vvEPA
United States Industrial Environmental Research
Environmental Protection Laboratory
Agency Research Triangle Park NC 2771 1
EPA-600/8-83-008
April 1983
Research and Development
Pollution Control
Technical Manual for
Koppers-Totzek
Based Indirect Coal
Liquefaction
-------
EPA-600/8-83-008
April 1983
POLLUTION CONTROL TECHNICAL MANUAL
FOR
KOPPERS-TOTZEK-BASED INDllMWoAL LIQUEFACTION
Contract
68-02-3137
Program Manager: Gregory G. Ondich
Office of Environmental Engineering and Technology (RD-681)
U. S. Environmental Protection Agency
401 M Street, SW
Washington, DC 20460
Project Officer: William J. Rhodes
Industrial Environmental Research Laboratory-RTP
Research Triangle Park, NC 27711
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NC 27711
U.S. Environmental Protection Agency
Region 5, Library (5PL-16)
230 S. Dearborn Street, Room 1670
Chicago, IL 60604
-------
DISCLAIMER
This Pollution Control Technical Manual was based on data obtained in EPA's
source characterization study at the Koppers-Totzek based plants at Modder-
fontein, S.A. and Ptolemais, Greece. The evaluation at Ptolemais was a joint
study supported by the Tennessee Valley Authority (TVA) and the Environmental
Protection Agency (EPA). Additional data sources used in this manual in-
cluded other EPA synfuels characterization studies, environmental impact
statements, published literature, and EPA supported engineering calculations.
No proprietary or confidential data appear or have been used in the prepara-
tion of this manual. Although this manual addresses the Koppers-Totzek
gasifier based technology, the process developer, Krupp-Koppers, GmbH, Essen
was not involved in the development of this manual. Thus, the manual does
not necessarily represent Krupp-Koppers1 engineering design data, material
balances, or operational information.
This document has been reviewed in accordance with U.S. Environmental
Protection Anency policy and approved for nublication. Mention of trade
names or commercial oroducts does not constitute endorsement or recommenda-
tion for use.
-------
FOREWORD
The purpose of the Pollution Control Technical Manuals (PCTMs) is to
provide process, discharge, and pollution control data in summarized form
for the use of permit writers, developers, and other interested parties.
The PCTM series covers a range of alternate fuel sources, including coal
gasification and coal liquefaction by direct and indirect processing, and
the extraction of oil from shale.
The series consists of a set of technical volumes directed at production
facilities based upon specific conversion processes. The entire series is
supplemented by a pollution control technology appendix volume which describes
the operation and application of approximately 50 control processes.
All PCTMs are prepared on a base plant concept (coal gasification and
liquefaction) or developer proposed designs (oil shale) which may not fully
reflect plants to be built in the future. The PCTMs present examples of
control applications, both as individual process units and as integrated con-
trol trains. These examples are taken in part from applicable permit appli-
cations and, therefore, are reflective of specific plants. None of the
examples are intended to convey an Agency endorsement or recommendation but
rather are presented for illustrative purposes. The selection of control
technologies for application to specific plants is the exclusive function of
the designers and permitters who have the flexibility to utilize the lowest
cost and/or most effective approaches. It is hoped that readers will be able
to relate their waste streams and controls to those presented in these
manuals to enable them to better understand the extent to which various tech-
nologies may control specific waste streams and utilize the information in
making control technology selections for their specific needs.
The reader should be aware that the PCTMs contain no legally binding
requirements or guidance, and that nothing contained in the PCTMs relieves
a facility from compliance with existing or future environmental regulations
or permit requirements.
Herbert L. Wiser
Acting Deputy Assistant Administrator
Office of Research and Development
U.S. Environmental Protection Agency
-------
ABSTRACT
The Environmental Protection Agency (EPA), Office of Research and Devel-
opment has undertaken an extensive study to determine synthetic fuel plant
waste stream characteristics and to evaluate potentially applicable pollution
control systems. The purpose of this and all other PCTMs is to convey this
information in a manner that is readily useful to designers, permit writers,
and the public.
The Koppers-Totzek (K-T)-based indirect liquefaction PCTM addresses the
K-T gasification technology as licensed by Krupp-Koppers, QnhH, Essen, West
Germany (GeselIschaft fur Kohle-Technologie for licensing within the U.S.)
and all intermediate process operations leading to each of three liquid pro-
duct configurations. The liquid product syntheses considered include Fischer-
Tropsch liquids, methanol, and Mobil M-gasoline. A single feed coal (Illinois
No. 6 bituminous) is utilized throughout the manual as the basis for illus-
tration, with the impacts of alternative coal ranks also described in the
text.
This manual proceeds through a description of the hypothetical base
plant, character!«s the waste streams produced in each medium, and discusses
the array of commercially available controls which can be applied to the base
plant waste streams. From these generally characterized controls, several ex-
amples are constructed for each medium in order to illustrate typical control
technology applications. Then, example control trains are constructed for
each medium, illustrating the function of integrated control systems. Control
and control system cost and performance estimates are presented, together with
descriptions of the discharge streams, secondary wast>^ streams, and energy
requirements. A summary of the gaps and limitations in the data base used to
develop this manual is presented, along with a listing of additional data
needs.
iv
-------
CONTENTS
FOREWORD • ill
ABSTRACT iv
FIGURES viii
TABLES x
GLOSSARY OF ACRONYMS xv
CONVERSION FACTORS xviii
ACKNOWLEDGEMENT xviii
SECTION 1 INTRODUCTION 1
1.1 Koppers-Totzek Based Indirect Liquefaction 2
1.2 Approach to Manual Development 4
1.2.1 Base Plant Definition 4
1.2.2 Control Technology Evaluation 5
1.3 Data Base 6
1.4 Manual Organization and Utilization 9
1.4.1 Manual Organization 9
1.4.2 Manual Utilization 10
SECTION 2 PROCESS DESCRIPTION OVERVIEW 13
2.1 Coal Feed Characteristics and Product Slate 17
2.2 Base Plant Description 20
2.2.1 Coal Preparation 20
2.2.2 K-T Coal Gasification 21
2.2.3 Gas Purification and Upgrading 22
2.2.4 Product Synthesis . 25
2.2.5 Auxiliaries 28
2.2.6 Fugitive and Miscellaneous Wastes 28
2.3 Base Plant Capital Investment and Annualized Operating
Costs 30
SECTION 3 PROCESS DESCRIPTION AND WASTE STREAM CHARACTERIZATION . . 33
3.1 Coal Preparation 42
3.2 Coal Gasification 52
3.3 Gas Purification and Upgrading 60
3.3.1 Gas Cooling and Dust Removal 60
3.3.2 NOX Reduction 69
3.3.3 Raw Gas Compression and Cooling 70
3.3.4 Cyanide Wash 74
3.3.5 Shift Conversion 78
3.3.6 Acid Gas Removal 83
3.3.7 Trace Sulfur Removal 89
-------
CONTENTS (Continued)
3.4 Product Synthesis 91
3.4.1 Methanol Synthesis 92
3.4.2 Fischer-Tropsch (F-T) Synthesis 96
3.4.3 Methane Co-production (Fischer-Tropsch Synthesis
Case) 98
3.4.4 Mobil M-gasoline Synthesis 100
3.4.5 Product Recovery and Upgrading 105
3.4.6 Waste Streams Generated by Synthesis Operations . 106
3.5 Products and By-Products Ill
3.5.1 Methanol Synthesis Product Ill
3.5.2 Fischer-Tropsch Liquid Products 113
3.5.3 Mobil M-gasoline Products 114
3.5.4 Substitute Natural Gas (SNG) 119
3.5.5 LP Gas 119
3.5.6 By-Product Sulfur (Stream 112) 119
3.6 Auxiliaries 121
3.6.1 Raw Water Treatment 121
3.6.2 Power Generation and Process Heating 128
3.6.3 Cooling Operations 134
3.6.4 Oxygen Production 137
3.6.5 Product and By-Product Storage 138
3.7 Fugitive and Miscellaneous Wastes 142
3.7.1 Fugitive Organic Emissions (Stream 241) 142
3.7.2 Non-Process/Intermittent Wastewater Streams . . . 143
3.7.3 Equipment Cleaning Wastes (Streams 242 and 305) . 146
3.8 Waste/Control Technology Index 147
SECTION 4 EVALUATION OF POLLUTION CONTROL TECHNOLOGY 163
4.1 Gaseous Medium 169
4.1.1 Acid Gases Containing Reduced Sulfur/Nitrogen,
Organics, and/or Carbon Monoxide 173
4.1.2 Combustion Gases 218
4.1.3 Organic and CO Containing Waste Gases 249
4.1.4 Fugitive Dust from Material Storage (Stream 200) . 254
4.1.5 Fugitive VOC Emissions 259
4.1.6 Fugitive Particulates from Material Conveying and
Processing 274
4.2 Aqueous Medium 276
4.2.1 Water Pollution Control Processes 285
4.2.2 Water Pollution Controls for Streams Containing
Predominantly Organic Constituents 309
4.2.3 Water Pollution Controls for Streams Containing
Predominantly Inorganic Constituents 320
4.2.4 Integrated Pollution Control Examples 335
vi
-------
CONTENTS (Continued)
4.3 Solid Waste Management 370
4.3.1 Solid Waste Control 375
4.3.2 Inorganic Ashes and Sludges 390
4.3.3 Recovered By-Products 404
4.3.4 Organic Sludges 405
4.3.5 Spent Catalysts and Sulfur Guard 408
SECTION 5
SECTION 6
Appendix A
Appendix B
DATA GAPS AND LIMITATIONS
REFERENCES
411
432
Costing A-1
Rectisol Acid Gas Removal Process B-l
VII
-------
FIGURES
Number Page
2-1 Simplified flow diagram for K-T based synthesis gas
production 14
2-2 Simplified flow diagram for conversion of synthesis gas to
liquids 15
3-1 Operations associated with synthesis gas production in K-T
based indirect liquefaction facilities 34
3-2 Synthesis operations associated with K-T based indirect
liquefaction facilities 35
3-3 Auxiliary processes associated with K-T based indirect
liquefaction facilities 36
3-4 Waste streams associated with coal preparation for a K-T based
indirect liquefaction facility - Illinois No. 6 coal 44
3-5 GKT's gasifier with waste heat boiler and slag extraction
system 53
3-6 Gas cooling and dust removal 61
3-7 Raw gas compression and cooling 72
3-8 Shift conversion 82
3-9 Two-stage selective Rectisol acid gas removal system 85
3-10 Flow diagram for the ICI methanol synthesis process 93
3-11 Fischer-Tropsch (Synthol) synthesis and product recovery ... 97
3-12 Methanation, C02 removal, and drying for SNG production ... 101
3-13 Flow diagram for Mobil M-gasoline synthesis and crude product
fractionation 102
3-14 Flow diagram for base plant raw water treatment system .... 124
4-1 Three stage Glaus plant with split flow option 176
4-2 The Stretford process 179
4-3 Example 1 - Glaus bulk sulfur removal with Beavon/Stretford
tail gas treatment 205
4-4 Example 2 - Glaus bulk sulfur removal with SCOT tail gas
treatment and incineration 210
4-5 Example 3 - Glaus bulk sulfur removal with Wellman-Lord tail
gas treatment 214
viii
-------
FIGURES (Continued)
Number Page
4-6 Integrated control examples - Mobil M or F-T synthesis case . 340
4-7 Integrated control examples for base plants producing crude
methanol - discharge to surface waters 352
4-8 Integrated control examples for base plants producing crude
methanol - discharge to surface impoundment or deep well
injection 353
4-9 Landfill design 383
4-10 Capital investment and annual ized unit cost for landfills . . 384
4-11 Surface impoundment design 387
-------
TABLES
Number Page
1-1 Completed and Ongoing Data Acquisition Programs at Coal
Gasification Facilities Sponsored or Co-Sponsored by the
EPA 8
2-1 Proximate and Ultimate Analyses of Base Plant Illinois No. 6
Coal 17
2-2 Estimated Product/By-Product Slate for K-T Base Plants ... 19
2-3 Capital Costs for Uncontrolled K-T Based Indirect Liquefaction
Plants 30
2-4 Annualized Costs for K-T Based Indirect Liquefaction Plants . 31
3-1 Flows of Major Streams for K-T Based Indirect Liquefaction
Facilities - Illinois No. 6 Coal 38
3-2 Characteristics of Illinois No. 6 Coal Selected for Use in
Indirect Liquefaction Base Plants 43
3-3 Coal Preparation Section Mass Flows - Methanol Synthesis Case 46
3-4 Estimated Fugitive Dust Emissions from Coal Storage Piles . . 47
3-5 Particulate Emissions from Coal Preparation (Illinois No. 6
Coal) 50
3-6 Material Flow Estimates for K-T Gasification (Illinois No. 6
Coal) 56
3-7 Results of K-T Slag Leaching Tests 57
3-8 Material Flow Estimates for Raw Gas Cooling and Dust Removal
Processes (Illinois No. 6 Coal) 63
3-9 Leachability of Selected Elements from K-T Dust and from
Illinois No. 6 Feed Coal 66
3-10 Characteristics of Slowdown from Cooling and Dust Removal . . 68
3-11 Material Flow Estimates for K-T NOX Reduction, Compression and
Cooling, and Cyanide Wash Processes (Illinois No. 6 Coal) . . 71
3-12 Characteristics of Primary Compression and Cooling Condensate
from South African Sub-bituminous Coal 74
3-13 Material Flow Estimates for K-T Shift Conversion and Acid Gas
Removal Processes (Illinois No. 6 Coal) 87
3-14 Methanol Synthesis Material Flow Estimates for K-T Gasification
(Illinois No. 6 Coal) 95
-------
TABLE (Continued)
Number Page
3-15 Fischer-Tropsch Synthesis Material Flow Estimates for K-T
Gasification (Illinois No. 6 Coal) 99
3-16 Mobil M-gasoline Synthesis Material Flow Estimates for K-T
Gasification (Illinois No. 6 Coal) 104
3-17 Components Reported in Commercial Methanol 112
3-18 Estimated Composition of Crude Methanol from Coal 113
3-19 Comparison of the Estimated Composition of Finished Indirect
Coal Liquefaction, Unleaded Gasolines, and Typical Petroleum
Gasolines 115
3-20 Distribution of Oxygenated By-Products from Fluid-Bed Fischer-
Tropsch Synthesis 116
3-21 Methanol Conversion Unit Feed and Product Composition .... 117
3-22 Composition of Raw Makeup Water 122
3-23 Estimated Makeup Water Quantity for a K-T Based Indirect
Liquefaction Plant (Illinois No. 6 Coal) 123
3-24 Raw Water Treatment Sludge (Stream 300) Production Rates and
Characteristics 126
3-25 Demineralizer Regeneration Wastewater Composition (Stream 301) 127
3-26 Boiler Mass Flow for Illinois No. 6 Coal - Methanol Synthesis
Case 130
3-28 Cooling System Makeup Water Requirements for a K-T Indirect
Liquefaction Plant 135
3-29 Estimated Characteristics of Cooling Tower Slowdown and Drift 136
3-30 Evaporative Emission Estimates for Product and By-Product
Storage 139
3-31 Composition of Evaporative Emissions from Gasoline Storage . . 141
3-32 Estimated Total Fugitive Organic Emissions 144
3-33 Drainage Estimate from Non-Process/Intermittent Streams . . . 145
3-34 Stream Index 148
3-35 Cross-Reference Index for Primary Waste Streams 154
3-36 Cross-Reference Index for Secondary Waste Streams 159
xi
-------
TABLES (Continued)
Number Page
4-1 Summary of Estimated Gaseous Waste Stream Characteristics in
K-T Based Indirect Liquefaction Facilities 170
4-2 Categorization of Gaseous Waste Streams According to Source
Type in K-T Indirect Liquefaction Facilities 171
4-3 Key Features of Bulk Sulfur Removal Processes 175
4-4 Key Features of Residual Sulfur Removal Processes 183
4-5 Comparison of Incineration Processes 188
4-6 Example 1 - Material Flow Estimates for Integrated Control
Employing Claus Bulk Sulfur Removal with Beavon/Stretford
Tail Gas Treatment 206
4-7 Example 1 - Costs of Integrated Claus Bulk Sulfur Removal
with Beavon/Stretford Tail Gas Treatment (1980 Basis) ... 208
4-8 Example 2 - Material Flow Estimates for Integrated Control
Employing Claus Bulk Sulfur Removal, SCOT Tail Gas Treatment,
and Incineration 211
4-9 Example 2 - Costs of Integrated Claus Bulk Sulfur Removal
with SCOT Tail Gas Treatment and Incineration (1980 Basis) . 213
4-10 Example 3 - Material Flow Estimates for Integrated Control
Employing Claus Bulk Sulfur Removal with Well man-Lord Tail
Gas Treatment 215
4-11 Example 3 - Costs of Integrated Claus Bulk Sulfur Removal
with Wellman-Lord Tail Gas Treatment (1980 Basis) 217
4-12 Combustion Modification Techniques for NOX Control 221
4-13 NOX Flue Gas Treatment Control Alternatives for Boilers . . 228
4-14 Key Features of Particulate Collection Equipment 234
4-15 Key Features of S02 Removal Processes 237
4-16 Composition of Flue Gas from the Dewatered Dust-Fired
Fluidized-Bed Boiler (Stream 413) 246
4-17 Key Features of Storage Pile Dust Control Technologies . . . 255
4-18 Repair Methods for Fugitive Emissions Reduction 264
4-19 Equipment Design/Modifications for Fugitive Hydrocarbon
Emissions Control 265
Xll
-------
TABLES (Continued)
Number Page
4-20 Storage Tank Emission Estimates 267
4-21 Estimated Incremental Costs for Storage of Synthetic Liquids 269
4-22 Fugitive Organic Emissions from Process Equipment 271
4-23 Capital and Annualized Costs for Fugitive Organic Emission
Controls 273
4-24 Summary of K-T Base Plant Wastewater Streams and Estimated
Characteristics 277
4-25 Categorization of Aqueous Waste Streams in K-T Gasification
Facilities 280
4-26 Control Processes Potentially Applicable to the Treatment of
K-T-Based Gasification Plant Wastewaters 286
4-27 Material Flow for Cooling Tower Concentration - Mobil M
Synthesis Case 314
4-28 Material Flow for Forced Evaporation-Mobil M Synthesis Case . 317
4-29 Material Flow for Cooling Tower Concentration and Forced
Evaporation - Water-Based Cyanide Wash Case 326
4-30 Material Flow for Cooling Tower Concentration and Forced
Evaporation - Methanol-Based Cyanide Wash Case 331
4-31 Characterization of Major Streams Combined for Common Treat-
ment - Mobil M Synthesis Case 338
4-32 Example 1 - Material Flow for Mobil M Synthesis Base Plant
Integrated Controls 342
4-33 Example 2 - Material Flow for Mobil M Synthesis Base Plant
Integrated Controls 345
4-34 Example 3 - Material Flow for Mobil M Synthesis Base Plant
Integrated Controls 347
4-35 Estimated Characteristics of Wastewater Streams Discharged
to Ultimate Disposal - Mobil M Synthesis Case 348
4-36 Costs of Integrated Control Examples - Mobil M Synthesis
Case 349
4-37 Characterization of Major Streams to be Combined for Treat-
ment - Crude Methanol Production Case 355
4-38 Example 4 - Material Flow for Crude Methanol Production Base
Plant Integrated Controls 357
xi ii
-------
TABLES (Continued)
Number Page
4-39 Example 5 - Material Flow for Crude Methanol Production Base
Plant Integrated Controls 358
4-40 Example 6 - Material Flow for Crude Methanol Production Base
Plant Integrated Controls 363
4-41 Example 7 - Material Flow for Crude Methanol Production Base
Plant Integrated Controls 364
'4-42 Example 8 - Material Flow for Crude Methanol Production Base
Plant Integrated Controls 365
4-43 Estimated Characteristics of Wastewater Streams Discharged
to Ultimate Disposal - Crude Methanol Production Case .... 366
4-44 Costs of Integrated Control Examples - Crude Methanol
Production Case 367
4-45 Summary of Solid Waste Streams from K-T-Based Indirect
Liquefaction Facilities 371
4-46 Summary of Solid Waste Management Technologies 376
4-47 Site-Specific Factors to be Considered for Land-Based
Disposal Options 377
4-48 Estimated Flow Rates for the Inorganic Ash and Sludge
Streams 390
4-49 Characteristics of Flue Gas and Spent Bed Media from FBC
Boiler 396
4-50 Estimated Capital Investment and Total Annualized Cost for
Burning Dewatered Gasifier Dust in FBC Boiler 397
4-51 Summary of Capital Investment and Total Annualized Cost for
Disposinq of Gasifier Dust in Surface Impoundment 399
4-52 Estimated Treatment/Disposal Cost for Biosludge 406
4-53 Estimated Flow Rates for Spent Catalysts and Sulfur Guard . . 408
5-1 Completed and Ongoing Data Acquisition Programs at Coal Gasi-
fication Facilities Sponsored or Co-Sponsored by the EPA . . 415
5-2 Data Gaps and Research Needs - Gaseous Medium 416
5-3 Data Gaps and Research Needs - Aqueous Medium 422
5-4 Data Gaps and Research Needs - Solid Medium 438
5-5 Data Gaps and Research Needs - Products/By-Products 431
xiv
-------
GLOSSARY OF ACRONYMS
ACP Ammonia from Coal Project
ADA Anthraquinone disulfonic acid
ADIP Shell-patented diisopropyl amine-based acid gas removal process
AGR Acid gas removal
BM Bureau of Mines
BOD Biochemical oxygen demand
COD Chemical oxygen demand
CRA Compression-refrigeration-absorption
CRF Capital Recovery Factor
DEA Diethanolamine
DIPA Diisopropanolamine
DOE Department of Energy
DOI Department of Interior
EGD Effluent Guidelines Division, Office of Water Regulations and
Standards, EPA
EP Extraction Procedure
EPA Environmental Protection Agency
EPRI Electric Power Research Institute
ERDA Energy Research and Development Administration
ESP Electrostatic precipitator
FGD Flue gas desulfurization
FGR Flue gas recirculation
FGT Flue gas treatment
F-T Fischer-Tropsch
GKT Gesellschaft fur Kohle-Technologie
HHV Higher Heating Value
IERL Industrial Environmental Research Laboratory
K-T Koppers-Totzek
LEA Low excess air
LHV Lower Heating Value
xv
-------
GLOSSARY OF ACRONYMS (Continued)
LNB Low NO burners
s\
LPG Liquified petroleum gas
MAP Moisture and ash free
MEA Monoethanolamine
MDEA Methyldiethanolamine
NFI Nitrogenous Fertilizer Industry (S.A.)
NOV Nitrogen oxides
A
NMHC Non-methane hydrocarbons
NPDES National Pollutant Discharge Elimination System
NSPS New Source Performance Standards
OAQPS Office of Air QUality Planning and Standards, EPA
OFA Overfire Air
OPTS Office of Pesticides and Toxic Substances, EPA
OSW Office of Solid Wastes, EPA
PCB Polychlorinated Biphenyls
PCTM Pollution Control Technical Manual
PNA Polynuclear Aromatics
POM Polycyclic organic matter
PSD Prevention of Significant Deterioration
RCRA Resource Conservation and Recovery Act
RL Reduced Load
ROM Run of Mine
SASOL South African Coal, Oil and Gas Corporation, Ltd.
SCOT Shell Claus Off-Gas Treatment
SCR Selective Catalytic Reduction
SNG Substitute Natural Gas
SNPA Societe Nationale des Petroles d'Aquitaine
SO Sulfur oxides
/\
TDS Total dissolved solids
xvi
-------
GLOSSARY OF ACRONYMS (Continued)
TEA Triethanolamine
TGT Tail gas treatment
TOC Total organic carbon
TSP Total suspended particulates
TSS Total suspended solids
TVA Tennessee Valley Authority
VOC Volatile Organic Compounds
W-L Wellman-Lord
xvn
-------
CONVERSION FACTORS
1.0 kg [kilogram]
1.0 Mg [megagram (metric ton)]
1.0 kg/min [kilogram per minute]
1.0m3 [cubic meter]
1.0 Nm3/hr [normal cubic meter
(at 0°C) per hour]
1.0 GJ [gigajoule]
1.0 MW [megawatt]
1.0 MJ/s [megajoule per second]
1.0 kWh [kilowatt hour]
1.0 MJ/Nm3 [megajoule per
normal cubic meter (at 0°C)]
o
1.0 g/Nm [gram per normal
cubic meter (at 0°C)]
1.0 kPa [kilopascal]
1.0 kmole
Prefixes
2.205 Ib [pound (mass)]
1.102 ton [short ton (2000 lb)l
132.3 Ib/hr [pound per hour]
264.2 gal [gallon]
37.32 scfh [standard cubic feet
(at 6QOF) per hour]
0.9479 x 106 Btu [British thermal
unit]
3.413 x 106 Btu/hr [British thermal
unit per hour]
3.413 x 106 Btu/hr [British thermal
unit per hour]
3413 Btu [British thermal unit]
25.40 Btu/scf [Btu per standard
cubic foot (at 60°F)]
0.413 gr/scf [grains per standard
cubic foot (at 60°F)]
0.00987 atmosphere
22.4 Nm3 (at 0°C and 1 atmosphere)
T - tera - 10
12
G - giga = 10'
M = mega = 10
k = kilo =
ACKNOWLEDGEMENT
Technical and background information for this Pollution Control Technical
Manual was prepared for the EPA by the Environmental Division, TRW, Inc.,
Redondo Beach, California, under Contract 68-02-3647. The TRW Project
Manager for this effort was Mr. R. Orsini.
xvi
-------
Section 1
Introduction
SECTION 1
INTRODUCTION
Future U.S. energy production envisions the development of an environ-
mentally acceptable synthetic fuels industry. As part of this overall effort,
the Environmental Protection Agency (EPA), Office of Research and Development,
has for the past several years undertaken extensive studies to determines syn-
thetic fuel plant waste stream characteristics and potentially applicable
pollution control systems.
The purpose of the Pollution Control Technical Manuals (PCTMs) is to
convey in a summarized and readily useful manner, information on synfuel
waste stream characteristics and pollution control technology as obtained
from studies by EPA and others. The documents provide waste stream character-
ization data and describe a wide variety of pollution controls in terms of
estimated performance, cost, and reliability. The PCTMs contain no legally
binding requirements, no regulatory guidance, and include no preference for
process technologies or controls. Nothing within these documents binds a
facility to accepting the suggested emission control process(es) in the ser-
vice^) indicated nor relieves a facility from compliance with existing or
future environmental regulations or permits.
The Pollution Control Technical Manuals consist of several discrete docu-
ments. There are six process-specific PCTMs and a more general appendix
volume which describes over fifty pollution control technologies. Application
of pollution controls to a particular synfuel process is described in each
process specific manual. The volumes currently contemplated are:
Pollution Control Technical Manual for Lurgi-Based Indirect
Coal Liquefaction and SNG
Pollution Control Technical Manual for Koppers-Totzek-Based Indirect
Coal Liquefaction
Pbllution Control Technical Manual for Exxon Donor-Solvent
Direct Coal Liquefaction
1
-------
Section 1
Introduction
Pollution Control Technical Manual for Lurgi Oil Shale Retorting
vn'th Open Pit Mining
Pollution Control Technical Manual for Modified In-Situ Oil
Shale Retorting Combined with Lurgi Surface Retorting
Pollution Control Technical Manual for TOSCO II Oil Shale
Retorting with Underground Mining
Control Technology Appendices for the Pollution Control Technical
Manuals
By focusing on specific process technologies, the PCTMs attempt to be
as definitive as possible on waste stream characteristics and control tech-
nology applications. This focus does not imply any EPA recommendations for
particular process or control designs. Those described in the manuals are
intended as representative examples of processes and control technologies
that might be used. The organization of the PCTMs from process description
through waste stream characterization and control technology evaluation pro-
vides the user with a number of alternative approaches. Permit writers must
be cautioned that these control technology configurations are not the only
ones suitable for a specific plant.
Control technology configurations presented in the PCTMs reflect pollu-
tant removal levels which are believed to be achievable with currently avail-
able control technologies based upon existing data. Since there are no
domestic commercial scale synfuels facilities, the data base supporting this
document is from bench and pilot synfuel facilities, developers' estimates,
engineering analyses, analogue domestic industries, and non-U.S. commercial
synfuel plants.
1.1 Koppers-Totzek Based Indirect Liqufaction
Indirect coal liquefaction links of two types of processes. One
produces the H2- and CO-rich synthesis gas from coal by gasification, while
the second produces a range of synthetic liquid products by reacting the H?
and CO components of a synthesis gas. A number of specific processes are
-------
Section 1
Introduction
available for the performance of both the gasification and synthesis steps.
The combination of gasification and synthesis steps is in constrast with
direct liquefaction, wherein any of several technologies can be used to
directly produce liquid products from coal.
This PCTM addresses indirect liquefaction facilities for the production
of synthetic fuels by means of Koppers-Totzek (K-T) coal gasification followed
by any of three alternative fuel product synthesis routes, including methanol,
Fischer-Tropsch, and Mobil M-gasoline. Facilities of this type utilize all
of the process operations normally associated with indirect liquefaction,
namely coal preparation; coal gasification and raw gas cleaning; shift con-
version and acid gas removal; and synthesis of the desired fuel product.
Auxiliary processes required to support these production operations are those
required for oxygen production, raw water treatment, process cooling, and
waste stream treatment. Depending upon the local availability and cost of
electric power, on-site auxiliary power generation facilities may also be
required.
The K-T process, as developed and licensed by Krupp-Koppers, GmbH,
Essen, West Germany (Gesellschaft fur Kohle-Technologie (GKT) in the U.S.),
is a commercially viable process which has been widely used outside the U.S.
to produce industrial fuel gas and synthesis gas from coal. To date, the
GKT K-T process has been used with a variety of coal feeds, ranging from
brown coals through lignite and bituminous ranks and encompassing the full
breadth of coking tendencies.
As a consequence of the recent formal separation of Koppers, Pittsburgh
from Krupp-Koppers, a new gasifier has been designed and is being marketed
by Koppers-Pittsburgh and Babcock and Wilcox as the KBW unit. While the KBW
gasifier does incorporate different approaches (relative to K-T) to both heat
recovery and dust separation, it does not appear to incorporate any features
which would alter the actual gasification conditions, reactions or the extent
-------
Section 1
Introduction
of those reactions. Therefore, while the KBW gasifier has riot been operated
commercially and no specific process, operational or environmental data on
the design were available for use in this manual, the Agency believes that the
statements and examples presented herein regarding the K-T gasifier are also
valid to a first approximation for the KBW gasifier.
All of the process operations discussions in this PCTM are largely based
upon foreign experience, with the exception of the U.S.-developed Mobil M-
gasoline process which is not yet in commercial operation. The methanol and
Fischer-Tropsch synthesis processes have been commercially applied to the pro-
duction of liquids from coal-derived synthesis gases. For purposes of this
document, all of these technologies are considered ready for commercial appli-
cation.
1.2 Approach to Manual Development
1.2.1 Base Plant Definition
In order to define the production operations and waste streams that would
be associated with representative integrated process facilities, an uncon-
trolled base plant was defined which incorporates the features seen in the K-T-
based plants which are either proposed, under construction, or currently in
operation. In this context, an uncontrolled base plant is one which has full
production capability (all of the equipment required to produce saleable pro-
duts) but no equipment to control pollutant discharges. Auxiliary processes
included in the base plant are those that render a facility essentially self-
sufficient in energy; i.e., one requiring only run-of-mine coal, raw water,
and various chemicals and catalysts as inputs. Illinois No. 6 bituminous coal
was examined as the primary feed to these facilities, with the impacts of
using different ranks of coals with various heating values, moisture, sulfur,
and ash contents being examined as variations to the primary coal. This
approach permitted estimation of total stream and consistent flow rates in
a process facility utilizing different feedstocks and served to define the
-------
Section 1
Introduction
range of uncontrolled (i.e., base plant) waste streams for which treatment
and/or disposal could be considered.
A base plant size corresponding to approximately 120 TJ/day of clean
synthesis gas produced was selected to be representative of the sizes of the
modules of the first plants that may be built in the U.S. The energy output
rate (after synthesis) of a plant of this size is equivalent to the energy
content of about 20,000 bbls/day of gasoline. Using various data sources
(discussed below), material flows and energy usages were estimated for all
base plant feedstock/synthesis process combinations.
1.2.2 Control Technology Evaluation
The PCTMs examine control alternatives from three viewpoints: first,
identifying several control technologies with their operating principles
and applicability to particular types of waste streams and defining the in-
herent performance limitations of these control technologies; second, by
using waste streams of given compositions as feeds to several of these con-
trol technologies, estimates of achievable control unit performance levels
and costs are illustrated; and third, alternative integrated control trains
are employed to treat base plant waste streams, thereby illustrating the
range of overall control and control economics for individual streams. Each
control technology utilized in the illustrative examples is further described
in the control appendix volume.
Since very limited data on the performance of controls are available
from operating synthetic fuels plants, many assumptions had to be made to
extrapolate the experience gained with the same control technologies in
related industries. These assumptions have been carefully documented in the
appropriate sections of this manual. Waste streams resulting from pollution
control process operations (secondary waste streams) were also identified
and controls for those streams described and illustrated. Cost estimates
for all controls were developed based upon published data and vendor-supplied
-------
Section 1
Introduction
estimates. These data were extrapolated to a 1980 cost basis to provide a
consistent basis for comparing the relative costs of alternate controls.
Base (uncontrolled) plant costs were extrapolated in a similar manner.
Users of this PCTM should recognize that there are two yery significant
limitations associated with the use of the data presented.
t First, no fully-integrated, well controlled commercial plants
of the type discussed in this manual have been constructed to
date. Thus, in using the data base presented here, users are
cautioned to take careful note of the documented limitations in
the data and assumptions made to resolve apparent differences
in data obtained from test facilities having widely differing
feedstocks, designs, operating characteristics, and site-specific
constraints.
t Second, it should be noted that this manual does not attempt to
address all of the issues that will be important in the selection
and design of environmental control systems for new synthetic fuels
facilities. Since this manual focuses on streams that tend to be
unique to synfuels facilities, streams that are covered by exist-
ing source-based regulations or that are similar to waste streams
routinely encountered in other industries for which regulatory
precedents already exist are recognized but not treated in depth.
Also, this PCTM focuses primarily on controls for point and
fugitive sources of pollution, and not on the environmental impacts
of those emissions.
Fugitive dust from coal storage and handling facilities may have
to be controlled to satisfy Prevention of Significant Deteriora-
tion (PSD) limitations associated with ambient particulate matter
concentrations. Selecting appropriate controls for this type of
emission requires more details on the use of technologies and site-
specific analyses than are included in this manual.
It should also be noted that this manual does not address issues
related to worker health and safety, noise, socioeconomic, or
ecological impacts.
1.3 Data Base
Since the early 1970's the EPA has sponsored a significant environmental
assessment program addressing technologies for producing synthetic fuels from
-------
Section 1
Introduction
coal. This work has involved a combination of theoretical studies and plant
data acquisition programs and has contributed to both the data and background
knowledge used in the development of this document. Table 1-1 lists the major
contributing data acquisition programs sponsored or co-sponsored by the EPA.
As indicated, the data encompass specific research projects, pilot-level
sampling and analysis projects, and source sampling of foreign and domestic
commercial production facilities.
The major sources of data used to define the types and characteristics of
uncontrolled synthetic fuels facility waste streams and to develop base plant/
process configurations were: (1) an EPA-sponsored test program of a K-T gasi-
fication facility at Modderfontein, S.A.; (2) an EPA- and TVA-sponsored test
program at the K-T facility at Ptolemais, Greece; (3) Linde/LOTEPRO research
and EPA (IERL and OAQPS) sponsored tests at Rectisol installations; (4) DDE-
sponsored gasoline-from-coal research studies conducted by Mobil Research
and Development Corporation; and (5) permit filings, environmental impact
statements, and design studies for various proposed K-T-based domestic syn-
thetic fuel facilities. In addition, data derived from applications of con-
trols in related industries such as the petroleum refining, natural gas
processing, by-product coking, electric utility, and coal preparation indus-
tries were relied upon heavily in determining control applicability and costs.
Uncontrolled waste stream characteristics were estimated using overall
material balance calculations and available compositional data from bench-,
pilot-, and commercial-scale facilities based on similar technologies. The
reader should recognize that K-T-based facilities built in the U.S. may con-
tain design features that will result in different uncontrolled waste stream
characteristics. Therefore, users of this manual should carefully consider
the design features of a particular facility before making judgments concern-
ing uncontrolled waste stream composition and the applicability and perfor-
mance of candidate control technologies for these streams.
-------
TABLE 1-1. COMPLETED AND ONGOING DATA ACQUISITION PROGRAMS AT COAL GASIFICATION FACILITIES SPONSORED
OR CO-SPONSORED BY THE EPA
oo
Facility
Medium/High Btu Gasification
and Indirect Liquefaction
Facilities (Foreign)
t Lurgi Gasification
- Kosovo, Yugoslavia
- SASOL, S.A.
- Westfield, Scotland
« Koppers-Totzek Gasification
- Modderfontein, S.A.
- Ptolemais, Greece
- Kutahya, Turkey
• Winkler Gasification
- Kutakya, Turkey
• Texaco Gasification
- Federal Republic of Germany
Low-Btu Gasification Facilities
(U.S.)
* Wellman Galusha
- Site No. 1
- Site No. 2
• Chapman/Wilputte
• Riley
• Stoic (Foster Wheeler)
Control Research Facilities
• Raw/Acid Gas Cleanup (Fluidized
Bed Gasifier)
• Wastewater Treatability Studies
t Pollutant Identification (Bench
Information Classification
Data acquisition
Plant visit and discussions
Plant visit and discussions
Data acquisition
Data acquisition (TVA & EPA)
Plant visit and discussions
Plant visit and discussions
Data acquisition (EPRI, TVA
& EPA)
Data acquisition
Data acquisition
Data acquisition
Data acquisition
Data acquisition (DOE & EPA)
North Carolina State Univ.
Univ. of North Carolina
Research Triangle Institute
Coal Used
Lignite
Low rank bituminous
Various
High volatile "B"
bituminous
111 . No. 6 bituminous
Lignite
Lignite
111 . No. 6 bituminous
Anthracite
Lignite
Low sulfur bituminous
Lignite
Western bituminous
Various
Various
Various
Products
Medium Btu gas
Various via indirect
1 iquefaction
Test center
Ammonia, methanol
Ammonia
Ammonia
Ammonia
Test center
Fuel gas
Test center
Fuel gas
Test
Fuel gas
Test center
Test center
Test center
Scale Gasifier)
• Ash Leaching Evaluations
Other Domestic Facilities
• Texaco Gasification
- Ammonia from coal plant, TVA
• Rectisol Acid Gas Cleanup
University of Illinois
Data acquisition (TVA & FPA)
Texaco, Wilmington, CA
Va-ious
111. No. 6 bituminous
(in shakedown)
Oil fired partial
oxidation
Test center
Ammonia
Process hydrogen
-------
Section 1
Introduction
1.4 Manual Organization and Utilization
1.4.1 Manual Organization
This Pollution Control Technology Manual is presented in two volumes.
The main text volume describes the processes, their associated waste streams,
and the pollutants potentially requiring control. It also provides descrip-
tions and illustrative examples of pollution control technologies. Detailed
information supporting the cost estimates cited in the main text can be
found in the appendix of the main text volume. In addition, detailed
discussions of control processes can be found in a separate appendix volume
that is common to all PCTMs.
The subsequent sections of this volume provide:
Section 2 An overview of the process operations discussed
in this manual.
Section 3 A description of the sources and characteristics
of the waste streams generated by those process
operations.
Section 4 An analysis of the performance capabilities and
costs of candidate control processes for waste
streams, including those generated by the control
processes themselves.
Section 5 A summary of the quality of the data base used for
the base plant development and control technology
analysis.
Section 2 will be most useful to readers seeking a general knowledge of
the characteristics of the gasification technology which this document addres-
ses. Detailed information about the characteristics of specific uncontrolled
waste streams is presented in Section 3. This section also describes how the
characteristics of those streams are likely to be impacted by differences in
feedstock (coal) characteristics, process design features, and plant operating
characteristics. The rationale for the selection of specific control processes
to serve as illustrative examples as well as an analysis of the expected
-------
Section 1
Introduction
performance of those controls is presented in Section 4. In light of the in-
tended use of this document, this Section is most critical because it presents
current estimates of the performance capabilities and costs of controls which
have been or could be proposed for use in the subject facilities. Potential
users of information presented in Section 4 should carefully note the data
limitations summaries presented in Section 5 and should utilize the control
appendices to establish the bases for adapting the PCTM base plant information
to the specific case at hand. These summaries are intended to give potential
users of this document a general feel for the quality of the data used to
estimate both uncontrolled base plant waste stream characteristics and con-
trol equipment performance and cost data.
1.4.2 Manual Utilization
PCTM use has been targeted for those individuals concerned with the pre-
paration of permits and the applications for them. As such, the interests of
these plant designers and permit writers will focus both on grouped waste
streams, with the common characteristics being the medium (i.e., air, water,
or solid waste streams) and/or major pollutant species involved, as well as
on individual streams which, because of their flow volume or constituents,
could significantly impact environmental design criteria.
The PCTM accommodates these interests in several ways. First, the entire
study is centered around a generalized uncontrolled "base" plant, which con-
tains all of the production operations to be found in the coal conversion
sections of K-T-based indirect liquefaction facilities; second, all waste
streams (including control residuals or secondary wastes) are categorized by
medium; and third, within each medium, the streams are grouped by major pol-
lutant constituents.
To illustrate the utilization of the PCTM by someone with a designer's or
permit writer's interest in a particular waste stream in a planned site-specific
facility, the initial step will, of course, be to locate the counterpart waste
10
-------
Section 1
Introduction
stream in the base plant. This can be done through the use of the Section 2
and 3 flow diagrams and the stream and cross reference indicies in Section 3.8.
Section 3 will provide both a full characterization of the counterpart base
plant stream, plus a description of (as applicable) other base plant streams
which may have to be analytically combined in order to better match the design
stream in question.
Section 3.8 contains a process/waste stream index and cross reference
indicies for waste streams. The process/waste stream index, Table 3-34,
provides stream numbers for many process streams and all waste streams, and
facilitates identification of streams discussed in the text. The cross refer-
ence index for primary waste streams, Table 3-35, lists: (1) all of the un-
controlled base plant waste streams and indicates the Section 3 subsection
that contains detailed characterization data for each stream, and (2) the
Section 4 subsection that contains information on potentially applicable pol-
lution controls for each stream. The cross reference index for secondary
waste streams, Table 3-36, provides similar information on the pollution con-
trols discussed in Section 4, the secondary waste stream(s) generated by
those controls, available controls for the secondary waste stream(s), and
references to the appropriate Control Technology Appendices.
Should the user's interests lie only with waste stream controls, Table
3-35 will direct him to the appropriate level of discussion in Section 4. J
General control technology descriptions are followed by specific discussions
of the performance, secondary waste streams, and costs associated with the
application of example control techniques to specific waste streams. For
example, Section 4.1.1.1 discusses the capabilities and costs of specific
control processes applied to the offgases from the acid gas removal unit.
Following the stream-by-stream discussions and examples of individual control
applications are illustrative examples of potential integrated control schemes
(i.e., multiple control techniques used in series) for those streams or com-
binations of streams which would normally be expected to use more than a
. 11
-------
Section 1
Introduction
single control (due to multiple pollutant loadings or other reasons). Similar
types of information are presented for water pollution controls and solid
waste management techniques in Sections 4.2 and 4.3, respectively. Additional
"how to use" information is presented at the beginning of Sections 4.1, 4.2,
and 4.3
If the user's interests involve a potential control residual or secondary
waste stream, these and the location of applicable PCTM discussions of control
alternatives can also be directly determined through the Table 3-36 cross
reference index for secondary waste streams. The user will find that cross-
media secondary wastes are generally discussed under the medium to which they
contribute; exceptions are those peculiar cases in which the character of the
secondary stream implies a potential for treatment in excess of that which
would be provided if it were combined with other streams. In these cases,
the control alternatives are discussed under the medium of origin.
Additional detail on the controls utilized in the Section 4 illustrative
examples can be found in the Control Technology Appendices volume. Discus-
sions of current commercial applications, performance, and the basis for both
the performance and cost estimates are described there.
Limitations in the available data base are discussed in Section 5. The
section is organized such that the data bases on both synfuel-unique primary
streams and on the applicable control unit alternatives are covered. The user
will find the counterpart base plant waste streams arranged by medium. This
section covers the main base plant waste streams and associated secondary
streams from potential control applications.
12
-------
Section 2
Process Overview
SECTION 2
PROCESS DESCRIPTION OVERVIEW
This section presents a brief description of the process operations and
non-pollution control auxiliary processes expected to be included in Koppers-
Totzek (K-T) based indirect liquefaction facilities. It also identifies the
major uncontrolled waste streams associated with those operations and pro-
cesses. The operations and processes described in this section comprise
what is called an "uncontrolled base plant" or "base plant" since they are
all required to produce marketable products and by-products. According to
this definition, the "base plant" excludes those processes whose primary
function is to treat waste streams to render them suitable for discharge or
reuse within the plant. Base plant process operations discussed here in-
clude coal preparation, coal gasification, gas purification and upgrading,
crude product synthesis and separation, and product upgrading. Auxiliary
operations discussed include process cooling, product storage, raw water
treatment, and oxygen production. Integrated K-T facilities are ordinarily
more than self-sufficient with respect to process steam requirements.
Further, on-site generation of electric power is not expected to be econom-
ical in the U.S. as compared with the purchase of power. However, on-site
auxiliary power generation using coal as boiler fuel is examined in this
document in an attempt to cover a broad range of possible plant configura-
tions. Capital investment and operating costs for the base plant are also
summarized.
Figure 2-1 presents a simplified block diagram of a K-T based synthesis
gas production facility. The process operations associated with the synthe-
sis of Fischer-Tropsch liquids, methanol, and gasoline-range hydrocarbons
via the Mobil M process are shown in Figure 2-2. These base plant flow
schemes are based upon published designs for existing facilities, and con-
ceptual and proposed designs. They are believed to reasonably represent the
13
-------
COAL PREPARATION OPERATION
COAL GASIFICATION OPERATION
RUN OF
MINE COAL
-
RAW COAL
STORAGE
^
i
COAL
CRUSHING,
DRYING
AND
PULVERIZING
1 / STEAM — >
j \ PREPARED COAL — *•
• | OXYGEN — *•
1 ' QUENCH WATER — *
COAL
GASIFICATION
1
1
1.
1
1
GAS PURIFICATION AND UPGRADING OPERATION
GAS COOLING
AND DUST
REMOVAL
NOX
REDUCTION
RAW GAS
COMPRESSION
AND COOLING
CYANIDE
WASH
BYPASS GAS
H2S
REMOVAL
SHIFT
CONVERSION
CO2
REMOVAL
TRACE
SULFUR
REMOVAL
SULFUR-FREE
SYNTHESIS
GAS
Figure 2-1. Simplified flow diagram for K-T based synthesis gas production
-------
SULFUR-FREE
SYNTHESIS
GAS
FISCHER-
TROPSCH
SYNTHESIS
METHANOL
SYNTHESIS
METHANOL
SYNTHESIS
MOBIL
M-GASOLINE
SYNTHESIS
PRODUCT
SEPARATION AND
UPGRADING
PRODUCT
SEPARATION AND
UPGRADING
PRODUCT
SEPARATION AND
UPGRADING
F-T SYNTHESIS
PRODUCTS
METHANOL
SYNTHESIS
PRODUCTS
MOBIL M-GASOLINE
SYNTHESIS
PRODUCTS
Figure 2-2. Simplified flow diagram for conversion of synthesis gas to liquids
-------
Section 2
Process Overview
configuration options that are likely to be incorporated into the first
generation facilities built in the U.S.
16
-------
Section 2
Feed Characteristics/Product Slate
2.1 COAL FEED CHARACTERISTICS AND PRODUCT SLATE
A wide variety of domestic coals are potential feedstocks for K-T based
synthetic fuels facilities. In general, specific characteristics of the coal
feedstocks will determine the characteristics of process and waste streams.
To date, the North Alabama Coal Gasification Consortium has performed large
scale K-T gasification tests only with an Illinois No. 6 coal. Thus, an
Illinois No. 6 coal was chosen for examination in this PCTM to provide a basis
for sizing and characterizing base plant process and waste streams and eval-
uating waste stream control options and costs. Characteristics of the base
plant Illinois No. 6 coal are summarized in Table 2-1. The effects of differ-
ing feed coal characteristics, particularly with respect to waste stream
generation rates and characteristics, are discussed qualitatively and, to an
extent consistent with available data, quantitatively in Section 3.
TABLE 2-1. PROXIMATE AND ULTIMATE ANALYSES OF BASE PLANT ILLINOIS NO. 6 COAL
Analyses Composition
Proximate Analysis, wt %
Moisture 10.2
Volatile matter 34.7
Fixed carbon 46.0
Ash 9.1
Higher Heating Value, MJ/kg (as received) 26.5
Ultimate Analysis, wt % (dry basis)
Carbon 71.5
Hydrogen 4.8
Nitrogen 1.4
Sulfur 3.1
Chloride 0.28
Ash 10.1
Oxygen (by difference) 9.0
17
-------
Section 2
Feed Characteristics/Product Slate
In developing the base plant material flow estimates, a fixed coal feed
rate of 6682 Mg per day (dry basis) to gasification corresponding to 120 TJ/
day synthesis gas was used. In addition to the gasifier coal requirements,
some K-T based synfuels facilities may include on-site coal-fired auxiliary
boilers for power generation. Coal requirements for any such boilers will
depend upon the amount of purchased electricity from off-site sources.
Boiler fuel requirements could also be offset by using high energy process
waste gases as fuel. Due to the large number of variables that affect the
auxiliary boiler coal requirements, it is difficult to estimate the quantity
of or need for boiler feed coal for each type of synfuels plant examined in
this manual. For facilities which are self-sufficient in energy, auxiliary
boiler energy requirements for electric power are expected to range from
about 4 to 26% of the coal energy input to the gasifier. It should be noted
that most plants are potentially self-sufficient with respect to steam and,
depending upon the cost of electric power, an on-site power boiler may not
be needed at all.
Typical upgraded product slates for the K-T based synfuels plants ex-
amined in this PCTM are summarized in Table 2-2. As indicated in this table,
the quantity and energy content of the products from each type plant varies
even tnough tne same quantity of synthesis gas is assumed in all cases
(i.e., the same coal feed rate to gasification for all cases). This is a
reflection of the different energy conversion efficiencies associated with
the three different synthesis operations examined. The only K-T based in-
direct liquefaction facility proposed in the U.S. (The North Alabama Coal
Gasification Consortium Project) is currently designed for methanol produc-
tion.
18
-------
TABLE 2-2. ESTIMATED PRODUCT/BY-PRODUCT SLATE FOR K-T BASED PLANTS*
Product/ Methanol Synthesis
By-Products Mg/day TJ/day
Gasoline
Diesel Oil
Fuel Oil
LPG
SNG
Alcohols 4710 113
Sulfur 202
Total 4910 113
Mobil M-Gasoline
Synthesis
Mg/day TJ/day
1750 81.0
236 11.7
202
2190 92.7
Fischer-Tropsch
Synthesis
Mg/day
992
201
58.9
74.6
798
171
202
2497
TJ/day
46.9
9.45
2.71
3.72
34.8
5.38
103.0
Coal feed rate to gasification is G682 Mg per day (dry basis) which
corresponds to 120 TJ/day of synthesis gas.
19
-------
Section 2
Base Plant
2.2 BASE PLANT DESCRIPTION
Base plant process operations consist of coal preparation, coal gasifi-
cation, gas purification and upgrading, crude product synthesis and separa-
tion, and product upgrading. In addition, the auxiliary processes required
to support a fully integrated, self-sufficient, liquid fuels production
facility would include raw water treatment, power generation, and oxygen
production. These processes and their associated waste streams are described
briefly in this section. Detailed descriptions are presented in Section 3.
2.2.1 Coal Preparation
The coal preparation operation in a K-T based synfuels facility will be
similar to those found in other coal-based plants such as pulverized coal-
fired power plants. Equipment is provided to receive, transport, and store
coal, and to prepare pulverized coal for gasification and consumption in on-
site power boilers. Coal is received by conveyor, train, barge, or truck and
is stored in either an active or inactive (emergency) storage pile, as neces-
sary. Coal from storage is prepared for gasification/combustion by screen-
ing, crushing, drying, and pulverizing to a size predominantly less than
0.1 mm. Prepared coal moisture contents of 1 to 2% are required for gasifi-
cation of bituminous coals and 8 to 10% are required for gasification of
lignites. Dried and sized coal is stored in silos and transported to gasi-
fiers and boilers as required.
Major waste streams associated with the coal preparation operation are
storage pile runoff; fugitive dust emissions from coal storage and transport;
and dust from coal screening, crushing, and pulverizing. Storage runoff
tends to contain high levels of suspended and dissolved solids and can be
quite acidic in the case of Midwestern and Eastern coals. Dust from coal
preparation consists of natural soil and overburden material as well as coal.
20
-------
Section 2
Base Plant
2.2.2 K-T Coal Gasification
The K-T process involves low pressure (slightly above atmospheric), en-
trained-flow slagging coal gasification in the presence of steam and oxygen.
Flame temperatures may range from 2000-2300K during gasification with re-
actor temperatures of 1510-1860K. The gasifier is a water-cooled steel
vessel with a refractory lining, which, in the most common two-headed config-
uration, resembles an ellipsoid with horizontally opposing burners at the
apices. The newest installations (three gasifiers each at Ramagundam and
Talcher, India) employ four-headed gasifiers which resemble two intersecting
ellipsoids with burners located 90 degrees apart at each of the four apices.
Coal is introduced continuously into the gasifier through screw feeders at
the burner heads and then entrained in a stream of low pressure steam and
high purity oxygen. Partial oxidation reactions occur rapidly within the
central portion of the gasifier; coal residence time is less than a second.
Raw product gas and entrained dust exit the gasifier vertically upward
through a waste heat boiler system producing high pressure saturated steam.
Molten slag exits the bottom of the gasifier and falls into a quench tank
where circulating cooling water causes it to shatter into granular form.
Slag is removed from the quench tank by a drag chain conveyor on which de-
watering occurs prior to subsequent slag disposal.
Waste streams associated with the gasification operation are quenched
slag and transient waste gases. K-T slag is a coarse, pebble sized material
which is physically stable and essentially inert. Quenched slag has about
the same composition as the parent coal ash and retains about 10% moisture.
Depending upon the quench water quality, and whether the slag is rinsed,
contaminants such as NhL and SCN~ may be present in the slag moisture.
Transient gases unsuitable for processing into synthetic liquids may be
generated for short periods of time (less than 1/2 hour) during startup,
shutdown, and unscheduled transient operations. These waste gases will vary
in composition but are similar to either gases from heavy oil combustion or
21
-------
Section 2
Base Plant
raw K-T product gas depending upon the gasifier conditions during the parti-
cular transient period.
2.2.3 Gas Purification and Upgrading
The gas purification and upgrading operation consists of (1) gas cool-
ing and dust removal; (2) NO reduction; (3) raw gas compression and cooling;
X
(4) cyanide washing; (5) shift conversion; (6) acid gas removal; and (7)
trace sulfur removal.
Gas Cooling and Dust Removal
Hot raw gas from the waste heat boiler is cooled and scrubbed of en-
trained dust in two stages by means of direct water contacting. Final de-
dusting is provided by a wet electrostatic precipitator. The collected dust
slurry is pumped to settling basins for thickening. Most of the clear water
overflowing the settlers is indirectly cooled and recycled. The settler
underflow is filtered to produce dewatered dust (up to 50% moisture) and
filtrate which is combined with a portion of the clarifier overflow as blow-
down for controlling the buildup of dissolved components within the washer
system.
Waste streams associated with the gas cooling and dust removal process
are dewatered dust and blowdown. Gasification dust consists primarily of
coal ash and unreacted carbon. The dust is combustible and has leachable
components. In addition, moisture associated with the dust will be similar
in composition to the washer blowdown, containing contaminants such as NH^,
CN~, SCN", S^, SOjj, Cl", and other species.
NOX Reduction
The NO reduction unit catalytically hydrogenates nitrogen oxides and
X
oxygen present in the raw synthesis gas to eliminate associated fouling on
compressor blades and in the acid gas removal system. Although such a unit
is currently in use at an operating K-T based facility, details regarding
22
-------
Section 2
Base Plant
unit performance and reaction chemistry are not available. The only waste
stream expected from this unit is spent cobalt molybdate catalyst which will
periodically require disposal.
Raw Gas Compression and Cooling
Koppers-Totzek technology involves coal gasification at essentially
atmospheric pressure. However, downstream operations such as cyanide wash,
shift conversion, acid gas removal, and liquid product synthesis are more
efficient and economical at elevated pressures. Therefore, raw gas from
cooling and dust removal is compressed to about 3 MPa prior to subsequent
treatment. The temperature rise of the gas during compression is controlled
by intercoolers and an aftercooler, consistent with materials limitations
and temperature requirements of downstream processes.
The principal waste stream from raw gas compression and cooling is com-
pression condensate. Contaminants expected to be present in the condensate
include NH^, Cl", S=, SCN", and CN".
Cyanide Wash
Hydrogen cyanide and any residual ammonia present in the raw synthesis
gas are removed by absorption in either water or cold methanol. In the case
of a water-based cyanide wash, rich wash water from the absorber is flashed
to atmospheric pressure yielding depressurized wash water and a sour flash
gas. In the case of a methanol-based cyanide wash, rich methanol from the
absorber is regenerated by depressurization and indirect heating to also
yield a sour flash gas. Water vapor coabsorbed with the cyanide is re-
covered from the process methanol by distillation.
Major waste streams associated with a water-based cyanide wash process
are depressurized wash water and sour flash gas. Depressurized wash water
is expected to contain CN", S~, and traces of NH». Sour flash gas from a
water-based cyanide wash is expected to consist primarily of CCu and H2S
23
-------
Section 2
Base Plant
with lower levels of HCN. The major waste stream associated with a methanol-
based cyanide wash process is sour flash gas. Sour flash gas from a methanol-
based cyanide wash is expected to consist primarily of H2S, CO, HCN, C02,
COS, H2, and methanol vapor.
Shift Conversion
Gases suitable for feed to methanol synthesis or hydrocarbon production
via Fischer-Tropsch synthesis should have somewhat greater than a 2:1 ratio
of \\2 to CO and no more than a few percent C02- The H2 to CO ratio in raw
K-T gas typically ranges from 1:2 to 1:2.5, well below the ratio required for
liquids synthesis. Thus, a shift conversion step is a necessary part of the
gas upgrading. All commercial scale K-T based coal gasification facilities
employ a shift conversion unit which follows raw gas sulfur removal and
precedes C02 removal, and this approach has been incorporated into the
base plant design. This approach enables the use of conventional iron-
chromium or copper-zinc shift catalysts. Also, due to the favorable
HpS to C02 ratio before shift conversion, it facilitates production of
an FLS-rich offgas for economic sulfur recovery.
Waste streams produced by shift conversion are spent shift catalyst
which periodically requires replacement and shift condensate blowdown which
is expected to be reused as makeup water to the gasification quench circuit.
Acid Gas Removal
Removal of hUS and other sulfur compounds present in the raw K-T gas
is necessary to prevent catalyst poisoning in subsequent shift conversion,
and methanol and Fischer-Tropsch synthesis operations. Bulk removal of C02
is necessary to obtain a composition meeting the stoichiometric requirements
for synthesis feed gas. There are several acid gas removal processes which
have been demonstrated in coal gasification or similar applications. However,
24
-------
Section 2
Base Plant
only the two-stage selective Rectisol process has been included in the base
plant design, since all commercial scale K-T based coal gasification facili-
tates utilize two-stage selective Rectisol units and the only K-T based in-
direct liquefaction facility proposed in the U.S. (The North Alabama Coal
Gasification Consortium Project) is also expected to use this process.
Rectisol is a physical absorption process using low temperature methanol
as a solvent. In two-stage selective Rectisol systems, sulfur compounds are
removed from the raw gas prior to shift conversion and subsequent C02 removal.
This facilitates high process selectivity due to the favorable HUS to CCU
ratio before shift conversion. Sulfur laden methanol from sulfide absorption
is enriched by flashing and stripping a portion of the absorbed C02 and then
regenerated in a hot stripper to produce a sulfur-rich offgas typically con-
taining 25 to 35% H2S. Carbon dioxide laden methanol from the C02 absorber
is regenerated by flashing and stripping with N2 to produce a C02-rich off-
gas. An additional waste stream from the Rectisol system is still bottoms
from a methanol/water distillation unit which controls moisture buildup in
the methanol solvent.
Trace Sulfur Removal
To protect synthesis catalysts from sulfur poisoning, zinc oxide guard
beds may be used following the Rectisol process to remove residual traces of
sulfur compounds. Ordinarily the Rectisol process can attain levels of less
than 0.1 ppmv total sulfur species in the synthesis feed gas, but ZnO provides
for temporary removal during periods of Rectisol process upsets. Periodi-
cally, sulfur guard material must be decommissioned and replaced. This gen-
erates a solid waste consisting of spent ZnO/ZnS.
2.2.4 Product Synthesis
Methanol synthesis and hydrocarbon production via Fischer-Tropsch (F-T)
synthesis can be represented by the following reaction:
25
-------
Section 2
Base Plant
CO + 2H2 -*• CH3OH + heat (Methanol Synthesis)
nCO + (2n + 0.5x) \\2 + CnH2n+x + nH20 + heat (F-T Synthesis)
where n ranges from 1 to about 20, x = 2 for paraffins and x = 0 for olefins.
The mix of F-T products obtained (i.e., the range of n and x values) is
dependent upon several factors including the reactor design, temperature,
pressure, and type of catalyst used. Synthesis gas usually contains some
C02 in addition to CO and Hp. Because synthesis catalysts are also active
for the hydrogenation of COo, the presence of C02 does not create problems
as long as the synthesis gas contains the proper ratio of H2/(CO + C02).
Methanol synthesis employs Cu/Zn-based catalysts at 470K and 3.5 to 7.0 MPa
while F-T synthesis proceeds over iron-based catalysts at 600K and 2.3 MPa
(fluidized bed reactors) or 500K and 2.7 MPa (fixed bed reactors).
Mobil M-gasoline synthesis from methanol can be represented as follows:
nCH3OH -> (CH2)n + nH20 (Mobil M-gasoline Synthesis)
This process employs zeolite-based catalysts and operates at about 570K and
2.2 MPa.
The crude liquid fuel products of methanol, F-T, and Mobil M-gasoline
synthesis processes will require upgrading (probably onsite) to yield final
products which are marketable as substitutes for petroleum-derived fuels.
This is particularly true for motor gasolines, where crude coal-derived
gasoline fractions would not meet octane requirements for the retain market
in the U.S. F-T and Mobil M-gasoline products could be upgraded by catalytic
alkylation of the C3~C4 fraction to yield gasoline-blend hydrocarbons and
commercial grade LPG by hydrotreating (in the F-T case) for destruction of
olefins and oxygenated organics, by catalytic reforming to produce more
cyclic and branched chain hydrocarbons, by C5/Cg isomerization to increase
26
-------
Section 2
Base Plant
the anti-knock quality of pentanes and hexanes, and by catalytic polymeriza-
tion to convert propene/butene fractions into higher molecular weight gaso-
line blending compounds. All of these upgrading processes will utilize con-
ventional petroleum refinery technology. Since the feed streams to these
upgrading processes in an indirect liquefaction plant are not expected to
have any unusual characteristics relative to current refinery experience,
waste streams generated during these upgrading operations are not expected
to present any unique treatment problems. For these reasons and due to the
multiplicity of possible options for product upgrading, waste stream charac-
teristics and pollution control alternatives for product upgrading processes
are not specifically discussed in this PCTM.
All of the synthetic liquid fuels synthesis processes generate a purge
gas containing compounds such as unreacted carbon oxides, hydrogen, methane,
and methanol. Several options are available to handle the purge gas in-
cluding use as an on-site fuel, reforming to generate additional synthesis
gas, or conversion of the residual hydrogen and carbon oxides into methane
to produce SNG. Because supplemental fuel may be required for power genera-
tion in all synthesis cases, use of these purge gases as an on-site fuel was
selected for analysis purposes in this PCTM. In actual practice, the deci-
sion regarding the disposition of synthesis purge gases involves site- and
design-specific considerations which are outside the scope of this manual.
A variety of waste streams are associated with liquid product synthesis
processes, exclusive of upgrading processes. Major waste streams from F-T
synthesis include spent F-T and methanation catalysts, methanation catalyst
decommissioning offgases, SNG dehydration offgases, carbon dioxide offgases,
condensates, and wastewaters. Major waste streams from methanol synthesis
include spent catalyst and synthesis condensate. Major waste streams
from Mobil M-gasoline synthesis include spent methanol and Mobil catalysts,
Mobil catalyst regeneration offgases, and wastewater.
27
-------
Section 2
Base Plant
2.2.5 Auxiliaries
The major additional auxiliary processing units required for self-suffi-
cient K-T based synfuels facilities are: (1) on-site boiler for power gen-
eration; (2) makeup water treatment facilities; (3) process cooling water
system; (4) liquid product/by-product storage facilities; and (5) oxygen
production unit.
The most significant potential source of waste streams from the auxiliary
processes is the boiler. The boiler flue gases are particularly important be-
cause the boilers will generally be coal-fired and are therefore potentially
major sources of SO , NO , and particulate emissions. In addition to flue
X X
gases, waste streams resulting from the boilers include blowdown condensates
and bottom ashes. It should be again noted that most plants are potentially
self-sufficient with respect to steam and, depending upon the local avail-
ability and cost of electric power, an on-site power boiler may not be needed.
The major waste streams from makeup water treatment are sedimentation
pond sludges, lime/soda softener sludges, and demineralizer regeneration
wastewaters from the boiler feedwater treatment unit. Evaporated volatiles,
drift, and cooling tower blowdown are the major waste streams from the cool-
ing water system. Evaporative emissions are the major waste streams from
product/by-product storage facilities.
2.2.6 Fugitive and Miscellaneous Wastes
In addition to the waste streams associated with specific processes,
there are three general categories of wastes which are of non-specific
origin. These categories are fugitive organic emissions, non-process/
intermittent wastewater streams, and equipment cleaning wastes. There are
many potential- sources of fugitive organic emissions in an indirect coal
liquefaction plant. These include pumps, compressors, valves, flanges, and
pressure relief devices. Non-process waste streams include fluid leaks from
sources such as pump seals, valves, and flanges. In addition, drainage
28
-------
Section 2
Base Plant
resulting from emergency process fluid discharges or process area washdown/
cleanup activities will contribute additional intermittent aqueous wastes.
The two primary sources of equipment cleaning wastes in an indirect lique-
faction facility are process equipment and boiler cleaning wastes.
29
-------
Section 2
Costs
2.3 BASE PLANT CAPITAL INVESTMENT AND ANNUALIZED OPERATING COSTS
In order to assess the relative impact of the costs of pollution con-
trols on the total plant cost, approximate costs for uncontrolled base
plants were developed. Both capital and operating costs for K-T plants were
estimated and are presented in Tables 2-3 and 2-4, respectively. Details
of the costing methodology are presented in Appendix A. These base plant
cost estimates are used subsequently in Section 4 to evaluate the relative
magnitude of costs for individual pollution control technologies; however,
since control systems for an integrated facility are not evaluated in this
manual, the total relative cost impact of pollution control has not been
evaluated.
TABLE 2-3. CAPITAL COSTS FOR UNCONTROLLED K-T BASED INDIRECT LIQUEFACTION
PLANTS*
Item
Installed cost
Contractors Overhead and Fee
Engineering and Construction
Contingency
Subtotal, Facility Cost
Interest during Construction
Working Capital
Total Capital Investment
Capital Costs
Methanol
603
18
151
121
893
201
17
1111
, 106 Dollars
Fischer-
Tropsch
714
21
178
143
1056
238
20
1314
(1980 Basis)
Mobi 1
M-Gasoline
657
20
164
131
972
220
_17
1209
Capital cost estimates are based upon estimates published in references
1, 2, 3, 4, 5, and 6. Published estimates were scaled to a plant capacity
of 6682 Mg dry coal per day to gasification. To the extent that they could
be identified, capital costs associated with pollution controls were deleted
from published cost estimates. Details of the costing methodology are pre-
sented in Appendix A.
30
-------
Section 2
Costs
TABLE 2-4. ANNUALIZED COSTS FOR K-T BASED INDIRECT LIQUEFACTION PLANTS*
Coal
Water
Other
Total
, Catalyst, and Chemicals*
Operating Costs*
Operating Cost
Capital Charges
Total Annual i zed Cost
Annual
Methanol
98
4
53
155
191
346
ized Cost,
Fischer-
Tropsch
114
4
63
181
226
407
106 Dollars
Mobi 1
M-Gasoline
95
4
51
150
208
358
*
Annual operating costs relating to "water, catalyst, and chemicals"
and "other operating costs" are based upon published cost estimates for
a K-T based methanol production facility (3). Published estimates were
scaled directly to a plant capacity of 6682 Mg dry coal per day to gasi-
fication. Insufficient details were available to enable adjustment, if
any is required, for F-T and Mobil M-gasoline synthesis cases or to deduct
the annual operating costs for pollution controls. Details of the costing
methodology are presented in Appendix A.
Installed costs are based upon published cost estimates for K-T based
methanol and hydrogen production facilities (3,4). Published cost data were
adjusted to reflect differences in costs for methanol, F-T, and Mobil M-
gasoline synthesis processes (1,2,3,5,6). To the extent that pollution con-
trol costs could be quantified, they were deleted from published installed
costs. The resulting installed cost estimates were then scaled to the base
plant capacity using a scaling exponent of 0.8. For analysis purposes, the
adjusted base plant installed costs were escalated to 1980 dollars using the
Chemical Engineering (CE) plant cost index. Total uncontrolled base plant
capital investments were estimated to be $1.1, $1.3, and $1.2 billion (1980
basis) for the methanol, F-T, and Mobil M-gasoline synthesis cases, respec-
tively. Differences among the estimated capital investment costs derive
31
-------
Section 2
Costs
primarily from differences in costs of the synthesis operation and, to a
lesser extent, differences in on-site boiler costs.
The total operating costs are based upon the annual coal cost and pub-
lished cost estimates for "water, catalyst, and chemicals" and "other opera-
ting costs" in a K-T based methanol production facility (3). Published
operating cost estimates were scaled directly on the basis of plant capacity
and escalated to 1980 dollars. Published cost estimates for "water, catalyst,
and chemicals" and "other operating costs" relate specifically to a methanol
production facility; however, no adjustment has been made for F-T and Mobil
M-gasoline synthesis cases, if any is required, since insufficient details
of these estimates are available to do so. Similarly, the annual operating
costs of pollution control equipment have not been deducted from cost esti-
mates for "water, catalyst, and chemicals" and "other operating costs" since
insufficient details of these estimates are available to do so.
It should be noted that annual coal costs and annualized capital charges
comprise about 84% of the total annualized cost. Therefore, uncertainties
associated with estimated costs for "water, catalyst, and chemicals" and
"other operating costs" are not expected to have a major impact on the esti-
mated total annualized cost. Total annualized costs were estimated to be
$346, $407, and $358 million for the methanol, F-T, and Mobil M-gasoline
synthesis cases, respectively.
32
-------
Section 3
Detailed Charac.
SECTION 3
PROCESS DESCRIPTION AND WASTE STREAM CHARACTERIZATION
This section defines the basic operations and auxiliary processes assoc-
iated with Koppers-Totzek (K-T) based indirect liquefaction facilities, and
the characteristics of major associated process and waste streams. Material
flow estimates are also presented to provide pertinent information relating
to the magnitudes of the various waste streams and their environmental con-
trol needs. Operations covered by this manual are shown schematically in
Figures 3-1 and 3-2. These operations include coal preparation, coal gasi-
fication, gas purification and upgrading, crude product synthesis and sep-
aration, and product upgrading. Auxiliary processes are shown in Figure 3-3
and include all of the operations which may be required to support a fully
integrated, self-sufficient, liquid fuels production facility. These pro-
cesses include raw water treatment, oxygen production, and product storage.
Integrated K-T facilities are ordinarily more than self-sufficient with
respect to steam requirements. Also, on-site power generation is not expected
to be economical in the U.S. as compared with the purchase of power. However,
on-site power generation using coal as boiler fuel is examined in this docu-
ment in an attempt to cover a broad range of possible plant configurations.
