Chlorine
Mercury
Selenium
Lead
Cadmium
- imony
Vanadium
Nickel
Zinc
Boron
pluorine
Chromium
Feed
Ibs/hr
1,610.0
0.081*
9-1
16.8
7-7
0.62
0.7
11.9
10.
0.
3l+.
lUO.O
1*2.7
1*90.0
10.5
.5
.7
.3
Final Residue
Ibs/hr
161.0
0.008
U.9
0.23
0.1*7
8.3
8.0
0.57
30.9
126.0
38.1*
1*1*1.0
10.5
Sour Water
Ibs/hr
1,1+1*9.0
0.076
6.7
12.1*
2.8
0.39
0.23
3.6
2.5
0.13
3.1*
lU.o
1+.3
1*9.0
227
-------
The importance of trace elements in system design is apparent from the
previous discussion. Unfortunately, the estimated fate of these elements is
based on scant available data. However, it does indicate that a number of
these elements are potentially troublesome. The appearance of so many trace
elements in the most troublesome of both the air and water-land columns of
Tables 10-12 and 10-14 indicates that these must not be discharged into the
stack nor be discharged in a water stream.
While the actual form of the trace element compounds is largely unknown,
it is likely that a water wash will be sufficient to control the air emis-
sions to within allowable limits. The ability of a system using high-tempera*
ture cleanup to deal with these elements is subject to question. Further,
the "dirty" water will require processing prior to disposal. In all likeli~
hood it will be used as quench water for slag in which case virtually all of
the trace elements could be concentrated in one location. By proper atten-
tion to leaching problems, this could be the most effective means of dealing
with the trace elements.
228
-------
REFERENCES
°~1. GLuskoter, H. J., et al: Trace Elements in Coal: Occurrence and
Distribution. EPA-600/7-77-064, (NTIS No. PB 270-922), June 1977.
10—o
•i- Fleming, D. K.: Purification of Intermediate Streams in Coal
Gasification. Clean Fuels from Coal Symposium II Papers, IGT,
Chicago, 111., June 1975.
"3. Attari, A., M. Mensinger, J. Pau: Initial Environmental Test Plan
for Source Assessment of Coal Gasification. EPA-600/2-76-259
(NTIS No. PB261-916), September 1976.
"^- Rice, James K. and Sheldon D. Strauss: Water Pollution Control in
Steam Plants, Power 20(4), April 1977, pp. 5-1 to 5-20.
~~5. Forney, Albert J., William P. Haynes, Stanley J. Gasior,
Glenn E. Johnson, and Joseph P. Strakey, Jr.: Analyses of Tars,
Chars, Gases, and Water Found in Effluents from the Synthane
Process, prepared for U.S. Energy Research and Development
Administration by Pittsburgh Energy Research Center, Pittsburgh,
Pennsylvania, November 1975.
lO-fi
Skrylov V. and R. A. Stenzel: Reuse of Wastewaters-Possibilities and
Problems, Proceedings of the Workshop for Industrial Process Design
for Pollution Control, New Orleans, Louisiana, October 15-17, 1974,
American Institute of Chemical Engineers, New York, New York, 1975.
lQ-7
Cleland, J. G. and G. L. Kingsbury: Multimedia Environmental Goals
for Environmental Assessment. Volume 1: EPA-600/7-77-136a and Volume
II: EPA-600/7-77-136b, (NTIS No. PB276-920), November 1977.
lO-.a
Talmage, S. S.: Health Aspects of Coal Conversion Technologies, pre-
sented at the Southeastern Regional Meeting of the American Chemical
Society, Gatlinburg, Tennessee, October 27-29, (1976).
l(Uq
Jahnig, C. E.: Evaluation of Pollution Control in Fossil Fuel Con-
version Processes Gasification: Section 7. U-Gas Process, EPA-600/
2-74-009-i, (NTIS No. PB247-226), September 1975.
229
-------
REFERENCES (Continued)
10-10. Attari, A.: Fate of Trace Constituents of Coal During Gasification.
EPA-650/2-73-004, (NTIS No. 223-001), August 1973.
10-11. Beckner, Jack L.: Trace Element Composition and Disposal of Gasifier
Ash, presented to the 7th Synthetic Pipeline Gas Symposium, Chicago,
Illinois, October 27-29, 1975.
10-12. Sinor, J. E., editor: Evaluation of Background Data Relating to New
Source Performance Standards for Lurgi Gasification, EPA-600/7-77-057,
(NTIS No. PB269-557), June 1977.
230
-------
SECTION 11
PERFORMANCE AND COST OF INTEGRATED SYSTEMS
^TRODUCTION
The performance and cost of eight integrated power systems are presented.
he merits of air- vs oxygen-blown operation for a BCR-type gasifier are com-
pared for both a low- and high-temperature sulfur removal system. The U-Gas
bed gasifier was evaluated at two steam feed rates with low-temperature
and at low steam feed rate with high-temperature cleanup.
molten-salt gasifier has integral desulfurization and only a single con-
juration was investigated. All fuel processing systems were integrated
*th a high-performance combined-cycle power generating system. Performance
. the resulting systems is summarized in Table 11-1 and generating costs are
8lven in Table 11-2.
The results show a reduced difference between high- and low-temperature
^sterns over that previously estimated. Also, little difference in perfor-
3nce and cost of power generation is seen between the air- and oxygen-blown
," type systems. Not shown here are the emission estimates. Sulfur emis-
lK°nS from the high-temperatures oxygen-blown system are relatively high (0.73
/106 Btu). Also, nitrogen oxide control may be difficult because of the
I8h combustion temperature associated with the medium-Btu gas from the
Vgen-blown systems
The improved performance of the air-blown systems with low-temperature
eanup should improve the viability of this concept for future clean power
Deration.
th
Discussion of the integration of the various systems is divided into
ree parts. These are low-temperature cleanup, high temperature cleanup,
the molten salt system. A versatile simulation system was used in esti-
t In8 performance. It permits schematic changes and parametric variations
be made with little difficulty. It is described in Reference 11-1.
231
-------
TABLE 11-1
SYSTEM PERFORMANCE SUMMARY
UJ
S3
Low -Temperature Cleanup
Gasification and Cleanup
Coal feed rate - Ib/hr
Oxidant/coal ratio
Steam/coal ratio
Transport gas/coal ratio
Gasifier exit temp-F
Gasifier jacket heat - Btu/lb coal
Gasifier press - psia
Raw gas heating value -Btu/SCF (HHV)
Cold gas efficiency - %
Clean gas temperature - F
Clean gas heating value - Btu/SCF
U-Gas
High Steam
700,000
3.013
-557
.053
1660
686
1*00
138.8
80.1
9l*l*
(HHV) 158.6
U-Gas
Low Steam
700,000
2.86k
.197
.053
1660
686
Uco
162
81.6
91*1*
169.7
BCR-Air
700,000
2.78
.11*1*
.088
1700
2ll*
1*00
171.2
83.0
926
177.9
BCR-OO
700,000
.591*
.598
.088
1700
180
400
270.7
85.7
926
332.1*
Molten
Salt
700,000
2.806
0
0
1800
'
1*00
151.1*
80.8
805
151*
High- Temperature Cleanup
U-Gas
Low Stm
700,000
2.861*
.197
.053
1660
686
1*00
162
81.6
1,000
160.9
BCR-Air
700,000
2.78
.ll*l*
.088
1,770
21 1*
1*00
171.2
83.0
1,000
170.0
BCR-02
700,000
.591*
• 598
.088
1700
180
1*00
270.7
85.7
1,000
268.0
Utilities
Boost compressor power - MW
Let-down turbine power - MW
Gasifier & Cleanup - MW
Boiler & Cooling tower - MW
Misc Plant losses - MW
Power System *
Gas Turbine exhaust temp - F
Stack temperature - F
32.6
27-9
16.3
6.5
31.0
_
26.7
16.9
6.7
27.8
_
29-7
15.8
6.6
22
28
20.5
lU.9
6.0
35-2
15.7
lU.2
5.7
31
_
13.1*
17.2
6.8
27.8
-
21.2
16.6
6.8
22
28
19-1*
15-6
6.3
1191*
290
1189
300
1187
315
1181
310
1191
299
1193
303
1190
309
1193
318
Gas turbine - MW
Steam turbine - MW
Total - MW
Bet plant output - MW
•Efficiency - Overall
* "SO-WS.TC system.
CO
si.1.
697
1*60
1157
107U
A29
712
1*78
1190
1109
.HH3
725
1*53
1178
1098
.U39
659
1*22 '
1081
101*5
.MB
6l8
1*07
: 1025
955
-U33
730
1*83
1213
111* 5
.U57
7l+3
1*65
1208
1136
.1*51*
69!*
1*1*0
1131*
1098
.1*39
C.a.s
- 2.6OO "E1
-ratio -
-------
TABLE 11-2
OVERALL POWER GENERATION COST SUMMARY
Low-Temperature Cleanup
High Temperature Cleanup
No
u>
Capital Costs - $10
Power System
Gasification & Cleanup
Total Capital Cost
n (HHV Coal)
Electricity Costs -
Owning Cost
Power Sys. Oper. Cost
G&C Oper. Cost
Fuel at $1.00/MMBtu
Total Mills /kWi
U-Gas
High Steam
Selexol
399
26k
663
.U29
Mills /kWh
18.37
2.27
3.61*
7.96
32.2^
U-Gas
Low-Stm
Selexol
398
2h3
6Ui
M3
17-76
2.27
3.35
7-70
31.08
BCR
Air
Selexol
hoh
253
657
.U39
18.20
2.30
3.^9
7.77
31.76
BCR
Oxygen
Selexol
Ul8
307
725
.Ul8
20.08
2.38
k.2k
8.17
3^.87
Molten
Salt
.00
283
683
-U33
18.92
2.28
3.91
7.88
32.99
U-Gas
Low-Stm
Conoco
361
182
5^3
. ^-57
15. OU
2.06
2.51
7-^7
27.08
BCR
Air
Conoco
377
220
597
.„*
16. 5U
2.15
3.0U
7.52
29-25
BCR
Oxygen
Conoco
.10
299
709
M9
19. 6U
2.3^
1^.13
7-77
33.88
-------
The power system, described in Section 7, combines an advanced gas
turbine with a high-performance steam cycle for full utilization of the
available fuel energy. The gas turbine has a pressure ratio of 18:1 and a
firing temperature of 2600 F. Steam cycle throttle conditions are 2400 psi
and 950 F with a single reheat to 950 F. Condenser pressure is 4 in. Hg.
The sensitivity of overall peformance to the careful matching of fuel
processing and power system characteristics can be best demonstrated by com-
paring the estimated efficiency of the BuMines-type gasifier integrated with
the high-performance power system to previous performance estimates that used
a high (24:1) pressure ratio gas turbine and a simple, nonreheat steam cycle.
Performance with the better steam system decreases to 35 percent compared to
the 37 percent previously estimated. The major difference between the BuMines
type gasifier and those presented here is the inability to use the sensible
heat in the fuel gas to raise steam. Because of this decrease in performance,
the BuMines-type gasifier was not included in the current detailed comparisons-
Low-Temperature Systems with Selexol Desulfurization
Flow sheets for the four gasif ier/low-temperature system combinations
using Selexol desulfurization are shown in Figures 11-1 through 11-4. Mater-
ial balances for the respective systems are given in Tables 11-3 through 11-6
and utility requirements in Tables 11-7 through 11-10. Schematically, the
first three gasif ier/cleanup configurations are virtually the same, differing
only in the point of steam bleed for the gasifier and in the way in which the
waste heat boiler is segmented. The oxygen-blown system differs in that it
includes a let-down turbine, air cooler, oxygen plant and oxygen compressor.
These replace the air cooler and boost compessor needed in the air-blown sys-
tem to raise the pressure from compressor discharge to gasifier inlet pressure-
The let-down turbine extracts enough power to drive the oxygen compressor when
operating with a back pressure of 100 psig, the operating pressure required
the air separation process. For both cost and performance reasons bleed air
from the gas turbine compressors was used as a source of compressed air for
the oxygen plant. The incremental cost of the gas turbine compressor is very
small compared to a unit dedicated only to the oxygen plant. From a perfor-
mance standpoint, the high efficiencies achieved in the gas turbine compressor
outweigh the advantages of intercooling normally associated with a separate
compressor .
The fuel gas processing schematic is identical for all four systems.
fuel gas is first cooled in a boiler to approximately 1200 F then sent to a
fuel gas regenerator. The 1200 F value was selected as the maximum tempera-
ture desired for the hot side of the regenerative heat exchanger. Dirty gas
leaving the regenerator is cooled further in a feedwater heater and subseque°
cooler where most of the water vapor is condensed and removed from the gas
stream.
A water wash is used for both ammonia and particulate removal. Nearly
100 percent of the ammonia is removed along with equal molar quantities of « *
and HxS. These tend to enhance ammonia removal but require separation so
the H2S can be sent to the sulfur recovery unit. A Phosam unit is used
234
-------
SYSTEM FLOW DIAGRAM- U-GAS/SELEXOL
COAL 700.000 LS HR
STRFAM MO
Q ) PHfSS PSIA
CONDfNSAI
-------
TABLE 11-3
MATERIAL BALANCE FOR U-GAS/'AIR-BLOWN/SELEXOL
U>
Stream
Comp M.W.
CO
cor
16. oU
2.016
28.01
LB/HR
MOLS/HR
H2S 3^.08
cos 60.07
NH 17.03
N2 28.02
02 32.00
H20 18.02
Coal
Char
TOTAL
700,000
700,000
LB/HR
MOLS/HR
LB/HR
MOLS/HR
1,592,999 56,852.2
U83,603 15,112.6
32,811 1,820.8 390,029 21.6UU.2
2, 109,^13 73,785.6 390.029 21.6UU.2
LB/HR
MOLS/HR
263
175
U38
-------
Stream
MATERIAL J3ALANCE FOR U-GAS/AIR-BLOWN/SELEXOL
6 1
Comp
CH
H
CO
co2
H S
COS
NH,
Np
H20
TOTAL
M.W.
16. oi+
2.016
28.01
1+1*. 01
3^.08
60.07
17.03
28.02
18.02
Stream
Co:ap
CH
H2
CO
CO
H S
COS
Mo
Np
H,0
M.W.
16. OU
2.016
28.01
1*1*. 01
3l+. 08
60.07
17.03
28.02
18.02
LB/HR
820
521*
8,1+90
6,285
31+1+
18
__
20,676
97
37,25U
9
LB/HR
568
61
2,179
77,136
25,968
61+9
—
3,51+2
7,1+10
MOLS/HR
51.1
259.7
303.1
11+2.8
10.1
0.3
— —
737-9
5.1+
1.510.U
MOLS/HR
35.1+
30.5
77.8
1,752.7
762.5
10.8
—
126.1+
1+11.2
LB/HR
61+, 186
1+0,995
661+,899
J+9l59l4l;
26', 562
1,899
715
1,619,595
252,507
3,163,302
10
LB/HR
711
1+76,700
MOLS/HR
1+,001.6
20, 331+. 6
23,737-9
11,178.0
779. ^
31.6
1+2.0
57,801.1+
ll+,012.6
131,919.1
MOLS/HR
1+1.8
26,1+53.9
LB/HR
63
1+0
656
1+83
26
1
1,598
7
2,878
,369
,1*71
,1*11+
,930
,133
,871+
—
,922
,571+
,687
11
LB/HR
711.8
MOLS/HR
3,950.
20,075.
23,1+35.
10,995.
766.
31.
57,063.
1+20.
116,738.
MOLS/HR
7
0
0
9
8
2
.-
6
3
5
LB/HR
62,801
1+0,1+10
65!+, 235
1+06,793
11+7
1,226
—
1,595,380
166
2,761,158
12
LB/HR
1,730
92
—
in. 8 i;
2,285
MOLS/HR
3,915.3
20,Ol+l+.5
23,357.2
9,21+3.2
1+.3
20.1+
—
56,937.2
9.2
113,531.1
MOLS/HR
39.3
2.7
—
0.2
126.8
TOTAL
117,513
3,207.3
1+77,1+11 26,1+95.7
711.8
1+1.8
169.0
-------
M
Co
oo
Stream
13
MATERIAL BALANCE FOR U-GAS/AIR-BLOWN/SELEXOL
ll* 15
Comp
CEh
H2
CO
co2
H2S
COS
so2
NH-
N,3
°2
H20
S
TOTAL
M.W.
16.04
2.016
28.01
It It. 01
3l*. 08
60.07
61*. 06
17.03
28.02
32.00
18.02
32.06
Stream
Comp
CH
H
CO
co2
HpS
COS
so2
NH3
N2
°2
S
M.W.