Stream numbers have been provided in the figures for major process streams
and all waste streams to clarify stream identification in subsequent sec-
tions. A stream identification listing and cross referencing indicies are
provided in Section 3.8. Figures 3-1 through 3-3 are based primarily upon
published designs for existing facilities (e.g., the AECI Limited facility
at Modderfontein, Republic of South Africa) and conceptual and proposed de-
signs (e.g., the North Alabama Coal Gasification Consortium Project). They
are believed to reasonably represent the configuration options that are likely
33
-------
FUGITIVE FUGITIVE
DUST DUST
DRVER
OFF GAS
FUGITIVE
DUST
FUGITIVE
DUST
CJ
PREPARED
COAL
/~\
RUN OF /T\ fe RAW COAL ^ COAL CRUSHING k -
MINE COAL V^/ STORAGE AND TRANSFER *"
„,„ ,. ~|
1 |
^
WASTE GAS
t
,208,
1
HAI DRviittr » COAL . . PULVERIZED
OAL DRYING » PULVERIZING * COAL STORAGE
., . * 1
V'
PREPARED COAL
TO BOILER
L H2S RICH
I OFF GAS
LEGEND
•_ ^ mm Indicates nte-rt
NU I C Trace iui'ui .Cp
aad'^MfeTO?!
itrent flow
t'eamo* RECTlSOL
"SPENT SULFUR
GUARD
Figure 3-1. Operations associated with synthesis gas production in K-T based indirect liquefaction
facilities
-------
oo
en
DEHYDRATION
OFF-GAS
FUEL GAS CATALYST REGENERATION/
TO BOILER DECOMMISSIONING OFF-GAS
t
WASTEWATER
CRUDE PRODUCTS
GASOLINE
MIXED BUTANES
PROPANE
CATALYST * -(232) (3)—». CONDENSATE
Figure 3-2. Synthesis operations associated with K-T based indirect liquefaction facilities
-------
EVAPORATIVE EMISSIONS PROM
PRODUCT STORAGE
CO
CTi
RAW
WATER '
METHANATION
CONOENSATE
HOLDING PONDS
(
RAW WATER
TREATMENT
i) d
+
SLUDGES
1
D
BOILER
FEEDWATER
MAKEUP
TREATMENT
/~\ ^
(301J >-R
\^ w
MAKEUP
WATER
(X6) ^E
l'^ D
COOLING
TOWER
BOILER
FEED WATER
DEMINERALIZER
REGENERATION
WATESWATER
PREPARED COAL-
EVAPORATION/
DRIFT
POWER
GENERATION
(SEE NOTEI
BOILER F"LUEGAS
BOILER SLOWDOWN
BOILER BOTTOM ASH
BOILER CLEANING WASTES
OXYGEN TO
GASIFIER
NOTE ON-SITE POWER GENERATION MAY
NOT BE REQUIRED, DEPENDING ON
THE AVAILABILITY AND COST OF
LOCAL ELECTRICITY
Figure 3-3. Auxiliary processes associated with K-T based indirect liquefaction facilities
-------
Section 3
Detailed Charac.
to be incorporated into the first generation K-T facilities built in the
U.S.
Flow estimates for major process and waste streams are presented in
Table 3-1 for facilities based on an Illinois No. 6 coal. The base plant is
sized for a fixed coal input to gasification. Therefore, the material flows
presented for gasification and gas purification and upgrading are identical
for all three synthesis alternatives considered. Since the overall thermal
efficiency of indirect liquefaction differs among synthesis routes, energy
to on-site auxiliary boilers for self-sufficient facilities will be strongly
dependent on the specific synthesis process. To a lesser extent, this is
also true with respect to material flows through coal preparation and raw
water treatment. Hence, process and waste streams associated with the power
boiler and most other auxiliary processes will be synthesis specific.
Numerical values presented in Section 3 are based upon both published
data and engineering estimates derived from published data and material
balance considerations. Material flow estimate tables in this section gen-
erally contain both published data and engineering estimates. Detailed
references are provided for published data in both the text and below each
table; values which are not referenced represent either summaries of published
data which cannot be referenced in summary format or engineering estimates.
37
-------
TABLE 3-1. FLOWS OF MAJOR STREAMS FOR K-T BASED INDIRECT LIQUEFACTION FACILITIES - ILLINOIS NO. 6 COAL'
CO
00
Component Flows, kmol/hr
»2
CO
co2
N2 + Ar
CH4
S
Total Dry Gas
Moisture in Gas
Total gas, kmol/hr
Total Aqueous, m /hr
Total Dry Solids, Mg/hr
Quench Quenched Dewatered Raw Gas From
Prepared Coal Steam Oxygen Water Slag Dust Washer Cooler
Stream 2 Stream 3 Stream 4 Stream 5 Stream 207 Stream 209 Stream 7
5829
13699
1907
300
22
259
8430 22019
2082 964
2082 8430 22982
2.8 93 1.0 30
278 9.3 30
Washer Cooler
Slowdown
Stream 210
322
(Continued)
-------
TABLE 3-1. (Continued)
Component Flows, kmol/hr
Synthesis
Feed
Stream 14
Fuel Grade
Methanol
Stream 108
Methanol
Disti llation
Wastewater
Stream 229
F-T SNG
Liquids and LPG
Streams 100-107
F-T
Wastewater
Stream 223
Mobil-M
Liquids + LPG
Streams 109-111
Mobil-M
Wastewater
Stream 233
"2
CO
13664
5719
CO,
610
CO
10
CH,
Ar
307
21
Total Dry Gas
Moisture in Gas
20320
Total Gas, kmol/hr
Total Aqueous, m /hr
Total Organic Liquids,
Mg/hr
20320
0.2
196
9.7
0.17
96
159
1.5
83
110
0.92
(Continued)
-------
TABLE 3-1. (Continued)
Component Flows, kmol/hr
H2
CO
co2
N2 + Ar
CH4
S
Total Dry Gas
Moisture in Gas
Compression HCN HCN Flash Gas HCN Flash Gas
Condensate Wash Water (Water Wash) (Methanol Wash)
Stream 211 Stream 215 Stream 21 4a Stream 21 4b
0.5
4.4
20 1.2
2.3 7.9
22 16
HgS-Ricn
Offgas
Stream 216
269
6
255
535
Shift C02-Rich
Slowdown Offgas
Stream 218 Stream 219
46
101
8875
1041
1.2
0.1
10064
Rectisol
Still Bottoms
Stream 220
Total gas, kmol/hr
Total Aqueous, in /hr
Total Dry Solids, Mg/hr
22
16
535
17
238
10064
(Continued)
-------
TABLE 3-1. (Continued)
Coal to
Boiler-t
Component Flows, kmol/hr Stream 30
N2 + Ar
co2
°2
NOX
so2
Total Dry Gas
Moisture in Gas
Total Gas, kmol/hr
o
Total Aqueous, rr /hr 3.2
Total Dry Solids, Mg/hr 78.7
Boiler Boiler Boiler ,
Air to Flue Gas1" Bottom Ashf Fly Ash1"
Boiler"1" Stream 302 Stream 304 Stream 423
24356 24388
10 4597
6455 1065
12
69
30821 30133
651 2581
31472 32713
1.8 7.6
For overview purposes, only major streams and their principal components have been
included. Detailed stream characterization data are presented subsequently in this
section.
These streams are representative of the estimated maximum boiler size for a K-T based
indirect liquefaction plant with a coal feed of 278 Mg/hr (dry basis) to the gasifier.
-------
Section 3
Coal Preparation
3.1 COAL PREPARATION
Run-of-mine (ROM) coal is received from the mine via conveyor and stored
in an active storage pile which holds a five-day supply. The five-day active
storage pile (40,000 Mg for methanol synthesis case) normally provides the
gasifier feed. Raw coal is also kept in an emergency storage pile contain-
ing a 30-day supply. The 30-day emergency pile (241,000 Mg for methanol
synthesis case) is built up over a period of time, when coal is not required
at the active storage pile, and is used if the coal supply from the mine is
interrupted for an extended period of time. Coal storage piles for the
Fischer-Tropsch synthesis case are approximately 16% larger than the coal
piles for the methanol synthesis case. However, coal piles for the Mobil
M synthesis case are approximately 4% smaller in size. These changes in size
are due to the different coal feeds to the boiler in each of the three syn-
thesis cases.
Because the K-T gasification process requires free flowing pulverized
coal, some level of drying during coal preparation is necessary for all coals.
The extent of drying which is required is coal specific, with residual mois-
ture levels of approximately 1 to 2% required for bituminous coals and 8 to
10% required for lignites. Recent tests in Greece (7) indicated that drying
of Illinois coal to a 1% moisture level might be required. The particle size
of the feed coals to the K-T gasifier is predominantly less than O.lmm. The
allowable portion of oversize coal is about 10% in the case of bituminous
coal, and 15 to 20% in the case of lignites (8) . The particle size of
boiler feed coal is typically 70% less than 0.075mm. Table 3-2 presents the
characteristics of the Illinois No. 6 coal under consideration.
A schematic diagram of the coal preparation plant is presented in Figure
3-4. Run-of-mine coal with a top size of 10.2 cm is transported to the pre-
paration plant by a belt conveyor. It is then transferred through the feed
42
-------
TABLE 3-2. CHARACTERISTICS OF ILLINOIS NO. 6 COAL SELECTED FOR USE IN INDIRECT
LIQUEFACTION BASE PLANT (9,10)
Moisture
Volatile Matter
Fixed Carbon
Ash
C
H
0
S
N
Heating Value
MJ/kg
Major and Minor Elements in
(%, on moisture-free whole
Al
Ca
Cl
Fe
K
Trace Elements
(ppm, on moisture-free whol
As
B
Be
Br
Cd
Co
Ce
Cu
F
Ga
Ge
As Revised Basis (wt %)
10.23
34.70
46.0
9.15
64.16
4.34
8.1
2.81
1.21
26.63
Coal
coal basis)
1 . 20 Mg
0.92 Ma
0.28 Si
1 . 50 Ti
0.16
e coal basis)
1.0 Hg
132 Mo
1.6 Mn
17 Ni
<0.4 P
4 Pb
20 Sb
12 Se
79 V
4.5 Zn
6.0
Dry Basis (wt %)
10.19
71.47
4.83
9.02
3.13
1.35
29.67
0.04
0.060
2.45
0.06
1.1
7
20
14
29
10
0.1
1.3
20
43
43
-------
LEGEND:
INDICATES THE
MAIN PROCESS
FLOW
INDICATES
EMISSIONS
FLOW
PARTICULATES
PARTICULATES
ROM COAL
STORAGE PILE
TO MAIN BAGHOUSE
RUNOFF
PARTICULATES (TO
MAIN BAGHOUSE)
SCREEN
CRUSHER
STORAGE
SILO
PARTICULATES
[205
PRODUCT
SILO
r
PULVERIZER
i
r
4-
i
r
THERMAL
DRYER
i
±
' i
-
PULVERIZER
' 1
r
1
PRODUCT
SILO
I
TOGASIFIER 4~ '
[204]
PARTICULATES
I
1 ^ TO BOILER
Figure 3-4, Waste streams associated with coal preparation for a K-T based indirect liquefaction
facility - Illinois No. 6 coal
-------
Section 3
Coal Preparation
hopper to the raw coal screens that make a size separation at 1.9 cm. The
oversize coal (10.2 cm x 1.9 cm) is conveyed to double roll crushers where it
is reduced to a top size of 1.9 cm. The undersize coal from the sizing
screens and the crusher product are transferred to storage silos by a belt
conveyor. Coal from the storage silo is transported to the pulverizer after
it is ground to 70% minus 20 mesh. The moisture content of the coal is re-
duced in the thermal dryer and the pulverizer from 10% to 1-2% for the K-T
gasifier. Coal to the boiler does not require thermal drying since the
pulverizers reduce the moisture content from 10% to a desirable 4%. The dry
ground coal from the pulverizers is transferred to the product silos by gravity
and then continuously fed to the boiler and K-T gasifier. Coal preparation
mass flow estimates for the methanol synthesis case are presented in Table
3-3.
Major waste streams associated with the ROM coal storage and handling
operations are particulate emissions from storage (Stream 200), storage pile
runoff (Stream 201), particulate emissions from crushing/screening/transfer/
pulverizing (Streams 202, 203, 204), and particulate emissions from prepared
coal storage and feed (Stream 205). Sources of these wastes streams are
shown in Figure 3-4.
The quantities of fugitive particulates generated by the five-day active
and 30-day emergency storage piles are shown in Table 3-4. Emissions from
coal storage piles have been investigated and results presented in several EPA
reports (11,12,13). Estimates for uncontrolled emissions from various acti-
vities were compiled from information given in the references indicated in
Table 3-4. Environmental assessment studies data showed relative high total
suspended particulate values within 200 meters of the coal and refuse piles
when compared to 24-hour primary and secondary ambient air quality standards
(14,15). Ambient concentrations decrease sharply at 500 to 600 meters downwind.
Particulate morphology tests showed that downwind particles were primarily a
quartz material rather than coal particles indicating that the source of
45
-------
Section 3
Coal Preparation
participates is activity around the piles, not the coal itself. The prepara-
tion plants were found to contribute no gaseous organic matter to the environ-
ment, other than trace amounts attributable to diesel truck activity around
the coal pile.
TABLE 3-3. COAL PREPARATION SECTION MASS FLOWS - METHANOL SYNTHESIS CASE
Stream
ROM Coal to
Screen*
Coal to Crusher*
Coal from Screen*
Coal from Crusher*
Coal to Storage
Silo*
Coal from Storage
Silo*
Coal to Product
Silos*
Coal to Gasifier*
Coal to Boiler1"
Hours per Day
of Operation
13
13
13
13
13
24
24
24
24
Dry
Mg/day
7225
4624
2601
4621
7225
7225
7225
6682
544
Illinois No..
Basis
Mg/hr
556
356
200
356
556
301
301
278
23
6 Coal
Moist Basis
Mg/hr
619
396
223
396
619
335
305
281
24
These streams increase by approximately 16% for the Fischer-Tropsch (F-T)
synthesis case and decrease by 3-6% for the Mobil M synthesis case.
Coal to boiler increases by 215% for the F-T synthesis case and decreases
by 48% for the Mobil M synthesis case.
46
-------
Section 3
Coal Preparation
TABLE 3-4. ESTIMATED FUGITIVE DUST EMISSIONS FROM COAL STORAGE PILES
Activity
Wind erosion*
Loading on+
Loading off*
Vehicular
activity5
TOTAL
Reference
12
13
13
13
Emission
Active Storage Piles
(5-day storage)
10
28
36
1
75
Estimates* (Mg/yr)
Emergency Storage Piles
(30-day storage)
57
28
36
6
128
Based on a respirable emission factor of 6.4 mg of dust per kg/yr of coal
stored
"^Assumed activity factor of 0.75, silt content of 0.5% and Thornthwaite's
precipitation index (PE) of 93 for S.W. Illinois
^Assumed activity factor of 0.77, silt content of 0.5% and PE of 93
§Assumed activity factor of 0.5, silt content of 0.5% and PE of 93
^Emissions will increase by 16% for the Fischer-Tropsch synthesis case and
decrease by 3.4% for the Mobil M synthesis case
Runoff streams originating from coal storage piles have been character-
ized in detail during environmental assessment testing programs at two coal
preparation plants (14,15). The assessment results indicated that runoff
water quality parameters complied with the most stringent state effluent
regulations for eastern and midwestern states.
Laboratory leaching tests with raw coal and coal refuse materials have
indicated that the types and quantities of pollutants released from coal
storage piles are similar to those produced by coal refuse piles (16).
Assuming that this similarity carries over to the pollutant loadings gene-
rated by actual refuse and coal piles, the information available from refuse
pile pollutant analysis provides a source of data. In addition, there are
47
-------
Section 3
Coal Preparation
some data available on actual effluents produced by high-sulfur coal storage
piles (17). Information for these two sources, coupled with what is known
about the composition of the subject coal, allows at least a range estimation
of the composition of the storage pile runoff produced.
Runoff from storage piles of Midwestern coals (such as the Illinois No.
6) and Eastern coals can be highly acidic, with pH values in the range of
2 to 4, if the runoff is allowed to remain in contact with the coal for long
periods of time. Total suspended solids during storm runoff can be as high
as 2300 mg/L. Sulfate concentrations may be in excess of 9000 mg/L. Iron
concentrations can range from 23 to 1800 mg/L, while manganese concentrations
can range from 1.8 to 45 mg/L. Other elements that are at concentrations of
potential concern include aluminum, mercury, arsenic, and zinc (17).
At facilities utilizing either sub-butiminous coal or lignite, runoff
from the coal storage piles would be expected to be close to neutral, with
a pH slightly above 7. Sulfate concentraions would probably be less than
1000 mg/L. Iron and manganese concentrations would be expected to be low,
less than 0.8 mg/L for iron and below 0.4 mg/L for manganese (18). The
dominant water contaminants are expected to be calcium and magnesium, with
concentrations in the ranges of 200 to 400 mg/L for calcium and 50 to 250
mg/L for magnesium (19). Total suspended solids levels would probably be
higher than those present in Illinois No. 6 runoff because of the tendency
of these coals to slake.
Published data on particulate emissions for coal operations are limited
and were taken from surface coal mining and ore mining operations, the crushed
stone industry, and the manufacturing of coke. Reported emission factors
(13,20) for specific operations within preparation plants are based on esti-
mates rather than actual data. There is no consensus on the best available
values for use with specific coal preparation operations. In one surface
mine study, an average uncontrolled particulate emission factor of 0.2 kg/Mg
48
-------
Section 3
Coal Preparation
of coal processed was used for loading and unloading activities in all modes
of transport (21). In another surface mine investigation, a factor of 0.05
kg/Mg of coal mined was used for coal loading and unloading operations,
respectively (20). The moisture contents and coal types used in these
studies were not specified, and thus no distinction can be made between
bituminous, sub-bituminous, and lignite coals. However, it is commonly
reported that western coals and lignites generate several times as much dust
as comparable amounts of eastern coal (22).
Table 3-5 provides estimates of emissions generated during the specific
operations, such as unloading, crushing, and screening. The streams assoc-
iated with the estimated emission values are shown in Figure 3-4.
In the absence of more specific data, simplifying assumptions were
necessary in preparing these estimates. First, the same average factors
were used for similar operations, such as screening, disregarding the effect
of particle sizes (i.e., the same factor is used for secondary and for terti-
ary screening). Since the reported emission factor was based upon combined
crushing/screening operations, only 50% of the published factor was used for
emissions from screening alone or crushing alone. In addition, the moisture
content of the coals from which the emission factors were generated was un-
known; therefore, corrections for moisture content were not applied.
The quantities of particulate emissions generated by the various opera-
tions associated with the storage and subsequent handling of the screened,
crushed coal are included in Table 3-5. As with crushing/screening emission
factors, no distinctions have been made for differences in particulate emis-
sions associated with lignite, sub-bituminous, and bituminous coals.
The emission factors used to calculate emissions at transfer points to
the gasifier and boiler feed equipment are the same as those used for trans-
fer points inside the preparation plant.
49
-------
TABLE 3-5. PARTICULATE EMISSIONS FROM COAL PREPARATION (ILLINOIS NO. 6 COAL)
en
o
Discharge
Stream
Number
202
202
202
202
202
203
203
203
203
204
204
204
205
205
205
205
206
Hours per
Day of
Emissions Source Operation
Loading
Transfer to screening
Screening
Transfer to crushing
Crushing
Transfer to silo
Transfer to silo
from screening
Transfer to storage
Inert gas purge
Transfer to pulverizer
and thermal dryer
Transfer to Pulverizer
from thermal dryer
Pulverizer
Transfer to product
silo
Transfer to boiler
Transfer to gasifier
Inert gas product silo
purge
Thermal dryer
13
13
13
13
13
13
13
13
13
24
24
24
24
24
24
24
24
Estimated Average
Emission Factors
Uncontrolled
kg/Mg
Negl igible
0.02 - 0.48
0.04*
0.02 - 0.48
0.04*
0.02 - 0.48
0.02 - 0.48
0.02 - 0.48
0.02 - 0.48-1-
0.02 - 0.48
0.02 - 0.48
Negligible
0.02 - 0.48
0.02 - 0.48
Negligible
0.02 - 0.48f
10
Reference
20
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
Feed Rate
As Received
Basis
Mg/hr
619
619
619
396
396
396
223
619
619
335
290
-
314
24
290
314
290
Uncontrol led
Emission
Rates
kg/hr
12.4
7.9
7.9
4.5
12.4
12.4
6.7
5.8
6.3
0.5
6.3
_
- 297*
25*
- 190*
16*
- 190*
- 107*
- 297*
- 297*
- 161*
- 139*
-
- 151*
- 11. 5§
-
- 151*
435
Emission factor as reported for combined "secondary crushing/screening operations" is 0.08 kg/Mg of coal
processed. Assumed contribution from crushing is equal to that from screening and is thus equal to 0.04
kg/Mg of coal processed.
fAssume same emission factor as that of transfer.
^Emissions from these streams increase by approximately 16?:'- for the Fischer-Tropsch synthesis and decreases
by 3-4% for the Mobil^M synthesis case.
Emissions from this stream increase by 215% for the Fischer-Tropsch synthesis case and decrease by 48%
for the Mobil-M synthesis case
-------
Section 3
Coal Preparation
Emissions from Thermal Dryers (Stream 206)
In a thermal dryer, hot gas from a furnace is forced past wet coal in
the drying chamber to volatilize coal moisture. Drying gas is generated by
combustion of coal or fuel gas. In the case of coal combustion, dryer off-
gases would be similar in composition to flue gases from the boiler system
with respect to gaseous pollutants such as SO , NOV, and CO. Particulate
X A
loadings would be higher due to entrainment of dried coal. Low Btu fuel
gases available for coal drying purposes at this type of facility are expected
to be essentially free of sulfur and nitrogen compounds. The gaseous pollu-
tants associated with use of these gases from thermal drying are NO and CO.
/\
The exact quantity of dryer offgases generated is dependent upon the quality
of dryer fuel used, ROM coal moisture, and residual coal moisture require-
ments. The composition of dryer offgases with respect to SO. NOV, and CO
X X
is dependent upon the quality of dryer fuel used, and is not unique to the
coal preparation operation; offgas compositions will be discussed in conjunc-
tion with auxiliary combustion processes (Section 3.6.2). It should be noted
that offgases from coal drying may contain low levels of volatile organic
compounds (VOC), although essentially no data are available regarding such
emissions.
The amount of coal particles entrained in the drying chamber exit gas
stream is significant. One study estimated an uncontrolled particulate emis-
sion factor from thermal dryers to be 10 kg/Mg of coal dried (23). In order
to recover these substantial quantities of dry airborne product coal, mechan-
ical collectors are generally included as process equipment for thermal dryers.
Uncontrolled fugitive emissions from the dryer were calculated and are
assumed to be those from the mechanical collector outlet. For a gasifier feed
rate of 290 Mg/hr uncontrolled emissions from the thermal dryer are estimated
to be 435 kg/hr.
51
-------
Section 3
Gasification
3.2 COAL GASIFICATION
Gasification consists of the partial oxidation of pulverized coal to
produce raw synthesis gas. The Koppers-Totzek (K-T) process is characterized
by low pressure (slightly above atmospheric), entrained-flow slagging coal
gasification, featuring rapid partial oxidation of pulverized coal in a
stream of oxygen and steam. Flame temperatures may range from 2000-2300K
during gasification, with reactor temperatures of 1500-1900K (7,24,25,26).
The gasifier is a double-walled, water-cooled steel vessel with a refractory
lining. Low pressure steam is generated in the water cooling system for
process use. While older commercial installations utilize a two-headed
gasifier configuration resembling an ellipsoid with horizontally opposing
burners at the apices, the newest installations (three gasifiers each at
Ramagundam and Talcher, India) employ four-headed gasifiers. The four-headed
configuration resembles two intersecting ellipsoids with burners located 90
degrees apart at each of the four apices. Essentially all ranks of dried
coal (1 to 8% moisture) with ash contents of up to 40% can be gasified.
Four-headed K-T gasifiers are capable of processing up to 800 Mg of coal per
day.
The K-T gasification process is depicted in Figure 3-5. A mixture of
dried, pulverized coal, low pressure steam, and high purity oxygen are con-
tinuously injected into the gasifier with injection speeds exceeding flame
propagation speeds to prevent flash back. Coal is introduced into the gasi-
fier through screw feeders at the burner heads. This system provides for
(1) a high degree of turbulence and mixing; (2) continuous ignition should
one burner become temporarily blocked; (3) flue gases being directed into
the center of the gasifier, thus minimizing hot spots in the refractory
lining; and (4) unreacted coal particles from one flame region being gasified
in the opposing region. Reaction between the coal and oxygen upstream of the
gasification zone is inhibited by maintaining moderate temperatures in the
mixing nozzles through circulation of cooling water. The gasification
52
-------
RAW QUENCHED
GAS TO DUST
REMOVAL
SATURATED
STEAM (10 MPa)
SATURATED STEAM (0.1 MPa)
QUENCH WATER
OXYGEN
PREPARED COAL
STEAM
QUENCHED
SLAG TO DISPOSAL
FEEDWATER
PREHEATER
FEED
WATER
VAPORIZER
WASTE
HEAT BOILER
CLINKER
OUTLET
QUENCH
WATER RETURN
SLAG QUENCH
WATER
SLAG QUENCH TANK
Figure 3-5. GKT's gasifier with waste heat boiler and slag extraction system
(26,27)
53
-------
Section 3
Gasification
reactions occur in a cocurrent stream of suspended coal particles, with raw
product gas and entrained particles exiting the gasifier vertically. Coal
residence time in the gasifier is less than one second.
Coals vary widely in terms of their reactivity or degree of carbon con-
version to useful gas (CO plus hL) as a function of temperature in a K-T
gasifier. Coals also differ greatly in their ash fusion temperatures and
molten ash viscosities. The temperature of gasification for a particular •
coal in a K-T gasifier is largely dictated by the coal ash properties since
a layer of slowly flowing molten slag must be maintained on the sides of the
gasifier to protect the refractory lining. Temperature control in the gasi-
fier is achieved by addition of moderating steam along with oxygen. Lignitic
coals commonly show higher reactivities (low gasification temperatures for
high carbon conversion) at the temperatures needed for proper slag flow.
Bituminous coals require higher gasification temperatures than lignites to
obtain a high degree of carbon conversion, and thus the ash (slag) behavior
in such coals is more critical. Generally, lower rank coals will show
higher carbon conversions in a K-T gasifier than higher rank coals. Unreacted
carbon along with a fraction of the original coal ash exits the gasifier as
dust entrained in the hot raw product gas (Stream 7).
That portion of the coal ash which impinges as molten slag on the gasi-
fier walls eventually exits the bottom of the gasifier into a quench tank.
Circulating cold water causes the slag to shatter into a granular form, and
a continuous conveyor system removes the granulated slag (Stream 207) from
the tank. The relative amounts of coal ash exiting the K-T gasifier as
slag or dust varies among coals. The dust-to-slag ratio (carbon free basis)
can be as low as 1:10 for some lignites, and as high as 2:1 for some bitu-
minous coals.
The composition of K-T gas has been measured at commercial facilities
gasifying Indian bituminous coal (28), South African sub-bituminous coal
54
-------
Section 3
Gasification
(o,24,29), and Turkish lignite (8,30). Gas characterization data are
also available from commercial-scale gasification tests performed with an
Illinois No. 6 coal (7) and with Char-Oil-Energy Development (COED) process
chars derived from Western Kentucky and Pittsburgh seam coals (25). These
data indicate that the bulk composition of K-T gas is determined by the
water-gas equilibrium and the relative amounts of coal, steam, and oxygen.
Raw gas consists of 55-65% CO, 25-30% H2> and 6-12% C02 on a dry basis.
Small amounts of CH. are present in K-T gas (ppmv levels) but no tars, oils,
phenols, or high molecular weight organics have been found. Coal sulfur is
largely gasified forming H2S and COS at a ratio of about 9:1. Traces of CS^
and S02 are also found. A fraction of the original coal sulfur will also be
contained in the slag and dust in the raw gas. Nitrogen in the coal is mostly
converted to elemental nitrogen (over 90%), although side reactions also lead
to the formation of NH^, HCN, and NO in the ppmv range. Limited data indi-
cate that ammonia levels are higher with low rank coals than with high rank
coals, but the data do not show a convincing similar trend in the case of HCN.
Water sprays are utilized at the gasifier outlet to reduce the gas tem-
perature to below the ash fusion temperature, approximately 1360-1500K.
This solidifies molten entrained dust to prevent its adherence to waste heat
boiler tubes. Raw product gas exiting the gasifier flows vertically through
a waste heat boiler system producing high pressure steam, thereby recovering
heat rejected by the gasifier, and cooling raw gas to approximately 570K.
Material flow estimates for the coal gasification operation are presented
in Table 3-6.
Quenched Slag (Stream 207) and Slag Quench Water
K-T slag is a coarse, pebble sized material which is physically stable
and essentially chemically inert. Due to the irregular surfaces/shapes
which form during slag solidification in the quench tank, the material can
be crushed and broken somewhat by handling and compacting. Quenched slag
retains about 10% moisture.
55
-------
TABLE 3-6. MATERIAL FLOW ESTIMATES FOR K-T GASIFICATION (ILLINOIS NO. 6 COAL)1
01
01
Prepared Coal Steam Oxygen
Stream 2 Stream 3 Stream 4
kmol/hr kg/hr kmol/hr kg/hr kmol/hr kg/hr
H,
CO
co2
CH4
H2S
COS
cs2
so2
so=/s,o;
so3
02 8261 264342
HCN
HSCN
NO
NH3
N2 65 1817
Ar 104 4145
Cl 22 780
F 1.3 22
Total Dry Gas
H20 156.1 2810 2082 37500
Ash 28400
Coal (MAP) 250000
C 16565 198970
H 13343 13450
S 272 8710
0 1519 24308
N 268 3760
Total- kmol/hr 2082 8430
kg/hr 281210 37500 270303
Temperature(K)
Pressure(MPa)
Quench Water Quenched Slag
Stream 5 Stream 207
kmol/hr kg/hr kg/hr
0.4
0.001
0.4
0.03
0.01
0.004
0.16
2.2
0.07
5134 92496 1037
9245
48
38
5134
92496 10371
300
0.1
Raw Quenched Gas
Stream 6
kmol/hr Vol "•
5829 26.4
13699 62.1
1910 8.7
22 998 ppmv
229 1.0
26 1166 ppmv
2 89 ppmv
0.5 21 ppmv
2.6 118 ppmv
0.07 3 ppmv
4.0 180 ppmv
196 0.9
1 04 0.5
22 998 ppmv
1.3 58 ppmv
22047
7919
18351 kg/hr
10805 kg/hr Dust
30 kg/hr
419 kg/hr
29966
672242
450
0.1
The number of significant figures shown in some cases do not represent the degree of accuracy and are retained for material balance
purposes only Nevertheless, slight imbalances do appear as a result of numerical rounding. Material flow estimates and stream composi-
tions are based upon published data and engineering estimates. Tabulated data are based upon references /, 8, 24, 2b, and M.
-------
Section 3
Gasification
Slag from the gasification of Illinois No. 6 coal has essentially the
same overall composition as that of the parent coal ash. The quenched slag
contains about 0.5% carbon by weight (7). Quenching of the slag is integral
with gasification and involves intimate contact with circulating water (e.g.,
from gas cooling and dust removal); hence slag from the quench tank is leached
of any readily soluble material. Table 3-7 summarizes the results of leach-
ing tests on quenched slag from K-T gasification of an Illinois coal at the
Nitrogenous Fertilizer Industry (NFI), Ptolemais, Greece (31). The NhU, SCN ,
and COD values shown in Table 3-7 reflect residuals derived from washer cooler
water used for slag quenching. Such species are not believed to be present
in measurable amounts when non-process water is used for quenching or for
rinsing the quenched slag.
TABLE 3-7. RESULTS OF K-T SLAG LEACHING TESTS (31)
Element
Ag
As
Ba
Cd
Cr
Hg
Pb
Se
NH3
SCN-
COD
Concentration
KCKA Lxtract AS
pH 5
<0.2
<8
<0.2
<0.1
<0.8
<0.004
<1
<8
in rng/kg SI
ag
TM Neutral Extract
pH 7
<0.2
<8
<0.2
<0.1
<0.8
<0.004
<1
<8
14
<2
100
57
-------
Section 3
Gasification
The slag generation rate in the example case is about 9300 kg/hr. About
30 kg of cooling (or quench) water are needed for each kg of slag. Since the
slag is essentially non-leachable, the quality of water leaving the quench
tank is determined primarily by the quality of input water. In existing K-T
plants the input water is the same as input to gas cooling (see Section 3.3.1)
since slag and wash waters are treated together in common solids settlers
and direct contact cooling towers. The slag quench flow in such systems
amounts to only a few percent of the total water flow. As mentioned pre-
viously, "clean" water may be used for slag quenching and/or slag may be
rinsed to minimize the slag loading of undesired species (e.g., CN or NhL).
Thus, the slag quench water is not viewed as a separate waste stream but
rather as an integral part of the overall wash circuit. In fact, the slag
quench tank may be a point of makeup to the water circuit. The very small
amounts of pollutant species actually contributed to the water during slag
quenching exit the system with cooling and dust removal blowdown.
Transient Waste Gases (Stream 208)
For short periods of time during startup, shutdown, and transient opera-
tion, gases are generated which are not of use in downstream processing.
Startup of a K-T gasifier generally features use of fuel oil and oxygen under
essentially stoichiometric conditions to accomplish gasifier heatup. The
combustion gases are vented to the atmosphere through startup stacks. Gen-
erally, less than one-half hour is needed to start a gasifier from a cold
state, and emissions during most of this period are representative of com-
bustion rather than gasification conditions. Startup of a gasifier which is
still warm from earlier operation can be accomplished in as little as five
minutes. During upsets or shutdowns, the gasifier system is flooded with
inert nitrogen to purge combustibles. Thus, the entire gas volume of the
gasifier/waste heat boiler system is vented to the atmosphere in just a few
seconds. These gases will contain all of the components of raw K-T gas
(Table 3-6) along with the purge nitrogen.
58
-------
Section 3
Gasification
At present, no data are publicly available relating to characteristics
of transient gases or to the quantities which might be encountered in K-T
facilities. All existing facilities are provided with vent stacks for such
gases, although no further control (e.g., flaring) is effected other than
discharge away from the immediate areas where personnel may be exposed. The
transient waste gases are generally viewed as more of an occupational safety
problem than as a pollution problem since the actual mass emissions repre-
sented are very small on an average basis. Since a relatively large volume
of such gases can be released in a very short period in a small area, K-T
plant designs reflect concern for safety of operating personnel.
59
-------
Section 3
Purification
3.3 GAS PURIFICATION AND UPGRADING
The gas purification and upgrading operation consists of: (1) gas cool-
ing and dust removal to remove some of the volatile contaminants (e.g.,
ammonia, chloride, fluoride, and some fraction of the cyanide) and all dust
from the raw K-T gas; (2) NO reduction for control of traces of NOV present
A X
in the raw gas; (3) raw gas compression and cooling to compress the raw K-T
gas to a pressure suitable for downstream operations; (4) shift conversion
to obtain the hydrogen to carbon oxides ratio required for liquid product
synthesis; (5) acid gas removal for removal of sulfur compounds and carbon
dioxide; and (6) removal of trace sulfur compounds which may be present in
the synthesis gas.
Contaminants removed from the raw K-T gas during purification and up-
grading operations become components of various waste streams. Thus, purifi-
cation and upgrading operations generate several of the most important waste
streams in an integrated facility from the standpoint of waste characteristics
and volumes. The ensuing sections provide details on the nature of specific
processes and their associated waste streams.
3.3.1 Gas Cooling and Dust Removal
Raw K-T gas from the waste heat boiler contains entrained dust which
must be removed prior to compression and downstream processing. Because
primary dust removal is effected by water washing, gas cooling is an integral
part of the dust removal operation. The gas cooling and dust removal opera-
tions are presented schematically in Figure 3-6. Hot raw gas from the waste
heat boiler is cooled and scrubbed of entrained dust in two stages. The
first stage consists of a washer cooler which reduces the gas temperature
and provides bulk dust removal by means of direct water scrubbing. Addi-
tional cooling and dust removal are subsequently achieved in Thiesen disinte-
grators and droplet separators. The gas temperature is reduced from about
570K to 300K during these washing operations. Blowers downstream of the
60
-------
RAW GAS TO
PURIFICATION
SLOWDOWN TO
WASTEWATER
TREATMENT
Figure 3-6. Gas cooling and dust removal
-------
Section 3
Purification
droplet separator compress the gas for transport to the raw gas holders.
The gas holders provide for a continuous, uniform gas feed rate to the down-
stream operations (e.g., compression). Final dedusting is provided by a wet
electrostatic precipitator (ESP) which can reduce the raw gas dust content
o
to less than 0.2 mg/m of raw gas (7).
Dust collected in the washer cooler, droplet separator, and wet ESP is
pumped to settling basins for thickening. Most of the clear water overflow-
ing the settlers is indirectly cooled and recycled (although all existing K-T
gasification facilities utilize cooling towers for wash water cooling, com-
mercial designs for U.S. facilities feature indirect cooling to eliminate the
potential for volatilization of pollutants such as NHL and HCN). The settler
underflow is filtered to produce dewatered dust (Stream 209) and filtrate
which is combined with a portion of the clarifier overflow as blowdown
(Stream 210) for controlling of dissolved component buildup. Makeup water
required to maintain the system water balance is added to the recycle wash
water stream to minimize corrosion in the washer cooler, disintegrator, drop-
let separator, and wet ESP. Material flow estimates for key streams in the
gas cooling and dust removal process are presented in Table 3-8. It should
be noted that dewatering of carbonaceous dusts using fuel oil to displace
dust moisture has been demonstrated in conjunction with coal gasification
and petroleum refining industries. This approach is not considered in this
manual because developers indicate that it is economically unattractive in
these applications due to the associated fuel oil consumption.
Waste streams from gas cooling and dust removal are the dewatered dust
(Stream 209), and cooling and dust removal blowdown (Stream 210). The dust
from gasification is typically disposed of in settling ponds. However, large
land requirements can be associated with such ponding. Further, in the case
of higher rank coals, gasification dust may contain 6% or more of the feed
coal carbon. Therefore, a filtration step has been included in the design
gas cooling and dust removal process to reduce both area requirements and/
62
-------
TABLE 3-8. MATERIAL FLOW ESTIMATES FOR RAW GAS COOLING AND DUST REMOVAL PROCESSES (ILLINOIS NO. 6 COAL)'
oo
Raw Quenched Gas
Stream 6
kmol/hr Vol %
H2
CO
co2
CH4
H2S
COS
cs2
so2
S20|
so3
504
HCN
HSCN
NO
NH3
N2
Ar
cr
F-
Total Dry Gas
H20
Dust
Total: kmol/hr
kg/hr
Temperature (K)
Pressure (MPa)
5829
13699
1910
22
229
26
2
0.5
2.6
0.07
4.0
196
104
22
1.3
22047
7919
29605
29966
672242
450
0.1
26.4
62.1
8.7
998 ppmv
1
1166 ppmv
89 ppmv
21 ppmv
118 ppmv
3 ppmv
180 ppmv
0.9
0.5
998 ppmv
58 ppmv
kg/hr
Raw Gas From
Washer Cooler
Stream 7
kmol/hr
5829
13699
1907
22
229
26
2
0.1
2.5
0.07
0.75
196
104
1
0.06
22019
964
22982
516351
300
0.1
Vol %
26.5
62.2
8.7
999 ppmv
1.0
1168 ppmv
89 ppmv
6 ppmv
113 ppmv
3 ppmv
34 ppmv
0.9
0.5
50 ppmv
3 ppmv
Dewatered Dust
Stream 209
kmol/hr
0.3
0.001
0.03
0.009
0.08
0.008
0.002
0.27
1.8
0.1
1639
1641
310
0.1
kg/hr
12
0.03
3.0
0.74
7.5
0.2
0.1
4.6
62.3
2.0
29520
29605
59217
Slowdown to
Wastewater Treatment
Stream 210
kmol/hr
2.9
0.009
0.3
0.1
0.85
0.08
0.02
2.9
19
1
17843
17870
322448
310
0.1
mg/L
403
1
103
25
254
7
4
156
2111
66
The number of significant figures shown in some cases do not represent the degree of accuracy and are
retained for material balance purposes only. Nevertheless, slight imbalances do appear as a result
of numerical rounding. Material flow estimates and stream compositions are based upon published data
and engineering estimates. Tabulated data are based upon Reference 7.
-------
Section 3
Purification
or improve the feasibility of dust combustion prior to disposal. For flow
estimating purposes, gasification dust from Illinois No. 6 coal has been
assumed to settle to about a 20% solids slurry and be further concentrated
to 50% solids by filtration, although a higher degree of dewatering (65%
solids) has been indicated in a similar design for K-T gasification of a sub-
bituminous coal (4).
Slowdown from the gas cooling and dust removal operation may be required
to maintain a system water balance and/or to prevent excessive buildup of
soluble components in the wash water. Thus, as a minimum, the blowdown
stream would remove that portion of the unreacted gasification steam and
quench water condensed during gas cooling which is not removed with the gasi-
fication dust. The blowdown of any additional water must be accompanied by
the addition of a corresponding amount of makeup water and serves to lower
the level of dissolved solids in the washer circuit. Because the quantities
of gasification steam, quench water, and water soluble ash/dust components
are coal specific and because the quantity of water removed from the washer
system with the dust is both coal and process specific, the quantity of blow-
down required is therefore both coal specific and process specific. The
blowdown quality is determined by the coal characteristics and to a lesser
extent, the makeup water characteristics.
Slag quench water may be incorporated into the gas cooling and dust re-
moval water circuit through the settling basin used for fine slag settling and
cooling purposes. Alternatively, the slag quench system could utilize a
separate settling/cooling circuit which would tend to minimize the level of
contaminants (particularly NH3, HCN, Cl~, and F~) present in the slag mois-
ture. In this case, slag quench water may be suitable for use as makeup
water to the gas cooling and dust removal circuit. In either case, specifics
of the slag quench/gas cooling and dust removal interconnection have little
effect on either the flow or quality of the dewatered dust (Stream 209) and
64
-------
Section 3
Purification
blowdown (Stream 210) streams since slag quench consumes a negligible amount
of process water and is not likely to contribute significantly to levels of
dissolved ammonia, cyanide, chloride, and fluoride.
Dewatered Dust (Stream 209)
Dust entrained in raw K-T gas consists of very small particles (mean
diameter about 0.03 mm) containing ash and ungasified carbon from the feed
coal. When the dust is removed by water washing, a water slurry is generated
which is normally sent to clarification. Clarifier underflow contains around
20% solids and can be dewatered further by simple gravity thickening in hold-
ing ponds or by mechanical means. The final moisture content of dewatered
dust depends largely upon the dust characteristics (particularly its carbon
content) and the method of dewatering employed. Regardless of the exact
moisture content, the dewatered material has a relatively low density when
compared to slag or combustion ash. It also has poor mechanical stability,
and upon rewetting, exhibits tendencies toward plastic flow and/or gravity
settling.
Elemental analyses of the dry dust from Illinois No. 6 coal gasification
indicate that the dry dust contains approximately 36% carbon, 0.1% hydrogen,
and 1.4% sulfur (7). Ash present in the dust has essentially the same com-
position as that of the coal ash, although it is substantially depleted of
volatile elements such as chlorine and fluorine. Ash in the dust corresponds
to approximately 67% of the total coal ash to gasification.
Since the dust particles are very small and experience thorough contact
with wash water for at least a few minutes, it is expected that most of the
readily soluble material in the dust will be leached. Further, the remain-
ing interstitial water in dewatered dust will contain constituents derived
from wash water used in the quench circuit. Table 3-9 presents the avail-
able data on the Teachability of selected elements in dry K-T dust, wet
(dewatered) K-T dust, and, for comparison purposes, the parent coal.
65
-------
Section 3
Purification
The dry dust was collected using a cyclone for test purposes only. The data
show that for Ag, Cr, Hg, Pb, and Se, readily soluble levels are near or be-
low analytical detection limits in all of these solid materials. Detectable
amounts of Cd, As, and Ba are Teachable from the dry dust. Based upon ele-
mental balance considerations, mass loadings of trace elements per unit weight
of dry material in extracts of dry dust would be expected to be greater than
or equal to those in extracts of wet dust. Boron is highly Teachable from
both the dry dust and the parent coal. Chromatographable organics, including
priority organic pollutants have not been detected in wet K-T dust (32,34).
TABLE 3-9. LEACHABILITY OF SELECTED ELEMENTS FROM K-T DUST AND FROM ILLINOIS
NO. 6 FEED COAL*
———————
Dry K-T Dust (31)* Wet K-T Dust (32)T
Element
Ag
As
B
Ba
Cd
Cr
Hg
Pb
Se
RCRA
(pH 5)
<0.2
<7
44
0.4
<0.14
<0.4
<0.004
<3
<16
ASTM
(PH 7)
<0.2
5.6
340
2.4
11
<0.4
<0.004
<3
<13
RCRA
(pH 5)
<0.2
0.04
—
<2
0.6
0.02
<0.004
0.26
<0.02
Illinois No.
RCRA
(PH 5)
<0.2
<8
58
1.0
2
<0.4
<0.004
<3
<8
6 Coal (33)
ASTM
(PH 7)
<0.2
<8
114
16
--
--
<0.004
--
--
Collected-by a cyclone for test purposes only; dry dust collection is not
employed in conventional plants
f34% solids basis
*Units are mg/kg solid
Not indicated in Table 3-9 are components which are associated with the
wash water as a result of scrubbing gaseous constituents from the raw gas
stream; these include Cl", F", CN" (and, indirectly, SCN"), NhJ, and S=.
66
-------
Section 3
Purification
The exact levels of these species are dependent upon the coal type and the
water management plan for the washer cooler. Wash water associated with the
dewatered dust is expected to have the same composition as cooling and dust
removal blowdown (Stream 210), discussed below.
Cooling and Dust Removal Slowdown (Stream 210)
If water quality were not a concern in the wash water circuit, it could,
in principle, be possible to eliminate an aqueous blowdown by internal re-
cycle. In such a case, water would only leave the system with dewatered
slag and dust. However, constituents derived from coal, particularly chlo-
rine, will concentrate in the wash water and limit the ability to internally
recycle all of the wash water due to corrosion and scaling considerations.
Further, in the case of high chlorine, low alkalinity ash coals, such as the
example coal in this document, pH control with caustic or lime would be
necessary in the wash circuit. The sodium or calcium added for pH control
will contribute to the dissolved solids load of the wash water, affecting
the quantity of blowdown needed. Thus, blowdown quantity is dependent upon
the coal characteristics and water quality requirements for the washer cooler
system.
For purposes of analysis, it has been assumed that levels of Cl exceed-
ing around 2000 mg/L would be unacceptable from a corrosion standpoint.
Given the Cl level in the subject coal (2800 ppm), a blowdown of 322 m /hr
was estimated. For coals with different Cl levels, the corresponding blow-
down would differ proportionally. The estimated characteristics of the blow-
down in the case of the subject coal are presented in Table 3-10. Principle
potential pollutants are NH^, CN", and reduced sulfur species. The net pro-
duction of these species is not thought to be greatly different among coals,
so that their concentrations will be similar for most coals when blowdown
rates are similar. Also, heavy elements such as As and Se will be present,
but levels are highly coal specific.
67
-------
TABLE 3-10. CHARACTERISTICS OF SLOWDOWN FROM COOLING AND DUST REMOVAL (7,34)
cr>
03
Major Constituents/Gross Parameters mg/L
NH* (as NH3) 156
CN" 7
SCN" 4
S202 103
S03 25
SO; 254
S= 1
C1" 2111*
Na+ 1000*
Ca++ plus Mg++ --f
TDS 4200**
pH 7§
,
Organic Constituents/Parameters
Total Chromatographable Organics
Oil & Grease
Phenols
Formate
COD
TOC
Trace Elements
F
B
As
Se
Ag
Ba
Cd
Cr
Hg
Pb
mg/L
1.3
<0.001
<0.1
100**
<10**
66#
100'
<1
<1
<0.02
0.04
<0.01
<0.01
<0.001
<0.03
Added in the form of NaOH to water system for pH control, also influenced by makeup water quality
fHighly influenced by makeup water quality as well as ash characteristics
Dependent upon clarifier design
^Without pH control the wash water would be acidic for high chlorine, low alkalinity ash coals
^Estimated from raw coal analysis and assuming that all of the amount in the raw coal will report to the
wash water
**
Calculated from contributing constituents
-------
Section 3
Purification
Since essentially no non-methane organics are produced in K-T gasifica-
tion, no measurable levels of chromatographable organics have been found in
K-T wastewaters. All COD and TOC in these wastes can be attributed to re-
duced sulfur and nitrogen species. Some suspended carbonaceous dust will
remain in the blowdown; the exact amount depends heavily upon the clarifier
and filter designs.
3.3.2 NOX Reduction
Operating data from Modderfontein (29) have indicated fouling of compressor
components and the acid gas removal unit which is attributed to the presence
of nitrogen oxides and 02 in the raw synthesis gas. Fouling in the compressor
resulted in reduced heat transfer in intercoolers, and, occasionally, vibra-
tion due to rotor imbalance. Fouling in the acid gas removal columns resulted
in reduced solvent circulation rates, and fouling of the heat exchangers
resulted in insufficient cooling capability to achieve the required degree
of gas purification. Generation of fouling deposits was attributed to the
following reactions:
Fe + H2S -* FeS + H2
4FeS + 7NO -v Fe^NO)^ + S (Roussin's salt)
2NO + 02 -> 2N02
2N02 + 2H2S -»• 2H20 + 2NO + 2S
Although several operating modifications had been employed to permit
stable process operation, final resolution was achieved by catalytic reduc-
tion of nitrogen oxides in the process gas. Reduction of nitrogen oxides is
achieved by passing compressed gas through a reactor containing a cobalt
molybdate catalyst. Therefore, a nitrogen oxide reduction reactor has been
included in the base plant design. For balance purposes it has been assumed
that NO in the raw gas is catalytically hydroyenated to produce N2 and H20;
69
-------
Section 3
Purification
data regarding this unit's performance and reaction chemistry are not publicly
available. Some degree of gas compression preceding the NO catalytic re-
A
actor would be required to reduce unit size and promote the desired reaction.
For simplicity, raw gas compression is shown to follow NO reduction in Figure
A
3-1 since this is where most of the compression would occur.
Material flow estimates for NOY reduction, raw gas compression, and HCN
A
wash steps are presented in Table 3-11. In addition to essentially complete
NOY destruction, some conversion of HCN to NH-, and CO is also known to occur
A 0
over the catalyst. For analysis purposes, and due to lack of publicly avail-
able data relating to such HCN conversion , the material flow in Table 3-11
indicates no HCN destruction in the subject unit.
Spent NOX Reduction Catalyst (Stream 212)
The only waste stream expected from this reactor is spent catalyst.
Since no experience has been documented for this reaction system, the quantity
of spent catalyst requiring disposal and its physical/chemical characteristics
are currently unknown. However, based upon experience with shift conversion
catalysts, approximately 80 Mg of spent catalyst would be generated every 3
to 5 years.
3.3.3 Raw Gas Compression and Cooling
Koppers-Totzek technology involves coal gasification at essentially
atmospheric pressure. However, the NO reduction, cyanide wash, shift con-
A
version, and acid gas removal (particularly in the case of physical absorp-
tion processes) processes are more efficient and economical at elevated pres-
sures. Therefore, raw gas is compressed to about 3 MPa prior to subsequent
treatment. The temperature rise of the gas during compression is controlled
by intercoolers and an aftercooler, consistent with compressor materials
limitations and temperature requirements of downstream processes. Raw gas
compression and cooling are depicted schematically in Figure 3-7. Waste
streams from raw gas compression and cooling are condensate (Stream 211) and
70
-------
TABLE 3-11. MATERIAL FLOW ESTIMATES FOR K-T NO REDUCTION, COMPRESSION AND COOLING, AND CYANIDE WASH
PROCESSES (ILLINOIS NO. 6 COAL)* x
H2
CO
co2
CH4
V
COS
cs2
so2
S2°3
S03
HCN
HSCN
NO
NH,
N2
Ar
Cl"
F"
Methanol
Total Dry Gas
H20
Total, kmol/hr
kg/hr
Temperature (K)
Pressure (MPa)
Raw Gas From
Washer Cooler
Stream 7
kmol/hr Vol I
5829 26.5
13699 62 2
1907 8.7
22 999 ppmv
299 1.0
Kaw Gas After
NOX Reduction
Stream S
kmol/hr Vol I
5829 26.5
13699 62 2
1907 8.7
22 999 ppmv
229 1.0
Compression
Condensate
Stream 211
kmol/hr mg/L
0.7 1900
0.03 48 7
26 1168 ppmv 26 1168 ppmv
2 89 ppmv
0.1 6 ppmv
2.5 113 ppmv
0.07 3 ppmv
0.75 34 ppmv
196 0.9
104 05
1 50 ppmv
0.06 3 ppmv
22019
964
22982
516351
300
0.1
The number of significant figures
numerical roundina. Material flow
2 89 ppmv
0.1 5 ppmv
2.5 113 ppmv
0.75 34 ppmv
196 0.9
104 0.5
1 50 ppmv
0 U6 3 ppmv
22019
964
22982
516349
shown in soire cases do
0.001 6.3
:0.0002 <1
0.006 8 9
0.004 14
0.75 735
1 2200
0.06 66
964
967
17454
300
0.1
not represent the
mnnsitinn*; nrf ha'
Ra»
Compressed Gas
Stream 9
kmol/hr Vol 1
5829 26.5
13699 62.2
1907 8.7
22 999 ppmv
229 1.0
26 1168 ppmv
2 89 ppmv
0. 1 5 ppmv
25 113 ppmv
0.07 3.2 ppmv
196 0.9
104 0.5
22018
22016
498898
420
3
degree of accuracy and
Compressed Gas HCN Wash Water
After HCN Wash Water Wash Case
Stream 10 Stream 25 5+
kmol/hr Vol % kmol/hr mg/L
5829 36.5
13699 62.3
1873 8 5 11.4 2111
22 1001 ppmv
225 1.0 1.2 176
26 1170 ppmv
2 89 ppmv
0.1 5 ppmv
0.34 16 ppmv 2.1 241
196 0.9
104 0.5
21979
9 13227
21988 13242
497444 238884
280 280
3 0 1
are retained for material balance
Cyanide Wash Flash Gas
Water Wash Case
Stream 214af
kmol/hr Vol %
19 7 89 3
2.3 10 7
0.011 499 ppmv
22
0.3
22
951
280
0.1
purposes only Nevertheless,
Cyanide Wash Flash
Methanol Wash Ca
Stream 21 4b*
kmol/hr Vol
0.49 3.0
4.4 26.
1 2 7.3
7.3 44
0.56 3.4
2.1 12.
0.3 1.8
16
16
524
310
0.2
slight Imbalances
Aqueous Cyanide
Gas Wash Still Bottoms
se Methanol Wash Case
Stream 213*
J kmol/hr mo/L
4
7
9 <0. 00001 10
0.00004 1000
0.07
0.07
1.3
310
0.1
do appear as a result of
Streams 215 and 214a exist only
^Streams 213 and 214b exist only
the water-based cyanide wash case.
the methanol-based cyanide wash case.
-------
RAW GAS
FROM
NOX
REDUCTION
RAW GAS
COMPRESSION
COOLING
DEPRESSURIZATION
OFF-GASES
RAW
COMPRESSED
GAS
CONDENSATE
DEPRESSURIZATION
CONDENSATE
Figure 3-7. Raw gas compression and cooling
-------
Section 3
Purification
condensate depressurization offgases. Characterization data are not avail-
able for condensate depressurization offgases; however, these offgases are
recycled back to the primary compression feed gas and are not a discharge
stream.
Raw Gas Compression and Cooling Condensate (Stream 211)
Characterization data for raw gas compression and cooling condensate
available from the K-T coal gasification facility operated by AECI Limited
at Modderfontein, Republic of South Africa, are summarized in Table 3-12.
These data relate to gasification of South African sub-bituminous coal and
are likely to be somewhat coal specific. Material balances of the Modder-
fontein data indicate that virtually all of the compression feed gas ammonia
was recovered in the compression condensate, while only 1 to 2% of the cyanide
was recovered as either cyanide or thiocyanate. Because raw gas ammonia and
cyanide levels are anticipated to be similar for bituminous and sub-bituminous
coals, the quality of condensate with respect to these species is also
expected to be similar.
The higher raw gas sulfide levels presented in the base case design as
compared with the Modderfontein data may result in higher sulfide levels in
the pressurized condensate, although depressurized condensate sulfide levels
are expected to be similar. It should be noted that higher sulfide levels
in the pressurized condensate could influence the levels of nonvolatile sulfur
species in the condensate (e.g., SCN", S^, SO^, and 50=), provided that
formation reactions for the nonvolatile sulfur species are rapid, if these re-
actions occur slowly, that is, occur after condensate decompression, the tab-
ulated levels of nonvolatile sulfur species would be appropriate for the
Illinois coal case. In any event, the condensate characterization data from
Modderfontein provide a good qualitative basis for estimating condensate
characteristics for systems utilizing other coals.
73
-------
Section 3
Purification
TABLE 3-12. CHARACTERISTICS OF PRIMARY COMPRESSION AND COOLING CONDENSATE
FROM SOUTH AFRICAN SUB-BITUMINOUS COAL (24)
PH 8.0 - 8.2
Total suspended solids, mg/L 0-12
Total dissolved solids, mg/L 170 - 260
Hardness, mg/L 46 - 60
Alkalinity, p-value (as CaC03 mg/L) o
m-value (as CaC03 mg/L) 2690 - 2990
COD, mg/L 559 _ 644
NH3> m9/L 900 - 973
CN-, mg/L 7.2 - 10.5
SCN", mg/L 10.g _ 17j
H2S, mg/L 43i9 _ 53-5
S2°3' m9/L 4.8 - 7.8
S03, mg/L <1
S0~, mg/L 49 _ 56
PO'3, mg/L 2 _ 3
3.3.4 Cyanide Wash
Because the base plant design includes the Rectisol process for acid
gas removal, minor gas constituents such as ammonia and hydrogen cyanide can
complicate operation of the acid gas removal unit. Ammonia and hydrogen
cyanide, which are very soluble in methanol, make methanol regeneration more
complex and result in additional steam requirements (,35). These contaminants
may be removed from the raw compressed gas by employing a prewash of either
74
-------
Section 3
Purification
water or cold methanol. A water wash unit is employed at the Modderfontein
facility (24), and wash water to gas ratios from this unit provided the
basis for the water wash system presented herein. Methanol-based cyanide
wash units have been employed in several applications (partial oxidation of
oil) and may be attractive for coal conversion applications where water con-
sumption is a major concern or to minimize cyanic wastewater generation.
In the case of a water-based cyanide wash, cyanide is absorbed at 280K
and 3 MPa. The rich wash water is subsequently flashed to atmospheric pres-
sure yielding depressurized wash water (Stream 215) and sour flash gas
(Stream 214a) waste streams. The depressurized cyanide wash water can be
partially recycled as gasifier quench water, combined with wastewaters from
the gas cooling and dust removal and raw gas compression and cooling pro-
cesses and sent to wastewater treatment, or treated separately. The sour
flash gas is processed through the sulfur removal/recovery system since its
sulfide content corresponds to approximately 1% of the total sulfur in the
gasified coal.
In the case of a methanol-based cyanide wash, cyanide absorption was
assumed to proceed at 270K and 3 MPa. The rich methanol is regenerated by
depressurization and indirect heating to yield a sour flash gas (Stream 214b).
The sour flash gas is processed through the sulfur removal/recovery system
since its sulfide content corresponds to approximately 3% of the total sulfur
in the gasified coal. Because of the rigorous hot regeneration required to
liberate the highly soluble cyanide from wash methanol, solvent losses with
the flash gas are anticipated to be relatively high, although no operating
data are available.
An additional waste stream associated with the methanol-based cyanide
wash is an aqueous cyanide wash still bottoms (Stream 213). Feed gas to the
cyanide wash unit contains moisture which would accumulate in the wash meth-
anol. This moisture is recovered by distillation of the regenerated methanol.
75
-------
Section 3
Purification
The composition of the compressed gas after cyanide wash (Stream 10) is
essentially independent of the cyanide wash absorbent used since, in either
case, only a small quantity of gas is absorbed in the cyanide washer. There-
fore, material flows in operations downstream of the cyanide washer (i.e.,
shift conversion, acid gas removal, and liquid product synthesis) are essen-
tially independent of the cyanide wash absorbent used. For these reasons,
material flow estimates for downstream operations will be provided only for
the water-based cyanide wash case.
Estimated characteristics and material flows of waste streams from the
cyanide wash process have been presented in Table 3-11.
Cyanide Hash Water (Stream 215)
The cyanide wash water stream exists only in the case of a water-based
cyanide wash system. No characterization data are available for this waste
stream. Therefore, the wash water quality has been estimated using gas solu-
bility data (25,3fc>). Publicly available data indicate that low pressure
aqueous wash systems approach Henry's Law equilibria for HCN and NH-, when
the influence of pH is considered (7,25). Therefore, theoretical HCN and
NH^ solubility data were utilized in conjunction with absorption/desorption
estimates. However, H-S absorption does not appear to be equilibrium con-
trolled and, therefore, empirical absorption data were utilized (7,25).
Some C02 is also absorbed in aqueous wash systems and theoretical solubility
data were used for purposes of analysis (this approach may overestimate the
quantity of C02 in the flash gas). These estimates indicate a depressurized
cyanide wash water containing 2111 mg/L C02, 241 mg/L HCN, and 176 mg/L H2$.
The flow rate of depressurized cyanide wash water has been factored from
2
nominal Modderfontein operating data (24) and is approximately 238 m /hr.
Sour Gas from Cyanide Uash Flash - Water Uash Case (Stream 214a)
No flow rate or characterization data are available for sour flash gas
from cyanide wash water depressurization. The sour gas flow rate and
76
-------
Section 3
Purification
composition have been estimated using gas solubility data. Sour flash gas
will be generated at a rate of approximately 22 kmol/hr and is composed of
89% C02, 11% H2S, and 499 ppmv HCN. The sulfur content of this sour gas, 2.35
kmol/hr, corresponds to nearly 1% of the total sulfur in the gasified coal.
Sour Gas from Cyanide Wash Flash - Methanol Wash Case (Stream 214b)
No flow rate or characterization data are available for sour flash gas
from cyanide wash methanol regeneration. The sour gas flow rate and composi-
tion have been estimated using gas solubility data (37). The methanol con-
tent of the sour gas has been derived from Rectisol operating data on the
basis of inorganic gases stripped during hot regeneration (38). Sour flash
gas will be generated at a rate of approximately 16 kmol/hr and is composed
of 45% H2S, 27% CO, 13% HCN, 7% C02, 3% COS, 3% H2> and 2% methanol. The
sulfur content of this sour gas, 7.3 kmol/hr, corresponds to nearly 3% of
the total sulfur in the gasified coal.
Losses of \\2 and CO in this stream are expected to be small owing to
their low solubility in methanol and the low methanol circulation rate re-
quired for cyanide removal. Carbon dioxide losses in this stream are small
due to the low C02 partial pressure prior to shift conversion. Thus, the
losses of H2, CO, and C02 in the sour gas correspond to only 0.008%, 0.03%
and 0.06% of their respective feed rates into the cyanide wash unit.
Cyanide Wash Still Bottoms - Methanol Wash Case (Stream 213)
Raw compressed gas contains a small amount of moisture which will be
removed during cyanide washing by absorption in methanol. This moisture is
recovered by distillation of regenerated methanol and will have a flow rate
3
of approximately 0.001 m /hr. No characterization data are available for
the aqueous cyanide wash still bottoms. However, HCN and methanol levels of
10 mg/L and 1000 mg/L, respectively, have been estimated as upper limit
concentrations based upon data for Rectisol still bottoms at Sasol (39).
77
-------
Section 3
Purification
3.3.5 Shift Conversion
Methanol synthesis and hydrocarbon production via Fischer-Tropsch (F-T)
synthesis can be represented by the following reactions:
CO + 2H2 CATALYST> CH3OH + heat (Methanol Synthesis)
nCO + (2n + ,5x)H2 CATALYST ) CnH2n+x + nH20 + heat (F-T Synthesis)
where n ranges from 1 to 20 and x = 2 for parafins and x = 0 for olefins.
Since feed gas to synthesis would usually contain small amounts of C02 in
addition to CO and H?, and synthesis catalysts are also active for the inter-
connecting water gas shift reaction (CO + HLO -> C0? + H?), the hydrogenation
of C0? may be represented as follows:
C02 + 3H2 CATALYST^ CH3OH + H20 + heat (Methanol Synthesis)
nC02 + (3n + .5x)H2 CATALYST) = CnH2n+x + 2nH20 (F-T Synthesis)
Although the theoretical stoichiometry for synthetic liquids production calls
for a ratio of 2 moles of H2 per mole of CO and 3 moles of H2 per mole of C02,
in practice a stoichiometric excess of about 3% is required (1). That is:
H2/(2CO + 3C02) = 1.03
Assuming a residual C02 concentration of about 3% in the synthesis gas,
an Hp to CO mole ratio of about 2.4:1 is required in the synthesis gas. The
H2 to CO ratio in compressed, raw K-T gas typically ranges from 1:2 to 1:2.5
(7,8,24,29), well below the ratio required for liquids synthesis. Thus, a
shift conversion step is a necessary part of the gas upgrading operations
for K-T based indirect liquefaction plants.
78
-------
Section 3
Purification
The shift conversion reaction, reaction of CO and water vapor to form
\\2 and CC^, is a mildly exothermic reaction which can be promoted by a variety
of catalysts:
CO + H20 = C02 + H2 + 41.2 kJ/mol
Shift conversion catalysts currently in use include iron-chromium, copper-
zinc, and cobalt-molybdenum catalysts. Iron-chromium catalysts are active
at 600K to 823K, and retain their activity and physical strength with sulfur
species concentrations of up to about 200 ppmv in the shift gas (40,41,42).
Copper-zinc catalysts are active at low temperatures (475K to 520K) and are
typically used for final CO conversion following high temperature shift con-
version. Copper-zinc shift catalysts lose activity due to poisoning by sul-
fur and chlorine compounds in the process gas at concentrations in the range
of 1 ppmv (40,41). The newest types of shift catalysts (cobalt-molybdenum
catalysts) maintain high activity over a wide range of temperatures (473K
to 810K) and are not affected by the presence of gaseous sulfur compounds.
Indeed, since cobalt-molybdenum catalysts are active in the sulfided form,
some hydrogen sulfide is required in the process gas to maintain catalyst
activity (43,44). Thus, iron-chromium or copper-zinc catalysts are appli-
cable if shift conversion is preceded by a sulfur removal process, while a
cobalt-molybdenum catalyst would be required if shift conversion precedes
acid gas removal.
The sequence in which shift conversion and acid gas removal are per-
formed is dependent upon a variety of design considerations including raw
gas temperatures, acid gas removal selectivity requirements, and catalyst
costs. K-T based coal conversion facilities have, in the past, employed a
shift conversion process which follows raw gas sulfur removal and precedes
carbon dioxide removal (e.g., the AECI Modderfontein facility). Such a con-
figuration facilitates achievement of highly selective sulfur recovery due
to the favorable H2S to C02 ratio prior to shift conversion. In addition,
conventional iron-chromium or copper-zinc shift conversion catalysts can be
79
-------
Section 3
Purification
utilized. All commercial K-T coal gasification facilities utilize this con-
figuration and the only K-T-based indirect liquefaction facility proposed in
the U.S. (The North Alabama Coal Gasification Consortium Project) is expected
to utilize this configuration also. This is, therefore, the approach which
has been incorporated into the base plant design. Because shift conversion
feed gas is the outlet gas from HpS removal and the shift conversion outlet
gas is the inlet to C02 removal, material flow estimates for shift conversion
are presented in the acid gas removal section (Section 3.3.6).
It should be noted that an additional gas compression step will likely
precede shift conversion (24). In contrast with raw gas compression, the
desulfurized gas is dry and hence no compression condensate results. Since
there would be no waste streams associated with this additional compression,
this process is not further considered in this manual.
To achieve the required H~ to CO ratio of about 2:1 using catalytic
shifting, two approaches are possible: (1) processing the entire raw gas
stream through a relatively low conversion efficiency reactor, and (2) pro-
cessing a portion of the raw gas stream through a higher conversion effici-
ency reactor and combining the shifted and unshifted (bypass) gases after-
ward. Based on actual operating experience, a residual CO concentration of
less than 3% can be obtained at 500K to 700K. Taking advantage of the high
conversion efficiency achievable, the split flow approach is preferred since
it provides costs savings associated with a smaller conversion reactor size
and a smaller C02 absorber. A tradeoff exists between bypass fraction and
degree of shift conversion. The exact bypass for a given facility is an
important consideration from a cost and process efficiency standpoint. For
analysis purposes, a bypass fraction of 20% has been assumed, a value which
is consistent with a single stage of shift conversion. In the extreme, two
stage shift conversion providing a 3% (or less) CO residual in the shifted
gas would enable a bypass fraction of about one-third. The magnitude of the
bypass fraction has essentially no impact upon waste stream generation in
80
-------
Section 3
Purification
the shift section. The shift conversion operation is presented schematically
in Figure 3-8.
Two waste streams associated with the shift conversion unit are spent
shift catalyst (Stream 217) and shift condensate blowdown (Stream 218).
Spent catalyst, in this case iron-chromium and/or copper-zinc, can either
be disposed of or reclaimed, depending upon the current metal cost and re-
clamation trends. Shift condensate blowdown is suitable for reuse within
the plant since it is low in dissolved solids although it may contain traces
of ammonia and methanol from acid gas removal.
Spent Shift Conversion Catalyst (Stream 217)
Typically, iron-chromium shift catalyst contains about 44% iron and 6%
chromium, and copper-zinc catalysts contain about 24% copper and 36% zinc
(40,41,42,45). There is very little interest in the steel industry and
specialized metals industry for recovery of iron or chromium from spent high-
temperature shift catalysts, and this material will probably be disposed of
directly. Spent chromium promoted copper-zinc catalysts are usually re-
claimed for their copper content, although reclaimers have difficulty separat-
ing out the zinc and the chromium (46,47). Spent alumina promoted copper-zinc
catalysts are usually reclaimed for both their copper and zinc contents (48).
There are essentially no publicly available leachate data for these materials.
However, few coal specific contaminants are expected to be present in spent
shift catalysts since shift conversion follows sulfur removal from the raw
gas. Spent shift catalyst generation rates are estimated to be approximately
70 Mg every 3 to 5 years.
Shift Condensate Blowdown (Stream 218)
Because shift conversion is preceded by cyanide wash and sulfur removal
operations, shift condensate is expected to be free of contaminants associated
with coal gasification. However, traces of methanol volatilized from the
sulfur removal operation and ammonia used for pH control may be present in
81
-------
co
RAW GAS
FROM
HCN WASH
SHIFT/C02 REMOVAL BYPASS GAS
i
, H2S REMOVAL 1
4 (REFER TO '_
[SECTION 3.3.6) '
I 1
k -\ii-i —
-6j) — *
STEAM »
1
SHIFT
COMVERTER(S)
j C02 RE
vjv 1 (REfER
1 SECTIOf
L.
r
MOVAL '
TO |.
< 3.3.6) |
SYNTHESIS
GAS
SPENT SHIFT
SHIFT CONDENSATE
CATALYST
Figure 3-8. Shift conversion
-------
Section 3
Purification
the condensate. A shift condensate blowdown is generated to control buildup
on dissolved solids in the condensate recycle system. Shift condensate
3
blowdown will be generated at a rate of approximately 3 m /hr. Ordinarily
this very small volume waste stream would be used as makeup to the raw gas
wash circuit. Hence, this stream is not considered further as a separate
waste.
3.3.6 Acid Gas Removal
Removal of hydrogen sulfide and other sulfur compounds present in the
raw K-T gas is necessary to prevent catalyst poisoning in subsequent shift
conversion (if sulfur sensitive catalysts are to be used) and methanol or
Fischer-Tropsch synthesis operations. Bulk removal of carbon dioxide is
necessary to obtain a composition meeting the stoichiometric requirements
for synthesis feed gas (see Section 3.4). There are several acid gas removal
processes which have been demonstrated in coal gasification applications or
similar applications. However, only the two-stage selective Rectisol process
has been included in the base plant design since most of the recently com-
missioned K-T coal gasification facilities utilize two-stage selective
Rectisol units and the only K-T based indirect liquefaction facility proposed
in the U.S. is also expected to use this process. A detailed description of
the Rectisol process and a summary of published performance data are pre-
sented in Appendix B and Control Technology Appendices.
Rectisol is an acid gas removal process which removes CCL, H?S, COS,
HCN, NH3, organic sulfur compounds, benzene, and gum-forming hydrocarbons
from synthesis gases by means of physical absorption in cold methanol. The
principle of operation is based upon the fact that these compounds, parti-
cularly the reduced sulfur species and C02, are very soluble at high pres-
sure in cold methanol and are readily recoverable by flash desorption.
Solubility coefficients (the ratios of solubility to partial pressure) of
H2S and C02 are higher than those of major product gases such as H2 and CO
and increase substantially with decreasing temperature while those of major
83
-------
Section 3
Purification
product gases are relatively temperature independent. For this reason,
Rectisol absorption columns operate at low temperatures, typically in the
range of 210 to 250K (37,49,50). Low temperature operation also reduces
solvent losses by reducing the partial pressure of methanol in the outlet
streams.
Because the solubilities of reduced sulfur species (e.g., H?S and COS)
in methanol are substantially greater than that of C02 at the same partial
pressure, the Rectisol process is capable of selective recovery of reduced
sulfur species versus C02; to some degree, this holds for all physical absorp-
tion solvents capable of absorbing reduced sulfur species and C02 almost
independently.
The two-stage selective Rectisol process is presented schematically in
Figure 3-9. In general, a small quantity of methanol is added to the raw gas
exiting the HCN washing step prior to cooling and H2S absorption, to prevent
icing. This step would not be required when a methanol-based cyanide wash
process is employed. Moisture in the feed gas is removed from the H2S
absorber in solution with methanol which is recovered by distillation.
Hydrogen sulfide and COS are absorbed from the feed gas using lean methanol
from the C0? regeneration column. Rich methanol from the 1-LS absorber is
flashed to liberate absorbed CO and H2 which is returned to the raw gas.
Additional flashing and stripping in the concentration column, with reabsorp-
tion of reduced sulfur species in lean methanol, produces an H^S-rich
methanol stream for hot regeneration and a CO^ offgas. Hydrogen sulfide is
recovered by stripping with methanol vapor in the hot regeneration column.
Carbon dioxide is removed from shifted process gas by absorption in
regenerated methanol. A small quantity of methanol is added to the shift
gas prior to cooling and COo absorption to prevent icing, and moisture in
the shift gas is removed from the C02 absorber in solution with methanol.
Shift conversion bypass gas may be injected into the upper portion of the C02
absorber. This serves as a convenient mixing point for shifted and bypass
84
-------
oo
en
FLASH GAS MAY ALSO 8E ROOTED TO THE
GAS HOLDERS OR RAW GAS COMPRESSION
SYSTEM RATHER THAN BEING COMPRESSED
IN A DEDICATED COMPRESSOR FOR RECYCLE
FLASH GAS FROM THE C02 tOAOED METH ANOL
FLASH MAY ALSO BE USED W
-------
Section 3
Purification
gas streams, and provides for trace sulfur removal from the bypass gas.
Rich methanol from the C02 absorber is flashed to recover absorbed HL.
Carbon dioxide is recovered by flashing and stripping with No in the C02
regeneration column. Material flow estimates for the shift conversion and
acid gas removal operations are presented in Table 3-13.
It should be noted that desulfurization prior to shift conversion en-
ables the use of conventional shift catalysts (e.g., iron-chromium and copper-
zinc). Also, due to the favorable H2S to C02 ratio before shift conversion,
it facilitates production of an hLS-rich offgas for economic sulfur recovery.
Shift conversion prior to acid gas removal would result in an increased con-
centration of COo in the H2S absorber feed gas; the associated decrease in
the H2S to C02 ratio is much less favorable for highly selective operation.
There are two gaseous waste streams and one liquid waste stream produced
by the selective Rectisol unit. The gaseous waste streams are the H2S-rich
offgas (Stream 216) which contains most of the sulfur present in the feed
coal to gasification and the C02-rich offgas (Stream 219). The H2S-rich off-
gas is combined with flash gas from the HCN wash and sent to sulfur removal/
recovery because of its high sulfur content, while the C02-rich offgas is
typically vented directly to the atmosphere. The liquid waste stream is the
Rectisol condensate/still bottoms (Stream 220) from the methanol/water dis-
tillation column.
H2S-Rich Offgas (Stream 216)
Approximately 94% of the total sulfur present in the feed coal to gasi-
fication is present in the H2S-rich offgas. The offgas is estimated to con-
tain about 50% C02, 42% H2S, 5% COS, 0.4% CS2, 340 ppmv HCN, and 200 ppmv
S02. The only organic compound present in significant quantities is ex-
pected to be process methanol. Data indicating the total methanol loss for
selective Rectisol systems are available (35,38), although data relating to
methanol losses in individual streams have not been published. Therefore,
86
-------
TABLE 3-13. MATERIAL FLOW ESTIMATES FOR K-T SHIFT CONVERSION AND ACID GAS REMOVAL PROCESSES (ILLINOIS
NO. 6 COAL)*"1"
00
Compressed Gas
from HCN Hash
Stream 10
H2
CO
co2
CH4
H2S
COS
cs2
Sf2
HCN
N2
»r
Methanol
Total Dry Gas
H20
Total . kmol/hr
kg/hr
Temperature (K)
Pressure (MPa)
kmol/hr
5829
13699
1873
22
225
26
2
0 1
0.34
196
104
21979
9
21988
497444
280
3
Vol %
26.5
62.3
8.5
0.1
1.0
1170 ppmv
89 ppmv
5 ppmv
16 ppmv
0.9
0 5
HjS Rich Offgas
Stream 216
kmol/hr Vo! %
269 50.2
225 42.1
26 48
2 0.37
01 216 ppmv
0.34 636 ppmv
6 1.2
7 1 3
535
535
21671
310
0.3
Shift Conversion Shift Conversion Shift Gas to Shift
Bypass Gas Feed Gas C02 Absorber Condensate
Stream 12 Stream 11 Stream 13 Stream 218
kmol/hr Vol % kmol/hr
1039
2443
321
3 9
0.004
35
18
Trace
3860
3860
86421
305
3
26 9 4790
63 3
8.3 1480
0 1 18
1 ppmv 0 02
0.9 161
0.5 85
Trace
17791
17791
398279
305
3
Vol I kmol/hr Vol % kmol/hr
26.9 12670 49.4
63.3 3377 13.2
8 3 9359 36.5
01 18 700 ppmv
1 ppmv 0 02 0.8 ppmv
09 161 06 Trace
0.5 85 0 3
Trace Trace
25671
55 167
25726 167
541238 3000
310 370
3 0.1
Nitrogen C02-Rich Rectisol tondensate/
Strip Gas Vent Gas Still Bottoms
Stream 15 Stream 215 Stream 220
kmol/hr kmol/hr Vol % kmol/hr mg/L
45 05
101 1.0
8875 tl 2
12 122 ppmv
0 05 5 ppmv
0 0^ 8 ppmv
0 0004 10*
1054 1041 10 3
0.04 1000
1054 10064
64
1054 10064 64
29534 422684 1160
310 305 340
0 4 0.1 0.1
Combined
Synthesis Gas
Stream 14
kmol/hr Vol I
13664 67.2
5719 28.1
610 3.0
21 0.1
0 02 1 ppmv
203 1 0
104 0.5
20320
20320
224718
305
3
The number of significant figures shown in some cases do not represent the degree of accuracy and are retained for material balance purposes only. Nevertheless, slight imbalances
do appear as a result of numerical rounding Material flow estimates are based upon published data and engineering estimates. Tabulated data are based upon References 1, 24, 35, 37, 38 and 39.
A two-stage Rectisol acid gas removal system is used with shift conversion following sulfur removal and preceding carbon dioxide removal
*This value represents the total cyanide plus thiocyanate.
-------
Section 3
Purification
for evaluation purposes, the entire methanol loss has been assumed to be
associated with hot regeneration, resulting in an estimated methanol concen-
of about 1% in the H2S-rich offgas. The H2S-rich offgas will be generated
at a rate of about 535 kmol/hr.
A somewhat higher level of selectivity has been incorporated into the
Rectisol material flow estimates than is typically reported for existing
units processing low-sulfur feed gases. Hydrogen sulfide concentrations
in the H?S-rich offgas from selective Rectisol units are generally in the 25
to 35% range for these applciations (35,37,38,51). However, these data are
from facilities with Rectisol feed gas H2$ to C02 mole ratios ranging from
about 1:18 to 1:66. Because of the relatively high sulfur content of the
design Illinois No. 6 coal (3.13% sulfur, dry basis), the design feed gas
H2S to C02 ratio is about 1:8. Thus, other variables being equal, a higher
selectivity would be expected. In any case, the available data indicate that
H?S concentrations of at least 25 to 35% can be obtained for a wide range of
feed coal sulfur contents.
CQ2-Rich Offgas (Stream 219)
C02-rich offgas is generated in the subject facility at about 10,000
kmol/hr and consists of 88% C02 with the remainder being largely N2. Small
amounts of H2, CO, methanol vapor, and sulfur compounds will also be present
in the offgas. The CO level in the C02-rich offgas is related to the CO
level in the shifted gas. The design in this manual is consistent with
single stage shift conversion. This results in a higher level of CO in the
shifted gas to the C02 absorber as compared to multi-stage shift conversion.
Thus, the estimated concentration of CO in the C02~rich offgas (i.e., 1% CO)
is higher than published concentrations for multi-stage shift conversion
systems. Published data relating to multi-stage shift conversion systems
indicate CO concentrations of 0.1% to 0.4% in the C02-rich offgas (35,37,38).
It should be recognized that CO (and H2) could be largely recovered by
88
-------
Section 3
Purification
flashing of the CCL-loaded methanol from the C02 absorber. This flash gas
could be recycled to the raw gas or used as fuel or reducing gas elsewhere
in the plant. Hence, the CO content of Stream 219 as shown in Table 3-13
probably represents an extreme case.
Methanol vapor will also be present in the COp-rich offgases, although
no data are publicly available at present to indicate the exact level of this
compound. For analysis purposes all methanol losses from the Rectisol pro-
cess are assumed to be associated with the H^S-rich fraction. Some portion
of this methanol will in actuality be contained in the CO^-rich offgas and
synthesis feed gas.
A few ppm each of H~S and COS will be present in the COo offgases.
Data from existing facilities indicate about 5 ppmv HoS and 8 ppmv COS.
Designs for several recent selective Rectisol plants specify less than 10
ppmv total sulfur in the CO^-rich offgas (52).
Rectisol Condensate/Still Bottoms (Stream 220)
Little characterization data are available for Rectisol condensate/
still bottoms from the methanol/water distillation. Based upon data from
SASOL (39), concentrations of methanol and total cyanide (cyanide and thio-
cyanate) in this stream are estimated to be less than 1000 mg/L and about
10 mg/L, respectively. The generation rate for this stream is estimated to
be 1 m3/hr.
3.3.7 Trace Sulfur Removal
Although the Rectisol process can routinely achieve sulfur levels well
below the ppmv required to protect synthesis catalysts, a guard bed material,
employed as insurance to prevent catalyst poisoning during upset or trans-
ient conditions. Zinc oxide, the most commonly proposed guard bed material,
can reduce hUS levels to below 0.01 ppmv. However, COS levels cannot be re-
duced below the 0.01 ppmv level. Zinc oxide beds are closed systems during
89
-------
Section 3
Purification
routine operation, and spent material is infrequently generated. Assuming
a sulfur absorption capacity of 66 g/kg ZnO, an average of about 80 Mg/yr
of spent guard material (Stream 218) would require disposal (53). If the
sulfur guard bed is sized for a 3-year operating life, about 230 Mg of spent
guard material would require disposal every 3 years.
90
-------
Section 3
Product Synthesis
3.4 PRODUCT SYNTHESIS
Methanol synthesis and hydrocarbon production via Fischer-Tropsch
(F-T) synthesis can be represented by the following reactions:
CO + 2H9 CATALYST > CH.OH + heat (Methanol Synthesis)
C. O
nCO + (2n + .5x)H2 CATALYST > CnH2n+x +nH20 + heat (F-T Synthesis)
where n ranges from 1 to 20 and x = 2 for parafins and x = 0 for olefins.
The mix of F-T products obtained (i.e., the range of n and x values) is
dependent upon several factors including the reactor design, temperature,
pressure, and type of catalyst used. Since feed gas to synthesis would
usually contain some C02 in addition to CO and H2 and synthesis catalysts
are also active for the interconnecting water gas shift reaction (CO + H20 ->
C02 + Hp), the hydrogenation of C02 may be represented as follows:
TATAI YST
C02 + 3H2 > CH3OH + H20 + heat (Methanol Synthesis)
nC02 + (3n + .5x)H2 CATALYST > = CnH2p+x + 2nH20 (F-T Synthesis)
Although the theoretical stoichiometry for synthetic liquids production
calls for a ratio of 2 moles of H per mole of CO and 3 moles of H per
mole of C02, in practice the following ratio is required:
H2/(2CO +3C02) = 1.03 (1)
The major difference in methanol and F-T synthesis is in the catalysts
used and temperatures and pressures employed. Methanol synthesis is accom-
plished over Cu/Zn-based catalysts at 473K and 3.5 to 7.0 MPa while F-T
91
-------
Section 3
Product Synthesis
synthesis proceeds over iron oxide-based catalysts at 603K (fluidized bed)
and 438K (fixed bed) (3).
The Mobil M-gasoline synthesis from methanol can be represented as
follows:
CATALYST
nCH3OH - >
The process employs a zeolite-based catalyst and operates at about 573K and
2.2 MPa (1).
For methanol, F-T, and Mobil M-gasoline synthesis processes, high con-
versions of synthesis gas are achieved only when gas recycle is employed due
to performance limitations of the catalysts. Complete recycle, however, is
not possible due to the buildup of nonreactive components in the system (e.g.,
N2, Ar, CH4). Thus, all synthesis processes produce a purge gas containing
inerts and lost CO and H2. Variations in process designs for synthesis
reactors reflect different approaches to heat recovery, maximum syngas con-
version, minimum recycle, and minimum purge. The discussions below provide
some detail about the subject synthesis processes.
3.4.1 Methanol Synthesis
Methanol production is a fully commercialized technology with a number
of firms offering conversion processes including Lurgi , ICI, Chem Systems,
Vulcan-Cincinnati, Mitsubishi, Nissui-Topsoe, and Selas-Pol imex (3). Figure
3-10 is a simplified flow diagram of the ICI process, one of the leading
commercial processes. In the ICI process, compressed synthesis feed is
mixed with recycle gas and heated by exchange with methanol product before
entering the synthesis reactor. The bulk of the reactor feed enters the top
of the reactor while a portion of the gas, which has bypassed the remaining
heat exchangers, is injected at various levels in the reactor. The cooler
"quench" gases serve as the main temperature control in the system. Crude
92
-------
RECYCLE GAS
SPENT
CATALYST
COMPRESSOR
-METHANOL'
SYNTHESIS
-REACTOR ^
PROCESS
WATER'
HOT
'WATER
DISTILLATION
OFFGAS
WASTEWATER
Figure 3-10. Flow diagram for the ICI methanol synthesis process
-------
Section 3
Product Synthesis
methanol vapors which exit the bottom of the reactor are cooled by feed/pro-
duct heat exchange and expansion in a turbo-expander before the methanol pro-
duct is condensed. Condenser overhead is partially recycled, with a purge
stream withdrawn from the system through an expansion turbine. Depressuriza-
tion gases from crude methanol pressure letdown are combined with these purge
gases for use as plant fuel.
In the Lurgi process, a leading commercial alternative to the ICI pro-
cess, the fixed bed reactor would be replaced by a boiling water jacketed
tube reactor with catalyst in the tubes (1). In the Lurgi case, isothermal
reactor operation is realized, and no gas quench is necessary. In all metha-
nol synthesis processes, large amounts of heat are recovered as medium pres-
sure steam.
Regardless of the specific process employed, all methanol synthesis
processes generate a continuous purge gas and an intermittent spent catalyst.
The purge gas (plus expansion and distillation gases) are useful as sulfur-
free fuel gases. When crude methanol is purified by distillation, a waste-
water stream is generated. The quantity of vastewater generated is alnoct
directly proportional to the quantity of CCu present in the synthesis gas,
and its quality (i.e., its content of methanol and higher alcohols) is depen-
dent upon the specific process employed.
Table 3-14 presents example material flow estimates for methanol syn-
thesis. Because a commercial facility may produce either crude methanol or
fuel grade methanol, or both, estimated compositions of both crude and fuel
grade methanol have been presented. As indicated by the data, a small amount
of purge gas is produced due to the inerts in the synthesis feed gas. Con-
version of carbon oxides to fuel methanol is about 97%. Crude methanol con-
tains a few percent water and small quantities of higher alcohols, dimethyl
ether, and low molecular weight hydrocarbons and is purified by distillation.
94
-------
TABLE 3-14. METHANOL SYNTHESIS MATERIAL FLOW ESTIMATES FOR K-T GASIFICATION (ILLINOIS NO. 6 COAL)11
on
Feed Gas
Stream 14'
Constituent kmol/hr Vol %
CO 5719 28.1
H2 13664 67.2
C02 610 3.0
CH4 21 0.1
N2 + Ar 307 1.5
CH3OH
(CH3)20
C2H5OH
C3H7OH
H20
Total 20320 100
Crude Kethanol
Stream 20
kmol/hr Wt %
0.02
0.02
1.5
0.01
0.02
6123
5.1
1.9
1.2
549
6682
<0.01
<0.01
0.03
<0.01
<0.01
95.1
0.12
0.04
0.03
4.8
100
Purge Gas
Stream 226
kmol/hr Vol %
106
818
58
20
304
4.0
0.08
0.05
1309
8.1
62.5
4.4
1.5
23.2
0.3
0.01
0.04
100
Expansion Gas
Stream 225
kmol/hr
2.5
7.5
11
0.8
3.3
4.6
0.1
0.06
30
Vol %
8.3
25.0
36.7
2.7
11.0
15.3
0.3
0.2
100
Fuel Grade
Hethanol
Stream 108
kmol/hr Wt %
0.01
0.02
1.3
0.01
0.02
6120
5.1
1.9
0.02
11
6139
<0.01
<0.01
0.03
<0.01
<0.01
99.7
0.12
0.04
<0.01
0.1
100
Distillation
Wastewater Off gas
Stream 229 Stream 228
kmol/hr Wt % kmol/hr
0.
0.
0.
0.
0.
3.0 1.0 0.
0.
0.001 <0.01
1.2 0.74
538 98.25
542 100 0.
003
003
2
002
003
1
002
35
Vol %
1.0
1.0
62
1.0
1.0
33
1.0
100
Material flow estimates are based upon published engineering studies (1,2).
-------
Section 3
Product Synthesis
3.4.2 Fischer-Tropsch (F-T) Synthesis
The F-T process can produce a wide range of products from methane to
heavy fuel oil. Generally, fluidized bed designs operating at higher tem-
peratures produce a lighter mix of products than fixed bed designs. For pur-
poses of analysis, it is assumed that the fluidized bed design similar to
that used at Sasol, S.A. (called the Synthol process) would be utilized in
the U.S. since major emphasis in synfuels production would be on light motor
fuels. However, even the Synthol process produces a range of products from
middle distillates to methane. An indirect liquefaction plant using F-T
synthesis could be designed to produce mostly liquid products by: (1) steam
reforming of methane and other light hydrocarbons for recycle and (2) cata-
lytic cracking of heavier oils. Such an approach, however, would result in
low overall thermal efficiencies due to extensive heat losses. Thus, a mix
of hydrocarbon products including SNG, LPG, gasoline, diesel fuel, alcohols,
and ketones may be a more practical scenario for U.S. facilities.
Figure 3-11 is a flow diagram of the Synthol process (1). Purified
synthesis gas and recycle gas are compressed together and heat exchanged
against hot reactor product. Synthesis gas is mixed with circulating iron
catalyst in the Synthol reactor where the synthesis reactions proceed. Re-
action heat is removed by hot oil circulating in tubes internal to the
reactor. Catalyst and vapor products are separated in a cyclone system and
catalyst solids are recycled. Crude product vapors are cooled in a hot wash
tower which uses cooled F-T recycle oil as the wash medium,, Heat is re-
covered via feed/product exchange and by generation of steam in waste heat
boilers.
Heavy oil condensate is sent to product fractionation while hot wash
tower overhead is sent to a cold wash tower for recovery of lighter oils.
In the cold wash tower, cool process water spray effects condensation of
96
-------
1JO
RECYCLE GAS
PURGE GAS TO
METHANATION
SULFUR FREE
SYNTHESIS GAS
OFF GASES TO SNG
CO PRODUCTION
WASTEWATER
Figure 3-11. Fischer-Tropsch (Synthol) synthesis and product recovery
-------
Section 3
Product Synthesis
most of the light hydrocarbons; oxygenated organics (alcohols, ketones,
acids) become dissolved in the aqueous condensate. Liquid light oil is
further washed with process water and sent to product fractionation. The
aqueous condensate is neutralized with lime and stripped to recover alcohols
and ketones. Stripper bottoms, containing mostly organic acids, constitute
the aqueous waste from the F-T process. Overhead vapors from the cold wash
tower are split into a recycle and a purge stream, with the latter sent to
catalytic methanation.
Not shown in Figure 3-11 is the catalyst preparation process. At
Sasol, S.A., catalyst is prepared on-site from iron ore via reduction with
fuel gas (39). Other transition metals are added as promoters during the
preparation step. It is not known whether F-T catalyst would be prepared
on-site (at least the ore reduction step) for U.S. facilities; or purchased
from a vendor. For simplicity, it is assumed that no preparation other than
mixing/packaging would be conducted on-site and thus no wastes are generated
by the preparation step.
Table 3-15 presents example material flow estimates for F-T synthesis.
About 15% of the heating value of F-T products is accounted for by purge gas
which is subsequently converted to SNG. Light hydrocarbons in the fractiona-
tion light gas would be recovered as LPG and a fuel gas (or as methanation
feed). F-T middle oil is about 80% gasoline range hydrocarbons and 20% diesel/
fuel oil hydrocarbons. Alcohols (and ketones) represent significant chemical
by-product(s). A relatively large amount of wastewater is generated in the
F-T process, since approximately one mole of water is produced for each mole
of carbon oxides reacted during hydrocarbon synthesis.
3.4.3 Methane Co-production (Fischer-Tropsch Synthesis Case)
Purge gas from light ends recovery in the F-T synthesis case contains
large amounts of methane along with CO and H2. An efficient way to recover
the fuel value of these purge gases is to convert residual hydrogen and
98
-------
TABLE 3-15. FISCHER-TROPSCH SYNTHESIS-MATERIAL FLOW ESTIMATES FOR K-T GASIFICATION (ILLINOIS NO. 6 COAL)*
IO
Purge Gases to
Feed Gas Methanation
Stream 14 Stream 25
H2
N2 + Ar
CO
co2
CH4
H20
C2H4
C2H6
C3H6
C3H8
C4H8
C4H10
VC7
C? + oils
Organic Acids
Total kmol/hr
Total kg/hr
kmol/nr vol % kmol/hr
13664 67.2 973
307 1.51 307
5719 28.1 61.4
610 3.00 396
20.8 0.102 612
12.4
118
177
240
40.0
122
15.7
108.88
21.7
20320 3204
78340
Vol %
30.4
9.57
1.92
12.3
19.1
0.386
3.67
5.52
7.49
1.25
3.82
0.491
3.402
0.678
Fractionation
Wastewater from Offgases to
Alcohol Recovery C02 Removal
Stream 223 Stream 224
kmol/hr Wt % kmol/hr
5.42
5.17
1.18
139
44.8
8820 98.9 12.3
61.2
103
65.4
6.31
0.08
0.01
0.03
0.22
29 1.1
8849 443
160600 15100
Vol %
1.22
1.17
0.267
31.4
10.1
2.78
13.8
23.2
14.8
1.42
0.002
0.003
0.006
0.05
Methanation Wastewater CO? Removal Dehydration
Condensate from C02 Offgas Offgas
Stream 236 Stream 235 Stream 239 Stream 240
kmol/hr Wt % kmol/hr Wt % kmol/hr Wt % kmol/hr Vol %
0.0287 0.01
0.794 0.291
0.0042 0.001
253 . 92.7
15.5 5.67
703 98.1 22.9 100 3.01 1.11 10.6 100
0.252 0.06
0.150 0.09
0.016 0.006
0.159 0.06
4.2 1.9
707.2 22.9 100 273 10.6 100
12910 413 11490 191
CO-Product SNG
Stream 107
kmol/hr Vol %
2.77 0.188
302 20.5
0.301 0.02
131 8.94
1028 69.9
6.35 0.432
1470
33240
Material flow estimates are based upon published engineering studies (1). Main products from Fischer-Tropsch
synthesis are: Blended gasoline (Stream 100) - 41320 kg/hr
C3 LPG (Stream 106) - 2699 kg/hr
C4 LPG (Stream 106) - 410 kg/hr
Diesel oil (Stream 101) - 8380 kg/hr
Heavy fuel oil (Stream 102) - 2454 kg/hr
Mixed alcohols (Streams 103,104,105) - 6995 kg/hr
-------
Section 3
Product Synthesis
carbon oxides in the gases to additional methane, producing high heating
value gas equivalent to pipeline gas (substitute natural gas or SNG). Metha-
nation involves the catalytic reaction of carbon oxides in the gases to
methane as indicated by the following reactions:
3H2 + CO = CH4 + H20 + heat
4H2 + C02 = CH4 + 2H20 + heat
Figure 3-12 is a simplified flow diagram for SNG production showing catalytic
methanation, C02 removal, and drying.
The methanation reactions which are carried out over nickel-based cata-
lyst at 573 and 753K and 7 MPa (54) are highly exothermic. In the fixed
bed design shown in Figure 3-12 temperature is controlled by recycle of
cooled product gas. Also, a large amount of steam is recovered in this pro-
cess. Water vapor formed during methanation is condensed in knockout drums
and the condensate is subsequently depressurized for reuse as a boiler feed
water. Methanator product gas is most commonly treated for CCL removal
using an amine-based acid gas removal system (e.g., monoethanolamine). Final
moisture removal is usually accomplished in a triethylene glycol (TEG)
absorber.
The material flow estimates for SNG co-production have been included
in Table 3-15.
Waste streams generated continuously during SNG co-production are metha-
nation and C02 removal, C02 containing offgases, dehydration offgas, metha-
nation catalyst, decommissioning offgases and spent methanation catalyst.
3.4.4 Mobil M-Gasoline Synthesis
The Mobil M-gasoline process is depicted in Figure 3-13 (1). Crude
methanol is vaporized by heat exchange with reactor product and fed to the
dimethyl ether (DME)reactor where it is catalytically converted to an
100
-------
CATALYST
DECOMMISSIONING
OFF-GAS
FRACTIONATION
OFF-GASES
DEHYDRATION
OFF-GASES
COMPRESSOR
COMBINED
CONDENSATES
Figure 3-12. Methanation, C02 removal, and drying for SNG production
-------
-5
O>
CO
I
CO
o
s:
Q.
to
-5
Ol
o
-s
o
cr
OJ
CO
O
Z5
05
t-h
3"
0)
O
-S
Q-
fD
O
Q.
QJ
O
QJ
rt
-------
Section 3
Product Synthesis
equilibrium mixture of methanol, DME, and water vapor. DME reactor product
is subsequently converted to hydrocarbons in M-gasoline reactors, with
temperature control obtained by recycling cooled purge gas from product
separation. The DME reactor inlet conditions are about 573K and 2.3 MPa
while M-gasoline inlet conditions are 603K. Product vapors from the M-gaso-
line reactors are cooled by methanol feed heat exchange, by generation of
steam in a waste heat boiler, and by air cooling. Crude liquid product is
separated in a knockout drum. Drum overhead is split into a recycle stream
and a purge stream used as plant fuel. The bulk of liquid product is sent
to the product fractionation unit for light ends recovery, with a small slip-
stream injected at the inlet to the boiler feed water heat exchanger to con-
trol Durene* crystallization. The aqueous condensate from the knockout drum
(Stream 233) constitutes the only continuous wastewater stream from the
process.
Both the DME catalyst and M-gasoline catalyst require periodic regenera-
tion. The DME catalyst accumulates coke slowly and requires regeneration
perhaps once or twice per year. The M-gasoline catalyst requires regenera-
tion about every two weeks to remove coke. Regeneration is accomplished
using No to purge hydrocarbons from the system followed by air injection.
Control of inlet CL level and injection of cooled recycle gas to the system
maintain combustion temperatures below 753K. Regeneration offgas is cooled
by exchange with fresh regeneration gas and by air cooling. Any water of
combustion is condensed in a knockout drum before depressurization and vent-
ing to the atmosphere. As depicted in Figure 3-13, five M-gasoline reactors
constitute a "train" with one reactor being regenerated while four are in
service. Thus, regeneration is a regular but intermittent process resulting
in the routine generation of an offgas.
Table 3-16 presents example mass flow calculations for Mobil M-gasoline
synthesis. As indicated in the table, a small methanol synthesis purge stream
*1,2,4,5-tetramethylbenzene (Durene) has a very high octane rating and is
desirable in product gasoline for that reason but it freezes at 353K.
103
-------
TABLE 3-16. MOBIL M-GASOLINE SYNTHESIS MATERIAL FLOW ESTIMATES FOR K-T GASIFICATION (ILLINOIS NO. 6 COAL)'
Crude Methanol Purge Gas Expansion Gas
Stream 20 Stream 226 Stream 225
Constituent kmol/hr Wt - fcmol/hr Vol i kmol/hr Vol
CO 0.02 <0.01 106 8.1 2.5 8.3
H2 0.02 <0.01 818 62.5 7.5 25.0
N£ + Ar 0.02 <0.01 304 23.2 3.3 11.0
C02 1.5 0.03 58 4.4 11 36.7
CH4 0.01 <0.01 20 1.5 0.8 2.7
' C2H4
C2H6
H20 549 4.8 0.05 0.004 0.06 0.2
CH3OH 6123 95.1 4.0 0.3 4.6 15.3
(CH3)20 5.1 0.12 0.08 0.01 0.1 0.33
C2HgOH 1.9 0.04
C3H?OH 1.2 0.03
C3H6
C3H8
1 C4H10
C4H8
n C4H10
i C5H12
C5H10
n C5H12
C6
C,HfiO
CH202
Coke
Total moles 6682 100 1309 100 30 100
Crude Product
Stream 21
kmol/hr
0.7
1.6
0.01
3.5
39.0
1.0
12.6
4.2
3.9
89.0
131.2
16.4
40.6
144.0
26.6
16.4
531
1061
Vol ':>
0.07
0.15
0.001
0.33
3.7
0.10
1.2
0.4
0.36
8.4
12.4
1.6
3.8
13.6
2.5
1.5
50.0
100
Fuel Gas
Stream 230
kmol/hr
0.3
0.9
0.006
0.3
6.8
0.06
0.6
0.2
0.08
1.6
1.2
0.1
0.3
0.5
0.09
0.05
0.3
13.4
Vol '
2.3
6.8
0.04
2.6
50.8
0.5
4.3
1.6
0.6
11.6
8.7
1.0
2.1
3.7
0.7
0.4
2.5
100
Condensate
Stream 233
kmol/hr Wt "
0.04
0.68 0.03
6099 99.0
1.35 0.10
7.92 0.42
10.0 0.42
6119 100
Material flow estimates are based upon published engineering studies (1).
-------
Section 3
Product Synthesis
is produced which accounts for less than 10% of the total product heating
value. Most of the remaining fuel value is recovered as gasoline and LPG.
Mobil M-gasoline offgas and fractionation light gas are used as plant fuels.
The aqueous condensate waste (Stream 233) contains the oxygenated organics
(ketones, acids) which are generated in small amounts in the Mobil M-gasoline
catalytic reactors.
3.4.5 Product Recovery and Upgrading
The crude liquid fuel products of methanol, F-T, and Mobil M-gasoline
syntheses may require upgrading on-site to yield final products which are
marketable as substitutes for petroleum-derived fuels and chemicals. This is
particularly true for motor gasolines, where crude coal-derived gasoline
fractions would not meet octane requirements for the retail market in the
U.S. F-T and Mobil M-gasoline products could be upgraded by catalytic
alkylation of the C3-C4 fraction to yield gasoline-blend hydrocarbons and
commercial grade LPG, by hydrotreating (in the F-T case for destruction of
olefins and oxygenated organics), by catalytic reforming to produce more
cyclic and branched chain hydrocarbons, by 05/05 isomerization to increase
the antiknock quality of pentanes and hexanes, and by catalytic polymeriza-
tion to convert propene/butene fractions into higher molecular weight gaso-
line blending compounds. All of these upgrading processes are standard
refinery technology. Further, waste streams generated during upgrading (e.g.,
alkylation sludges, condensates) are not expected to present treatment prob-
lems other than those encountered in existing refineries since feed streams
to upgrading processes have no unique characteristics differing from current
refinery experience. In fact, the absence of sulfur and nitrogen in the syn-
thesis products elminates waste streams such as sour waters and gases often
generated in the upgrading of petroleum fractions. Control of refinery
waste streams has been discussed in several reports (55,56,57) and therefore
will not be addressed in this manual. For these reasons and due to the multi-
plicity of possible options for product upgrading, the "plant boundary" chosen
105
-------
Section 3
Product Synthesis
for purposes of this manual includes only crude product separation/fractiona-
tion. The one exception to this defined plant boundary is the fugitive
organic emissions inventory which includes emissions from all upgrading pro-
cesses (refer to Section 3.7).
Tables presented previously in Section 3.4 have shown the characteristics
and flow rates of crude synthesis products consistent with the plant boundary
defined for purposes of this manual. Table 2-2 in Section 2 presented typical
upgraded product slates for K-T based facilities. This mix of upgraded pro-
ducts is assumed in Section 3.6 for purposes of estimating evaporative emis-
sions from product storage.
3.4.6 Haste Streams Generated by Synthesis Operations
Synthesis operations do not generate any continuous gaseous waste
streams. All purge gases are valuable as sulfur-free fuels or as SNG feed.
However, periodic regeneration or decommissioning of catalysts results in
the generation of offgases. In addition, spent catalysts requiring dis-
posal constitute an intermittent solid waste. Aqueous wastes are routinely
generated by methanol, F-T, Mobil M-gasoline, and SNG synthesis. In all but
the SNG case, these wastewaters contain oxygenated organics requiring treat-
ment. SNG condensates contain only dissolved gases and are reusable as
boiler feed water after degassing. Water produced during methanol synthesis
is found in the crude methanol fuel product.
Spent Methanol Synthesis Catalyst (Stream 227)
No data are currently available in the public domain relating to the
characteristics or quantity of spent methanol catalyst. For the subject
facility a catalyst inventory of about 300 Mg having a useful life of 3-5
years is assumed. Therefore, based on these assumptions, the average annual
spent catalyst rate is 60 to 100 Mg.
106
-------
Section 3
Product Synthesis
Spent F-T Catalyst (Stream 222)
As with methanol catalyst, no data are currently available on properties
of spent catalyst. Further, amounts and types of metals other than iron
which are used in formulations of fresh catalysts are proprietary. Based on
data for the Sasol, S.A. facility, about 5200 Mg/year of spent F-T catalyst
would be generated (1).
Spent Mobil M Catalyst (Stream 232)
Mobil M catalysts are zeolite-based (synthetic clay-like) materials. No
data are publicly available at present on the characteristics of these cata-
lysts. It is estimated that the subject facilities would generate about 80
Mg/year of spent DME catalyst and about 300 Mg/year of spent M-gasoline
catalyst (1).
Mobil M Synthesis Catalyst Regeneration Offgas (Stream 231)
Based on design data contained in a Mobil report (1) concerning the
number of catalyst vessels, regeneration frequency, regeneration duration,
and offgas volume, it is anticipated that catalyst regeneration will occur
over a period of about 3800 hours per year with an average offgas flow rate of
3
approximately 4600 Nm /hr. Pollutants of concern in the offgas stream which
may require control are VOC, carbon monoxide, and particulate matter. No data
are currently available on the composition of the subject waste gases.
Spent Methanation Catalyst (Stream 238)
Nickel-based methanation catalysts are eventually deactivated by phy-
sical degradation of crystal size and by chemical accumulation of poisons
such as sulfur. No data are currently available on the Teachability of
nickel from spent methanation catalyst. The average generation rate of spent
catalyst is estimated at about 40 Mg/year.
Methanation Catalyst Decommissioning Offgas (Stream 237)
Methanation catalyst contains nickel in reduced form and is thus
107
-------
Section 3
Product Synthesis
pyrophoric in nature. Prior to removal of spent catalyst from the bed, the
material is oxidized with air in a controlled manner to convert nickel to
its oxide. No information is available on the characteristics of the cata-
lyst decommissioning offgas.
Methanol Synthesis Condensate (Stream 229)
Based on the calculations in Table 3-14, methanol synthesis condensate
has the following characteristics:
3
Production rate 10 m /hr
Methanol 10,000 mg/L
Ethanol 5 mg/L
Propanol 7,400 mg/L
COD 33,000 mg/L
TOC 7,400 mg/L
The condensates will contain very low inorganic TDS levels, no sulfur or
nitrogen species, and no significant levels of trace elements. The volume
of this stream is determined by the C02 level in the feed gas which is, in
turn, determined by Rectisol design. The exact composition of methanol
systems condensate will vary with the specific process employed.
F-T Wastewater (Stream 223)
Condensates generated by the F-T product separation step have the
following estimated characteristics (1) (see Table 3-15):
3
Production rate 160 m /hr
Organic acids 10,800 mg/L
COD 12,000 mg/L
TOC 4,300 mg/L
Although the condensates would inherently have low inorganic TDS levels, dur-
ing product separation the F-T condensates are neutralized with lime or
caustic to allow distillation of alcohols and ketones while leaving acids
108
-------
Section 3
Product Synthesis
in the aqueous phase. Thus, the wastewaters would also have high levels of
alkalinity and Ca or Na . No significant levels of trace elements are
expected.
Mobil M-Gasoline Wastewater (Stream 233)
Condensates generated by the Mobil product separation step have the
following estimated characteristics (1,2) (see Table 3-16):
o
Production rate 110 m /hr
Formic acids 4,200 mg/L
Acetone 4,200 mg/L
Cg + hydrocarbons 1,000 mg/L
COD 14,000 mg/L
TOC 4,000 mg/L
The waste will contain very low levels of inorganic TDS, no sulfur or nitro-
gen compounds, and no significant levels of trace elements. The estimates
for gross parameters or constituents may be lowered when chemical recovery
is practiced. Also, the specific catalyst will affect condensate quality.
Methanol synthesis condensate is not generated separately in Mobil M-gasoline
facilities. Water produced in methanol synthesis is ultimately found in
Mobil M-gasoline wastewater.
Methanation Condensate (Stream 236)
Water contained in crude SNG is condensed under pressure and contains
about 10 mg/L CH4 and 20,000 mg/L of C02 (58). About 12,910 kg/hr of con-
densate are generated. Since this stream contains essentially no dissolved
solids, it is reusable as boiler feed water after depressurization and air
or N2 stripping to remove dissolved gases. In an integrated facility, metha-
nation condensate would be considered an internal process stream rather than
a waste stream.
109
-------
Section 3
Product Synthesis
C02 Offgas from SNG Purification (Stream 239)
In most cases, residual C02 in the methanation product (crude SNG)
would be removed to obtain a pipeline quality gas. Since amine processes
for C02 removal will also remove some CO, H2, and hydrocarbons, the offgas
will contain these constituents in the following approximate amounts (1,58):
C02 93 vol %
CO 15 ppmv
H2 105 ppmv
CH4 1 vol %
C2H. 550 ppmv
C2H6 925 ppmv
C3H6 582 ppmv
C^Hg 58 ppmv
H20 5.8 vol %
The total flow rate is estimated to be approximately 273 kmol/hr.
Dehydration Offgas (Stream 240)
The triethylene glycol regenerator offgas contains very small amounts of
methane and the glycol solvent. No data are available to quantify these con-
stituents. The offgas flow rate is around 1000-2000 kg/hr and consists
mostly of water vapor.
C02 Removal Wastewater (Stream 235)
The fractionation offgas sent to the C02 removal unit in an F-T synthe-
sis facility is estimated to contain appreciable quantities of water. As a
result, wastewaters are generated in the C02 removal step. The flow rate of
this waste stream is estimated at approximately 400 kg/hr. Composition data
are not available for this waste stream, but it should contain only minor
quantities of dissolved gases such as C02. As such it could be combined
with the methanation condensate for reuse within the facility.
110
-------
Section 3
Products/By-Products
3.5 PRODUCTS AND BY-PRODUCTS
The products/by-products considered in this section include those pro-
duced as substitutes for petroleum-derived fuels or chemicals and sulfur re-
covered as a result of air and water pollution control. The available data
on the composition of each product and by-product are presented, and those
substances or classes of substances which would be considered toxic are
identified. However, it should be noted that product and by-product specia-
tion data are generally limited. Data presented herein should not be con-
strued as an adequate basis for evaluating the potential environmental risks
associated with products and by-products.
3.5.1 Methanol Synthesis Product
Methanol is currently produced primarily from natural gas (59). The
composition of the crude product varies somewhat depending upon such factors
as the specific synthesis process used, the synthesis pressure and tempera-
ture, and the hydrogen to carbon monoxide ratio in the synthesis feed (59).
The primary reaction in methanol synthesis is:
CO + 2H2 CH3OH + heat
However, a number of side reactions also take place which introduce impuri-
ties into the crude methanol product. A representative, but not exhaustive,
list of impurities which can be expected in crude methanol is presented in
Table 3-17. An additional, and highly toxic, potential impurity not shown
in the table is iron carbonyl . Under certain conditions the formation of
this compound has been observed in the compression and synthesis sections
of methanol plants (60).
Table 3-18 shows an estimated composition of a crude methanol made from
coal. As with a natural gas feed the amounts and types of impurities present
will vary somewhat depending upon the specific synthesis process used and
111
-------
Section 3
Products/By-Products
process conditions. However, water is expected to be the largest single
impurity (5%) with all others comprising less than 1%. It should also be
mentioned that several methods for purifying methanol are currently available
(59), and the degree of methanol purification will largely be determined by
user needs. Thus, coal-derived methanol in commerce may range in purity from
about 95% pure in the case of the crude product to 99.85% pure for Grade AA
methanol.
TABLE 3-17. COMPONENTS REPORTED IN COMMERCIAL METHANOL (59)
Compounds
1. Dimethyl Ether
2. Acetaldehyde
3. Methyl Formate
4. Diethyl Ether
5. n-Pentane
6. Propionaldehyde
7. Methyl Acetate
8. Acetone
9. Methanol
10. Isopropyl Ether
11. n-Hexane
12. Methyl Propionate
13. Ethanol
14. Methyl Ethyl Ketone
15. t-Butyl Alcohol
16. n-Propanol
17. n-Heptane
18. Water
19. Methyl Isopropyl Ketone
20. Acetal
21. Isobutanol
22. n-Butyl Alcohol
23. Isobutyl Ether
24. Diisopropyl Ketone
25. n-Octane
26. Isoamyl Alcohol
27. 4-Methyl Amy! Alcohol
28. n-Amyl Alcohol
29. n-Nonane
30. n-Decane
112
-------
Section 3
Products/By-Products
TABLE 3-18. ESTIMATED COMPOSITIONS OF CRUDE METHANOL FROM COAL* (2)
Compound Concentration
CH3OH ~^%
C2H5OH, C3H7OH and
C4HgOH 2800 ppm
(CH3)20 150 ppm
Nonmethane HCs 600 ppm
H20 5.0%
*The composition of crude methanol is highly process specific. These esti-
mates are based upon the ICI synthesis process.
3.5.2 Fischer-Tropsch Liquid Products
The crude Fischer-Tropsch synthesis product is primarily composed of
straight-chained paraffinic and olefinic hydrocarbons (61). Minor quanti-
ties of aromatic, naphthenic, and branched-chain hydrocarbons are also pre-
sent, along with small amounts of oxygenated compounds such as alcohols,
aldehydes, ketones, and acids, most of which have fewer than five carbon
atoms (62). The crude F-T product (Stream 22) can be refined into several
different products including LPG (Stream 106), gasoline (Stream 100), diesel
oil (Stream 101), heavy oil (Stream 102), methanol (Stream 103), acetone and
methyl ethyl ketone (MEK) (Stream 104), and heavy alcohols (Stream 105).
Much of the chemical composition data presented here is based on analyses
of products from the commercial-scale Fischer-Tropsch synthesis plant which
is currently operating in Sasolburg, South Africa.
Limited data on the composition of gasolines, diesel oils, and heavy
oils from F-T synthesis indicate that they are essentially nitrogen-and-
sulfur-free (63). Crude Fischer-Tropsch gasoline requires upgrading prior
113
-------
Section 3
Products/By-Products
to its use as a motor fuel, and Table 3-19 shows the estimated chemical com-
position, by compound class, of the finished Fischer-Tropsch gasoline. The
aromatics content (17%) is lower than that of typical petroleum gasolines
(23-36%). The saturates content is similar to that of petroleum-derived
gasoline, but the olefins content is much higher. The estimated Reid Vapor
Pressure of 69 kPa for finished Fischer-Tropsch gasoline is within the range
of values (48 to 100 kPa) for typical petroleum gasolines (64).
Table 3-20 shows the distribution of the oxygenated by-products from
fluid bed Fischer-Tropsch synthesis before further refining. At the SASOL
plant, the aldehydes are hydrogenated, and methanol is reported to be used
onsite as Rectisol solvent makeup. Ethanol, propanol, butanol, pentanol,
acetone, MEK, and a higher alcohol fraction are distributed commercially (63).
The SASOL operators also convert the propylene and butylene from the light
ends recovery to gasoline by polymerization over a solid phosphoric acid
catalyst. The propane and butane are sold as LPG (65).
3.5.3 Mobil M-Gasoline Products
The crude Mobil M-gasoline synthesis product (Stream 21) is fractionated
into a gasoline (Stream 109), mixed butanes (Stream 110) and propane (Stream
111). Table 3-21 which shows the expected composition of the synthesis pro-
duct, representing an average yield over 14 days of operation before catalyst
regeneration (1).
Table 3-19 presented the estimated composition of the crude Mobil M-
gasoline. It can be seen from this table that the olefins content of the
Mobil M-gasoline is higher than that of petroleum gasoline, while the satu-
rates and aromatics contents are within the ranges found in petroleum gaso-
lines. The benzene content of the finished Mobil M-Gasoline is also reported
to be less than the one percent by volume which is typical of petroleum gaso-
lines (64). As was mentioned earlier, iron carbonyl could be present in
trace quantities in the methanol feed. It is, however, expected that any
114
-------
TABLE 3-19. COMPARISON OF THE ESTIMATED COMPOSITION OF FINISHED INDIRECT COAL LIQUEFACTION, UNLEADED
GASOLINES, AND TYPICAL PETROLEUM GASOLINES
Fischer-Tropsch
Unleaded Gasoline
Component (Refs. 1,63)
Saturates
Olefins,
Aromatics
, vol %
vol %
, vol %
Sulfur, wt %
Nitrogen,
wt %
63
20
17
<1 ppm
<1 ppm
Finished
Unleaded Mobil M-
Gasoline (Ref. 1)
60
11
29
<1 ppm
<1 ppm
Crude
Mobil M-Gasoline
(Ref. 1)
56
13
30
<1 ppm
<1 ppm
Petrol eum- Deri ved
Gasolines
(Refs. 66,67,68)
56 -
4 -
23 -
0.014 -
0.05 -
69
8
36
0.417
0.49
-------
Section 3
Products/By-Products
carbonyls in the methanol feed would be trapped by the gasoline synthesis
unit's zeolyte catalyst and thus would not be present in the gasoline product.
TABLE 3-20. DISTRIBUTION OF OXYGENATED BY-PRODUCTS FROM FLUID-BED FISCHER-
TROPSCH SYNTHESIS (63)*
Component
Acetaldehyde
Propionaldehyde
Acetone
Methanol
Butyraldehyde
Ethanol
MEK
i-Propanol
n-Propanol
2-Butanol
Dimethyl Ketone-Methylpropyl
Ketone
i-Butanol
n-Butanol
n-Butylketone
2-Pentanol
n-Pentanol
Cg + alcohols
Wt %
3.0
1.0
10.6
1.4
0.6
55.6
3.0
3.0
12.8
0.8
0.8
4.2
4.2
0.2
0.1
1.2
0.6
The sum of reported component weight percentages does not equal 100%.
116
-------
TABLE 3-21. METHANOL CONVERSION UNIT FEED AND PRODUCT COMPOSITION (1)
(Basis: 100 kmol Methanol in Feed)
Component Names
Coke (as CHQ g)
Acetone
Formic Acid
Methanol
Dimethyl ether
Water
Carbon Monoxide
Carbon Dioxide
Hydrogen
Methane
Ethane
Ethene
Propene
Propane
n-Butane
i -Butane
Butenes
n-Pentane
i-Pentanes
Pentenes
Cycl opentane
Methyl cycl opentane
n-Hexane
i-Hexanes
Hexenes
Methyl cycl ohexane
n-Heptane
i -Heptanes
Heptenes
Feed
.000
.000
.000
100.000
.000
7.529
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
Product
.128
.129
.163
.000
.000
106.951
.017
.053
.040
.746
.193
.018
1.474
.064
.665
2.155
.270
.268
2.353
.435
.047
.214
.113
1.991
.297
.062
.026
.786
.292
117
(Continued)
-------
TABLE 3-21. (Continued)
Component Names
1,3-Dicyclopentane, cis
i-Octanes
Octenes
n-Propyl cycl opentane
n-Nonane
i-Nonanes
Nonenes
n-Butyl cycl opentane
i-Decanes
Decenes
Benzene
Tol uene
Ethyl benzene
m- + p-Xylenes
o-Xylene
1, 2, 4-Trimethyl benzene
1 , 3 , 5-Trimethyl benzene
p-Ethyl tol uene
i-Propyl benzene
1,2,4, 5-Tetramethyl benzene
1,2,3, 5-Tetramethyl benzene
1,2,3, 4-Tetramethyl benzene
p-Di ethyl benzene
Penta-Methyl benzene
2-Methyl naphtha! ene
Total kmol
Weight (kg)
Feed
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
107.529
3,339.846
Product
.233
.228
.301
.299
.015
.084
.116
.071
.024
.045
.036
.280
.070
.876
.240
.818
.034
.292
.014
.436
.063
.023
.198
.068
.017
124.832
3,339.846
118
-------
Section 3
Products/By-Products
3.5.4 Substitute Natural Gas (SNG)
The primary constituent of the SNG product (Stream 107) is methane with
smaller quantities of H,,, CO, CO,,, N2, and Ar. Standards for pipeline gas
generally require that the CO content be less than 1000 ppmv, and it is
expected that the crude SNG product would be sufficiently upgraded to meet
this criterion. Trace quantities of metal carbonyls may be produced during
catalytic methanation, during gasification, or by reaction of CO and Ni or
Fe in piping. Nickel carbonyl at concentrations of about 0.01 ppmv was found
in product gas from the Lurgi gasifier at Westfield, Scotland (60). Operators
of the Lurgi gasifier at Sasol, South Africa, reported that carbonyls were
not present in measurable concentrations in the product gas from their faci-
lity (69). Recent data from the Kosovo, Yugoslavia, Lurgi facility indicate
that carbonyls are not present to any appreciable extent in the desulfurized
process gases leaving the Rectisol unit (70). Carbonyl formation trends are
expected to be similar for K-T and Lurgi gasification systems.
3.5.5 LP Gas
The LP gas (Stream 106) from F-T synthesis will consist primarily of
propane and butane with smaller quantities of ethane, methane, and short-
chain olefins. As discussed in Section 3.4.5, it is not expected that coal-
derived LP gas will be significantly different in chemical composition than
petroleum-derived LP gas. However, this has not been verified with product
composition data from a commercial-scale production facility.
3.5.6 By-Product Sulfur (Stream 112)
Elemental sulfur is recovered as a by-product in the treatment of con-
centrated acid gases for air pollution control. The recovered sulfur may be
contaminated with a number of impurities. When the Stretford process is
employed, the by-product sulfur contains traces of vanadium, thiosulfate,
and thiocyanate. Claus sulfur may contain carbonaceous materials to the
119
-------
Section 3
Products/By-Products
extent that the by-product is at times termed "black sulfur." Depending on
site-specific factors, the sulfur may be saleable or may need to be disposed
as a waste.
120
-------
Section 3
Auxiliaries
3.6 AUXILIARIES
A number of non-pollution control auxiliary operations are associated
with a self-sufficient K-T indirect liquefaction facility. Included in these
are raw water treatment, power generation, process cooling, oxygen production,
and product/by-product storage. These support operations are sources of
additional waste streams which in many cases would be combined and treated
with the wastes generated in the main process train. The sections below pro-
vide brief descriptions of the most important auxiliaries and define the
expected characreristies of the major waste streams.
3.6.1 Raw Hater Treatment
The source and characteristics of the raw water assumed for the K-T
indirect liquefaction plant location are presented in Table 3-22. The com-
ponent concentrations presented represent average annual conditions. Varia-
tions in raw water characteristics, while important in the design of a faci-
lity, are not addressed here because they do not greatly affect the charac-
teristics of the plant's most environmentally significant waste streams.
Makeup water quantity and composition that determine the raw water treat-
ment load for a K-T based indirect liquefaction facility depend primarily
upon the plant location, feed coal composition, synthesis route, and the
extent of condensate/wastewater reuse achievable. An accurate estimate of
the makeup water requirements would require detailed site specific water and
energy balances around the plant which are beyond the scope of this manual.
However, rough approximations of the makeup water requirements have been made
based upon the major water/steam consuming operations within the facility.
These operations are process cooling, coal gasification and slag quench,
raw gas cooling and dust removal, cyanide wash (water wash case), and shift
conversion.
Water/steam consumption for each of these operations has been incorporated
into their respective material flow estimates presented throughout this section
121'
-------
Section 3
Auxiliaries
and are summarized in Table 2-23. The largest single consumer of makeup
water is the cooling water system (Section 3.6.3). Makeup requirements to
the cooling system are based upon published overall thermal efficiency data
for an Illinois No. 6 coal and different synthesis routes. Makeup water
requirements to the gas cooling and dust removal operation (Section 3.3.1)
are primarily dependent upon the soluble components in raw gas and dust,
particularly chlorides, and to coal moisture which is related to coal rank.
This requirement could decrease by as much as a factor of about three depend-
ing upon the feed coal composition. The other tabulated makeup water re-
quirements are also dependent upon feed coal characteristics, but to a lesser
extent, and are expected to have relatively little effect upon the total
makeup water requirements.
TABLE 3-22. COMPOSITION OF RAW MAKEUP WATER
Constituent
HC03
S04
Ca++
Mg++
Na+
cr*
TDS
Si02
PH
Assumed Source of Raw Water
Ohio River @ Grand Chain, 111.
(Ref. 71)
Concentration (mg/L)
no
60
36
9
30
15
250
6.5
8.1
*
Estimated from TDS minus major constituents and equivalence of cations and
anions
122
-------
Section 3
Auxiliaries
TABLE 3-23. ESTIMATED MAKEUP WATER QUANTITY FOR A K-T BASED INDIRECT
LIQUEFACTION PLANT (ILLINOIS NO. 6 COAL)
Process Makeup Water Quantity,
Cooling Water 820 - 1207
Cyanide Wash 0 - 240
Gas Cooling and Dust Removal 230
Shift Conversion 160
Gasification 40
Boiler Bottom Ash Removal 13-80
Flue Gas Desulfurization 10 - 60
(Power Generation)
TOTAL 1270 - 2020
Raw water makeup requirements can be reduced by the selective reuse of
wastewaters. Systems with high makeup flow requirements, such as the cool-
ing water, cyanide wash, and gas cooling and dust removal systems, offer the
greatest potential for cost effective wastewater recycle/reuse. These sys-
tems have lower makeup quality requirements and will allow recycle of some
wastewaters without extensive pretreatment. The same is true of the ash
sluicing and flue gas desulfurization systems. The high quality makeup
requirements to boilers make it difficult to recycle all but the highest
quality wastewaters (e.g., methanation/compression condensates, boiler
blowdown, and selected reclaimed wastewaters) to the boiler feedwater treat-
ment system.
Figure 3-14 shows a raw water treatment scheme for producing makeup
water in the base plant. Systems with lower quality requirements may choose
to withdraw water after any step in the treatment process that meets their
requirements. For example, cooling tower makeup could be supplied directly
from the sedimentation/equalization ponds in cases where raw water is low in
123
-------
RECLAIMED
WASTEWATER
METHANATION
CONDENSATE
BOILER
SLOWDOWN
RAW
WATER
SEDIMENTATION
AND
EQUALIZATION
MAKEUP WATER TO
GAS COOLING AND DUST REMOVAL,
CYANIDE WASH, AND GASIFIER
QUENCH
COAGULATION
AND
CLARIFICATION
r
FILTRATION
RAW WATER |
TREATMENT I
SLUDGES '
I
COOLING
TOWER
MAKEUP
DEMORALIZATION
BOILER
FEEDWATER
MAKEUP
REGENERATION
CHEMICALS
REGENERATION
WASTEWATER
NOTE:
DASHED LINES INDICATE
FLOW ALTERNATIVE
BACKWASH
Figure 3-14. Flow diagram for base plant raw water treatment system
-------
Section 3
Auxiliaries
hardness, alkalinity, and suspended solids. Where any of these constituents
are present at high concentrations, raw water may require further treatment
by softening, coagulation, clarification, or filtration prior to its use in
the cooling water system.
The sedimentation and equalization step in Figure 3-14 includes with-
drawal of water and storage in a reservoir. This storage provides a reliable
supply of water to the facility that is independent of river flow, reduces
the impact of daily water quality variations, and allows sedimentation of
silts and other suspended material.
The raw water leaving the reservoir is treated in a sludge contact
clarifier followed by a filter and a demineralizer. The clarifier is fed
with lime, a coagulant, and a polymer to coagulate and/or flocculate fine
suspended solids. The treated water then passes through a sludge bed of
previously formed floe. This sludge contact enhances the agglomeration and
settling characteristics of flocculated particles. An added benefit of this
unit is a partial reduction in the calcium alkalinity. Boiler blowdown is
recycled to the raw water treatment loop at this point to provide a mechanism
for the removal of residual boiler feedwater treatment chemicals prior to
demineralization. A filter is also provided to protect the demineralizers
from solids carryover from the clarifier.
Demineralization is accomplished in two steps. Strong acid and base
ion exchange units are used as the primary treatment step, and a mixed bed
polisher is used as the secondary step. The waste streams generated as a
result of this treatment scheme are discussed below.
Raw Water Treatment (Clarifier) Sludges (Stream 300)
Excess sludge consisting of calcium carbonate, magnesium hydroxide, and
coagulated particulate matter is removed from the clarifier. Table 3-24
presents expected sludge production rates for varying suspended solids con-
centrations and estimated characteristics of this waste. The impact of
125
-------
Section 3
Auxiliaries
suspended solids on the volume of sludge produced can be significant relative
to that of the chemical precipitates. However, even with high input TSS
levels, these sludges are low volume wastes which can easily be disposed of
independently or with gasifier or boiler ash or with flue gas desulfurization
sludges. This sludge could also be used as a neutralizing/scrubbing agent
to supply a portion of the scrubber makeup alkalinity requirements for a
throwaway FGD process.
TABLE 3-24. RAW WATER TREATMENT SLUDGE (STREAM 300) PRODUCTION RATES AND
CHARACTERISTICS
Sludge Production Rate
Suspended Solids Concentrations (mg/L) _ (Mg/hr)* _
0+ 0.135 - 0.176
10 0.172 - 0.224
50 0.332 - 0.432
Sludge Constituents Sludge Compositions (wt
CaC03 38.0
Mg(OH)2 2.0
60.0
Sludge is assumed to be 40% solids by weight
tClarifier sludge flows for the 0 mg/L suspended solids case derived from
calculations presented in Water and Waste Treatment Data, Permutit
Company, Inc., 1961
^Sludge composition figures are for case involving 0 mg/L of suspended
solids
Demineralizer Regeneration Wastewaters (Stream 301)
Table 3-25 summarizes the estimated characteristics of demineralizer
regeneration wastes from the strong acid and strong base exchangers. The
regeneration wastes from the mixed bed polisher will be comparable in dis-
solved solids to the compositions shown, but their flow will be intermittent
126
-------
Section 3
Auxiliaries
and small compared to those from the primary ion exchange step. Regeneration
wastewater flow rates (Stream 301) are calculated as 9% of the total water
flow rate into the demineralizer. The boiler feedwater (Stream 33) makeup
is assumed to be approximately 180 m /hr. At this makeup rate, the demin-
eralizer regeneration wastewater flow rate for Illinois No. 6 coal is approxi-
mately 18 m /hr. Wastewater flow rates will vary by coal type and synthesis
process in the same manner as the raw water makeup requirements for the
boiler feedwater system. In addition, in an actual plant the regeneration
wastewaters may be produced on an intermittent basis and require flow equali-
zation prior to treatment.
TABLE 3-25. DEMINERALIZER REGENERATION WASTEWATER COMPOSITION* (STREAM 301)
Constituent
HCOg
OV/ n
4
Ca++
Mg++
Na+
cr
TDSf
Si02
PH
Regenerated Wastewater for
Illinois No. 6 Coal Case
0
5,037
156
89
1,758
167
7,207
72
1.7
*
All units are mg/L except pH. 33% regeneration efficiency and the use of
H?SO* and NaOH as regeneration chemicals assumed.
t
Total dissolved solids calculated as a sum of the ions except Si02.
127
-------
Section 3
Auxiliaries
3.6.2 Power Generation and Process Heating
Self-sufficient indirect liquefaction facilities would require boilers
for power generation and heaters for various process units including coal
drying and pulverizing. In addition to boilers and heaters, other auxiliaries
could include dedicated gasifiers for producing low heating value fuel gas,
electrical generating units, and gas turbines. The size of the boiler would
be determined primarily by the extent to which electric drivers are used
versus steam or gas turbine drivers. Where electrical drivers are used,
steam demand for electricity production may differ (qualitatively and quanti-
tatively) from the case where steam drivers are used. When gas turbine
drivers are used, steam requirements would be greatly reduced. A large
number of possible auxiliary configurations exist, and it is beyond the scope
of this manual to perform the detailed engineering required to assess all of
these configurations. For purposes of analysis, electric drivers were
assumed to be employed to the maximum practical extent, and the electrical
energy needed is generated on-site.
A pulverized coal-fired boiler is assumed to supply the facility with
all steam not produced in process waste heat boilers (e.g., gasifier, shift,
synthesis). The boiler and associated systems are of conventional design
using a regenerative air heater which preheats the combustion air to 533K
by exchange with flue gas and an economizer to preheat boiler feedwater to
588K. Flue gas exits the system at 450K and bottom ash at 811K. Boiler
thermal efficiency is approximately 90% (LHV) for Illinois No. 6 coal.
Steam is used for process purposes (gasification, shift) for direct heating,
in turbine drivers for motive power (e.g., compressors), and for generation
of electric power. For certain purposes, steam superheating may be necessary.
As indicated earlier, 278,400 kg/hr of Illinois No. 6 coal (dry basis)
is being gasified to produce approximately 113 TJ/day of fuel grade methanol
and other products (methanol synthesis case). An overall plant thermal
128
-------
Section 3
Auxiliaries
efficiency of 52.4% has been assumed for a K-T indirect liquefaction plant
employing methanol synthesis (3). The amount of boiler coal needed for
self-sufficient facility operation is then approximately 22,650 kg/hr (dry
basis) for a K-T based methanol plant. The quantity of coal feed to the
boiler will vary depending upon the type of coal being fired and the overall
plant efficiency. The overall plant efficiency is different for different
synthesis processes. Both Mobil M and Fischer-Tropsch syntheses are assumed
to be less efficient than methanol synthesis, approximately 44.8 and 40.0%,
respectively.
Coal requirements for hot gas generation associated with coal drying and
pulverizing operations are included implicitly in the overall plant thermal
efficiency. Therefore, using the overall thermal efficiency in estimating
the coal feed to the boiler, coal which may actually be consumed in the coal
preparation unit necessarily reports as boiler feed coal. The amount of coal
which would be combusted in-conjunction with the coal preparation operations
(assuming coal is burned rather than fuel gas) is dependent upon a variety
of factors including the ROM coal moisture and the residual coal moisture
requirements for coal gasification. For the subject Illinois No. 6 coal,
the coal consumption in coal preparation is estimated to be about 2500 kg/hr
(as received basis) which corresponds to about 10% of the feed rate to the
boiler in a K-T based methanol plant. Because coal combustion related to
coal preparation is not unique, no attempt has been made to differentiate
between flue gases associated with coal preparation and those associated
with steam and power generation.
Mass flow calculations for a pulverized coal fired boiler associated
with a K-T based methanol plant utilizing Illinois No. 6 coal are presented
in Table 3-26.
129
-------
TABLE 3-26. BOILER MASS FLOW FOR ILLINOIS NO. 6 COAL* - METHANOL SYNTHESIS CASE
co
o
Components
Gaseous
N2
HjO
co2
°2
so2
N02
CH. (and other
hydrocarbons)
CO
CHjCHO (and other
aldehydes)
Ar
Total
Solid
C
H (non water)
0 {non water)
S
N
Ash
Water
Total
Molecular
or Atomic
Weight
28.01
18.015
44.0098
31.9988
64.0588
46.0055
16.0426
28.0104
44.0530
39.948
--
12.011
1.0079
15.9994
32.06
14.0067
--
18.0152
-
Fuel Feed
Stream 30
Weight % kg/hr knol/hr
70.757 17854.1 1486.48
4.786 1207.65 1198.18
8.933 2254.1 140.89
3.099 781.97 24.39
1.334 336 61 24.032
10.091 2546.26
1.00 252.33 11.01
100 25233 2887.98
Air Feed
mole t, kg/hr knol/hr
76.47 213989.96 7639.77
2.069 3723.7 206.7
0.032 140.7 3.197
20.513 65567.5 2049.06
0.915 3651.8 91.414
100 287073.7 9990.1
Flue Gas Discharge
Stream 302
mole X kg/hr kr.iol/hr
73.67 214273.7 7649.9
7.89 14760.2 819.33
14.05 64221.7 1459.26
3.26 10821.4 338.18
0.21 1406.2 21.951
0.036 173.603 3.773
23ppmv 3.785 0.236
43ppmv 12.616 0.45
0.14ppmv 0.063 0.00143
0.88 3651.8 91.414
100 309325 10384.5
321.37 26.76
62.56 1.95
2037.01
2420.94 28.71
Bottom Ash
Stream 304
kg/hr kmol/hr
35.71 2.97
15.63 0.49
509.25
560.59 3.46
Boiler mass flows will increase
represent engineering estimates
by 215% for the Fischer-Tropsch synthesis case and decrease by 48% for the Mobil M synthesis case. Tabulated naterial flows
-------
Section 3
Auxiliaries
An integrated facility would also have a number of small gas-fired
heaters serving various process units (e.g., startup heaters). Such heaters
would likely utilize sulfur-free waste gases from synthesis/fractionation
operations as the most convenient fuel. Since the contribution of small
heaters to sulfur and particulate emissions is expected to be minimal, pol-
lution control alternatives (other than for NOX) for these small heaters are
not discussed in this report.
Boiler Flue Gases (Stream 302)
Table 3-26 also contains the estimated composition of the combustion
flue gases for the Illinois No. 6 coal case. In addition to the high load-
ings of S02 and total particulates, these flue gases contain both particulate
and volatile trace elements derived from the coal. NO emissions (as N09)
X c.
were assumed to be controlled by boiler design and were estimated to be 260
ng/J. New pulverized coal fired boiler designs include some type of NOX con-
trols. However, if boiler design does not incorporate NOX controls, uncon-
trolled NO emissions are expected to range from 280 ng/J to 430 ng/J. In
A
the case of the boiler in Table 3-26, this would result in an increase of
8 to 65% in NO emissions. In an integrated facility, the flue gas would
A
represent one of the major uncontrolled gaseous waste streams in terms of
pollutant loading and volume. It should be noted, however, that combustion
emissions are not unique to indirect liquefaction facilities and generally
present no new problems for emissions control over those encountered in elec-
tric utility or industrial applications.
Boiler Bottom Ash (Stream 304) and Fly Ash (contained in Stream 302)
Although data on the characteristics of bottom and fly ash for Illinois
coals are available, no data were available for the subject Illinois No. 6
coal but its bulk composition is expected to reflect the major inorganic ele-
ments found in the raw coal (see Table 3-2). Table 3-27 summarizes available
data on the maximum levels of various constituents which have been reported
131
-------
Section 3
Auxiliaries
in ash slurry waters from coal-fired boilers.
TABLE 3-27. COMPARISON OF ASH AND ASH SLURRY MAXIMUM CONCENTRATIONS (72)
As
Ba
Cd
Cr
Pb
Hg
Se
Ag
F-
cr
Cu
Fe
Mn
so4
Zn
Fly
Ash
ppm
1,700
13,900
250
7,400
1,600
22
500
50
624
25,000
3,020
289,000
4,400
13,000
Bottom
Ash
ppm
40
4,000
250
270
35
4
7.7
25
100
1,800
720
204,000
720
950
Slurry
mg/L
0.12
3.0
0.052
0.17
0.2
0.026
0.05
0.02
16.2
2,415
0.45
11.0
1.1
2,300
2.7
Boiler Slowdown (Stream 303)
The quality of the boiler blowdown wastewater stream will be dictated
by the boiler drum operating pressure. In this analysis a boiler drum opera-
ting pressure of 10.3 MPa is assumed for all synthesis process cases.
Maximum silica concentrations and specific conductivity concentrations
allowed in boilers at this pressure are 2.0 mg/L and 150 micromhos/cm,
132
-------
Section 3
Auxiliaries
respectively. As the pressure of the blowdown stream is reduced to atmo-
spheric pressure, 62% of the blowdown will flash to steam leaving a stream
with a specific conductivity of approximately 400 micromhos/cm. Assuming
a one to one ratio between specific conductance and total dissolved solids,
this results in a maximum dissolved solids concentration of 400 mg/L. The
dissolved solids will contain varying quantities of phosphates or other
treatment chemicals, trace metals, and contaminants resulting from condenser
leakage. This water can either be reused directly (e.g., as cooling tower
makeup) or returned to the boiler feedwater pretreatment system as discussed
in Section 3.6.1. It is assumed that 1% of the steam made is lost as blow-
down from the steam drum. In the case of the K-T based methanol plant this
is equivalent to 2145 kg/hr.
An energy efficient indirect liquefaction plant will produce large quan-
tities of low pressure steam from process waste heat boilers. The specific
conductivity of the blowdown from these low pressure boilers can range as
high as 2000-5000 micromhos/cm for boilers in the 4.1 MPa to 6.9 MPa range.
After flashing to atmospheric pressure, the blowdown concentrations can be
two to three times higher. The design, economic, and site-specific considera-
tions involved in determining the quantities of low pressure and high pres-
sure steam generated in the model plants are beyond the scope of this study.
Supply steam pressure, low pressure boiler makeup water treatment, and
cascading reuse of high pressure boiler blowdown in the low pressure boiler
system are all considerations that must be evaluated in terms of the overall
water management plan at a specific site.
These blowdown flow rates can increase dramatically if severe condenser
leakage occurs. However, since operation under severe leakage conditions
cannot be tolerated for a long period of time, the impact of these upset
conditions should be minimal.
133
-------
Section 3
Auxiliaries
3.6.3 Cooling Operations
In an indirect liquefaction facility, a certain portion of the input
energy to the plant will be rejected as waste heat. The exact amount of
heat lost will be a function of both process design and operating practices
and will be highly plant-specific. Further, the cooling water evaporation
rate will be a function of the amount of wet versus dry cooling used at a
given site. This, in turn, will be affected by design decisions based upon
climatic factors and raw water costs. Since detailed designs and heat
balances were not developed for each of the indirect liquefaction facilities
addressed in this manual, some simplifying assumptions were made to develop
representative cooling system loads.
Table 3-28 summarizes the discharge rates expected to arise from cool-
ing tower operations. The energy rejection rate from the cooling system
was obtained by assuming an overall conversion efficiency of coal energy (HHV)
to useful product energy of 52.4% in K-T indirect liquefaction plants and
assuming 40% of the unrecovered thermal energy in the feed coal is rejected
through evaporative cooling in a cooling tower (63). Under these assumptions,
the energy rejected through the cooling tower will not be significantly
affected by the different coal types employed. However, the energy rejec-
tion rate will be affected by the different overall efficiencies of the
Methanol, Fischer-Tropsch, and Mobil M synthesis processes, and the varia-
tion in the energy rejection rates is directly reflected in the calculated
cooling water evaporation rates for each model plant.
Discharges from cooling systems consist of: Slowdown water (Stream
307), evaporative losses (Stream 306), which include evaporated water, en-
trained water (drift), and stripped gases. Table 3-29 summarizes the cooling
tower blowdown and drift characteristics for a cooling system operating at
5 cycles of concentration for a K-T plant using the raw waters described in
Section 3.6.1 for cooling tower makeup. The blowdown/drift characteristics
134
-------
Section 3
Auxiliaries
are intended to be typical of the concentrations expected for the Grand
Chain, Illinois location where raw water is used as makeup. These charac-
teristics do not necessarily represent optimum conditions for any given site,
TABLE 3-28.
COOLING SYSTEM MAKEUP WATER REQUIREMENTS FOR A K-T INDIRECT
LIQUEFACTION PLANT*
WaterRate (m3/hr) for Illinois No. 6 Coal Case
Stream Stream Methanol Mobil M Fischer-Tropsch
Name Number Synthesis Synthesis Synthesis
Cooling water 306
evaporation
rate (E)
Drift loss (D) 306
(0.01% of circu-
lation rate)+
Cooling water 307
blowdown (BD)
flow rate (five
cycles of con-
centration)*
660
3.7
161
737
4.1
180
966
5.4
236
Assumed cooling water inlet temperature 314K, outlet temperature 303K,
cooling water flow rate = 36,980 m^/hr
^Calculation based on reference 74
^Cycles of concentration (CC) is calculated by the following equation:
CC = (E + BD + D)/(BD + D)
In addition to the concentrations of inorganic dissolved solids shown,
scale and corrosion control additives would also be present in the blowdown/
drift. The control additives, especially chromate- and zinc-based inhibitors,
will be a consideration in the treatment of cooling tower blowdown. Treat-
ment for removal of these inhibitors will be discussed in Section 4.