16. oi*
2.016
28.01
It It. 01
3&. 08
60.07
61*. 06
17.03
28.02
32.00
18.02
32.06
LB/HR
379
1*1
l,i*51
80,688
279
1*33
1*93
3
1*8,802
23,9^0
156,509
17
LB/HR
1,608,257
9,997,139
1,586,880
502,81*8
MOLS/HR
23.6
20.lt
51.8
1,833.1*
8.2
7.2
7.7
0.2
1,71*1.7
1,328.5
5,022.7
MOLS/HR
36,51*3
35^,886
>*9,590
27,905
LB/HR
2U,865
18
LB/HR
MOLS/HR
775-6
775.6
MOLS/HR
13,
3,015,000 167,350
3,015,000
LB/HR
379
1,375
81,110
18
3
1*8,802
1*,208
135,895
19
LB/HR
16
MOLS/HR
23.6
1*9.1
1,81*3.0
2 ppm
0.3
0.2
LB/HR
10,188,976
233.5
3,891.1* 13,261,608
20
MOLS/HR
LB/HR
MOLS/HR
361,696
96,089
1*57,785
MOLS/HR
2,98U,785 165,670
165,670
-------
BAI^LffCE fOfl U- GAS /AIR-BLOWN /SELEXOL,
Stream
21 22 o
'
16. Ok
H2 2.016
CO 28.01
C02 Itli . 01
H2S 3li.o8
cos 60.07
so2 6i+. 06
NH3 17.03
N2 28.02
02 32.00
H20 18.02 3,126,316 173,530 396,892 22,030 2,729,^2li 151,500 1^,308 8 010
S 32.06
g TOTAL 3,126,316 173,530 396,892 22,030 2,729,U2U 151,500 1^,308 5,010
-------
SYSTEM FLOW DIAGRAM U-GAS/ LOW STEAM/ SELEXOL
COAL 700.000 LB HR
STREAM NO
S3
-P-
o
COAL STORAGE
AND HANDLING
V
A
COAl
PROCFSStNG
\/ TRANSPORT GAS
nilFNCHWATF-R • —
-------
Stream
Comp
H,
CO
co2
H2S
COS
NH3
N2
°2
H20
Coal
Char
M.W. LB/HR
16. Oil
2.016
28.01
UIt.01
3k. oQ
60.07
17-03
28.02
32.00
18.02
700,0
TOTAL
MATERIAL BALANCE FOR U-GAS/AIR-BLOWN/SELEXOL - LOW STEAM
1 ^ 3
MQLS/HR LB/HR MOLS/HR LB/HR MOLS/HR
l,5lli,078 5^.036.6
^59,608 l
31,215 1,732.2 137,795 76U6.8
LB/HR MOLS/HR
256
175
700,000
2,00l*,901 70,130.5 137,795 76U6.8
-------
TABLE 11-U (Cont'd)
Strsam
MATERIAL BALANCE FOR U-GAS/AIR-BLOWB/SELEXOL - LOW STEAM
5 6 7
Comp
^. ~
c\
c8
co2
H2S
COS
fflo
U
H20
TOTAL
to
(0 Q+T.O
M.W.
16. Ol*
2.016
28.01
M*. 01
3l*. 08
60.07
17-03
28.02
18.02
iom
LB/HR
7988
1*56
10,901.5
3,1*10.8
31*7.6
36.0
__
20,611.5
97.3
36,660
c
MOLS/HR
1*9.8
226.2
389.2
77.5
10.2
0.6
—
735.6
5.U
l,l*9U. 5
3
LB/HR
59,727
31*, 123
815,815
256,680
26,008
2,1*89
529
1,51*2,581*
98U
2,802,939
1
MOLS/HR
3,723.6
16,926.
29,125.8
5,832.3
763.1
1*1.1*
31.1
55,052.9
3,6o6.2
115, 102. U
0
LB/HR
58,927.8
33,666.8
80^,912.2
252,018.9
25,566.8
2,U50.9
—
1,521,970.7
7,157-5
2,706,670
MOLS/HK
3,673.8
16,699.8
28,736.6
5,726.1*
750.2
1*0.8-
—
5^,317.3
397-2
110,31*2.1
11
LB/MK
58,1*00
33,616
802,21*0
211,81*7
ll*3
i,6oi*
—
1,518,560
162
2,626,572
MULS/ UK
3,61*0.9
16, 67!*. 5
28,6Ul.2
1*,813.6
1*.2
26.7
—
5^,197-0
9.0
108,007-1
12
Court
CHU
H2
CO
co2
H2S
COS
H20
TOTAL
M.W.
16. Ol*
2.016
28.01
1*1*. 01
3l*. 08
60.07
17-03
28.02
18.02
LB/HR
527.7
51.0
2,672.2
1*0,172.3
25,^23.7
81*7.0
3,370.8
1,679.5
MOLS/HR
32.9
25-3
95-1*
912.8
71*6.0
lU.l
120.3
93.2
2,bUb
LB/HR MOLS/HR
LB/HR
MOLS/HR
LB/HR
MOLS/HR
526.2 30.9
^76,699.3 26,U53-9
U77,225.5 26,U8U.8
526.2
526.2
30.9
30.9
1289.5
61.3
3.U
1693.9
30U8.1
29-3
1.8
0.2
9^.0
125.3
-------
Stream
MATERIAL BALANCE FOR U-GAS/AIR-BLOWN/SELEXOL - LOW STEAM
13
15
16
Comp
CH
RZ
CO
co2
H2S
COS
S0o
2
Np
H20
Si o
•P* "
LO
TOTAL
M.W.
16. oi*
2.016
28.01
W*. 01
s^.os
60.07
61*. 06
17.03
28.02
18.02
32.06
Stream
Comp
OTJ
/i
H
CO
C02
H2S
COS
HHo
Np
°2
M.W.
16. oi*
2.016
28.01
1*1*. 01
31*. 08
60.07
17.03
28.02
32.00
18.02
LB/HR
351.28
3U.1
1781.1*
1*3,578.7
272.6
636.7
595.8
3.U
1*8,062.7
18,1*23.6
113,7^0.28
17
LB/HR
1,632,9^7
10,1*09,885
1,722,176
1*31,939
MOLS/HR LB/HR MOLS/HR
21.9
16.9
63.6
990.2
8.0
10.6
9.3
0.2
1715.3
1022 . 1*
21*, 1*71.1* 763.3
3,858.1* 2U, 1*71.1* 763.3
18
MOLS/HR LB/HR MOLS/HR
3T,10U
369,538
53,818
23,970 2,389,035 165,910
LB/HR
351.3
—
1,781.1*
1*1*, 027. 6
21*
_ —
3
1*8,062.7
3,222.0
97,1*72.0
19
LB/HR
20U,355
MOLS/HR
21.9
—
63.6
1,000.1*
2 ppm
o.i*
-.—
0.2
1,715.3
178.8
2,980.6
MOLS/HR
11,31*3
LB/HR MOLS/HR
10,606,005 376,500
3,200,61*0 100,020
13,806,61*5 1*76,520
20
LB/HR . MOLS/HR
2, 751*, 773 152,907
TOTAL
ll*,186,9^7 1*81*,1*30 2,989,035 165,910 20l*,355 11,31*3 2,75^,773 152,907
-------
Stream
Comr
CO
CO,-
COS
M.W.
16. Ql*
2.016
28.01
3l*. 08
60.07
17-03
28.02
18.02
TOTAL
Stream
Com
H2
CO
CO,
COS
N
M.W.
16. oi*
2.016
28.01
1*1*. 01
31*.08
60.07
17-03
28.02
18.02
TABLE 11-1* (Cont'd)
MATERIAL BALANCE FOR U-GAS/AIR-BLOWN/SELEXOL - LOW STEAM
21 22 23
LB/HR MOLS/HR LB/HR MOLS/HR LB/HE MOLS/HR
3,107,189 172,1*30
3,107,189 172,1*30
25
LB/HE
MOLS/HR
ll*l*,668 8,030
ll*l*,668 8,030
26
LB/HR MOLS/HR
2,962,1*88 i6i*,i*oo
2,962,1*88 16!*,1*00
27
LB/HR MOLS/HR
2U
LB/HR
LB/HR
MOLS/HR
136,381 7,570
136,381 7,570
MOLS/HR
TOTAL
-------
COAL roo.anaiH'iin
SYSTEM FLOW DIAGRAM BCR/AIFt—BLOWN/SELEXOL
480 PSIG STM
,
572
1
rjtR
TUBE
COO
EXCHA
O
SINE
JNG
NGER
ISO
~
/"KP~
/ TURB
50 PSIG STM -»-
7
950
REHEAT
TURBINE
COOLING
TOWER
Q « 2487
\/ I
CONDENSER
P 4 IN HG
1 '
J54J 1187
H.P. STM
FROM PROCESS
REHEATER
SUPERHEATER
REHEATFR
BOILER
ECONOMIZER
TO FFTI)
HEATEFi.S
CONDFNSATE
I
CO
-------
IS)
•p-
TABLE 11-5
MATERIAL BALANCE FOR BCR/AIR-BLOWE/SELEXOL
*
Stream l 2 3
Comp M.W. LB/HR MOLS/HR LB/HR MOLS/HR LB/HR MOLS/HR LB/HR
MOLS/HR
CE^
H2
CO
co2
H2S
cos
NH3
N2
°2
H20
Coal
Char
TOTAL
16. Oh
2.016
28.01
UU.01
3^.08
60.07
17.03
28.02
32.00
18.02
700,000
700,000
1,1*69,803 52,U55.5
hh6,208 13, 9l*U. 0
10,051 557.8 120,960 6,712.5 h ,319
2,880
1,926,062 66,957.3 120,960 6,712.5 7,199
* Coal is dried to 2 percent moistirre prior to firing; flow at the dried condition is 68^,286 l"b/hr.
-------
MATERIAL BALANCE FOR BCR/AIR-BLOWN/SELEXOL
N3
Stream
Com
^
CO
co2
COS
H20
TOTAL
' •
16.01+
2.016
28.01
1+1+.01
3l+. 08
60.07
17.03
28.02
18.02
..
LiDf n.K
1,391
711
20,753
596
102
139
61,813
Stream
Conp
CH^
H2
CO
CO
H2S
COS
N,3
H2°
TOTAL
M.W.
16. oU
2.016
28.01
1+1*. 01
3l+. 08
60.07
17.03
28.02
18.02
LB/HR
529
1+6
2,927
20,126
23,836
1,532
3,259
5,7H+
57,969
MOLS/HR
86.7
352.5
71+0.9
78.3
17.5
1.7
1,237.5
7.7
2,522.8
-
MOLS/HR
33.0
22.7
101+.5
1+57.3
699.1+
25.5
116.3
317-1
1,775.8
LB/HR
60,551
30,875
901,631
11+9,1+6?
25,591+
1,506,167
39,871
2,726,81+7
LB/HR
8,101+
37l+, 220
382,321+
MOLS/HR
3,775.0
32,189.6
3,396.1
751.0
75-6
1+78.8
53,753.3
2,212.6
111,91+6.9
10
MOLS/HR
1+75.8
20,766.9
21,21+2.7
LB/HR
59,160
30,161+
880,881+
126,273
23,968
1+, 1+1+0
1,1+71,1+93
5,867
2,602,21+9
LB/HR
8,029
8,029
MOLS/HR
3,688.3
ll+, 962.1
31,1+1+8.9
2,869.2
703.2
73.9
52,515.8
325.6
106,587.0
11
MOLS/HR
1+71.1+
1+71.1+
u
LB/HR
58,631
30,118
877,962
106,11+8
133
2,908
1,1+68,231+
153
2, 51+1+, 287
12
LB/HR
19,71+3
1,029
51
26,OU8
1+6,871
MOLS/HR
3,655.3
ll+, 939.1+
31, 31+1+. 6
2,1+11.9
3.9
1+8.1+
52,399.5
8.5
10M11.5
MOT S /H"R
1 iwj_to / niA
1+1+8.6
30.2
3.0
1,1*1+5.5
1,927.3
-------
TABLE 11-5 (Cont'd)
MATERIAL BALANCE FOR BCR/AIR-BLOWW/SELEXOL
Stream
Comp M.W.
LB/HR
13
MOLS/HR
15
16
CHU
H2
CO
co2
H2S
COS
so2
NH
V
H20
S
TOTAL
16. Ol*
2.016
28.01
Hit. 01
3l*. 08
60.07
61*. 06
17.03
28.02
32.00
18.02
32.06
•F- Stream
00
Comp
H
CO
co2
HpS
COS
so2
NH-a
N2
°2
H2
S
M.W.
2.016
28.01
1*1*. 01
3^.08
60.07
6k. 06
17.03
28.02
32.00
18.02
32.06
353
30
1,9^9
1+2,258
1*98
1,021
1,012
32
!*7,673
1+5,203
11,0,029
17
LB/HR
1,61*7,910
10,686,061*
1,803,261*
l*Ol,U32
22.0
15.1
69.6
960.2
lU.6
17.0
15.8
1.9
1,701.1*
2,508.5
5,326.1
MOLS/HR
37, kkk
379,31*2
56,352
22,277
353 22.0
"*
1.0U5 37-3
1*1*, 388 1,008.6
— 2 ppm
1*8 ~0.8
— —
32 1.9
1*7,673 1,701.1*-
3,179 176.1*
2U, 178 75U. 1
2l*,178 75U.1 96,718 2,9U8.1*
18 . 19
LB/HR MOLS/HR LB/HR MOLS/HR
2,827,338 156,900 808,917 1*1*, 890
.LJJ->/ im j.'jw±ju/ ni\
10,875,282 386,059
3,281,920 102,560
Ik, 157, 202 1*88,619
20
LB/HR MOLS/HR
1,990,128 110,1*1*0
H95:M-5
2,821,338 1.56,900
802,115 UU,522
1,983,001 110.068
-------
fOff
Stream 21 22 23 2i
Comp M.W. LB/HR MOLS/HR LB/HR MOLS/HR LB/HR MOLS/HR LB/HR MOLS/HR
CHjj 16.01*
H2 2.016
CO 28.01
C02 UU.01
H2S 3^. 08
cos 60.07
S02 6U.06
M 17.03 *
N2 28.02
0 32 00
H20 18^02 2,750,392 152,630 199,661 11,080 2,750,392 152,630 202,500 11,2UO*
s 32.06
.^ TOTAL
Includes steam for deaerator
-------
SYSTEM FLOW DIAGRAM BCR/OXYGEN-BLOWN/ SELEXOL
COAL 700.000 LB HR
COALSTOHAbf
AND HANDLING
V
0
Si AC; AMD
COAl
PROCrSSING
IHANSPOH] Cl
OUENCHWATFR i—
SIRf AM NO
-------
MATERIAL BALANCE FOR BCR/OXYGEN BLOWN/SELEXOL
LB/HR MOLS/HR
LB/HR
MOLS/HR
LB/HR
MOLS/HR
LB/HR MOLS/HR
to
CH^
H2
CO
co2
H2S
cos
NH
N2
°2
H20
Coal
Char
TOTAL
16. OU
2.016
28.01
UU.01
3H . 08
60.07
17.03
28.02
32.00
18.02
700,000
700,000
7,302 260.6
1*08,627 12,769.6
1*18,320 23,2lU.2 3,667
2,U1*5
1*15,929 13,030.2 Iil8,320 23,2lU.2 6,112
* Coal is dried to 2 percent moisture prior to firing; flow at the dried condition is 68^,286 Ib/hr.
-------
TABLE 11-6 (Cont'd)
MATERIAL BALANCE FOR BCR/OXYGEN BLOWN/SELEXOL
Stream
Comp M.W. LB/HR
7
MOLS/HR
LB/HR
MOLS/HR
LB/HR
MOLS/HR
LB/HR
MOLS/HR
CHU
H2
CO
co2
H2S
COS
NH3
N2
H20
16. Ol*
2.016
28.01
1+1*. 01
3l+. 08
60.07
17.03
28.02
18.02
3,1+07
2,093
3U,8lU
19,361+
1,237
210
—
1+1*6
191
212.1*
1,038.2
1,21*2.9
1*140.0
36.3
3.5
—
15.9
10.6
72,531
UU.5U9
7Ul,108
1*12,356
26,255
1+,620
8,56l
9,^26
19l+, 380
1+,521.9
22,097.6
26,1*58.7
9,369.6
770.1*
76.9
502.7
336.1*
10,786.9
69,120
1*2,1*56
706,297
372,276
23,938
l*,l*0l*
—
8,980
3,896
•^
1*,309.2
21,059.1*
25,215.9
8,1*58.8
702.1*
73.3
—
320.5
216.2
68,500
1*2,391
703,91*7
312,933
133
2,881*
»
8,961
85
1*,270.6
21,027.1*
25,132.0
7,100.5
3-9
1*8.0
—
319.8
1*.7
TOTAL
61,762 2,999.8
1,513,786 7!*, 921.1 1,231,367 60,355-7
1,139, 8314 57, 916.9
Stream
Comp
CH.
*4
H2
CO
CO
HpS
COS
NH3
No
H00
M.W.
16. oi*
2.016
28.01
1*1*. 01
3l+. 08
60.07
17.03
28.02
18.02
9
LB/HR
619
65
2,350
59,339
23,805
1,520
—
20
3,811
MOLS/HR
38.6
32.0
83.9
1,31*8.3
698.5
25.3
—
0.7
211.5
LB/HR
8,5C
266, 7C
10
11
12
MOLS/HR
LB/HR
MOLS/HR
LB/HR
MOLS/HR
20,720
1,080
1*70.8
31.7
TOTAL
91,529 2,1*38.8
1*99.6
275,208 15,299.8
8,^55
8,1*55
1+96.5
1*96.5
53 3.1
27,338 1,517.1
1*9,191 2,022.7
-------
Stream
Comp M.W.
13
15
16
LB/HR
MOLS/HR
LB/HR
MOLS/HR
LB/HR
MOLS/HR
LB/HR
MOLS/HR
CHU
H2
CO
co2
H2S
COS
so2
NH_
N,3
°2
HO
S
16. ok
2.016
28.01
Mi. 01
3^.08
60.07
6U.06
17.03
28.02
32.00
18.02
32.06
TOTAL
S3
Ul
u>
Comj
CH
H
CO
co2
HpS
COS
so2
NHo
3
°2
S
Stream
j M.W.