135
-------
Section 3
Auxiliaries
TABLE 3-29. ESTIMATED CHARACTERISTICS OF COOLING TOWER SLOWDOWN AND DRIFT*
Slowdown/Drift Constituents
Discharge Concentration for
the Illinois No. 6 Coal Case
(mg/L except pH)
HCOs
S0§
Ca++
Na+
CT
Total Dissolved Solidst
Si02
pH
110
650
180
45
150
75
1210
33
8.0
Blowdown/drift composition estimated using raw water makeup chemistry from
Section 3.6.1. Concentrations are presented for operation at 5 cycles of
concentration.
fTotal dissolved solids is the sum of the ions except Si02
The number of cycles of cooling water concentration which can oe
achieved in the cooling tower is largely dependent upon the total dissolved
solids (particularly CaCOj and CaS04) of makeup water to the system, and a
higher makeup water quality permits a larger number of cycles. Also, in-
creasing the number of cycles of concentration has the effect of decreasing
the blowdown rate for a given system. For calculation purposes, blowdown
flow rates are based on cooling system operation at 5 cycles of concentration.
Operation at lower cycles of concentration to allow for discharge to a sur-
face water or operation at higher cycles in order to minimize wastewater
volume and treatment equipment costs may be considered on a site-specific
basis.
136
-------
Section 3
Auxiliaries
The cooling operation at an indirect liquefaction facility can be a
critical factor in the disposal and reuse of process wastewaters. The
use of treated process wastewaters as cooling tower makeup water will have
a significant effect on raw water makeup requirements and discharge stream
pollutant concentrations. The impact of potential emissions in both blow-
down and drift resulting from the reuse of treated wastewaters will be
discussed in Section 4 under air and water pollution control.
3.6.4 Oxygen Production
Oxygen required by the K-T gasification processes is assumed to be pro-
duced by standard cryogenic air separation units. Oxygen purities assumed
in the gasification process mass flow calculations are based on published
data for the subject coal, and no attempt has been made to force consis-
tency. It should be noted that the purity of oxygen utilized in the gasi-
fication process affects the quantity of the purge stream from the down-
stream synthesis process and thereby affects synthesis efficiency. There-
fore, a tradeoff exists between the energy required to produce high purity
oxygen and the efficiency of the synthesis process. An analysis of optimum
oxygen purities is beyond the scope of this manual.
In an air separation plant, air is compressed to 0.68 to 0.72 MPa and
cryogenically cooled to facilitate distillation of oxygen, nitrogen, and
noble gases (75). The oxygen stream, containing small quantities of nitro-
gen and argon, is compressed and sent to the gasifiers. Air and oxygen
compressors can either be steam, gas, electric driven, or a combination
thereof. Most of the separated nitrogen, containing small quantities of
oxygen, water, and carbon dioxide, is vented to the atmosphere. However
a portion of the nitrogen may be utilized as an inert gas for blanketing
coal storage and transfer operations or as stripping gas for solvent regen-
eration in acid gas treatment. The quantity of condensate resulting from
air compression depends upon atmospheric humidity and therefore is highly
137
-------
Section 3
Auxiliaries
variable. Condensate contains only dissolved gases and can be utilized as
a supplement to the plant's high quality water supply.
Oxygen requirements for the design K-T gasification system are 132,200
kg/hr for Illinois No. 6 coal, and may increase by approximately 10% for
lower rank coals (23).
Production of oxygen does not directly generate environmentally signifi-
cant waste streams, since chemical reactions do not take place in the air
separation process nor are any chemicals added to the process streams. A
gaseous waste stream containing mostly nitrogen and a liquid condensate
are produced, but these streams are essentially pollutant free. Reducing
the compressed air temperature prior to cryogenic cooling results in in-
creased drift and blowdown from the cooling towers because of increased
interstage compressor cooling required. Emissions indirectly associated with
the compressors are dependent upon the type of power drive (either steam,
gas, or electric).
3.6.5 Product and By-Product Storage
The expected production rates of upgraded liquid and gaseous fuels for
the K-T indirect liquefaction plant are presented in Section 3.5. For pur-
poses of estimating potential emissions from the storage of liquid products,
a 15-day capacity is assumed in all cases. Table 3-29 provides a summary of
the storage capacities, vessel types, and estimated uncontrolled mass emis-
sion rates for the various liquids. The more volatile products (e.g., LPG)
are stored in pressure vessels and have no routine evaporative emissions.
Methanol and gasoline are stored in floating roof tanks while diesel oil and
fuel oil are stored in fixed-roof tanks.
Data on the components of evaporative emissions associated with the
storage of coal derived liquid fuels are generally lacking. However, limited
data are available on evaporative emissions associated with petroleum gasoline
138
-------
TABLE 3-30. EVAPORATIVE EMISSION ESTIMATES FOR PRODUCT AND BYPRODUCT STORAGE*
Product
Methanol
Methanol
Methanol
Gasoline
Gasoline
Diesel Oil
Fuel Oil
No. of
Tanks
1
2
2
1
1
1
1
Capacity (m^)
Float Roof
(3200)
Float Roof
(45,000)
Float Roof
(46,500)
Float Roof
(22,000)
Float Roof
(39,000)
Fixed Roof
(3600)
Fixed Roof
(940)
Tank Diameter
(m)
18.2
62.5
64.0
43.6
53.3
19.5
11.6
Synthesis Case
Fischer-Tropsch
Methanol
Mobil M
Fischer-Tropsch
Mobil M
Fischer-Tropsch
Fischer-Tropsch
Vapor
Pressure
kPa
17.3
8.83
17.3
8.83
17.3
8.83
49.78
33.85
49.78
33.85
0.080
0.046
0.00059
0.00026
Uncontrolled Mass
Emission Rate*
(kg/hr)
8,740
6,630
59,790
45,360
61,250
46,470
21,790
20,430
28,950
27,140
800
500
9
5
Calculations based on information contained in AP-42 (76)
The higher values represent the month with maximum average emissions (July). The lower number
represents the average annual values.
-------
Section 3
Auxiliaries
storage as indicated in Table 3-31. In addition to the paraffins and olefins
listed in the table, aromatics are expected to be present in evaporative
emissions in the 1 to 1000 ppm range. Since the compositions of F-T and
Mobil M-gasolines are not dramatically different from those of petroleum
gasolines, the gross characteristics of evaporative emissions are expected
to be similar.
Product sulfur storage may result in H?S emissions due to the evolution
of dissolved sulfide. H~S may be liberated from the liquid sulfur either
inadvertantly during storage/handling operations or as a result of intentional
liquid sulfur degassing to produce a sulfide-free product. Intentional de-
gassing may be employed to minimize the fire and toxicity hazards potentially
associated with liquid sulfur handling. Data are not publicly available
relating to the magnitude of such potential H-S emissions or to typical con-
trol practices in the sulfur industry. In any case, such emissions are not
unique to coal gasification and hence are not further addressed in this
manual.
140
-------
TABLE 3-31. COMPOSITION OF EVAPORATIVE EMISSIONS
Section 3
Auxiliaries
FROM GASOLINE STORAGE (77)
Compound
Methane
Ethyl ene
Ethane
Propylene
Propane
Isobutane
Isobutylene
n-butane
cis-2-butane
Isopentane
n-pentane
Hexanes
Heptanes
Octanes
Vol %*
1
0
0
0
0
3
1
3
0
6
2
0
0
0
.13
.001
.15
.03
.82
.03
.12
.53
.84
.18
.89
.44
.16
.17
Balance is air
141
-------
Section 3
Fugitive Wastes
3.7 FUGITIVE AND MISCELLANEOUS WASTES
In this section gaseous and liquid emissions from process equipment,
emergency process discharges and washup/cleanup activities are discussed.
Emission estimates for these streams are based upon petroleum refining experi-
ence.
3.7.1 Fugitive Organic Emissions (Stream 241)
There are many potential sources of gaseous fugitive emissions in an
indirect coal liquefaction plant. These sources include: pumps, compressors,
in-line process valves, pressure relief devices, open-ended valves, sampling
connections, flanges, agitators, and cooling towers. Extensive tests and
measurements for fugitives have been performed at petroleum refineries (78).
As a result of this testing, average emission factors have been developed for
fugitive emission sources such as pump seals, compressor seals, valves, etc.
(79). These factors can be applied to synfuel plants because plant opera-
tions following synthesis are quite similar to those in petroleum refining.
Although the plant boundary as stated in Section 3.4.5 excludes upgrading
processes, these processes were included in estimating fugitive organic emis-
sions since these processes are the major contributors to fugitive organic
emissions in a K-T based indirect liquefaction facility.
Fugitive emissions estimates were made by estimating the number of each
type of emission source and applying the appropriate emission factor with
no adjustment for size, pressure, or flow rate. The number of pumps, com-
pressors, and process units for synfuel plants was estimated by correlating
equipment lists to the proper size synfuel plant or adjusting the equipment
counts reported in conceptual designs. Equipment spares were counted in
determining the number of pumps and compressors, because it was assumed that
spares usually contain fluid under pressure.
The process streams associated with each piece of equipment were clas-
sified with respect to percent hydrocarbon content and hydrocarbon type.
142
-------
Section 3
Fugitive Wastes
Different factors were used for liquid streams containing light and heavy
hydrocarbons. Liquid streams containing C2 through Cg hydrocarbons, naphtha,
and other aromatic hydrocarbons were classified as light. Kerosene, diesel
oil, and other heavy hydrocarbons were classified as heavy. All emission
factors shown assume 100% hydrocarbon content, so all emission factors except
those for compressors were multiplied by the actual hydrocarbon content for
each process stream. Streams containing less than 10% hydrocarbons were
neglected and those containing greater than 80% were considered to contain
100%. Gaseous streams were classified as either hydrocarbon or hydrogen
depending on which was present in a greater percentage. The compressor seal
emission factors for these two classifications were used as reported and not
adjusted for percent hydrocarbon content.
Results of these fugitive emission calculations for K-T based indirect
liquefaction facilities are given in Table 3-32. The data indicate that
methanol plants are expected to have considerably lower fugitive organic
emissions than Fischer-Tropsch plants. In-line valves are the single biggest
contributor to total fugitive organic emissions for all facilities.
3.7.2 Non-Process/Intermittent Wastewater Streams
Fugitive process fluid leaks from sources such as pump seals, valves,
and flanges will generate a "composite" waste stream with a highly variable
flow and composition. In addition, drainage resulting from emergency process
fluid discharges or process area washdown/cleanup activities will contribute
additional intermittent aqueous wastes. All of these wastes would normally
be collected in a process or oily waste sewer and routed to a common treat-
ment system. The estimated flow rates of the combined wastewaters from these
types of sources in the subject plants are summarized in Table 3-33.
These process drainage calculations are based upon refinery experience
and are estimated as 2% of the total raw water makeup to the plant (80).
Since the process drainage flow rate is based on the total plant raw water
143
-------
TABLE 3-32. ESTIMATED TOTAL FUGITIVE ORGANIC EMISSIONS
Pump Seals*
Light Liquid Service
Heavy Liquid Service
In-Line Valves
Gas Service
Light Liquid Service
Heavy Liquid Service
Safety Relief Valves
Vapor Service
Compressor Seals*
Hydrocarbon
Hydrogen
Flanges
Drains
Totals
Uncontrolled
Emission
Factor
(kg/hr)
0.154
0.029
0.027
0.011
0.00023
0.086
1.28
0.10
0.00025
0.07
Uncontrol
Fischer-
Tropsch
16.74
0.52
34.72
37.02
0.14
24.77
12.8
1.8
1.09
10.94
140.5
led Emission Rates
(kg/hr)
Methano'l
3.17
0.23
7.78
7.09
0.06
5.16
2.88
--
1.18
2.45
30.0
Mobil M
7.09
0.38
15.39
15.68
0.09
11.35
1.54
--
2.42
4.90
58.8
Uncontrolled emission factors for pumps
emissions from each pump and compressor
compressor seal.
and compressors represent
and not from each pump seal and
144
-------
Section 3
Fugitive Wastes
makeup rate, it will vary with coal type and synthesis process as discussed
in Section 3.6.1.
TABLE 3-33. DRAINAGE ESTIMATE FROM NON-PROCESS/INTERMITTENT STREAMS
Drainage Flow Rate
Stream m3/hr
Process drain effluent (Stream 315)* 32 - 42
Storm drain effluent (Stream 314) 47
—.- _______________________
Flow rate based on 2% of total raw water makeup to the plant
Both the flow rate and composition of these wastewaters will vary widely
among the different plants. Of course, good housekeeping and mainten-
ance practice will minimize these flows. These wastewaters will contain oil
and grease, dissolved organics, dissolved inorganics, and particulate matter
in widely varying concentrations. Because the characteristics of these waste-
waters will be site-specific and highly variable with time, no composition
estimates were developed for these streams. The treatment of these streams
and their impact on facility costs and energy usage will be discussed in
Section 4.
Storm runoff water flow rates will be a function of the surface drain-
age area of the plant site and the annual rainfall. For calculation purposes,
a plant site drainage area of 100 surface acres was used with an annual rain-
fall rate of 1 m/yr (81). The average storm runoff flow was estimated to be
47 m3/hr (refer to Table 3-33).
The composition of storm runoff from the area will vary with the fre-
quency of a rainfall occurrence and with time during a rainfall occurrence.
The major contaminants potentially requiring control are total suspended
145
-------
Section 3
Fugitive Wastes
solids, oil, and grease. Because of the variable, site-specific nature of
the composition of this stream, no attempt has been made to develop a de-
tailed composition.
3.7.3 Equipment Cleaning Wastes (Streams 242 and 305)
The two primary sources of equipment cleaning wastes at an indirect
liquefaction facility are process equipment (Stream 242) and boiler cleaning
wastes (Stream 305). Process equipment cleaning wastes will result from
periodic cleanup or maintenance of equipment such as heat exchangers, pumps,
and pressure vessels. The volume of cleaning waste generated will be deter-
mined by the vessel volumes, frequency of cleanup, cleaning agent used, and
rinsing requirements. Cleaning waste discharges are intermittent, short
duration, high flow rate occurrences.
Boiler cleaning wastes will be generated on a one to four year cycle
depending on plant maintenance practices. The large volume of the boiler
can result in cleaning waste dumps in excess of 3800 m over short periods
of time. Boiler cleaning wastes will probably be the largest single source
of cleaning wastes at an indirect liquefaction facility.
The composition of equipment cleaning wastes will vary with the clean-
ing agent used and the material being removed. Cleaning of process equipment
generally includes the removal of oils, sludges, and waxy materials using
alkaline solvents. Boiler cleaning is undertaken to remove inorganic (scal-
ing) materials and metal corrosion products with acidic and alkaline clean-
ing agents. Rinse volumes from both process and boiler cleaning wastes con-
tain lower contaminant concentrations than the cleaning wastes themselves
but can amount to 2 to 5 times the volume of the cleaning waste. Treatment
of these wastes is difficult because of their complex composition. Despite
the intermittent and large volumes of waste generated, when considered on an
annual average basis, cleaning wastes are produced at relatively low flow
rates compared with other wastewaters generated in an indirect liquefaction
facility.
146
-------
Section 3
Stream Index
3.8 WASTE/CONTROL TECHNOLOGY INDEX
The preceding parts of this section have provided a general description
of K-T based synfuels facilities and test data and engineering estimates
characterizing the uncontrolled or primary waste streams expected. Section 4
of this manual presents information on the available control techniques for
these primary waste streams and illustrative examples of both individual con-
trol technologies and integrated systems of control technologies applied to
specific streams. As will be discussed in Section 4, residuals or secondary
waste streams are generated as a result of the application of some control
technologies; control of secondary waste streams is discussed in conjunction
with the illustrative examples. A summary of the process streams and primary
and secondary waste streams discussed in this manual is presented in Table
3-34.
To aid users in finding characterization data and control technology
information for any waste stream addressed in this manual, a cross reference
index was developed. This index is presented in Table 3-35 and indicates
where characterization data can be found in Section 3 and where control tech-
nology information can be found in Section 4 for each primary waste stream.
The waste streams in Table 3-35 are grouped by the operation or auxiliary
process from which they originate and then further grouped within each opera-
tion by waste medium. Similar types of information on secondary waste
streams are presented in Table 3-36. The entries in Table 3-36 are not meant
to imply that those streams will necessarily be found in K-T based facilities,
but that if the control techniques listed are used, then those streams will
be produced.
147
-------
TABLE 3-34. STREAM INDEX
Main Process Streams (Numbers 1-99)
Coal Preparation
1. Run of mine coal
2. Prepared coal to gasifier
Gasification
3. Gasifier steam
4. Oxygen
5. Quench ring water to gasifier
6. Raw quenched gas from waste heat boiler
Gas Purification and Upgrading
7. Raw gas from washer cooler
8. Raw gas after NOV reduction
/\
9. Raw compressed gas
10. Compressed gas after cyanide wash
11. Desulfurized gas to shift conversion
12. Desulfurized shift conversion bypass gas
13. Shift gas to C0~ absorber
14. Synthesis gas
15. Nitrogen strip gas
Product Synthesis
20. Crude methanol
21. Crude Mobil M-gasoline products
22. Crude Fischer-Tropsch synthesis products
23. Methanation product gas
24. C02-free SNG
25. Purge gas to methanation
(Continued)
148
-------
TABLE 3-34. (Continued)
Auxiliaries
30. Prepared coal to boiler
31. Plant raw water
32. Cooling tower makeup water
33. Boiler feed water
Product and By-Product Streams (Numbers 100-199)
Fischer-Tropsch Synthesis
100. Gasoline
101. Diesel oil
102. Heavy oil
103. Methanol
104. Ketones
105. Heavy alcohols
106. LPG
107. SNG
Methanol Synthesis
108. Fuel grade methanol
Mobil M-gasoline Synthesis
109. Gasoline
110. Mixed butanes
111. Propane
By-Products
112. Sulfur
Discharge Streams from the Main Process Train (Numbers 200-299)
Coal Preparation
200. Fugitive dust from raw coal storage
(Continued)
149
-------
TABLE 3-34. (Continued)
Coal Preparation (continued)
201. Raw coal storage runoff
202. Fugitive dust from coal screening and crushing
203. Fugitive dust from screened/crushed coal transfer
204. Fugitive dust from coal pulverizing
205. Fugitive dust from pulverized coal storage and feeding
206. Emissions from thermal dryers
Gasification
207. Quenched slag
208. Transient waste gases
Gas Purification and Upgrading
209. Dewatered dust
210. Cooling and dust removal blowdown
211. Raw gas compression and cooling condensate
212. Spent NO reduction catalyst
A
213. Aqueous still bottoms from methanol based cyanide wash
214. Sour flash gas from cyanide wash
215. Cyanide wash water
216. H2S-rich offgas
217. Spent shift catalyst
218. Shift condensate blowdown
219. C02-rich offgas
220. Rectisol condensate/still bottoms
221. Spent sulfur guard
Product Synthesis
222. Spent F-T catalyst
223. F-T wastewater
224. Fractionator light gas
225. Methanol synthesis expansion gas
(Continued)
150
-------
TABLE 3-34. (Continued)
Gas Purification and Upgrading (continued)
226. Methanol synthesis purge gas
227. Spent methanol synthesis catalyst
228. Methanol distillation offgas
229. Methanol distillation wastewater
230. Fuel gas from Mobil synthesis
231. Mobil synthesis catalyst regeneration/decommissioning offgas
232. Spent Mobil catalyst
233. Mobil synthesis condensate
234. Mobil fractionator offgas
235. C02 removal condensate
236. Methanation condensate
237. Methanation catalyst regeneration/decommissioning offgas
238. Spent methanation catalyst
239. COp offgas from SNG purification
240. Dehydration offgas
Miscellaneous Waste Streams
241. Fugitive organic emissions
242. Equipment cleaning wastes
Discharge Streams from Auxiliary Processes (Numbers 300-399)
Raw Water Treatment
300. Raw water treatment sludges
301. Demineralizer regeneration wastewater
Power Generation
302. Flue gases from power generation
303. Boiler blowdown
304. Boiler bottom ash
305. Boiler cleaning wastes
(Continued)
151
-------
TABLE 3-34. (Continued)
Cooling Tower
306. Cooling tower evaporation/drift
307. Cooling tower blowdown
Product/By-Product Storage
308. Evaporative emissions from methanol storage
309. Evaporative emissions from gasoline storage
310. Evaporative emissions from diesel oil storage
311. Evaporative emissions from heavy oil storage
312. Evaporative emissions from ketones storage
313. Evaporative emissions from heavy alcohols storage
Miscellaneous
314. Storm drain effluent
315. Plant process drains effluent
Discharge Streams from Pollution Control Processes (Numbers 400-499)
Coal Preparation
400. Coal particulates from coal preparation
Gasification
No secondary waste streams are associated with this process area.
Gas Purification and Upgrading and Product Synthesis*
401. Catalyst regeneration offgas from Claus process
402. Claus spent catalyst
403. Claus sulfur
404. Stretford oxidizer vent gas
405. Sour condensate from Beavon/Stretford process
(lontTnueBl
*Secondary waste stream numbers 401 to 414 result from control processes
treating waste streams from gas purification and upgrading operations only,
Secondary waste stream numbers 415 to 422 result from control processes
treating combined waste streams from gas purification and upgrading, and
product synthesis
152
-------
TABLE 3-34. (Continued)
Gas Purification and Upgrading and Product Synthesis (continued)
406. Stretford solution purge
407. Beavon/Stretford spent catalyst
408. Beavon/Stretford sulfur
409. SCOT sour condensate
410. SCOT spent catalyst
411. Wellman-Lord acidic wastewater
412. Wellman-Lord thiosulfate/sulfate purge
413. Flue gas from fluidized bed boiler
414. Spent bed media from fluidized bed boiler
415. Activated sludge solid waste
416. Sulfide/cyanide offgas
417. Filtration backwash
418. Denitrification waste sludge
419. Evaporation/drift from cooling tower concentration
420. Regeneration offgas from granular activated carbon
421. Offgas from liquid waste incinerators
422. Offgas from solid waste incinerators
Auxiliaries
423. Boiler fly ash from ESPs/fabric filters
424. FGD sludge from lime/limestone scrubbing
425. Thiosulfate/sulfate purge from Wellman-Lord FGD process
426. Sulfur from Wellman-Lord FGD process
153
-------
TABLE 3-35. CROSS-REFERENCE INDEX FOR PRIMARY WASTE STREAMS
Stream Identification
PCTM Section Reference
Waste Characterization Control Techniques
en
Coal Preparation
Gaseous Waste Streams
Fugitive dust emissions from raw coal
storage piles (Stream 200)
Crushing/screening/transfer/pulverizing
dust (Streams 202, 203, 204)
Particulate emissions from prepared coal
storage and feeding (Stream 205)
Emissions from thermal dryers (Stream 206)
Aqueous Waste Streams
Raw coal storage pile runoff (Stream 201)
Coal Gasification
Gaseous waste Streams
Transient waste gases (Stream 208)
Solid Waste Streams
Quenched slag (Stream 207)
Section 3.1
Section 3.1
Section 3.1
Section 3.1
Section 3.1
Section 3.2
Section 3.2
Section 4.1.4
Section 4.1.6
Section 4.1.6
Section 4.1.2.1
Section 4.2.1
Section 3.2
Section 4.3.2.1
(Continued)
-------
TABLE 3-35. (Continued)
Stream Identification
PCTM Section Reference
Waste Characterization
Control Techniques
en
en
Gas Purification and Upgrading
Gaseous Waste Streams
Sour flash gas from cyanide wash - water
wash case (Stream 214a)
Sour flash gas from cyanide wash - methanol
wash case (Stream 214b)
H2S-rich offgas (Stream 216)
C02-rich offgas (Stream 219)
Aqueous Waste Streams
Cooling and dust removal blowdown
(Stream 210)
Raw gas compression and cooling
condensate (Stream 211)
Cyanide wash water (Stream 215)
Cyanide wash still bottoms - methanol
wash case (Stream 213)
Shift condensate blowdown (Stream 218)
Rectisol condensate/still bottoms
(Stream 220)
Section 3.3.4
Section 3.3.4
Section 3.3.6
Section 3.3.6
Section 3.3.1
Section 3.3.3
Section 3.3.4
Section 3.3.4
Section 3.3.5
Section 3.3.6
Sections 4.1.1.2
and 4.1.1.5
Sections 4.1.1.2
and 4.1.1.5
Sections 4.1.1.1
and 4.1.1.5
Section 4.1.1.3
Section 4.2.3.3
Section 4.2.3.2
Section 4.2.3.1
Section 4.2.1
Section 4.2.1
Section 4.2.1
(Continued)
-------
TABLE 3-35. (Continued)
Stream Identification
PCTM Section Reference
Waste Characterization
Control Techniques
en
01
Solid Waste Streams
Dewatered dust (Stream 209)
Spent NOV reduction catalyst (Stream 212)
/\
Spent shift conversion catalyst
(Stream 217)
Spent sulfur guard (Stream 221)
Product Synthesis
Gaseous Waste Streams
Mobil synthesis catalyst regeneration
offgas (Stream 231)
Methanation catalyst decommissioning
offgas (Stream 237)
C09 offgas from SNG purification
(Stream 239)
Dehydration offgas (Stream 240)
Liquid Waste Streams
Methanol distillation wastewater
(Stream 229)
F-T wastewater (Stream 223)
Section 3.3.1
Section 3.3.2
Section 3.3.5
Section 3.3.7
Section 3.4.6
Section 3.4.6
Section 3.4.6
Section 3.4.6
Section 3.4.6
Section 3.4.6
Section 4.3.2.2
Section 4.3.5
Section 4.3.5
Section 4.3.5
Section 4.1.3.2
Section 4.1.3.2
Section 4.1.3.1
Not evaluated
Section 4.2.2.3
Section 4.2.2.2
(Continued)
-------
TABLE 3-35. (Continued)
Stream Identification
Waste Characterization
PCTM Section Reference
Control Technique?
en
Mobil synthesis condensate (Stream 233)
Methanation condensate (Stream 236)
Solid Waste Streams
Spent methanol synthesis catalyst
(Stream 227)
Spent F-T catalyst (Stream 222)
Spent Mobil catalyst (Stream 232)
Spent methanation catalyst (Stream 238)
Auxiliaries
Gaseous Waste Streams
Boiler flue gases (Stream 302)
Cooling tower evaporative losses
(Stream 306)
Evaporative emissions from product and
by-product storage (Streams 308 through 313)
Fugitive organic emissions (Stream 241)
Aqueous Waste Streams
Demineralizer regeneration wastewaters
(Stream 301)
Section 3.4.6
Section 3.4.6
Section 3.4.6
Section 3.4.6
Section 3.4.6
Section 3.4.6
Section 3.6.2
Section 3.6.3
Section 3.6.5
Section 3.7.1
Section 3.6.1
Section 4.2.2.1
Section 4.2.1
Section 4.3.5
Section 4.3.5
Section 4.3.5
Section 4.3.5
Section 4.1.2.1
Not evaluated
Section 4.1.5.1
Section 4.1.5.2
Section 4.2.1
(Continued)
-------
00
TABLE 3-35. (Continued)
PCTM Section Reference
Stream Identification _ Waste Characterization Control Techniques
Boiler blowdown (Stream 303) Section 3.6.2 Section 4.2.1
Cooling tower blowdown (Stream 307) Section 3.6.3 Section 4.2.1
Process drain effluent (Stream 315) Section 3.7.2 Section 4.2.1
Storm drain effluent (Stream 314) Section 3.7.2 Section 4.2.1
Equipment cleaning wastes (Streams 242 Section 3.7.3 Section 4.2.1
and 305)
Solid Waste Streams
Raw water treatment sludges (Stream 300) Section 3.6.1 Section 4.3.2.8
Boiler bottom ash (Stream 304) Section 3.6.2 Section 4.3.2.3
-------
TABLE 3-36. CROSS-REFERENCE INDEX FOR SECONDARY WASTE STREAMS
Control Technology/Secondary Waste
Waste
Characterization
PCTM Section Reference
Control Technology
Control
Technique
Appendix^
Air Pollution Control
Claus Process
Catalyst regeneration offgas (Stream 401)
Spent catalyst (Stream 402)
Sulfur (Stream 403)
Beavon/Stretford Process
Stretford oxidizer vent gas (Stream 404)
Sour condensate (Stream 405)
Stretford solution purge (Stream 406)
Spent catalyst (Stream 407)
Sulfur (Stream 408)
Shell Claus Offgas Treatment (SCOT) Process
Sour condensate (Stream 409)
Sections 4.1.1
and 4.1.1.1
Sections 4.1.1
and 4.1.1.1
Sections 4.1.1
and 4.1.1.1
Sections 4.1.1
and 4.1.1.1
Sections 4.1.1
and 4.1.1.1
Section 4.1.1
and 4.1.1.1
Not evaluated
Section 4.3.5
Section 4.3.3.1
Not evaluated
Section 4.2.3.4
Section 4.2.3.4
Section 4.3.5
Section 4.3.3.1
Section 4.2.3.4
A-6
A-9
A-8
(Continued)
-------
TABLE 3-36. (Continued)
PCTM Section Reference
Control Technology/Secondary Waste
Waste Control Control Technology
Characterization Technique Appendix*
CTl
o
Spent catalyst (Stream 410)
Wellman-Lord Process
Acidic wastewater (Stream 411)
Thiosulfate/sulfate purge (Stream 412)
Electrostatic Precipitators
Boiler fly ash (Stream 423)
Limestone Scrubbing
FGD sludge (Stream 424)
Wellman-Lord FGD Process
Thiosulfate/sulfate purge (Stream 425)
Sulfur (Stream 426)
Coal Particulate^ Dry Collectors^
Coal particulate (Stream 400)
Sections 4.1.1
and 4.1.1.1
Sections 4.1.1
and 4.1.1.1
Sections 4.1.1
and 4.1.1.1
Section 3.6.2
Section 4.1.2
Section 4.1.2
Section 4.3.5
Section 4.2.3.4
Section 4.2.3.4
Section 3.1
Section 4.3.2.4
Section 4.3.2.5
Section 4.2.3.4
Section 4.3.3.1
Section 4.3.3.2
A-10
A-13
A-20
A-10
A-ll.A-12
(Continued)
-------
TABLE 3-36. (Continued)
PCTM Section Reference
Control Technology/Secondary Waste
Waste Control Control Technology
Characterization Technique Appendix*
cr>
Water Pollution Control
Activated Sludge
Sulfide/cyanide offgas (Stream 416)
Waste sludge (Stream 415)
Filtration
Backwash (Stream 417)
Biological Denitrification
Waste sludge (Stream 418)
Cooling Tower Concentration
Evaporation/drift (Stream 419)
Granular Activated Carbon
Regeneration offgas (Stream 420)
Incineration^
Flue gas (Stream 421)
Section 4.2.4
Sections 4.2.2.1,
4.2.3.1, 4.2.3.2
Section 4.1.1.4
Section 4.3.4
Section 4.2.2.1 Section 4.2.2.1
Sections 4.2.3.1 Section 4.3.4
and 4.2.3.2
Sections 4.2.2.1,
4.2.3.1, 4.2.3.2
Not evaluated
Section 4.2.2.1 Section 4.2.2.1
Section 4.2.2.1 Section 4.2.2.1
B-10,B-23
B-4
B-23
B-18
B-15
B-17
(Continued)
-------
TABLE 3-36. (Continued)
PCTM Section Reference
Control Technology/Secondary Haste
Waste
Characterization
Control
Technique
Solid Waste Management
Fluidized Bed Combustion Boiler
Boiler flue gas (Stream 413)
Spent bed media (Stream 414)
Incineration
Flue gas (Stream 422)
Section 4.3.2.2
Section 4.3.2.6
Section 4.3.4.1
Section 4.1.2.3
Section 4.3.2.6
Section 4.3.4.1
o>
ro
The Control Technology Appendices for the PCTMs are compiled in a separate volume.
-------
Section 4
Pollution Control
SECTION 4
EVALUATION OF POLLUTION CONTROL TECHNOLOGY
At the present time, no K-T based indirect liquefaction plants are opera-
ting in the United States, although the K-T process is widely used in other
countries for the production of hydrogen (primarily for ammonia) and fuel gas.
The emphasis on pollution control which has been incorporated into designs
for facilities abroad is generally less than that which is anticipated for
U.S. facilities. Thus, directly applicable performance data for most pollu-
tion control technologies are quite limited. The potential applicability of
most pollution control technologies to waste streams identified in Section 3
has, therefore, been inferred from their use in similar applications in indus-
tries such as petroleum refining, coke production, natural gas processing,
coal cleaning, and electrical utilities. This section provides an evaluation
of the control methods which may be adapted from other industries and from
general pollution control practice, and definition of the principal limita-
tions of these controls methods in K-T based synfuels plants. Control alter-
natives evaluated include process modifications (relative to existing designs)
in addition to add-on controls.
Approach
In this section a wide variety of potentially applicable pollution con-
trol technologies are discussed. In addition, illustrative examples of the
application of individual control technologies to specific waste streams as
well as the application of integrated systems of control technologies to
specific waste streams are presented for each waste medium (e.g., gaseous,
aqueous, and solid waste).
Descriptions of the pollution control technologies presented in this
section are based upon more detailed descriptions provided in the Control
Technology Appendices. Performance data for control technologies have been
163
-------
Section 4
Pollution Control
obtained primarily from the open literature and have been supplemented by
vendor-supplied data in some cases. The capabilities of various controls
have not usually been assessed on a design-specific basis, but rather upon
a generalized basis derived from test results and/or engineering studies of
the subject technologies. Example performance levels used for evaluation
purposes encompass most of the published data. Therefore, only limited data
referencing is provided in Section 4; detailed references are available in
the Control Technology Appendices.
In many cases, performance can only be estimated in terms of control of
major constituents (e.g., hydrogen sulfide) or gross parameters (e.g.,
COD) since often no information is available for removal efficiencies of
specific substances. Further, even in those cases where substance-specific
performance information exists for a control technology, accurate or complete
characterization of the waste stream requiring control may be lacking. In
the final analysis, the capabilities of controls can only be accurately eval-
uated by testing at operating facilities or at smaller units from which data
can be confidently extrapolated to commercial size. The performance estimates
in this document are believed to reflect the best information publicly avail-
able based on actual experience and on engineering analysis.
In providing example applications of pollution control technologies,
waste streams unique to K-T based synfuels facilities and large volume/high
loading waste streams have been emphasized. The source and characteristics
of these waste streams have been detailed in Section 3, and those characteris-
tics of principal significance with regard to the application of pollution
control technologies are reiterated in this section. It should be noted,
however, that Section 3 does not reflect the design of a specific facility,
but incorporates key features of a number of existing and proposed facilities.
Some of the waste streams identified in Section 3 may not be found in all
facilities. Further, in a specific facility, some streams encountered may
differ significantly in size and characteristics from analogous streams
164
-------
Section 4
Pollution Control
discussed in this section, and controls other than those cited in the examples
may be more appropriate. For these reasons the reader is encouraged to con-
sider design-specific waste stream characterization data whenever they are
available and to use the detailed Control Technology Appendices for estimat-
ing the applicability and performance of specific controls to waste streams.
The control examples in this section emphasize the Illinois No. 6 coal and the
the base plant defined in Section 3. Effects of coal characteristics and
design modifications are discussed in those cases where either may signifi-
cantly influence the control performance or cost.
Reliability of pollution control processes in U.S. coal conversion
facilities will depend largely upon the time required to determine the
optimum operating conditions and the emphasis placed upon operation and
maintenance. In essentially any industry, the introduction of a new process
or modification of an older process for application to new streams meets with
unexpected problems relating to both design features and operating practices.
Thus, some shakedown period must be expected where process performance,
efficiency, and on-stream time will improve. Once a process is properly char-
acterized for a specified application, reported reliability in terms of on-
stream time and performance levels may still vary among facilities. Some-
times the specific design is a factor, but reliability also reflects the
emphasis placed upon operation and maintenance. A properly trained operating
crew and a regular maintenance regimen can significantly improve the reli-
ability of well designed pollution control equipment.
Because most of the potentially applicable pollution control technologies
have not been employed in synthetic fuels facilities, few directly related
reliability data are available. Further, the overall characteristics and
variability of waste streams in coal conversion facilities are often signif-
icantly different from those encountered in other industries. As a result,
reliability data accumulated in other industries may not be directly
165
-------
Section 4
Pollution Control
applicable to coal conversion processes. Published reliability data for con-
trol technologies in other industrial applications are summarized in the Con-
trol Technology Appendices. However, in this section, reliability is dis-
cussed primarily in terms of those factors which have a major influence upon
process operation and stability.
Organization
The pollution control technology evaluations are presented according
to the medium to which the technologies apply. Technologies applying to
gaseous, aqueous, and solid waste media are discussed in Sections 4.1, 4.2,
and 4.3, respectively. Included in each of these sections are (1) a summary
of waste stream characteristics which are significant with respect to the
application of pollution controls (detailed characterization estimates are
presented in Section 3), (2) a brief description of the performance and
costs of potentially applicable pollution control technologies, (3) examples
of the performance and cost of individual pollution control technologies
applied to specific waste streams, and (4) examples of the performance and
cost of integrated pollution control systems.
Gaseous waste streams may be categorized according to the principal
pollutants which are present and, in general, different controls or groups
of controls are applicable to each category. Therefore, the technology
descriptions and control examples presented in Section 4.1 (Gaseous Medium)
are by waste stream categories or source types to which they apply.
Source type categorizations may also be made for aqueous and solid
wastes. Waste streams in these media, however, often lend themselves to
treatment by control technologies which may be applicable to several indivi-
dual source types or to combinations of source types. Therefore, in Sections
4.2 and 4.3 (Aqueous Medium and Solid Waste Management, respectively) poten-
tially applicable pollution control technologies are discussed at the begin-
ning of the section prior to the presentation of control examples.
166
-------
Section 4
Pollution Control
Costing Methodology
Capital and operating cost estimates have been developed for the control
processes discussed in this section. These cost estimates are based pri-
marily on estimates contained in non-proprietary published literature. The
estimates are provided to give the reader an indication of the costs of con-
trols that are applicable to K-T based synthetic fuels plants. It was beyond
the scope of this manual to develop detailed engineering designs necessary
for highly accurate cost estimation.
There are three general factors that lead to uncertainties in the cost
estimates provided. These are related to the assumptions used to develop
material and energy balances, the level of accuracy of the published cost
data used, and the general methodology used to apply the acquired cost data
to the control processes addressed in this manual.
Material and energy balances were derived mainly from commercial synfuels
tests, from data from analog industries, and from results of engineering cal-
culations, as described in Section 3. The level of accuracy in specifying
the flow rates and quality of input streams to controls will affect the
accuracy of the resulting cost estimates.
Sources of cost data used in this manual are published costs for pro-
cesses applied to similar streams in related industries, costs from published
detailed design studies, and vendor quotes. The accuracy of cost data taken
from published sources is influenced by the details of the design upon which
the cost was based, the cost methodology, and the degree of similarity of
the streams. Also, the accuracy of the estimates and the components included
in the cost estimates (e.g., contingency reserves and working capital), are
not always provided in the reference. Thus, extrapolation of these costs to
the stream being treated in this manual will also introduce uncertainties.
167
-------
Section 4
Pollution Control
The costing methodology used in this manual (see Appendix A for details)
also introduces some uncertainties. Other estimators may have used differ-
ent factors or weighted them differently. In addition, available cost esti-
mates were adjusted to a 1980 basis using the Chemical Engineering plant
cost annual index. It is also possible that recent advances in the state-
of-the-art are not reflected in some of the resulting cost estimates.
As a result of the above influences, the accuracy of the cost informa-
tion presented will vary from control to control. However, the cost infor-
mation presented is believed to be adequate for the use intended.
Capital costs presented are total depreciable investment costs. Included
in the total depreciable investment costs are (1) installed equipment costs,
(2) indirect installation charges (including construction and engineering
costs, contractor fees, and contingency), and (3) interest during construc-
tion. Total annualized costs presented include (1) labor and maintenance,
(2) raw materials, utilities, chemicals, and catalysts, (3) overhead charges,
and (4) capital related charges (including interest on working capital, taxes,
insurance, and capital recovery). The same methodology was used to calculate
capital and annualized costs for both the base plant and pollution controls.
Details of that methodology and other pertinent assumptions and bases which
were used to develop the cost estimates are presented in Appendix A.
168
-------
Section 4
Gaseous Medium
4.1 GASEOUS MEDIUM
Gaseous waste streams, or uncontrolled gaseous emissions, originating
from the main process train and from non-pollution control auxiliary processes
in a K-T based gasification facility, were identified in Section 3. Char-
acteristics of these streams and additional gaseous waste streams generated
by pollution control processes are summarized in Table 4-1. In terms of
volume and pollutant loading, the most important streams in the subject
facilities are Rectisol acid gases, flash gases from cyanide washing, and
combustion flue gases. Pollutants of primary concern in these streams are
reduced sulfur species, S02, HCN, CO, and particulates. These pollutants
may also be present in intermittent transient waste gases generated during
startup, shutdown, and transient operations. Smaller volume waste gases sucn
as regeneration/decommissioning offgases, COp-rich offgases from SNG purifica-
tion, condensate depressurization offgases, fugitive organic emissions, and
evaporative emissions may contain non-methane organics (or VOC-volatile
organic compounds), CO, or particulate emissions. An additional source of
particulate emissions is the coal preparation operation.
The waste streams in Table 4-1 may be regrouped into two broad catego-
ries, those which are unique to gasification or synthesis operations and those
which are not unique. Non-unique streams are associated with auxiliary
operations within an integrated facility. In Section 3, gaseous wastes were
identified and characterized from the standpoint of their origin in an inte-
grated plant. Table 4-2 is a regrouping of these waste streams according to
the major types of potential pollutants which they contain. Streams unique
to K-T facilities are primarily those containing reduced sulfur, non-methane
organics, and smaller amounts of HCN and NH3> These streams are: (1) Rectisol
acid gases, (2) flash gases from the cyanide wash, and (3) gasifier transient
waste gases. Another unique stream which contains VOC, CO, and/or particulates
is the Mobil M-catalyst regeneration offgas. Non-unique streams common to
169
-------
TABLE 4-1. SUMMARY OF ESTIMATED GASEOUS WASTE STREAM CHARACTERISTICS IN K-T BASED INDIRECT LIQUEFACTION
FACILITIES
Stream Name
Constituent Concentration (volume percent, dry basis) Flow Rate (kraol/hr dry basis
NfT HCN fjfjParticulate unless otherwise specified)*
VOC
CO
Streams from Main
Process Train
Dust from coal preparation
(streams 200, 202 to 206)
Transient waste gases
(stream 208)
Sour gas from cyanide wash
flash - water wash case
(stream 214a)
Sour gas from cyanide wash
flash - methanol wash case
(stream 214b)
HjS-nch Rectisol offgas
(stream 216)
C02-rich Rectisol offgas
(stream 219)
Streams from synthesis
processes
Mobil synthesis catalyst
regeneration offgas
(stream 231)
Methanation catalyst
decommissioning offgas
(stream 237)
002-nch offgas from SNG
purification (stream 239)
Fugitive organic emissions
(stream 241)
Streams from auxiliary
processes
Flue gases from power
Evaporative emissions from
product storage (stream 308
to 313)
Streams from pollution
control processes'
Fugitive emissions from waste-
water treatment (stream 416)
Flue gas from K-T dust incin-
eration (stream 413)
3.1
0.0005 0.0008
Present
-62 -0.02 -0.01
0 05
3 -- 2 26
5 0.02 1
Present 1
15
0.06
Present '
Present
0.2
Present
0.2 0.01 0.03
Present
Present 100-2200 kg/hr (maximum rate)
70-1400 kg/hr (average rate)
Present Unknown but small on an
average basis
22
16
535
10,064
Present 200 maximum, 100 average
Present Unknown but very infrequent
273
30-140 kg/hr
0.04 10 g/m
Present Present
02 10 g/m
5400-32711
2.5-7.5 kg/hr average
2.8-9.3 kg/hr max.
Unknown
7419
For a plant with an input to the gasifier of 278 Mg/hr dry Illinois No. 6 coal.
-------
TABLE 4-2. CATEGORIZATION OF GASEOUS WASTE STREAMS ACCORDING TO SOURCE TYPE IN K-T INDIRECT
LIQUEFACTION FACILITIES
Source Type
Stream Name and Origin
Factors Affecting Flow Rate
and Pollutant Loading
Acid gases containing
reduced sulfur/nitrogen,
organics and CO
Combustion gases
Organic and CO
containing gases
Fugitive dust
Fugitive VOC
Fugitive particulate
Rectisol acid gases
(stream 216)
Flash gases from cyanide
wash (stream 214)
Boiler flue gases
(stream 302)
Process heater flue
gases
Flue gas from K-T dust
incineration (stream 413)
Catalyst regeneration/
decommissioning offgases
(stream 237)
C02-rich offgas from SNG
purification (stream 239)
Dust from coal storage
(stream 200)
Product storage evapora-
tion emissions (streams
308 to 313)
Process equipment
fugitives (stream 241)
Particulate from coal
handling and preparation
(streams 202 to 206)
Coal sulfur; coal rank; acid gas removal selec-
tivity
Coal sulfur; cyanide wash solvent; process pres-
sure; coal rank may influence HCN formation
Overall plant thermal efficiency; coal ash and
sulfur; boiler design and efficiency
Tend to be design/synthesis specific; not related
to coal parameters
Coal rank; coal ash and sulfur; combustor design
and efficiency; dust feed rate
Tends to be design specific; not related to coal
parameters
Tends to be design specific; not related to coal
parameters
Coal physical properties (particle size, moisture,
density, etc.); site-specific climatological
factors
Synthesis process; product slate
Synthesis process; upgrading steps employed;
plant design
Coal physical properties; coal preparation opera-
tions and layout
-------
Section 4
Gaseous Medium
all K-T plants would include coal combustion flue gases, coal preparation
dust, product storage emissions, and process VOC emissions.
In each of the subsections that follow, controls which may be applicable
to the above gaseous wastes are identified. For those technologies for which
data are available, the expected performance and costs are provided. Techno-
logy descriptions are provided at the beginning of each subsection as follows:
• Section 4.1.1 provides descriptions of bulk sulfur removal,
tailgas treatment, and incineration technologies.
• Section 4.1.2 provides descriptions of N0x removal,
particulate removal, and S02 removal technologies.
• Section 4.1.3 refers to the Section 4.1.1 incineration technology
description.
• Section 4.1.4 provides descriptions of technologies for suppression
of fugitive dust from coal storage piles.
• Section 4.1.5 provides descriptions of fugitive VOC control
technologies.
• Section 4.1.6 provides descriptions of technologies for suppression,
capture, and collection of fugitive particulates from material
conveying and processing.
In many cases control of gaseous pollutants involves both inherent process
design features and tradeoffs among processes. Further, some waste gases are
combined for treatment rather than handled separately. Accordingly, example
approaches to the control of pollutants in integrated facilities are pro-
vided. Sufficient detail is included in these examples such that overall
emissions reductions and costs can be seen. In the discussions below,
emphasis is on the unique streams rather than on those streams for which in-
formation about control may be found in other documents.
-------
Section 4
Acid Gas - Red. S/N, Org., CO
4.1.1 Acid Gases Containing Reduced Sulfur/Nitrogen, Organics, and/or Carbon
Monoxide
As discussed in Section 3, K-T based coal gasification plants will
employ the Rectisol process for removal of sulfur compounds, C02, and HCN
from the washed product gas. Regeneration of the methanol solvent produces
waste gases enriched with these acid gases. Since CO and hL are soluble in
cold methanol, Rectisol acid gases will also contain these constituents as
well as traces of methanol vapor. Rectisol acid gases are by far the largest
sources of uncontrolled sulfur and CO emissions in a K-T facility.
An additional acid gas stream is generated in conjunction with the
cyanide wash operation preceding the acid gas removal unit. Flash gases
from the cyanide wash are enriched in FLS and HCN. In methanol-based cyanide
wash systems, flash gases may also contain CO, COS, and methanol.
Approaches to treatment of the sulfur-rich acid gases are aimed pri-
marily at removal of reduced sulfur species by bulk sulfur removal followed
by tail gas treatment. Partial or essentially complete control of HCN, CO,
NH^, and methanol can be realized either as an integral part of the sulfur
control approach or through a separate add-on step such as incineration.
The approach for controlling CO emissions in sulfur-free acid gas streams
involves incineration.
173
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Bulk Removal
Bulk Sulfur Removal
To date only three processes have seen any significant commercial appli-
cation for the removal of hLS from acid or fuel gases, namely the Claus, Stret-
ford, and Giamarco-Vetrocoke processes. These are the only processes examined
here due to their commercial status, reliable operation, applicability to
the wide range of sulfur contents in the acid gases, and, in the cases of the
Claus and Stretford processes, availability of operating information and
capital cost data. In addition to these processes, alternatives involving
incineration with S02 removal may be applicable for facilities using low sul-
fur coals. It is recognized that a few other processes are available or have
been proposed; however, it is unlikely that processes other than Claus, Stret-
ford or Giamarco-Vetrocoke will be utilized in first generation coal gasifica-
tion facilities in the U.S. In existing applications, Stretford units are
favored economically over Claus units for feeds containing only a few percent
sulfur, although Claus plants have operated successfully on feeds containing
as low as 8% H2$. The Giamarco-Vetrocoke process is generally applicable to
feed streams with hLS concentrations of up to 1.5%. Table 4-3 summarizes
the key features of these bulk sulfur removal processes.
The Claus process is a dry, high temperature process in which H^S is
catalytically reacted with S02 to form elemental sulfur. There are two
common versions of the process: "straight through", and "split flow".
In the "straight through" mode, sufficient air is added to oxidize
one-third of the H2S to S02. The "split flow" mode, shown in Figure 4-1, is
often used when H2S levels in the feed gas are below 25% by volume. The
acid gas is split into two streams and one-third of the input acid gas is
combusted in a reaction furnace to form S02. Heat is recovered from the gas
before it is recombined with the other two-thirds of the feed. The combined
stream then enters a series of converter stages where elemental sulfur is
produced. Regardless of the Claus mode, the number of stages determines
sulfur removal efficiency; Claus units of 3-stage design can achieve overall
174
-------
TABLE 4-3. KEY FEATURES OF BULK SULFUR REMOVAL PROCESSES
Principle of
Operation
Claus
Catalytic oxidation
of H2S and SO? to
elemental sulfur.
Components Removed H2S, COS, RSH, VQC,
Stretford
Liquid phase oxidation of HoS
to elemental sulfur in an
alkaline solution of metavana-
date and anthraquinone disul-
fonic acid.
H
?S, HCN, and CH..SH.
^
Giamarco-Vetrocoke (G-V)
Liquid phase oxidation of H^S
to elemental sulfur in sodium
carbonate and arsenate/
arsenite solution.
H2S, COS, and
lncineration/S02 Removal
Oxidation of reduced sulfur and
organics, followed by S02 removal
using either regenerative or
throwaway FGD technologies.
H2S, COS, RSH, VOC, and CO.
CJ1
Efficiency
Feed Stream
Requirements/
Restrictions
By-Products
Secondary
Waste Streams
Over 95X total S,
other combustibles
partially destroyed*
Streams containing
H2S levels much
below 10- require
enrichment prior
to processing.
Organics cause com-
bustion control
problems and "grey"
sulfur.
Elemental sulfur,
Spent catalyst and
catalyst regenera-
tion decommissioning
offgas^
As low as 1 ppmv H.S but no
removal of non H-S^sulfur.
High HCN loading should be
reduced prior to processing to
prevent excessive solution
purge.
Elemental sulfur.
Oxidizer vent gas and purge
solution.
Maximum 1.5" H2S in feed.
Elemental sulfur which may
require arsenic removal.
Oxidizer vent gas and
arsenate/arsenite wash
water.
As low as TOO ppmv VOC in incinerated
gas and up to 992 total sulfur
removal.
In principle, gases with any level
of H2S or sulfur compounds could
be incinerated and subsequently
treated via FGD. Other components
cause no problem.
Either CaS04, concentrated S02, or
throwaway lime sludges are gen-
erated by FGD units.
Some condensate and scrubber
sludge.
Reliability/
Limitations
Effects of High
C02 in Feed
NH3 and HCs may
cause catalyst plug-
ging and variable
sulfur recovery.
Can adversely affect
sulfur removal
ability of the
process.
Process does not remove COS,
RSH, or organics, HCN forms
nonregenerable salts in scrub-
bing solution.
High C02 concentrations will
decrease absorption efficiency
by lowering solution alka-
linity. Increased absorber
tower height and addition of
caustic are required.
Hazardous nature of arsenic
solution may cause handling and
safety problems.
Little or no effect ,
FGDs systems have varying degrees of
reliability and generally have lower
on stream factors than process units.
Capital Costs $?b to $180 x 10 per $110 to $270 x 103 per Mg
Mg sulfur/day capacity sulfur/day capacity, depending
depending on both primarily upon total flow.
total flow and sulfur
content.
No cost data available.
$700 to $1700 x 10J per Mg sulfur/day
capacity depending upon total flow
and degree of sulfur removal.
General Comments
Applicable only to
acid gases from
selective AGR system.
Hydrocarbon removal
from feeds may be
necessary.
1 ppmv H^S in tail gas is
possible, however higher limits
are proposed when high levels
of other reduced sulfur species
are present in tail gas.
Limited data available. Haz-
ardous nature of arsenic solu-
tion makes application unlikely
in large U.S. facilities.
FGD process has usually been
applied to combustion flue gases
containing less than 5000 ppmv S02
and achieving about 90^ control.
Performance and cost data for higher
SO? feeds achieving 99°= control are
limited.
-------
en
HP
STEAM
t
CLAUS,
FEED
-^
'
REACTION
_ FURNACE
F
t
BFW
1 N
J\
*i • • |
Yd
AIR
BLOWER
i
>
HP
STEAM
LP
STEAM
1 t
> /
p
CONDENSER
NO. 1
1
BFW
AIR
1
r
~^
CONVERTER I
NO. 1 I
.
X
/
J
REHEATER
NO. 1
1
0
HP
STEAM
LP
STEAM
it
P
CONDENSER
NO. 2
t
BFW
i
r
('
"^
CONVERTER \
NO. 2 I
S
J
REHEATER
NO. 2
•\
0
HP
STEAM
LP
STEAM
I t
/
i
0
CONDENSER
NO. 3
t
BFW
T
r
V
/*
J
REHEATER
NO. 3
1
TAILGAS
A k
LP
STEAM
-t
CONDENSER
NO. 4.
T LIQUIC
BFW SULFU
,- t
SULFUR PIT
Figure 4-1. Three stage Claus plant with split flow option
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Bulk Removal
removal efficiencies of over 95%. Gaseous sulfur species distribution in
Claus tail gas in high CCL applications is approximately 60% FLS, 30% S02,
9% COS, and 1% CSp, although exact levels can vary depending upon the feed
gas composition and Claus plant operation. Elemental sulfur as both vapor
and entrained mist can contribute 20-50% to the total sulfur in Claus tail
gases, depending primarily on the level of h^S in the Claus feed and the
effectiveness of mist eliminators. The relative contribution of elemental
sulfur to total sulfur in Claus tail gas generally increases as HLS content
of Claus feed gas decreases.
In the "straight through" mode of Claus operation, organics, HCN, and
NhU in the feed are largely converted to carbon dioxide, water vapor, and
elemental nitrogen. Such components are not ordinarily of concern unless
levels exceed perhaps 1% vol. each. Organics make control of combustion stoi-
chiometry and temperature more difficult and can lead to an off-color by-
product sulfur containing elemental carbon. HCN at high levels causes corro-
sion throughout the process while NH3 can form deposits which plug/deactivate
Claus catalysts. The organics problem is usually solved by limiting their
content in the Claus feed. HCN at high levels can be removed (converted
to NH3) prior to entering the Claus furnace using Claus or shift type cata-
lysts under reducing conditions. Ammonia at high levels requires either bulk
removal prior to entering the Claus unit or special design to minimize
deposition of ammonia salts.
With feed gases dilute in H^S or other combustibles, flame stability
can be a problem in "straight through" operation. One method for control
of the sulfur combustion process in Claus plants with feeds containing less
than 25% H2S or more than 30% C02 has been the use of the "split flow" mode.
In this process, any organics, HCN, and NH^ in Claus feed gases would be
only about one-third destroyed in the split flow mode unless streams contain-
ing high levels of these constituents are specifically routed to the combus-
tion furnace of the Claus plant. Generally, organics present the most
177
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Bulk Removal .
difficult problem for split flow Claus plants, leading to carbon contaminated
sulfur. A portion of the input organics, HCN, and NHg to split flow Claus
plants may be present in Claus tail gases prior to incineration.
One problem potentially associated with "split flow" operation is the
decomposition of olefins and aromatics to elemental carbon which contaminates
the product sulfur. In such cases, other approaches to combustion control
have been utilized. First, combustible gases such as CO, H£, or CH^ are
added to the feed or are combusted separately to provide sufficient heat to
enable operation in the straight through mode. Liquid sulfur has also been
used. Second, and of particular attractiveness in coal gasification facil-
ilities, is the use of oxygen (Og) or enriched air. Use of 02 not only im-
proves flame stability, but also decreases the inert volume through the Claus
and any subsequent tail gas treatment units. Since unit sizes and associated
costs are flow dependent, savings can be realized. Oxygen is expensive to
generate, but much of the capital cost is already absorbed in a gasification
facility which would necessarily have a large on-site oxygen plant. Finally,
the acid gas and/or air can be preheated before being fed to the Claus
burner, using steam or flue gas from fuel combustion as the heating media.
The Claus process produces spent catalyst and catalyst regeneration
off-gases where catalyst regeneration is used. The Claus catalyst has an
estimated life of at least two to three years. Regeneration of catalyst is
performed intermittently at a few facilities when the efficiency of the proc-
ess drops below acceptable "levels. However, no data regarding regeneration
frequency, duration, or offgas characteristics are available.
The original Stretford process (as developed by the British Gas Corpora-
tion) is a liquid-phase oxidation process using an aqueous solution of sodium
vanadate and anthraquinone disulfonic acid (ADA) in which H2$ is both absorbed
and converted to sulfur. Figure 4-2 is a simplified flow diagram of the Stret-
ford process. The H^S is absorbed in either a packed tower (or contacted in
178
-------
ACID
GAS '
AIR
TAIL GAS
C. W.
ABSORBER
VENT
PRIMARY
OXIDIZER
i
H2S RICH GAS
WATER CHEMICALS
SOLVENT
MAKE-UP
SECONDARY
OXIDIZER
RECOVERED CHEMICALS
WATER
CENTRIFUGE
STEAM
SULFUR
MELTER
SEPARATOR
REDUCTIVE
INCINERATION
PURGE SOLUTION
SULFUR
Figure 4-2. The Stretford process
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Bulk Removal
a venturi scrubber) and then oxidized to sulfur by the sodium vanadate. Re-
duced vanadium is then oxidized by the ADA Solution. ADA is regenerated with
air in the oxidizer tanks where elemental sulfur is removed as a froth. A
continuous solution purge is required to remove the buildup of sodium thio-
sulfate and sodium thiocyanate. Until recently, disposal or treatment of the
solution purge containing thiosulfate, thiocyanates, and small amounts of
vanadium salts was required. In 1973, a reductive incineration process was
developed which converts the purge solution into a gas stream containing H^S,
water vapor and a solid residue containing soda ash and reduced vanadium
salts The salts are returned to the Stretford process as make-up chemicals
and the F^S-rich gas and water vapor are recycled to the absorber as
shown in Figure 4-2. Thus, the reductive incineration process recovers
expensive chemicals while effectively attaining a "zero" discharge of solu-
tion purge.
Recently, modifications of the original Stretford process have been
developed. One modified version of the Stretford process has been used at the
SASOL Lurgi coal gasification complex in South Africa. At SASOL, severe plug-
ging problems have occurred in the Stretford towers which apparently relate to
the high CO^ levels in the Stretford feed compared to feeds in other services.
Preliminary information indicates that sulfur deposition is primarily respon-
sible. SASOL has modified the original Stretford unit, presumably substituting
a different absorbent while saving the bulk of the existing equipment.
The Stretford process generates two waste streams, the oxidizer vent
gas and the purge solution. The purge solution is treated via the reductive
incineration process where sulfur is recovered as HLS for recycle to the
absorber and sodium and vanadium salts are recovered for reuse. The oxidizer
vent gas is expected to consist primarily of air, carbon dioxide, and water
vapor.
The Giamarco-Vetrocoke H2S removal process is a liquid phase oxidation
process using an absorbent solution of alkali arsenates/arsenites in which
180
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Bulk Removal
hydrogen sulfide is both absorbed and converted to elemental sulfur. Sodium
carbonate is the alkali usually applied for removal of large quantities of
sulfur because of its relatively low cost. The Giamarco-Vetrocoke process
is applicable to gas streams containing up to 1.5% hydrogen sulfide and can
reduce hydrogen sulfide levels to 0.5 ppmv or less. The hydrogen sulfide is
absorbed at pressures from 0.1 to 7.5 MPa by countercurrent absorption. Rich
solution from the absorber is subsequently oxidized in an atmospheric pres-
sure, air-blown column to produce regenerated solution and elemental sulfur.
Product sulfur is recovered by froth flotation, filtered, and washed. Based
upon limited available data, the only waste streams generated by this process
are wash water from the sulfur washing operation and oxidizer vent gas.
Characterization data are not available for these streams although the wash
water will contain arsenate/arsenite absorber solution.
In principle acid gases can be directly incinerated to convert all
organics, CO, and reduced sulfur and nitrogen species to C02, H20, N2 and SOp.
S02 removal from the incinerated gas could then be accomplished using any one
of a number of available FGD processes (see Section 4.1.2). Generally, such an
approach is unattractive for several reasons. Throwaway FGD systems create
large solid waste disposal problems and recovery type FGD systems often feature
Claus or Claus type (e.g., the Allied process) processes for elemental sulfur
recovery. There is little to be gained in these approaches over direct use
of the Claus process. FGD systems are also not demonstrated for gases con-
taining over about 0.5% S02, and although there appear to be no inherent tech-
nical limitations prohibiting designs for much higher S02 levels, such systems
are expected to be several times more costly than Claus or Stretford plants
applied directly to crude acid gases. However, for facilities using very low
sulfur coals, the direct incineration approach might be viable due to high
costs associated with enrichment to obtain an H2S-rich acid gas suitable for
Claus processing.
181
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Tail Gas Treatment
Tail Gas Treatment
A number of processes are commercially available for treatment of sulfur
plant tail gases or other waste gases containing low levels of reduced sulfur
species. Table 4-4 summarizes the key features of the most prominent of these
processes. The processes can be categorized in three generic types:
1) Conversion of sulfur species to H2S followed by its removal - this
includes processes such as the Beavon/Stretford and SCOT processes.
2) Conversion of sulfur species to sulfur dioxide (S02) by incineration,
followed by S02 removal - this includes processes such as the Well-
man-Lord or Chiyoda Thoroughbred 102.
3) Extensions of the Claus process - this includes processes such as the
IFP-Clauspol 1500, Sulfreen or BSR/Selectox processes.
Processes in the first category involve catalytic reduction of oxidized
sulfur species to hUS followed by HLS removal from the gas stream by solvent
absorption. In general, the designs of these processes are influenced by the
high levels of C02 in the feed gas. High C02 levels reduce the efficiency of
catalytic reduction of COS and CSp to H^S and impair the effectiveness of the
h^S removal/recovery systems.
Both the Beavon/Stretford and the SCOT processes are commercially avail-
able catalytic processes which are potentially applicable to sulfur plant tail
gases in indirect coal liquefaction facilities. These processes feature two
sections: a hydrogenation section to convert sulfur species in the gas to
HpS and an H^S absorption section. In the hydrogenation reactor, a reducing
gas is added to the feed gas and the combined gas stream is passed over a
cobalt molybdate catalyst. The hydrogenation/hydrolysis reactions occur in the
catalyst bed reducing the sulfur species to HLS.
The Beavon process employs a Stretford unit for hLS absorption and
elemental sulfur production. In contrast, the SCOT process employs an
alkanolamine scrubbing system for hLS absorption. The absorbing solution is
then regenerated resulting in a HpS-rich acid gas which is ordinarily returned
182
-------
TABLE 4-4. KEY FEATURES OF RESIDUAL SULFUR REMOVAL PROCESSES
Beavon
SCOT
Incineration Plus SO,
Removal (FGD) i
Sulfreen
Principle of
Operation
Components Removed
Catalytic reduction of sulfur
compounds to H2S, followed by
integrated Stretford process.
H2S, COS, CS2> and
Sulfur species are
catalytically reduced to
H2$; HjjS is scrubbed in
a regenerable amine
system.
Incineration (an on-site
boiler or separate
incinerator) followed by
SO? removal (e.g., Uellman-
Lord).
H2S, COS, and CS.,, S02. 503, also removes HCs,
CH3SK, NH3, and HCN.
Gas phase continuation of
Claus reaction at a low
temperature.
H-S, S02> COS, and CS2-
00
co
Efficiency
Feed Stream
Requirements/
Restrictions
By-Products
Secondary Waste
Streams
Reliability/
Limitations
Over 99.9% total sulfur re-
moval in combination with the
Claus plant or can attain
equivalent of 50 ppm total
sulfur in tail gas (not
including reducing gases).
None.
Elemental sulfur.
Sour condensate, oxidizer
vent gas, solution purge,
and spent catalyst.
Has only been applied to
Claus process tail gases.
Over 99.9;» total sulfur
removal in combination with
the Claus plant or can
attain equivalent of 250
ppm total sulfur in tail
gas (will vary depending
on C02 and H?_S concentra-
tion in specific applica-
tions) .
None.
Concentrated H2S.
Sour condensate and
spent catalyst.
Requires further treat-
ment and/or recycle to
Claus.
Up to 99' total sulfur re-
moval from Claus tail gas
or 50 ppm SOj in tail gas
and complete removal of
other compounds.
None.
Sulfur or sulfuric acid from
the Wellman-Lord recovery
FGD process.
Sour condensate and solution
purge.
Up to 99° sulfur removal
in combination with Claus
plant. Can exoect a
typical total sulfur level
of 2500 ppm in tail gas.
Optimum performance requires
H2S:S02 ration of 2:1.
Elemental liquid sulfur.
Spent catalyst.
Solid wastes may be generated Has only been applied to
by throwaway FGD processes. Claus process tail gases.
Effects of High
CO- in Feed
Reduces conversion efficiency
of catalyst and decreases H?S
absorption in Stretford
solution.
Reduces conversion
efficiency of catalyst
and efficiency of
alkanolamine system.
None.
No effect.
Capital Costs
General Comments
$20 to $60 x 103 per Mg/day
of S at Claus plant.
Exact ppm limit achievable in
coal gasification application
is not known. Vendor believes
100 ppm is attainable.
$20 to $60 x 10J per Mg/
day of S at Claus plant.
Off-gas from amine
scrubber is not as low in
total sulfur as ReSvon
process.
$40 to $110 x 10 per Mg/day
of S at Claus plant.
On-site boiler/FGD system Much higher residuals in tail
is the most likely candidate gas than Beavon process.
Installing a separate incin-
erator and FGD would not be
as economically feasible.
(Continued)
-------
TABLE 4-4. (CONTINUED)
Cleanair
IFP Claus 1,500
IFP-2
BSR/Selectox
00
Principle of
Operation
Components Removed
Efficiency
Feed Stream
Requirements/
Restrictions
By-Products
Secondary Waste
Streams
Reliability/
Limitations
Effects on High
C02 in Feed
General Comments
Catalytic reduction of
sulfur compounds to t^S
followed by a continua-
tion of the Claus reaction
and Stretford process.
H2S, COS, CS2, and S02>
Reduces sulfur to less than
250 to 300 ppm S02 equiva-
lent in effluent.
H2$:S02 ratio can vary up
to 8 to 1 without affect-
ing efficiency; designed
specifically for Claus
tail gas e
Elemental sulfur.
Spent catalyst„
Has only been applied to
Claus process tail gases.
Reduces conversion effi-
ciency of catalyst;
decreases HoS absorption
in Stretford solution.
Cannot attain as low a
residual sulfur level in
tail gas as Beavon process.
Liquid phase continuation
of Claus reaction at a low
temperature.
H2S, and S02.
Reduces sulfur species in
Claus tail gas to 1500
ppa as S02-
H2S:S02 ratio must be main-
tained in the range of 2.0
to 2.4.
Elemental sulfur.
Spent catalyst o
Has only been applied to
Claus process tail gases.
No effect.
Cannot attain as low a
residual sulfur level in
tail gas as Beavon process.
Incineration of tail gas
followed by ammonia scrub-
bing. Solution is evapor-
ated to produce a concen-
tration S02 stream which
is returned to the Claus
plant.
COS, CS2, and H2S.
Reduce^ suhur species in
Claus tail gas to less than
500 ppn.
H2S:S02 ratio must be main-
tained in the range of 2.0
to 2.4.
Concentrated S02.
Spent catalyst •
Has only been applied to
Claus process tail gases.
No effect.
Cannot attain as low a
residual sulfur level in
tail gas as Beavon process.
Catalytic reduction of sulfur
compounds to H2S, followed
by oxidation of H2S to sulfur
over Selextox catalyst.
H2S, S02, COS, and CS2.
Up tu 99.5*. total sulfur re-
moval equivalent to 750 ppmv
S02 i" the incinerated off-
gas.
o
ratio must be main-
tained in the range of 2.0
to 2.4. HC and NH, should not
be in the feed.
Elemental liquid sulfur,
Spent Beavon and Selectox
catalyst, and sourcondensate.
Has only been applied to
Claus plant tail gas.
Reduces conversion efficiency
of BSR catalyst.
Higher sulfur emissions than
Beavon process ,
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Tail Gas Treatment
to the parent Claus plant for treatment. The alkanolamine scrubbing system
ultimately limits the SCOT'S capabilities because the solvent is only
partially selective for H^S over CO,,. Thus, where feeds contain large amounts
of C0?, it is more difficult to generate an FUS stream suitable for Claus
processing while simultaneously obtaining a tail gas stream with a low level
of total sulfur. In high C02 applications, vendors of the Beavon process
report that levels of less than 100 ppmv total sulfur can be achieved, while
vendors of the SCOT process report less than 350 ppmv total sulfur (Claus
plant tail gas bases).
Inherent limitations of selectivity in the amine absorption step of the
SCOT process place a lower limit of about 200 ppmv of H2S which will be present
in SCOT tail gases. In comparison, the Stretford unit following the Beavon
reactor can remove H2S to below 10 ppmv H2S. Both systems would result
in 50-100 ppmv tail gas COS in high C02 applications. The higher levels of
HLS in SCOT tail gases in existing applications have necessitated that the
gases be incinerated to minimize odor problems while Beavon tail gases do not
generally require incineration. However, no Beavon/Stretford units are
currently used in high C02 service, while at least four SCOT units have
sucessfully operated on feeds with C02 levels above 40%.
The secondary category of processes involves incineration of the waste
gas followed by SOo removal. Such processes are capable of achieving levels
as low as 150 ppmv of S02 in tail gas. One of the more prominent processes,
the Wellman-Lord process, removes S02 with a sodium sulfite solution. Sub-
sequent regeneration of the absorbent generates a concentrated SOo stream
which would be recycled to the parent Claus plant. Two approaches to
incineration can be employed, either separate incineration (with added fuel
where needed) or incineration in an onsite boiler. Either option would
net similar results from a sulfur oxidation standpoint.
Processes in the third category are used exclusively for Claus plant
tail gases and are capable of approximately 80% recovery of tail gas sulfur.
185
-------
Section 4
Acid Gas - Red. 5/N, Org., CO
Tail Gas Treatment
These processes are extension of the Claus process and therefore require a
2:1 ratio of HUS to S02 for proper operation. The limited sulfur removal
capabilities of these processes result in sulfur concentrations of not less
than 1000 ppmv. To date, these processes have not been proposed for coal
gasification applications in the U.S.
186
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Incineration
Incineration
As discussed above, partial or total control of HCN, CO, NHU, and VOC
can be achieved during bulk sulfur removal or sulfur tail gas treatment.
The combustion section of a Claus unit largely destroys these constituents,
but only in that portion of the gas passing through the burner. Cyanide can
also be destroyed over Claus type catalysts used ahead of the Claus plant
itself. HCN is removed from feed gases in the Stretford process, forming
SCN" which leaves the systems with the aqueous blowdown. NhU is also par-
tially removed by Stretford solvent while organics, CO, and COS are not re-
moved by Stretford. Catalytic sulfur tail gas treatment systems achieve at
least partial control of any residual HCN contained in sulfur plant tail
gases. Both Beavon and SCOT catalytic sections are expected to achieve a
high degree of conversion of HCN to NHo and CO. Hydrocarbons, CO, and NH^
contained in the feed to Beavon or SCOT units or added/generated within such
units will be present in their tail gases.
Incineration of these tail gases is an effective approach to controlling
residual hydrocarbons, CO, NH^, and reduced sulfur species. In addition,
incineration is essentially the only alternative for controlling CO emissions
associated with COo-rich acid gases. SCOT tail gases are ordinarily incin-
erated to minimize odor problems arising from residual H2S. Beavon tail
gases (with lower H^S levels) are not ordinarily incinerated, but an incin-
eration step can be added for control of organics, CO, and/or NHo if neces-
sary. Sulfur dioxide tail gas treatment processes such as the Wellman-Lord
inherently achieve control of VOC, CO, and NI-L as part of the incineration
step. Hence, no further control for these constituents is ordinarily neces-
sary with SOo removal processes.