16.0
2.016
28.01
Mi. 01
3^.08
60.07
61i.o6
17.03
28.02
32.00
18.02
32.06
lilli
^3
1,566
82,221;
1*98
1,015
999
36
Mi, 726
176,165
17
LB/HR
1,608,389
10,387,630
2,135,328
533, 55^
25.8
21.3
55.9
1,868.3
Ik. 6
16.9
15.6
2.1
1,593.3
2,kQ2.0
6,095.8
MOLS/HR
36,5^6
370,722
66,729
29,609
lilli 25.8
—
852 30. k
8U,055 1,909.9
2 ppm
U8 0.8
—
36 2.1
M;, 633 1,592.9 11,930,663 ^25,791
3,6l9,8Uo 113,120
U,087 226.8
2U,192 75^.6
2^,192 75^.6 13^,125 3,788.7 15,550,503 538,911
18 19 20
LB/HR MOLS/HR LB/HR MOLS/HR LB/HR MOLS/HR
2,877,073 159,660 1,106,^28 6l,lfOO 1,527,915 8*1,790
TOTAL
13,056,512 U67>060 2,877,073 159,660 1,106,^28 6l,Uoo 1,527,915 8U,790
-------
Stream 21 22
Comp
CHU
H2
CO
co2
H2S
COS
so2
NH
V
°2
H20
S
M.W. LB/HR MOLS/HR LB/HR
16. OU
2.016
28.01
Mt.Ol
3^.08
60.07
6k. 06
17.03
28.02
32.00
18.02 2,517,935 139,730 1*96,992
32.06
MOLS/HR
27,580
TABLE 1.1-6 (Cont'd)
MATERIAL BALANCE FOR BCR/OXYGEN BLOWN/SELEXOL
23 2h
LB/HR MOLS/HR LB/HR MOLS/HR
10
2,517,935 139,730 6l
TOTAL 2,517,935 139,730 1+96,992 27,580 2,517,935 139,730 6l,UU8 3,UlQ
-------
TABLE 11-7
UTILITIES - U-GAS/SELEXOL/AIR-BLOWN
Coal Gas Scrubbing Acid Gas Ammonia Sulfur
Preparation Gasification & Gas Cooling Removal Recovery Recovery Total
Steam #/hr
1*20 psig (sat) 390,030 6,900 396,930
100 psig (sat)
50 psig (sat) 1U3,300 55,700 (5^,600) lU*,l*00
Power - kW 2,500 810 225 22,765 575 1,060 27,935
NJ
01 Cooling Water - gpm 32,680 30,2Uo 805 125 3^,730
Cooling Duty MMBtu/hr 326.8 302.lt 805 1-25 638.5
-------
TABLE 11-8
UTILITIES - U-GAS/SELEXOL/AIR-BLOWN - LOW STEAM
fO
Ul
Coal
Preparation
Steam #/hr
420 psig (sat)
100 psig (sat)
50 psig (sat)
Power - kW
Cooling Water - gpm
Cooling Duty MMBtu/hr
2,500
Gasification
137,795
810
Gas Scrubbing Acid Gas Ammonia Sulfur
& Cooling Removal Recovery Recovery Total
6,900
135,^00 55,700
225
10,300
103
21,500 575
28,600 805
286.0 8
(5^,600) 136,500
1,060 26,670
125 33,090
1.3
-------
UTILITIES BCFt/SELEXOL - AIR-BLOWN
Coal Feed - 700,000 #/hr 111. #6
10
Ul
•vl
Steam #/hr
1+20 psig (sat)
50 psig (sat)
Power - kW
Cooling Water - gpm
Cooling Duty Mvffitu/hr
Coal
Preparation
^,360
Gas Scrubbing
Gasification & Cooling
121,000
1,960 80
5,H80 7,800
78
Acid Gas
Removal
131,500
20,1*00
27,770
277-7
Ammonia
Recovery
78,700
5^, 000
1,850
650
6.5
Sulfur
Recovery
(5^,600)
1,060
125
1.3
Total
199,700
130,900
29,710
36,560
3^7.7
-------
TABLE 11-10
UTILITIES BCR/SELEXOL - OXYGEN-BLOWN
NJ
Ln
00
Steam #/hr
1*20 psig (sat)
50 psig (sat)
Power - kW
Cooling Water - gpm
Cooling Duty MMBtu/hr
Coal
Preparation Gasification
1*18,320
1*,360 1,260
5,^80
Gas Scrubbing Acid uas
& Cooling Removal
7!*, 800
80 11,930
21,100 15,800
211 158
Ammonia
Recovery
78,700
1*1,200
1,850
290
2.9
iDUJ-iur
Recovery
(5^, 600)
1,060
125
1.3
Total
1*97,020
61,1*00
20,5^0
2l*,590
373.3
-------
to separate the I^S and recover the ammonia for possible use as a byproduct.
B°th low- and reheat-pressure steam are required by the Phosam process (see
APpendix A). A concentrated slurry of char recovered from the particulate
scrubber is returned to the gasifier and recycled to extinction.
Gas leaving the water wash is sent to the Selexol acid-gas removal system.
Peration of that unit is also described in Section 6. Steam bled from the
Power system is used in the stripper. Acid gas from the Selexol stripper
juries between 24 and 40 percent in H2S concentration, the particular value
°eing a direct function of the amount of C(>2 in the fuel gas. A small per-
Cfintage of the CC>2 (.16 percent) is absorbed and finds its way into the acid
&as. To this stream~~is added the H2S recovered from the water scrub and the
c°robined total is sent to the Glaus plant for sulfur recovery. Low-pressure
team is raised in the burner and sulfur condensers of that process. The oper-
ational details of the Glaus plant and Beavon tail gas cleanup process are as
in Section 6. Nearly all of the sulfur is recovered and emissions from
part of the process are relatively insignificant.
The clean fuel gas exiting the Selexol process is reheated to approxi-
925 F by heat exchange with the raw gas and delivered to the gas
turbine fuel control.
. The major source of heat for the steam cycle is the waste heat recovery
Oiler, although some steam is raised at reheat pressure by heat exchange with
^e turbine cooling air. In this case, boiler is a misnomer in that virtually
.1 the steam is raised in the gasifier jacket and/or the heat recovery boiler
11 the gasifier exit stream.
The reheater is shown in two sections. This minimizes the pinch problem
"at would occur in the downstream unit of a series of combinations of reheater
J^d superheater. An alternative is to arrange them in parallel in the gas
*owstream. In this arrangement, reheat-pressure stream is available at bleed
^mperature, reheat temperature or an intermediate temperature as required for
. e gasifier. In the air-blown systems, very little, if any, steam is raised
i? the waste heat boiler in the gas turbine exhaust. However, in the oxygen-
1-own system, the smaller fuel gas flow rate results in reduced quantities of
.^t and, therefore, less steam is available from the waste heat boiler. In
•Us system, it is necessary to raise both high- and reheat-pressure steam in
^e gas turbine exhaust waste heat boiler. This is done in order to utilize
j. °f the available heat in the exhaust stream. As a result, temperature
inferences are as small as practical, making for a larger and more complex
6 exchanger. This shows up in steam cycle cost for the oxygen-blown sys-
(It is possible that slight modifications in the steam cycle could
e the size of the waste heat boiler but the changes would not materially
*ect study conclusions.)
The steam turbine is conventional in all respects except for bleed pro-
lsions. Because of the large amount of waste heat available from the process
j^ gas turbine exhaust, bleed for feedwater heating is unnecessary. The con-
Stlser pressure of 4 in. Hg is compatible with cooling tower use. In general,
6 combined cycle is not very sensitive to condenser pressure and a higher
259
-------
level could be helpful if it were desired to reduce water use rate in the
cooling tower.
As expected, the two air-blown systems with low steam feed rates had sim-
ilar (.443 and .439) efficiencies. The U-Gas system with high steam feed was
over one point less and the oxygen-blown system approximately two points less.
Molten Salt System
With the exception of the fuel processing system, operation is quite sim-
ilar to the other low-temperature systems. Both gasifier and fuel processing
systems are described in Section 6. The system flow diagram is shown in Fig-
ure 11-5 the material balance in Table 11-11 and utility requirements in Tab-
le 11-12.
The molten salt gasifier does not require steam feed. However, the salt
recovery process needs 200 psi steam for calcination of the sodium bicarbonate.
Also, some low-pressure steam is used to preheat the sodium carbonate solution
prior to entering the CC^-^S absorber. This steam is raised in the
Glaus plant. Also, part of the calciner steam requirement is raised in the
Glaus plant. The remainder is raised by heat exchange with the turbine cooling
air and superheated in the waste heat boiler. Some of that steam is used for
the feedwater pump drive.
A further difference between the molten salt and the other low-tempera-
ture systems results from the fact that the fuel gas is not cooled below its
dew point. In the Selexol systems, the lower heat capacity of the dry gas
on the cold side of the regenerative heat exchanger results in an elevated
exit temperature on the hot side with the remainder of the sensible heat used
for feedwater heating. In this system, it was necessary to use some of the
heat from the gas turbine exhaust waste heat boiler for feedwater heating.
This increased the temperature differences in this waste heat boiler thereby
reducing boiler size and cost. Quite possibly, a slight improvement in effi~
ciency could be achieved by the use of regenerative feedwater heating up to
250 F. However, it would not materially affect the results or conclusions of
the study.
High-Temperature Systems
The three systems using Conoco desulfurization are shown in Figures 11~«»
11-7, and 11-8. Material balances are given in Tables 11-13, 11-14, and ll-l5
and utility requirements in Tables 11-16, 11-17, and 11-18. The same diffe-
rences between air- and oxygen-blown systems appear here as in the low-temper*
ture systems. While all systems required the use of the gas turbine exhaust
waste heat boiler for feedwater heating, the oxygen-blown design again proved
more difficult and a stack temperature lower than 318 F was not achievable.
The fuel gas flow path for these systems is quite simple. Raw gas from
the gasifier goes directly to the desulfurizer except in the air-blown BCR
case. There, the desulfurizer cannot operate above 1600 F due to the low C02
partial pressure in the gas stream. (The temperature limitation of 1000 F i9
to prevent calcining of the absorbent during desulfurization.) In that syste
260
-------
MOL TEN SAL TSYSTEM FLOW DIAGRAM
COAL-700.000 LB/HR
MAKE-UP
SODIUM _
CARBONATE
-------
TABLE 11-11
MOLTEN SALT SYSTEM
to
Stream
Component
Coal - as rec'd
Na CO
Ua S
NaOH
Ash
Sand
Sulfur
Total Solids
and Melt
1
LB/HR
700,000
2
LB/HR
169,565
3
LB/HR
30,730
LB/HR
2,606
700,000
170,01+7
30,730
2,670
5
LB/HR
128,682
8,885
72,590
6k
25^,769
6
LB/HR
19,6l6
19,6l6
-------
TABLE 11-11 fCont 'd)
MOLTEN SALT SYSTEM
NJ
O
to
Comp
^
CO
co2
H2S
COS
so2
N2
°2
H20
Na^Ct
Stream
M.W.
16. 04
2.016
28.01
44.01
34.08
60.07
64.06
28.02
32.0
18.02
33 105.99
TOTAL
Comp
CHll
H2
CO
co2
H2S
COS
so2
°2
Na0CO
Stream
M.W.
16.04
2.016
28.01
44.01
34.08
60.07
64.06
28.02
32.0
18.02
105-99
7
LB/HR MOLS/HR LB/HR
25,585.4
29,333.0
843,101.
144,040.3
3,558.0
426.5
1,506,988 53,782.6 1,513,968.2
457,504 14,297.
44,958.1
2,607.4
1,964,492 68,079.6 2,607,770.9
11
LB/HR MOLS/HR LB/HR
1,468,129.6
1,588.7
9,157,269 326,812 9,0o6,o48.3
2,780,040 86,876 i, 448,960.0
365,247.4
p.
-------
ro
TABLE 11-11 (Cont'd)
MOLTEN SALT SYSTEM
Stream 15
Comp
CH
E2
CO
co2
H2S
COS
so2
N2
°2
Na2C03
TOTAL
Stream
Comp
J —
NaoCOo
NaHCOo
NaHS
H00
M.W.
16. OU
2.016
28.01
UU.01
3lt. 08
60.07
6U.06
28.02
32.0
18.02
105.99
M.W.
105-99
81*. 01
56.06
18.02
LB/HR
68,888.9
211.3
350.25
105.6
69,556.05
16
LB/HR
U.8UO
9,873
2,150
81,388
MOLS/HR
1,565.3
6.2
12.5
3.3
1,587.3
MOLS/HR
1*5.67
117-5
38.35
U,5l6.5
17 18 19
LB/HR MOLS/HR LB/HR MOLS/HR LB/HR MOLS/HR
2l*U,303 2, 301*. 9 321,1*22 3,032.6 209,7^7 1,978.9
136,922 1,629.83 65,696 782 235,1*39 2,802.5
66,868 1,192.79 32,817 585- U 37,677 672
l,8ll*,H2 100, 672. lU 1,760,601 97,702.6 1,758,770 97, 600.9
TOTAL
98,251
U,718.02 2,262,205 105,799-66 2,180,536 102,102.6 2,2U1.633 107,835-7
-------
TABLE 11-11 (Cont'd)
MOLTEN SALT SYSTEM
Stream
20
21
22
Comp.
NagCO
NaHCO
NaHS
H20
TOTAL
S3
Ul
Comp.
NagCO
NaHCO
NaHS
H20
M.W.
o 105.99
3 81*. 01
56.06
18.02
Stream
M.W.
3 105.99
3 81*. 01
56.06
18.02
LB/HR
57,716
217,397
37,708
l,6ll,0l*6
1,923,867
LB/HR
2,1*OU,1*56
MOLS/HR
5UU.5
2,587.8
672.6
89,1*03.2
93,208.1
21*
MOLS/HR
133,1*32.6
LB/HR
56,656
213,1*06
37,016
1,581,1*71
1,888, 5U9
LB/HR
619,927
MOLS/HR
53^.5
2,51*0.2
660.3
87,762
91,1*97
25
MOLS/HR
31*, 1*02. 2
LB/HR
1,060
3,991
693
29,575
35,319
LB/HR
128,030
MOLS/HR
10.00
U7.50
12.36
l,6Ul.23
1,711.09
26
MOLS/HR
7,101*. 9
23
LB/HR
MOLS/HR
2,1*28,766
-------
TABLE 11-12
UTILITY SUMMAEY - MOLTEN SALT SYSTEM
Coal Gasification & Ash Removal Sulfur
Handling Gas Cleanup Salt Recovery Recovery Total
Steam = Ib/hr, 200 psi - 182,700 (37,800) lUl+,900
50 psi - 11,UOO (ll,UOO)
Power - kW 2,500 100 12,200 850 15,650
Cooling Water - GPM Uo,100 120 U0,220
Cooling Duty - MMBtu/hr U01 1.2 kQ2
Make-up Na2CO - TPD 369 369
-------
SYSTEM FLOW DIAGRAM U-GAS/ LOW STEAM/ CONOCO
COAL- 700,000 LB/HR
SLAG AND
OUFNCHWATFR
SYSTFM
-------
TABLE 11-13
Stream
Comp.
CHj^
«2
CO
co2
H2S
COS
N>
00
NH_
N2
°2
E^O
Coal
Char
TOTAL
M.W.
16.0*4.
2.016
28.01
U4.01
3^.08
60.07
17.03
28.02
32.00
18.02
LB/HR MOLS/HR
700,000
700,000
MATERIAL BALANCE FOR U-GAS/LOW STEAM/CONOCO
1 2 3
LE/HR MOLS/HR LB/HR MOLS/HR
1,51^,078 5M35-6
1*59,608 lU,362.7
31,215 1,732.2 137,795 7,6^6.8
2,00^,901 70,130.5 137,795 7,6^6.8
-------
Stream
Comp. M.W.
BALAffCE fO_R IS-GAS/LOW STEAM/COffOCO
MOL/HR
MOL/HR
MOLS/HR
7
MOLS/HR
CO
co2
COS
NH3
TOTAL
16. (A
2.016
28.01
44.01
34.08
60.07
17.03
28.02
18.02
Stream
Comp.
°2
so2
NO
TOfAT
M.W.
32.0
28.02
44.01
18.02
64.06
30.01
Stream
Comp.
COp
V
°2
so2
TOTAL
M.W.
44.01
18.02
34.08
32.0
28.02
64.06
788
452
io,73l
3,908
7
7
20,343
1,016
37,252
8
LB/HR
3,172,192
10,1*55,607
13,627,799
1 P
LB/HR
2,711
366
3,688
5776?
49.1
224.2
383.1
88.8
0.2
0.4
726.0
56.4
1528.2
MOLS/HR
99,131
373,148
^72,279
MOLS/HR
61.6
20.3
108.2
190.1
59,727
34,123
815,815
256,680
26,008
2,489
529
1,542,584
64^84
2,802,939
9
LB/HR
1,692,768
10,260,420
1,715,800
5ll,94l
1,268
921
14,183,118
13
LB/HR
382,900
55,363
3,050
441,313
3,723.6
16,926.0
29,125.8
5,832.3
763.1
4i.4
31.1
55,052.9
3,606.2
115,102.4
MOLS/HR
52,899.0
366,182.0
38,986.6
28,409.6
19.8
30.7
486,527.7
MOLS/HR
8,700.3
3,072.3
89.5
11,862.1
59,727
34,279
813,648
296,275
562
210
530
1,542,582
77,045
2,824,858
10
LB/HR
404,822
90,403
3,050
498,275
14
LB/HR
14,352
47,275
61,627
3,723.6
17,003.3
29,048.5
6,732.0
16.5
3-5
31.1
55,052.9
4,275.5
115,886.9
MOLS/HR
9,198.4
5,016.8
89.5
14, 304.7
MOLS/HR
448.5
1,687.2
2,135.7
58,939
33,827
802,917
292,367
555
210
523
1,522,239
76,029
2,787,606
11
LB/HR
380,189
77,836
26,8l8
484,843
15
LB/HR
72
1,306
47,275
314
48,967
3,674.5
16,779-1
28,665.4
6,643.2
16.3
3.5
30.7
54,326.9
4 21Q 1
T ,d-l-^ . J_
114,358.7
MOLS/HR
»
8,638.7
4,319.4
7«£ Q
(Q'o^y
13,745-0
MOLS/HR
4.0
4o.8
1,687.2
4 9
1,736.9
-------
TABLE 11-13 (Cont'd)
MATERIAL BALANCE FOR U-GAS/LOW STEAM/CONOCO
Stream
16
17
18
M
O
19
Comp.
H20
co2
S
TOTAL
M.W.
18.02
44.01
32.06
Stream
Comp. M.W.
MgCOg-CaCOo
MgO-CaS
MgO • CaCOo
Inert
Solids
H20
TOTAL
184.1
112.46
l4o.4i
100.0
18.02
Stream
Comp.
MgO'CaS
MgO-CaCO.,
Inert
Solids
TOTAL
M.W.
112.46
i4o.4i
100.0
LB/HR
25,670
25,670
20
LB/HR
25,338
2,3^0
33,836
61, 514
24
LB/HR
12,168
4,142
16,310
MOLS/HR
800.7
800.7
MOLS/HR
137.7
23.4
1,877.7
2,038.8
MOLS/HR
108.2
29.5
137.7
LB/HR
37,283
37,283
21
LB/HR
608,634
206,894
117,050
935,578
MQLS/HR
2,069
2,069
MOLS/HR
5,412.0
1,473-5
1,170.5
8,056.0
LB/HR
2,663
35,455
38,118
22
LB/HR
699,231
93,780
117,050
910,061
MOLS/HR
147.8
805.6
953.4
MOLS/HR
6,217.6
667.9
1,170.5
8,056.0
LB/HR MOLS/HR
35,135 1,949.8
35,135 1,949.8
23
LB/HR MOLS/HR
25,338 137-7
2,34o 23.4
27,678 161.1
-------
SYSTEM FLOW DIAGRAM— BCFt/AIFt —BLOWN/ CONOCO
COAL- 700.000 LB/HR
TRFAM NO
' \s
MAKE-UP
DOLOMITE
-------
TABLE
NJ
-J
NJ
Stream
Cpnrp
H2
CO
co2
H2S
COS
NH3
N2
°2
H~0
M.W.
16. ou
2.016
28.01
tt.01
3U.08
60.07
17.03
28.02
32.00
18.02
Coal
Char
TOTAL
MATERIAL BALANCE FOR BCR/AIR BLOWN/CONOCO
LB/HR
MOLS/HR
700,000
TOO,000
LB/HR
1,1*69,803
10,051
1,926,062
MOLS/HR
52,U55-5
13,9^.0
557.8
66,957.3
LB/HR
MOLS/HR
120,960 6.712.5
120,960 6,712.5
-------
ro
~j
u>
MATERIAL BALANCE FOR SCR/AIR BLOWN/CONOCO
6
Comp
CH
E2k
CO
co2
H2S
COS
NHo
N2
H2°
TOTAL
Comp
°2
N2
co2
H20
so2
WO
H2S
TOTAL
Stream
M.W.
16. 04
2.016
28.01
44.01
34.08
6o.07
17.03
28.02
18.02
Stream
M.W.
32
28.02
44.01
18.02
64.02
30.01
34.08
LB/HR
1,355
717
19,809
4,788
—
—
182
33,697
957
61,505
LB/HR
3,282,240
10, 818, 214
—
—
—
—
—
14,100454
4
MOLS/HR
84.5
355.6
707.2
108.8
—
__
10.7
1,202.6
53.1
2,522.5
8
MOLS/HR
102,570
386,089
—
—
—
—
—
488,659
5
LB/HR
60,551
30,875
901,631
149,462
25,594
4,542
8,154
1,506,167
39,871
2,726,847
9
LB/HR
1,794,080
10,623,223
1,733,158
455,239
640
—
—
lit, 606, 340
MOLS/HR
3,775.0
15,314.9
32,189.6
3,396.1
751.0
75.6
478.8
53,753.3
2,212.6
111,946.9
MOLS/HR
56,065
37,913
39,381
25,263
10
—
—
499,849
LB/HR
60,551
32,043
885,396
214,038
324
150
8,154
1,506,167
42,789
2,749,612
LB/HR
—
409,350
91,4lU
—
3,084
503,848
10
MOLS/HR
LB/HR
MOLS/HR
3,775.0
15,894.5
31,6lO.O
4,863.4
9-5
2.5
478.8
53,753.3
2,374.5
112,761.5
MOLS/HR
9,301.3
5,072.9
90.5
23,766.0
59,196
31 , 326
865,587
209,250
324
150
7,972
1,472,471
41,832
2,688,108
LB/HR
384,440
78,707
27,118
490,265
3,690.5
15,538.9
30,902.8
4,754.6
9.5
2.5
468.1
52,550.7
2,321.4
110, 239. Q
11
MOLS/HR
8,735-3
4,367-7
795-7
13,898.7
-------
Stream
TABLE 11-lU (Cont'd)
MATERIAL BALANCE FOR BCR/AIR BLOWN/CONOCO
12 13 1U
15
Comp
co2
H20
H2S
°2
N2
S00
£.
TOTAL
M. W.
M.01
18.02
3U.08
32.0
28.02
6k. 06
Stresim
Conrp
H 0
co2
S
TOTAL
M. W.
18.02
UU.Ol
32.06
LB/HR
2,7^2
371
3,728
—
—
•V _
6,8la
16
LB/HR
—
25,959
25,959
MOLS/HR
62.3
20.6
109. u
—
—
___
192.3
MOLS/HR
—
809.7
809.7
LB/HR
387,182
55,981
3,08U
—
—
_«
UU6,2U7
17
LB/HR
37,703
—
—
37,703
MOLS/HR
8,797.6
3,106.6
90.5
—
—
_ _
11,99^.7
MOLS/HR
2,092.3
—
—
2,092.3
LB/HR
_
—
—
lU.509
^7,791
^.^
62,300
18
LB/HR
2,692
35,851
— .
38,5>*3
MOLS/HR
_
—
—
U53.U
l-,705-6
^^
2,159.0
MOLS/HR
1U9.U
8lU.6
—
96U.O
LB/HR
72
—
132
^7,791
31^
^8,309
19
LB/HR
35,526
—
—
35,526
MOLS/HR
U.o
—
Ul.2
1,705.6
U.9
1,755-7
MOLS/HR
1,971.5
—
—
1,971.5
-------
MATERIAL BALANCE FOR BCR/AIR BLOWN/CONOCO
K5
vj
Ul
Stream
Comp M.W.
MgCoo 'CaCOo 181+.01
MgO -CaS 112. h6
MgO -CaCo3 lUo.1+1
Inert Solids 100.0
H20 18.02
TOTAL
Stream
Comp M.W.
MgO 'CaS 112.16
MgO 'CaCo,, ll+O.l+l
Inert Solids 100.0
TOTAL
^u
LB/HR
25,6lU
—
—
2,370
34,211
62,195
_^ i
2U
LB/HR
12,303
U,18U
2,370
18,857
MOLS/HR
139-2
—
—
23.7
1,898.5
2,o6i.U
MOLS/HR
109-^
29-8
23.7
162.9
dL
LB/HR MOLS/HR
_
612,300 5,4l;i|.6
213,11^ 1,507.8
118,360 1,183.6
9U3,77lt
1,887, 5W 8,11+6.0
LB/HR MOLS/HR
^2
LB/HR MOLS/HR
._
703,910 6,259.2
98,736 703.2
118,360 1,183.6
—
921,006 8,iU6.0
LB/HR MQLS/HR
23
LB/HR MOLS/HR
25,6l4 139.2
2,370 23.7
162.9
LB/HR
MOLS/HR
-------
SYSTEM FLOW DIAGRAM- BCR/ OXYGEN-BLOWN/ CONOCO
COAL- 700.000 LB/HR
-:1RFAM NO
-------
MATERIAL BALANCE FOR BCR/OXYGEN BLOWN/CONOCO
Stream
Comp M.W.
CHU
H2
CO
co2
H2S
COS
NH3
N2
°2
H20
Coal
Char
16.01+
2.016
28.01
1+1+.01
3l+. 08
60.07
17.03
28.02
32.00
18.02
1
LE/HR
MOLS/HR
700,000
LB/HR
MOLS/HR
7,302 260.6
1408,627 12,769.6
LB/HR
1+18,320
MOLS/HR
LB/HR
23,2ll+.2 7,805
MOLS/HR
2,876
1,810
28,772
18,726
116
30
339
373
179.3
897.7
1,027-2
1*25.5
3.4
0.5
19.9
13.3
1+33.1
TOTAL
700,000
1+15,929 13,030.2 1+18,320 23,2ll+.2 60,81+7
2,999-9
-------
Is}
-J
00
TABLE 11-15 (Cont'd)
MATERIAL BALANCE FOR BCR/OXYGEN BLOWN/CONOCO
Stream
Comp M.W.
LB/HR
MOLS/HR
LB/HR
MOLS/HR
LB/HR
MOLS/HR
LB/HR
MOLS/HR
CH,
H2
CO
C02
H2S
COS
WE3
W2
H20
02
TOTAL
16. 04
2.016
28.01
44.01
34.08
60.07
17.03
28.02
18.02
32.00
72,531
44,51*9
741,108
412,356
26,255
4,620
8,561
9,426
194,380
1,513,786
Stream
Comp
CH1+
H2
CO
C02
H2S
COS
UH3
M.W.
16.04
2.016
28.01
44.01
34.08
60.07
17.03
LB/HR
1,740,860
W2 28.02 10,134,049
H20
02
TOTKL
18.02
32.00
737,883
2,o45,Uo8
lU ,65& ,200
14,521.9
22,097.6
26,458.7
9,369.6
770.4
76.9
502.7
336.4
10,786.9
74,921.1
9
MOLS/HR
39,556
361,672
ltO,948
63,919
506 ,095
72,531
45,652
725,776
472,373
2,887
697
8,561
9,476
196,872
1,534,775
10
LB/HR
377,351
2,81*2
84,267
H6H,U6o
4,521.9
22,645.0
25,911-3
10,733.3
84.7
11.6
502.7
336.4
10,925.2
75,672.1
MOLS/HR
8,574.2
83.4
14,676.3
13,333.9
69,655
43,843
697,004
453,646
2,771
667
8,222
9,053
189,068
1,473,929
LB/HR
35li}391
22,012
725,552
1,101,955
4,31*2.6
21,747.3
24,884.1
10,307.8
81.3
11.1
482.8,
323.1
10,492.1
72,672.2
11
MOLS/HR
8,052.5
733.5
4,026.2
12,812.2
11,692,634 417,296
3,547,520 110,860
15,240,154 528,156
12
LE/HR MOLS/HR
2,531 57-5
3,439 100.9
3l*2 19.0
6,312 177-4
-------
MATERIAL BALANCE FOR BCR/OXYGEN BLOWN/CONOCO
Stream
Comp
S02
S
H2
CO
co2
H2S
COS
KH3
N2
H20
02
TOTAL
M.W.
64.06
32.06
2.016
28.01
44.01
34.08
60.07
17.03
28.02
18.02
32.00
Stream
Comp
H2
CO
co2
H2S
COS
NHo
H2
H20
S
TOTAL
M.W.
16. 04
2.016
28.01
44.01
34.08
6o.07
17.03
28.02
18.02
32.06
13
LB/HR MOLS/HR
15
16
356,921 8,110.0
2,8U2 83.U
51,606 2,863.8
Ull.369 11,057.2
17
LB/HR MOLS/HR
3^,755 1,928.7
3^,755 1,928.7
LB/HR
33,052
2,483
35,535
MOLS/HR
LB/HR
295
MOLS/HR
U.6
751.0
137.8 32,762
.8 32,762
1,818.1
1,818.1
LB/HR
23,930
44,073
13,379
57,452
LB/HR
1,572.9
418.1
1,991.0
18
MOLS/HR
44,073
67
1,216
45,651
19
LB/HR
1,572.9
3-7
38.0
1,619.2
MOLS/HR
. 23,930
LB/HR
MOLS/HR
7U6.U
MOLS/HR
-------
TABLE 11-15 (cont'd)
MATERIAL BALANCE FOR BCR/OXYGEN BLOWN/CONOCO
Stream
20
21
22
23
10
00
Comp
Inert
Solids
H20
TOTAL
M.W.
18U.01
112. k6
lUo.Ul
100.0
18.02
LB/HR
23,627
2,180
31,51*9
57,356
MOLS/HR LB/HR MOLS/HR LB/HR
128.1*
567,383 5,01*5.2 651,81*1
192,867 1,373.6 87,1*19
21.8 109,120 1,091.2 109,120
1,750.8
1,901.0 869,370 7,510.0 81*8,380
MOLS/HR LB/HR MOLS/HR
23,627 128.1*
5,796.2
622.6
1,091.2 2,180 21.8
7,510.0 25,807 150.2
Stream 2k
Comp
Inert Solids
M.W.
112.1*6
lUo.Ui
100.0
LB/HR
11,3^7
3,861
2,180
MOLS/HR LB/HR MOLS/HR LB/HR
100.9
27-5
21.8
MOLS/HR LB/HR MOLS/HR
TOTAL
17,388
150.2
-------
to
oo
Steam - #/hr
1*50 psig (Sat)
100 psig (Sat)
Power - kW
Cooling Water-GPM
Cooling Duty MM Btu/hr
Export Heat
MM Btu
Temp - F
TABLE 11-16
UTILITY SUMMARY - BCR/CONOCO - AIR-BLOM
Coal
Prep. Gasifier
120,960
^,360 1,960
5,500
55.0
Acid Gas
Removal
35,1+25
1+70
1,225
12.3
62.1
300
Sulfur C0p
Recovery Supply
63,1*55
It, 260 9,975
6,720 5,800
67.2 58.0
1*9.8
350
Sour
Water
Stripper
8,400
186
65
0.7
Total
156,385
71,855
21,211
19,310
193
-------
TABLE 11-17
UTILITY SUMMARY - BCR/CONOCO - 0 - BLOWN
Steam - #/hr
U50 psig (SAT)
100 psig (SAT)
Power - KW
Cooling Water-GPM
Cooling Duty-MM Btu/hr
Export Heat -
MMBtu
Sour
Coal Acid Gas Sulfur C02 Water
Prep. Gasifier Removal Recovery Supply Stripper
Ul8,320 31,7^0
58,505 7,900
U,360 1,260 U35 3,930 9,200 175
5,500 1,130 6,200 5,350 60
55 11.3 62. 53.5 -6
Total
^50,060
66,1*00
19,360
18,2HO
182
57.2
Temp - F
300
350
-------
UTILITY SUMMARY - BCR/CONOCO - AIR-BLOWN
NO
CD
U)
Coal
Prep.
Acid Gas
Gasifier Removal
Steam - ?/hr
U50 psig (SAT)
100 psig (SAT)
Power - KW
Cooling Water - GPM
Cooling Duty - MM Btu/hr
Export Heat -
MM Btu/hr
Temperature - F
137,795
2,500 810
35,050
U65
1,210
12.1
300
Sulfur
Recovery
I* ,220
6,650
66.5
^9.2
350
co2
Supply
62,760
9,865
5,750
57.5
Sour
Water
Stripper
8,^00
185
65
0.6
Total
172, 8k
71,160
18, 0^5
13,675
137
NOTE: Based on 700,000 #/hr coal feed rate.
-------
a heat recovery boiler is used to drop the raw gas temperature by 100 F. The
net effect of the reaction in the absorber is endothermic. In addition, heat
leaks and the effect of the recirculated stone combine to cause an approximate-
ly 100 F drop in the gas temperature from inlet to outlet of the desulfurizer.
Following desulfurization, gas temperature is dropped to 1000 F by rais-
ing high-pressure steam in a waste heat boiler. The steam cycle arrangement
is the same as in the low-temperature systems; however, utility requirements
of the Conoco system differ somewhat.
High-pressure steam is needed to rehumdify the gas stream to the acceptor
regenerator. Low-pressure steam is needed in the C02 supply system which
removes C02 from the stack gas. The other need for low-pressure steam is in
the sour water stripper. Overall, the process exports heat to the power sys-
tem. Heat is available from the acid gas stream that must be cooled before
being fed to the liquid phase Glaus plant. Also, the waste heat from the
sulfur combustor is available.
In the earlier studies, C02 for use in acceptor regeneration and conver-
sion of the spent dolomite was removed at pressure from the cool transport gas.
In the present study, transport gas flow rate and C02 concentration are so
low as to preclude use of that stream for a CC>2 supply; therefore, a stack
gas bleed stream was selected for use. Less than 3 percent of the stack gas
is required so that the resultant change in mixed mean temperature is insig-
nificant when the bleed stream is returned to the stack.
Performance of the high-temperature systems, as expected, showed a some-
what higher overall efficiency than the low-temperature counterparts. However,
the advantage was markedly reduced over previous studies where as much as a
six point differential had been noted. The comparison between air- and oxygen-
blown operation seems to be relatively unaffected by the cleanup system. A
slight advantage for the oxygen-blown gasifier with Conoco cleanup was noted
due apparently to the increased water vapor in the raw gas from the oxygen-
blown gasifier.
The reduced differential between high- and low-temperature desulfuriza-
tion is of special interest because it results from operation of the gasifier
at conditions favorable to combined-cycle performance. Operation under low-
steam feed rates has yet to be demonstrated and there is some concern over the
ability to control reactor conditions without large quantities of excess steam-
Of importance, however, is the fact that there is more than one way to achieve
improved overall power plant generating efficiency.
SYSTEM COST ESTIMATES
The following paragraphs discuss the costs of the eight integrated sys-
tems and the means by which they were determined. All capital costs are in
mid-1977 dollars. Where necessary, costs have been escalated to that time by
established procedures (References 11-2, 11-3, and 11-4). In determining the
cost of electricity, a relatively simple procedure was used. Yearly capital
charges are taken to be 17 percent of the total investment cost. A load
284
-------
of 70 percent was assumed and coal cost taken to be $1.00/10° Btu.
°ased upon prior studies, yearly maintenance. and operating costs were esti-
Nated to be 3.5 percent of the combined-cycle power system cost and 8.5
Percent of the fuel processing system cost.
Sources of cost data are discussed below but, in general, gasifier and
processing costs for systems with Selexol cleanup were prepared by Fluor
^ngineers and Constructors; Selexol costs by Allied Chemical; and power sys-
tems costs by United Technologies Research Center. Costs for the molten salt
sYstem were developed previously at Pullman Kellogg (Reference 11-5) and
*°r the Conoco desulfurizat ion process by Conoco (Reference 11-6) under con-
tract to the EPA.
A summary of fuel processing system costs is given in Table 11-19 for
low-temperature and in Table 11-20 for high-temperature cleanup systems. Power
system costs are given in Table 11-21. While total costs are on the same
.asis, individual costs for the various parts of the fuel systems include
costs for engineering and supervision during construction but do not
any funds for contingency. For the power systems, indirect costs are
n°t included in the individual items.
Contingencies have been applied to each group of costs. These vary based
°n judgment of the overall development status of the equipment. For the power
^sterns a contingency of 15 percent was used while the factor was higher for
^e fuel systems, 21 percent for low- and 25 percent for high-temperature
sys tems.