Advantages and disadvantages of the various incineration technologies
aimed primarily at control of volatile organic compounds are summarized in
Table 4-5. Generally, a greater degree of control is obtained with high
187
-------
TABLE 4-5. COMPARISON OF INCINERATION PROCESSES
Type of Incineration
Thermal Incineration
via Separate Incinerator
Thermal Incineration
in Fuel Fired Boiler
Advantage
Disadvantage
Costs (Total Depreciable Capital)
CO
oo
Catalytic Incineration
Flaring
Can handle all types of waste gases.
Reliable and simple operation is
common. VOC/CO control and oxida-
tion of sulfur compounds simulta-
neously.
Sulfur and particulates can be
removed in the associated electro-
static precipitator and flue gas
desulfunzation (FGD) units when
these are integral with the boiler.
The fuel required for steam boiler
incineration is less than that of
a separate incinerator for wastes
with low heating values.
Requires less fuel than thermal
incineration, although heat re-
covery may not be as high. Waste
gases with very little combustible
material can often be incinerated
catalytically without supplemental
fuel.
Simple to operate, least expensive
alternative, especially for trans-
ient and small volume waste gases.
High supplemental fuel costs for
streams with low heating value,
control is a problem with streams
of varying flow and composition.
In most cases, this option is more
capital intensive than a separate
incinerator; however, extent of
heat recovery is generally greater
with boilers than with incinerators.
Subject to control problems with
varying waste gas flow rates and
compositions.
Cannot handle large quantities of
particulates; they will gradually
coat the catalyst and reduce its
efficiency. Some catalyst can be
easily poisoned by sulfur compounds
and elements such as arsenic and
lead. High levels of hydrocarbons
can raise catalyst to excessive
temperatures and shorten the useful
life of the catalyst. Temperature
control is also a problem with
streams of varying flow and com-
position.
Destruction efficiencies much lower
than for thermal or catalytic incin-
eration. Performance data are
generally lacking.
On basis of kmols/hr of flow: flow
range of 0.3 to 3.0 x 103 kmol/hr
a. no heat recovery: S140 to S870
b. primary heat recovery: $190 to S1000
c. primary and secondary heat recovery:
$225 to S1200
Incremental boiler capital costs are
$2000-S3000/kg mole of incremental flue
gas compared to coal combustion on a heat-
ing value basis. Incremental ESP and FGD
costs are an additional S2000-S2500/kg
mole incremental flue gas.
S400 to S2200 per kmol/hr of flow for
a flow rate range of 40 to 7100 kmol/nr
40 to 100 ft elevated flares for flow
rate range of 800 to 6500 kmol/hr -
$25 to $125 per kmol/hr
Ground flares for flow rate o^ 100 to
1000 kmols/hr - $800 to $2700 per
kmol/hr
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Incineration
temperature incineration in either a fuel-fired boiler or a separate incin-
erator (either thermal or catalytic) than can be achieved through the use of
flares. The main combustion zone of a gas incinerator is engineered such
that the gases are maintained at a minimum temperature of HOOK for a mini-
mum of 0.5 seconds. This results in nearly complete destruction of volatile
organic compounds, reduced sulfur compounds, organic aerosols, CO, Nf-U, HCN,
and particulate matter consisting primarily of combustible material.
Thermal incineration may also be effected in a boiler where a minimum
combustion temperature of 1500K and a minimum residence time of 0.5 seconds
are typical design parameters. This approach results in a degree of pollu-
tant destruction similar to that which would be achieved in a specially
engineered incinerator. In tail gas or C02-rich offgas treatment applica-
tions, using the boiler as an incinerator results in an increase in the
capital and annualized operating costs of captive ESP and FGD units, since
costs of the pollution control units are flow rate dependent (even if the
increased gas flow to the ESP/FGD unit contains no dust or sulfur dioxide),
and these waste gases will not be of sufficient heating value to displace
the primary boiler fuel.
Catalytic incineration is not likely to be an attractive alternative
for control of carbon monoxide and VOC in tail gases from sulfur recovery
units due to the presence of sufficient reduced sulfur compounds to inter-
fere with or degrade catalysts. However, catalytic incineration of CO-rich
flash gases from Rectisol has considerable promise, and is featured in the
recent design of at least one U.S. coal gasification facility under construc-
tion (based on Texaco gasification).
189
-------
Section 4
H2S-Rich Rectisol Gases
4.1.1.1 H2S-Rich Rectisol Offgas (Stream 216) - Individual Stream Contro"!
Details of pollution control alternatives applicable to the H-S-rich
Rectisol offgas (Stream 216) are discussed in this section. Although
this discussion focuses primarily upon individual stream controls, the cost
estimates provided reflect the fact that, in a K-T based gasification faci-
lity, h^S-rich Rectisol offgas would most probably be combined with the
relatively small volume cyanide wash flash gases (Stream 214) prior to treat-
ment. The likelihood of such stream combination derives from the high sulfur
concentrations of both of these streams, and the high percentage of the gasi-
fied coal total sulfur present in the cyanide wash flash gases (about 1 to
3%). This approach is necessary in order to provide realistic cost esti-
mates for the bulk sulfur removal, tail gas treatment, and incineration con-
trols discussed in this section.
For evaluation purposes, the H2S-rich Rectisol offgas have been
assumed to contain 50% carbon dioxide, 42% hydrogen sulfide, 5% carbonyl sul-
fide, 0.4% carbon disulfide, 636 ppmv cyanide, and 200 ppmv sulfur dioxide.
The main organic compound present in significant quantities is expected to be
methanol at about a 1% level (see Section 3.3.6). The corresponding flow
rate for this stream is about 535 kmol/hr. It should be noted that the exact
hydrogen sulfide level in the H2S-rich Rectisol offgas would be determined
based upon engineering/economic tradeoff studies of Rectisol enrichment costs
versus bulk sulfur removal and tail gas treatment costs. The most economic
sulfide level is likely to be coal-specific and, in particular, vary with
coal sulfur content and sulfur recovery/pollution control alternatives.
190
-------
Section 4
H2S-Rich Rectisol Gases
Bulk Sulfur Removal
Bulk Sulfur Removal_
In the base plant, the H^S-rich Rectisol offgas contains approximately
255 kmol/hr total sulfur, and is the primary feed gas to the bulk sulfur re-
moval unit. An additional 2 to 8 kmol/hr of total sulfur is likely to be
present in the feed gas to bulk sulfur removal, depending upon whether a
water-based or methanol-based cyanide wash system is employed (refer to Sec-
tions 3.3.4 or 4.1.1.2 for cyanide wash flash gas compositions). Thus, a
total feed rate to the bulk sulfur removal unit of about 549 kmol/hr, includ-
ing 257 to 263 kmol/hr total sulfur (approximately 47% total sulfur), is
expected from the uncontrolled base plant. Such a stream is suitable for
bulk sulfur removal by the Claus process since it has a relatively high sul-
fur concentration and does not contain appreciable levels of contaminants
such as organics, HCN, or NH^.
As discussed previously, Claus units of 3-stage design can achieve over-
all sulfur removal efficiencies of over 95%. Assuming a 95% sulfur removal
efficiency, a Claus tail gas sulfur loading of 13 kmol/hr would be expected.
The gaseous sulfur species distribution in Claus tail gas in high C0? appli-
cations is approximately 60% H2S, 30% S02, 9% COS, and 1% CS2- It should be
noted that up to 50% of the Claus tail gas total sulfur (S02 equivalent basis)
may actually be present as either sulfur vapor or entrained elemental sulfur.
However, this would not appreciably alter the tail gas sulfur species distri-
bution (e.g., for 50% S02 equivalent sulfur as Sg, the species distribution
would become 53% H2S, 27% S02, 8% COS, 1% CS2, and 11% Sg). The correspond-
ing concentration of total sulfur (S02 equivalent) in the Claus tail gas
would be 1.5 to 1.6% (dry basis). Some degree of removal of minor constitu-
ents present in the Claus feed gas such as methanol, CO, and HCN would be
achieved in addition to sulfur removal. It is expected that about one-third
or more of the methanol and CO present in the feed gas would be combusted
resulting in tail gas concentrations of less than 0.6% and 0.4%, respectively.
However, CO is apparently generated within the Claus reactor, and concentrations
191
-------
Section 4
H2S-Rich Rectisol Gases
Bulk Sulfur Removal
of CO in the 1 to 2% range are typically reported. Feed gas cyanide, present
at concentrations of up to about 0.4%, is expected to be almost totally des-
troyed during Claus processing but may contribute to low levels of NhU in
the tail gas. For the subject Claus plant, a dry tail gas flow rate of about
800 to 900 kmol/hr would be expected with 200 to 250 kmol/hr of reaction
moisture.
For a Claus plant of the capacity specified above, total capital invest-
ment and annualized costs would be about $8.7 million and $0.4 million,
respectively. However, the cost of the Claus unit is dependent upon both the
feed gas flow rate and sulfur content. For example, if the selectivity
achieved by the Rectisol unit was reduced to produce a Claus feed gas with
15% sulfur for the same coal sulfur content, the Claus feed gas rate would
increase by a factor of about three and the total capital investment for the
Claus plant would increase by about 40%. Alternatively, with the same
Rectisol selectivity (i.e., 47% sulfur in the Claus feed) for a feed coal
having one-half the total sulfur content initially considered, the Claus feed
rate would decrease by about one-half and the total capital investment for
the Claus plant would decrease by about 40%.
The principal secondary waste stream generated by the Claus process is
spent catalyst (Stream 402). The subject Claus plant would have a bauxite
or alumina catalyst inventory of approximately 50 Mg which would periodically
require disposal. Assuming a catalyst life of 5 years, the average spent
catalyst generation rate would be 10 Mg per year. Claus sulfur is produced
at a rate of approximately 7.9 to 8.3 Mg/hr and may also require disposal
depending upon its quality and market considerations.
192
-------
Section 4
^S-Rich Rectisol Gases
Tail Gas Treatment
Tail Gas Treatment
In the base plant, tail gas from bulk sulfur removal would contain
approximately 1.5% total sulfur and less than 0.6% methane and 1 to 2% CO,
on a dry gas basis. Traces of NFU may also be present as a result of HCN
destruction during bulk sulfur removal. A tail gas flow rate of about 840
kmol/hr dry gas and 300 kmol/hr water vapor is expected. A number of pro-
cesses are available for recovering residual sulfur from tail gases from bulk
sulfur removal processes (refer to Section 4.1.1). For purposes of discus-
sion, the Beavon/Stretford, SCOT, and Wellman-Lord processes will be considered.
The Beavon/Stretford process is reported to be capable of reducing
residual sulfur concentrations to about 100 ppmv (dry tail gas basis) with
a 9:1 ratio of COS to H^S. This corresponds to a total sulfur emission rate
of less than 0.1 kmol/hr from tail gas treatment. Minor constituents such
as methanol and NHg present in the tail gas from bulk sulfur removal are not
expected to be affected by the Beavon/Stretford process. An external source
of reducing gas may be required for the catalytic hydrogenation of S0? and
elemental sulfur. Reducing gas may be added directly to tail gas treatment
in the form of H2 and CO-rich gas (e.g., flash gases from Rectisol) or
may be generated by substoichiometric combustion of organic-rich fuel gas
derived from synthesis operations. The extent to which this reducing gas
dilutes the residual sulfur concentration from tail gas treatment depends
upon the reducing gas quality and the amount of air required for stable com-
bustion. For this reason, performance of Beavon/Stretford units is typically
reported on a dry tail gas basis. Residual reducing gas will contribute to
CO concentrations in the effluent from the Beavon/Stretford process although
quantitative data are not available. The total capital investment and annual-
ized costs of the subject Beavon/Stretford unit would be approximately $7.4
million and $1.4 million, respectively.
Secondary waste streams from the Beavon/Stretford process are: (1) sour
reactor effluent condensate (Stream 405), (2) Stretford solution purge (Stream
193
-------
Section 4
H2S-Rich Rectisol Gases
Tail Gas Treatment
406), (3) Stretford oxidizer vent gas (Stream 404), and (4) spent catalyst
from the Beavon reactor (Stream 407). Sour condensate is expected to contain
dissolved sulfide and traces of NH^, and would be generated at a rate of up
3
to 4 m /hr. Stretford solution purge for control of thiosulfate and thiocyanate
buildup may be present as a waste stream containing vanadium and sodium salts
as thiosulfate, sulfate, carbonate, and thiocyanate. The solution purge rate
is approximately 0.1 to 0.2 Mg/hr and purge solution may either be regenerated
or discarded (refer to Section 4.1.1). Oxidizer vent gas would consist primarily
of air, water vapor, and (^ but may contain traces of NhU. Insufficient data
are available for estimating the flow rate of oxidizer vent gas. The cobalt
molybdate hydrogenation catalyst inventory is approximately 15 Mg and would
require periodic replacement. Assuming a catalyst service life of three years,
the average spent catalyst generation rate would be 5 Mg/year. Beavon/
Stretford sulfur is generated at a rate of approximately 0.4 Mg/hr and may
require disposal, depending upon its quality and market considerations.
The SCOT process is reported to be capable of reducing residual sulfur
concentrations to about 250 ppmv (dry tail gas basis). This corresponds to
a total sulfur emission rate of 0.2 kmol/hr from tail gas treatment. Minor
constituents such as methanol and NH3 present in the tail gas from bulk sul-
fur removal are not expected to be affected by the SCOT process. As dis-
cussed in conjunction with the Beavon/Stretford process, supplemental reduc-
ing gas would be required, and the reducing gas quality/generation mode
would influence the concentration of sulfur species in the tail gas treatment
effluent but not the sulfur effluent rate. Residual reducing gas will con-
tribute to CO concentrations in the treated gas from the SCOT process although
quantitative data are not available. The total capital investment and
annualized costs of the subject SCOT unit would be approximately $6.3 million
and $1.9 million, respectively.
Secondary waste streams from the SCOT process are sour water (Stream 409)
and spent catalyst (Stream 410). The sour water is expected to contain dissolved
194
-------
Section 4
I^S-Rich Rectisol Gases
Tail Gas Treatment
3
sulfide and traces of NH^, and would be generated at a rate of 5 m /hr. The
cobalt molybdate hydrogenation catalyst inventory is approximately 15 Mg, and
would require periodic replacement. Assuming a catalyst service life of 5
years, the average spent catalyst generation rate would be 3 Mg/year.
The Wellman-Lord process is reported to be capable of reducing residual
sulfur concentrations to about 250 ppmv SOo (dry tail gas basis). This
corresponds to a total sulfur emission rate of 0.2 kmol/hr from tail gas
treatment. Minor constituents such as methanol, CO, and NHo present in the
tail gas from bulk suflur removal are expected to be destroyed during tail gas
incineration. Combustion of the tail gas requires supplemental fuel and air,
and results in an increase in the molar flow rate through the tail gas treat-
ment unit. The magnitude of this increase and its related SOn dilution depend
upon the quality of the supplemental fuel and the stoichiometric excess of
air. Assuming methane as the incineration fuel with 20% excess air, effluent
from the subject Wellman-Lord unit would contain about 150 ppmv S02. The total
capital investment and annualized costs for this Wellman-Lord unit would be
approximately $11 million and $3.7 million, respectively.
Secondary waste streams from the Wellman-Lord process are acidic waste-
water from combustion gas quenching (Stream 411) and thiosulfate/sulfate by-
product purge (Stream 412). The acidic wastewater typically has a pH value
3
between 1 and 2, and would be generated at a rate of 3 m /hr. By-product
purge consists primarily of sodium salts such as sulfite, pryosulfite, sul-
fate, and thiosulfate with approximately 29% water. By-product purge is
generated at a rate of 90 kg/hr.
195
-------
Section 4
H2S-Rich Rectisol Gases
Incineration
Incineration
Waste gases from tail gas treatment may be incinerated to control odors
due to the presence of H2$ and to destroy CO. As discussed in Section 4.1.1,
thermal incineration, catalytic incineration, and flaring are employed in
various commercial applications. Thermal incineration in a dedicated incin-
erator is an approach often used in conjunction with tail gas treatment units
and will be considered in this section.
In the base plant, gases from tail gas treatment processes not having
an integral incineration step are generated at a rate of approximately 880
kmol/hr. The sulfur content of these gases ranges from 100 to 250 ppmv with
H2S concentrations ranging from 10 to 200 ppmv. The level of combustibles
is expected to be inadequate to support combustion, and supplemental fuel
would be required for incineration. Using desulfurized, unshifted synthesis
gas as a typical fuel and an incineration temperature of HOOK, about 1230
to 1300 kmol/hr of incineration gas would be generated. The concentration
of S02 in the effluent would be 70 to 180 ppmv, depending upon the input
sulfur concentration and the amount of excess combustion air provided for
incineration. Concentrations of CO in the combustion gas are likely to be
below 100 ppmv although quantitative data are limited. No secondary waste
streams are generated by thermal incineration processes. The total capital
investment and annualized costs for the subject thermal incinerator would
be approximately $0.42 million and $0.63 million, respectively.
196
-------
Section 4
Cyanide Wash Flash Gas
4.1.1.2 Cyanide Wash Flash Gases (Stream 214) - Individual Stream Control
Generation rates and compositions of flash gases from cyanide wash
operations are dependent upon whether a water-based or methanol-based cyanide
wash is employed (see Section 3.3.4). In the case of a water-based cyanide
wash, the estimated flash gas generation rate is 22 kmol/hr, consisting of
89% C02, 11% H2$, and 449 ppmv HCN (Stream 214a). In the case of a methanol-
based cyanide wash, the estimated flash gas generation rate is 16 kmol/hr,
consisting of 45% H2S, 27% CO, 13% HCN, 7% C02, 3% COS, and 2% methanol
(Stream 214b). Due to the relatively high sulfur concentrations of these
streams and since they may contain 1 to 3% of the gasified coal's total sul-
fur, these gases would probably be treated in bulk sulfur removal with the
larger volume H2S-rich Rectisol offgas (Stream 216). Control of cyanide
wash flash gases would be achieved as previously discussed in conjunction
with controls applicable to the H2S-rich Rectisol offgas (Section 4.1.1.1).
197
-------
Section 4
C02-Rich Rectisol Gases
4.1.1.3 C02-Rich Rectisol Offgas (Stream 219) - Individual Stream Control
For purposes of analyses, the combined C02-rich Rectisol offgas (Stream
219) has been estimated to consist primarily of C0? with 13 ppmv total sulfur
and 1% CO (see Section 3.3.6). The indicated sulfur level in this stream is
below the levels attainable with existing sulfur removal processes, and no
further treatment for sulfur removal is warranted. In the event that CO
control would be required for this stream, the Rectisol unit could be de-
signed to further reduce the CO concentration. Carbon monoxide is sparingly
soluble in methanol, and CO absorbed from K-T product gas at high pressures
can be largely recovered as a separate CO-rich C02 stream by flashing laden
methanol from C02 absorption to an intermediate pressure before regeneration
for bulk CO- removal. The CO-rich offgas can be subsequently incinerated
for control of CO emissions; this approach has been proposed in the U.S. for
one Texaco-based coal gasification facility employing the selective Rectisol
process. Alternatively, the CO-rich offgas may be recompressed for addition
to the Rectisol product gas if the resulting C02 load is compatible with
synthesis requirements. This latter alternative may involve design modifica-
tions for both the shift conversion and C02-removal operations to ensure an
appropriate synthesis gas composition. Finally, the CO-rich offgas may be
utilized within the facility as either fuel or reducing gas. Any of these
alternatives would offer advantages over "add on" control of CO in the com-
bined C02-rich offgas stream due to the energy penalty associated with incin-
eration of high volume, low heating value gases.
With regard to controlling potential CO emissions associated with the
C02-rich Rectisol acid gas, catalytic incineration of a CO-rich offgas will
be examined in this section. It should be noted that constraints associated
with the control of CO emissions would influence the design of the Rectisol
acid gas removal unit and, possibly, the shift conversion unit. This influ-
ence may result in an increased cost for these units which should be attri-
buted to pollution control. However, due to the absence of specific design
198
-------
Section 4
C0?-Rich Rectisol Gases
and cost information for the Rectisol and shift conversion units, these
costs will not be addressed in this section. Therefore, control costs pre-
sented in this section may be lower than the actual cost incurred in an inte-
grated facility.
199
-------
Section 4
C02-Rich Rectisol Gases
Incineration
Incineration
To evaluate catalytic incineration as a CO control alternative, a CO-
rich COo stream containing 20% CO and 80% COo has been assumed to be obtain-
able from the Rectisol unit. The flow rate of this stream would be approxi-
mately 500 kmol/hr. No supplemental fuel is required for incineration of
this waste gas in a catalytic incinerator. With a 99% combustion efficiency,
CO emissions from the incinerator would be 1 kmol/hr at an effluent concen-
tration of about 1000 ppmv. The total capital investment for this catalytic
incinerator would be approximately $574,000 which corresponds to 0.05% of
the base plant capital investment. The associated annualized capital charge
is offset by steam credits from heat recovery and a net annualized credit of
$54,000 results.
200
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Sec. Strms. - Other Media
4.1.1.4 Secondary Streams from Other Media
As will be discussed in Section 4.2.4, certain volatile species may degas
or be stripped from wastewaters during treatment. This problem is most likely
to occur with the polysulfide addition and activated sludge processes. The
mass flow of volatile species in the offgas is uncertain, but is expected to
be quite low, and volatiles would primarily consist of hydrogen sulfide with
lesser amounts of hydrogen cyanide and ammonia. Such fugitive emissions would
primarily represent a potential odor problem.
If control of volatile emissions (sulfide/cyanide offgas - Stream 416)
from water pollution control operations is warranted, incineration may be a
viable destruction alternative. One approach to control would involve pro-
viding air-blanketed enclosures for water pollution control units which may
emit odorous volatiles. The enclosure vent gas could then be used as combustion
air in on-site incinerators or boilers. Such an approach would obviate the
need for a dedicated incineration unit, but would require the purchase and
maintenance of enclosures, fans, and ducting. Control costs would primarily
depend upon factors such as the specific water pollution control units
requiring control, the location of the incineration unit(s) best suited for
such an application, and the purge rate of blanketing air required to maintain
slightly subatmospheric pressures within the water pollution control enclosures.
Enclosure of activated sludge units with induced draft collection of over-
head gases for incineration is practiced at a number of domestic/industrial
sewage treatment plants in the U:S. These treatment plants are generally
much larger than those which would be found in K-T based facilities; therefore,
induced draft collection appears feasible from both an economic and technical
standpoint. Large ponds which might be utilized in K-T facilities as "polishing1
steps would be difficult to enclose. However, most volatiles would evolve from
upstream treatment units which could be more economically enclosed. Costs
associated with control of volatile species from wastewater treatment operations
201
-------
Section 4
Acid Gas - Red. S/N, Orq., CO
Sec. Strms. - Other Media
will not be addressed in this manual due to large uncertain!ties in emissions,
their sources, and the site/design-specific nature of costing factors.
202
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Integrated Control
4.1.1.5 Integrated Control Examples
In this section, examples of combined and sequential control of waste
streams are evaluated from the standpoint of the overall emissions reductions
achieved and costs incurred. The selection of specific control examples for
evaluation in this section is not intended to imply that other technologies
could not provide equivalent or better performance with similar or even lower
costs. Specific technologies are selected to cover the types of alternatives
which are under consideration for facilities in the U.S. Selection of
integrated controls will be based upon specific design requirements and local
conditions and can only be made by designers and regulatory authorities in-
volved in a specific project.
Acid gas streams containing reduced sulfur/nitrogen compounds, organics,
and CO and which would logically be combined prior to sequential control
treatment are the H2S-rich Rectisol offgas (Stream 216) and the cyanide
wash flash gases (Stream 214). Three sequential control alternatives for the
combined waste gas stream will be examined in this section.
• Claus bulk sulfur removal with Beavon/Stretford tail gas treatment
0 Claus bulk sulfur removal with SCOT tail gas treatment, and
incineration
t Claus bulk sulfur removal with Wellman-Lord tail gas treatment
These alternatives have been demonstrated in refinery applications and are
currently under consideration for application to the North Alabama Coal
Gasification Consortium Project (82). Detailed descriptions of the controls
included in these alternatives are presented in the PCTM Pollution Control
Technology Appendices.
Example 1 - Claus Bulk Sulfur Removal with Beavon/Stretford Tail Gas Treatment
This example illustrates treatment of the combined hLS-rich Rectisol off-
gas (Stream 216) and the flash gas from a methanol-based cyanide wash unit
(Stream 214b) in a Claus bulk sulfur removal unit with Beavon/Stretford tail
203
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Integrated Control
gas treatment. The control system is presented schematically in Figure 4-3,
and material flow estimates are presented in Table 4-6. The Claus process
could operate in either a "split flow" mode, with the cyanide-rich flash gas
passing entirely through the burner for HCN destruction, or in a "straight
through" mode. The sulfur concentration of the combined feed gas (47% total
sulfur) is sufficiently high to enable "straight through" processing. Also,
the HCN level of 0.4% in the combined feed gas is not expected to require
special approaches for HCN destruction.
The Claus unit is reported to achieve 95% sulfur removal and have
a residual sulfur distribution of 60% H2S, 30% S02, 9% COS, and 1% CS2. It
should be noted that up to 50% of the residual sulfur (S0? equivalent basis)
may actually be present as either sulfur vapor or entrained elemental sulfur,
which would slightly reduce the tail gas volume. One-third of the feed gas
CO and methanol have been assumed to be destroyed during Claus processing,
although higher methanol destruction efficiencies may result depending upon
the specifics of the combustion process. Carbon monoxide may actually be
generated in the Claus process since some facilities report tail gas concen-
trations of 1 to 2% CO with essentially CO-free Claus feed gases. Complete
HCN destruction has been assumed; some low level NH-, emission may result from
HCN destruction although quantitative data are not available.
The Beavon/Stretford unit is reported to be capable of reducing tail
gas residual sulfur concentrations to about 100 ppmv (dry Claus tail gas
basis) with a 9:1 ratio of COS to H2S. A CO-rich fuel gas has been used in
the example for reheat of the Claus tail gas and catalytic reduction of S0?.
The resulting sulfur emission rate is 0.08 kmol/hr from tail gas treatment.
Methanol present in the Claus tail gas is not expected to be affected by the
Beavon/Stretford process. While no residual reducing gas in the Beavon
effluent has been indicated in Table 4-6, concentrations of CO ranging from
250 to 670 ppmv in the effluent have been reported.
204
-------
no
o
on
H2S-RICH
OFFGAS
FLASH GAS
FROM —
METHANOL-
BASED
CYANIDE WASH
STRETFORD
OXIDIZER
VENT GAS
CLAUS BULK
SULFUR REMOVAL.
i
_L
402'
- '
t
SPENT CATALYST
CLAUS TAIL GAS
FUEL/REDUCING
GAS
BEAVON/STRETFORD
TAIL GAS TREATMENT
SULFUR
SOUR
CONDENSATE
SPENT
CATALYST
STRETFORD
SOLUTION
PURGE
BEAVON/STRETFORD TAIL
GAS TO ATMOSPHERE
Figure 4-3. Example 1 - Claus bulk sulfur removal with Beavon/Stretford tail gas treatment
-------
TABLE 4-6. EXAMPLE 1 - MATERIAL FLOW ESTIMATES FOR INTEGRATED CONTROL EMPLOYING
CLAUS BULK SULFUR REMOVAL WITH BEAVON/STRETFORD TAIL GAS TREATMENT*
ro
o
01
Rectisol Cyanide Wash Flash Gas
H2S-Rich Offgas Methanol Wash Case
Stream 216 Stream 214b
kmol/hr Vol kmol/hr Vol
H
CO
co2
COS
cs2
so2
HCN
'2
Methanol
Sulfur
Total Dry Gas
H20
Total
kmol/hr
kg/hr
271 50.8
222 41.6
25 4.7
2 0.4
0.1 218ppmv
0.3 636ppmv
6 1.2
7 1.3
533
533
21544
0.49 3.0
4.4 26.4
1.2 7.3
7.3 44.7
0.56 3.4
2.1 12.9
0.3 1.8
16
16
524
Combi ned
Feed
kmol /hr
0.49
4.4
272
229
26
2
0.1
2.4
6
7
549
549
22068
Claus
Gas
Voi -
892ppmv
0.8
49.5
41.7
4.7
0.3
211 ppmv
0.44
1.2
1.3
Stretford Sour
Claus Sulfur Beavon/Stretford Sulfur Condensate
Claus Tail Gas Stream 403 Fuel Gas Tail Gas Stream 408 Stream 405
kmol/hr Vol kg/hr kmol/hr Vol kmol/hr Vol kq/hr kq/hr
0.33 0.04 33 26.9
2.9 0.3 78 63.3
310 36.8 12 9.8 402 35.6
7.8 0.9 0.008+ 7ppmv
1.2 01 0.075f 74ppmv
0.07 78ppmv
3 9 0.5
512 60.7 721 63 9
4.7 0.6 4.7 0.4
7857 411
843 123 1128
227 47 4027
1070 123 n75
32926 7857 2784 38893 411 4027
The number of significant figures shown in some cases do not represent the degree of accuracy and are retained for material balance purposes only
Nevertheless, slight imbalances do appear as a result of numerical rounding Material flow estimates are based upon published data and engineering
estimates. Performance data and references are detailed in the Control Technology Appendices.
Residual sulfur levels are based upon 100 ppmv total sulfur (Claus tail gas basis).
^Some level of residual reducing gas would be present in the Beavon/Stretford effluent. Concentrations of CO in the effluent ranging from 250 to
670 ppmv have been reported.
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Integrated Control
Secondary waste streams from the control system are: (1) spent Claus
catalyst (Stream 402), (2) sour condensate (Stream 405), (3) Stretford solu-
tion purge (Stream 406), (4) Stretford oxidizer vent gas (Stream 404), and
(5) spent Beavon catalyst (Stream 407). The Claus unit would have a bauxite
or alumina catalyst inventory of approximately 50 Mg. Assuming a catalyst
life of 5 years, the average spent Claus catalyst generation rate would be 10
Mg/year. Sour condensate is expected to contain dissolved sulfide and traces
o
of NH3 and would be generated at a rate of about 4 m /hr. Stretford solution
purge (Stream 406) for control of thiosulfate and thiocyanate buildup may be
present as a waste stream containing vanadium and sodium salts as thiosulfate,
sulfate, carbonate, and thiocyanate. The solution purge rate is approximately
0.1 to 0.2 Mg/hr. Purge solution may either be regenerated or discarded
(refer to Section 4.1.1). Oxidizer vent gas would consist primarily of air,
water vapor, and C02 but may contain traces of NH3. Insufficient data are
available for estimating the flow rate of oxidizer vent gas. The cobalt
molybdate Beavon hydrogenation catalyst inventory is approximately 15 Mg.
Assuming a catalyst service life of three years, the average spent catalyst
generation rate would be 5 Mg/year. In addition, sulfur from the Claus and
Beavon/Stretford processes (Streams 403 and 408) is generated at a combined
rate of 8.3 Mg/hr and may require disposal, depending upon sulfur quality and
market considerations.
Cost estimates for the Claus and Beavon/Stretford units are summarized
in Table 4-7. Estimates are presented on two bases: (1) total capital in-
vestment and annualized costs and (2) total capital investment and annualized
costs as percentages of the uncontrolled base plant costs for the methanol
synthesis case. All costs are presented on a 1980 basis. The total
capital investment was found to be similar for the Claus and Beavon/Stretford
units. However, annualized Claus costs were estimated to be about 70% lower
than those for the Beavon/Stretford, primarily as a result of large credits
associated with steam generated during Claus processing, and fuel gas require-
207
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Integrated Control
TABLE 4-7. EXAMPLE 1 - COSTS OF INTEGRATED CLAUS BULK SULFUR REMOVAL
WITH BEAVON/STRETFORD TAIL GAS TREATMENT (1980 BASIS)
Total Capital Investment
106 Dollars
% of
Base Plant*
Annual i zed Cost
106 Dollars
0.41
% of
Base Plant*
Beavon/
Stretford Unit
7.4
1.4
Total
16.1
1.45
1.8
0.52
Expressed as a percentage of the cost for an uncontrolled methanol synthesis
facility.
ments for the Beavon/Stretford process. Capital and annualized costs for the
integrated control system represent approximately 1.4% and 0.5% of the respec-
tive costs for the uncontrolled methanol synthesis base plant.
Performance and cost of the Claus/Beavon/Stretford control system are
dependent upon both the feed gas flow rate and sulfur content. For example,
if the selectivity of the Rectisol unit were reduced to produce a Claus feed
gas of 15% total sulfur for the same coal sulfur content, the Claus feed gas
flow rate would increase by a factor of about three and the Beavon/Stretford
tail gas flow rate would increase by a factor of about two. Therefore, emis-
sions from the Beavon/Stretford would also increase by a factor of about two.
The resulting increases in capital costs for Claus and Beavon/Stretford units
are estimated to be about 40% and 60%, respectively. Alternatively, with the
same Rectisol selectivity (i.e., 47% sulfur in the Claus feed) for a coal
having one-half the total sulfur content initially considered, the Claus feed
208
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Integrated Control
gas rate and the Beavon/Stretford tail gas flow rate would decrease by about
one-half. Therefore, emissions from the Beavon/Stretford would also decrease
by about one-half. The resulting decreases in capital costs for Claus and
Beavon/Stretford units are estimated to be about 40% and 35%, respectively.
Example 2 - Claus Bulk Sulfur Removal with SCOT Tail Gas Treatment and
Incineration
This example illustrates treatment of the combined FUS-rich Rectisol
acid gas (Stream 216) and the flash gas from a methanol-based cyanide wash
unit (Stream 214b) in a Claus bulk sulfur removal unit with SCOT tail gas
treatment and thermal incineration. The control system is presented schemat-
ically in Figure 4-4, and material flow estimates are presented in Table 4-8.
Assumptions relating to Claus unit performance are idential to those pre-
sented for Example 1 although in this example the SCOT recycle gas repre-
sents an additional Claus feed gas stream.
The SCOT unit is reported to be capable of reducing Claus tail gas resi-
dual sulfur concentrations to about 250 ppmv (dry Claus tail gas basis). A
CO-rich fuel gas has been used in this example to reheat the Claus tail
gas, catalytic reduction of SO-, and subsequent incineration. The resulting
sulfur emission rate is about 0.2 kmol/hr S02 from tail gas treatment and
incineration. While methanol present in the Claus tail gas is not expected to
be affected by SCOT tail gas treatment, essentially complete destruction would
be achieved during incineration.
Secondary waste streams from the control system are: (1) spent Claus
catalyst (Stream 402), (2) sour condensate (Stream 409), and (3) spent SCOT
catalyst (Stream 410). As in Example 1, spent Claus catalyst would be gen-
erated at an average rate of approximately 10 Mg/yr assuming a 5 year catalyst
life. The sour condensate is expected to contain dissolved sulfide and traces
3
of NhL and would be generated at a rate of 4 m /hr. The cobalt molybdate SCOT
hydrogenation catalyst inventory is approximately 15 Mg. Assuming a catalyst
209
-------
HgS-RICH
ACID GAS
FLASH GAS
FROM
METHANOL-
BASED
CYANIDE WASH
SCOT RECYCLE GAS
CLAUS BULK
SULFUR REMOVAL
I
(To?)
t
SPENT
CATALYST
403
CLAUS TAILGAS
FUEL/REDUCING
GAS
SULFUR
SCOT TAIL GAS
TREATMENT
SOUR
CONDENSATE
SCOT TAIL GASJ
SPENT
CATALYST
FUEL GAS
THERMAL
INCINERATION
INCINERATED
GAS TO
ATMOSPHERE
Figure 4-4. Example 2 - Claus bulk sulfur removal-with SCOT tail gas treatment and incineration
-------
TABLE 4-8. EXAMPLE 2 - MATERIAL FLOW ESTIMATES FOR INTEGRATED CONTROL EMPLOYING
CLAUS BULK SULFUR REMOVAL, SCOT TAIL GAS TREATMENT, AND INCINERATION*
H2
CO
co2
H2S
COS
cs2
so2
HCN
N2
Methanol
Sulfur
Total Dry Gas
H20
Total
kmol/hr
kg/hr
Offgas
Stream 216
krol/hr
271
222
25
2
0 1
0 3
6
7
533
533
21544
Vol ,
50 8
41.6
4.7
0.4
218ppmv
636ppmv
1 2
1.3
The number o! significant figures
slight imbalances do appear as a
data and references are detailed
Cyanide ^ash
Flash Gas
Methanol hash Case Combined Claus
Stream 214b SCOT Recycle Gas Feed Gas (
kmol/l
0.49
4.4
1 2
7.3
0 56
2 1
0.3
16
16
524
shown in
result of
in the Coi
ir Vol '. kmol/hr Vol = kmol/hr Vol * k
3.0 0 28 0.6 0.8 0 1
26 4 2.4 50 6.8 1 1
7.J 32 66 8 304 50.9
44 7 13 27 6 242 40 5
34 26 4.4
2 03
0 1 168ppmv
12 9 2.4 0.4
6 1.0
18 7 1.2
48 597
2 2
50 599
1984 24052
some cases do not represent the degree of accuracy and
numer'cal rounding. Material flow estimates are based
Claus Sour
:laus Tail Gas Stream 403 SCOT Fuel Gas SCOT Tail Gas Stream 409 Fuel Gas Tail Gas
mol/hr Vol kg/hr kmol/hr Vol . kmol/hr Vol t kq/hr kmol/hr Vol ". kmol/hr Vol
0 46 SOOppnw 38 26.9 40 26 9
4.1 05 89 63 3 94 63.3
340 37 7 14 9 8 411 34 5 15 9.8 520 33 6
8.1 0 9 0 2f 168ppm»
12 01 0.04* 34ppfflv
0 07 78ppmv
4105 0.2 129ppmv
538 59 7 776 65 1 1029 66 4
4 7 0.5 4.7 0 4
8260
90! 141 1192 149 1549
242 50 3978 90
1220 8260 141 1244 149 1639
35295 2569 40889 3978 2715 5335
are retained for material balance purposes only Nonetheless
upon published data and engineering estimates. Performance
fResidual sulfur levels are based upon 250 ppmv total sulfur (Claus plant tail gas basis).
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Integrated Control
life of 5 years, the average spent catalyst generation rate would be 3 Mg/yr.
In addition, Claus sulfur is produced at a rate of about 8.3 Mg/hr and may
require disposal, depending upon sulfur quality and market considerations.
Cost estimates for the Claus, SCOT, and incineration units are summarized
in Table 4-9. Estimates are presented on two bases: (1) total capital in-
vestment and annualized cost, and (2) total capital investment and annualized
cost percentages of the uncontrolled base plant cost for the rnethanol synthe-
sis case. All costs are presented on a 1980 basis. The total capital invest-
ment for the combined SCOT/incineration system was estimated to be nearly 30%
lower than the capital investment for the Claus unit. However, annualized
Claus costs are estimated to be about 80% lower than those for the combined
SCOT/incineration system. This is primarily due to large credits associated
with steam generated during Claus processing, and fuel gas requirements for
tail gas treatment and incineration. Capital and annualized costs for the
integrated control system represent approximately 1.4% and 0.7% of the res-
pective costs for the uncontrolled methanol synthesis base plant.
The effects of feed gas flow rate and sulfur content upon the control
system's performance and cost have been discussed in Example 1. Similar con-
clusions can be drawn with respect to the effects of these parameters upon
the Claus/SCOT/incineration control system.
Example 3 - Claus Bulk Sulfur Removal with Wellman-Lord Tail Gas Treatment
This example illustrates treatment of the combined H2S-rich Rectisol
acid gas (Stream 216) and the flash gas from a methanol-based cyanide wash
unit (Stream 214b) in a Claus bulk sulfur removal unit with Wellman-Lord
tail gas treatment. The control system is presented schematically in Figure
4-5, and material flow estimates are presented in Table 4-10,. Assumptions
relating to Claus unit performance are identical to those presented in
Example 1, although in this example the Wellman-Lord recycle gas represents
an additional Claus feed stream.
212
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Integrated Control
TABLE 4-9. EXAMPLE 2 - COSTS OF INTEGRATED CLAUS BULK SULFUR REMOVAL
WITH SCOT TAIL GAS TREATMENT AND INCINERATION (1980 BASIS)
Total Capital Investment
2T6T
106 Dollars Base Plant*
Annualized Cost
106 Dollars
% of
Base Plant*
Claus Unit
9.2
0.44
SCOT Unit
6.3
1.9
Incineration
Unit
Total
0.42
15.9
1.44
0.63
3.0
0.86
Expressed as a percentage of the cost for an uncontrolled methanol synthesis
facility.
The Wellman-Lord unit is reported to be capable of reducing Claus
tail gas residual sulfur concentrations to about 250 ppmv (dry Claus tail
gas basis). In this example, the corresponding Wellman-Lord tail gas sulfur
concentration is about 150 ppmv. A methane-rich fuel gas has been used for
Claus tail gas incineration in this example consistent with the design upon
which the Wellman-Lord flow estimates are based. In an integrated facility
other fuels may be preferable. The resulting sulfur emission rate is about
0.2 kmol/hr S02 in the Wellman-Lord tail gas. Minor constituents such as
methanol, CO and NH3 which may be present in the Claus tail gas are expected
to be effectively destroyed during the incineration step of tail gas treatment,
Secondary waste streams from the control system are: (1) spent Claus
catalyst (Stream 402), (2) sour condensate (Stream 411), and (3) thiosulfate/
213
-------
WELLMAN-LORD RECYCLE GAS
ro
H2S-RICH
ACID GAS
FLASH GAS
FROM
METHANOL-
BASED
CYANIDE WASH
GLAUS BULK
SULFUR REMOVAL
T
CLAUS TAIL GAS
FUEL GAS
SPENT
CATALYST SULFUR
WELLMAN-LORD
TAIL GAS TREATMENT
(4U)
t
SOUR
CONDENSATE
WELLMAN-LORD
-TAIL GAS TO
ATMOSPHERE
THIOSULFATE/SULFATE
PURGE
Figure 4-5. Example 3 - Claus bulk sulfur removal with Wellman-Lord tail gas treatment
-------
TABLE 4-10. EXAMPLE 3 - MATERIAL FLOW ESTIMATES FOR INTEGRATED CONTROL EMPLOYING
CLAUS BULK SULFUR REMOVAL WITH WELLMAN-LORD TAIL GAS TREATMENT*
Cyanide Wash
H2$-Rich Flash Gas
Offgas Methanol Hash Case Wellman-Lord Combined Claus
Stream 216 Stream 214b Recycle Gas Feed Gas
kmol/hr Vol I kmol/hr Vol * kmol/hr Vol % kmol/hr Vol %
f\>
. — i
cn
H2
CO
co2
H,S
COS
cs2
so2
HCN
N2
°2
CH4
C2H6
Methanol
Sulfur
Total Dry Gas
H20
Total
kmol/hr
kg/hr
271
222
25
2
0.1
0 3
6
7
533
533
21544
0.49 3 0
4.4 26.4
50.8 1.2 7.3
41.6 2.3 44.7
4.7 0 56 3.4
0.4
218ppmv 13
636ppmv 2.1 12.9
1.2
1.3 0.3 1.8
16 13
7
16 19
524 938
0.49 0 09
4.4 0.8
272 48.4
229 40.7
26 4.6
2 0.4
100 13 2.3
2.4 0.4
6 1.1
7 1.2
562
7
569
23006
Claus .
Sulfur
Claus Tail Gas Stream 403 Fuel
kmol/hr Vol '. kg/hr kmol/hr
0.34 0.04
3.1 0.4
302 38.5
8.1 1.0
1.2 0.2
0.07 87ppmv
4.1 0.5
461 58.7
35
2.2
5 2 0.7
785 8260 37
232
1017 37
31270 8260 623
Sour
+ Wellman-Lord Condensate
Gas Feed Gas Stream 411
Vol % kmol/hr Vol 'i kg/hr
352 27.0
14 1.0
920 70.4
20 1.5
94
6
1306,
327 3297
1682
48651 3297
Hellman-Lord
Tail Gas
kmol/hr Vol "
352
0.2*
920
20
1292
143
1435
44484
27.2
147ppnw
71 2
1.5
The number of significant figures shown in some cases do not represent the degree of accuracy and are retained for material balance purposes only. Nonetheless,
slight imbalances do appear as result of numerical rounding Material flow estimates are based upon published data and engineering estimates. Performance data
and references are detailed in the Control Technology Appendices.
fA methane-rich fuel gas has been used in this example consistent with the design upon which the Wellman-Lord flow estimates are based. In an integrated
facility other fuel gases may be preferable.
*The S0? emission rate is based upon 250 ppmv total sulfur (Claus plant tail gas basis).
-------
Section 4
Acid Gas - Red. S/N, Org., CO
Integrated Control
sulfate by-product purge (Stream 412). As in Example 1, spent Claus catalyst
would be generated at an average rate of approximately 10 Mg/yr, assuming a
5-year catalyst life. The sour condensate typically has a pH value between
3
1 and 2 and would be generated at a rate of 3 m /hr. By-product purge con-
sists primarily of sodium salts such as sulfite, pyrosulfite, sulfate, and
thiosulfate with approximately 29% water. By-product purge is generated at a
rate of 90 kg/hr. In addition, Claus sulfur is produced at a rate of about
8.3 Mg/hr and may require disposal, depending upon sulfur quality and market
considerations.
Cost estimates for the Claus and Wellman-Lord units are summarized in
Table 4-11. Estimates are presented on two bases: (1) total capital invest-
ment and annualized costs, and (2) total capital investment and annualized
cost as percentages of the uncontrolled base plant cost for the methanol
synthesis case. All costs are presented on a 1980 basis. The total capital
investment was found to be similar for the Claus and Wellman-Lord units.
However, annualized Claus costs are estimated to be about 90% lower than
those for the Wellman-Lord unit. This is primarily due to large credits
associated with steam generated during Claus processing, and fuel gas require-
ments for tail gas incineration. Capital and annualized costs for the inte-
grated control system represent approximately 1.8% and 1.2% of the respective
costs for the uncontrolled methanol synthesis base plant.
The effects of feed gas flow rate and sulfur content upon the control
system's performance and cost have been discussed in Example "I. Similar con-
clusions can be drawn with respect to the effects of these parameters upon
the Claus/Wellman-Lord control system.
216
-------
TABLE 4-11. EXAMPLE 3 - COSTS OF INTEGRATED CLAUS BULK SULFUR REMOVAL
WITH WELLMAN-LORD TAIL GAS TREATMENT (1980 BASIS)
Total Capital Investment
, % of
10° Dollars Base Plant*
Annuali zed Cost
, % of
10° Dollars Base Plant*
Claus Unit 9.2 0.44
Wellman-Lord 11 3.7
Unit
Total 20 1.8 4.1 1.2
*
Expressed as a percentage of the cost for an uncontrolled methanol synthesis
facility.
217
-------
Section 4
Combustion Gases
4.1.2 Combustion Gases
Combustion of many types of fossil fuels produces a gas stream that
contains undesirable amounts of S09, NO, and/or particulates. Sulfur dioxide
c. X
is formed rapidly in the combustion process when sulfur contained in the fuel
reacts with oxygen in the air. Variations in the combustion process are not
effective in reducing S02 emissions. Rather, sulfur must be removed from
the fuel or, once formed, S02 must be removed from the exhaust gas.
The generation of NO from air-fed fuel combustion processes occurs by
/\
two separate mechanisms, namely thermal NO formation and fuel NOV formation.
X X
Thermal NOV results from the thermal fixation of molecular nitrogen and
A
oxygen in the combustion air and is sensitive to flame temperatures and to
local concentrations of oxygen. Fuel NOV is created from the oxidation of
X
chemically-bound nitrogen in the fuel being combusted. Fuel NO formation
X
is strongly affected by the rate of mixing of the fuel and air and by the
local oxygen concentration. Approximately 95% of nitrogen oxides from com-
bustion are emitted as NO (83,84).
The particulates generated during combustion result mainly from the ash
content of the fuel. The magnitude of these emissions in the flue gas is a
function of combustion unit design and ash content of the fuel. Particulate
emissions are very low when oil-based fuels are used and are negligible with
gaseous fuels.
Combustion gases will also contain CO and very small amounts of unburned
organics (including polycyclic compounds). Concentrations of these compon-
ents are a function of both the fuel burned and the design and operation of
the combustion unit. Trace elements present in the fuel may also be present
in flue gases. Ordinarily, however, no controls are applied to combustion
gases specifically for control of CO, organics, or trace elements. Some
degree of trace elements control is achieved as part of particulate removal.
218
-------
Section 4
Combustion Gases
The major sources of flue gases from a K-T based indirect liquefaction
facility are the power boiler, process heaters, gas turbines, and secondary
combustion gas streams from other media. In this section control technologies
applicable to the removal of nitrogen oxides, particulates, and sulfur oxides
from such stationary sources are discussed.
219
-------
Section 4
Combustion Gases
NOV Control
X
NOX Control
NOX pollution control techniques are of two types: (1) those that
limit nitrogen conversion to NOY by modifying combustion characteristics and
/\
(2) flue gas treatment techniques (i.e., removal of NO after it is formed).
X
Combustion modification techniques are the most widely used techniques.
They can achieve from 25 to 60% reduction in NOV emissions. Some of the
X
common combustion modification techniques are (1) low excess air, (2) staged
combustion, (3) flue gas recirculation, (4) reduced load, (5) low NO burners,
X
and (6) ammonia injection. The key features and unit costs of these techni-
ques are discussed in Table 4-12.
Low excess air level in the furnace has generally been found to be an
effective method for NO control. In this technique, the combustion air is
/\
reduced to the minimum amount required for complete combustion while main-
taining acceptable furnace cleanliness and steam temperature. With less
oxygen available in the flame zone, both thermal and fuel NO formation are
X
reduced. In addition, the reduced air flow lowers the quantity of flue gas
released resulting in an improvement in boiler efficiency.
Staged combustion produces overall fuel-rich conditions during the first
couple of seconds and promotes the reduction of NO to N~. Various methods
to achieve this are available. Overfire Air and Burners Out of Service are
two techniques generally used on coal fired boilers. Details regarding their
performance and applicability are provided in Table 4-12.
Flue gas recirculation (FGR) consists of recycling a portion of the flue
gas back to the primary combustion zone. This reduces NO formation by lower-
ing the bulk gas temperature and oxygen concentration. This technique,
however, is effective only on oil and gas fired boilers.
Load reduction can be used to decrease NOV emissions. Thermal NOV for-
X X
nation generally increases as the volumetric heat release rate or combustion
220
-------
TABLE 4-12. COMBUSTION MODIFICATION TECHNIQUES FOR NOX CONTROL
Control
Technique
Description
of
Technique
Efficiency
(as % NOX
Reduction)
Type of
Fuel Fired
Range of
Application
Low Excess Air
(LEA)
ro
Staged Combustion
Overfire Air
Injection (OFA)
Staged
Combustion Air
(LEA + OFA)
Combustion air is reduced
to the minimum amount re-
quired for complete com-
bustion while maintaining
proper stream temperature.
Injection of air above the
top burner level through
OFA ports together with a
reduction in air flow to
the burners (staged com-
bustion).
Reduction of under grate
air flow and increase of
overfire air flow.
Fuel rich firing burners
with secondary combustion
air ports.
Injection of secondary
air downstream of the
burner(s) in the direction
of the flue gas path.
0 - 25
5 - 25
0 - 28
0-24
5 - 35
5 - 30
5 - 25
20 - 50
17 - 44
5 - 46
Pulverized coal
Stoker coal
Residual oil
Distillate oil
Natural gas
Pulverized coal
Stoker coal
Residual oil
Distillate oil
Natural gas
Excess Q£ lowered to
5.2% on the average.
Excess 02 limited to
5-6% minimum.
-\
Excess 02 can be
reduced to <3%.
Excess 02 can be
reduced to <3%.
Burner 02 can be as
low as stoichiometric
Excess 02 limited to
5% minimum.
70-90% burner stoichio-
metry can be used with
proper burner installa-
tion.
70-90% burner stoichio-
metries can be main-
tained. Applicable to
all units, however,
requires extensive equip-
ment modification.
(Continued)
-------
TABLE 4-12. (CONTINUED)
Control
Technique
Low Excess Air
(LEA)
Type of
Fuel Fired
Pulverized
Coal
Stage of
Development
Available but implemented
on a limited basis only.
Secondary
Waste
None
Capital :
Operating:
Cost
$440 to $550/MW heat input
0 to 8 mills/103 kg steam
Limitations and Comments
Limited by increase in CO,
HC, and particulate emis-
sions. Increase in boiler
efficiency may be achieved
as a benefit.
Stoker coal
Residual Oil
Distillate Oil
Natural Gas
IV.
no
Staged Combustion
Overfire Air
Injection (OFA)
Staged
Combustion
Air (LEA + OFA)
Pulverized
Coal
Stoker Coal
Residual Oil
Distillate Oil
Natural Gas
Available now but need
R&D on lower limit of
air.
Available.
Available.
Available but not
demonstrated.
Most stokers have OFA
ports as smoke control
but may need better air
flow control devices.
Technique is applicable
on packaged and field-
erected units. However,
not commercially avail-
able for all design types.
Technique is still
experimental especially
for small firetube and
watertube units.
None
None
None
None
None
None
None
Capital:
Operating:
Capital:
Operating:
Capital:
Operating:
Capital:
Operating:
Capital:
Operating:
Capital:
Operating:
Capital:
Operating:
$600 to $1850/MW heat input
13 to 57 mills/103 kg steam
$460 to $2400/MW of heat
input
<83 mills/103 kg steam
$580/MW of heat input
$800 to $940/MW heat input
80 to 85 mills/103 kg steam
$600 to $800/MW heat input
24 to 32 mills/103 kg steam
$870 to $5150/MW of heat
input
123 to 370 mills/103 kg
steam
$1070/MW of heat input
117 mills/ICr kg steam
Danger of overheating grate,
clinker formation, corrosion,
and high CO emissions.
Added benefits include in-
crease in boiler efficiency.
Limited by increase in CO,
HC, and TSP emissions.
Generally practical because
of increase in boiler effici-
ency. Best NOX reductions
reported for large multi-
burner units.
Limited by possible increase
in slagging and corrosion.
Excess air may be required
to ensure complete combus-
tion thereby decreasing
efficiency.
Overheating grate, corrosion,
and high CO emissions can
occur if under grate air
flow is reduced below accept-
able level as in LEA.
Best implemented on new units.
Retrofit is probably not fea-
sible for most units espe-
cially packaged ones.
Found to be less effective
on firetube boilers than
watertube boilers. Generally
less effective for gas-fired
units.
(Continued)~
-------
TABLE 4-12. (CONTINUED)
Control
Technioup
Description
of
Technique
Efficiency
(as % NOX
Reduction)
Type of
Fuel Fired
Range of
Application
Staged Combustion
Air and Fuel
Rich Firing
Flue Gas
Recirculation
(FGR)
One or more burners
fired on air only.
Remainder of burners
firing fuel rich.
Recirculation of the
flue gas to the burner
windbox.
rv>
ro
CO
27 - 39
0 - 20
15-30
58 - 73
48 - 86
Pulverized coal
Pulverized coal
Residual oil
Distillate oil
Natural gas
Reduced Load
Reduction of fuel and
air flow to the burner.
Up to 45%
Average 15%
Pulverized coal
Stoker coal
Boilers must have a
minimum of 4 burners,
or designed with excess
burners.
A maximum of 25% of the
flue gas can be recir-
culated.
Up to 25-30% of flue gas
recycled. Can be imple-
mented on all design
types.
Flue gas recirculation
rates possible up to 45%.
Technique is applicable
to all boiler types except
ones equipped with ring
burners.
Load may be reduced to
25% of capacity.
Load may be reduced to
25%.
(Continued)
-------
TABLE 4-12.
(CONTINUED)
Control
Technique
Staged Combustion
Air and Fuel
Rich Firing
Type of
Fuel Fired
Pulverized
Coal
Stage of
Development
Available, but engineering
refinement is needed prior
to implementation.
Secondary
Waste Costs
None Not available
Limitations and Comments
Load reduction required in
most cases. Possible in-
creased slagging and corro-
Flue Gas
Recirculation
Pulverized
Coal
ro
ro
Reduced Load
Residual Oil
Distillate Oil
Natural Gas
Pulverized
Coal
Stoker Coal
Not offered because the
method is comparatively
ineffective.
None
Not available
Available. Requires
extensive modifications
to the burner and wind-
box .
Available now. Best suited
for new boilers. Retrofit
application would result
in extensive burner modi-
fications.
Available but not imple-
mented because of negative
operational impacts.
Available.
None
None
None
None
Capital: $1070 to S5150/MW of
heat input
Operating: 196 to 438 mills/103 kg
of steam
Capital:
$870 to $1070/MW of
heat input
Not available
Not available
sion. New boiler design
will incorporate the re-
quired number of burners.
Flue gas recirculation
lowers the bulk furnace gas
temperatures and reduces C>2
concentration in the com-
bustion zone. Requires in-
stallation of flue gas re-
circulation ducts, fans,
insulation, etc. flay cause
combustion instability.
Best suited for new units.
Costly to retrofit. Possible
flame instability at high
FGR rates.
Flame instability problem
is not severe except for
ring burners. Minor burner
modifications can guarantee
stable flames. Most effec-
tive on watertube units.
Best used with increase in
firebox size for new boilers.
Load reduction may not be
effective because of increase
in excess 0£.
Only for stokers that can
reduce load without in-
creasing excess air. Not
a desirable technique be-
cause of loss in boiler
efficiency.
(Continued)
-------
TABLE 4-12. (CONTINUED)
ro
rv>
en
Description Efficiency
Control of (as % NOx Type of
Technique Technique Reduction) Fuel Fired
Reduced Load 33% decrease to Residual oil
25% increase*
31% decrease to Distillate oil
17% increase*
32% decrease to Natural gas
82% increase*
Low NOX Burners New burner designed 45 - 60 Pulverized coal
to utilize controlled
air-fuel mixture.
20 - 50 Residual oil
20 - 50 Distillate oil
all boilers.
20 - 50 Natural gas
NHj Injection Injection of NH3 into 40 - 60 Pulverized coal
convective section of
the boiler.
40 - 70 Residual oil
40 - 70 Distillate oil
40 - 70 Natural gas
Range of
Application
Applicable to all boiler
types and sizes. Load
can be reduced to 25% of
maximum.
Tests to 20% of rated
capacity. Applicable to
all units.
Prototypes are limited
to size ranges >30 MW.
New burners described
generally applicable to
all boilers.
More specific information
needed.
NH3 injection rate limited
to NH3
NO = ]'5
Applicable for large
package and field-erected
watertube boilers.
Not feasible for firetube
boilers.
Apparent increases in NOX are indicated by limited test data and need confirmation.
Such increases could be due to higher oxygen levels at reduced loads relative to oxygen
levels at design load.
(Continued)
-------
FABLE 4-12. (CONTINUED)
Control
Technique
Reduced Load
Type of
Fuel Fired
Residual Oil
Distillate Oil
Stage of
Development
Available now as a retrofit
application. Better imple-
mentation with improved
firebox design.
Secondary
Waste Costs
None Not Available
Limitations and Comments
Technique not effective when
it necessitates an increase
in excess 02 levels, RL is
possible to implement in new
designs as reduced combustion
intensity (enlarged furnace
plan area).
Low NO Burners
Natural Gas Technique available. How-
ever, retrofit application
is not feasible due to
initial low load factor of
industrial units.
Pulverized Development stage prototypes
Coal are available from major
boiler manufacturers.
None
None
Not Available
Capital: S800 to S940/HW of heat
input
Operating: 80 to 85 mills/103 kg
steam
ro
ro
01
Residual Oil Commercially offered but
Distillate Oil not demonstrated.
Natural Gas
Pulverized
Coal
Commercially offered but
not demonstrated.
Commercially offered but
not demonstrated.
None
None
Ammom urn
Bisulfate
Capital. S860 to S5150/MW of heat
input
Operating: 123 to 376 mills/10J kg
steam
Capital- S860 to S1070/MW of heat
input ,
Operating: 106 to 117 mills/10 kg
steam
Capital: S4800/MW of beat input
Operating: 247 mills/103 kg steam
Residual Oil
Distillate Oil
Commercially offered but
not demonstrated.
Ammom urn
Bisulfate
Capital: S4940 to S9770/MW of
heat input
Operating 266 to 433 mills/103 kg
of steam
Not available for natural gas
Least effective on firetube
boilers because of lower
combustion intensity. Appli-
cable for new watertube units
with increased firebox size.
Low NOX burners could maintain
the furnace in an oxidizing
environment to minimize slag-
ging and reduce the potential
for furnace corrosion. More
complete carbon utilization
results because of better coal/
air mixing in the furnace.
Lower 02 requirements may be
obtained with all the combus-
tion air admitted through the
burners.
Specific emissions data from
oil fired industrial boilers
equipped with LNB are lacking.
Specific emissions data from
gas fired industrial boilers
equipped with LNB are lacking.
Limited by furnace geometry.
Performance is sensitive to
flue gas temperature and
residence time at optimum
temperatures. By-product
emissions such as ammonium
bisulfate could cause opera-
tional problems.
Some increased maintenance
of air heater/economizer
parts might be necessary
when burning high sulfur oil.
Technique is very costly.
Should have fewer problems
when firing natural gas.
Natural Gas
Not available.
-------
Section 4
Combustion Gases
N0v Control
A
intensity increases. Reduced combustion intensity can be brought about by
load reduction by either derating the boiler or using an enlarged firebox.
Low NOV burners have been developed primarily for reducing NO emissions
A A
from utility boilers. Their principal characteristics are reduced flame tur-
bulance, delayed fuel air mixing, and establishment of fuel-rich zones where
combustion initially takes place. It is now standard practice for all utility
boilers to come equipped with low NO burners.
X
The process of injecting ammonia was developed by Exxon Research and
Engineering Company. This technique acts by reducing NO to elemental nitro-
gen and oxygen with NhU at flue gas temperatures ranging from approximately
1070 to 1290K. However, optimal NO reduction occurs over a very narrow
temperature range, around 1240K +50K.
The cost of combustion modification techniques for controlling NO emis-
A
sions depends upon (1) the additional hardware required and (2) any changes
in operational procedures that may increase the cost of steam production.
Cost estimates for combustion modification techniques are provided in Table
4-12. For more detailed information other EPA studies should be consulted
(83,84).
Flue gas treatment techniques have been proposed for control of NO
A
emissions to levels significantly below those achievable by combustion modi-
fication techniques. Although large scale flue gas treatment schemes have
not been proved commercially in the U.S., these techniques are being applied
in Japan. The key features of some of these processes are provided in Table
4-13.
Of the processes listed in Table 4-13, the Selective Catalytic Reduction
(SCR) system, using ammonia to react with NO , is perhaps the most promising.
X
SCR commercial units are being applied on many gas and oil-fired boilers in
Japan. Full scale tests of coal-fired boilers are scheduled in mid-1982;
two units are operating and several more are due online. Considerable
227
-------
TABLE 4-13.
NOV FLUE GAS TREATMENT CONTROL ALTERNATIVES FOR BOILERS*
A
Control
Technique
Description
of
Technique
Principle
of
Operation
Efficiency
(% as NOX
Reduction)
Appl
icability
Selective
Catalytic
Reduction (SCR)
- Fixed Packed
Bed Reactors
Utilizes NH3 to
selectively reduce
NOX to N2.
Reactor contains
ring shaped catalyst
pellets packed in
fixed bed.
Up to 90%
Applicable only
to flue gas streams
containing particu-
late emissions of
less than 20 mg/Nm3.
ro
ro
CO
- Moving Bed
Reactors
- Parallel Flow
Reactor
Absorption-
Oxidation
SCR - NOX/SOX
Removal
Utilizes NH3 to
selectively reduce
NOX to N2.
Utilizes NH3 to
selectively reduce
NOX to N2.
Removes NOX from
flue gas by absorb-
ing the NO or NOX
into a solution
containing an oxi-
dant which converts
the NOX to a
nitrate salt.
Utilizes NH3 to
catalytically reduce
NOX after SOX is
absorbed and reacted
with catalyst.^
Reactor contains
catalyst (rings or
pellets) gravity-fed
mechanically-screened,
and returned to reactor.
Reactor contains a
special catalyst
arrangement (honey-
comb, parallel plate,
or tubes).
Use of gas/liquid
contactors. Perfo-
rated plate and packed
towers accomplish NOX
absorption by generat-
ing high gas/liquid
ratio.
Reactor and catalyst
removes both NOX and
S02» uniquely designed
parallel flow reactor
used to avoid parti-
culate problems.
Up to 90%
Up to 90%
The relative
insolubility of
NO in water will
prohibit a high
efficiency.
80% NOX reduc-
tion, 90% SOX
reduction
(theoretical).
Applicable only to
flue gas streams con-
taining particulates
at less than 1 g/Nm3.
Testing currently
under way for high
particulate (>1 g/Nm3)
flue gas.
No published informa-
tion available.
Should be applicable
to high particulate
flue gas.
(Continued)
-------
TABLE 4-13. (CONTINUED)
Control
Technique
SCR - Fixed
Packed Bed
Reactors
SCR - Moving
Bed Reactors
ro
ro
SCR - Parallel
Flow Reactor
Absorption-
Oxidation
SCR - NOX/SOX
Removal
Stage of
Development
Secondary Wastes
Costs
Limitations and Comments
Commercially avail-
able only for natural
gas-fired boilers at
this time.
Spent catalyst
Has been applied in
Japan to several oil-
fired industrial and
utility boilers.
Had been applied in
Japan to several oil-
fired industrial and
utility boilers.
Applicability to coal-
fired boilers cur-
rently being tested
by EPA.
No coal-fired tests
have been made.
Spent catalyst
Spent catalyst
N05 salts in
wastewater
20MW estimate:
Capital: $155/kW (coal)
$84/kW (oil)
$32/kW (gas)
Operating: 2.5 mills/kWh (coal)
2.3 mills/kWh (oil)
1.4 mills/kWh (gas)
20MW estimate:
Capital: $110/kW (coal)
$84/kW (oil)
Operating: 2.4 mills/kWh (coal)
2.5 mills/kWh (oil)
20MW estimate:
Capital: $53/kW (coal)
$46/kW (oil)
Operating: 1.8 mills/kWh (coal]
20MW estimate
Capital: $597/kW (coal
Operating: 9.6 mills/kWh (coal)
No continuous coal-
fired NOX removal
test data for NOX/
SOX systems are
available.
Spent catalyst
20MW estimates
Capital: $567/kW (coal & oil)
Operating: 6 mills/kWh (coal
& oi 1)
Although it is possible
to install a hot ESP to
reduce the particulate
level to 20 mg/Nm3, this
is expensive and not always
effective. Thus, fixed bed SCR
systems are not considered
for application to coal-
fired boilers.
Although it is possible
to install a hot ESP to
reduce the particulate
level to 1 g/Nm3, this is
expensive and not always
effective. Thus, moving bed
SCR systems are not con-
sidered for application
to coal-fired boilers.
Greatly reduces particulate
impaction as gas flow is
parallel to catalyst surface.
Unreacted NH^ downstream can
react with S02 or 503 to
form ammonium bisulfate or
the NH3 could enter FGD and
ESP equipment.
The presence of particulates
requires a prescrubber.
The presence of S0£ requires
FGD pretreatment. Increased
NOX concentration requires
a larger column height and
increased oxidant concentra-
tion. Nitrate salts formed
as a secondary pollutant.
System is not affected by
changes in the boiler gas
flow rate or particulate
concentrations. Changes in
NOX concentration because
of boiler load changes may
be compensated for by con-
ventional control system
used with the NH3 injection
equipment.
(Continued)
-------
TABLF 4-13. (CONTINUED)
Control
Technique
Description
of
Technique
Principle
of
Operation
Efficiency
(% as NOX
Reduction)
Applicability
ro
CO
o
Adsorption Adsorbed NOX is reduced
N0x/S0v to N2 while S02 is re-
Removal duced and condensed to
elemental S.
Electron Beam
Radiation
N0x/S0x
Removal
A dry process that
utilizes an electron
beam to bombard the
flue gas, thereby
removing NOX and SOo-
The adsorption process
removes NOx and S02
from flue gas by
absorbing them onto
a special activated
char.
Flue gas is taken from
the boiler air pre-
heater and passed
through a cold ESP to
remove particulate.
NH3 is added and the
gas is then bombarded
with an electron beam.
40-60% NOX
reduction and
80-95% SOX
reduction.
Removal effici-
ency will de-
crease as NOX
and S02 increase.
May be applicable
to handle coal.
By-product (ammonium
nitrate and ammonium
sulfate powder) treat-
ment technology needs
to be more fully de-
veloped before commer-
cialization.
Adsorption-
Reduction
NOX/SOX
Removal
Oxidation-
Absorption-
Reduction
NOX/SOX
Oxidation-
Absorption
Simultaneously removes
NOX and S02 from flue
gas by absorbing them
into a scrubbing
solution.
Simultaneously removes
NOX and S02 from flue
gas by oxidizing NO to
N02 and then absorbing
N02 and S02 into a
scrubbing solution.
Excess 03 is used to
selectively oxidize
NOX to N205.
Based on the use of
Chelating compounds
complexed with iron to
"catalyze" the absorp-
tion of NOX.
Based on the use of gas-
phase oxidants, either
03 or C102, to selec-
tively oxidize NO to
N02.
formed by oxida-
tion is absorbed into
aqueous solution and
concentrated to form
a 60% HN03 by-product.
60-70% NOX reduc-
tion and 90% S02
reduction.
90% NOX reduction
and 95% S02 reduc-
tion for oil-fired
tests.
Not available.
Applicable only to
high sulfur coals.
Not applicable to low
sulfur coals.
May be applicable to
handle high particu-
late flue gas.
(Continued)
-------
TABLE 4-13. (CONTINUED)
Control
Technique
Stage of
Development
Secondary Wastes
Costs
Limitations and Comments
ro
co
Adsorption
NOX/SOX
Removal
Electron Beam
•Radiation
NOx/SOx
Removal
Adsorption-
Reduction
NOX/SOX
Removal
Oxidation-
Absorption-
Reduction
NOX/SOX
Oxidation-
Absorption
Presently in the
prototype unit stage
of development.
No coal-fired tests
have been performed
at this time.
Preliminary testing
stage of development.
Prototype stage of
development. No
coal-fired flue gas
tests have been per-
formed at this time.
One coal-fired test
has been performed
with no published
information.
Ash for disposal
Ammonium nitrates
and sulfates
Sulfate and
nitrate salts
and gypsum
N05 or N-S salts
or NH3 based com-
pounds in waste-
water
N0§ salts in
wastewater gypsum
20MW estimates:
Capital: $257/kW (coal)
Operating: 2.7 mills/kWh (coal)
20MW estimates:
Capital: $241/kW (coal)
Operating: N/A (coal)
20I1W estimates:
Capital: $493/kW (coal)
$223/kW (oil)
Operating: 8.8 mills/kWh (coal
6.4 mills/kWh (oil)
20MW estimates:
Capital: $278/kW (oil)
Operating: 7.6 mills/kWh
Not available
Very complex process. Numerous
process steps involve hot
solids handling with numerous
mechanical problems possible.
N0x/S02 removal will drop off
drastically at low radiation
doses based on oil-fired pilot
tests. Sulfate and nitrate
salts as well as other ionic
species formed as by-products.
Requires large absorbers with
high liquid rates. Absorbing
solution is highly corrosive;
and sulfate and nitrate salts
formed as secondary pollutants.
Costly gas-phase oxidants
create secondary wastewater
pollution problems. The use
of C102 introduces a chloride
pollutant problem.
Production of nitrate salts
poses a potential secondary
pollution problem. Corrosion
problems.
-------
Section 4
Combustion Gases
NO Control
X
advances have been made; however, some significant technical and economic
questions must be answered before widespread application of SCR units can
occur.
232
-------
Section 4
Combustion Gases
Particulate Removal
Particulate Removal
The choice of the particulate collection equipment depends upon a number
of factors: the properties of the materials such as particle size and physi-
cal and chemical characteristics; the concentration and volume of the parti-
culate to be handled; the temperature and humidity of the gaseous medium;
and most importantly the collection efficiency required.
There are four basic types of particulate collection equipment: (1)
cyclones, (2) fabric filters/baghouses, (3) venturi scrubbers, and (4) electro-
static precipitators. The key features and unit costs of these collection
devices are presented in Table 4-14.
Cyclones are generally employed for the removal of bulk particulates
(usually greater than 4 microns in size) and, in many cases, precede other
control devices. The unit installed costs of cyclones are relatively low -
3
approximately $212/m per minute.
Baghouses have very high particulate removal efficiencies and lend
themselves to applications involving small or intermittent gas flows. Bag-
houses, however, have high pressure drops (in comparison with electrostatic
precipitators) and cannot ordinarily handle wet gases, gases containing oily
materials, or gases having temperatures in excess of 573K. The unit installed
cost for a typical baghouse is about $300/m per minute.
Venturi scrubbers can generally handle gases having temperatures higher
than those which can be handled by fabric filters, can operate at high pres-
sure, can tolerate wet and dry gases, and can be very efficient for the
removal of submicron particles. In contrast to devices in which the parti-
cles are collected in dry form, venturi scrubbers generate a wet slurry
which is more voluminous and generally more difficult to dispose of. Unit
•3
installed costs for venturi scrubbers are approximately $250/m per minute.