Process ing Systems with Selexol Desulfurization
For this group of systems, cost estimates were prepared by Fluor. They
based on data developed previously under contract to EPRI and adjusted to
e parameters of this study. An example of the costing basis is given in
Ppendix B for the air-blown BCR system. The data provided by Fluor covered
*l major fuel processing equipment with the exception of the Selexol system.
The cost of that unit was estimated by the developer, Allied Chemical,
scaled as a function of gas flow rates. The costs appear to be consistent
data that have subsequently been obtained from Fluor. Other costs
9ssigned to the fuel processing section include the bleed-air cooling heat
e*changer and boost compressor, an incremental cost for cooling water (cooling
ower pius circulating system) and boiler make-up water treatment facilities
^°r treating gasifier steam), and "other support facilities". This last cat-
e8ory includes waste water treatment, gas flaring, instrument air, etc. For
oxygen-blown system, an auxiliary compressor was included for startup of
e air separation plant.
Salt System Costs
For the molten salt system, cost information was provided by Pullman
gg (Reference 11-5). The estimate is considered to be a rough order of
nitude number and required escalation and scaling to be comparable with
285
-------
TABLE 11-19
FUEL PROCESSING COST SUMMARY
LOW-TEMPERATURE CLEANUP
Coal Handling
Air Separation
Oxidant Feed & Cooling
Gasification & Ash Handling
Gas Cooling & Scrub
Acid Gas Removal
Sulfur Rec. & Tail Gas Cleanup
Sour Water Treating
Cooling & Make-up Water
Other Support Facilities
Aux. Og Plant Compressor
Subtotal
Contingency
Escalation During Constr.
Subtotal
Interest During Constr,
Total Cost
Costs in Thousands of Dollars - Mid 1977
Unit Cost - $/kW 26k
U-Gas
EPRI
16
1*
30
53
29
10
1
12
8
166
35
27
229
53
283
Data
,725
,2Ul
,699
,005
,000
,891*
,853
,06U
,1*00
,881
,01*5
,986
,912
,31*0
,252
U-Gas
Low Stm
16
1*
30
,725
,01*2
,699
53,005
27
10
1
5
8
158
33
26
218
50
269
,500
,891*
,853
,51*7
,1*00
,665
,320
,608
,593
,71*
,307
BCR
BCR
Air
28
3
1*0
31
26
10
7
1*
10
163
31*
27
225
52
277
,287
,861
,353
,oi*o
,600
,891*
,377
,66l
,1*00
,1*73
,329
,im*
,216
,250
,1*66
28
1*1
7
35
19
16
10
7
11
10
2
188
39
31
260
60
320
°2
,287
,155
,622
,oi*o
,605
,600
,89^
,377
,027
,1*00
,21*5
,752
,638
,651*
,01*1*
,330
,37^
Molten
Salt
*137,
10,
_
2,
8,
159,
33,
26,
219,
50,
900
260
1*68
1*00
028
396
717
11*1
81*1
269,982
21*3
253
307
283
* Total cost for gasifier and cleanup system
286
-------
TABLE 11-20
^Handling
** Separation
**ant Feed and Cooling
asification & Ash Handling
Cooling
Gas Removal
culate Removal
^fur Recovery
°2 Supply
Sing & Makeup Water
V Support Facilities
^Uy ~
^' 02 Plant Compressor
Subtotal
FUEL PROCESSING COST SUMMARY
HIGH-TEMPERATURE CLEANUP
Constr.( -168)
Subtotal
est During Const. (.232)
Total Cost
ts in thousands of dollars
Cost - $/kW
U-Gas
Low Stm
16,725
U,oU2
30,699
9,720
l6,UUU
8,529
12,279
3,676
U,UU9
8,1*00
BCR-Air
28,287
3,861
1*0,353
10,160
16, W*
8,313
12,279
3,676
H,l*5l
10,UOO
BCR-02
28,287
7^622
35,OliO
7,700
16, hkk
6,210
12,279
3,676
10,516
io,Uoo
39,15U
20T>92°
- Mid 1977
182
138.22U
23,222
2°2'913
U7,0T6
2U9'989
220
181,57*
30,5oU
6!,8UO
299
287
-------
TABLE 11-21
POWER SYSTEMS COST SUMMARY
Item
3l+l Site and Structures
3l*3 Gas Turbine
3l*l* Gas Generator
312 Boiler Plant
3ll+ Steam Turbogenerator
3l+5 & 353 Ace. Elect. Equip.
3l+6 Misc. Pwr Plant Equip.
Adj. Direct Const. Costs
Other Expenses
Direct Const. Costs
Indirects
Escalation
Total Investment Cost
Interest During Const.
Grand Total
U-Gas
Selexol
10,577
57,763
15,289
72,165
37,51+1
I8,5l>7
559
212,1+ Itl
18,589
231,030
80,861
35,690
31+7,581
80,639
1*28,220
U-Gas
Selex
(Low Stm)
10,771
58,871*
15,618
75,698
38,029
19,187
568
2l8,7l*5
19,11*0
237,885
83,260
36,71*9
357,891*
83,031
1+1+0,925
BCR-Air
Selexol
10,702
59,831
15,903
77,609
36,876
18,853
563
220,337
19,279
239,6l6
83,866
37,017
360,1*99
83,636
HI* ,135
BCR-02
Selexol
10,UU5
55,396
U,l*56
82,01+6
36,51+1*
17,523
5Ul
216,951
18,983
235,931*
82,577
36,1*U8
35^,959
82,350
1*37,309
Molten
Salt
9,1*87
51,571
13,556
63,1+91+
37,735
13,169
528
189,51*0
16,585
206,125
72..11*1*
31,1+6U
309,733
71,858
381,591
U-Gas
Conoco
Low-Stm
10,825
60,086
16,126
60,062
38,01+3
19,1+63
573
205,178
17,953
223,131
78,096
3l+, 1+70
335,697
77,882
1*13,579
BCR-Air
Conoco
10,838
61,062
16,298
66,953
37,360
19,391
571
212,1*73
18,591
231,061+
80,872
35,696
31+7,632
80,651
1+28,283
BCR-02
Conoco
10,620
57,698
15,202
81+, 058
36,957
18,299
55^
223,388
19,51+7
2^2,935
85,027
37,529
36 5, >491
81*, 791*
1+50,285
Costs are in thousands of dollars - Mid 1977
Unit cost - $/kW 399 398
1+01+
1+00
361
377
1*10
-------
°ther data. Therefore, the degree of confidence in the data is low. However,
the cost in $/kW is quite consistent with the other low-temperature systems.
Those items not covered in the estimate provided by Pullman Kellogg were
Scaled on a consistent basis with the other low-temperature systems.
cjjgh^ Temperature Fuel Processing System Costs
For these systems, coal handling and gasification costs are the same as
°r the low-temperature systems. Conoco costs were obtained from data pre-
sented in Reference 11-6. However, because gas cooling requirements differ
Slgnif icant ly, that equipment was removed from the Conoco process and treated
SeParately. Costs were assumed to vary with the 0.7 power for sulfur removal,
8ulfur recovery and make-up C02 sections. Data contained in Reference 11-7
y Stone and Webster were helpful in estimating costs of hot gas handling
6cluipment, i.e., the high-pressure boiler and particulate filter. Inform-
ation contained there correlated well with other sources (References 11-8 and
*l~9) and was used for those estimates.
While the degree of confidence in the costs of the three high-temperature
systems is not high, they are valuable for comparative purposes. The U-Gas
ess shows a cost advantage in both coal preparation and gasification over
BCR air-blown system. These two areas account for most of the difference
the two. In comparing the air- and oxygen-blown BCR systems, all
c°sts are similar with the exception of the air-separation plant. In the
low-temperature comparison, the cost of air-separation was offset in large
"^asure by reduced gas cooling and desulfurizat ion costs. Here, desulfuriza-
I0n costs were based directly on sulfur removal rate as the desulfurizat ion
ess is not aided by increased partial pressure of H2§ and the major
of the Conoco system is devoted to regeneration of the dolomite. Thus,
e capital cost and overall generating cost for the high-temperature oxygen-
lo system does not show any significant advantage over the low-temperature
System Costs
Power system costs are relatively constant on a per kilowatt basis.
hey are summarized in Table 11-21. Using the air-blown BCR Selexol system
a® an example, the costing breakdown is presented in Appendix C. As men-
i in the discussion of performance, the use of a fixed steam cycle (i.e.,
5 steam cycle whose operating conditions have not been optimized for this
Particular application) tends to increase the size of the gas turbine exhaust
Waste heat boiler in the oxygen-blown systems. Also, because it was neces-
Sary to use the gas turbine exhaust waste heat boiler for feedwater heating in
5 e two air-blown Conoco systems, temperature differences were increased and
.i-ler costs reduced significantly. This suggests that some further optimiza-
is possible, but it is doubtful that the resulting improvement would
y change the study's conclusions.
289
-------
The power systems presented in this report differ significantly as
compared to those of earlier studies (References 11-1 and 11-10). The changes
are due to the decision to use a moderate (18:1) pressure ratio gas turbine so
that a 2400 psi reheat steam cycle could be utilized. This results in im-
proved cycle efficiency and an overall reduction in generating costs, despite
the fact that the steam generation fraction is increased. This has been dis-
cussed previously in Section 7. Turbogenerator costs, as compared to Ref-
erence 11-10, increase but not out of proportion to the increased output.
In fact, the unit price is slightly less after taking into account escalation
and the reduction in cost due to increased size. However, waste heat boiler
sizes are up by 50 to 60 percent and when combined with revised unit cost data
that show a higher than average escalation rate, waste heat boiler costs have
increased approximately by a factor of two. The overall effect is a slight
increase in unit cost of the equipment associated with steam power generation.
Unfortunately, the picture is clouded by the fact that some of the steam
associated costs are assigned to the gasification and cleanup portion of the
system. These are primarily the costs of steam generation in the gasifier,
the fuel gas heat recovery boiler, and boost compressor cooling boiler. If
these were included, the unit cost of the steam power generation would prob-
ably decrease, or at worst remain constant, because of the improved efficiency
achieved with a reheat cycle.
Gas turbine manufacturing costs appear to follow overall escalation
factors quite closely. Other considerations, however, indicate that current
mark-up factors may be insufficient when applied to advanced, high-temperature,
high-performance engines. These improvements generally tend to reduce or at
least not increase the unit manufacturing cost. On the other hand the devel-
opment costs needed to produce the improvements are sharply increased. The
number of units over which these costs can be spread is one of the major
uncertainties. Also, the availability of government funding for development
of these improvements is questionable. Because of these uncertainties, recent
studies have indicated that estimates of gas turbine prices (based on the
availability of advanced technology) should be increased by some 70 percent
over those used in previous studies although the inflation rate for this area
has been approximately 15 percent.
Current conditions are such that it is prudent to raise the total of
indirect costs from 23 to 30 percent. These consist of a contingency allow-
ance plus engineering and construction supervision costs. The overall result
is a relatively large increase in the unit cost of the power system. However,
it is balanced by an improved system efficiency and the resulting higher out-
put tends to reduce the unit cost of fuel processing.
290
-------
REFERENCES
ll~l Robson, F. L., A. J. Giramonti, W. A. Blecher and G. Mazzella. Fuel
Gas Environmental Impact - Phase Report, EPA 600/2-75-078, November
1975.
I-I--2 Handy-Whitman Index of Public Utility Construction Costs, Bulletin 102.
Whitman, Reguardt, and Associates, Baltimore, Maryland.
H-3 Economic Indicators. Bi-Weekly Issues of Chemical Engineering.
ll~4 Nelson Cost Indexes of Refinery Construction. Oil and Gas Journal
Issues.
1~5 Private Communication. Pullman Kellogg Division of Pullman Inc.,
April 1976.
^"•6 Curran, G. P., B. J. Koch, B. Paske, M. Pell and E. Gorin: High-
Temperature Desulfurization of Low-Btu Gas.EPA-600/7-77-031,
(NTIS No. PB271-008), April 1977.
*""7 Jones, C. H. and J. M. Donahue: Comparative Evaluation of High and
Low Temperature Gas Cleaning for Coal Gasification - Combined Cycle
Power Systems. EPRI AF-416, April 1977.
1-8 Private Communication. Combustion Power Co.,May 1977.
^"9 Guthrie, K. M.: Process Plant Estimating, Evaluation and Control.
Craftsman Book Co. of America, 1974.
*""10 Robson, F. L. , W. A. Blecher and C. B. Colton: Fuel Gas Environmental
Impact. EPA-600/2-76-153, June 1976.
291
-------
APPENDIX A
WATER TREATMENT AND REUSE
The following appendix presents a discussion of the water requirements,
treatment and reuse for the integrated power plants described in the current
study.
WATER REUSE FOR COGAS SYSTEMS
High water requirements for coal conversion plants impose certain con-
straints on water utilization. The water use and reuse system should be
designed to make maximum utilization of scarce water without damaging
the environment. The water for the plant will be provided from the reservoir,
which will be withdrawn from a local source. The water supply system will be
capable of delivering sufficient water to match the plant requirements. The
commercial plant will probably have to meet zero discharge criteria by the
year 1985. None of the water will then be returned to any stream or stream
bed. The COGAS plant water reuse system is shown in Figure A-l for the
BCR/Air-Blown/Selexol, BCR/Oxygen-Blown/Selexol, and U-Gas/Air-Blown/Selexol
processes. The flowsheet is greatly simplified and is intended to show the
overall plant water balance or overall recirculating systems. The water
balances for the above three processes are given in Tables A-l to 3. The
numbers represent combined flow rates of cooling water for condenser and
gasification process units. Cooling water requirements for the gasification
process units are given separately at the bottom of each table. The cooling
water requirements for BCR and IGT processes employing Conoco cleanup are
shown in Tables A-4 to 6. The cooling water requirements for the gasification
units, acid gas removal, sulfur recovery, carbon dioxide supply, and sour
water stripper are shown at the bottom of each table.
The processes will use two separate cooling tower systems. The first
system of cooling towers will supply water to the surface condenser. The sec-
ond system will supply cooling water to the gasification system. The make-
up water requirements for the BCR/Air-Blown/Conoco, BCR/Oxygen-Blown/Conoco
and IGT/Air-Blown/Conoco combined cooling systems are about 6,460, 6,150, and
292
-------
COMBINED CYCLE-POWER PLANT - WATER TREATMENT AND REUSE
10
00
k
CO
RAW WATER STORAGE
6243.7
5234.1 (j)
6232.2
RROSION INHIBITOR COOLING TOWER
DEFLOCULANT
CHLORINE
DRIFT EVAP
3 53 .4
303.2 (R) (U)
346.3
6183.7 MAKE-UP " '
5306.9 COOLING/—X
6060.8 WATER QJ/ C
RbStRVOIR . - .....
828.6
S
DOLING
YSTEM
rTo\ 1146.2
888.6 ^ ~ 13S1'9
1533.3
PHOSPHATE
HYDRAZINE
AMMONIA
848.3
1025.4 (5)
1463.3
BOILER FEED
353,351
303 254 (~~\ CIRCULATING
346',333 ^ WATER
ORATION
5300.3
4548-8
5195.0
RECYCLE WATER TO COOLING TOWER
(T|) 530.0
454.9
519.5
COOLING TOWER °'"
' »- 436.7
SLOWDOWN SETTLING 493.2
cvcTPE7ER BOILER BLOWDOWN ®
COGAS 252 ^
... . c-rn (c
WATER 731
20,1 (3)! 20.1 ](T
24.0 ^>. 24.0 JL^
35 0 / ^ 35.0 r ^
f .,, , , . CC
^ UJ UJ
0° 20
2z 02
* < -r * OcAt-RATOR * Z _•
00
>
!
-------
TABLE A-l
WATER BALANCES FOR THE BCR/AIR-BLOWN/SELEXOL
PROCESS OVERALL WATER BALANCE (SEE FIG. A-l)
Stream
No. Description Flows, Ib/hr,
1 Raw water 3,120,600
2 Raw water to demineralizer UUU,l20
3 Acid regenerant and rinse water 10,050
U Alkaline regenerant and rinse water 10,05°
5 Boiler feed water to process U23,98°
6 Demineralizer blowdown 20,09°
7 Water to slag handling 28,8UO
8 Water associated with slag disposal 28,8UO
9 Make-up water to cooling tower from stripping 125,95°
10 Boiler blowdown 32,0^0
11 Make-up cooling water 3,090,610
12 Cooling tower circulating water 176,60^,880
13 Drift losses from cooling water 176,630
lU Evaporation losses from cooling tower 2,6^9,09°
15 Cooling tower blowdown 26^,89°
16 Make-up water to cooling tower from settling tank 256,15°
17 Settling tank blowdown to slag handling 8,75°
18 Water from neutralization tank to slag handling 20,100
19 Recycle water to cooling tower as make-up water
Cooling Water Requirements for Process Units
1. Gasification 2,739,^°°
2. Gas scrubbing 1,265,99°
3. Acid gas removal 13,879,^5°
U. Ammonia recovery 323,87
5. Sulfur recovery 62,9°°
294
-------
TABLE A-2
1
2
3
\
5
6
T
8
9
10
U
la
13
11*
is
16
1?
18
19
WATER BALANCES FOR THE BCR/OXYGEN-BLOWN/SELEXOL
PROCESS OVERALL WATER BALANCE (SEE FIG. A-l)
Description
Raw water
Raw water to demineralizer
Acid regenerant and rinse water
Alkaline regenerant and rinse water
Boiler feed water to process
Demineralizer blowdown
Water to slag handling
Water associated with slag disposal
Make-up water to cooling tower from stripping
Boiler blowdown
Make-up cooling water
Cooling tower circulating water
Drift losses from cooling tower
Evaporation losses from cooling tower
Cooling tower blowdown
Make-up water to cooling tower from settling tank
Settling tank blowdown to slag handling
Water from neutralization tank to slag handling
Recycle water to cooling tower as make-up water
Flow Ib/hr
2,616,000
536,1*90
12,000
512,600
23,990
83,090
83,090
281*,890
31,81*0
2,652,1*10
151,566,350
151,51*0
227,360
2,273,1*90
218,260
211,650
9,100
23,990
572,870
Water Requirements for Process Units
Acid gas removal
Ammonia recovery
Sulfur recovery
7,892,81*0
ll*3,9l*0
62,980
295
-------
TABLE A-3
WATER BALANCES FOR THE IGT/AIR-BLOWN/SELEXOL
PROCESS OVERALL WATER BALANCE (SEE FIG. A-l)
Stream
Mo. Description Flow Ib/hr
1 Raw water 3,11^,850
2 Raw water to demineralizer 766,3^0
3 Acid regenerant and rinse water 17,^90
U Alkaline regenerant and rinse water 17,^90
5 Boiler feed water to process 731,360
6 Demineralizer blowdown 3^,990
7 Water to slag handling U5,380
8 Water associated with slag disposal (5U$ solids) ^5,380
9 Make-up water to cooling tower from stripping 365,350
10 Boiler blowdown 66,320
11 Make-up cooling water 3,029,200
12 Cooling tower circulating water 173,097,230
13 Drift losses from cooling tower 173,080
lU Evaporation losses from cooling tower 2,596,1*60
15 Cooling tower blowdown 259,660
16 Make-up water to cooling tower from settling tank 2^9,000
17 Settling tank blowdown to slag handling 10,^00
18 Water from neutralization tank to slag handling 3^,990
19 Recycle water to cooling tower as make-up water 680,680
Cooling Water Requirements for Process Units
1. Gas scrubbing 1,780,290
2. Acid gas removal 151,139,950
3. Ammonia recovery l;02,8Uo
296
-------
TABLE A-U
COOLING WATER REQUIREMENTS FOR
BCR/AIR-BLOWN/CONOCO PROCESS
Description Flow, Ib/hr
Her feed waters to demineralizers 2^2 820
^e-up cooling water 3,229,'300
°°ling tower circulating water 181; 531 iUo
VaPoration losses from cooling tower 2,767 970
°°ling tower Slowdown 276 800
lft losses from cooling tower ^.Qh 530
Water Requirements for process Gasification Units
removal 612,260
Q Ur removal 3,358,660
o2 supply . 2,898,8^0
^ water stripper 32,^90
2,7^8,900
297
-------
TABLE A-5
COOLING WATER REQUIREMENTS FOR
BCR/OXYGEN-BLOWN/CONOCO PROCESS
Description Flows. Ib/hr
Boiler feed water to demineralizers 570,010
Make-up cooling water 3,073,130
Cooling tower circulating water 175,607»^5°
Evaporation losses from cooling tower 2,63^,110
Cooling tower "blowdown 263,^10
Drift losses from cooling tower 175j6l<->
Cooling Water Requirements for Process Gasification Units
Acid gas removal 56^,77°
Sulfur recovery 3,098,760
C02 supply 2,673,930
Sour water stripper 29,99°
298
-------
TABLE A-6
COOLING WATER REQUIREMENTS FOR
IGT/AIR-BLOWN/CONOCO PROCESS
Description Flows, lb/hr
feed water to demineralizers 289 093
cooling water 3, 336, ^ 57
"°oling tower circulating water 190,65^,765
losses from cooling tower 2 859 821
9 •/ f 9
tower blowdown 285,982
losses from cooling tower 190, 65)+
Water Requirements for Process Gasification Units
gas removal 604,758
Sulfur recovery 3,323,670
2,873,850
water stripper 32,U87
299
-------
6,675 gpm, respectively. The make-up water to the cooling tower for gasifi-
cation systems for the above three processes is about 340, 220, and 240 gpm,
respectively. The total cooling water supply (recirculating water) for the
gasification systems for each process is BCR/Air-Blown/Conoco, 13,675;
BCR/Oxygen-Blown/Conoco, 12,740; and IGT/Air-Blown/Conoco, 19,310 gpm.
The molten salt process, which is different from the above processes,
requires a cooling water supply to the power system condenser of about
243,750 gpm. A separate cooling tower supplies 45,700 gpm of cooling water
to the gasification system and off-site facilities.
The BCR/Air-Blown/Selexol process and the water system associated with
it are described below. The other two processes, BCR/Oxygen-Blown/Selexol
and IGT/Air-Blown/Selexol, operate similarly with the exception of differing
flow-rate values. Water will enter the raw storage pond at 7,810 gpm. The
lined reservoir will hold 170 million gallons, about a two-week holdup. The
process will require 6,240 gpm, and approximately 20 percent of the total
inflow to raw storage pond, 1,250 gpm, will consist of auxiliary water
requirements such as mine and offsite needs, and storage pond evaporation
losses.
Boiler Feedwater
The process water from the raw storage ponds is broken into two major
streams. One of the streams, having a flowrate of about 889 gpm, flows to the
demineralizer; the second goes to the cooling towers. The demineralized
water is used for generating the necessary steam for the process. Steam is
utilized in the gasifier to supply hydrogen and is used in the ammonia
recovery unit for the tower reboilers. Sour water strippers require a steam
flow of about 1.0 to 1.2 Ib per gallon of sour water entering the unit. About
4.5 percent of the water entering the demineralizers is used as regenerant and
rinse water. A high steam flow rate is required to drive the steam turbines
of the COGAS system. However, most of the steam is condensed and recycled to
the boilers. Only 64 gpm of fresh water will be needed to make-up the boiler
blowdown. Modern, high-pressure boilers have blowdown rates of 0.1 to 1.0
percent of the total steam generated.
Cooling Water
Most of the cooling water make-up is the fresh water from the reservoir.
Of 6,184 gpm water required for the cooling tower make-up, 5,355 gpm comes
from the reservoir. This water will be passed through cold lime treating if
the total dissolved solids concentration is high. Since lime treatment for
the water is not essential in this study, it is not included in the flowsheets.
The rest of the total cooling tower make-up water, about 13 percent, is the
water recovered and recycled from the process itself.
Raw gas from the gasifiers will contain unreacted steam, part of which
originated from the moisture in the coal and part of which was input for
300
-------
gasificaton. The raw gas is free of any tars and oils, but may contain trace
amounts of organics. The gasifier raw gas will be cooled in a heat recovery
system and further scrubbed with water to remove ammonia.
The scrubber water will then be stripped to recover ammonia. Part of this
treated water is reused in the scrubber and the rest is passed to the cooling
tower. Organics that may be present will be degraded in the cooling tower.
fhe water from the stripping section accounts for 4 percent of the total
make-up.
The second reuse system in the process consists of the boiler and
cooling tower blowdown treatment. Blowdown from all boilers in the plant
will be collected and reused as make-up to the cooling tower. All the
Boilers in the plant will run to a limit of the allowable concentration of
silica. Because cooling towers can operate at a higher silica limit than
Boilers, blowdown from the boilers can be used in the cooling towers. The
cooling tower blowdown water will have a high concentration of dissolved and
suspended solids. The suspended solids are removed in a settling tank. When
the excess water from stripping section and boiler blowdown water is combined
with the cooling tower blowdown, the total dissolved solids concentration of
the combined recycle water is less than cooling tower blowdown alone, and can
^e used in the cooling tower without treatment.
Cooling water is required for removing excess heat from process units
s"ch as acid gas removal, sulfur recovery and ammonia recovery. A temperature
rise of about 15 to 20 F is allowed for the water circulating through the
heat exchangers. The heated water is cooled in the cooling towers and then
recirculated in the heat exchangers. A separate cooling tower will be used
*°r this purpose.
jjgg Handling System
The coal entering the gasifiers will be about 8.7 percent ash. This
^on-reactive ash will go through the gasifier and leave in the form of
The slag, at a high temperature, is cooled by water sprayed at
bottom of the gasifier. The cooled slag is further quenched in a
quench tank. Rinse water from the demineralizers will be treated
the neutralization tank. This stream and the blowdown from the settling
can be used to quench the slag. The slag-water mixture can be further
c°ncentrated in the clarifier. The water recovered will then be recycled to
^e quench tank.
S°UR WATER STRIPPING
The raw gas from the gasifier is scrubbed in a water scrubber. The
Scrubber may be packed-bed tower or a tray tower. Almost all ammonia and
^a*t of the H2S and C02 are absorbed in the water. Phenols, cyanides and
°r8anics may be present in the raw gas and also will appear in the sour
^ter. Ammonia, a valuable by-product, is recovered by stripping. Since
Ammonia stripping is an integral part of the gasification process, two
c°nmionly used stripping processes, the two-stage All-Distillation Process and
^e Phosam-W Process, are described below.
301
-------
Sour Water Strippers
Sour water stripping is used to remove hydrogen sulfide and ammonia
from sour waters. In the gasification plants, ammonia is mostly neutralized
by carbon dioxide. Ammonia must be removed from the sour water to a level
low enough to conform to the effluent regulations for wastewater. It should
be collected separately because it interferes with the Glaus sulfur recovery
process. Since sulfur is a saleable commodity, it should be collected in a
salable form. H2S cannot be vented to the atmosphere; instead, it must be
collected and sent to the sulfur recovery plant. Sour water may also contain
organic acids, sulfide oxidation products, and phenols and cyanides, some of
which interfere with ammonia and hydrogen sulfide separation. Phenols, if
present in high concentrations, are removed by solvent extraction prior to
stripping. However, phenols are not present in the raw gas of the gasifica-
tion processes considered in this study. If present, their concentration
will be low.
A conventional sour water stipper is shown in Figure A-2. In general,
the stripper contains 10 to 20 actual trays. In some cases, packed towers
are employed. Heat integration is common between the feed and bottom stream,
and the bottoms are either reboiled or stripped with live steam at a rate
usually in the range 10-24 Ib steam/100 Ib water. The stripper overhead is
condensed to produce an off-gas suitable for further treatment in a sulfur
plant and a reflux which is either returned to the top tray in the tower or
the feed drum. Because of its low solubility, hydrogen sulfide can be
stripped more readily than ammonia. A temperature of 230°F will remove
more than 90 percent of ammonia. However, 90 percent of hydrogen sulfide
removal requires a temperature of only 100°F. ^ In a single stage (tower)
sour water stripping operation, ammonia must be separated from the C02 and
H2S vapors. This may be done by absorbing the ammonia in sulfuric acid,
which will produce crystalline ammonium sulfate for sale. Another procedure
is to absorb the ammonia in phosphoric acid or ammonium phosphate.
Two-Stage All-Distillation Process
•
This proprietary process, developed by Chevron Research, uses two
stripping columns for the recovery of high-purity ammonia and hydrogen
sulfide from sour waters as separate by-products. The process has been in
successful commercial use since 1966 and a number of commercial installations
are in operation.
The process flow diagram is shown in Figure A-3. The stripped water from
the process is 99.9 percent water by weight with less than 50 ppm ammonia and
10 ppm H2S. The H2S stream is suitable for feed to plants producing
sulfur or sulfuric acid without further treatment. High purity ammonia can
be produced in either aqueous or anhydrous liquid form, depending on the
desired end-use market.
A typical performance feed rate of 134 gpm is given below (the numbers
represent the stream numbers on Figure A-3):
302
-------
CONVENTIONAL SOUR WATER STRIPPING PROCESS
FIG. AU2
SOUR FLASH
GAS
SOUR
WATER
(RAW
PEED)
SOUR GAS
i
REFLUX AND
FEED SURGE
COOLING WATER
AIR-COOLED
CONDENSER
STRIPPED WATER
ok)
FEED
STRIPPER
STEAM
BOTTOMS
78-02-138-7
303
-------
TWO STAGE ALL-DISTILLATION PROCESS
HYDROGEN SULFIDE
00
I
o
M
\
FOUL WATER
SULFIDE
STRIPPER
DEGASSER
SURGE
TANK
LJd
COOLING WATER
(ACCUMULATOR)
EN-
3
I
X
I
ST
"<
»-
••Mil
WIMO
5TRIP
EAM
1
^
IMIA
3ER
(-
STEA
3J
x-s.
», 1
O
HYDROGEN-SULFIDE/AMMONIA RECYCLE
AMMONIA
SCRUBBERS
LIQUID
AMMONIA
COMPRESSOR
STRIPPED WATER
Jj
CO
-------
Feed (Stream-1)
Feed rate
Dissolved material, by weight
Dissolved ammonia, by weight
Dissolved t^S, by weight
Dissolved hydrocarbons
j>t ripped H7S (Stream-2)
Generation rate
H2S by weight
H20 by weight
NH3 by weight
Stripped NH^ (Stream-3)
Generation rate
NH3 by weight
H20 by weight
H2S by weight
^tripped Water (Stream-4)
H20 by weight
H2S by weight
NH3 by weight
Flash Gas (Stream-5)
Contains light hydrocarbons like CH2,
Heat Requirements
-W Process
134 gpm
3.8 percent
3-5.5 percent
4-5.5 percent
20.6 TPD
99.9 percent
0.1 percent
30 ppm
10.3 TPD
99.9 percent
0.1 percent
1 ppm
99.9 percent +
10 ppm
50 ppm
3.9 x iQ5 Btu
The Phosam-W process, developed by U.S. Steel, removes ammonia from
~WaterS Containin8 acid 8ases by using phosphoric acid as a stripping
rather than the commonly used sulfuric acid. Phosphoric acid, with
bound hydrogen atoms, can bind ammonium ions tightly enough for absorp-
, but loosely enough for ammonia recovery via stripping.
Figure A-4 is a flowsheet of the Phosam-W process. Feed, consisting of
e with dissolved ammonia and acid gas is preheated in two streams, and
ombined to enter the stripping section of the superstill. The low-pressure
si8^ striPping section removes both ammonia and acid gases from the
r. The superstill bottoms, with an ammonia content of less than 200
W s recvce to the preheater to heat part of the wastewater feed.
irect steam, at 60 psig, provides most of the energy input to the super-
1U stripper.
305
-------
PHOSAM-W PROCESS FOR AMMONIA SEPARATION
OJ
o
FRACTION-
ATOR FEED
TANK
CD
O
10
\
*»
-------
Vapor from the superstill stripper section rises, entering the absorber
section of the superstill. The vapor is contacted with a spray of cool
ammonium phosphate solution. Scrubbing continues in the tray section of the
at>sorber where gas and lean ammonium phosphate solution contact counter-
currently.
A solution exchanger in the bottom of the absorber purges acid gases from
the rich solution. The rich solution is then pumped to the stripper.
Overhead vapors from the stripper transfer energy to the rich solution before
lt enters the stripper.
The high pressure stripper shifts the equilibrium of the stripping
teaction, thereby removing more ammonia from the rich solution. The resulting
solution is recycled to the tray section of the superstill absorber.
at 600 psig provides the heat for the stripper reboiler.
The vapor exiting the top of the stripper is 10 to 20 weight percent am-
. The vapor is fed to a two-stage condenser. The first condenser stage
"eats the stripper feed. Cooling water absorbs heat energy in the second
cotidenser stage. The resulting condensate passes to the fractionator feed
'ank from which it is pumped to the fractionator. The aqua-ammonia condensate
ls distilled into a 99.99 percent pure ammonia overhead fraction and a 0.05
Percent ammonia bottoms fraction. Accumulation of trace quantities of acid
&ases and organics in the fractionator is prevented by adding caustic sodium
ydroxide solution to the system, thereby assuring a high purity product.
water exiting the bottom of the fractionator is pumped to the bottom of
superstill to provide energy.
: Water Treatment Configuration and Cost
The generation, pretreatment and stripping of sour water for
combined cycle systems are shown in Figure A-5. Part of the unreacted
steam condenses while passing through the heat recovery system and is
8eParated from the gas in water knock-out drums. The gaseous mixture is
^en scrubbed with recycle water to remove ammonia. Ammonia removal is
"^essary to reduce the total plant NOX producton during low-Btu fuel gas
c°«ibustion. With the ammonia, some H2S and CC-2 also dissolve in the
Drubbing water. Entrained char particles will be removed by the scrubbing
ater. Removal of char particles is necessary to limit absorption on the
urbo-generator blades and to reduce particulate emissions during low-Btu gas
c°mbustion. Water collected in the knockout drums and scrubber water is
through gravity settling tanks to remove char particles. The solids-
stream is decanted, filtered, then passed to the stripping columns.
condensate from the acid gas removal section also will be sent to the
Dipping columns. H2S first will be stripped, then sent to the sulfur
ecovery process. Ammonia is stripped off in the separate column and can be
Covered as aqueous ammonia by quenching it with water. About 1 to 2 pounds
^ steam will be required for H2S and ammonia removal. The treated water
*-ll be free of suspended particles, ammonia and hydrogen sulfide. Part of
6 treated water is recycled to the scrubber and the rest is utilized in
°°ling towers.
307
-------
SOUR WATER GENERATION AND TREATMENT
LOW-BTU
PRODUCT GAS
CO
O
00
-j
CO
o
WATER
K.O.
1066
839
561
RECYCLE WATER
GASTOSELEXOL
ACIDGAS REMOVAL
WATER
SCRUBBER
80
389
505
1134
1228
1051
CHAR
SETTLING
230
185
7
SOLIDS
THICKNING
NOTE:
THE NUMBERS REPRESENT WATER
FLOW RATES IN GALLONS PER
MINUTE FOR BCR/AIR, NCR/
OXYGEN, AND IGT/AIR PRO-
CESSES USING LOW TEMPERATURE
SELEXOL CLEANUP.
18
11
1
CHAR TO
GAS>F\ER
VENT GAS
ACID GAS
REMOVAL SULFUR
RECOVERY
NH-
TO COOLING
TOWERS
STEAM
1-1.2 LB/GALLON OF FEED
Tl
-------
The characterization of the inlet and outlet streams as shown in
Figure A-5 is given in Tables A-7 to 9. The characteristics of the streams
are based on the data published in the Office of Coal Research (OCR) report
on coal gasification for combined cycle pilot plants. The cyanide, phenol,
and organic concentrations of the sour water in this report were based on
Complex calculations of thermal equilibrium determining trace compounds in
the gasifier produce gas. All of these components were assumed, as a worst
case, to be absorbed into the plant sour water.