233
-------
TABLE 4-14. KEY FEATURES OF PARTICULATE COLLECTION EQUIPMENT
Control
Device Operating Principle
High Parti culates removed from gas
Efficiency stream by imparting a centri-
Cyclone fugal force. The inertia of
to the walls where they fall
to the bottom of the cyclone
for removal .
Fabric Filter Fabric filter material is
(Baghouse) arranged in a tubular shape
with the particulate laden gas
stream passing through the
filter. Particulate removal
primarily results from the
buildup of collected material
on the dirty-air side of the
filter. The filter is per-
iodically cleaned by mech-
anical shaking or a pressur-
ized reverse air flow.
ro
^ Venturi Removal of particulates from
Scrubber a gas stream by intimate con-
tact with multiple jet streams
of scrubbing water and drop-
lets. Agglomerated particu-
lates are subsequently removed
in a centrifugal and/or mist
el imi nator
Electrostatic A negative electrical charge
Precipitator is imparted to the particu-
lates and they are collected
on positively charged plates.
Collected material is removed
by periodically rapping or
vibrating the collection
plates .
Removal Loading
Efficiency Range Limitation
(weight -) (g/Nm3)
50 to 80= for >5 _m. >2.4
80 to 95 for 5 to
20 urn.
98.5 to 99.5. for >0.24
0.25 to 0.5 ,m.
99 to 99.5: for
0.75 to 1 jn.
99.5 to 99.9' for
3 Lm.
99.95i for 3 urn.
50 to 92.5- for >0.5
0.25 .m.
60 to 98'- for
0.5 jn.
70 to 99', for
0.75 urn.
90 to 99.6?, for
3 um.
95 to 99'; for 0.24
0.1 LIU.
90 to 96'- for
0.5 jin.
95 to 99v for
1 utn.
99 to 99.9', for
5 jin.
Pressure
Drop Range Reliability
(cm FLO) or Other Limitations
7 to 20 Cannot effectively remove
parti culates smaller than
5 to 25 Plugging problems will
result if condensation
occurs on filter media or
if hygroscooic-matenal is
collected. Temperture
limit varies with type of
filter media used, maximum
is 560 K.
13 to 250 Reliability may be limited
by scaling, fouling, or
corrosion. Scrubbing
liquor blowdown may require
treatment or contain poten-
tially valuable material
not directly recoverable.
0.5 to 2 5 Not applicable to combus-
tible or potentially
explosive mixtures. Par-
ticulates to be collected
must have suitable elec-
trical resistivity to
facilitate efficient re-
moval. Used in low pres-
to gas streams with temper-
atures below 700 K.
Secondary
Waste
Collected
particulates
Collected
particulates
Scrubbing
liquor blow-
down and
wet slurry
Collected
particulates
Installed
Costs
About $212/m3/min
for total in-
stalled system.
About S282/m3/m1n
for total system.
About S250/m3/min
(increasing cost
with increasing
removal effici-
ency).
About $250 to
S530/m3/min
(increasing cost
with increasing
removal efficiency).
General Comments
High reliability due to
simple operating prin-
ciple with no moving
parts . Low energy
requi rements .
High particulate removal
efficiency. High instal-
lation cost. Large scale
requl red.
High particulate removal
efficiency. Capable of
treating streams with
wide ranges in tempera-
ture (no limitation for
high temperatures),
pressure , and gas compo-
sition. High efficiencies
require high energy con-
sumption.
High particulate removal
efficiency, especially
the sub-micron range.
High capital and instal-
lation cost. Very low
pressure drop. Suitable
for high temperature or
large volume applications
High electrical consump-
tion. Sensitive to parti-
culate resistivity.
-------
Section 4
Combustion Gases
Particulate Removal
Electrostatic precipitators are high efficiency participate removal
devices, have low pressure drops, are capable of handling large volumes of
gases, and can tolerate high feed gas temperatures. Electrostatic precipi-
tators, however, are not generally suitable for applications to gases above
atmospheric pressure and are not economical for treating small or intermittent
gas flows (such as those resulting from material handling dust collection
systems). Unit installed costs range from $250 to $530/m per minute.
235
-------
Section 4
Combustion Gases
S02 Removal
Removal
Several flue gas desulfurization (FGD) processes are commercially avail-
able. They are basically of two types: (1) throwaway systems which produce a
waste sludge, and (2) regenerable systems which produce a usable sulfur by-
product and regenerate the sorbent. Common examples of throwaway systems
are the lime/limestone system, dry scrubbing, double alkali scrubbing, fly
ash alkalinity scrubbing, and Chiyoda Thoroughbred 121. Typical examples
of regenerable systems are the Wellman-Lord and magnesium oxide processes.
Key features of some of these scrubbing systems are presented in Table 4-15.
The lime/limestone scrubbers are the most commonly employed throwaway
systems for electric utility applications. In the lime/limestone process,
solid lime or limestone is pulverized and mixed with water to form a scrubber
liquor which contacts the flue gas in an absorption tower where calcium sul-
fate and calcium sulfite are formed. The resulting slurry is removed from
the system and treated, and the sludge resulting from slurry treatment is dis-
posed of. Scrubbing solution is recovered and recycled to the absorption
tower. S02 removal efficiencies can approach 90% by carefully balancing the
many chemical reaction parameters involved in the lime/limestone FGD process.
Of the throwaway FGD systems available, lime/limestone offers the least com-
plex system and equipment, the easiest pH control, and the cheapest raw
materials. Operating experience has indicated that careful attention to system
control is important for successful operation of lime/ limes tone FGD systems.
Of all the commercially available regenerable FGD systems, Wellman-Lord
is the most extensively used. A venturi prescrubber often precedes the
Wellman-Lord absorber to remove residual particulates from the flue gas and
avoid ash accumulation in the absorber. Sulfur dioxide is absorbed by an
alkaline sodium sulfite solution to produce primarily sodium bisulfite. This
bisulfite-rich solution is then pumped to a forced-circulation vacuum
236
-------
TABLE 4-15. KEY FEATURES OF S02 REMOVAL PROCESSES
Feature
Lime/Limestone
Scrubbing
Double Alkali Scrubbing
Chiyoda
Thoroughbred 121
Wellman-Lord
Dry Scrubbing
Alkalinity Scrubbing
Principle
Feed Stream
Requirements
ro
Liquid phase absorption
of S02 in lime or lime-
stone slurry.
Particulates can be re-
moved in an ESP or
fabric filter to achieve
99*% at lowest energy
consumption. Fly ash
may be removed in a
venturi where the fly
ash contains signifi-
cant alkalinity. A
scrubber can be
used for both high
particulate and SO?
removal.
Liquid phase absorption
of SO? in a sodium
hydroxide, sodium sulfite,
sodium bisulfite, sodium
sulfate,and sodium carbon-
ate solution. Regenera-
tion of the sodium sul-
fite/bisulfite with lime
in a reactor. A dilute
mode can be used for con-
centrations of 200-1500
ppm SO? and where less
than 252 oxidation of sul-
fite to sulfate occurs.
Concentrated mode can be
used for concentrations
of 1000-8000 ppm S02.
Excessive particulates
should be removed in an
ESP, fabric filter or
venturi. 02 should be
less than 7% for con-
centrated mode.
Liquid absorption of SOp
in a single vessel,
where limestone addition
and dissolution, air oxi-
dation, and gypsum preci-
pitation occur.
Particulates and chlorides
should be removed from in-
let flue gas if byproduct
gypsum is to be sold.
Liquid phase absorption
of SO? in a sodium bi-
sulfite, sodium sulfite,
and sodium carbonate
solution. A rich S02
is produced by evapora-
tion, which is then pro-
cessed in a Claus unit
to produce elemental
sulfur or in a sulfuric
acid plant.
Particulates and chlo-
rides must be removed
from flue gas.
Process involves the use
of a spray dryer which
contacts the flue gas
with an aqueous alkaline
material and produces a
dry product. System
involves two stages: 1st
stage-spray dryer; 2nd
stage-dry particulate
collector which removes
flue ash and reaction
product fron flue gas.
Inlet S02 concentration
should not exceed 1000
ppmv.
Process involves a two
stage venturi-spray tower
absorber utilizing the
fly ash alkalinity for
502 removal. Hydrated
dolomitic lime, (Hg(OH>2
and Ca(OH)2> is also used
to achieve an outlet S02
of 43 ng/J.
Venturi is used to remove
particulates and a portion
of S02-
Absorbent Slaked lime or 200-300
mesh limestone 6-12%
slurry circulated.
Product/ Gypsum can be produced
Waste with forced oxidation.
Calcium sulfate/sul-
fite can be produced
with 50-70% solids.
Efficiency 90?; removal can be
obtained generally for
low and high sulfur
coals. Higher removals
can be obtained with
higher L/6 and pressure
drop, and to some extent
scrubber type. 90Z and
greater can be obtained
for low sulfur coals.
95% removal for high
sulfur coals when adipic
acid is used. Commer-
cially demonstrated in
over 30 FGD units.
Sodium hydroxide, sodium
sulfite/bisulfite, and
a small amount of sodium
sulfate.
Filter cake with 60-70i
solids, primarily calcium
sulfite and calcium sul-
fate.
90-99' removal can be
obtained for low and high
sulfur coals. Concentrated
mode has been demonstrated
at Louisville Gas &
Electric's 200 MW coal
fired boiler. Smaller
industrial units (General
Motors) have been operated
in the di1ute mode.
Limestone slurry.
Gypsum (CaSO^HjO) with
less than 20" moisture
content.
90/V SO^ removal or outlet
S02 equal to 300 ppmv.
Process has been demon-
strated at Gulf Power's
Scholz station - 20MW
prototype.
Concentrated sodium
sulfite/bisulfi te.
Concentrated S02 p
stream (90% S02).
Process has been demon-
strated in a NIPSCO 11 BMW
coal fired boiler. Can
remove up to 95% S02-
Lime slurry or sodium
carbonate solution.
Sodium sulfite-sodiurn
sulfate, calcium sulfite/
sulfate.
70,, 502 removal using
lime as absorbent. 80^
S02 removal using sodium
absorbent.
Fly ash alkalinity and
hydrated lime (calcium and
magnesium hydroxide).
Sludge consists of fly ash,
gypsum (CaSOa-2H20),
Mg(OH>2, small amount of
calcium sulfite.
85-90% removal of S02
commercially demonstrated
in Montana Power's Colstrip
1 & 2. 96+* expected in
Colstrip units 3 & 4.
(Continued)
-------
TABLE 4-15. (CONTINUED)
Process
Feature
Cost*
Lime/Limestone
Scrubbi ng
Capital - S90/kW to
S185AW.
Chiyoda
Double Alkali Scrubbing Thoroughbred 121
Capital - $101/kW to Capital - 5160/kW.
S163/kW.
Wellman-Lord
Capital - $138/kW to
$265/kW.
Dry Scrubbing
Capital - $23/kW to
$47/kW.
Fly Ash
Alkalinity Scrubbinq
Not available.
PO
CO
cx>
Advantages
Disadvantages
Lower capital cost and
O&M costs. SO? and
particulates can be
removed simultaneously.
Relatively simple proc-
ess. Reliability is
90-95 .
Process produces approxi-
mately 2 times (by wt)
sludge as ash collected.
Sludge can be thixo-
tropic. Sludge quanti-
ties can be reduced by
forced oxidation.
Lower capital and OJM
costs. S02 can be
removed to very high
efficiencies (99 ).
Reliability is 90-95 .
Conventional process
equipment.
Process produces 1.5
times (by wt) fi1ter
cake as collected ash.
316SS material of con-
struction may be re-
quired to prevent
corrosion and pitting.
Capital and OSM costs
appear to be competitive
but data limited to pro-
totype experience. Poten-
tial saleable gypsum by-
product.
Process has not been
demonstrated commercially
in a 100MW or larger unit.
Commercially demonstrated.
Process produces saleable
product sulfur with a
Claus unit or sulfuric
acid. Lower potential for
scaling than calcium
system.
High utility costs (steam)
compared to other systems.
Special metallurgy may be
required. System required
to process S02 to sulfur
or sulfuric acid.
Lower projected
operating and capital
costs. Dry product.
Process will not be
commercially demon-
strated until late
1982. Product disposal
could be a problem
when sodium salts are
used as absorbent.
Commercially demonstrated
in 300MW units. Sludge con-
tains little calcium sulfite
which improves dewatering
and therefore reduces
settled water content.
Less potential for scaling.
Process is generally applied
to coal fired boilers which
burn high alkalinity coal.
-------
Section 4
Combustion Gases
S02 Removal
evaporator where it is indirectly heated by steam to convert the bisulfite
to sulfite and gaseous SC^. A portion of the sodium sulfite is also con-
verted irreversibly to sodium sulfate and thiosulfate which must be purged
from the system, requiring a makeup of NaOH or NaCOo. The Wellman-Lord
process can achieve over 90% SC^ removal.
Dry scrubbing experience to date has been limited, although systems
that have been operated show much promise, especially for low- and medium-
sulfur coals (79). The Spray Drying process is the only dry scrubbing pro-
cess currently being offered commercially. In this process the absorbent
solution, usually either lime or soda ash, is atomized and sprayed into the
incoming flue-gas stream to increase the liquid/gas interface and promote
mass transfer between the SCL and the slurry droplets. Simultaneously, the
thermal energy of the gas stream evaporates the water in the atomized drop-
lets to produce a dry, powdered mixture of sulfite/sulfate and unreacted
reagent. When used in combination with fabric filters these systems have
performed extremely well. The fabric filter collects the particulates and
also recovers some of the expensive reagent which is reused. In addition,
unused reagent that cakes on the fabric is available to react with more SC^
as flue gas passes through it.
FGD costs for boilers in synfuel plants depend upon the amount of
sulfur emissions control required. This may vary depending upon the amount
of sulfur in the fuel. FGD cost data have been developed by the EPA for
electric utility steam generating units ranging in size from 25 MW to 1000
MWe. The cost variations are principally governed by (1) size of the boiler,
(2) coal used, (3) averaging time over which the plant must meet S02 limita-
tions, and (4) the level of control maintained. Capital investment and
annual operating costs for lime/limestone and Wellman-Lord FGD systems are
listed in Table 4-15. These costs are for a 500 MW unit boiler burning
3.5% sulfur bituminous coal capable of achieving 90% removal.
239
-------
Section 4
Boiler Flue Gases
4.1.2.1 Boiler Flue Gases (Stream 302) - Individual Stream Control
Steam requirements for process and electric power generation purposes
for the K-T based methanol plant are assumed to be met by the boiler coal
feed rate of 25,230 kg/hr (as received basis). A material balance for the
Illinois No. 6 coal fired boiler was previously presented in Table 3-26. The
offgases from the boiler were estimated to have a flow rate of 10,385 kmol/hr
and contain 0.21 and 0.036% by volume of SOp and N02, respectively. The
uncontrolled particulate emissions were estimated to be 2421 kg/hr. These
emission rates were estimated to increase by 215% for the Fischer-Tropsch
synthesis case and decrease by 48% for the Mobil-M synthesis case.
In this section, details of pollution control alternatives applicable
to the 25,230 kg/hr Illinois No. 6 coal fired boiler are examined. However,
it is possible that some of the steam requirements of the pulverized coal
fired boiler may be offset by the combustion of dewatered dust from gasifica-
tion. If dewatered dust is burned to generate steam, coal requirements of
the pulverized coal fired boiler for the K-T based methanol plant will de-
crease by approximately 8900 kg/hr. Details regarding combustion of dewatered
dust are provided in Section 4.3. For the purposes of discussion in this
section, the offset in steam production from dewatered dust combustion will
be ignored.
The control of NO , S0?, and particulate emissions from the Illinois No.
A C-
6 coal fired boiler is expected to present no unique problems over those
encountered in the electric utility and other industries which use coal fired
units. The applicability, performance, and costs of these controls for the
pulverized coal fired boiler are discussed below.
240
-------
Section 4
Boiler Flue Gases
NO Control
X
NOY Control
A
As discussed earlier, NOV control in boilers is achieved through both the
A
design and operation of the combustion units to minimize its formation. For
new pulverized coal fired boilers this is achieved primarily by the use of
either low NOV burners (LNB) or overfire air (OFA). Boiler manufacturers have
A
used LNB or OFA as standard equipment on new boilers since the early 1970's
(85). Therefore, there are no incremental costs for NO control equipment
A
since no new boilers can be purchased from U.S. manufacturers without these
controls, regardless of whether the boiler is being built in the U.S. or
abroad. However, retrofitting NOV control equipment on older boilers does
A
have cost implications. These costs would be site specific and will also de-
pend upon the type of NOX control technique utilized. Since new K-T gasifica-
tion facilities will employ new boiler units, retrofit costs for NO controls
X
are not discussed in this manual. However, these costs can be obtained from
several EPA documents (83,84).
241
-------
Section 4
Boiler Flue Gases
Particulate Removal
Particulate Removal
Particulate loading in boiler flue gases is a function of many variables
such as the type of coal being fired, the amount of ash in the coal, and the
boiler design. Typically, 60 to 80% of the feed coal ash is emitted as fly
ash. For the Illinois No. 6 coal under consideration in this document, the
uncontrolled particulate emission rate for the methanol synthesis case was
estimated to be 2421 kg/hr.
Usually electrostatic precipitators or baghouses are used for particu-
late emission control from coal fired boilers. In utility applications
electrostatic precipitators are more widely used than baghouses at the pre-
sent time.
The only secondary waste streams from particulate control devices are
the collected particulate from the boiler flue gas (Stream 423). Typically
99% of the uncontrolled particulates in"the flue gas stream are collected.
Varying degrees of control for trace elements are also achieved with partic-
ulate control, depending primarily on the volatility of the individual
elements and on the efficiency of particulate control. For the size K-T
based methanol plant under consideration, this waste dust will be generated
at a rate of approximately 2400 kg/hr. In the case of both the ESP and bag-
house the collected particulates are dry. Therefore, it is possible that
fugitive particulate emissions may be generated during transfer and convey-
ing from control device to solid waste storage. The control of fugitive
particulate emissions during conveying and transfer is discussed in Section
4.1.6.
Costs for ESPs depend primarily upon the flue gas particulate loading
and the flue gas flow rate. To achieve outlet loadings of less than 45 ng/J
for the base plant boiler, over 99% particulate removal is desired. For a
combustion flue gas flow rate of 10,385 kmol/hr, the capital investment to
achieve outlet particulate loadings of 14 ng/J and 43 ng/J are estimated to
242
-------
Section 4
Boiler Flue Gases
Particulate Removal
be $6.1 million and 4.8 million, respectively. The annualized costs for the
same particulate loadings are estimated to be $1.2 million and $0.9 million,
respectively. These costs correspond to 0.55 and 0.43% of the base plant
capital investment and 0.35 and 0.27% of the base plant annualized costs.
The limited cost data on ESPs indicate that ESP capital investment costs
vary proportionally with flue gas flow rate provided all other variables
remain the same.
243
-------
Section 4
Boiler Flue Gases
S02 Removal
S02 Removal
As discussed previously, a number of competitive FGD processes are cap-
able of achieving similar S02 reductions. However, lime/limestone and Wellman-
Lord FGD processes are the most widely used S02 control processes in the
industry today. The costs for these systems vary greatly depending upon the
boiler size, coal sulfur content, and the degree of SO.? removal desired.
The pulverized coal-fired boiler associated with a K-T based methanol
plant utilizing Illinois No. 6 coal has approximately 21.95 kmol/hr (0.21%)
of S0? in the flue gas. Assuming that 83% S02 removal is desired (outlet
concentration of 370 ppmv), a lime-limestone FGD system will require a
capital investment of approximately $23.3 million with annual operating costs
of about $5.32 million. Alternatively, it is estimated that a Wellman-Lord
system will require a capital investment of $22.5 million with annual opera-
ting costs of $4.9 million. These costs correspond to 2.1% of the base plant
capital investment and 2.7% of the base plant annualized cost for the lime/
limestone FGD. In the case of the Wellman-Lord, the costs are 2% of the base
plant capital investment and 2.5% of the base plant annualized cost. Capital
investment and annual operating costs would increase by approximately 5% had
90% SOo removal been the design target.
Sludges and brines are generated by the lime/limes tone and Wellman-Lord
FGD systems. For the specific size K-T based methanol plant under considera-
tion, the lime/limestone FGD system is estimated to produce 10,722 kg/hr of
moist waste sludge (Stream 424). Alternatively, a Wellman-Lord FGD system is
estimated to produce 550 kg/hr of sulfur product (Stream 426) and a thiosulfate/
sulfate by-product purge stream (Stream 425) of 150 kg/hr.
244
-------
Section 4
Heater Flue Gases
4.1.2.2 Process Heater Flue Gases - Individual Stream Control
Process heaters in a K-T gasification facility are expected to utilize
either offgases or liquid products from synthesis/fractionation operations.
The offgases and liquid fuels are essentially free of sulfur, particulate,
and fuel bound nitrogen. Therefore, the contribution of process heaters to
sulfur and particulate emissions is minimal. Any NO emissions generated
X
are a function of combustion design, and combustion modification techniques
discussed previusly are applicable.
Control of NOX emissions from process heaters has been investigated in
recent years. Initial data indicate that load variation, staged combustion,
and low NOV burners can be used to control NC" emissions. Cost information
A X
on these techniques is very site-, size-, and fuel-specific. The number and
size of process heaters for a K-T gasification facility depend upon the
details of process heat requirements. Since it is beyond the scope of this
document to perform the detailed engineering required to assess these require-
ments, no detailed cost estimation for the control of NO emissions from
X
specific process heaters was performed. However, cost information from
vendors indicates that low NOX burners may cost approximately $0.95/MJ when
applied to process heaters (86).
245
-------
Section 4
Combustion Gases
Sec. Strms. - Other Media
4.1.2.3 Secondary Streams from Other Media - Fluidized-Bed Boiler Flue Gases
(Stream 413) - Individual Stream Control ~~~
As discussed in Section 3.3.1, a filtration step has been included in
the design for the gas cooling and dust removal process. In the case of bitu-
minous coal gasification, this filtration step may be desirable to reduce
pond area and/or improve the feasibility of dust combustion prior to disposal.
In this section details of pollution control alternatives applicable to the
combustion flue gases from the fluidized bed combustion of dewatered dust (see
Section 4.3) are discussed. The flow rate of the flue gas from the fluidized-
bed boiler is estimated to be 7419 kmol/hr. It is estimated to contain 0.18%
S02 by volume and 1846 kg/hr of particulates. NOX emissions are estimated
to be about 150 ng/J. The composition of the flue gas is presented in Table
4-16.
TABLE 4-16. COMPOSITION OF FLUE GAS FROM THE DEWATERED DUST-FIRED FLUIDIZED-
BED BOILER (STREAM 413)*
Components
N2f
H20
co2
°2
S09
c.
Ar
Solids*
Flue Gas
mole %
59.5
23.9
12.0
3.7
0.18
0.71
100.0
Composition (Dry Basis)
kg/hr
123,664
31,916
39,278
8,746
838
2,109
206,551
1846
kmol/hr
4415
1773
892
273
13
53
7419
*Flue gas flow rates are based upon engineering estimates.
NOX is approximately 150 ng/J which is equivalent to 56 kg/hr (as
^Before final particulate control.
246
-------
Section 4
Combustion Gases
Sec. Strms - Other Media
NOX Control
Almost all the data from fluidized-bed boilers indicates NOX emission
levels of less than 301 ng/J (87). For a fuel containing only 36% carbon by
weight (dry), these emissions will be even lower, around 150 ng/J. Since these
emission levels are relatively low for coal fired boilers, no NOV controls
A
for the fluidized bed boiler are likely to be used.
Particulate Control
Particulate matter emitted from the combustion section of a fluidized-
bed boiler consists of fly ash from the coal, unburned carbon, and elutriated
bed material. A primary cyclone is used to collect larger particles contain-
ing the most significant carbon concentration for circulation back to the
fluidized-bed combustor or to a separate carbon burnup cell. A secondary
cyclone of higher efficiency can also be used to collect smaller particles
for disposal as ash. Approximately 90% of the particulate matter is captured
prior to final particulate control.
Final particulate control after primary and/or secondary cyclones is
performed by use of conventional systems previously discussed in Section
4.1.2. The systems can consist of electrostatic precipitators, fabric
filters, scrubbers, or cyclones. They can be operated as hot-side or cold-
side units (upstream or downstream of final heat recovery), except for fabric
filters which must be installed cold-side to prevent excessive fabric deteri-
oration. Existing fluidized bed units typically use either fabric filters
or electrostatic precipi tenors (87).
Costs for particulate control were estimated assuming the particulate
control device used is an electrostatic precipitator. For 99.7% particulate
removal, the capital investment for the ESP unit was estimated to be $4.4
million for a flow rate of 7419 kmol/hr (refer to Table 4-16). Annualized
costs were estimated to be $0.9 million. These costs correspond to 0.4% of
the base plant capital investment and 0.26% of the base plant annualized cost.
247
-------
Section 4
Combustion Gases
Sec. Strms. - Other Media
SC>2 Removal
Sulfur dioxide emissions are a major concern in conventional coal-fired
boilers. However, by using fluidized-bed technology, SCL emissions can be
reduced by up to 90% or more by direct addition of sorbent into the bed. The
coal is burned in the presence of lime sorbent. The S0£ reacts with the
calcium oxide and excess oxygen forming calcium sulfate. The degree of S02
capture is strongly dependent on the calcium to sulfur molar feed ratio.
Other factors which affect the sulfur capture efficiency of the system are
the reactivity of the sorbent, the particle size of both sorbent and coal,
gas residence time in the bed, the feed mechanism and material distribution
in the bed, and temperature. These parameters can be adjusted to obtain the
maximum S02 removal for the system at a particular Ca/S molar feed ratio.
Past data indicate that an average Ca/S molar feed ratio of between about
2.5 and 4 will achieve a reduction of 90% in S02 emissions, depending upon
the sorbent reactivity; however, higher ratios may be needed if sorbents of
low reactivity are used (87).
Capital investment and annualized costs for controlling SOp emissions
are included in the costs of the fluidized-bed boiler. Capital investment
and annualized costs for the fluidized boilers are discussed in Section 4.3
Past data indicate that the percent increase in costs for fluidized boilers
capable of achieving 85% S02 removal over uncontrolled conventional boilers
(49 MW. capacity) burning 0.9% sulfur coal is approximately 5 to 7% (87).
248
-------
Section 4
Org. and CO Waste Gas
4.1.3 Organic and CO Containing Haste Gases
There are two major waste streams which are free of sulfur compounds but
contain appreciable levels of organics and/or CO: (1) C02 offgas from SNG
purification (Stream 239) in F-T synthesis; and (2) Mobil M-gasoline synthesis
catalyst regeneration/decommissioning offgases (Stream 231). In addition, an
offgas containing CO may be generated as a result of decommissioning methana-
tion catalyst (Stream 237) in facilities employing F-T synthesis. In the
event that these waste gases warrant control of organic and/or CO emissions,
controls such as thermal or catalytic incineration may be applicable. Flaring
is typically employed for small volume, intermittent gases or gases resulting
from upset conditions and would not be appropriate for treatment of SNG
purification offgas or the Mobil catalyst regeneration/decommissioning off-
gases. Key features of these incineration processes are discussed in Section
4.1.1.
In an integrated facility, waste gases discussed in this section might
be combined with other waste gases for incineration in a common incinerator.
However, because of the limited characterization and generation rate data
for waste gases, cost estimates in this section are presented in terms of
dedicated incineration units. Therefore, these control costs represent upper
limit costs for control of these streams due to the economy of scale which
might result if waste gases can be combined for incineration.
249
-------
Section 4
C02 Offgas from SNG Purif.
4.1.3.1 C02 Offgas from SNG Purification (Stream 239) - Individual Stream
Control
Facilities employing F-T synthesis with SNG co-production will generate
a CC^-rich offgas (Stream 239) on a continuous basis as a result of SNG puri-
fication. This offgas will consist primarily of CCL and would be free of
sulfur compounds although it may contain appreciable concentrations of
organics and CO. For purposes of evaluation, the offgas has been estimated
to contain 1% CH^, 0.2% non-methane hydrocarbons, and traces of CO (see Section
3.4.6). The total flow rate of this stream was estimated to be approximately
273 kmol/hr. Control of potential C02 offgas emissions by incineration is
discussed in this section.
250
-------
Section 4
C02 Offgas from SNG Purif.
Incineration
Incineration
The CCL offgas from SNG purification is estimated to contain approxi-
mately 4.3 kmol/hr total hydrocarbons (expressed as ChL). Thermal incinera-
tion of organic-containing waste gases and liquids is reported to result in
flue gas concentrations ranging from <5 to 70 ppmv total hydrocarbons (ex-
pressed as CHj, and 7 to 89 ppmv CO. Assuming flue gas concentrations of
30 ppmv total hydrocarbons and 50 ppmv CO, thermal incineration of the C02
offgas using a low Btu supplemental fuel gas would result in emissions of
approximately 0.01 kmol/hr total hydrocarbons (expressed as CH^) and 0.02
kmol/hr CO. The total capital investment for this thermal incinerator would
be approximately $296,000. Annualized costs would be $100,000. These costs
correspond to 0.02 to 0.03% of the base plant capital investment and the
annualized base plant costs. Similar performance and costs would be asso-
ciated with the catalytic incineration alternative.
251
-------
Section 4
Regen/Decom Offgases
4.1.3.2 Regeneration/Decommissioning Offgases (Streams 231 and 237) -
Individual Stream ControT
There are two sources of offgases resulting from catalyst regeneration
and/or decommissioning. Facilities employing Mobil M-gasoline synthesis will
generate an offgas from regeneration/decommissioning of the synthesis catalyst
(Stream 231), while facilities employing F-T synthesis with SNG co-production
may generate an offgas when decommissioning the methanation catalyst (Stream
237). The Mobil M-gasoline synthesis catalyst requires periodic regeneration
for coke removal. For the subject facility, it has been estimated that re-
generation will occur over a period of 3800 hours per year, and offgases will
be generated at an average rate of 200 kmol/hr. This offgas is likely to con-
tain hydrocarbons and/or CO. Methanation catalysts are not typically regen-
erated but, because they tend to be pyrophoric in the active state, they are
oxidized during decommissioning. No data which enable the estimation of
methanation catalyst decommissioning offgas flow rates or characteristics
are available.
Control of potential Mobil M-gasoline catalyst regeneration/decommission-
ing offgas emissions by incineration is discussed in this section. Control of
offgases from methanation catalyst decommissioning will not be addressed due
to data limitations. However, in principle, control of organic/CO emissions
associated with methanation catalyst decommissioning could be achieved by
thermal or catalytic incineration or by flaring.
252
-------
Section 4
Regen/Decom Offgases
Incineration
Incineration
Thermal incineration of organic-containing waste gases and liquids is
reported to result in flue gas concentrations ranging from <5 to 70 ppmv
total hydrocarbons (expressed as CH^) and 7 to 89 ppmv CO. Although char-
acterization data are not available for offgas from Mobil M-gasoline synthe-
sis catalyst regeneration, hydrocarbons and/or CO may be present in the in-
cinerator feed gas due to the offgas composition or the fuel gas composition.
Assuming fuel gas concentrations of 30 ppmv total hydrocarbons and 50 ppmv
CO, thermal incineration of the Mobil M-gasoline synthesis catalyst regen-
eration/decommissioning offgas would result in emissions of approximately
0.01 kmol/hr total hydrocarbons and 0.02 kmol/hr CO. These estimates are
based upon the assumptions that the offgas would not have significant heat-
ing value and that a low Btu fuel gas would be employed as supplemental fuel.
The capital investment for this thermal incinerator would be approximately
$289,000. The annualized costs would be $90,000. These costs correspond to
0.02% to 0.03% of the base plant capital investment and the annualized base
plant costs, respectively. Similar performance and costs would be associated
with the catalytic incineration alternative.
253
-------
Section 4
Fugitive Dust Material Storage
4.1.4 Fugitive Dust from Material Storage (Stream 200)
Open or partially enclosed storage piles are often used for storage of
bulk materials not affected by precipitation or slight contamination such as
coal, sand, gravel, clay, and gypsum. The material may be stored for a short
time with a high turnover rate to accommodate surges in daily or weekly rates
of sequential processes or may provide a long-term reserve for emergency
supply or to meet cyclical seasonal demands.
Most dust arises from stockpile areas as the material is dumped from the
conveyor or chute onto the pile and as material is reclaimed from the pile.
During periods of high wind speeds or low moisture, wind erosion of the sur-
face may also cause emissions.
In coal gasification/liquefaction plants fugitive dust is emitted by
coal and solid waste storage piles. The techniques used to control these
emissions are not unique to liquefaction plants and are widely used in in-
dustries that require large scale material storage. The most commonly used
techniques are vegetative stabilization, chemical stabilization, capping,
stacked segregation, water spraying, and confinement. Surface protection
methods such as vegetative stabilization, chemical stabilization, capping,
and stacked segreation are primarily used on reserve storage piles since
these piles are subject to minimal disturbances. Active storage piles gen-
erally require either water spraying or confinement to control dust emissions.
The key features and unit costs for these techniques are listed in Table
4-17.
Vegetative stabilization involves planting an appropriate ground cover
or shrub over the pile to be stabilized. A soil cap may be required to sup-
port vegetation. The efficiency of vegetative cover in reducing wind erosion
is dependent on the density and type of vegetation that can be grown. For
applications such as stabilizing tailing piles, the use of vegetative stabi-
lization decreases emissions approximately 65%. When vegetative stabilization
254
-------
TABLE 4-17. KEY FEATURES OF STORAGE PILE DUST CONTROL TECHNOLOGIES
Method
Control Principle
Control
Effectiveness*
Reliability/
Special Problems
Unit Costs*
Other
Pollutants
Generated
Vegetative
Stabilization
Covering pile with sod.
Approximately 65%
reduction over un-
stabilized pile; and 90%
if chemical stabilizer
is also used.
IV)
tn
en
Chemical
Wetting Agents
Modify surface tension
properties to improve
effectiveness of water
sprays.
Up to 90% reduction
in dust losses-,*
Crusting Agents
Organic binders combine
with particles to form
tough crust on surface.
Up to 90% reduction
in dust losses.
Requires frequent $2.70/m2
watering.
Handling of sod during
reclamation operations
is cumbersome and
expensive.
3. Upper layer of stored
material is contami-
nated with soil.
1. Piping may require heat $.33-.77/Mgf
tracing when freezing
is a concern.
2. Can cause corrosion
problems in equipment
exposed to sprays,
3. May increase material
degradation*
4. Effects are short-term,
1. May increase chances $.55- . 72/m
of spontaneous com-
bustion, especially
in piles subject to
stockpile and reclaim
operations.
2. Crust tends to break
up during heavy rains.
Soil dust from
earth covering,
Volatiles which
depend on wetting
agent utilized.
Volatiles which
depend on crust-
ing agent
utilized.
(Continued)
-------
TABLE 4-17. (Continued)
Method
Capping
Control
Control Principle Effectiveness*
Paving with earth or Up to 100%c f
asphalt or cover with
polyethylene.
Reliability/
Special Problems
1 . Both coverings may
increase chances of
spontaneous combustion.
2. Polyethylene presents
severe handling prob-
lems and is also not
practical in high wind
climates.
Unit Costs*
$.49/m2 for
asphalt.
$1.96/m2 for
polyethylene.
Other
Pollutants
Generated
Soil dust from
earth covering.
ro
en
Stacked
Segregation
Water
Spray
Confinement
Coating surface of com-
pacted storage pile with
layer of select, medium
sized material,,
Spray application of
210-250 L/Mg to reduce
dusting.
No data available<•
Approximately 50%
reduction in losses.
Enclosure of active
storage pile in a totally
enclosed barn or silo
with point source dust
control equipment on
building vents.
Up to 99% reduction
in losses.
Either deliveries of differ- Not available
ent sized material must be
coordinated or both sizes
must be readily available
from storage.
1. Piping may require heat
tracing when freezing
is a concernc
2. May increase degradation*
3. Frequent re-treatment
necessary.
Requires extreme care
when storing potentially
explosive materials.
Not available
none
none
$110/Mg of
stored material
$1 million to $3
million per silo
depending on
size.
none
*Cost and control efficiency data obtained from Reference 88
^Data obtained from Reference 89
*Data obtained from Reference 90
-------
Section 4
Fugitive Dust Material Storage
is used in conjunction with a chemical stabilizer, this efficiency increases
to approximately 90%.
Chemical stabilization to decrease fugitive dust emissions involves the
application of wetting or crusting agents. Wetting agents are used to provide
better wetting of fines and longer retention of moisture. They also reduce
the water surface tension allowing the fines to be wetted with a minimum
amount of water. This treatment protects stockpiled material until the added
moisture is removed by heat and wind. Some of these agents remain effective
for weeks or months without additional rewatering depending on local condi-
tions. Crusting procedures involve the use of bunker C crude oil, water
soluble acrylic polymers, or organic binders. These materials are sprayed
on the surface of the storage pile, coating the top layer of particles with
a thin film. This film causes the particles to adhere to one another, form-
ing a tough durable crust which is resistant to wind and rain. As long as
the crust remains intact, the storage pile is protected from wind losses.
Capping involves the paving of the surface area of the storage pile with
asphaltic compounds or earth or covering the pile with polyethylene tarpaulins,
Usually a slurry of wood pulp and asphalt or road tar is sprayed over the
surface of the pile. The covering is usually about 3 mm thick. Polyethylene
tarpaulins are also used, however, they are cumbersome to handle when there
are high wind speeds or when a large size storage pile is to be capped.
Another effective means of controlling dust emissions from coal storage
piles is the stacking of coarse material on the surface of a properly com-
pacted pile. For instance, a 0.152 m layer of fine coal (6.4mm x 0 mm) on
the top and sides of the coal storage pile can be anchored in place by a
0.102 m layer of larger size coal (24 mm x 0 mm) placed on top of the fine
coal. The larger size coal has better weathering characteristics compared
to the smaller sized coal.
257
-------
Section 4
Fugitive Dust Material Storage
Water spraying is another common method of dust suppression. Dust con-
trol by water spraying is usually obtained by placing spray nozzles at stra-
tegic locations over the stockpile area. The spraying operation is simple in
that it only involves the operation of a pump. Water requirements for large
volume operations vary from 210 to 250 liters/Mg of material. Such systems
are, however, prone to freezeups during winter months. Also, the added mois-
ture can create handling problems during reclamation and subsequent proces-
sing.
Enclosure of the coal storage pile is generally the most effective means
of reducing fugitive dust emissions, because it allows the emissions to be
captured. However, enclosures can be very expensive, since they have to be
designed to withstand wind and snow loads and meet requirements for interior
working conditions. An alternative to enclosure of all material is to screen
the material prior to storage, sending the oversize material to open storage
and the fines to enclosures.
258
-------
Section 4
Fugitive VOC Emissions
Evaporative Emissions
4.1.5 Fugitive VOC Emissions
There are many sources of fugitive VOC emissions in a synthetic fuel
plant. These emissions can be categorized as (1) evaporative emissions that
result from the storage of liquid products and by-products and (2) VOC emis-
sions that result from fluid leaks from plant equipment.
Evaporative Emissions
Evaporative emissions from storage tanks storing volatile liquids result
from temperature changes which cause the vapor pressure of the stored liquid
to vary, causing vapor emissions. The minimum accepted standard for storage
of VOC is the fixed roof tank. It is designed to operate at only slight
internal pressure or vacuum and is susceptible to emissions from thermal
expansion and other mechanisms by which vapors are produced.
Emissions from fixed roof tanks can be reduced by minimizing diurnal tem-
perature variations (e.g., placing tanks underground), proper setting and main-
tenance of pressure vacuum vents, and leak prevention efforts. Significant
controls can be effected by either floating a cover on the surface of the
stored liquid or by replacing the fixed roof storage tank by a floating roof
storage tank.
Floating roof tanks successfully limit hydrocarbon losses by eliminating
the ullage into which stored material vaporizes. This is accomplished by
floating a rigid deck or roof on the surface of the stored liquid thus
eliminating air space and preventing the formation of organic vapor above the
liquid surface. To effectively control emissions, the floating roof employs
primary and secondary seals to shelter the liquid surface from the atmosphere.
Control efficiencies of greater than 90% are achievable by floating roof tanks.
Vapor processing units can also be used to control VOC emissions from
fixed roof storage tanks. Some of the vapor processing techniques available
are carbon adsorption, thermal oxidation, refrigeration, compression-
259
-------
Section 4
Fugitive VOC Emissions
Evaporative Emissions
refrigeration-absorption, and compression-refrigeration-condensation. Cataly-
tic oxidation is not typically employed in this application.
The carbon adsorption vapor recovery unit uses beds of activated carbon
to remove VOCs from the air-vapor mixture. These units generally consist of
two vertically positioned carbon beds and a carbon regeneration system. The
air-vapor mixture enters the base of one of the adsorption columns, and the
VOC components are adsorbed onto the activated carbon as the gases ascend. Ad-
sorption in one carbon bed occurs for a specific timed cycle before switch-over
to desorption. The nearly saturated carbon bed is then subjected to vacuum,
steam, or thermal regeneration, or a combination of these methods, and the VOCs
are stripped from the bed. Vacuum regenerated units recover VOCs by absorption
in a product stream which circulates between the control unit and product
storage. The air and any remaining VOCs exiting from the absorber are passed
again through the absorbing bed and exhausted to the atmosphere. Steam re-
generated units condense the VOC-water mixture and return the separated product
to storage. Some vacuum regenerated systems remain in operation for up to two
hours after loading activity ceases, in order to collect any residual vapors
in the system and to assure complete regeneration of the carbon beds.
Thermal oxidation units rely upon burning VOC vapors to produce non-
polluting combustion products. Vapors are piped either to a vapor holder or
directly to the oxidizer unit. When a vapor holder is used, operation of the
oxidizer begins when the holder reaches a preset level and ends when the holder
is empty or at the lower preset level. With no vapor holder in the system,
the oxidizer is energized by means of pressure in the vapor line, or by an
electrical signal produced by manual activation. In some cases propane is
injected into the vapor stream to keep the VOC level above the explosive level.
Refrigeration type recovery units remove VOCs from an air-vapor mixture
by straight refrigeration at atmospheric pressure. Vapors displaced from
storage tanks enter a condenser section where methylene chloride "brine" is
260
-------
Section 4
Fugitive VOC Emissions
Evaporative Emissions.
pumped through the finned tube sections of a heat exchanger. Brine tempera-
ture in this section ranges from 190K to 210K. Some units contain a pre-
cooler section (glycol and water solution circulating at 274K) to remove most
of the water from the gases prior to the main condenser. There are no com-
pression stages in this type of unit. The condensed product is collected and
pumped to one of the product storage tanks. The cold collection surfaces are
periodically defrosted by pumping warm (305K) trichlorethylene through the
condenser. This defrost fluid is kept warm by heat salvaged from the refri-
geration equipment. Recovered water passes to a waste storage tank or
gasoline-water separator. The defrost cycle takes from 15 to 60 minutes,
depending on the amount of ice accumulated on the finned-tubes.
In a compression-refrigeration-absorption (CRA) vapor recovery system,
the vapors from the product storage tanks are first passed through a saturator
which sprays liquid product into the air-vapor gas stream. This ensures that
the VOC concentration is above the explosive range. The saturated gas mixture
is stored in a vapor holder until, at a preset level, it is released to the
control unit. The vapor holder is usually a special tank containing a bladder
with variable volume and constant pressure. A product storage tank with a
lifter roof can also function in this capacity.
The first stage of CRA processing is a compression-refrigeration cycle in
which water and heavy VOCs are compressed, cooled, and condensed. The uncon-
densed vapors move into a packed absorber column where they are contacted by
chilled product (277K) drawn from product storage and absorbed. The fresh
product stream is used first in the saturator, then it passes through an
economizing heat exchanger as it enters the absorber. The rich absorbent also
passes through the heat exchanger before being pumped back to storage. The
operation of the control system is intermittent, starting when the vapor
holder is filled and stopping when it has emptied or at its lower preset level.
Cleaned gases are vented from the absorber column to the atmosphere.
261
-------
Section 4
Fugitive VOC Emissions
Fugitive Organic Emissions
A vapor recovery system employing a compression-refrigeration-condensa-
tion unit makes use of a vapor holder to store accumulated air-vapor mixture
and a saturator for ensuring that the VOC concentration is above the explosive
range. The unit is activated and begins processing vapors when the vapor
holder has filled to a preset level. Incoming saturated air-vapor mixture is
first compressed in a two-stage compressor with an intercooler. Condensate
is withdrawn from the intercooler prior to compression in the second stage.
The compressed vapors then pass through a refrigeration-condenser section
where they are returned along with the intercooler condensate to a product
storage tank. Cleaned gases are exhausted from the top of the condenser.
Costs for vapor processing units vary with the type of product and the
product throughput. In the case of gasoline, capital investment costs for
these units range from $152,000 to $270,000 for a gasoline throughput of 380
m /day. These costs increase by 15% when the gasoline throughput increases by
150% (91).
Costs for internal floaters range from approximately $4,000 to $40,000
for storage tanks with diameters of 5 and 30 m, respectively. Secondary
seals are estimated to cost $75 per linear meter.
Fugitive Organic Emissions
There are many potential sources of fugitive organic emissions that
result when process fluid (either liquid or gaseous) leaks from plant equip-
ment in a typical gasification/liquefaction synthetic fuel plant. Some of
these are: pumps, compressors, in-line process valves, pressure relief
devices, open-ended valves, sampling connections, flanges, agitators, and
cooling towers.
There are two basic methods which have been used to control fugitive
organic emissions:
262
-------
Section 4
Fugitive VOC Emissions
Fugitive Organic Emissions
1) leak detection and repair methods and
2) equipment specification.
Leak detection methods include individual component surveys, area (walk-
through) surveys, and fixed point monitors. In an individual component survey
each fugitive emission source (pump, valve, compressor, etc.) is checked for
VOC leakage. The source may be checked for leakage by visual, audible, olfac-
tory, soap bubble, or instrument techniques. Visual methods are particularly
effective in locating liquid leaks. Escaping vapors from high pressure leaks
can be audibly detected, and leaks of odorous materials may be detected by
smelling the odor. Perhaps the best method of identifying leaks of VOC from
equipment components is by using portable detection instruments. By sampling
and analyzing the air in close proximity to the leak, the hydrocarbon concen-
tration of the sampled air can be determined. The leak rate from the source
can be roughly estimated since relationships exist between monitoring concen-
trations and mass emission rates.
An area survey (also known as a walk-through survey) requires the use of
a portable hydrocarbon detector and a strip chart recorder. The procedure
involves carrying the instrument within one meter of the upwind and downwind
sides of process equipment and associated fugitive emission sources. An in-
crease in observed concentration indicates leaking fugitive emission sources.
The instrument is then used for an individual component survey in the suspected
leak area.
Fixed point monitors are automatic hydrocarbon sampling and analysis in-
struments positioned at various locations in the process unit. The instruments
may sample the ambient air intermittently or continuously. Elevated hydro-
carbon concentrations indicate a leaking component. As in the walk-through
method, an individual component survey is required to identify the specific
leaking component in the area. For this method, the portable hydrocarbon
detector is also required.
263
-------
Section 4
Fugitive VOC Emissions
Fugitive Organic Emissions
Reduction of fugitive emissions from the identified leaking components
is effected by repair methods. In many cases, perfect repair will not be
achieved; however, effective repair can substantially reduce emissions from
the leaking component. Typical repair methods employed on the various com-
ponents are listed in Table 4-18.
TABLE 4-18- REPAIR METHODS FOR FUGITIVE EMISSIONS REDUCTION
Component Repair Method
Pumps and compressors Tighten packing gland
Relief valves Manual release of the valve
may improve the seat seal
In-line valves Tighten packing gland
Lubricate plug type valves
Inject sealing fluid into process
valves requiring repair
Flanges Replace flange gaskets
The second method used to control fugitive emissions is by equipment
specification. Typical equipment specifications used are listed in Table
4-19.
Costs for repair methods will depend upon the complexity of the compon-
ent undergoing repair. The major costs of maintenance and repair methods are
labor related. In the case of equipment specifications costs will depend
upon the component being replaced. Typically, double mechanical seals cost
$815/pump (installed). Flush oil systems for double mechanical seals cost
$1500/pump.
264
-------
TABLE 4-19. EQUIPMENT DESIGN/MODIFICATIONS FOR FUGITIVE HYDROCARBON EMISSIONS CONTROL
CTl
CJ1
Pumps
Compressors
Pressure Relief Devices
Open-Ended Valves
In-Line Valves
- improve seal at the junction of moving shaft and stationary casing
- use seal less pumps
- use double mechanical seals
- use closed vent systems around seal areas
- improve seal at the junction of moving shaft and stationary casing
- use double mechanical seals
- use closed vent systems around seal areas
- use rupture disks upstream from the safety/relief valve
- use resilient seal or "o-ring" relief valves
- use closed vent systems to transport valve discharge to control
devices
- install a cap, plug, flange, or a second valve to the open end of
the valve
- use diaphram and bellows seal type valves
-------
Section 4
Fugitive VOC Emissions
Storage Emissions
4.1.5.1 Product/By-Product Storage Emissions
Various types of vessels are employed to store petrochemical products.
The suitability of a specific tank design depends on the vapor pressure that
the stored product exerts at ambient conditions and the type of storage
desired. The floating roof tank is widely used for control of volatile
organic compounds such as gasoline when the true vapor pressure is in the
range of 10 to 80 kPa at storage conditions. Low vapor pressure VOCs (<10
kPa) are stored in fixed roof tanks. Therefore, it was assumed that methanol
and gasoline products would be stored in floating roof tanks and diesel fuel
and fuel oil in fixed roof tanks. Uncontrolled emissions estimates from
these tanks have previously been discussed in Section 3.6.5.
Emissions from floating roof tanks consist primarily of standing storage
losses and wetting losses. These losses are greatly reduced by the addition
of secondary seals. The most widely used approach for VOC control is the addi-
tion of secondary seals to existing floating roof tanks. The secondary seal is
generally of a resilient fabric (e.g., loop seals) or a pliable material such
as a treated rubber. Its flexibility allows it to maintain contact in places
where the shell might be slightly out of round as well as in areas where
rivet heads project from the shell wall. Upon descent of the roof, these
seals wipe down the film left behind by the primary seal. These seals also
reduce standing storage evaporative losses since they form a second seal above
the vaporized product which has diffused past (or permeated through) the pri-
mary seal. Not only do they form a second barrier for the vapor, they also
seal this vapor off from the effects of moving air. As a result, secondary
seals are effective control devices which, when used on floating roof tanks,
can reduce overall emissions by as much as 98% (refer to Table 4-20).
Fixed roof tanks consist of a steel cylindrical shell with a permanently-
affixed roof. The roof design may vary from cone-shaped to flat. Of pre-
sently employed tank designs, the fixed roof tank is the least expensive to
266
-------
TABLE 4-20. STORAGE TANK EMISSION ESTIMATES
CTl
—1
Product
Methanol
Methanol
Methanol
Gasoline
Gasol ine
Diesel
Fuel Oil
Roof Type
Floating
Floating
Floating
Floating
Floating
Fixed
Fixed
Capacity
(m3)
3,200
45,000
46,500
22,000
39,000
3,600
940
Diameter
(m)
18.2
62.5
64
43.6
53.3
19.5
11.6
Synthesis Case
Fischer-Tropsch
Methanol
Mobil M
Fischer-Tropsch
Mobil M
Fischer-Tropsch
Fischer-Tropsch
Assumed
Vapor
Pressure
(KPa)
17.3
8.83
17.3
8.83
17.3
8.83
49.78
33.85
49.78
33.85
0.08
0.046
0.00059
0.0026
Mass Emission
Uncontrolled
(kg/yr)
8,740
6,630
59,790
45,360
61,250
46,470
21,790
20,430
28,950
27,140
800
500
9
5
Rate (kg/yr)*
Control! edt
(kg/yr)
173
125
1226
896
1256
918
1471
1208
1955
1605
39.2
38.8
1.05
0.45
Avg. Control
Efficiency
98.0
98.1
97.9
98.0
97.9
98.0
93.2
94.1
93.2
94.1
95.1
92.2
90
91
t
Calculations based on information contained in AP-42 (Reference 76). The
maximum emissions (July). The lower number represents the average annual
Southwest Illinois.
Floating roof tanks were assumed to have both primary and secondary seals.
internal floaters with closure seals.
higher values represent the month with
values. Location was assumed to be
Fixed roof tanks were assumed to have
-------
Section 4
Fugitive VOC Emissions
Storage Emissions
construct and is generally considered as the minimum acceptable standard for
storage of petroleum liquids. Fixed roof tank emissions are most readily
controlled by the installation of internal floating roofs. An internal float-
ing roof tank is essentially a fixed roof tank with a cover floating on the
liquid surface inside the tank which rises and falls with the liquid level.
Calculations indicate that emission reductions of more than 90% are achieved
by retrofitting fixed roof tanks with internal floating roofs. Other control
technologies such as vapor processing systems can be also used to effect 90%
control. However, internal floating roof covers are widely used because of
their simplicity and their low annual operating and capital costs. Controlled
emissions from fixed roof tanks are listed in Table 4-20.
Annual costs of applying secondary seals and internal floaters to storage
tanks for the K-T based indirect liquefaction facility being discussed in
this report are listed in Table 4-21. Secondary seals were assumed to cost
$130 per linear meter (92). The cost of the aluminum internal floating roof
cover varies expotentially with the storage tank diameter and ranges from
$5500 for a 5m diameter tank to $49,300 for a 30m diameter tank (92).
The capital investment and annualized costs for controls on fixed and
floating roof tanks was estimated to be from $25,500 to $69,600 and $5,900
to $10,300, respectively. However, annualized costs are decreased consider-
ably because of savings due to product recovered by use of these controls
(refer to Table 4-21). These costs correspond to 0.002 to 0.005% of the base
plant capital investment. The annualized costs are negligible when compared
to base plant costs.
268
-------
TABLE 4-21. ESTIMATED INCREMENTAL COSTS FOR STORAGE OF SYNTHETIC LIQUIDS*
ro
CTl
Type of
Liquid
Methanol
(3,200 m3)
Methanol
(45,000 m3)
Methanol
(46,500 m3)
Gasoline
(22,000 m3)
Gasoline
(39,000 m3)
Diesel Oil
(3,600 m3)
Fuel Oil
(940 m3)
Synthesis
Case
Fischer-Tropsch
Methanol
Mobil M
Fischer-Tropsch
Mobil M
Fischer-Tropsch
Fischer-Tropsch
Type of
Control
Secondary
seal
Secondary
seal
Secondary
seal
Secondary
seal
Secondary
seal
Internal
floater
Internal
floater
Annual i zed
Control
System Costs
1,900
5,900
6,000
4,200
5,000
5,200
3,500
Annual
Product
Savings
1,800
12,100
12,300
7,000
9,300
150
1
Net
Annual
Cost
100
(6200)f
(6300)t
(2800)t
(4300 )f
5050
3500
Controlled
Emissions
Costs $/kg/hr
0.013
0
0
0
0
8.3
560
t
250,000 kg/hr (MAF) coal feed to gasifier basis
Parenthetic cost data represent credits
-------
Section 4
Fugitive VOC Emissions
Fugitive Organic Emissions
4.1.5.2 Fugitive Organic Emissions from Process Equipment (Stream 241)
As discussed earlier, process equipment such as pumps, compressors, in
line valves, pressure relief devices, open-ended valves, etc., are prone to
leakage and thus are sources of fugitive organic emissions. Two methods
can be employed to control these emissions. A labor intensive method
involving leak detection and constant repair and maintenance can be used,
and/or replacement of leaking equipment by leak-free equipment. Obviously,
if equipment specification in addition to extensive leak detection and
repair is performed, greater reduction in fugitive organic emissions is
achieved.
Two approaches to reduce fugitive organic emissions are generally used.
In the first approach leak detection and repair methods (as suggested in the
VOC leak control techniques guideline document for the petroleum industry) can
be applied (93). Here, leak detection is accomplished by checking equipment
components for emissions of VOC at specified intervals using a portable VOC
detection instrument to sample and analyze the air in close proximity to the
potential leak area. A measured VOC concentration greater than some pre-
determined level would indicate a leak that would require equipment repair.
Inspection of all equipment has to be performed on a regular basis.
Controlled emission estimates for the K-T based indirect liquefaction
facility under consideration were made assuming that the aforementioned
approach was employed. Emission reductions of approximately 70% were esti-
mated for the three synthesis cases as shown in Table 4-22. These estimates
were based on a detection level of 10,000 ppm, weekly inspections of light
liquid pump seals, monthly inspection of all other equipment, and open-ended
valves were required to be sealed with a cap plug or another valve. Capital
investment and annualized costs were estimated to range from $41,000 to
$145,000 and $251,000 to $102,000, respectively (refer to Table 4-23). These
270
-------
TABLE 4-22. FUGITIVE ORGANIC EMISSIONS FROM PROCESS EQUIPMENT
ro
Pump Seals*
Light Liquid Service
Heavy Liquid Service
In-Line Valves
Gas Service
Light Liquid Service
Heavy Liquid Service
Safety Relief Valves
Vapor Service
Compressor Seals*
Hydrocarbon
Hydrogen
Flanges
Drains
Totals
Uncontrolled
Emission
Factor"1"
(kg/hr)
0.154
0.029
0.027
0.011
0.00023
0.086
1.28
0.10
0.00025
0.07
Uncontrolled Emission Rates
(kg/hr)
Fischer-
Tropsch
16.74
0.52
34.72
37.02
0.14
24.77
12.8
1.8
1.09
10.94
140.5
Methanol
3.17
0.23
7.78
7.09
0.06
5.16
2.88
--
1.18
2.45
30.0
Mobil M
7.09
0.38
15.39
15.68
0.09
11.35
1.54
--
2.42
4.90
58.8
Controlled Emission
(kg/hr)
Fischer-
Tropsch
0 - 4.19
0 - 0.44.
3.47
9.63
0.14
9.41
0 - 3.84
0-0.86
1.09
6.35
30.1 - 39.4
Methanol
0-0.79
0 - 0.19
0.78
1.84
0.06
1.96
0 - 0.86
--
1.18
1.42
7.2 - 9.1
Rates
Mobil M
0 - 1.77
0-0.32
1.54
4.08
0.09
4.31
0 - 0.46
--
2.42
2.84
15.3 - 17.8
* ~~~ " ~~ — ~~
Uncontrolled emission factors for pumps and compressors represent emissions from each pump and compressor and
not from each pump seal and compressor seal.
fThese factors are averages for all sizes of the items indicated. Data are not sufficient at present to allow
emission rates to be related to equipment sizes.
-------
Section 4
Fugitive VOC Emissions
Fugitive Organic Emissions
costs correspond to 0.004 to 0.011% of the base plant capital investment and
0.007 to 0.025% of the base plant annualized costs.
The second approach goes a step beyond the first in that it relies on
equipment specification in addition to leak detection, repair, and main-
tenance. Monitoring requirements are similar to those for the first approach
except in cases where equipment specification eliminates the need for monitor-
ing. Typical equipment specifications can consist of caps for open-ended
valves, rupture disks on gas service relief valves, and double mechanical
seals with a seal oil flushing system on pumps. In addition, compressor seal
areas and degassing vents from oil reservoirs seals can be connected to a
control device with a closed vent system. As a result of these equipment
specifications, fugitive emissions from pumps, safety/relief valves, com-
pressors, and sampling connections can be completely controlled.
On applying these specifications to the K-T based indirect liquefaction
facility, an emission reduction of approximately 80% can be achieved. The
capital investment and annualized costs for these controls are estimated to
be from $0.3 million to $1.9 million and $0.09 million to $0.75 million,
respectively, as listed in Table 4-23. These costs correspond to 0.03 to
0.14% of the base plant capital investment and 0.03 to 0.19% of the annualized
base plant cost.
272
-------
TABLE 4-23. CAPITAL AND ANNUALIZED COSTS FOR FUGITIVE ORGANIC EMISSION CONTROLS
Fi scher-Tropsch
Cost Item
Capital
Cost ($)
Annual ized
Cost ($)
Leak Detection,
Repair, and
Maintenance
145,000
102,200
Equipment
Specification
1,876,300
754,200
Methanol
Leak Detection,
Repair, and
Maintenance
41 ,400
24,900
Equipment
Specification
311,900
93,900
Mobi
Leak Detection,
Repair, and
Maintenance
69,700
43,600
1 M
Equipment
Specification
531,400
178,000
-------
Section 4
Fugitive VOC Emissions
Fugitive Organic Emissions
4.1.6 Fugitive Particulates from Material Conveying and Processing
Material transfer and conveying operations are common to nearly all
processing industries. Equipment includes belt conveyors, screw conveyors,
bucket elevators, vibrating conveyors, and pneumatic conveyors. The type
of conveying equipment varies with the application and is determined pri-
marily by the quantity and characteristics (size, specific gravity, moisture
content, etc.) of the material being handled, the transfer distance and ele-
vation, and the working conditions. Loss of material from conveyors occurs
primarily at the feeding, transfer, and discharge points due to spillage or
wind. The majority of particulate emissions are generally from spillage and
mechanical agitation of the material at transfer points.
Material from storage piles is generally crushed, screened, and pulver-
ized prior to transfer to the boiler or gasification plants. Fugitive dust
generated during this process is typically controlled by either wet suppres-
sion techniques or dry particulate collection systems.
Fugitive particulate control systems utilizing a wetting agent consist
of pre-engineered modules which incorporate both water handling components
and automatic spray controls. A typical spray solution contains 1,000 to
4,000 parts of water to one part of a wetting agent. The rate of spray
application is about 4 to 8 liters/Mg of material. This rate of application
results in an increase of total surface moisture by about 0.5 to 1.0%.
In wet dust suppression, the fugitive particulates are first confined by
a curtain of moisture droplets. Then the wetting of dust takes place by con-
tact and penetration with moisture droplets. Finally, agglomerates are
formed by contact with other droplets and settling takes place because of
the additional weight of the other droplets. Wet suppression techniques can
cost from $0.33 to $0.77/Mg of material treated depending upon the wetting
agent utilized.
274
-------
Section 4
Fugitive VOC Emissions
Fugitive Organic Emissions
Dry particulate collection systems consist of enclosures to contain the
particulates, ductwork and exhaust systems to convey the particulate laden
air, and particulate collectors to separate the particulate from the air.
Typically, hoods are used to capture particulate emissions at transfer points.
Conveyors generally have a half cover which provides dust containment and also
shields the conveyor from wind, rain, and snow. The type and size of enclosure
depends upon the particulate source. Data on ductwork velocities needed for
particulate capture for different source types are readily available (94).
Dust collectors that are applicable to the collection of the captured parti-
culate are: (1) venturi scrubbers, (2) electrostatic precipitators, (3)
fabric filters, and (4) dry centrifugal collectors. These have been discussed
previously in Section 4.1.2.
275
-------
Section 4
Aqueous Medium
4.2 AQUEOUS MEDIUM
A variety of wastewater streams are generated by a K-T gasification facil-
ity. Individual streams and their characteristics are summarized in Table
4-24. The shift condensate (Stream 218) and methanation condensate (Stream
236) contain only carbon dioxide and, therefore, probably require little or
no treatment prior to being recycled to the plant as makeup water or otherwise
handled. These two streams are not considered in this section. All other
wastewater streams are classified in Table 4-25 as either streams of an in-
organic or organic source type. The inorganic streams are further identi-
fied according to the predominating species present: reduced volatile,
reduced nonvolatile, and oxidized inorganics.
This section considers the control of only those wastewater streams that
significantly influence the wastewater treatment approach required for the
entire K-T gasification facility. In Section 4.2.1, a summary of control
technologies potentially applicable to these wastewater streams is presented.
In Section 4.2.2 and 4.2.3, performances of individual controls applied to
specific wastewater streams of either the organic or inorganic source type
are discussed.
The overall control of wastewater streams generated by a K-T gasifica-
tion facility generally involves more than one control process applied to
either an individual stream or to a composite of several streams. Perfor-
mance and costs of individual controls are most meaningfully assessed in the
context of integrated trains, but exact definition of these integrated trains
can be made only when data specific to a particular facility are available.
However, certain combinations of controls are believed to be likely, and
integrated control examples are therefore presented in Section 4.2.4 along
with corresponding performances and costs. These combinations are intended
as examples only and are not to be construed as EPA recommendations.
276
-------
TABLE 4-24. SUMMARY OF K-T BASE PLANT WASTEWATER STREAMS AND ESTIMATED CHARACTERISTICS
ro
Flowt
Characteristic* m3/hr CT CN" SCN- S~ S^/SOj
Wastewater Streams from
ROM Coal Storage
Pile Runoff
(Stream 201)
- Methanol Synthesis
- Mobil M Synthesis
- Fisher-Tropsch
Synthesis
Wastewater Streams from
Gas Cooling/Dust
Removal Blowdown
(Stream 210)
Raw Gas Compression
and Cooling Conden-
sate (Stream 211)
Cyanide Wash Water
(Stream 215)
Cyanide Wash
Still Bottoms
(Stream 213)
Shift Condensate
(Stream 218)
Rectisol Condensate/
Still Bottoms
(Stream 220)
Coal Preparation
151 -5
146
174 - -
Gas Purification and Upgrading
322 2111 7 4 1 128
17.4 2200 8.9 14 48.7 6.3
239 - 241 Present 176
0.001 - 10 Present
180 - -
1.2 - 10 -
„ Trace
NH, TDSff COD TOD TSS Elements
Present Present
Present Present
Present Present
156 4000 113 4 Present Present
735 - 124 7
500 111
65-
-
6 5 -
(Continued)
-------
TABLE 4-24. (Continued)
Flowt Trace
Characteristic* nvVhr CT CN~ SCN" S= S2°3/S03 NH3 TDS COD TOC Tss Elements
Wastewater Streams from
Methanol Distillation
Condensate
(Stream 229)
F-T Wastewater
(Stream 223)
Mobil M Wastewater
(Stream 233)
Methanation
Condensate
(Stream 236)
PO
co Wastewater Streams from
Demineral izer
Regeneration
Wastewater
(Stream 301)
Synthesis of Liquid Fuels
10 - - - 33000 4400
160 - - - - - - Present 12000 4300 Present
110 - - - - - - - 14000 4000
12.9 ----_._.
Additional
18 167 - - - - - 7200 _ - - -
Boiler Slowdown 2.1
(Stream 303)
Cooling Tower Blow-
down (Stream 307)
- Methanol Synthesis 161
- Mobil M Synthesis 180
- F-T Synthesis 236
400 Present Present Present Present
Present
Present
Present
(Continued)
-------
TABLE 4-24. (Continued)
no
Flowt
Characteristic* m3/hr Cl" CN" SCN" S=
Miscellaneous Waste Streams
Process Equipment -
Cleaning Wastes
(Stream 242)
Boiler Cleaning -
Wastes (Stream 305)
Storm Runoff 47 _.
(Stream 314)
Plant Process Drain 32-42 - - -
(Stream 315)
Wastewater Streams from Air Pollution Control Processes
Wellman-Lord Con- 3
densate (Stream 411 )
Beavon Condensate 4 - . Present
(Stream 405)
SCOT Condensate 4 - - - Present
(Stream 409)
Stretford Solution 2-3 - - Present
Purge (Stream 405)
Flue Gas Desulfuri-
zation Purge
(Stream 425)
- Mobil M case 6.1
- F-T case 37.5 - -
S-O./SO^ NH, TDS# COD
f. J 3 J
Present Present
Present Present
- - - Present
Present
Present - Present Present
Present - Present
Present - Present
Present - Present Present
Present - Present Present
Present - Present Present
Trace
TOC TSS Elements
Present Present Present
Present Present Present
Present Present Present
Present Present Present
Present
-
.
Present
Present Present
Present Present
*A11 concentrations are mg/L.
'''For a plant with an input to the gasifier of 278 mg/hr of dry'Illinois No. 6 coal.
^Present but not readily quantified.
A dash indicates that this is not an important constituent/parameter in this stream.
*TDS is Total Dissolved Solids; COD is Chemical Oxygen Demand; TOC is Total Organic Carbon; TSS is Total Suspended
Solids.
-------
TABLE 4-25. CATEGORIZATION OF AQUEOUS WASJE STREAMS IN K-T GASIFICATION FACILITIES
Source Type
Stream Name
Factors Affecting Flow Rate
Pollutant Loading
and
Inorganic-predominantly
reduced nonvolatiles
ro
§
Inorganic-predominantly
reduced volatiles
Gas cooling/dust removal
blowdown (Stream 210)
Wellman-Lord condensate
(Claus tail gas treatment)
(Stream 411)
Stretford solution purge
(Stream 405)
Flue gas desulfurization
purge (Stream 425)
Cyanide wash water
(Stream 215)
Compression condensate
(Stream 211)
Beavon condensate
(Stream 405)
SCOT condensate
(Stream 409)
Species other than NH3 are believed to be gener-
ally independent of coal type. Ammonia produc-
tion generally increases with decreasing coal
rank. TDS and Cl" levels in the circulating
wash water determine the blowdown flow and these
constituents are related to the Cl" and ash
content of the coal.
Flow is design specific but not necessarily
related to coal rank. Loading is related to
coal sulfur content.
Flow and loading are related to coal sulfur
content.
Flow and loading are related to coal sulfur
content and boiler size.
Flow and loading are qualitatively similar for
most coal types.
Flow and loading are qualitatively similar for
most coal types.
Flow is design specific but not necessarily
related to coal rank. Loading is related to
coal sulfur content.
Flow is design specific but not necessarily
related to coal rank. Loading is related to
coal sulfur content.
(Continued)
-------
TABLE 4-25 (Continued)
Source Type
Stream Name
Factors Affecting Flow Rate and
Pollutant Loading
Inorganic-predominantly
oxidized species
rv>
oo
Organic
Cooling tower blowdown
(Stream 307)
Coal storage pile runoff
(Stream 201)
Demineralizer regenera-
tion wastewater
(Stream 301)
Boiler blowdown
(Stream 303)
Boiler cleaning wastes
(Stream 305)
F-T synthesis condensate
(Stream 223)
Mobil M synthesis conden-
sate (Stream 233)
Methanol distillation
wastewater (Stream 229)
Cooling tower blowdown is not dependent on coal
type. Flow and loading are related to the
makeup water quality, the amount of wet versus
dry cooling in the base plant, the climatic con-
ditions at the facility, cycles of concentra-
tion, and quantity of inhibitory chemicals added.
Coal type and conditions of wastewater contact
with coal (e.g., residence time and temperature).
Rainfall rates, coal storage, and washing
practices.
Makeup water flow and characteristics.
Size of the boiler and boiler operating pressure.
Plant operating, maintenance, and boiler clean-
ing practices; boiler size.
Flow and loading are independent of coal type
fed to the gasifier. Type of organics depend on
the operating conditions and by product recovery.
Flow and loading are independent of coal type.
Flow depends on C02 in the synthesis gas.
(Continued)
-------
TABLE 4-25 (Continued)
Source Type
Stream Name
Factors Affecting Flow Rate and
Pollutant Loading
Organic (continued)
Rectisol still bottoms
(Stream 220)
Cyanide wash still
bottoms (Stream 213)
Process equipment clean-
ing wastes (Stream 242)
Storm runoff (Stream 314)
Flow and loading are not specific to coal type.
Flow and loading are not specific to coal type.
Waste stream source; plant operating, maintenance
and equipment cleaning practices.
Climatic conditions; housekeeping and maintenance
practices.
rvs
00
Plant process drain
(Stream 315)
Total plant water makeup; housekeeping and main-
tenance practices.
-------
Section 4
Aqueous Medium
The following wastewater streams are believed to most directly influence
the overall wastewater treatment approach taken at a K-T gasification facil-
ity: the gas cooling/dust removal blowdown (Stream 210) of the reduced non-
volatile inorganics source type; the cyanide wash water (Stream 215) and the
compression condensate (Stream 211) of the reduced volatile inorganics source
type; and the three synthesis condensates (F-T, Stream 223; Mobil M, Stream
233; and methanol distillation, Stream 229) of the organic source types. The
waste streams containing primarily inorganic contaminants (Streams 210, 211,
and 215) are generally considered unique to K-T indirect liquefaction facil-
ities. Streams similar to the synthesis condensates (Streams 223, 229, and
233) would be generated regardless of the type of gasifier employed. The gas
cooling/dust removal blowdown is a major source of wastewater flow and con-
tains significant concentrations of ammonia, chloride, TDS, and reduced sul-
fur species other than sulfide. For base plant designs where cyanide is re-
moved in the gas upgrading section by a water-based wash (see Section 3.3.4),
the cyanide wash water is another major source of wastewater flow. This
stream contains most of the cyanide that was originally in the gasifier off-
gas and sulfide. Condensate from the raw gas compression and cooling circuit
is a low flow stream but contains significant loadings of ammonia, chloride,
and cyanides. The methanol distillation condensate and the F-T and Mobil M
synthesis condensates contain varying loads of organics, typically ketones,
organic acids, and/or methanol.
All other wastewater streams generated by the K-T gasification facility
should not influence the overall wastewater treatment approach. Each would
be routed to existing controls designed for other, more heavily loaded waste-
water streams. These streams would contribute only a small fraction of the
total load and flow such that the performance and economics of individual
controls would be affected only slightly.