The capital cost for ammonia separation by the Two Stage Distillation
Process and the Phosam-W Process for the three processes is shown in Table
^"•10. These costs are based on the cost figures estimated for different
throughput by Water Purification Associates (Reference A-l).
BOILER FEEDWATER TREATMENT
Raw water initially will be pumped to a storage reservoir. The reservoir
also serves as a flow equalizer and a clarifier which removes the suspended
solids present in the natural water. At this stage the water going to the
Process is split into two streams, the smaller going to the demineralizers
aild the larger one to the cooling towers. The amount of demineralized water
tequired depends on the total steam consumption of the process. Steam is
tequired for gasification, sour water stripping and sulfur recovery. Boiler
and demineralizer blowdowns have to be considered when estimating the boiler
feed water stream. In general, the raw water to demineralizers can be
6stimated by summing up the total steam consumption and the boiler and
blowdowns.
The deminerlizers are used in the treatment scheme to remove ionic
®Pecies and silica from boiler feedwater by exchanging these species with
^drogen, sodium and hydroxyl ions on solid resins. The resins are then
tegenerated by H2S04, NaCl and NaOH, enabling their continued use and
teuse over long periods. The most widely used ion exchange materials are
tesins of styrene-divinyl-benzene copolymers. The ion exchange resins play
911 important role in determining the process economics and demineralization
CaPabilities . A summary of pertinent data for conventional ion exchange
*esins (Reference A-2) is presented in Table A-ll. Also, a conventional
en»ineralizer system is shown in Figure A-6.
There are a number of schemes that can be considered to optimize water
^mineralization. Two schemes are presented in Figure A-7. In Scheme 1,
.a++, Mg++, HC03 and Si02 are removed by a hot lime treatment. Ca++
further removed by a softening resin. Demineralization is carried out in
?Uccessive steps with strong acid and weak base exchange followed by a mixed
ed- In Scheme 2, the water is first treated with a weak acid ion exchange
to remove Ca4"*", Mg++ and some Na+, and to replace these ions with
The hydrogen ions release CC>2 from HC03~, which is removed in the
ifier. Additional removal of Ca++ is required with a softening resin.
.Mineralization requires a strong acid resin followed by a weak base resin.
strong base resin is used for removal of Si02 and a mixed bed system is
*ed for final polishing. Scheme 2 makes more efficient use of regeneration
and costs less. A return condensate polishing system will be
309
-------
TABLE A-T
FLOW CHARACTERIZATION FOR BCR/AIR-BLOWN/SELEXOL SOUR WATER TREATMENT
(SEE FIG. A-5)
B
u>
i—"
o
STREAM
COMPONENT
H.O
H2S
NH3
co2
Phenols
Organics
Cyanides
Suspended Solids
Non-C ondens ible s
LB/HR
235,720
1,030
8,15*1
19,7^3
3
12
7
2,880
38
WT PPM
—
2,811
22,260
58,898
8
33
19
7,862
103
LB/HR
125,950
3
6
3
1
U
l
l
—
WT PPM
—
25
50
25
< 8
< 30
< 5
< 5
—
TOTAL
267,389
125,963
-------
TABLE A-8
FLOW CHARACTERIZATION FOR BCR/OXYGEN-BLOWN/SELEXOL SOUR WATER TREATMENT
(SEE FIG. A-5)
A B
STREAM
COMPONENT
H20
H2S
NH3
co2
Phenols
Organics
Cyanides
Suspended Solids
Non-Condens ible s
LB/HR
613,750
1,080
8,390
20,716
k
20
12
2,1*1*5
61*
WT PPM
—
1,670
12,980
32,01*1*
6
30
18
3,780
100
LB/HR WT PPM
284,890
6 21
12 1*2
6 21
1.7 < 6
< 30
< 5.
< 5
—
TOTAL 646,481 284,916
-------
TABLE A-9
FLOW CHARACTERIZATION FOR IGT/AIR-BLOWN/SELEXOL SOUR WATER TREATMENT
(SEE FIG. A-5)
B
STREAM
COMPONENT
H20
H2S
NH3
COo
UJ ^
NS
Phenols
Organics
Cyanides
Suspended Solids
Non-Condensibles
LB/HR
525,290
337
T01
6,28b
5
25
10
175
35
WT PPM
—
632
1,315
11,793
9
1*5
18
315
63
LB/HR WT PPM
365, 351*
1 3
2 6
1 3
< 5
< 3
< 5
< 5
—
TOTAL 532,867 365,358
-------
TABLE A-10
COST ESTIMATION OF SOUR WATER TREATMENT
COGAS Process
Throughput,
106 gal/day
BCR/Air-Blown/Selexol BCR/Qxygen-Blovn/Selexol IGT/Air-Blo-wn/Selexol
1.5
Tvo Stage Distillation Process
Total Equipment
Cost, 10S $
Phosam-W Process
Total Equipment
Cost, 10° $
Operating Costs for Phosam-W Process
Capital charges including maintenance
at 17* yr (330 days/yr)
Energy at $1.80/106 Btu
Caustic soda
Phosphoric acid
3.9
3.1
3.06
O.OU
0.01
5.05
1.7
U.2
3.1
Costs ($/thousand gallons)
1.76
3.06
0.0*1
0.01
l.U
3.5
2.6
1.33
3.06
O.OU
0.01
k.kh
-------
TABLE A-ll
SUMMARY OF PERTINENT DATA FOR CONVENTIONAL ION EXCHANGE RESINS
Ion Exchange Resins
Cation Exchange Resins
(Hydrogen or Acid Form)
Sulfonic Acid
(Strong Acid)
Carboxylic Acid
(Weak Acid)
Flow Rate
Discussion
All feed water cations
are removed
Cation removal is limited
to quantities equivalent
to the amount of weak anion
('bicarbonate) that is present
in the feed water.
[(m3/day)/m2] (gpm/ft3) Chemical
Anion Exchange Resins
(Hydroxyl or Base Form)
Strong Base Anions of "both strong and weak
Type I - Gel Type mineral acids are subject to
Type II - Gel Type greater physical degradation
Type I - Marco Process than are macro porous resins
Type II - Marco Process
Weak Base
Aliphatic Amine
Styrene Copolymer
Anions of strong mineral
acids are removed. Anions
of weak mineral acids such as
C02, Si02 are not removed.
I+-5
8-10
H2S01|
HC1
HC1
Regeneration
(lb/ft-s)
Typical Ion
Typical Capacity
(Kg/m3!(Kgr/ft^)
80
176
160
160
112
5
11
10
10
7
25
57
69
137
137
11
25
30
60
60
6
6
6
6
6
8
NaOH
NaOH
NaOH
NaOH
NaOH
NHUOH
NaOH
NHl^OH
Ba0CO-3
61*
61*
61+
61*
59
51
79
1*8
1*3
6U
l+.O
I+.O
i*.o
i*.o
3.7
3.2
'4.9
3.0
2.7
U.O
25
1*8
23
1+1
66
66
66
1*8
U8
H8
11
21
10
18
29
29
29
21
21
21
-------
FIG.A-6
CONVENTIONAL TWO-BED ION EXCHANGE PROCESS
WATER
INLET
TREATED-WATER OUTLET
AIR
'CONDUCTIVITY
CELL . i
DILUTE ACID
TO
WASTE
AN ION
EXCHANGER
'TO *• DILUTE CAUSTIC
WASTE
78-02-138-11
315
-------
BOILER FEED WATER TREATMENT SCHEMES
1
r
HOT LIME
SOFTENING
>
r
SOFTENING
IX
^
r
STRONG-ACID
IX
>
r
WEAK-BASE
IX
>
r
MIXED-BED
IX
Co
SCHEME I
I
V
WEAK-ACID
IX
i
1
V
DEGASIFIER
\
Y
Na
SOFTENING
\
r
STRONG-ACID
IX
i
T
WEAK-BASE
IX
^
r
STRONG-BASE
IX
T
MIXED-BED
IX
!
i
— >-
IX = ION EXCHANGER
SCHEME 2
oo
o
M
-------
required, but this is inexpensive. A raw water analysis is shown in Table
A-12, the performance characterisitics of Scheme 2 are shown in Table A-13.
The deionized water will be used to generate steam at various pressures.
The gasifier requires steam at 420 psig. Acid gas removal and ammonia
recovery utilize steam at 50 psig and at 420 psig. The steam turbines will
require steam at pressures above 1000 psig. The quality requirements for
boiler feedwater at different pressures are shown in Table A-14. The deionized
Water produced by the Scheme 2 as mentioned above meets these boiler feedwater
quality requirements.
Boiler Feedwater Treatment Cost
A cost summary for Scheme 2 demineralization, including the cost of chem-
icals for the three processes considered, is shown in Table A-15. The
following costs are given for regenerating chemicals (^ = cents): H2S04,
3.8 4H\>; NaOH, 8.3 ^/lb; and NaCl, 2.0 ^/lb. The treatment costs were esti-
"•ated by using the cost data published by Water Purification Associates
(Reference A-l) (the treatment by Scheme 2 costs about $3.2 x 105 for a
flow rate of 869,000 Ib/hr including resin replacement).
jteverse Osmosis for Boiler Feedwater Treatment
Reverse osmosis systems combined with detnineralizers have proven to be
cost effective in power plant feedwater use. Such a combination scheme for
boiler feedwater treatment is used by some utility companies with an observed
savings in the operating cost of more than one-third the conventional system.
With the reverse osmosis unit operating upstream, the demineralizers produce
five to ten times more deionized water between regenerations, and require
less manpower to operate and maintain. In addition, the chemical requirements
are lowered, thereby waste-disposal problems are minimized. Ion exchange
resins operate with improved performance and extended life.
Figure A-8 shows the conventional demineralizer system and the combined
reverse osmosis (R0)/demineralizer system, with a flow rate of 420,000 gpm;
Table A-16 compares the cost (Reference A-3) of the two systems. The savings
^•n chemical costs alone justifies application of the combined system.
A new approach to boiler feedwater treatment is the use of RO without
Demineralizers. Such a system has been operating for more than two years at
a Texas petrochemical plant. The system, which feeds the boiler operating at
600 psi, produces deionized water from brackish well water containing 550 ppm
Dissolved solids. The 110,000 gpd system provides a 90 percent rejection at
^5 percent recovery. The operating costs have ranged between 60 cents and 70
cents per thousand gallons.
DOLING TOWER WATER TREATMENT
The total make-up water to the cooling tower is primarily fresh water
**om the source, but also includes recycled water from the settling tank,
"°iler blowdown and the stripped water from the ammonia stripping section.
317
-------
TABLE A-12
RAW WATER ANALYSIS
Component , Composition as CaCOg
pH 1.6
Total Dissolved Solids ItOO mg/1
Bicarbonate (HCO) 180 mg/1 ihl .6
2-
Sulfate (SO^ ) 90 mg/1 93.6
Chloride (Cl~j 170 mg/1 239. T
Nitrate (NO ) U.2 mg/1
Calcium (Ca) 52 mg/1 130.0
Magnesium (Mg) lit mg/1 57. ^
Sodium and Potassium (Na, K) 85 mg/1 iW.l
Iron (Fe) 0.7 mg/1 1.3
Silica (Si02) 8.8 mg/1
Dissolved Oxygen (Og) 9-8 mg/1
Ammonia (NHo) 2.5 mg/1
Specific conductivity at 25°C 1.1 x 10"^ mho
318
-------
TABLE A-13
PERFORMANCE CHARACTERISTICS FOR SCHEME 2
BOILER FEED WATER DEMINERALIZATION*
After Weak
• Acid IX and
Ot"iponent Degasifier
* Mg 20
* 69
l* 170
\ 0
\ 90
% 9
After
Softening
0
99
170
0
90
9
After Strong Acid
Weak-Base
Demineralization
0
hk
0
0
1
9
Acid Strong-Base
IX Silica Removal
0
u
0
0
0
-------
TABLE A-lU
QUALITY REQUIREMENTS FOR BOILER FEEDWATER^
Boiler Feedwater
Quality of water prior to the addition of
substances used for internal conditioning
Characteristics
Silica (Si02)
Aluminum (Al)
Iron (Fe)
Manganese (Mn)
Calcium (Ca)
Magnesium (Mg)
Ammonia (NH, )
<+
Bicarbonate (HC03>
Sulfate (SO^)
Chloride (Cl)
Dissolved solids
Copper (Cu)
Zinc (Zn)
Hardness (CaCC>3)
Free mineral acidity (CaCO.j)
Alkalinity (CaC03)
pH, units
Color, units
Organics :
Methylene blue active
substances .
Carbon tetrachloride
extract .
Chemical oxygen demand (CO
Dissolved oxygen (0-)
Temperature, F
Suspended solids
Low
pressure
0 to 150
psie
30
5
1
0.3
(2)
(2)
0.1
170
(2)
(2)
700
0.5
<3)
20
(3)
140
8.0-10.0
(2)
1
1
5
2.5
(2)
10
Industrial
Inter-
mediate
pressure
150 to 700
PSiR
10
0.1
0.3
0.1
(3)
(3)
0.1
120
(2)
(2)
500
0.05
(3)
(3)
(3)
100
8.2-10.0 8
(2).
1
1
5
0.007
(2)
5
High
pressure
700 to 1,500
0.7
0.01
0.05
0.01
(3)
(3)
0.1
48
(2)
(2)
200
0.05
(3)
(3)
(3)
z.0
.2-9.0
(2)
0.5
0.5
0.5
0.007
(2)
(2)
Electric
Utilities
1,500 to
5.000 psiK
0.01
0.01
0.01
(4)
(A)
(4)
0. 7
(4)
(A)
(A)
0. 5
0.01
(A)
(3)
(3)
(3)
8.8-9.2
(2)
(3)
(3)
(3)
0.007
(2)
(3)
(1) Unless otherwise indicated, units are ng/1 and values that normally should
not be exceeded. No one water will have all the maximum values shown.
(2) Accepted as received (if meeting total solids or other limiting values):
has never been a problem at concentrations encountered.
(3) Zero, not detectable by test.
(4) Controlled by treatment for other constituents.
320
-------
u>
ro
TABLE A-15
COST SUMMARY FOR DEMIWERALIZATION OF BOILER FEED WATER
Process
Deminer-
alizer
Capacity
(gpm)
Deminer-
alizers
Annual
Operating
(Cost $) Kr>i
Chemicals
Required Ib/hr
30) NaOH NaCl
Total
Operating
Cost $
BCR/Air-Blown 886.5 163,250 160 76 U8 275,520
Selexol
BCR/Oxygen-Blovn 1073.^ 197, MO 19^ 91 58 325,
Selexol
IGT/Air-Blovn 1533.2 282,2l|0 277 131 83
Selexol
-------
CONVENTIONAL DEMINERALIZER AND RO/DEMINERALIZER SYSTEMS FOR BOILER
FEEDWATER TREATMENT
PRIMARY
SECONDARY
to
N>
No
DUAL MEDIA
CLARIFIER FILTER
FEED
, SLUDGE
STORAGE TANK
CATION ANION CATION ANION
WASTE
PRODUCT
CLARIFIER
DUAL MEDIA
FILTER
FEED
CARTRIDGE
FILTER
O
SLUDGE
STORAGE TANK
REVERSE
OSMOSIS
CATION ANION CATION ANION
-
BRINE
WASTE
PRODUCT
I
oo
-------
TABLE A-16
COST COMPARISON OF CONVENTIONAL DEMINERALIZER AND
R/0 DEMINERALIZER SYSTEMS
Conventional
RO/Demineralizer
C*Pital
^bor
^•ntenance
^treatment
Chemicals
Mineralization,
Chemicals
Wn
Vies
*ste Chemicals
Vr
Hai
$/yr
$160,230
96,000
50,000
•,6.460
176,680
32,310
81,300
12,1*30
651,1*10
$1000/gal
1.015
0.608
0.317
0.291*
1.091*
0.205
0-515
0.079
U.13
$/yr
$168,1*30
96,000
50,000
1*6,1*1*0
15,880
8,950
30,880
2,570
1*0,270
1*59,1*20
$1000/gal
1.067
0.608
0.317
"•**
0.101
0.057
0.196
0.016
0.255
2.91
!)
^t based on water containing 960 ppm TDS.
323
-------
Cooling tower water is treated to prevent scaling, fouling, microbial growth
and corrosion. The control limits for cooling tower circulating water are
given in Table A-17.
The raw water may contain a high concentration of HC03, sometimes
referred to as "alkalinity". Removal of bicarbonate prior to using the water
in cooling towers will prevent scaling. The simplest way to accomplish
this is to add sulfuric acid. One pound of HCO^ requires 0.8 Ib
H2S04 for treatment. Although simple, this is an expensive treatment, in
that it causes a waste of biocides and anticorrosion chemicals in the cooing
tower blowdown.
Chemical precipitation is also a feasible pretreatment option for fresh
water. This method reduces dissolved solids comprised of heavy-metal or
hardness ions. Lime, soda-ash or caustic is added in softeners to precipitate
the hardness. Cold lime softening is commonly employed. This restricts
frequent cooling tower blowdowns and conserves the water conditioning chemicals.
The impact of pretreating cooling tower make-up is summarized in
Table A-18. This table presents chemical analysis of lime-softened water,
and recirculating water at 6 and 13 cycles of concentration. The quality of
make-up water shown in this table is different from the quality of raw water
considered. The analyses for both cases favor sulfuric acid addition for
control of calcium carbonate. Calcium sulfate solubility, therefore, limits
the cooling water concentration. Note that the scaling potential of calcium
sulfate (as indicated by the solubility product of total calcium and sulfate
in solution) is identical for each case by virtue of softening pretreatment.
Operation at 13 rather than 6 cycles of concentration in this example reduces
the tower blowdown by approximately 57 percent.
Cooling Tower Make-up Water Treatment Costs
The cost of cooling water by simple sulfuric acid treatment and by lime
pretreatment for the processes considered is summarized in Table A-19. The
following cost data has been used.
Sulfuric Acid Treatment
H2S04 required for removal of 1 Ib of HC03 = 0.8 Ib.
H2S04 cost =3.8 cents/pound
Biocides cost = $1/10^ gallons of blowdown
Lime Treatment
Lime cost =2.7 cents/pound
Biocides cost = $0.50/10^ gallons of blowdown
Anticorrosion chemicals • $0.50/10^ gallons of blowdown
324
-------
TABLE A-1T
CONTROL LIMITS FOR COOLING TOWER
CIRCULATING WATER COMPOSITION.
Limits
l£ameters
Minimum Maximum
Saturation Index
War Stability Index
5H
Calcium, mg/1 as CaC03
?°tal iron, mg/1
, mg/1
, mg/1
, mg/1
e, mg/1
, mg/l
Ca-) ' (SOI+), product
dissolved solids, mg/1
micromhos/cm
pended solids, mg/1
+0.5
+6.5
6.0
20-50
+1.5
+7.5
8.0
300
koo
0.5
0.5
0.08
1
5
150
100
500,000
2,500
It, 000
100-150
Remarks
Nonchromate programs
Nonchromate programs
Nonchromate program
Chromate program
For pH 7-5
For pH 7-5
Both calcium and sulfate
expressed as mg/1
CaCO-a
325
-------
TABLE A-18
IMPACT OF SOFTENING COOLING TOWER MAKE-UP WATER
Constituent
Calcium
Magnesium
Sodium
Bicarbonate
Carbonate
Sulfate
Chloride
Silicate
Total
Hardness
PH
Turbidity
KSGaS01+*
*K =
scaSOU _
Plant
As Makeup
CaC03 232
CaC03 135
Na+ 117
CaC03 129
CaC03
SO^ 3U5
Cl~ 91*
Si02 IT
CaC03 36T
8-8.3
5-20
-
solubility product of
(cca++) (csoj)
Following
Primary
Treatment
1*9
121
175
0
35
355
9^
10
170
8-9
5-10JTU
-
Circulating Waters
Concentrations
at 6
1392
810
702
35
-
2685
56U
-
2202
6.5-7
-
3-73xl06
at 13
6UO
1580
2U09
35
-
5821
1221
-
2200
6.5-7
-
3.73xl06
calcium sulfate
where (Cp ++) = total calcium in solution, ppm as CaC03
) = total sulfate in solution, ppm as
326
-------
Derating
reatment
ime
featment
TABLE A-19
COOLING TOWER WATER TREATMENT COST
BCR/Air-Blown/
Selexol
62.3
38.6
BCR/Oxygen-Blown/
Selexol
52.lt
36.5
IGT/Air-Blown/
Selexol
57-1
36.6
327
-------
Lime Treatment (Cont'd)
Flocculation chemicals » $0.30/1000 gallons of blowdown
Dispersant chemicals = $0.15/1000 gallons of blowdown
Table A-19 shows that the operating cost of cooling water treatment
by lime pretreatment is about 60-70 percent of the cost of sulfuric acid
treatment. However, lime treatment requires an initial capital investment of
about one million dollars for the clarifiers and the side stream filter.
Sidestream Softening of Cooling Water
The amount of cooling tower blowdown can be greatly reduced by using
side-stream treatment, such as lime softening and filteration. This method,
shown in Figure A-9, treats the most concentrated water in the cooling tower.
As a result, the amount of water treated is significantly less than the
pretreatment of the make-up water. The system consists of an upflow, solids-
contact reactor/clarifier. This is followed by filteration, and the treated
water is recirculated to the cooling tower. Calcium and magnesium bicarbonates
and silica concentration are reduced by lime treatment. Calcium and magnesium
bicarbonates and silica concentration are reduced by lime treatment. Calcium
and magnesium sulfates can be removed by adding soda ash together with lime.
Treatment of the blowdown for removal of zinc, chromate, and other potential
pollutants is much more practical when applied to a small flow. The perfor-
mance of a typical system having a circulating water flow rate of 500,000 gpm
and a side stream flow rate of 2000 gpm shows that the treatment reduces
hardness to 80-120 mg/1, silica to 10-20 mg/1, and suspended solids to
10-20 mg/1. Lime treated effluent is recarbonated and then filtered to
1-2 mg/1 of suspended solids. Sludge from the clarifiers is settled, de-
watered on a filter and used for landfill.
328
-------
SIDESTREAM SOFTENING OF COOLING WATER
to
I
o
ro
co
CO
I
MAKE-UP WATER-
EVAPORATION AND
" DRIFT
CONDENSER
SLIP- STREAM
SOFTENER
TREATING
CHEMICALS
SOFTENER
SLUDGE
jj
CO
-------
REFERENCES
A-l Goldstein, D. J. and David Yung: Water Conservation and Pollution
Control in Coal Conversion Processes. EPA-600/7-77-065,
(NTIS No. PB269-568), June, 1977.
A-2 Strauss, Sheldon D.: Water Treatment. Power, 117 (6) S1-S24 (1973)
A-3 Truby, Randy: Reverse Osmosis for Boiler Feedwater Treatment.
Power Engineering, December, 1976, pp.58-60.
330
-------
APPENDIX B
FUEL PROCESSING COST BASIS
Fuel processing costs for the low-temperature system were estimated
by Fluor. The costs were estimated on a major component basis and an
indication of the degree of detail considered is given in Table B-l. That
table lists the major components that were considered in preparing an
overall plant capital cost estimate.
-------
TABLE B-l
EQUIPMENT LIST FOR AIR-BLOWN BCR PROCESS
Unit 10 - Coal Preparation and Handling
a. Coal Handling - one train required
Item Description
10-BN-l Unloading Hopper
10-CV-l Belt Conveyor
10-CV-2 Belt Conveyor
10-TR-l Tripper
10-ME-l Stacking System
10-ME-2 Reclaiming system
10-CV-3 Belt Conveyor
b. Coal Pulverization - one train required
10-CV-4 Belt Conveyor
10-BN-2 Pulverizer Feed Hopper
10-1-V-l Transport Gas
10-2-V-l Heater
10-2-ME-3 Pulverization System
10-2-ME-3
10-BN-3A, B, C & D Pulverized Coal Storage Silos
332
-------
Unit 20 - Gasification and Ash Handling
a- Gasification - five trains required (four operating and one spare)
with each train consisting of the following:
Item
Vessels
20-1-V-1A & B
20-1-V-2
20-1-V-3A & B
20-1-V-4
20-1-V-5
2Q-1-V-6A & 6B
20-1-V-7A & 7B
Reactors
20-1-R-l
Pumps
20-1-P-l
Compressors
20-1-C-l
Storage Bins
20-1-BN-l
Mechanical Equipment
20-1-ME-l
Ash Handling - one train required
Cooling Tower
20-CT-l
Bins
20-BN-2
Description
Coal Feed Lock Hoppers
High Pressure Coal Feed Hopper
Slag Lock Hoppers
Char Separator
Char Cooler
Char Transport Drums
Slag Hoppers
Gas ifier
HP Quench Water Pump
Transport Gas Compressor
Coal Feed Surge Bin
Bag Filter
Cooling Tower
Dewatering Bin
333
-------
b. Ash handling - one train required (continued)
Item Description
a.
Tanks
20-TK-l
20-TK-2
Pumps
20-P-2
20-P-3
20-P-4
20-P-5
20-P-6
Transfer Tank
Settling Tank
Quench Water Supply Pump
Quench Water Return Pump
Settling Tank Bottoms Pump
Jet Ejector Water Pump
Transfer Tank Bottoms Pump
Unit 21 - Gas Cooling and Particulate Removal
Gas Cooling and Particulate Removal - high temperature gas cooling
section consisting of 21-1-E-l, 21-1-E-2, 21-1-E-3, 21-1-V-l and
21-1-P-l have four operating and one spare trains and remainder
of system has four operating trains and no spare. Each train
consists of the following:
Vessels
21-1-V-l
21-1-V-2
21-1-V-3
21-1-V-4
Exchangers
21-1-E-l
21-1-E-2
21-1-E-3
21-1-E-4
21-1-E-5
Pumps
21-1-P-l
21-1-P-2
HP Steam Drum
Particulate Scrubber
K.O. Drum
Ammonia Absorber
HP Boiler
Gasifier Effluent/Product Gas Exchange*
BFW Preheater
BFW Make-Up Heater
Demineralized Water Preheater
HP Boiler Circulation Pump
Particulate Scrubber Circulation Pump
334
-------
Char Recovery - one train required
Item
Pumps
21-1-P-3
Mechanical Equipment
21-ME-l
ii .
22 - Acid Gas Removal System
Data supplied by Allied Chemical Co.
it_23 - Sulfur Plant and Tail Gas Treating
Sulfur Plant
A total of three parallel trains (two operating and one spare)
required with each train consisting of the following:
Description
Char Slurry Recycle Pump
Secondary Char Recovery Unit
5.
Furnace
23-1-H-l
Reactors
23-1-R-l
23-1-R-2
Exchangers
23-1-E-l
23-1-E-2
Blowers
23-1-BL-l
Pumps^
23-1-P-l
Sump
23-1-S-l
Sulfur Furnace and Waste Heat Boiler
Converter No.l
Converter No.2
Condenser No. 1
Condenser No. 2
Air Blower
Sulfur Transfer Pump
Sulfur Sump
335
-------
b. Tail Gas Treating
A total of three trains required (two operating and one spare) with
each train consisting of the following:
Item
Heater
23-1-H-2
Reactor
23-1-R-3
Vessels
23-1-V-l
23-1-V-2
Exchangers
23-1-E-3
23-1-E-4
23-1-E-5
Blowers
23-1-BL-2
Pumps
23-1-P-2
23-1-P-3
23-1-P-4
23-1-P-5
Tanks
23-1-TK-l
23-1-TK-2
Description
Tail Gas Heater
SC>2 Converter
Absorber/Quench Vessel
Sulfur Decanter
Reactor Effluent Cooler
Quench Circulating Cooler
Regenerated Solution Cooler
Air Blower
Quench Water Circulating Pump
Regenerated Solution Pump
Froth Slurry Pump
Filter Cake Pump
Oxidizer Tank
Froth Trank
336
-------
Unit 24 - Sour Water Treating and Ammonia Recovery
One train required.
Item
Vessels
24-V-l
24-V-2
24-V-3
24-V-5
24-V-6
Tanks
24-TK-l
24-TK-2
Exchangers
24-E-l
24-E-2
24-E-3
24-E-4
24-E-5
24-E-6
24-E-7
24-E-8
24-E-9
24-E-l0
24-E-l1
24-E-l2
Description
Superstill
Ammonia Stripper
Ammonia Fractionator
Fractionator Feed Tank
Flash Drum
Phosphoric Acid Storage Tank
Caustic Storage Tank
Lean Solution Cooler
Solution Exchanger
Stripper Condenser
Condensate Subcooler
Fractionator Condenser
Superstill Reboiler
Absorber Cooler
Bottoms Cooler
Stripper Reboiler
Fractionator Reboiler
Reflux Condenser Cooler
Feed Preheater
24-P-l
24-P-2
24-P-3
24-P-4
24-P-6
24-P-7
Absorber Circulation Pump
Rich Solution Pump
Fractionator Feed Pump
Fractionator Reflux Pump
Superstill Bottoms Pump
Reflux Condenser Pump
337
-------
APPENDIX C
POWERPLANT COST ANALYSIS
Previous studies demonstrated the need for a capability to provide
cost estimates for future power systems that not only would enable power
stations with different operating characteristics to be considered, but
also would allow sensitivity studies to be undertaken for differing tech-
nical characteristics of the selected systems. As a result, members of the
technical staff at UTRC, in cooperation with cost estimators at Burns &
Roe, Inc., developed a set of correlations based on historical data which
enable the total installed cost of COGAS power stations to be estimated
in-house with a high degree of confidence. The procedure consists of a set
of equations, grouped by Federal Power Commission (FPC) account numbers, which
can be used to calculate the installed costs of major power station equipment.
Minor component equipment items are combined in logical groups, the
costs of which are also estimated from correlation equations. The calcu-
lation procedure requires input data which are normally calculated as part
of a routine technical analysis for electric power stations. Examples of
such data include: gross and net station output power; steam turbine
output power; gas turbine output power; cooling water flow rates; sub-
component input power requirements; and station efficiency. Because the
cost correlations have been developed in sufficient detail, the need to
resort to expensive preliminary system layout drawings is eliminated.
As part of the overall analysis, the costs of a few major components
are estimated by methods other than the correlations described since these
alternative methods make it possible to determine these specific equipment
costs in even greater detail. Two such components are the large, industrial
gas turbines and waste heat boilers, the costs of which are calculated
using sophisticated computer program analyses proprietary to UTRC.
Furthermore, the costs of large, "standard" items such as steam turbo-
generators and electric generators are obtained directly from published
catalog data, which are corrected by appropriate price multiplier factors
to maintain consistency with the latest industry experience.
338
-------
A detailed list of all system components whose prices are capable of
being estimated directly is presented in Table C-l for all applicable FPC
power station account categories. Not all items are used for each COGAS
system considered since often there are specific characteristics of each
station design which distinguish it from other stations. In the UTRC pro-
cedure, allowances are also made for such items as station start-up, tem-
porary buildings, transportation during construction, special tools required
during a particular power station contract, engineering, contingency, esca-
lation (except where included in the cost of selected major system components),
and interest during construction. Examples of typical resulting costs are
8hown to the right of each category. The example used is the air-blown BCR
type gasifier with low-temperature cleanup which was the basic system of
lnterest in this study.
339
-------
TABLE C-l
POWER SYSTEM COST BREAKDOWN
FPC No. 341: Structures and Improvements
Site Preparation 1,209,000
Administration Building 805,000
Gas Turbine Building 1,088,000
Turbogenerator Building 3,738,000
Circulating Water System 1,438,000
Stack 1.724.000
10,002,000
FPC No. 312 Boiler Plant Equipment
Waste Heat Boiler 56,365,000
Boiler Feed Pumps 741,000
Boiler Feedwater Tank and Deaerator 87,000
Demineralizer 906,000
Condensate Storage Tank 39,000
Misc. Pumps 156,000
Piping 12,265,000
Insulation for Piping 981,000
Controls 992,000
72,532,000
FPC No. 314 Turbogenerator Unit
Turbogenerator 23,616,000
Condenser, Tube & Ejector 1,585,000
Condenser Vacuum Pumps & Motor 221,000
Condenser Pumps & Drive Motor 80,000
Cooling Tower 7,550,000
Circulating Water System 1,412,000
34,464,000
340
-------
TABLE C-l (Cont 'd)
Power System Cost Breakdown
FPC No. 343 Gas Turbine
Gas Turbine 46,038,000
Air Start System 180,000
Coupling to Generator 125,000
Lube Oil Purification System 137,000
Lube Oil Filtration 112,000
Turbine Air Precooler 603,000
Service Air Compressor 160,000
Breeching 3,540,000
Expansion Joints 2,064,000
Inlet Air Filters 472,000
Turbine Enclosure Air Cooler 120,000
Emergency Cooling Water System 12,000
Misc. Pumps & Tanks 28,000
Control Panels 300,000
Computer 300,000
Fuel Piping 1,501,000
Fuel Pipe Insulation 225,000
55,917,000
FPC No. 341 Generator for Gas Turbine
14,863,000
FPC No. 345 Accessory Electrical Equipment
353 Station Equipment
17,620,000
FPC No. 346 Misc. Powerplant Equipment
526,000
341
-------
1. REPORT NO. 2.
EPA-600/7-78-088
4. TITLE AND SUBTITLE
Fuel Gas Environmental Impact: Final Report
7. AUTHOR(S)
F.L.Robson and W.A. Blecher (UTRC); and
V. B. May (Hittman Associates)
9. PERFORMING ORGANIZATION NAME AND ADDRESS
United Technologies Research Center
Silver Lane
East Hartford, Connecticut 06108
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
June 1978 ____-— -
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORI ""
1O. PROGRAM ELEMENT NO.
EHE623A ^
11. CONTRACT/GRANT NO.
68-02-2179
|J* \ ft n 1 « Q /^ K 1 0 / */ ^7
A XlldJL v / I w™ JLw / 1 1 ___*•— -"
14. SPONSORING AGENCY CODE
EPA/600/13 ^
TECHNICAL REPORT DATA
(Please read Iiittnictions on the reverse before completing)
15. SUPPLEMENTARY NOTES ffiRL 7RTP project .off leer is Thomas W. Petrie,
919/541-2708. EPA -600/7 6 -15 3 and -/75-078 are earlier, related r
Mail Drop 61,
eports.
_^ —*^r
The report gives results of continued investigation and further definition °
the potential environmental and economic benefits of integrated coal gasification/g85
cleanup/combined gas and steam cycle power plants. Reported refinements in
operating characteristics lower heat rates and reduce emissions from previous
values. An expanded study of plant environmental intrusions includes a look at
tially hazardous trace elements. Comparisons made of integrated plants using ail"
and oxygen-blown gasifiers favor air-blowing. Careful theoretical design of plants
with low temperature sulfur cleanup reduces to marginal levels the performance
and cost advantages of plants with high temperature cleanup. If gasifier steam f®6"
rates are kept low in all but fixed bed types, choice of gasifier among other major
generic types is not critical to achieving attractive systems using low temperature
cleanup. Excessive thermal NOx emissions may be avoided by departing from con*
ventional combustor designs. Fuel NOx and particulates still pose problems with
of high temperature cleanup. Sulfur removal to very low levels is possible with
integrated systems, but cost rises rapidly as it becomes necessary to remove
of the COS as well as the H2S.
16. ABSTRACT
7.
KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
Pollution
Coal Gasification
Gas Purification
Fuels
Toxic ity
Trace Elements
Desulfurization
Nitrogen Oxides
Dust
Chemical Analysis
13. DISTRIBUTION STATEMENT
Unlimited
b.lDENTIFIERS/OPEN ENDED TERMS
Pollution Control
Stationary Sources
Environmental Impact
Fuel Gas
Part iculate
19. SECURITY CLASS (This Report)
Unclassified
20. SECURITY CLASS (This page)
Unclassified
c. COS AT I ricld/W
13B c
13H c
07A 1
21D
06T
06A
-— -
21. NO. OF PAGt=>
357
22. PRICE
EPA Form 2220O (9-73)
------- |