The remaining wastewater streams probably do not influence the facility-
wide wastewater treatment approach, and are generally not unique to K-T
283
-------
Section 4
Aqueous Medium
gasification facilities. These streams include the Beavon (Stream 405) and
SCOT (Stream 409) condensates of the reduced volatile inorganics source
type; the Wellman-Lord condensate (Stream 411), Stretford solution purge
(Stream 405), and flue gas desulfurization purge (Stream 425) of the reduced
nonvolatile inorganics source type; the cooling tower blowdown (Stream 307),
ROM coal storage pile runoff (Stream 201), demineralizer regeneration waste-
water (Stream 301), boiler blowdown (Stream 303), and boiler cleaning wastes
(Stream 305) of the oxidized inorganics source type; and the process equip-
ment cleaning wastes (Stream 242), storm runoff (Stream 314), and plant pro-
cess drain (Stream 315) of the organic source type. Certain of these streams
have small flows/loadings and would be routed to existing controls designed
for larger wastewater streams with no significant effect on the performance
or economics of the control. Others, such as the ROM coal storage pile run-
off or the cooling tower blowdown, have large flows/loadings but have char-
acteristics that are not amenable to treatment by the controls for other
streams. Streams of this type would probably be handled by separate controls.
None of these streams are unique to K-T gasification facilities, and their
control has been well developed in parallel industries, particularly petro-
leum refining and electric utilities.
284
-------
Section 4
Aq. Med. Gen. Control
4.2.1 Water Pollution Control Processes
This section presents general information for a variety of water pollu-
tion control processes which are potentially applicable to the treatment,
disposal, and/or reuse of wastewaters from K-T-based gasification plants.
The information presented in this section is derived from industrial appli-
cations and laboratory tests with various wastewaters. The water pollution
control technologies discussed do not represent an all-inclusive list of
processes, as others may be available or may be developed. Rather, these
processes represent control alternatives which have been commercially
applied in analogous applications or have potential for commercial applica-
tion. Table 4-26 lists the treatment processes which are potentially appli-
cable to wastewaters from the K-T facility and presents summary information
about each. This section expands on the summary information for some of the
controls.
The water pollution control processes discussed include techniques for:
• removal of suspended solids, tars, and oils;
0 removal of bulk organics;
t removal of dissolved gases;
• removal of dissolved inorganics;
• removal of dissolved organics;
• removal of residual organics;
• volume reduction; and
• residual disposal.
A more detailed discussion of each may be found in the Control Technology
Appendices.
285
-------
TABLE 4-26.
CONTROL PROCESSES POTENTIALLY APPLICABLE TO THE TREATMENT OF K-T-BASED
GASIFICATION PLANT WASTEWATERS
00
CTi
Technology
Removal of Suspended
Gravity separation
- 'API oil separator
- parallel plate
Coagulation/
flocculation
Air flotation
- dissolved air
- induced air
Fil tration
Technology Principle
Solids, Tars, and Oils
Provision of adequate
residence time in a
stagnant vessel to
allow suspended solids
or immiscible fluids
to separate into
lighter and heavier
than water phases
Use of agents to pro-
mote the coalescence
of fine suspended
solids and adsorption
of tars and oils;
generally used in con-
junction with a gravity
separation process
Use of air bubbles to
promote the disengage-
ment of 1 ighter-than-
water materials from
solution
Passing wastewater
throujii suitable
filter medium; fil-
ter material dis-
carded or cleaned
by backflushing
Components
Removed
Suspended solids,
tars, and oils
Promotes removal
of finely dis-
persed particles
Suspended oils
and solids
Depends on filter
medium; both
coarse and fine
structure mate-
rials are used
industrially
Removal
Efficiency
Depends upon
design, 10-50 i
removal of TSS
typical , 60-991
for oils
Outlet suspended
solids concen-
tration to 10
mg/L possible,
oils removal of
60-95%
Depends on char-
acteristics of
source and treat-
ment process, TSS
removal of 20-75*=,
oil removals of
75-85i
TSS removals of
30-90+',, oil
removals of
65-90:,
Feed
Requirements/
Restrictions
Minimum feed
stream
turbulence
A wide range of
commercial floc-
culants are
available
Air requirements
depend upon waste
characteristics
Filter media
By-Products
and Secondary Estimated
Waste Streams Capital Cost
Recovered oils
(Sp. gr. <1),
sludges/solids
(Sp. gr. ^1)
Same as gravity
separation
Recovered oils,
entrained solids
Filter back- $1700 per m3/hr
wash; spent
filter media
Comments
Incorporated into the
tar/oil separation sys-
tem design in all exist-
ing Lurgi gasffication
plants.
Widely used in water
treatment system to
remove fine solids.
Of limited usefulness
in treating gasifica-
tion plant wastewaters
Proposed for use as pol-
ishing step for phenolic
water downstream of tar/
o: i separation; "Sticky"
tars/oils may cause prob-
lems with filter plugging
and regeneration.
Removal of Bulk Organics
Solvent extraction
- Phenosolvan
- Chem-Pro
Extraction of organics
from wastewater via
contact with an iimis-
cible solvent. Byprod-
uct organic 1 iquids
recovered from the sol-
vent in a separate
regeneration step.
Phenols, TOC,
BOD, COD, other
'
Phenosol van .
- monohydric
ohpnol 99 ^t
- polyhydric
phenol 60'
- organic acids
155,
Chem-Pro
- monohydric
phenol 99.8°;
- dihydric
phenol 95;.
- trihydric
phenol 90-95'.
- other organics
50+
Sensitive to
suspended matter
makeup solvent
Crude phenols,
filter back-
f liter media
Phenosolvan is the process
used in all major Lurgi
•--•••- • *- -* -
Chem-Pro process is com-
mercially applied in phenol
formaldehyde resin manu-
facturing plants. First
application in 1961 to
recover phenol from coke-
oven waste 1 iquor.
(Continued)
-------
TABLE 4-26 (Continued)
ro
CO
Technology
Wet air oxidation
Removal of Dissolved
Steam stripping
- Phosam-W
- Chevron WWT
Inert gas stripping
Vacuum distil lation
Selective
absorption
Technology Principle
Direct reaction of 0,,
with wastewater in a
closed, pressurized
vessel at elevated
temperatures
Gases
Increasing temperature
and providing a posi-
tive flow of inert
material (steam)
through the waste-
water; removes vol-
atile organKS and
inorganics with
overhead steam
Same as with steam
except that an inert
gas such as N2, air or
C02 is used as the
stripping medium
Low pressure, low
temperature stripping
process
Used in conjunction
with one of the above
processes to selec-
tively absorb
stripped gases
Components
Removed
Same as biological
oxidation except
better destruction
of cyanides and
other difficult
to treat organics
can be achieved
NH-,, acid gases
(CO,, H2S, HCN),
light hydrocar-
bons (phenols)
Same as for steam
stripping
Same as for steam
stripping
NH, (with acidic
solutions), acid
gases (with basic
solutions)
Removal
Efficiency
90+T removal of
COO is possible
in a system with
a residence time
of 1 hour or
greater
95-99S removal
of "free" ammo-
nia and acid
gases typical ;
hydrocarbon
removal varies
with volati lity
of stripped
components
Removal to:
150 mg/L NH3
1 mg/L HjS , C02.
95* HCN removal
Removal to:
50 mg/L NH3
5 mg/L H?S
Same as for
steam
Same as for
steam
9Q+1 removal of
acid gases or
NH, is typical
Feed
Requirements/
Restrictions
Air or oxygen,
heat if auto-
thermic reaction
conditions are
not present
Feed preheat
can be used to
reduce steam
requirements ;
acid/caustic
for pH adjust-
ment optional
Generally oper-
ates at lower
temperature than
steam stripping
process
Same as for inert
gas stripping
Makeup acid/
caustic
By-Products
and Secondary Estimated
Waste Streams Capital Cost
Vent gases con-
taining CO, CO™
1 ight hydrocar-
bons, NH,, sulfur
species
Stripped gases,
uncondensed steam
Stripped gases,
stripping gas
Stripped gases
Recovered NHj
or acid gases,
rich sorbent
Coninents
Piomising but not proven
in this appl ication,
fairly rigorous materials
of construction
requi r ements
Acid/caus t i c addition can
be used to improve the
efficiency of the strip-
ping process
The presence of an inert
stripping gas as a diluent
may make it more difficult
to handle or further treat
the gases stripped from
the wastewater
High energy requirements;
not cost competitive in a
plant where stripping
steam is readily available.
This is the basis for sev-
eral commercially proven
processes for recovering
high purity aninonia.
(Continued)
-------
TABLE 4-26 (Continued)
ro
oo
oo
Technology
Technology Principle
Components Removal
Removed Efficiency
Feed
Requi rements/
Restrictions
By-Products
and Secondary Estimated
Waste Streams Capital Cost
Comments
Removal of Dissolved Inorganics
Ion exchange
Chemical
precipitation
Polysulfide
addition
Activated sludge
Biological
denitrification
See also: chemical
Substitution of H*
Na+, OH-, or Cl- ions
for other ionic species,
exchange resins regen-
erated with acid, base.
or salt solutions
Use of agents to pro-
mote the precipitation
of inorganic solids
from wastewaters
Cyanide converted to
thiocyanate by
polysulfide
Microorganisms mediate
oxidation of ammonia
and thiocyanate
Microorganisms reduce
nitrate to molecular
nitrogen in the pro-
cess of oxidizing
methanol or some other
organic
oxidation, membrane separation
Heavy metal, F-, 90+% removal for
CN", scaling spe- most ions, regen-
cies, NH3 eration frequency
is a key
parameter
Ca, Mg, heavy Varies with waste
metals, stream constitu-
alkalinity ents. Typical
removals:
Cd 2% Ni 50%
Cr 40* Pb 5%
Cu 20% Se 10%
Hg 20% In 25%
CN' Varies
NH3, SCN' 90+% for NH,,
SCN-
NO" 90+3.
, and forced evaporation.
Regenerants ,
replacement
resins
Lime, polymer,
and soda ash
may be required.
Possible chem-
ical require-
ments for pH
control .
Air or oxygen,
supplemental
nutrients may be
required; rela-
tively constant
feed temperature
and pollutant
loadings
required.
Nutrients; con-
stant loading
conditions
Spent regencr- . $5700 per kg
ants and resins, NH3-N per day
treated water
Sludge contam-
inated with
heavy metals
Residual poly- $530 per m3/hr
sulfide;
increased 1%
concentration
due to NH| as
counter ion to
polysulfide
Biological $4400 per
oxidation equivalent kg
sludge NHo-N per day
Biological $2300 per kgN
oxidation per day
sludge
Most effective as a pol-
ishing process. Clearly
applicable to boiler
feedwater treatment needs.
Of limited use in treating
process watewaters contain:
ing high concentrations of
organics or dissolved solids
Generally followed by fil-
tration and/or activated
carbon adsorption.
This process utilized at
petroleum refineries to
control cyanide- induced
scaling. No experience
exists at the wastewater
treatment process level.
•
Used extensively to treat
wastewaters of a wide range
of characteristics and
sources.
Experience exists in both
municipal and industrial
applications.
(Continued)
-------
TABLE 4-26. (Continued)
ro
oo
Technology
Removal of Dissolved
Biological
oxidation
- act. sludge
- trickling filter
- rotating
biological
contactor
- lagoons
- high purity
oxygen (HPSAS)
Wet air oxidation
Ultrafiltration
Anaerobic digestion
Technology Principle
Organics
Biological conversion of
the carbonaceous organic
matter in wastewater to
cell tissue and various
gaseous end products.
Direct reaction of On
with wastewater in a
closed, pressurized
vessel at elevated
temperatures
Forcing wastewater
through semiperme-
able membrane under
pressure
Reduction of organics
in closed vessel at
moderate temperatures
to form CH^ and other
hydrocarbons, digestion
process relies upon
metabolic processes of
anaerobic organisms
Components
Removed
TOC, BOD, COD,
some inorganic
pollutants
Same as biologi-
cal oxidation
except better
destruction of
cyanides and
other difficult
to treat organics
can be achieved
Most effective
with high molec-
ular weight
organics
TOC, BOD, COD
Removal
Efficiency
Varies with waste
stream character-
istics. Typical
removals:
BOD 60-90=-
COD, TOC 80'-
total phenol 9K
org acids 95',
oil S grease 70*
CN- 70;
Over 90% removal
of COD is possible
in a system with a
residence time of
1 hour or greater.
Actual efficiency
heavily dependent
upon waste char-
acteristics.
Up to 955,
removal for total
organics
Unknown - this
process has not
been applied to
these types of
waste streams
Feed
Requirements/
Restrictions
Air or oxygen,
supplemental
nutrients may be
required; rela-
tively constant
feed temperature
and pollutant
loadings required
to minimize
"shocks" to
system
Air or oxygen;
heat if auto-
thermic reaction
conditions are
not present
Filter media
Some supplemen-
tal nutrients
may be required
By-Products
and Secondary Estimated
Waste Streams Capital Cost
Biological oxi- Activated
dation sludge sludge-
$700 per kg BOO
per day
Vent gases con-
taining CO, CO,,
light hydrocar-
bons, NH-i , sulfur
species
Spent filter
media; concen-
trated
wastewater
Waste gases
(combustible)
Comments
This is the basis for the
treatment of coke oven
wastewaters
Promising but not proven in
this application; fairly
rigorous materials of con-
struction requirements.
May be attractive as a pre-
concentration step prior to
wastewater incineration.
Kinetic limitations and
process control problems
could be substantial .
(Continued)
-------
TABLE 4-26 (Continued)
f\>
<£>
o
Technology
Removal of Residual
Activated carbon
adsorption
Chemical oxidation
Thermal oxidation
( Inc i neratior ^
Cooling tower
oxidation
Volume Reduction
Me- brane separation
- reverse osmosis
- elect rodi a lysis
Technology Principle
Orqanics
Adsorption of orgamcs
in water by activated
carbon or polymeric
resin, powdered acti-
vated carbon has been
used in conjunction
with biological pro-
cesses (above) with
some success in the
organic chemical
industry
Reaction of organics
in wastewater with
ozone, peroxides or
chlorine-based
ox idants
Combustion of orgamcs
ai r oxidation (and
stripping) of orgamcs
and dissolved gases
Use of seiupem.eable
membrane and pressure
to separate water from
its dissolved
constituents
Use of selective
anion- or cation-
permeable membranes
with electric field to
separate mineral ions
from water
Components
Removed
Most effective
with phenols;
some heavy metal
removal expected
TQC, BOD, COD.
oxidi zable
inorganics
All oxidizable
orgamcs
TOC, COD, BOO,
phenols and
other orgamcs,
NH3
Relative rejec-
tion efficiencies
of the various
soluble species
wi 1 1 be deter-
mined by membrane
characteristics
and conditions of
process operation
Removal
Efficiency
Varies with
waste stream
charactenst ics.
Typical removal s :
BOD 60%
COD 80
TOC 70
Phenols 99.9
org acids 70
CfT 50
SCN- 50
tars 99
oils 99
High removals
achievable depend-
ing upon condi-
tions of operation
Essentially com-
plete destruction
of organics in
properly designed
system
Unknown - this
process has been
tested using
5ASOL waste-
waters on a pi lot
scale but results
are not available
90-95 rejection
of dissolved
caits. Reduc-
tions in dis-
solved organics
and BOD of up
to 99-
Feed
Requirements/
Restrictions
Low TSS, pH 6-9;
high TOC loadings
are less amenable
to treatment.
Oxident
Supplemental
fuel , preconcen-
tration wil )
improve perfor-
mance and lower
supplemental
fuel
requi renents
Sensitive to
suspended matter,
oils, high TDS
Membrane
By-Products
and Secondary Estimated
Waste Streams Capital Cost
Spent adsorbent $260 per kg COD
regeneration per day
off -gases
Vent gases, $850 per m^/hr
wastewater, and (Cl2 and lime
reaction basis)
products
Flue gases $300,000 per
m3/hr
Bl owdown/dnft
Spent membrane
material , recov-
ered water.
brine
Comments
Probably more effective as
a polishing rather than a
bulk orgamcs removal
process.
Chlorine-based oxidants
may cause problems with
treated wastewater.
This will be the most
effective process for
removing orgamcs but
the supplemental fuel
requirements may be
substantial .
Treated wastewater use ir
small refinery cooling
towers has be practiced -
75-80 reduction in waste-
water volume is common
May be useful as precon-
centration step prior to
further treatment or ulti-
mate disposal of waste-
water. Membrane scaling
and fouling with organics
may limit the applicability
of this technology to the
treatment of process
condensates.
(Continued)
-------
TABLE 4-26 (Continued)
Technology
Forced evaporation
Cooling tower
concentration
Residual Disposal
Deep well Injection
Surface Impoundment
Co-disposal
Surface discharge
Technology Principle
Thermally induced evap-
oration of excess waste-
water, condensate recov-
ery optional
Wastewater used as
partial makeup to the
cooling tower and
thereby concentrated
into the blowdown.
Wastes are pumped into
subsurface geological
formations where they
are isolated from all
surface and ground-
water supplies
Wastes are held in
a containment basin
Ash is quenched with
wastewater then han-
dled by a solid
waste disposal
technique
Wastes are conveyed to
and mixed with a nat-
Components Removal
Removed Efficiency
All nonvolatile
species will
remain in brine
All nonvolatile
species concen-
trated into the
blowdown
Entire stream IOC'S
All nonvolatile 100%
species remain as
residual in the
impoundment
Entire stream lOCrt
Entire stream 100*
Feed
Requirements/
Restrictions
None
Feed character-
istics limited by
corrosion, seal-.
ing, and biologi-
cal fouling
Injected fluids
must be filtered
to 5 micrometers
and have a low
organic content
to prevent plug-
ging. Wastes
must not precip-
itate in the well
or when mixed
with subsurface
fluids. Volume
reduction prior
to injection is
often
economical .
Concentrations
of volatile
species may need
to be low to pre-
vent loss to the
atmosphere
Concentrations
of volatile
species may need
to be low to pre-
vent loss to the
atmosphere
Restrictions are
site-specific
By-Products
and Secondary Estimated
Waste Streams Capital Cost
Recovered $86,000 per
condensate, non- m^/hr feed
condensible
gases, waste
brine
Evaporation/
drift
None $160,000 per
m3/hr
Loss of $52,000 per
volatile nrVhr
species
Loss of
volatile
species
None
Comments
Very stringent materials
requirements due to
potential corrosion.
Forced evaporation 1s
energy intensive.
Applicability highly site
specific.
Possibly large land
requirements may limit
applicability.
Presence of certain chem-
ical species in the quench
water may yield an ash that
requires handling as a haz-
ardous material .
Assimilative capacity of
receiving body must be
ural water source
investigated.
-------
Section 4
Aq. Med. Gen. Control
Sus. Sol., Tars, Oils
4.2.1.1 Processes for Removal of Suspended Solids, Tars, and Oils
A variety of control processes have been summarized in Table 4-26 that
are applicable to waste streams having high concentrations of suspended
solids, tars, and/or oils. The process or combination of processes selected
for a particular waste stream would depend on the concentration of the con-
taminant species and its physical properties such as size distribution, spe-
cific gravity, and surface properties. This section summarizes available
information for filtration as an example of a control process for removal of
suspended solids, tars, and oils. .
Filtration is a well-established technology that finds nearly industry-
wide application. The process consists of passing a waste stream by either
gravity or pressure through a bed of inert material which physically retains
the solids suspended in the flow. Various materials and combination of
materials have been used for filter media including sand, anthracite, acti-
vated carbon, natural/synthetic fibers, resins, and garnet. Performance
varies with different media materials, and the material giving optimal per-
formance must generally be determined by direct testing of the subject waste
streams.
During the course of filtration, the bed becomes increasingly loaded
with suspended solids resulting in a corresponding increase in the hydraulic
headloss. At some point, the bed must be backwashed to return it to a condi-
tion approximating its original, unused state. About 2 to 10% of the filter
throughput is needed for backwash, and this amount is stored during the filter
run. Thus, suspended material contained in the waste stream is concentrated
by a factor of 10 to 50.
Suspended solids removal by filtration typically range from 30 to 70
percent. Exact performance can be determined only by testing the waste
stream to be treated.
292
-------
Section 4
Aq. Med. Gen. Control
Bulk Organics
4.2.1.2 Processes for Removal of Bulk Organics
For waste streams having high organic loadings, it is sometimes advan-
tageous to apply a process capable of achieving gross organic removal.
These processes can typically achieve high percentage removals, but since
the feed stream is heavily loaded, an appreciable concentration of the con-
taminant species remains in the effluent. Other processes that are capable
of reducing the contaminants to the low levels possibly required for ultimate
disposal can follow these, if desired.
Processes for removal of bulk organics are presented in Table 4-26.
These processes are not expected to be generally applied to any of the waste-
water streams generated by a K-T gasification facility.
293
-------
Section 4
Aq. Med. Gen. Control
Dissolved Gases
4.2.1.3 Processes for Removal of Dissolved Gases
Dissolved gases such as HCN, I^S, and NhU can be removed from waste-
waters by several processes. Some of these processes are summarized in Table
4-26. Most are designed to handle waste streams having high loadings of dis-
solved gases. High percentage reductions are possible in such cases, but the
concentration of dissolved gas remaining typically requires additional con-
trols of the waste stream. Residual levels achievable with techniques such
as steam or inert gas stripping are similar to those already found in K-T
gasification wastewaters (except cyanide wash water).
In the case of cyanide wash water (Stream 215) which is expected to con-
tain several hundred ppm each of HCN and H^S, stripping with inert gas may
be applicable. Potential stripping gases which would probably be available
onsite include waste nitrogen from an air separation plant or waste CC^ from
a selective Rectisol unit. The HCN laden offgas could be incinerated or sent
to a Claus plant for treatment.
294
-------
Section 4
Aq. Med. Gen. Control
Dissolved Inorganics
4.2.1.4 Processes for Removal of Dissolved Inorganics
A variety of processes summarized in Table 4-26 are applicable 'to waste
streams requiring removal of dissolved inorganics. Processes such as chemical
precipitation, membrane separation, and chemical oxidation are capable of
removing simultaneously a wide variety of inorganic species. Others are nar-
rowly applicable to one or a few specific species. These processes include
ion exchange using a clinoptiloli te resin for ammonia removal, polysulfide
addition for cyanide removal, activated sludge for ammonia or thiocyanate
removal, and biological denitrification for nitrate removal. The removal
mechanisms vary widely between individual processes. Polysulfide addition,
activated sludge, biological denitrification, and ion exchange using a clino-
ptiloli te resin are presented herein as examples of processes for removal of
dissolved inorganics.
Polysulfide Addition
Cyanide can be removed from wastewater streams by converting it to thio-
cyanate. The conversion is prompted by adding a sulfur source as polysulfide,
with only uncomplexed cyanide thereby removed. The product, thiocyanate,
would likely need to be removed, but unlike cyanide, it is easily biodegraded.
Recent bench-scale studies have demonstrated that thiocyanate removals exceed-
ing 95% can be made by a biological system where only the thiocyanate plus
ammonia are available as growth substrates.
Petroleum refiners have for some years used polysulfide to convert
cyanide to thiocyanate to protect process equipment from cyanide-induced cor-
rosion. No commercial precedent exists for promoting the reaction at the
wastewater treatment process level, but EPA-sponsored preliminary lab studies
(unpublished) suggested that the cyanide to thiocyanate conversion could be
successfully made if the polysulfide is supplied in large excess of its stoi-
chiometric requirement. The reaction rate seems to be sufficiently slow to
require large reactor volumes to supply the needed residence time.
295
-------
Section 4
Aq. Med. Gen. Control
Dissolved Inorganics
The concentration of cyanide in the influent has only limited effect on
the process feasibility. With increasing influent concentrations of cyanide,
the reaction kinetics become increasingly favorable. It is the concentration
of cyanide in the effluent that determines the residence time and therefore
the feasibility of the reaction. Preliminary EPA sponsored studies (unpub-
lished) suggest that the process may be impractical for cyanide removals to
less than about 10 mg/L.
Polysulfide addition therefore appears usable only for waste streams
where the initial cyanide concentration is greater than about 30 mg/L and some
benefit can be identified in reducing this concentration to about the 10 mg/1
level. Since additional cyanide removal by some other process would probably
be required, the feasibility of polysulfide addition would be determined by
the economic tradeoffs between it being used in combination with some other
process versus the other process being used alone.
Activated Sludge
Certain dissolved inorganics, particularly ammonia and thiocyanate, can
be biologically oxidized even in systems where these species are the only
growth substrates available for microorganisms. Most of the contaminants are
oxidized to supply energies for metabolism while the rest are assimilated as
new cell material. Microorganisms may be viewed as catalysts that mediate
the oxidation of material that would otherwise oxidize very slowly or not at
all.
A variety of physical configurations can be used to provide contact be-
tween the microorganisms and the contaminants to be removed, but suspended
growth or activated sludge systems are perhaps most common. The cells are
dispersed throughout a reactor volume where they contact the incoming waste
stream on a continuous basis. Long enough residence time is provided for the
microorganisms to degrade the contaminants to an acceptable level. Effluent
296
-------
Section 4
Aq. Med. Gen. Control
Dissolved Inorganics
from the reactor is routed to a clarifier where the suspended cells are
separated from the bulk fluid. Concentrated cells are drawn from the under-
flow of the clarifier, and a fraction are wasted equal to their increase in
mass resulting from cell growth in the reactor. The rest, constituting the
much larger fraction, are recycled to the reactor. By this process of cell
wastage and recycle, a constant.mass of cells can be maintained in the reactor
for a controlled period longer than the hydraulic residence time. The period
selected is one that is in phase with the optimal life cycle of the cells.
The cell residence time is therefore the key process variable as it controls
both the overall system performance and stability.
Activated sludge processes designed to oxidize ammonia to nitrate are
well established in both industrial and municipal applications. Ammonia
destruction at the 98% level are reportedly possible. Recent bench-scale
studies have indicated greater than 95% oxidation of thiocyanate and
ammonia for biological systems fed a waste stream containing only these two
species. This is especially significant since K-T gasification wastewaters
consist primarily of cyanides, thiocyanate, ammonia, and more readily oxidiz-
able species such as S~, SO^, and S203-
As is the case for all processes utilizing the growth reactions of
microorganisms, activated sludge processes are subject to upsets when the
characteristics of the feed stream vary beyond certain limits. However,
activated sludge processes designed to remove ammonia and thiocyanate may
be especially prone to upset. In general, most organics that are biodegrad-
able can be oxidized by a wide variety of microbial species in an activated
sludge system. Each specie has its particular range of conditions to which
it is tolerant, and the individual ranges overlap, forming a much wider
spectrum of tolerance than that represented by any one specie. However,
ammonia and thiocyanate are oxidized by only a very few, specialized micro-
organisms. A narrow range of tolerable conditions is represented by these
few species, and when conditions fall outside this range, the system fails.
297
-------
Section 4
Aq. Med. Gen. Control
Dissolved Inorganics
Another problem potentially affecting reliability is associated with
operation of the clarifier. Conditions that lead to malfunction of the
clarifier are not fully understood. However, where there is relatively con-
stant feed and where the system is otherwise carefully operated, activated
sludge should be a reliable process.
Biological Dentrification
In dentrification, nitrate is removed by being reduced to molecular
nitrogen which can be vented directly to the atmosphere. Physically the
system resembles that of activated sludge, but the removal mechanism is some-
what different. When oxygen is not available, most microorganisms can sub-
stitute nitrate as the terminal electron acceptor in their growth and energy
reactions. Nitrate is thereby reduced to molecular nitrogen in the denitri-
fication reactor not by being acted on directly but by participating in the
oxidation of some other organic substrate. If not otherwise available, a
substrate (or carbon source) must be supplied, and methanol is a common
choice. Other organic-laden wastes may be used as the carbon source to
advantage in a K-T facility.
While less experience exists for denitrification than for activated
sludge systems, the process is considered well developed. Nitrate removals
of up to 90% are reported in the open literature.
Denitrification is usually a reliable process due largely to the diverse
population of microorganisms that are cultured in the reactor. Most reli-
ability problems relate to settleability of biosolids to shock loadings, or
to very cold weather.
Ion Exchange
Ion exchange involves a reversible interchange of ions in solution with
other ionic species bound to a solid ion exchange medium. Clinoptilolite is
a naturally occurring ion exchange material that will selectively exchange
298
-------
Section 4
Aq. Med. Gen. Control
Dissolved Inorganics
its bound sodium or calcium ions for ammonium ions in the contacting solution.
Upon initial loading of fresh clinoptilolite, nearly all ammonia is removed
from the waste stream. With time, fewer exchange sites are available and
increasingly less ammonia can be removed. When the concentration of ammonia
escaping the process increases to an unacceptable level, the resin is replaced
and regenerated for reuse by contacting it with a strong brine solution.
Ammonia rich brine from regeneration is air stripped in a closed loop system.
The stripper off-gas is scrubbed by an acid wash to recover the nitrogen as
ammonium sulfate before the stripping gas is recycled.
Several complete ammonia removal/recovery systems are in operation for
municipal-strength wastewaters. One is a system that has been operated suc-
cessfully by the Tahoe-Truckee Sanitation Agency. No major problems have
reportedly occurred with either the ion exchange units or the regeneration/
ammonia recovery units. No known precedent exists for higher strength waste
streams. Preliminary lab studies sponsored by EPA suggest that ion exchange
would be a feasible process for treating waste streams having ammonia concen-
trations much higher than that normally associated with municipal wastewaters.
It is expected that ammonia can be removed to the 1 mg/1 level or less from
a waste stream containing ammonia at the 100 to 200 mg/1 level.
299
-------
Section 4
Aq. Med. Gen. Control
Dissolved Organics
4.2.1.5 Processes for Removal of Dissolved Organics
A variety of control processes are applicable to wastewater streams
requiring removal of dissolved organics. Activated sludge is a commonly
used process and is presented as an example in this section.
An activated sludge system designed to remove organics is identical in
principal to that discussed for removal of inorganic species (refer to Sec-
tion 4.2.1.4). Most biodegradable organics can be oxidized by a-variety of
microbial species. Performance varies widely depending on the characteristics
of the wastewater stream, but BOD removals of 80 to 90% or better are typical.
Nitrogen and phosphorous in forms usable by the microorganisms are required as
growth nutrients, and these elements are assimilated as new cell material.
Microorganisms that utilize organic substrates can sometimes oxidize or
otherwise remove a variety of inorganic species concurrently. Even some
species such as cyanide that are normally nondegradable can be oxidized.
Cyanide removals exceeding 90% have been reported in laboratory-scale tests.
The exact performance of an activated sludge system can be determined
with certainty only following bench-scale or pilot testing of the subject
wastewater stream. Removal of most inorganic species usually cannot be
assessed accurately until after the system is in place.
300
-------
Section 4
Aq. Med. Gen. Control
Residual Organics
4.2.1.6 Processes for Removal of Residual Organics
Residual organics occur in wastewater streams due largely to performance
limitations of upstream controls. When a biological system is the upstream
control, residual organics consist of biodegradable material and a larger
fraction of refractory organics that are resistant to biological oxidation.
Activated carbon, chemical oxidation, and thermal oxidation are presented
herein as examples of processes that can remove residual organics. Most of
these systems can also remove residual inorganic species.
Activated Carbon Adsorption
Activated carbon adsorption is a widely used method of removing residual
organics from industrial wastewaters. This technology may be applied by one
of two methods: 1) the use of powdered activated carbon in conjunction with
biological treatment or 2) the direct contact of granular activated carbon
with contaminated wastewaters. The direct contact method using granular
activated carbon is considered herein.
The activated carbon is contained in a packed bed through which the waste
stream is routed under either gravity or pressure flow conditions. Dissolved
organics (as well as possibly other dissolved species) are removed from the
waste stream by being attached to adsorption sites on the carbon. The adsorp-
tive capacity of the bed is a function of the organic loading, the contact
time in the bed, and the affinity of the carbon for the organics.
The carbon bed requires regeneration when its adsorptive capacity is
reached. Systems requiring more than 225 to 450 kg of carbon per day can be
economically regenerated on-site. Regeneration is a thermal process that re-
quires approximately 10 MJ per kilogram of carbon regenerated. Approximately
2 to 10 percent of the activated carbon is lost due to physical attrition and
oxidation. Both the thermal regeneration energy and makeup carbon required
to replenish regeneration losses are predominant factors in the overall eco-
nomics of carbon treatment processes.
301
-------
Section 4
Aq. Med. Gen. Control
Residual Organics
Removal of any contaminant species is highly stream- and carbon-specific.
A particular species may be sorbed by the carbon differently with varying
concentrations and characteristics of other species present in the waste
stream. Designs for activated carbon systems must therefore rely heavily on
direct testing of the actual wastewater stream.
Chemical Oxidation
Many chemical species are less reactive or otherwise less objectionable
to the environment when they are oxidized to their highest state. The process
of chemical oxidation consists of driving these reactions by adding a suitable
oxidizing agent.
Chemical oxidation is typically carried out in specially designed, closed
contactors when the oxidant is a gas and in stirred reactors when the oxidant
is a liquid. A variety of chemical oxidants can be used, but ozone, chlorine,
chlorine dioxide, and hydrogen peroxide have widest application.
Both organic and inorganic species are oxidized during chemical oxidation.
The extent of removal is highly dependent on the oxidant used, concentration
of the contaminant, complexity of the molecule to be oxidized, reaction time,
extent of reactor mixing, pH, and temperature.
Chemical oxidation is most commonly used to remove low level concentra-
tions of chemical species. For the stronger oxidants such as ozone and chlo-
rine, generally all materials in the waste stream are attacked, and each spe-
cies is removed approximately in order of its amenability to oxidation. Thus
if chemical oxidation is used with the intention to remove a particular spe-
cies, say cyanide, the presence of other oxidizable material in the waste
stream may greatly increase the demand for the oxidant above that which would
otherwise be required.
The equipment used for chemical oxidation with C12 and lime is typically
simple and therefore expected to be highly reliable. While no known experience
302
-------
Section 4
Aq. Med. Gen. Control
Residual Organics
exists for treating wastewaters from coal gasification facilities, the process
has been used extensively in other applications. Chemical oxidation and the
closely related process of disinfection have shown reliable performance treat-
ing municipal strength wastewaters over many years of service. Similar per-
formance is expected in coal gasification applications.
Thermal Oxidation
Thermal oxidation or incineration is a high temperature process for the
destruction of a variety of wastewater contaminants. Wastewater incineration
can be considered a combination of evaporation, pyrolysis, and oxidation
although oxidation is the primary process leading to the ultimate destruction
of most toxic pollutants. Pyrolysis, or destructive distillation, is a
process in which heat breaks down the waste material into simpler components
which can either be recovered or oxidized more easily than the original
material . Oxidation, or combustion, promotes the reaction of waste components
or pyrolysis products with oxygen to form such products as carbon dioxide,
water, and oxidized inorganics such as sulfates and nitrates.
Thermal oxidation is capable of treating a variety of organic-laden
wastes. Wastes having a high energy content are self-sustaining, that is,
their oxidation liberates sufficient energy to raise the waste temperature to
the temperatures required to sustain the oxidation reactions and drive them
to completion. Low energy wastes require addition of auxiliary fuel. Pre-
concentration of a low energy aqueous waste may also be used to reduce or
eliminate the need for an auxiliary fuel.
Thermal oxidation is capable of achieving essentially complete destruc-
tion of a wide range of otherwise difficult to treat organics. Most organics
can be completely destroyed at 1300K with two seconds residence time. Tests
with various pesticides such as DDT, malathion, and chlordane have shown 99.96
to 99.99 percent destruction. Tests at commercial hazardous waste incinera-
tors have shown over 99.99 percent destruction of PCBs. Many incinerators have
been operated reliably for years without shutdowns.
303
-------
Section 4
Aq. Med. Gen. Control
Volume Reduction
4.2.1.7 Processes for Volume Reduction
Most volume reduction processes are capital and energy intensive, but
their application can sometimes be justified by the economics of ultimate
disposal. An exception is cooling tower concentration which has little or no
associated capital charge since the cooling tower is an integral part of the
base plant. Cooling tower concentration and forced evaporation are presented
as examples of volume reduction processes.
Cooling Tower Concentration
Based upon the availability of cooling water at the plant site and the
overall wastewater treatment strategy, process wastewater may be used as par-
tial makeup to the cooling towers. This practice results in a reduction in
wastewater volume due to evaporative concentration. Concentration increases
of 4 to 5 times that of inlet water are common for the blowdown from con-
ventional cooling towers, representing a 75 to 80% volume reduction. If
organics are present in the feed, some will be oxidized in the tower by both
biological activity and direct combination with oxygen.
In some applications, cooling tower concentration may not be feasible.
Potential problems include excessive biological growth and fouling, corrosion,
scaling, and loss of volatile species. Each of these can be corrected by
appropriate pretreatment, but at some point the additional cost of pretreat-
ment steps exceed the value of benefits derived from cooling tower concentra-
tion.
Forced Evaporation
Forced evaporation of a wastewater stream can be accomplished by one of
two principal systems. Of these, vapor compression evaporation has lower
energy costs and will be discussed for purposes of this manual.
In the vapor compression evaporation process, vaporization energy is
supplied by a mechanical compressor. The compressor raises the temperature
304
-------
Section 4
Aq. Med. Gen. Control
Volume Reduction
and pressure of the vapor from a tubeside evaporator, and steam is condensed
on the shell side of the evaporator, boiling more water. Seed crystals are
maintained in the evaporator to prevent scaling by the supersaturated liquor.
The evaporation overhead may be of sufficient quality to recycle directly
to the plant. The blowdown is typically 2 to 10% of the total wastewater
flow and contains the original contaminant species concentrated into this
small flow.
Vapor compression evaporators are sensitive to dissolved gases and vola-
tile organic compounds in the feed. Volatile components will appear in the
overhead and may require control. Such components may also affect recovered
water quality and limit its direct reuse within the facility.
305
-------
Section 4
Aq. Med. Gen. Control
Residual Disposal
4.2.1.8 Methods of Residual Disposal
Most commonly, treated wastewaters are discharged to a receiving body
for ultimate disposal. Constraints of such discharges are highly site-specific
and beyond the scope of this manual. There are several other methods that
can be used to dispose of a waste stream after it has been handled by appro-
priate upstream controls. Deep well injection, surface impoundment, and co-
disposal with ash are presented as examples.
Deep Hell Injection
Deep well injection has been used to dispose wastes that are difficult
to treat such as those containing high total dissolved solids or those con-
taining organics like tert-butyl alcohols (nonbiodegradable) and ketones (not
amenable to activated carbon adsorption). The disposal system consists of a
surface facility for pretreatment, a well, and a disposal zone. Many types of
formations can, under favorable circumstances, have sufficient porosity and
permeability to accept large quantities of injected liquid wastes. In prac-
tice, most wells have been constructed to inject into sand or sandstone and
limestone or dolomite. Detailed geological and engineering studies are re-
quired to determine the suitability of a potential site.
The injection depth is generally selected to provide adequate separation
from potable subsurface water. In every case, injection zones for disposing
of hazardous wastes must be below the deepest underground source of drinking
water. Wells are usually cased and cemented to prevent the upward migration
of fluids that are injected through tubing in a packer set immediately above
the injection zone. Injection pressures are set to ensure that neither frac-
ture nor fluid migration occur.
The major advantages of deepwell injection are: 1) it is an ultimate
disposal method, 2) it requires little land, and 3) it removes the waste from
contact with air, surface water, usable ground water, and the surface of the
ground. Major disadvantages include: 1) long-term effects are largely
306
-------
Section 4
Aq. Med. Gen. Control
Residual Disposal
unknown and difficult to predict, 2) control over the waste may possibly be
lost after injection, 3) favorable geologic conditions may not be available
in the vicinity of the waste source, 4) pretreatment adds to the expense of
waste disposal, and 5) wastes with high organic loading may cause plugging
of the well formation.
Deepwell injection has seen limited use in the disposal of industrial
wastes for over 25 years. However, a large number of wells inject oil field
brines. Performance data for a number of the industrial wells has shown
failures as well as successes. Problems have included migration of waste
to usable aquifers as a consequence of fracturing, faults in confining
strata, or defects in well casings.
Surface Impoundment
Surface impoundments are widely used to treat/dispose of industrial
wastes. These impoundments (also known as holding basins, lagoons, oxidation
ponds, settling basins, and evaporation ponds) can be either natural or man-
made reservoirs to which wastewaters, slurries, or sludges are discharged.
Retention time in the impoundment provides for natural evaporation, settling
of solids, biological decomposition of organics, and possible loss of volatile
components of the wastes. If properly designed and operated, minimal environ-
mental contamination should occur.
The suitability of a surface impoundment depends on site-specific factors
For example, in order to successfully serve as an evaporation pond, a surface
impoundment must be located at a site having a sufficiently high net evapora-
tion rate. A major drawback to their use is the need for relatively large
areas of land.
Co-disposal of Wastewater Streams with Ash
Co-disposal is an ultimate disposal method wherby the wastewater stream
is sorbed by ash, then the ash is handled by an appropriate solid waste dis-
posal method. The feasibility is determined by the quantity of ash generated,
307
-------
Section 4
Aq. Med. Gen. Control
Residual Disposal
its sorptive capacity, and the volume reduction needed to match the flow of
the wastewater stream to that which can be sorbed by the ash. All constitu-
ents in the wastewater stream become entrained in the ash, except those that
might volatilize in the course of wastewater/ash mixing.
308
-------
Section 4
Aq. Med. Strm. Control
Organic
4.2.2 Water Pollution Controls for Streams Containing
Predominantly Organic Constituents
Wastewater streams that contain predominantly organic constituents are
the condensates from the synthesis operation: Mobil M synthesis condensate
(Stream 233), F-T synthesis condensate (Stream 223), or methanol distillation
condensate (Stream 229). Control processes applicable to these streams are
discussed in this section. In all likelihood, none of these streams would be
treated individually, but as a consequence of economics, each would be combined
with others for common treatment. Therefore, the following discussion of in-
dividual stream controls is necessarily in the context of composite streams.
None of the water pollution control technologies discussed in this sec-
tion would be singularly applied to a particular composite waste stream. Vari-
ous combinations of controls would be used, and the wastewater characteris-
tics input to an individual control would not necessarily be those of the raw
composite flow but those modified by other, upstream processes. The empha-
sis in this section, however, is on individual controls, and as such, inputs
(and therefore performances) are estimated as those being most likely to
occur.
Neither the flow nor the loading o* wastewater streams from the synthesis
sections are specific to the type of coal fed to the gasifier. However,
waste stream quality and quantity originating from the gasification section
are specific to both coal type and process design, and these streams contri-
bute to the composite flow of which synthesis condensate is a part. In this
sense, coal type and gasification train design affect the choice and cost of
water pollution controls. A more detailed discussion of the effect of coal
type on wastewater streams is provided in Section 4.2.3.
309
-------
Section 4
Aq. Med. Strm. Control
Mobil M Syn. Condensate
4.2.2.1 Mobil-M Synthesis Condensate (Stream 233)
The waste stream from the Mobil M synthesis section (Stream 233) would
probably be combined with the compression condensate (Stream 211) and the
gas cooling/dust removal blowdown (Stream 210) from the gasifier section for
common treatment. As discussed in Section 3.3.4, base plant designs may uti-
lize either a methanol- or water-based wash to remove cyanide from the raw gas,
In cases where a water-based wash is used, an additional wastewater stream
would be generated which could also be combined with the three waste streams.
This section considers individual water pollution control technologies
that are applicable to the Mobil M synthesis condensate where it is part of
a composite flow described for base plants utilizing either a methanol- or
water-based cyanide wash. Each technology has been discussed in Section 4,2.1,
Only details specific to the subject composite flow are presented herein.
Activated Sludge
Activated sludge could be used to remove dissolved organics and other
contaminant species from the composite waste stream. For facilities utilizing
a water-based cyanide wash, organics expressed as BOD are estimated to be
3
1100 mg/L in a flow of about 688 m /hr. Activated sludge by one or more
stages is expected to remove BOD to about the 35 mg/L level with all reduced
sulfur species removed to 1 mg/L.
For facilities utilizing a methanol-based cyanide wash, the Mobil M con-
3
densate would be part of a smaller composite flow, about 449 m /hr. Without
the dilution of the cyanide wash water, the BOD concentration in the composite
flow would be higher at about 1700 mg/L (but same loading) and removed to
about the 40 mg/L level. All other species would be removed to approximately
the same extent as those cited for the water-based, cyanide wash case.
Much of the ammonia needed to supply nitrogen as a growth nutrient would
be present in the waste stream. However, supplemental ammonia as well as
phosphorus and possibly trace amounts of other materials would be required,
310
-------
Section 4
Aq. Med. Strm. Control
Mobil M Syn. Condensate
Due to the usual variability of waste stream characteristics, precise addi-
tion of ammonia would not be practical, and some overdosing would be required.
All ammonia added in excess of that required for cell growth would appear in
the activated sludge effluent.
Cyanide is normally considered toxic to microorganisms, but when it
occurs as part of a larger matrix of biodegradable organics, the resulting
microbial population can often acclimate to the stress of cyanide and oxidize
or otherwise remove some of it. Greater than 90% removal has been recorded
in carefully controlled, laboratory-scale studies. The exact removal that
would be realized is highly uncertain and can be determined only once the
system is in place.
A secondary waste stream would be generated consisting of biological
solids drawn from the underflow of the clarifier (Stream 415). This stream
would have a solids content of about 1 to 2% with a dry solids flow of about
12,700 kg/d for base plant designs using a water-based cyanide wash. For
the methanol-based case, the dry solids flow would be about 12,300 kg/d. Fol-
lowing concentration to about 20 to 40% dry solids, the solid waste stream
would be handled by one of the methods discussed in Section 4.3.
A secondary waste stream consisting of volatile species escaping the acti-
vated sludge reactor (fugitives) may also be generated (Stream 416). This
stream is considered further in Section 4.2.4.
Total capital investment for activated sludge is estimated to be $18.3
million with a total annualized cost of $3.9 million.
Filtration
Filtration could be used to remove suspended solids from the composite
wastewater stream. For base plant designs utilizing a water-based cyanide
wash, the flow of the composite stream is expected to be about 688 m /hr,
3
while a flow of about 449 m /hr is expected for the methanol-based case. For
311
-------
Section 4
Aq. Med. Strm. Control
Mobil M Syn. Condensate
both cases, the filter would be utilized following upstream treatment processes,
and an influent suspended solids concentration of about 30 mg/L is expected
with most of this material consisting of biological solids. Filtration should
remove the suspended solids to less than 10 mg/L.
A secondary waste stream of filter backwash (Stream 417) would be gener-
ated on an intermittent basis. This stream would range between 2 and 10% of
the influent flow and would contain most of the material that was removed dur-
ing the preceding filter run. The filter backwash is a small volume stream,
but its high loading prevents it from being routed directly to other control
processes for treatment. The backwash would therefore be routed to either its
own or some other flow equalization facility for gradual release to other
water control processes.
Total capital investment and total annualized cost for filtration are
estimated to be $1.2 and $2.5 million for the water-based cyanide wash case
and $0.78 and $1.8 million for the methanol-based cyanide wash case.
Granular Activated Carbon
Granular activated carbon could be used to remove residual organics fol-
lowing upstream treatment processes. For the water-based cyanide wash case,
the flow of the composite wastewater stream would be about 688 m /hr with a
COD of about 680 mg/L. COD removals to approximately 70 mg/L are expected.
For base plant designs utilizing a methanol-based cyanide wash, the flow of
3
the composite stream would be about 449 m /hr, and a COD of about 1060 mg/L
would be reduced to approximately 100 mg/L.
Regeneration of activated carbon by standard thermal methods would gen-
erate an offgas (Stream 420). Control of this stream would probably be a
function integral with the regeneration facility design.
Total capital investment and total annualized cost for the granular acti-
vated carbon process are estimated to be $3.1 and $1.5 million for the water-
312
-------
Section 4
Aq. Med. Strm. Control
Mobil M Syn. Condensate
based cyanide wash case and $3.0 and $1.6 million for the methanol-based
cyanide wash case.
Cooling Tower Concentration
Cooling tower concentration could be utilized to reduce the volume of the
composite wastewater stream to be handled for disposal. The feasibility of
this control would have to be evaluated on an individual case basis. Certain
species in the subject waste stream including biodegradable organics, chloride,
IDS, calcium, and sulfate may cause excessive fouling by biological growth,
corrosion, or scaling.
Loss of volatile species stripped from the wastewater stream is a sec-
ondary waste stream (Stream 419) and another factor that may limit the feasi-
bility of cooling tower concentration. In particular, ammonia and hydrogen
cyanide may be lost to the atmosphere, but the exact amount as well as the
amount that would be permissible in a given location is uncertain.
For base plant designs using a water-based cyanide wash, the composite
3 3
wastewater stream could supply 688 m /hr of the 920 m /hr makeup flow to the
cooling tower. All species in the original v/astewater stream would be concen-
3
trated by a factor of 3.8 into the 180 m /hr blowdown, assuming minimal losses
by volatilization or drift. The corresponding makeup contribution and concen-
tration factor for the methanol-based case would be 449 m /hr and 2.5 respec-
tively.
Characteristics of the composite wastewater stream used as partial make-
up to the cooling tower are those that have been modified by upstream treat-'
ment processes. These feed and the resulting blowdown characteristics are
summarized in Table 4-27. Since the cooling tower is considered part of the
base plant, no capital costs associated with environmental controls are assumed
to be required.
313
-------
TABLE 4-27. MATERIAL FLOW FOR COOLING TOWER
CONCENTRATION - MOBIL M SYNTHESIS CASE
Base Plant
Cyanide Wash:
Characteristic*
Flow (m3/hr)
NH3
CN~
S=
S20=3
S0=3
SCN"
COD
BOD5
Cl"
TDS
TSS
Water- based
Makeup*
688
10
9-35
<1
<1
<1
<1
680
35
1040
1900
30
Slowdown"1"
180
38
34-130
<1
4
4
4
2600
130
3900
7200
110
Methanol -based
Makeup*
449
10
1
<1
<1
<1
<1
1060
42
1600
2900
30
Slowdown*
180
25
2
<1
2
2
2
2650
105
4000
7200
75
*A11 concentrations are mg/L and reflect published data and engineering
estimates. Detailed performance data and references are contained in
the Control Technology Appendices.
+This stream combined with another source to supply the total makeup
requirement of 920 nr/hr.
314
-------
Section 4
Aq. Med. Strm. Control
Mobil M Syn. Condensate
Forced Evaporation
Forced evaporation could be used to decrease the volume of the composite
wastewater flow by concentrating the contaminant species into a fraction of
the original volume. Condensate from forced evaporation may contain com-
pounds volatilized during water evaporation. Volatilized components may in-
fluence condensate reuse alternatives, depending upon which compounds are
volatilized and their concentrations in the condensate. For material balance
purposes, volatilization was neglected since no data are available to support
assumptions relating to species losses from the feed. The feasibility of this
control process would have to be evaluated on an individual case basis.
Forced evaporation would be preceeded by other control processes for dis-
solved gas or volatile organics removal. Estimated characteristics of the
composite wastewater stream input to forced evaporation and its blowdown are
presented in Table 4-28. The estimated blowdown quality represents the worst
case for the specified feed and volume reduction since species volatilization
has been neglected. Total capital investment is estimated to be $13.7 million
with a total annualized cost of $3.5 million.
Incineration
Incineration could be used to remove residual organics following various
pretreatment and concentration steps. For the subject composite wastewater
stream, the flow to be incinerated would be first decreased by upstream con-
o
trols to about 18 m /hr for both the methanol- and water-based cyanide wash
cases. Incineration would reduce COD and BOD in the composite flow from
approximately 26,000 and 1000 mg/L to 25 and 1 mg/L, respectively.
Incineration would generate both a liquid and a gaseous secondary waste
stream. The gaseous stream is flue gas (Stream 421) from combustion of
auxiliary fuel and waste stream constituents. This stream would be controlled
by equipment that is an integral part of the incinerator facility. The liquid
stream is the flue gas quenching/scrubbing system blowdown. This stream would
315
-------
Section 4
Aq. Med. Strm. Control
Mobil M Syn. Condensate
have a low flow and contain mainly inorganic salts with some particulate mat-
erial . It would be routed to other processes for treatment or combined
directly with the effluent stream.
Total capital investment for incineration is estimated to be $4.8 mil-
lion with a total annualized cost of $2.3 million.
316
-------
TABLE 4-28. MATERIAL FLOW FOR FORCED EVAPORATION-
MOBIL M SYNTHESIS CASE
Base Plant
Cyanide Wash:
Characteristic*
Flow (m3/hr)
NH3
CN~
S=
S2°3
S03
SCN"
COD
BOD5
cr
TDS
TSS
Water-based
Input
180
38
34-130
<1
4
4
4
2600
130
3900
7200
110
Slowdown*
8
380
340-1300
<1
40
40
40
26000
1300
39000
72000
1100
Methanol-based
Input
180
25
2
<1
2
2
2
2650
105
4000
7200
75
Slowdown
18
250
20
<1
20
20
20
26500
1050
40000
72000
750
*A11 concentrations are mg/L and reflect published data and engineering
estimates. Detailed performance data and references are contained in
the Control Technology Appendices.
"^Slowdown quality represents the worst case for the specified feed and
volume reduction since species volatilization has been neglected.
317
-------
Section 4
Aq. Med. Strm. Control
F-T Syn. Condensate
4.2.2.2 F-T Synthesis Condensate (Stream 223)
The characteristics of the F-T synthesis Condensate (Stream 223) are
somewhat different than those of the Mobil M condensate (refer to Section
3.4.6), but a similar treatment approach would be involved. Applicable
control processes are discussed in Section 4.2.2.1 with costs and per-
formances varying only slightly from those presented as a primary result of
flow rate differences. For example, an activated sludge system designed to
treat the F-T synthesis condensate is estimated to require a total capital
investment of $20.9 million, an increase of about 14% over the equivalent
Mobil M activated sludge system.
318
-------
Section 4
Aq. Med. Strm. Control
Methanol Syn. Condensate
4.2.2.3 Methanol Distillation Condensate (Stream 229)
The methanol distillation condensate (Stream 229) combined with gasifi-
cation wastewaters contains sufficient loading of organics to be handled
analogously to the Mobil M or F-T synthesis cases. However, the nitrogen
(as ammonia) in the inorganic streams would greatly exceed the amount needed
for biological oxidation of the organics. The methanol condensate could also
be handled by slight modification to control technologies discussed in Section
4.2.3.1 for base plant cases where cyanide is removed from the raw gas by a
water wash, and in Section 4.2.3.2 for the methanol wash case.
319
-------
Section 4
Aq. Med. Strm. Control
Inorganic
4.2.3 Water Pollution Controls for Streams Containing
Predominantly Inorganic Constituents
Wastewater streams that contain predominantly inorganic constituents
are the cyanide wash water (Stream 215), primary compression and cooling
condensate (Stream 211), and the gas cooling and dust removal blowdown
(Stream 210). It is not likely that any of these streams would be treated
separately, but each would be combined with the others, for reasons of econo-
mics, into a composite flow. Paralleling the approach taken in Section
4.2.2, each stream is discussed in this section in the context of a composite
flow.
320
-------
Section 4
Aq. Med. Strm. Control
Cyanide Wash Water
4.2.3.1 Cyanide Wash Water (Stream 215)
A number of control processes are applicable to the cyanide wash water
(Stream 215). Of these, only polysulfide addition would be applied exclusi-
vely to this stream. All other processes would be applied to a composite
wastewater stream consistinq of the combined flows of the cyanide wash water
(Stream 215), the compression condensate (Stream 211), and the gas cooling/
dust removal blowdown (Stream 210). Each of the control technologies con-
sidered in this section has been discussed in Section 4.2.1, and only details
specific to the subject composite flow are considered in the following sub-
sections.
Polysulfide Addition
Polysulfide could be added to the cyanide wash water to convert cyanide
to thiocyanate. The cyanide wash is estimated to have a flow of 239 m /hr
with a cyanide concentration of 240 mg/L. Less than about 10 mg/L cyanide
is expected to remain following polysulfide addition to a flash mix reactor
and subsequent reaction in a plug flow reactor. This performance is estimated
assuming that little of the cyanide is present in a complexed form, an expec-
tation based on estimated very small concentrations of potentially complex-
forming cations in the subject waste stream.
Preliminary EPA sponsored lab studies (unpublished) indicate that the
conversion of cyanide to thiocyanate can be accomplished at the wastewater
treatment process scale. However, excess polysulfide of approximately double
its stoichiometric requirement may be needed to achieve reasonably rapid
reaction rates. Otherwise, residence times and therefore reactor volumes
may assume unreasonable dimensions.
In some applications, it may be possible to dose polysulfide directly
to the cyanide wash circuit (or alternatively to the gas cooling/dust removal
circuit). If feasible, substantial cost savings would be realized. Most
capital requirements would be eliminated as a small modification to the cyanide
321
-------
Section 4
Aq. Med. Strm. Control
Cyanide Wash Water
wash system would replace the large treatment wastewater reactors. More
importantly, it is expected that the polysulfide demand would be reduced.
Nearly complete conversions of cyanide are realized in refinery applications
where approximately stoichiometric doses of polysulfide are made at the out-
let of process equipment. It is therefore expected that by dosing the poly-
sulfide directly to the cyanide wash circuit or to the gas cooling/dust re-
moval circuit, nearly complete cyanide conversions could likewise be realized
with a minimum of polysulfide added.
Another method of effecting cyanide conversion by polysu'ifide might be
to add the polysulfide in conjunction with another wastewater treatment pro-
cess. A likely prospect is the activated sludge process, but this configura-
tion has never been tried.
The cyanide to thiocyanate conversion has generally been identified as
requiring alkaline reaction conditions. Following polysulfide addition, the
subject waste stream would be combined directly with the compression conden-
sate and the gas cooling/dust removal blowdown, or if sufficent acidity is not
available in these streams, the cyanide wash would be first neutralized with
acid. In either case, the pH of the cyanide wash will be depressed to approx-
imately neutral and unreacted polysulfide may precipitate. In some cases,
this precipitate may be compatible with downstream controls. If not, a filter
or clarifier would be added.
Total capital investment for polysulfide addition is estimated to be
$130,000 with a total annualized cost of about $730,000.
The type of coal fed to the gasifier is expected to have little effect
on the mass flow of cyanide produced in the gasifier section. The flow and
cyanide concentration of the cyanide wash will therefore by generally inde-
pendent of coal type, and the cost of polysulfide addition will remain con-
stant as well.
322
-------
Section 4
Aq. Med. Strm. Control
Cyanide Wash Water
Activated Sludge
Activated sludge would be utilized primarily to oxidize ammonia and thio-
cyanate, although other species would be removed as well. The subject com-
posite wastewater stream would contain about 260 mg/L of ammonia and 210 mg/L
3
of thiocyanate, at 578 m /hr. Following activated sludge, about 8 mg/L am-
monia and less than 1 mg/L thiocyanate are expected to remain.
A secondary waste stream of biological solids (Stream 415) at about 1%
concentration would be drawn from the underflow of the clarifier at the rate
of about 1300 kg/d dry solids. Following concentration to 20 to 40% solids
by a filter press or other solids handling equipment, this waste stream would
be handled by appropriate solid waste disposal techniques discussed in
Section 4.3.
Total capital investment is estimated to be $12.6 million with a total
annualized cost of about $2.7 million.
Denitri'fication
Denitrification could be utilized to remove nitrate. The process would
be proceeded by other controls such that the composite wastewater stream
3
would contain a nitrate concentration of about 245 mg/L at 578 m /hr. Deni-
trification is expected to remove nitrate to about the 25 mg/L level.
A secondary waste stream of biological solids (Stream 418) at about 1%
concentration would be drawn from the underflow of the clarifier at the rate
of about 1800 kg/d dry solids.
Total capital investment is estimated to be $6.1 million with a total
annualized cost of about $2.2 million.
Filtration
Details of filtration are identical to those presented in Section 4.2.2.1.
The composite wastewater stream has a flow of about 578 m /hr. The total
323
-------
Section 4
Aq. Med. Strm. Control
Cyanide Wash Water
capital investment was estimated to be $1.0 million with a total annualized
cost of about $220,000.
Chemical Oxidation
Chemical oxidation could be used to remove low level concentrations of
cyanide that remain in the composite wastewater stream following treatment
by other controls. However, since cyanide is a slowly oxidized species, most
other reduced chemical species would be removed concurrently, increasing the
oxidant demand. The composite wastewater stream with a flow rate of 578
3
m /hr would probably contain less than 10 mg/L of both cyanide and ammonia.
Both would be removed to less than 1 mg/L.
Total capital investment is estimated to be $720,000 with a total annu-
al ized cost of $230,000.
Ion Exchange for Ammonia Removal
Ion exchange using a clinoptilolite resin could be utilized to remove
3
ammonia. When applied to the 578 m /hr composite wastewater stream contain-
ing 260 mg/L ammonia, ion exchange is estimated to be capable of reducing
the ammonia concentration to less than 10 mg/L.
Total capital investment is estimated to be $12.9 million with a total
annualized cost of about $3.5 million.
Cooling Tower Concentration
Cooling tower concentration could be utilized to reduce the volume of the
composite wastewater stream to be handled by downstream processes and disposal
methods. Since the subject waste stream is essentially free of organics, foul-
ing by biological growth is not expected to be a problem. However, other
characteristics of the waste stream may be limiting, and the applicability of
cooling tower concentration would therefore have to be determined on an indi-
vidual case basis.
324
-------
Section 4
Aq. Med. Strm. Control
Cyanide Wash Water
The makeup required for the cooling tower (methanol synthesis case) is
3 3
825 m /hr. The composite waste stream at 578 m /hr could supply most of this
with the balance supplied by other sources. With blowdown at 161 m /hr,
chemical species in the original wastewater stream would be concentrated by
a factor of 3.6, assuming minimal loss of species by volatilization and
drift. The characteristics of the composite waste stream when used as cool-
ing tower makeup, and therefore of the blowdown, will vary depending on the
upstream control processes applied. The makeup and blowdown characteristics
summarized in Table 4-29 are estimated as worst case examples. Since the
cooling tower is part of the base plant, no costs associated with environ-
mental controls have been estimated for it.
Forced Evaporation
Forced evaporation could be used to decrease the volume of the composite
wastewater stream. Forced evaporation would also permit recovery of a high
quality steam suitable for recycle to the base plant. The feasibility of
this control process would have to be evaluated on an individual case basis.
As with cooling tower concentration, the characteristics of the input to
forced evaporation and its blowdown would vary with upstream controls. Esti-
mated worst case characteristics are presented in Table 4-29.
Total capital investment is estimated to be $13.7 million with a total
annualized cost of about $3.5 million.
325
-------
TABLE 4-29. MATERIALS FLOW FOR COOLING TOWER CONCENTRATION AND FORCED
EVAPORATION - WATER-BASED CYANIDE WASH CASE
Cooling Tower
Concentration
Characteristic*
Flow (m3/hr)
NH3
CN"
s=
S2°3
S03
SCN"
cr
TDS
NO^-N
TSS
Make-up1"
578
260
7.5
2
58
14
210
1200
2200
245
30
Slowdown
161
933
27
<1*
205
50
740
4500
7900
880
110
Forced
Evaporation
Feed
161
933
27
<1
205
50
740
4500
7900
880
110
Bl owdown
16
9330
270
<1
2050
500
7400
45000
79000
8800
1100
*A11 concentrations are mg/L and reflect published data and engineering
estimates. Detailed performance data and references are contained in
the Control Technology Appendices.
"'"This stream combined with another source to supply the total makeup
requirement of 825 m^/hr.
S= is lost through both oxidation and stripping.
326
-------
Section 4
Aq. Med. Strm. Control
Compression Condensate
4.2.3.2 Raw Gas Compression and Cooling Condensate (Stream 211)
The compression condensate (Stream 211) could be combined with the cya-
nide wash water (Stream 215) and the gas cooling/dust removal blowdown
(Stream 210) for common treatment. In this context, control technologies
applicable to the compression condensate are identical to those presented
for the cyanide wash water, Section 4.2.3.1 with exception of polysulfide
addition. The characteristics of all waste streams and the performances and
costs of all controls are the same.
As discussed in Section 3.3.4, some base plant designs may use a methanol-
rather than water-based wash to remove cyanide from the raw gas. Where this
option is exercised, the compression condensate would probably be combined
only with the gas cooling/dust removal blowdown for common treatment since no
cyanide wash water would be generated. The control technologies discussed in
Section 4.2.3.1 would still be applicable to this composite flow, but with
different waste stream characteristics, certain performances and costs would
be different. This section considers only aspects of the control technologies
applied to the subject composite stream (i.e., compression condensate and
gas cooling/dust removal blowdown) that differ from those already presented
in Section 4.2.3.1.
Activated Sludge
The composite wastewater stream is estimated to contain about 190 and
4.5 mg/L of ammonia and thiocyanate, respectively, and has a flow rate of
339 m /hr. Ammonia would be removed to about 6 mg/L and thiocyanate removed
to less than 1 mg/L. A biological solids waste stream (Stream 415) would be
generated at a rate of 270 kg/d dry solids in a 1 to 2% solids slurry. This
solid waste stream would be concentrated to 20 to 40% solids by a filter
press or some other control, then disposed of by a method described in Sec-
tion 4.3.
327
-------
Section 4
Aq. Med. Strm. Control
Compression Condensate
Total capital investment is estimated to be $6.7 million with a total
annualized cost of about $1.5 million.
The composition of the compression condensate and therefore the charac-
teristics and costs of applicable controls are affected by the type of coal
fed to the gasifier. As discussed in Section 3.2.1, ammonia formation during
gasification tends to increase with decreasing coal rank. Therefore the
quantity of ammonia in the compression condensate and/or the gas cooling/dust
removal blowdown, which also contributes to the composite flow, would like-
wise increase for lower rank coals, and there would be a corresponding increase
in the total capital investment and total annualized cost of water pollution
controls.
Coal chloride content has little effect on the characteristics of the
compression condensate since halogens are almost completely removed during
gas cooling and dust removal. However the characteristics of the gas cooling/
dust removal blowdown and therefore the characteristics of the subject com-
posite stream are dependent on coal type. These aspects are considered in
Section 4.2.3.3.
Biological Denitrification
Denitrification would be applied to the composite waste stream following
3
upstream treatment processes. From a flow of 339 m /hr, the concentration of
nitrate could be decreased from about 140 mg/L to 15 mg/L. Total capital
investment is estimated to be $3.2 million with a total annualized cost of
$980,000.
A biological solids waste stream (Stream 418) would be generated at a
rate of about 590 kg/d dry solids in a 1 to 2% solids slurry. Following
dewatering to 20 to 40% solids, the waste stream would be disposed of by a
solid waste disposal method described in Section 4.3.
328
-------
Section 4
Aq. Med. Strm. Control
Compression Condensate
Filtration
Total capital investment for filtration of the composite stream of 339
m3/hr is estimated to be $590,000 with a total annualized cost of about $150,000.
All other aspects of filtration are identical to those presented in Section
4.2.2.1.
Chemical Oxidation
When applied to the composite wastewater stream following upstream treat-
ment processes, the chemical oxidation feed would contain approximately 6 and
o
7 mg/L of ammonia and cyanide, respectively, and has a flow rate of 339 m /hr.
Each of these species would be reduced to less than 1 mg/L. Total capital
investment is estimated to be $160,000 with a total annualized cost of about
$350,000.
Ion Exchange
When applied to the composite wastewater stream, the clinoptilolite-based
ion exchange system would be loaded with about 190 mg/L of ammonia at 339
m /hr. The effluent from this process is expected to contain less than 10
mg/L ammonia. Total capital investment is estimated to be $8.3 million with
a total annualized cost of about $1.8 million.
Cooling Tower Concentration
3 3
The composite wastewater stream could supply 339 m /hr of the 825 m /hr
3
required for cooling tower makeup. With blowdown at 161 m /hr, constituents
in the composite stream would be concentrated by a factor of 2.1. As described
in Section 4.2.3.1, the characteristics of the subject composite flow used for
makeup water and therefore characteristics of the blowdown would vary accord-
ing to upstream controls. The makeup and blowdown characteristics summarized
in Table 4-30 are estimated worst case examples. No capital cost associated
with environmental controls is assumed for cooling tower concentration since
the cooling tower is part of the base plant.
329
-------
Section 4
Aq. Med. Strm. Control
Compression Condensate
Forced Evaporation
As with cooling tower concentration, the characteristics of the input to
forced evaporation and its blowdown would depend on upstream controls. Esti-
mated worst case characteristics are presented in Table 4-30. Total capital
investment is estimated to be $13.7 million with a total annualized cost of
about $3.5 million.
330
-------
TABLE 4-30. MATERIAL FLOW FOR COOLING TOWER CONCENTRATION
AND FORCED EVAPORATION - METHANOL-BASED
CYANIDE WASH CASE
Characteristic*
Flow (m3/hr)
NH3
CN"
S=
S2°3
so-
SCN
cr
TDS
NO^-N
TSS
Cooling Tower
Concentration
Make-upt
339
190
7
3.5
98
24
4.5
2100
3800
140
30
Slowdown
161
400
15
<1*
205
50
9.5
4400
8000
300
60
Forced
Evaporation
Feed
161
400
15
<1
205
50
9.5
4400
8000
300
60
Bl owdown
16
4000
150
<1
2050
500
95
44000
80000
3000
600
*A11 concentrations are mg/L and reflect published data and engineering
estimates. Detailed performance data and references are contained in
the Control Technology Appendices.
''This stream combined with another source to supply the total makeup require-
ment of 825
=I=S= is lost by both oxidation and volatilization.
331
-------
Section 4
Aq. Med. Strm. Control
Gas Cool/Dust Removal
4.2.3.3 Gas Cooling and Dust Removal Slowdown (Stream 210)
The gas cooling and dust removal blowdown (Stream 210) could be combined
with the compression condensate (Stream 211) for common treatment. Applicable
control technologies and corresponding characteristics are discussed in
Section 4.2.3.2. For base plant designs where cyanide is removed from the
raw gas by a water-based wash, the cyanide wash water (Stream 215) would also
be combined with the two streams cited. Applicable control technologies and
corresponding characteristics are discussed in Section 4.2.3.1.
The characteristics of the coal fed to the gasifier affect the charac-
teristics of the gas cooling/dust removal blowdown in one of two general ways.
First, the mass of ammonia produced in the gasifier tends to increase with
decreasing coal rank. Also, higher sulfur levels generally result from coals
having higher sulfur contents, although certain secondary effects such as ash
alkalinity may have more of an affect on the sulfur contained in wastewater
streams. The level of cyanide in wastewater streams is affected only slightly
by characteristics of the coal.
Second, coal type can affect the quantity of water blown down from the
gas cooling/dust removal circuit. A minimum blowdown is required represent-
ing the difference between the water entering the circuit with the raw gas
and that leaving the system with the cooled gas and dust (after dewatering).
Most of the water contained in the raw gas is the result of the gas being
quenched at the gasifier outlet to reduce its temperature to below the ash
fusion temperature so that fouling of the waste heat boiler does not occur.
The amount of water added depends on the gasification temperature (a function
of coal rank and ash properties) and to a lesser extent on the mass of ash to
be quenched.
Blowdown requirements are also influenced by the chloride content of the
coal. Chloride in the coal is almost entirely gasified and is recovered in
the gas cooling and dust removal operation. If the concentration of chloride
332
-------
Section 4
Aq. Med. Strm. Control
Gas Cool/Dust Removal
in the gas cooling/dust removal circuit exceeds the level considered allow-
able for operating reasons, makeup water must be added to dilute the chloride
concentration, and an increase in blowdown results.
The following two examples demonstrate the effect of coal chloride.
For the base case utilizing the Illinois No. 6 coal described in Section 3 and
for no makeup water added to the gas cooling/dust removal circuit, the chlo-
ride would establish itself in the circuit at about 6000 mg/L. For purposes
of this manual, it was assumed that operating requirements of the gas cooling/
dust removal circuit impose a chloride concentration limitation of about
2100 mg/L. Makeup water was, therefore, added to dilute the chloride result-
ing in an increase in blowdown by a factor of about 2.8. If an actual design
assumes a significantly different concentration of chloride in the gas cooling/
dust removal circuit, the blowdown will change accordingly. The concentration
of all species for any composite wastewater stream of which the blowdown is
a part would be affected, but the mass loading of all species would remain
essentially unchanged. Some modification of individual pollution controls
might be needed, but the overall approach to wastewater control and the costs
would not be greatly affected.
The blowdown flow rate may also be dependent upon the chloride content
of the feed coal. For a system gasifying essentially the same Illinois No. 6
coal but which has one quarter the chloride content (0.07% dry basis) indi-
cated for the design coal, no makeup would be required to the circuit (assum-
ing a maximum permissible chloride level of 2100 mg/L), and the blowdown
would decrease by a factor of 2.8.
333
-------
Section 4
Aq. Med. Strm. Control
Secondary Waste Streams
4-2.3.4 Secondary Waste Streams from Other Media
Secondary waste streams from other media are limited to condensates and
purge streams from tail gas treatment processes and purge streams from flue
gas desulfirization processes. Estimated flows from these controls are:
3 "?
4 m /hr from Beavon (Stream 405) and SCOT (Stream 409); 3 m /hr from Wellman-
Lord (Stream 411); and 37.5 m /hr from flue gas desulfurization tStream 425)
in the F-T synthesis case and 6.1 m /hr in the Mobil M case. Additionally,
3
a waste stream of about 2 to 3 m /hr would be generated by the Beavon process
consisting of Stretford solution purge (Stream 405). Each of these waste-
water streams would either be combined with other streams for common treatment
or be treated separately by controls that are well developed in petroleum
refining applications.
334
-------
Section 4
Aq. Med. Int. Ex.
4.2.4 Integrated Pollution Control Examples
Previous sections have considered individual water pollution controls
applicable to the principal wastewater streams generated by a K-T gasification
facility. Section 4.2.2 considered streams having predominantly organic con-
stituents and Section 4.2.3 considered streams having predominantly inorganic
constituents. Details of individual controls or combinations of controls
have not yet been proposed for K-T-based gasification facilities in the U.S.
Except where otherwise noted, individual controls have been selected on the
basis of controls commonly utilized in parallel industries. This section
presents examples of how individual controls might be combined to treat waste-
water streams prior to their ultimate disposal. The approaches considered in
this section are not all inclusive but only serve to indicate how available
controls may be applied to the subject wastes.
The controls selected in each example are largely determined by the ulti-
mate disposal method being considered and the type of synthesis section uti-
lized by the base plant. The specific waste stream quality required by each
disposal method is uncertain and would be determined on a site-specific basis,
but certain general requirements are associated with each, and these are
incorporated in the integrated control examples.
Three ultimate disposal methods are considered: discharge to surface
waters, surface impoundment, and deep well injection. In some cases, co-
disposal of the wastewater with ash from the gasifier may also be a viable
method (see Section 4.2.1.8). Potential sources of ash are incinerated dust
from the gas cooling/dust removal blowdown, boiler bottom ash, and boiler fly
ash. This ash is expected to sorb 20% water. Additionally, slag generated
by the gasification section can sorb about 10% water. The total estimated
capacity of wastewater that can be sorbed by ash is 6.7 m /hr. However, this
capacity may vary greatly, depending primarily on the ash content of the feed
coal, the slag/dust and ash partitioning during gasification, and the feasi-
bility of resource recovery by incineration of gasifier dust.
335
-------
Section 4
Aq. Med. Int. Ex.
It is not expected that the ash and slag generated by the gasifier would
provide sufficient capacity to dispose of the entire wastewater volume. Co-
disposal is therefore not considered further in this manual. However, while
only single ultimate disposal methods are considered for each integrated con-
trol example in this section, several methods may be combined to satisfy the
needed capacity. In this context co-disposal may be a viable option.
The type of synthesis operation utilized in the base plant affects the
wastewater treatment approach required for the entire K-T plant. Both the
F-T and Mobil M synthesis sections generate a waste stream that is heavily
loaded with organics. Fuel grade methanol synthesis generates a much smaller
volume of organic-loaded waste to the total wastewater stream, while crude
methanol synthesis generates no waste stream from the synthesis operation.
All other significant waste streams in the K-T facility contain principally
inorganic constituents. Therefore, K-T plants that produce crude methanol
generate, facility-wide, only waste streams having primarily inorganic con-
stituents. Fuel grade methanol K-T plants generate only a small load of
organics compared to the inorganics load of a combined wastewater, and F-T
and Mobil M synthesis plants generate both heavily loaded organic and in-
organic waste streams when gasification and synthesis wastes are combined.
Somewhat different water pollution control approaches appear most advantageous
for each of these three cases. Each case is considered separately in the
following section.
336
-------
Section 4
Aq. Med. Int. Ex.
4.2.4.1 Treatment of Waste Streams from Base Plants Utilizing F-T or
Mobil M Synthesis
Both F-T and Mobil M synthesis operations generate a condensate having
high concentrations of organics. These organics are relatively simple species
including short chain hydrocarbons, ketones, and organic acids. It is expected
that the F-T and Mobil M synthesis condensates will be highly amenable to bio-
logical oxidation: BOD was estimated to be 70% of COD. However, there are
two problems in applying a conventional biological system to these waste
streams. First, the concentration of organics exceeds the maximum normally
handled by a suspended growth biological system such as activated sludge.
Dilution of the waste stream by recycling the activated sludge effluent
would require excessive pumping and increased capital costs. Other biologi-
cal process reactors, such as fixed film reactors (trickling filters and
others), could handle the high organic concentrations, but it is expected that
following these processes, the organics concentration would remain relatively
high so that an activated sludge process to effect more complete removal might
still be advisable. Additionally, this approach does not address the second
problem which is lack of growth nutrients. The subject waste stream does not
contain nitrogen, phosphorus, sulfur, and trace amounts of other materials
required for cell growth.
The waste streams generated by the gasifier section (Section 4.2.4) con-
tain, among other species, ammonia and sulfide but no organics. By combining
these waste streams with a synthesis condensate, the ammonia would supply
most of the nitrogen and the sulfide would supply all of the sulfur required
to support cell growth. The concentration of organics in the combined stream
would be diluted to a level suitable for an activated sludge system. Char-
acteristics of the individual waste streams from the gasifier section, the
Mobil M condensate, and the combined streams are presented in Table 4-31.
There are two cases of combined streams representing the two base plant
designs for cyanide wash discussed in Section 3.3.4. Where cyanide is removed
337
-------
TABLE 4-31. CHARACTERIZATION OF MAJOR STREAMS COMBINED FOR COMMON TREATMENT - MOBIL M SYNTHESIS CASE
co
to
Characteristic*
Flow (m3/hr)
NH3
CN"
S=
S2°3
S0=
SCN-
COD
BOD5
cr
TDS
Gas Cooling/
Dust Removal
Blowdown
Stream 210
322
156
7
1
103
25
4
--
--
2111
4000
Compression
Condensate
Stream 211
17.4
735
8.9
48.7
6.3
--
14
--
--
2200
— ™*
Cyanide
Wash
Water
Stream 215
239
--
241
176
--
--
--
—
--
--
~ ™
Compos
Mobil M Base Plant
Condensate Water-
Stream 233 basedt
110 688
92
87
63
48
12
2.5
14000 2200
6860 1100
1040
1900
ite
Cyanide Wash
Methanol-
based*
449
140
5
2.6
74
18
3
3400
1700
1600
2900
*A11 concentrations are mg/L.
tCombined streams: gas cooling/dust removal blowdown, compression condensate,
cyanide wash water, and Mobil M condensate.
^Combined streams: gas cooling/dust removal blowdown, compression condensate,
and Mobil M condensate.
-------
Section 4
Aq. Med. Int. Ex.
from the base plant by a water wash, a wastewater stream is generated con-
tributing to a combined flow of 688 m /hr. Where cyanide is removed by a
methanol-based wash essentially no wastewater is generated by this source,
and the combined flow is 449 m /hr.
Presented in Figure 4-6 are three examples of integrated controls appli-
cable to a composite waste stream consisting of the Mobil M condensate
(Stream 233) combined with the wastewater streams from the gasifier section.
Identical examples could be used for the F-T synthesis case with changes
required only to reflect the different flow and organic loading of the F-T
synthesis condensate. Neither performance nor cost characteristics would
differ greatly.
A feature common to all integrated pollution control examples is flow
equalization. Design performance of pollution control processes can be real-
ized only when the waste stream characteristics vary gradually and within a
generally narrow range. For some processes, particularly those based on the
activity of microorganisms, equalization is especially critical. Under con-
ditions of no more than slightly varying waste stream characteristics, micro-
bial populations can usually be acclimated to both concentrations and types
of chemical species that would otherwise not be degradable. If significant
fluctuations in the input occur, there is no chance to acclimate the micro-
bial culture, or worse, the entire system might fail requiring a complete,
new start-up.
Certain waste streams in the K-T plant other than those from the gasi-
fier or synthesis sections may also be routed to the equalization facility.
These waste streams would contribute only minor flows and loadings to down-
stream controls. The exact requirements of the equalization facility are
beyond the scope of this manual. Its design would take into consideration the
variability of all waste stream inputs to be handled by the integrated controls,
339
-------
EXAMPLE 1
SYNTHESIS CONDENSATE
STREAM 233 OR 223 *
COMPRESSION CONDENSATE
STREAM 211 *
GAS COOLING/DUST REMOVAL BLOWDOWN
STREAM 210 *
CYANIDE WASH WATER*
STREAM 215 ^
fe
FLOW
EQUALIZATION
ACTIVATED
SLUDGE
k
ML 1 HA 1 IUIM
GRANULAR
ACTIVATED
CARBON
DISCHARGE TO
SURFACE WATERS
oo
-e»
o
EXAMPLE 2
SYNTHESIS CONDENSATE
STREAM 233 OR 223 P
COMPRESSION CONDENSATE
STREAM 211 *
GAS COOLING/DUST REMOVAL BLOWDOWN
STREAM 210 *
FLOW
EQUALIZATION
ACTIVATED
SLUDGE
COOLING
TOWER
CONCENTRATION
CYAWDE_W_ASH_WAT ER*
STREAM 215
DISCHARGE TO
SURFACE IMPOUNDMENT
EXAMPLE 3
SYNTHESIS CONDENSATE
STREAM 233 OR 223 w
COMPRESSION CONDENSATE
STREAM 211 *
GAS COOLING/DUST REMOVAL BLOWDOWN
STREAM 210
k
FLOW
FDI 1 A 1 1
ZATION
5^
ACTIVATED
SLUDGE
^
COOLING
TOWER
CONCEN-
TRATION
^
FORCED
ATION
INCINER-
ATION
CYANjp.E WASH WATER'
STREAIVTT15
DISCHARGE TO
DEEP WELL
INJECTION
*This stream exists only for base plants where cyanide is removed from the raw gas by a water-based wash.
Figure 4-6. Integrated control examples - Mobil M or F-T synthesis case
-------
Section 4
Aq. Med. Int. Ex.
Example 1. Integrated pollution controls in Example 1 of Figure 4-6
are an example of how the waste stream might be treated prior to ultimate
disposal fay direct discharge to surface waters. The controls are activated
sludge, filtration, and granular activated carbon. The characteristics of
the effluent from each control process are presented in Table 4-32 and the
estimated characteristics of the wastewater stream discharged to surface
waters by this illustrative example as well as by the other examples are
presented subsequently in Table 4-35.
The activated sludge process would be designed primarily to remove
organics. Since relatively simple organic species are involved, better than
95% removal is expected through one or more stages. Thiocyanate, thiosulfate,
and sulfite would be concurrently oxidized.
Ammonia would be removed in the activated sludge process by being assim-
ilated as new cell material. Because the nitrogen available in the composite
flow is somewhat less than that required, extra nitrogen as ammonia would be
added. All ammonia in excess of that required for cell growth would appear
in the effluent. Therefore the concentration of ammonia in the effluent
would depend on the degree of control exercised over its addition. While
addition of ammonia representing its exact requirement is ideally possible,
in practice the continuously varying organics concentration in the influent
would necessitate excess ammonia addition.
Air or oxygen to support biological oxidation is supplied to the reactor
under highly turbulent conditions. This maximizes the mass of oxygen trans-
ferred to the water, keeps the contaminants and microbial floes dispersed
throughout the reactor, but provides conditions that are highly favorable to
stripping of volatile species. At the slightly alkaline reactor conditions
needed for optimal microorganism growth, much of the sulfide would occur in
its potentially volatile form, hydrogen sulfide (H?S). While some of the sul-
fide would be oxidized by direct combination with oxygen and some biologically
341
-------
TABLE 4-32. EXAMPLE 1 - MATERIAL FLOW FOR MOBIL M SYNTHESIS BASE PLANT INTEGRATED CONTROLS
(FIGURE 4-6)
co
Base Plant
Cyanide Wash:
Characteristic
Flow (m3/hr)
NH3
CN"
S=
S2°3
S03
SCN"
COD
BOD5
ci-
TDS
TSS
Composite Activated Sludge Filtration
Water- Methanol- Water- Methanol- Water- Methanol -
based based based based based based
*
688
92
87
63
48
12
2.5
2200
1100
1040
1900
--
449 688 449 688 449
140 10 10 10 10
5 9-35 1 9-35 1
2.6 <1 <1 <1 <1
74 <1 <1 <1 <1
18 <1 <1 <1 <1
3 <1 <1 <1 <1
3400 680 1060 680 1060
1700 35 42 35 42
1600 1040 1600 1040 1600
2900 1900 2900 1900 2900
30 30 <10 <10
Granular Activated Carbon
Water- Methanol -
based based
688 449
10 10
9-35 1
<1 <1
<1 <1
<1 <1
<1 <1
70 100
21 28
1040 1600
1900 2900
<10 <10
*A11 concentrations are mg/L and reflect published data and engineering estimates. Detailed
performance data and references are contained in the Control Technology Appendices.
-------
Section 4
Aq. Med. Int. Ex.
oxidized, the greatest fraction would probably be stripped from solution.
Since this hydrogen sulfide is transferred to the atmosphere at near ground
level, even low level emissions may cause odor problems.
Stripping may also occur with cyanide. At the near neutral conditions
found in the reactor, nearly all the cyanide would exist as the potentially
volatile hydrogen cyanide species (HCN). It is not certain how much cyanide
would be volatilized and how much would be oxidized biologically. For base
plant designs utilizing a methanol-based cyanide wash, both the cyanide and
sulfide concentrations in the composite wastewater stream would be much lower
than in the water-based case resulting in less potential for stripping. In
either case, if stripping of volatile species (Stream 416) is identified as
a problem, the activated sludge reactor could be covered. Air circulated
through the overhead space could be covered. Air circulated through the over-
head space could be routed to the boiler or to an incinerator.
Cyanide can be oxidized or otherwise removed by activated sludge systems
where the cyanide occurs as part of a matrix of predominatly organic species.
Greater than 90% cyanide removal has been recorded in the literature for care-
fully controlled laboratory-scale reactors. The exact cyanide removal that
could actually be realized in an operating system is highly uncertain. For
purposes of this manual, 60 to 90% removal through one or more stages of
activated sludge has been assumed. If this removal is not sufficient or is
determined to be unattainable, much of the cyanide can be converted to the
easily degradable thiocyanate species by adding polysulfide prior to the acti-
vated sludge process. Polysulfide addition is not expected to be required in
this application and therefore is not included in cost or performance esti-
mates. Details of polysulfide addition are presented in Section 4.2.1 and
4.2.3.
The activated sludge process is followed by filtration to remove sus-
pended solids, especially cells that escape the activated sludge clarifier.
343
-------
Section 4
Aq. Med. Int. Ex.
Although after filtration, the waste stream could possibly be discharged to
surface waters, it is expected that residual/refractory organics that could
not be oxidized by the activated sludge process will remain. Where it is
determined that these need to be removed, a granular activated carbon system
could be added. Granular activated carbon would also remove some trace metals
if these are present.
Example 2. Integrated pollution controls in Example 2 are illustrative
of those that might be used to treat the waste stream prior to ultimate dis-
posal by surface impoundment. Estimated characteristics of the wastewater
stream discharged to surface impoundment are presented in Table 4-35. The
controls are activated sludge and cooling tower concentration. Effluent char-
acteristics for each control are presented in Table 4-33.
All aspects of the activated sludge process are identical to those pre-
sented in Example 1. Following treatment by activated sludge, the waste
stream could possibly be suitable for discharge to a surface impoundment;
however, the capital requirements of surface impoundment can be decreased
greatly by reducing the volume of water to be held. The cooling tower would
be used for this purpose with the waste stream combined with raw water (or
recycle water) to supply the cooling tower makeup.
Cooling tower concentration is expected to be very attractive since this
system is a capital requirement of the base plant regardless of its source of
makeup water. Problems associated with using low quality makeup water to the
cooling tower have been discussed elsewhere (Section 4.2.1). Fouling by
biological solids escaping the activated sludge process or growing within the
tower can be controlled by biocides. Alternatively, the biological solids as
well as any other suspended material can be removed by filtering the activated
sludge effluent. Where scaling such as by calcium sulfate or corrision, par-
ticularly by chloride, are problems, various pretreatments would be needed.
However, at some point the cost of additional treatment to prepare the waste
stream suitable for cooling tower concentration would exceed any benefit of
344
-------
TABLE 4-33. EXAMPLE 2 - MATERIAL FLOW FOR MOBIL M SYNTHESIS BASE PLANT INTEGRATED CONTROLS
(FIGURE 4-6)
Base Plant
Cyanide Wash:
Characteristic*
Flow (m3/hr)
NH3
CN"
S=
£ S2°3
en =
so3
SCN"
COD
BOD5
cr
TDS
TSS
Composite
Water- Methanol -
based based
688
92
87
63
48
12
2.5
2200
1100
1040
1900
--
449
140
5
2.6
74
18
3
3400
1700
1600
2900
--
Activated Sludge Cooling Tower Concentration
Water- Methanol- Water- Methanol -
based based based based
688 449 180
10 10 38f
9-35 1 34-130*
<1 <1 <1
<1 <1 4
<1 <1 4
<1 <1 4
680 1060 2600
35 42 130
1040 1600 3900
1900 2900 7200
30 30 110
180
25f
2*
<1
2
2
2
2650
105
4000
7200
75
*A11 concentrations are mg/L and reflect published data and engineering estimates. Detailed
performance data and references are contained in the Control Technology Appendices.
tNo ammonia volatilization is assumed.
*Loss by volatilization/biological oxidation uncertain - no losses assumed.
-------
Section 4
Aq. Med. Int. Ex.
reducing its volume for downstream handling. The feasibility of this system
will, therefore, have to be determined on an individual case basis once the
compositions of the waste streams are known with some certainty.
Blowdown from the cooling tower will contain the original constituents
in the makeup stream but concentrated by a factor that is the ratio of makeup
flow to blowdown. Some losses will result from volatilization or entrainment
in the drift (Stream 419), but the exact amounts are uncertain and must be
evaluated on an individual case basis. The blowdown stream would be routed
to ultimate disposal by surface impoundment.
Example 3. Integrated pollution controls in Example 3 of Figure 4-6 are
illustrative of those that might be utilized prior to ultimate disposal of
the wastewater stream by a combination of recycle and deep well injection.
Estimated characteristics of the wastewater stream that is deep well injected
are summarized in Table 4-35. The controls are activated sludge, cooling
tower concentration, forced evaporation, and incineration. Effluent charac-
teristics for each control process are presented in Table 4-34.
Example 3 differs from Example 2 only in the handling of the cooling
tower blowdown. The blowdown is routed to forced evaporation where the orig-
inal waste stream constituents are concentrated into a low flow stream. Cer-
tain characteristics of the waste stream may limit the usability of forced
evaporation; these considerations have been addressed in Section 4.2.1. The
forced evaporation overhead is virtually free of contaminants and would be
recycled for use as high quality makeup water within the facility. The
forced evaporation concentrate would be deep well injected (or alternatively
surface impounded) after it has been incinerated to remove high concentrations
of organics that would otherwise foul the injection formation.
Costs for integrated pollution controls shown in Figure 4-6 are presented
in Table 4-36. All costs depend on the type of cyanide wash used in the base
plant design. For each integrated example, costs are presented in total
346
-------
TABLE 4-34. EXAMPLE 3 - MATERIAL FLOW FOR MOBIL M SYNTHESIS BASE PLANT INTEGRATED CONTROLS
(FIGURE 4-6)
Base Plant
Cyanide Wash :
Cooling Tower
Composite Activated Sludge Concentration Forced Evaporation Incineration
Water- Methanol- Water- Methanol- Water- Methanol- Water- Methanol- Water- Methanol -
based based based based based based based based based based
Characteristic*
Flow (m3/hr)
NH3
CN"
S=
S2°3
S03
SCN"
COD
BODC
0
cr
TDS
TSS
688
92
87
63
48
12
2.5
2200
1100
1040
1900
--
449
140
5
2.6
74
18
3
3400
1700
1600
2900
--
688 449 180
10 10 38
9-35 1 34-130
<1 <1 <1
<1 <1 4
<1 <1 4
<1 <1 4
680 1060 2600
35 42 130
1040 1600 3900
1900 2900 7200
30 30 110
180
25
2
<1
2
2
2
2650
105
4000
7200
75
18
380
340-1300
<1
40
40
40
26000
1300
39000
72000
1100
18
250
20
<1
20
20
20
26500
1050
40000
72000
750
18
0.2
0.3-1
<1
40
40
0.04
26
1
39000
72000
—
18
0.1
0.02
<1
20
20
0.02
26
1
40000
72000
—
*A11 concentrations are mg/L and reflect published data and engineering estimates. Detailed
performance data and references are contained in the Control Technology Appendices.
-------
TABLE 4-35. ESTIMATED CHARACTERISTICS OF WASTEWATER STREAMS DISCHARGED
TO ULTIMATE DISPOSAL - MOBIL M SYNTHESIS CASE
00
Integrated Controls
Ultimate Disposal :
Base Plant Cyanide
Characteristic*
Flow (m3/hr)
NH3
CN"
S=
S2°3
S03
SCN"
COD
BOD5
cr
TDS
TSS
Example 1 Example 2
Surface Surface
Waters Impoundment
Wash: Water Methanol Water Methanol
688 449 180
10 10 38
9-35 1 34-130
<1 <1 <1
<1 <1 4
<1 <1 4
<1 <1 4
70 100 2600
21 28 130
1040 1600 3900
1900 2900 7200
<10 <10 110
180
25
2
<1
2
2
2
2650
105
4000
7200
75
Example 3
Deep Well
Injection
Water Methanol
18
0.2
0.3-1
<1
40
40
0.04
26
1
39000
72000
"
18
0.1
0.02
-------
TABLE 4-36. COSTS OF INTEGRATED CONTROL EXAMPLES - MOBIL M SYNTHESIS CASE
Total Capital Total Annual ized % of Base Plant
Investment* Cost* Capital Cost
Base Plant Cyanide Wash
Integrated Control Example
Example 1
Control s :
oo Additional
10 Control:
Example 2
Controls:
Example 3
Control s :
Activated Sludge
Filtration
Granular Activated Carbon
Activated Sludge
Surface Impoundment
Activated Sludge
Forced Evaporation
Incineration
Deep Well Injection
Ultimate Disposal Water-
Technique based
Surface waters 19.5
18.3
1.2
3.1
Surface Impoundment 26.6
18.3
8.3
Deep well injection 39.4
18.3
13.7
4.8
2.6
Methanol -
based
19.1
18.3
0.78
3.0
26.6
18.3
8.3
39.4
18.3
13.7
4.8
2.6
Water-
based
6.5
3.9
2.6
1.5
5.6
3.9
1.7
10.3
3.9
3.5
2.3
0.56
Methanol- Water-
based based
5.7 1.6
3.9
1.8
1.6
5.6 2.2
3.9
1.7
10.3 3.2
3.9
3.5
2.3
0.56
Methanol -
based
1.6
2.2
3.2
*Mil1 ion dollars
-------
Section 4
Aq. Med. Int. Ex.
capital investment and total annualized cost for both individual controls and
the entire integrated system; and percent of base plant cost for the entire
integrated system. Costs of additional controls that are possibly applicable
have also been included.
350
-------
Section 4
Aq. Med. Int. Ex.
4.2.4.2 Treatment of Waste Streams from Base Plants Producing
Crude Methanol
The integrated wastewater control examples presented thus far have been
designed to treat a highly organic waste stream from a synthesis section (F-T
or Mobil M) combined with strictly inorganic waste streams from the gasifier
section. The resulting composite flow is highly amenable to treatment by
a conventional activated sludge process; the high organic concentrations are
diluted to an appropriate level for biological oxidation, and the inorganic
species from the gasifier wastewaters are biological oxidized, utilized as
growth nutrients and thereby assimilated as new cell material, or are other-
wise removed.
This water management strategy is not usable for K-T facilities where
crude methanol is produced since no. high volume waste stream with orqanics
equivalent to that generated by F-T or Mobil M synthesis would exist. The
composite waste stream would consist exclusively of inorganic species, partic-
ularly ammonia, cyanide, thiocyante, and a variety of reduced sulfur species.
There is no known precedent for treating a waste stream having significant
concentrations of each of the cited species. Due to these unique character-
istics, some novel treatment approaches are worth considering. Two examples
of integrated controls are presented in Figure 4-7 for cases where ultimate
disposal is by discharge to surface waters, and examples for cases where
ultimate disposal is by surface impoundment or deep well injection are pre-
sented in Figure 4-8. Each addresses the two base plant designs of cyanide
removal by a water or methanol wash.
Cyanide can be partially destroyed in biox processes. Microorganisms
in an activated sludge process can be acclimated to oxidize cyanide but only
where the cyanide is part of a larger matrix dominated by biodegradable
organics. However, cyanide can be removed from stricly inorganic wastewater
streams by converting it to thiocyanate then oxidizing the thiocyanate.
351
-------
EXAMPLE 4
CYANIDE
WASH
WATER*
1
STREAM
215
POLYSULFIDE
ADDITION
COMPRESSION CONDENSATE
STREAM 211
GAS COOLING/DUST
REMOVAL SLOWDOWN
STREAM 210
FLOW
EQUALIZATION
DISCHARGE
i 1 TO SURFACE
'CHEM.CAL ! WATERS .
OXIDATION | r
I I
EXAMPLE 5
CYANIDE
WASH
WATER*
STREAM
215
COMPRESSION CONDENSATE
STREAM 211
GAS COOLING/DUST
REMOVAL SLOWDOWN
ACTIVATED |_J FILTRA- |_J
SLUDGE |~n TION In OXIDATION
I I I I 1
DISCHARGE
TO SURFACE
WATERS
STREAM 210
"This stream exists only for base plants where cyanide is removed from the raw gas by a water-based wash.
Figure 4-7. Integrated control examples for base plants producing crude methanol - discharge to surface
waters
-------
EXAMPLE 6
GO
cn
OJ
STREAM 215
COMPRESSION CONDENSATE
STREAM 211
GAS COOLING/DUST REMOVAL SLOWDOWN
STREAM 210
EXAMPLE 7
CYANIDE
WASH
WATER"
STREAM 215
COMPRESSION CONDENSATE
STREAM 211
GAS COOLING/DUST REMOVAL SLOWDOWN
STREAM 210
EXAMPLE 8
CYANIDE
WASH
WATER*
STREAM 215
COMPRESSION CONDENSATE
STREAM 211
GAS COOLING/DUST REMOVAL SLOWDOWN
STREAM 210
DISCHARGE TO
SURFACE
IMPOUNDMENT
CLARIFIER I >|
I 1
A ALTERNATE
T r
DISPOSAL
DISCHARGE
TO DEEP WELL
INJECTION
'This stream exists only for base plants where cyanide is removed from the raw gas by a water-based wash.
Figure 4-8. Integrated control examples for base plants producing crude methanol - discharge to
surface impoundment or deep well injection
-------
Section 4
Sq. Med. Int. Ex.
Laboratory scale studies have demonstrated that thiocyanate can be biologi-
cally oxidized even where it and ammonia are the only usable growth substrates
for the microorganisms. Sulfur, when added as polysulfide, would drive the
conversion.
Assuming excess polysulfide addition, the influent and desired effluent
concentrations of cyanide determine the reaction time required for the thio-
cyanate conversion. In general, the effluent concentration is more critical
since large increases in reaction time are required to achieve small decreases
in the effluent cyanide concentration at effluent levels less than about 10
mg/L cyanide. Therefore, polysulfide addition does not appear to be economical
for waste streams such as the gas cooling/dust removal blowdown (Stream 210)
and the compression condensate (Stream 211) having less than 10 mg/L cyanide.
However, both the cyanide concentration and loading of the cyanide wash water
(Stream 215) are expected to be well over an order of magnitude greater than
that of the other two streams. By treating the cyanide wash water with poly-
sulfide separately, the reaction can be accomplished under kinetically more
favorable conditions than if all three streams had first been combined. Since
the cyanide wash water would then be combined with the gas-cooling/dust re-
moval blowdown and compression condensate and since both of these streams con-
tain cyanide at about the 10 mg/L level, the cyanide wash water would be
treated by polysulfide addition to a corresponding level. If it is desired
to decrease the cyanide concentration in the composite stream to a lower level,
downstream controls could be applied.
For each integrated control example (Figures 4-7 and 4-8) where cyanide
is removed from the base plant by a water wash, polysulfide addition would
be desirable as a pretreatment step. Characteristics of the cyanide wash water
(Stream 215) before and after polysulfide addition are presented in Table 4-37.
The characteristics of the compression condensate (Stream 211) and of the gas
cooling/dust removal blowdown (Stream 210) are also presented in Table 4-37
354
-------
TABLE 4-37. CHARACTERIZATION OF MAJOR STREAMS TO BE COMBINED FOR TREATMENT - CRUDE METHANOL
PRODUCTION CASE
OJ
en
on
Cyanide
Nash
Water
Characteristic* Stream 215
Flow (m3/hr) 239
NH3
CN" . 241
S= 176
S2°3
S03
SCN"
cr
TDS
Gas Cooling/
Cyanide Wash Dust Removal
Pretreated with Slowdown
Polysulfide Stream 210
239 322
360 156
8 7
<1 1
103
25
495 4
2111
4000
Compression
Condensate
Stream 211
17.4
735
8.9
48.7
6.3
--
14
2200
--
Composite
Base Plant Cyanide Wash
Water- Methane! -
Based1" Based*
578
260
7.5
2
58
14
210
1200
2200
339
190
7
3.5
98
24
4.5
2100
3800
*A11 concentrations are mg/L.
^Combined streams: cyanide wash water pretreated by polysulfide addition,
gas cooling/dust removal blowdown, and compression condensate.
^Combined streams: gas cooling/dust removal blowdown and compression condensate.
-------
Section 4
Aq. Med. Int. Ex.
as are the characteristics of the composite flows from base plants using a
water-based or methanol-based cyanide wash.
As was the case for integrated pollution controls for a base plant having
a Mobil M or F-T synthesis section, each of the examples (Figures 4-7 and 4-8)
for the subject crude methanol production case would use flow equalization.
Example 4, Figure 4-7. Integrated pollution controls in Example 4 of
Figure 4-7 are illustrative of those that might be utilized prior to ultimate
disposal of the composite wastewater stream by direct discharge to surface
waters. The controls are activated sludge, denitrification, and filtration.
Characteristics of the effluent from each control are summarized in Table 4-38,
and the estimated characteristics of the wastewater stream discharged to sur-
face waters by this illustrative example as well as by the other examples
presented in Figures 4-7 and 4-8 are presented in Table 4-43.
In the activated sludge process, both ammonia and thiocyanate would be
oxidized. An oxidation product of each of these species is nitrate; it would
be biologically reduced to molecular nitrogen in a following denitrification
reactor.
As discussed for Example 4 of Figure 4-6, it may be advisable to cover
and vent to an incinerator both the polysulfide and the nitrification reactors
to prevent loss of the potentially volatile hydrogen sulfide and hydrogen
cyanide species. Following denitrification, loss of volatile species is not
expected to present problems.
Following deniitrification, the waste stream is filtered to remove sus-
pended solids that escape the denitrification clarifier. At this point,
the waste stream could possibly be discharged to surface waters. If addi-
tional treatment is desirable, particularly to further reduce the cyanide con-
centration, chemical oxidation can be included.
Example 5, Figure 4-7. Integrated pollution controls in Example 5 of
Figure 4-7 are also aimed toward ultimate disposal of the composite wastewater
356
-------
TABLE 4-38. EXAMPLE 4 - MATERIAL FLOW FOR CRUDE METHANOL PRODUCTION BASE PLANT INTEGRATED CONTROLS
(FIGURE 4-7)
co
en
Base Plant
Cyanide Wash:
Composite
Water- Methanol -
based based
Nitrification Denitrification Filtration
Water- Methanol- Water- Methanol- Water- Methanol -
based based based based based based
Characteristic*
Flow (m3/hr)
NH3
CN"
S=
S90^
u. O
S03
SCN"
ci-
TDS
NO~-N
TSS
578
260
7.5
2
58
14
210
1200
2200
2200 >3800 >2200 >3800 >2200 >3800
245 140 24 15 24 15
<50 <50 30 30 <10 <10
*A11 concentrations are mg/L and reflect published data and engineering estimates. Detailed
performance data and references are contained in the Control Technology Appendices.
-------
TABLE 4-39. EXAMPLE 5 - MATERIAL FLOW FOR CRUDE METHANOL PRODUCTION BASE PLANT INTEGRATED CONTROLS
(FIGURE 4-7)
Base Plant
Cyanide Wash :
Characteristic*
Flow (m3/hr)
NH3
CN"
S=
s?o"
CO *- °
on =
09 S03
SCN"
ci-
TDS
NO--N
TSS
Composite
Water- Methanol -
based based
578
260
7.5
2
58
14
210
1200
2200
--
--
339
190
7
3.5
98
24
4.5
2100
3800
--
--
Ion Exchange Activated Sludge Filtration
Water- Methanol- Water- Methanol- Water- Methanol -
based based based based based based
578
12
7.5
2
58
14
210
>1200
>2200
--
--
339 578 -- 578
12 12 -- 12
7 7.5 — 7.5
3.5 <1 -- <1
98 <1 -- <1
24 <1 — <1
4.5 <1 -- <1
>2100 >1200 -- >1200
>3800 >2200 — >2200
60 — 60
30 - <10
*A11 concentrations are mg/L and reflect published data and engineering estimates. Detailed
performance data and references are contained in the Control Technology Appendices.
-------
Section 4
Aq. Med. Int. Ex.
stream by direct discharge to surface waters. Estimated characteristics of
the wastewater stream discharged to surface waters are presented in Table
4-43. For the methanol-based cyanide wash case, ion exchange is the primary
control. Activated sludge, filtration, and chemical oxidation can also be
included if additional treatment is desired. Estimated performance of indi-
vidual controls is presented in Table 4-39.
As is the case for all examples in Figure 4-7 and 4-8, where the cyanide
wash water occurs, it is pretreated by polysulfide addition. This reaction
must be completed at alkaline conditions, but downstream pollution control proc-
esses require near neutral or only slightly alkaline conditions. If the
acidity of the streams with which the cyanide wash would be combined is not
sufficient to produce a composite flow having a stable, near neutral condition,
the cyanide wash should be first neutralized by acid addition. In any case,
under neutral conditions, residual polysulfide may precipitate. This pre-
cipitate is not expected to cause significant problems with the nitrification
process of Example 4, but for the following ion exchange process of this
example, plugging and fouling would result. For purposes of this manual, the
cyanide wash is assumed to be neutralized following polysulfide addition so
that, if needed, a filter or clarifier can be dedicated separately to this
stream.
Ion exchange using an exchange resin of clinoptilolite would remove
ammonium ions from the waste stream and replace them with sodium ions. All
known experience to date with this system is associated with removing ammonia
from low-strength municipal wastewaters. However, preliminary, EPA supported
studies strongly suggest that the process would be equally feasible for a
waste stream having the characteristics of the K-T composite stream.
For base plant cases using a methanol-based cyanide wash, the composite
wastewater stream may be suitable for discharge to surface waters following
ion exchange. If additional treatment is determined necessary, chemical oxi-
dation can be included. For the case of the water-based cyanide wash, a
359
-------
Section 4
Sq. Med. Int. Ex.
biological treatment process followed by a filter would be included following
ion exchange to oxidize the thiocyanate that had been converted from cyanide
by upstream polysulfide addition.
Example 6. Figure 4-8. Examples of integrated controls presented in
Figure 4-8 are illustrative of cases where the ultimate discharge technique
is either surface impoundment or complete recycle with deep well injection.
Estimated characteristics of the wastewater stream discharged to surface im-
poundment or deep well injected are presented in Table 4-43 for each of the
three examples. Each example utilizes cooling tower concentration to reduce
the volume of water to be handled for disposal. The cooling tower blowdown
would be disposed of by surface impoundment, or if recovery of water for
reuse is desired, the blowdown would be routed to forced evaporation. The
forced evaporation overhead would contain only trace amounts of contaminants
and would be recycled for use as makeup water within the facility. Concen-
trate from this process would be disposed of by deep well injection or sur-
face impoundment.
Since cooling tower concentration is common to each of the examples in
Figure 4-8, upstream control processes are determined by the requirements of
this step. Problems with scaling, fouling, and corrosion that may limit the
feasibility of cooling tower concentration have been discussed in Section
4.2.1. It has been assumed that no extraordinary pretreatments would be
required to correct these problems. Control processes are assumed only to
remove volatile species that may be lost to the atmosphere during the con-
centration process.
In general, it is not certain how much of a particular volatile species
would be lost to the atmosphere during cooling tower concentration. Addi-
tionally, different losses may be acceptable at different locations. It is
assumed, however, that the concentration of cyanide in the cooling tower
makeup would have to be at low levels since at the expected pH operating
range nearly all the cyanide would exist as the potentially volatile hydrogen
360
-------
Section 4
Aq. Med. Int. Ex.
cyanide species. Therefore, all cases where cyanide is removed from the base
plant by a water wash, polysulfide addition is included as a pretreatment step.
Example 6 represents a case where no treatment of the composite waste-
water stream is included prior to its being used as cooling tower makeup.
In some facilities, this approach might be feasible; however, this approach
is not very likely due to expected volatilization of ammonia and sulfide in
the cooling tower. Estimated characteristics of the composite wastewater
stream following cooling tower concentration and forced evaporation are pre-
sented in Table 4-40.
Example 7. Figure 4-8. It is very likely that the composite wastewater
stream would be treated for removal of volatile species prior to being routed
to cooling tower makeup. Example 7 utilizes activated sludge. Estimated
characteristics of the subject wastewater stream through all control processes
are presented in Table 4-41. All other details are identical to those pre-
sented in Example 4 of Figure 4-7. Where ft is determined that even higher
quality makeup water is desired, filtration can be added to remove suspended
solids and chemical oxidation added to oxidize ammonia, cyanide, and other
remaining reduced species.
Example 8, Figure 4-8. Like Example 7, Example 8 prepares the subject
wastewater stream for use as makeup to the cooling tower. The control pro-
cess for this example is ion exchange. Details are identical to those pre-
sented in Example 5 of Figure 4-7. Where additional treatment is needed,
chemical oxidation could be added. Characteristics of the subject wastewater
stream through all control processes are presented in Table 4-42.
Costs for integrated pollution controls for each example presented in
Figures 4-7 and 4-8 are summarized in Table 4-44. All costs depend on the
type of cyandie wash used in the base plant design. For each integrated
example, costs are presented as total capital investment and total annualized
cost for both individual controls and the entire integrated system; and as a
361
-------
Section 4
Aq. Med. Int. Ex.
percentage of base plant cost for the entire integrated system. Costs of
additional controls that are possibly applicable are also presented.
362
-------
TABLE 4-40. EXAMPLE 6 - MATERIAL FLOW FOR CRUDE METHANOL PRODUCTION BASE PLANT INTEGRATED CONTROLS
(FIGURE 4-8)
Base Plant
Cyanide Wash:
Characteristic*
Flow (m3/hr)
NH3
CN-
S=
GO S 0=
S 2°3
S03
SCN"
ci-
TDS
Composite
Water- Methanol-
based based
578
260
7.5
2
58
14
210
1200
2200
339
190
7
3.5
98
24
4.5
2100
3800
Cooling Tower
Concentration
Water- Methanol-
based based
161
933
27
<1
205
50
740
4500
7900
161
400
15
<1
205
50
8
4400
8000
Forced Evaporation1"
Water- Methanol-
based based
16
9330
270
<1
2050
500
7400
45000
79000
16
4000
150
<1
2050
500
80
44000
80000
*A11 concentrations are mg/L and reflect published data and engineering estimates. Detailed
performance data and references are contained in the Control Technology Appendices.
^'Required for recycle only.
-------
TABLE 4-41. EXAMPLE 7 - MATERIAL FLOW FOR CRUDE METHANOL PRODUCTION BASE PLANT INTEGRATED CONTROLS
(FIGURE 4-8)
GO
cr>
Base Plant
Cyanide Wash :
Characteristic*
Flow (m3/hr)
NH3
CN"
s=
S2°3
S03
SCN"
Cl"
TDS
NO~-N
TSS
Composite Activated Sludge
Water- Methanol- Water- Methanol -
based based based based
578
260
7.5
2
58
14
210
1200
2200
-
•"
339 578 339
190 10 7
7 7.5 7
3.5 <1 <1
98 <1 <1
24 <1 <1
4.5 <1 <1
2100 1200 2100
3800 >2200 >3800
245 140
30 30
Cooling Tower
Concentration
Water- Methanol -
based based
161
36
27
<1
4
4
4
4300
>7900
880
110
161
15
15
<1
2
2
2
4400
>8000
300
60
Forced
Water-
based
16
360
270
<1
40
40
40
43000
> 79000
8800
1100
Evaporation*
Methanol-
based
16
150
150
<1
20
20
20
44000
> 80000
3000
600
*A11 concentrations are mg/L and reflect published data and engineering estimates. Detailed
performance data and references are contained in the Control Technology Appendices.
^Required for recycle only.
-------
TABLE 4-42. EXAMPLE 8 - MATERIAL FLOW FOR CRUDE METHANOL PRODUCTION BASE PLANT INTEGRATED CONTROLS
(FIGURE 4-8)
Composite
Base Plant Water- Methanol-
Cyanide Wash: based based
Characteristic*
Flow (m3/hr) 578
NH3 260
CN~ 7.5
S= 2
# S2°3 58
01 J
S03 14
SCN" 210
cr 1200
TDS 2200
339
190
7
3.5
98
24
4.5
2100
3800
Ion Exchange
Water- Methanol-
based based
578
12
7.5
2
57
14
210
>1200
>2200
339
12
7
3.5
98
24
4.5
>2100
>3800
Cooling Tower
Concentration
Water- Methanol-
based based
161
43
27
<1
205
50
740
>4300
>7900
161
25
15
<1
205
50
8
>4400
>8000
Forced Evaporation"1"
Water- Methanol-
based based
16
430
270
<1
2050
500
7400
> 43000
> 79000
16
250
150
<1
2050
500
80
>45000
>80000
*A11 concentrations are mg/L and reflect published data and engineering estimates. Detailed
performance data and references are contained in the Control Technology Appendices.
''"Required for recycle only.
-------
TABLE 4-43. ESTIMATED CHARACTERISTICS OP WASTEWATER STREAMS DISCHARGED TO ULTIMATE
DISPOSAL - CRUDE METHANOL PRODUCTION CASE
CO
en
en
Integrated Controls
Figure:
Ultimate Disposal:
Base Plant
Cyanide Wash:
'Characteristic*
Flow (m3/hr)
NH3
CN-
S=
S2°3
S0°
SCIT
ci-
TDS
N03"-N
TSS
: Example 4
4-7
Surface Waters
Water Methanol
578 339
8 6
7.5 7
-1 <1
<1 <1
•1 <1
1 2100
.2200 >3800
60
<10
Example 6
4-8
Surface
Impoundment
Water Methanol
161 161
933 400
27 15
<1 <1
205 205
50 50
740 8
4500 4400
7900 8000
-
-
Example 6
4-8
Deep Well
Injection
Water
16
9330
270
<1
2050
500
7400
45000
79000
-
-
Methanol
16
4000
150
<1
2050
500
80
44000
80000
-
-
Example 7
4-8
Surface
Impoundment
Water
161
36
27
<1
4
4
4
4300
.7900
880
110
Methanol
161
15
15
<1
2
2
2
4400
>8000
300
60
Example 7
4-8
Deep Well
Injection
Water
16
360
270
<1
40
40
40
43000
.79000
8800
1100
Methanol
16
150
150
<1
20
20
20
44000
.80000
3000
600
Example 8
4-8
Surface
Impoundment
Water
161
43
27
<1
205
50
740
.4300
.7900
-
--
Methanol
161
25
15
<1
205
50
8
.4400
.8000
—
-
Example 8
4-8
Deep Well
Injection
Water
16
' 430
270
43000
>79000
-
-
Methanol
16
250
150
<1
2050
500
80
>45000
>80000
--
-
All concentrations are mg/L.
-------
TABLE 4-44. COSTS OF INTEGRATED CONTROL EXAMPLES - CRUDE METHANOL
PRODUCTION CASE
Integrated Control Example
Example 4, Figure 4-7
Controls: Polysulfide
Activated Sludge
Dem trif i cation
Filtration
Additional
Control: Chemical Oxidation
Example 5, Ficure 4-7
Controls: Polysulfide
Ion Exchange
Activated Sludge
Filtration
Additional
Controls: Filtration
Chemical Oxidation
Example 6, Figure 4-8
Controls: Polysulfide
Surface Impoundment
Forced Evaporation
Deep Well Injection
Example 7> Figure 4-8
Controls: Polysulfide
Activated Sludge
Surface Impoundment
Forced Evaporation
Deep Well Injection
Additional
Controls. Filtration
Chemical Oxidation
Example 8, Figure 4-8
Controls: Polysulfide
Ion Exchange
Surface Impoundment
Forced Evaporation
Deep Well Injection
Additional
Controls: Filtration
Chemical Oxidation
Total Capital
Investment*
Ultimate Disposal Water- Methanol-
Method based based
Surface Waters 19.8
0.13
12.6
6.1
1.0
0 72
Surface Waters 18.8
0.13
12.9
4 8
1 0
0.42
0.72
Surface Impoundment 8.4
Deep Well Injection 16.4
0.13
8.3
13 7
2.6
Surface Impoundment 21.0
Deep Well Injection 29 0
0.13
12.6
8.3
13 7
2.6
1.0
0 72
Surface Impoundment 21.4
Deep Well Injection 29.3
0.13
12.9
8.3
13.7
2.6
0.42
0.72
10 5
-
6.7
3,2
0.59
0.16
8.3
-
8.3
-
-
-
0.16
8.3
16.3
-
8.3
13.7
2.6
15.0
23.0
-
6.7
8.3
13.7
2.6
0.59
0.16
16.6
24.6
-
8.3
8.3
13.7
2.6
-
0.16
Total Annual i zed Percent of Base
Cost* Plant Cost
Base Plant Cyanide Wash
Water- Methanol- Water- Methanol-
based based based based
5.8
0.73
2.7
2.2
0.22
0.23
5.6
0.73
3.5
1.1
0.22
0.11
0.23
2.4
4.8
0.73
1.7
3.5
0.56
5.1
7.5
0.73
2.7
1 7
3.5
0.56
0.22
0.23
5.9
8.3
0.73
3.5
1.7
3.5
0.56
0.11
0.23
2.6 1.8
-
1.5
0.98
0.15
0.12
1.8 1.7
-
L.8
-
-
0.35
1.7 0.8
4.1 1.5
-
1.7
3.5
0.56
3.2 1.9
5.6 2.5
-
1.5
1.7
3.5
0.56
0.15
0.12
3.5 1.9
5.9 2.6
-
1.8
1.7
3.5
0.56
-
0.35
0.9
0.7
0.8
1.5
1.4
2.0
1.5
2 2
Million dollars
367
-------
Section 4
Aq. Med. Int. Ex.
4.2.4.3 Treatment of Waste Streams from Base Plants Producing Fuel Grade
Methanol
For base plant cases where fuel grade methanol is produced, an organic
waste stream would be generated by the methanol distillation step. However,
when this waste stream is combined with gasification wastewaters, the total
stream would probably not contain a high enough organic loading to be handled
by integrated controls analogous to those of the F-T and Mobil M synthesis
cases. Instead, controls would be similar to those in the crude methanol case
with some modifications. Appropriate modifications of applicable integrated
control examples of Figures 4-7 and 4-8 are considered in this section.
Example 4, Figure 4-7. The integrated controls presented in Example 4 of
Figure 4-7 would be applicable to base plant cases where fuel grade methanol
is produced. For base plant designs where cyanide is removed from the raw gas
by a water-based wash, the methanol distillation condensate (Stream 229) could
be routed to the denitrification reactor to provide the source of organics
for reduction of nitrate to molecular nitrogen. For base plant designs where
cyanide is removed by a methanol-based wash, the organic loading of the metha-
nol distillation condensate would exceed that needed by the denitrification
reactor. The excess could be routed upstream to the nitrification reactor.
In either case, capital costs are expected to change by little, but annualized
costs would decrease by about $814,000 and $296,000 for the two respective
cases.
Example 5, Figure 4-7. Example 5 of Figure 4-7 would be applicable to
plants producing fuel grade methanol, but unlike the crude methanol case dis-
cussed earlier, the activated sludge process would be needed for both the
water- and methanol-based cyanide wash cases. The methanol distillation
condensate (Stream 229) would be routed to the activated sludge process, and
most of the nutrient requirements of this biological system would be met by
bleeding ammonia through the upstream, clinoptilolite-based ion exchange
process.
368
-------
Section 4
Aq. Med. Int. Ex.
Example 6, Figure 4-8. For Example 6 of Figure 4-8 to be applicable to
fuel grade methanol plants, an activated sludge system would probably have to
be added. This example would then be essentially identical to the following,
Example 7.
Example 7, Figure 4-8. Example 7 of Figure 4-8 would be directly appli-
cable to base plants producing fuel grade methanol. The methanol distillation
condensate (Stream 229) could be routed to the activated sludge reactor.
Capital and annualized costs would be affected only slightly.
Example 8, Figure 4-8. To be applicable to base plants producing fuel
grade methanol, an activated sludge reactor would be added to Example 8 of
Figure 4-8 to follow the ion exchange process. The methanol distillation con-
densate (Stream 229) could be routed to the activated sludge reactor, and most
nutrient requirements for this system would be met by bleeding some ammonia
through the upstream ion exchange process. Capital and annualized cost changes
would be reflected by the requirements of the added activated sludge process.
369
-------
Section 4
Solid Waste Management
4.3 SOLID WASTE MANAGEMENT
Many solid waste streams including ashes and sludges are generated in a
K-T based indirect liquefaction facility. The available control techniques
that are applicable to these streams are identified and evaluated in this
section. In comparison with air and water pollution control operations,
solid waste management techniques available for a K-T indirect liquefaction
facility are fewer and also more site specific. In addition, the quantities
and characteristics of some solid waste streams (e.g., brines and sludges) are
affected by the processes selected for air and water pollution control. Be-
cause of this, solid waste management at a K-T indirect liquefaction facility
is not an isolated issue but rather an element in the total program for pollu-
tion control.
The sources and factors affecting the characteristics of the solid waste
streams generated in K-T-based indirect liquefaction facilities are summarized
in Table 4-45. Of the streams listed, dewatered dust from gasification opera-
tion is by far the largest volume stream. Depending on the synthesis process
used, it comprises 45% to 64% of the total solid waste generated from the plant.
Flue gas desulfurization (FGD) sludge from a lime/limestone process and gasifier
slag are the next largest volume streams, comprising up to 35% of the total
waste generated.
Coal type has considerable impact on the characteristics of the solid
wastes generated, and thus the control required. Major coal properties affect-
ing the solid waste control approaches include: ash content, sulfur content,
and ash acidity/alkalinity. Coals with higher ash or sulfur contents will
result in the generation of larger quantities of gasifier ash or FGD sludges,
and thus will require larger control facilities.
The gasifier and boiler ash generated may be acidic or alkaline. Alka-
line fly ash has been used in existing power plants to mix with FGD sludges
before disposal. Mixing these two may result in a material structurally more
370
-------
TABLE 4-45. SUMMARY OF SOLID WASTE STREAMS FROM K-T BASED INDIRECT
LIQUEFACTION FACILITIES
Stream
Pollutants of
Potential Concern
Factors Affecting Waste
Stream Characteristics
From Main Process Train
Quenched Gasifier Slag
(Stream 207)
Dewatered Gasifier Dust
(Stream 209)
Spent Shift Catalyst
(Stream 217)
Leachable trace
elements, inorganic
compounds
Leachable trace
elements
Feed coal characteristics,
gasifier operating condi-
tions quench water
characteristics.
Catalyst composition,
life
Spent Sulfur Guard
(Stream 218)
Leachable trace
elements
Sulfur guard composition,
AGR effluent gas
characteristics.
Spent Methanation
Catalyst (Stream 238)
Spent NOX Reduction
Catalyst (Stream 212)
Spent F-T Catalyst
(Stream 222)
Spent Methanol Catalyst
(Stream 227)
Spent Mobil M Catalyst
(Stream 232)
From Auxiliary Processes
Boiler Bottom Ash
(Stream 304)
Raw Water Treatment
Sludges (Stream 300)
Leachable trace
elements
Leachable trace
elements
Leachable trace
elements
Leachable trace
metals
Catalyst life, decommission-
ing procedure, synthesis
gas characteristics.
Catalyst life, regenera-
tion frequency, synthesis
gas characteristics
Feed coal characteristics,
boiler operating condi-
tions, slurry water
characteristics.
Raw water characteristics,
treatment system design
and operation.
(continued)
371
-------
TABLE 4-45. (Continued)
Stream
Pollutants of
Potential Concern
Factors Affecting Waste
Stream Characteristics
From Pollution Control
Boiler Fly Ash
(Stream 302)
FGD Sludges from Boiler*
(Stream 424)
Spent Claus Catalyst
(Stream 402)
Spent Beavon Catalyst
(Stream 407)
Spent SCOT Catalyst
(Stream 410)
Fly Ash from FBC Boiler
(Stream 413)
Spent Sorbent from FBC
Boiler (Stream 414)
Biological Sludge from
Wastewater Treatment
(Stream 415)
Leachable trace
elements potential
dust emissions
Leachable trace
elements
Leachable trace
elements
Leachable trace
elements potential
dust emissions
Leachable trace
elements
Leachable organics
and trace metals
Feed coal characteristics,
boiler operating condi-
tions, slurry water
characteristics.
Coal sulfur content,
process design and
operation.
FGD
Catalyst life, regenera-
tion frequency, acid gas
characteristics.
Feed coal characteristics,
boiler operating condi-
tions, sorbent carryover.
Feed coal characteristics,
FBC boiler operating
conditions, sorbent
characteristics.
Wastewater treat-
ment system design
and operation.
Recovered Sulfur
(Streams 403 and 408)
Vanadium and cyanides
(with Beavon/Stretford
only)
Coal sulfur
bulk sulfur
efficiency.
content
removal
Collected Dust from
Particulate Control
(Stream 400)
Potential dust
emissions
Dust collection process
design and operation.
For the wet limestone process.
372
-------
Section 4
Solid Waste Control
suitable for landfill ing. The acidity/alkalinity nature of ashes may affect
their leaching characteristics and thus influence control practices.
The solid waste streams listed in Table 4-45 can be classified according
to four waste type categories (source types) which are based on the nature of
the waste. These waste types are: inorganic ashes, recovered by-products
which prove to be unsalable, organic sludges, and spent catalysts. Several
control techniques are potentially applicable to these streams.
In general, solid waste control techniques aim at containing the entire
waste stream. Thus, the performance of these techniques, in terms of removal
or control efficiencies, are generally 100 percent. However, unless designed
and operated properly, secondary waste streams with undesirable characteris-
tics may be generated and migrate away from the site employing the technique.
For example, runoff can contaminate surface water and percolating water can
contaminate ground water. The significance of this depends upon the nature
of the species which might be leached out of the solids by the runoff/perco-
lation. Thus, in selecting solid waste management techniques, the major
evaluation criteria are whether the technique is applicable and economically
feasible, and whether the secondary waste streams are suitablly contained.
Based upon current techniques practiced in the synfuel and other indus-
tries, together with those being considered by proposed synfuel plants, the
bulk of the solid waste from K-T facilities will likely be disposed of on
land. Land-based disposal techniques are by far the most site-specific tech-
niques. The suitability of the site, as well as the design and operation
of the facility would depend on the site location, transportation costs,
hydrogeologic conditions, and many other factors. In short, a detailed
analysis of the specific site is an important element of the overall control
technique evaluation process.
Land disposal (e.g., landfill, surface impoundment, land treatment) will
be subject to regulations promulgated by EPA pursuant to the Resource
373
-------
Section 4
Solid Waste Control
Conservation and Recovery Act (RCRA), covering the generation, transport,
treatment, storage, and disposal of sol id wastes. Requirements concerning a
solid waste can vary significantly, depending upon whether the waste is
determined to be "hazardous" or "non-hazardous" as defined by RCRA regula-
tions. In this section, no attempt is made to judge whether the various
individual waste streams will be determined to be hazardous or not. Rather,
treatment and disposal techniques are presented which would cover the range
of possibilities, whether the waste is hazardous or non-hazardous.
Another technique potentially applicable to some solid waste streams -in
addition to land disposal - is incineration. If a waste which is determined
to be "hazardous" is proposed for incineration, the incinerator will have to
be designed and operated in accordance with RCRA requirements.
The available techniques that.may apply to the K-T solid waste streams
are identified and evaluated in this section. Since no specific site is
being considered, a general overview of these techniques is first presented.
This is followed by an evaluation of the applicable controls to each individ-
ual stream under each waste source type. The emphasis is on identifying
the applicability of the technique. For the reasons stated above and because
the characteristics of many of the solid waste streams are not known, it is
not possible to evaluate the optimum design and operation of these techniques
in the PCTM; optimum design/operation will vary with the site. It is assumed
that solid waste management facilities are captive, i.e., they only handle
waste from the K-T plant.
374
-------
Section 4
Sol id Waste Control
4.3.1 Solid Haste Control
Several control techniques are potentially applicable for management of
solid waste streams. These techniques are summarized in Table 4-46. As
shown, they can be broadly divided into three control categories according to
their functions. The three control functions are treatment, reuse/resource
recovery, and disposal. Treatment may involve specific chemical/physical pro-
cesses for preparing the waste to meet certain reuse/resource recovery speci-
fications or to stabilize the waste for disposal. Reuse/recovery is one form
of ultimate or final control for the waste. This approach is usually waste
specific, highly dependent on market availability and cost tradeoffs, and may
require specific treatment of the waste. Disposal is another form of ultimate
control for the waste. Most disposal techniques are land-based techniques and
thus are highly site-specific. The major site-specific factors that affect the
design, operation and cost of land-based techniques are summarized in Table
4-47. The following is a brief description of the individual techniques.
375
-------
TABLE 4-46. SUMMARY OF SOLID WASTE MANAGEMENT TECHNOLOGIES
oo
^J
01
Technology
Treatment
Fixation/
Incineration
Reuse/Resource
Recovery
Reuse,
Resource
Recovery
Landfill
Surface
Impoundment
Land
Treatment
Deep Well
I njection
Description
Chemicals are added to
stabil ize or sol idify
the wastes
Organic Wastes undergo
destruction to reduce
volume and toxicity
Waste is utilized- in a
manufacturing process*
construction application,
or is processed for reuse
in original application,
or valuable components
are recovered from waste
as byproduct for marketing.
Site is designed, con-
to totally contain waste.
or artificial liners,
leachate collection and
treatment systems , and
groundwater monitoring
system
Site is excavated or
diked to form pond to
natant is syphoned off
and treated or allowed
to evaporate
Waste is treated by
incorporation into the
land according to speci-
fic procedures
Wastes are pumped
through wells into
appropriate formations
generally several thou-
sand feet below the
surface
Operations
Considerations
Wide variety of wastes
can be processed, the
feasibil ity of sol idi -
fying a particular
waste may differ with
different process
Each type of organic
waste may require dif-
ferent operating char-
quired
User for waste must be
located, cessation of
reuse requires imme-
diate alteration of
management techniques,
thereby, necessitati ng
long terra contracts
Wide variety of wastes
visions must be made
Similar to landfill
Only 1 imited types
and mass of organic
wastes can be managed
Wastes frequently
must be treated
before injection
Rel labil ity or
Limitations
Limited commercial
experiences
Only organic wastes
can be processed
Markets for wastes are
1 imited and economic
viabil ity is heavily
influenced by distance
to market
Waste will be con-
tained subject to
adequate site opera-
arid absence of ex-
treme hydrogeologi -
cal changes or earth
movements
Simi 1 ar to 1 andfil 1
Heavily dependent
upon weather condi-
tions
Injection rates
frequently 1 imited
by receiving forma-
tion
Equipment
May include mix-
ing chamber,
pumps, metering
devices, storage
tanks, chemical
feed systems
Waste transpor-
tation machin-
ery , incinerator
Transportation
vehicl es
Earth moving
equipment, waste
handl i ng machin-
1 ifts and trucks
Machinery to move
waste to site.
lines
Waste moving
machinery, usual -
ly trucks, and
waste incorpor-
ation machinery,
usually bul Idozers
and discs
Pumps , i njection
wells
Waste
Generated
Solidified
waste
Ash; air
pol 1 ution con-
trol residues
None
Surface runoff
and 1 eachate
Supernatant
and leachate
Possibly sur-
face runoff
and 1 eachate
Residues from
the waste
treatment pro-
cess
General
Comments
Most processes
are appl icabl e
only to small
waste streams
Process is energy
intensive
In general, these are
the most environmentally
acceptable management
techniques.
Site location and
design dependent
upon hydrogeolo-
gi cal conditions,
be marie for site
care after cessa-
tion of operations
Waste may have to
be renoved when
if waste remains.
long term site
care and mainte-
nance program
must be esta-
blished
Site location
dependent on
soil condi tions ,
provisions must
be mac*e for long
term site care
Some states pro-
hibit deep wel 1
injection
-------
TABLE 4-47. SITE-SPECIFIC FACTORS TO BE CONSIDERED FOR LAND-BASED DISPOSAL
OPTIONS
Climatological
t Wind conditions (e.g., speed, directional flux, dilution factors, humidity,
temperature)
0 Precipitation (e.g., annual precipitation, storm intensity, snow contribu-
tions)
0 Evapotranspiration rate (e.g., season variations)
Geologic Factors
• Physiographic features (e.g., runoff coefficient, slope, drainage patterns,
erosional features)
• Surface and subsurface geology (e.g., outcrops, bedrock features, strike
and dip of the bedrock, rock composition)
t Soil types (e.g., CEC capacity, texture, permeability, stratification,
homogenous vs. heterogenous deposition, chemical composition, percent of
humic material)
• Seismic factors (e.g., ground shaking or rupture)
Hydrogeologic Factors
• Drainage patterns
• Stream flow (e.g., velocity, perennial vs. intermittent, effluent or
influent source)
• Surface waters (e.g., tidal effects, recharge vs. discharge points)
• Vadose zone (e.g., depth, moisture content, hysteresis patterns, storage
capacity)
0 Groundwater (e.g., depth, number of aquifers and relationships, confined or
artesian, nature of confining layer(s), capillary fringe characteristics)
0 Piezometric surface (e.g., streamline flux patterns due to seasonal or event
related phenomena, influence of recharge/discharge areas, streamline anomalies)
0 Water quality (e.g., background vs. undersite vs. downgradient, water uses -
consumptive, irrigation, recreation, point source contributors and their
respective hydrogeologic pathways)
0 Floodplain (100 year flood) (e.g., aerial flooding limits, degree of localized
streamline pattern reversal, erosional consequences)
0 Wetlands (e.g., recharge vs. discharge source, wetland/groundwater continuity
and pathway)
0 Recharge and discharge areas (e.g., proximity of disposal area, volume of
flow)
Land Use Factors
0 Historic significance 0 Demographic setting
0 Transportation corridor (access) 0 Geopolitical impact
0 Beneficial uses 0 Ultimate land use
0 Cost
377
-------
Section 4
Solid Waste Control
Treatment
4.3.1.1 Treatment
For the purpose of this manual, only two solid waste treatment techniques
are considered. These are fixation/encapsulation, and incineration. Other
techniques such as dewatering, neutralization, etc. are either considered as
part of the base plant operation, or part of the pollution control processes
in other media (air or water), depending on the origin of the waste.
Fixation/Encapsulation
Fixation and encapsulation are treatment processes which stabilize or
solidify waste constituents, or enclose the waste within other materials.
Fixation processes generally combine the concepts of solidification (the
alteration of the characteristics of a waste to attain desired structural
characteristics) and stabilization (the immobilization of waste constituents
by chemical reactions to form insoluble compounds or by entrapment within an
inert polymer or stable crystal lattice). Depending on the principal
chemical agents used, fixation processes can be categorized as cement-based,
lime-based, thermoplastic organic polymer-based, and glassification techni-
ques. Encapsulation processes involve enclosing the waste in a coating or
jacket of an inert, relatively impermeable material so that contact between
the waste and water is prevented. Regardless of the specific chemicals used,
typical fixation process operations involve mixing the chemical with the
waste in a reactor at specific temperature and for a specific time period.
The end product is the fixed waste. In the case of encapsulation, bulk
wastes are enclosed in a stabilizing shell or container rather than being
intimately mixed with a stabilizing agent.
For economic reasons, these techniques have only been applied to small
volume waste streams or streams which are prone to pozzolanic reactions.
Chemical requirements for fixing the latter type of streams are generally
low. FGD sludge is one example of this type of stream. Several proprietary,
cement- or lime-based fixation techniques have been used in fixing FGD sludges.
378
-------
Section 4
Solid Waste Control
Treatment
Typical unit costs are reported to be $10 to $17 per metric ton of sludge
fixed. It is recommended that, before implementing this technique to a
specific waste, detailed treatability studies with various chemical additives
be performed to (1) establish that the waste is treatable, (2) select the
optimal process, and (3) minimize the cost (95).
Incineration
Incineration is a controlled thermal decomposition process which reduces
the weight and volume of the waste by converting many component elements of
organic matter into gaseous forms. The extent of volume and weight reduction
is dependent upon the waste characteristics, the incineration process, and
the specific equipment used. Incineration is also a viable detoxification
process if the toxicity results from the structure of the organic material as
opposed to the properties of the elements it contains. The end products of
incineration include carbon dioxide, water, ash, and other inorganic com-
pounds. Incineration has been applied to various industrial wastes including
refinery wastes, sewage sludge, paper mill waste liquor, pharmeceutical
wastes, and organic chemical wastes. The common types of incinerators used
for solid waste disposal include rotary kiln, multiple hearth, and fluid bed
reactor. The annualized unit cost for a 61 GJ/hr capacity rotary kiln in-
cinerator is estimated to be $270/Mg,
379
-------
Section 4
Solid Waste Control
Reuse/Resource Recovery
4.3.1.2 Reuse/Resource Recovery
Reuse or resource recovery of waste streams is desirable from environ-
mental standpoint because of direct waste reduction and perhaps displacement
of other resource requirements. Potential adverse environmental impacts
associated with disposal of the waste are eliminated although other impacts
may arise as a result of the reuse/recovery process utilized. This control
approach is highly waste specific and is constrained by the availability of
markets or uses for the waste.
The economics of reuse/resource recovery are sensitive to site-specific
factors such as transportation costs and some general factors such as the
prices of the recovered/reusable materials and the cost of preparing the
waste for reuse/resource recovery. The feasibility of this control should be
thoroughly analyzed for each individual facility before implementation.
380
-------
Section 4
Solid Waste Control
Disposal
4.3.1.3 Disposal
The bulk of the solid wastes from a K-T facility are likely disposed
on land. Three potentially applicable techniques for these streams are
discussed in this section. These are landfill, surface impoundment, and
land treatment.
Landfills
Landfills have been widely used for the disposal of municipal refuse and
a range of industrial wastes. In landfilling, waste is brought to the dis-
posal site by truck or conveyor, spread in layers, and compacted with heavy
equipment. In most municipal landfills the waste is covered with a thin
layer of soil at the end of the working day. The process is repeated until
the desired depth is reached or the available area is filled. A final cover
of soil is then added. The finished site is either revegetated or prepared
for other end uses.
Landfill can be accomplished in unexcavated depressions (the area-fill
method) or in excavated sites (the trench-fill method). These can be natural
sites or man-made sites such as coal mines. There are two major concerns in
landfill design and operation. Runoff from landfill sites may contaminate
surface water, and percolation from sites, after passing through the waste
pile, may contaminate groundwater. Runoff/surface water contamination may
be prevented by grading the site and by containment of runoff. Diversion
channels should be incorporated into the initial design of the landfill and
constructed before the site begins accepting waste. This prevents surface
runoff water from entering the site and generating leachate.
Migration of leachate from the site can be controlled by lining the land-
fill with clay, concrete, asphalt, or plastic. Liners will often be required
if the solid waste is considered to be hazardous and may sometimes also be
desirable if the waste is nonhazardous. The choice of an appropriate liner
381
-------
Section 4
Solid Waste Control
Disposal
or liners will depend on site-specific climatologic, geologic, and hydrogeo-
logic factors, as well as on the compatability of the liner and the waste to
be contained and the relative cost of compatible liners. A leachate collec-
tion and treatment system may also be necessary. Such systems consist of
perforated pipes and sumps placed in a layer of permeable sand at the bottom
of the fill. After being pumped out of the landfill, the collected leachate
may be treated in the gasification facility's wastewater treatment system or
in a separate treatment system.
In the absence of any judgment concerning whether or not a given waste
might ultimately be determined to be hazardous, and in order to remain inde-
pendent of site-specific factors, two landfill designs are considered in the
PCTM. These two designs cover the range from the simplest set of conditions
(nonhazardous waste, favorable hydrogeologic and other site conditions that
preclude the need for liners) to the most complex (hazardous waste, unfavor-
able hydrogeologic and other site conditions which necessitate double liners).
The two landfill designs are presented in Figure 4-9. For the purposes of
this assessment the lined landfill design assumes an upper liner consisting
of 1 m of clay and a lower liner of 0.76 mm synthetic material. Both land-
fill designs assume the completed fill will be 30 m above the original land
surface with a slope of 3:1. Both landfills will have a final cover consist-
ing of 0.5 m sand and 0.3 m of clay. The most complex, hazardous case would
also include provisions for closure and post-closure care, monitoring, record-
keeping and other requirements.
The total capital investment and total annualized unit costs as a func-
tion of the site capacities for the two designs are presented in Figure 4-10.
To be consistent with cost estimates performed for the base plant and the
air and water media, no land cost is included.* The capital investment
*About 1,100,000 m2 of land is needed per 106 Mg/yr of waste generated.
Assuming a land cost of $5,000/10,000 m2 the capital investments for the
lined and unlined landfills, as presented in Figure 4-10, will be increased
by more than nine percent. Additional land may be required for road construc-
tion, buffer zone, buildings, etc., which will incur additional costs.
382
-------
a) LANDFILL-DOUBLE LINER
b) LANDFILL-NO LINER
CLAY LINER
DRAINAGE
LAYER
CO
GO
ORIGINAL
GROUND
LEVEL
ORIGINAL
GROUND
LEVEL
NATIVE SOIL
i
SYNTHETIC
LINER
LEACHATE COLLECTION
SYSTEM
CLAY LINER
UNSATURATED
ZONE
LEACHATE
DETECTION
SYSTEM
UNSATURATED
ZONE
GROUNDWATER
Figure 4-9. Landfill design
-------
OJ
CO
-p.
CO
o
a.
< 4
o
LEGEND
•DOUBLE LINED LANDFILL
UNLINED LANDFILL
A = GASIFIERSLAG
B = A + BOILER ASH
+ FGDSLUDGE
+ RAW WATER TREAT-
MENT SLUDGE
+ BIOSLUDGE
C= B +ASH FROM
INCINERATING
DEWATERED
DUST
2 3 4 5 10 20 30 40 50
TOTAL WASTE GENERATION RATE, 104 Mg/Yr
Figure 4-10. Capital investment and annualized unit cost for landfills
11
10
7 K
o
0
I-
Q
LU
N
100
O
-------
Section 4
Solid Waste Control
Disposal
developed includes site preparation cost (e.g., clearing and scrubbing, ground-
water monitoring and collection, liners), final cover and revegetation cost,
and landfill equipment cost. It was assumed no excavation is required. The
annualized cost includes labor, fuel, and amortized capital costs, but does not
include hauling cost and other costs such as administrative, closure/post-
closure, and liability costs. These other costs would depend on the classifi-
cation of wastes under RCRA. EPA has estimated that for a 50,000 Mg/yr
commercial hazardous waste landfill, administrative and other compliance
costs amount to $9/Mg. Hauling cost is a function of distance between the
plant and the disposal site. It is estimated that the unit cost for a round-
trip distance of 5 and 15 km are $2 and $4/Mg, respectively.
Surface Impoundments
Surface impoundments have been utilized widely by municipalities and
industries to process or dispose of waste liquids, sludges, and slurries.
Like landfill sites, the impoundments can be in natural depressions or in
excavated areas. Earthen dikes are usually constructed around the impoundment
area. Wastes are transported hydraulically to the impoundment. The wastes
deposit at the bottom of the impoundment, and the supernatant may be removed
and treated for discharge or recycle or allowed to evaporate.
When the surface impoundment has been filled with waste, the site may
be closed in one of two ways: the waste may be left in place and covered to
prevent erosion and the infiltration of precipitation, or it may be removed
from the impoundment site for further treatment or final disposal in a landfill.
If wastes are left in place, the site becomes a landfill (subject to any
requirements pertaining to a landfill), and a long-term site care and mainten-
ance program will need to be established. The cost per unit volume for sur-
face impoundments are expected to be similar to those of landfills with similar
depths. However, since surface impoundments generally are used for the dis-
posal of wet, not yet dewatered wastes, a larger area may be required per
385
-------
Section 4
Solid Waste Control
Disposal
unit mass of dry solids, resulting in higher disposal costs (i.e. some of the
disposal cost will be for "disposing" of water). Disposal cost could be re-
duced if the wet wastes were dewatered first, but there would be costs assoc-
iated with dewatering. Similarly, disposal costs may be reduced if appreci-
able natural dewatering occurs within the surface impoundment due to settling
and evaporation. This trade-off in cost is highly dependent upon waste char-
acteristics and site-specific factors.
Leachate migration from surface impoundments is controlled in much the
same way leachate is controlled from landfills. Diversion structures prevent
runoff from surrounding terrain from entering the site; liners of in-place or
compacted soils or synthetic materials retard leachate migration down into
the soil and groundwater. As in the case of landfills, the PCTM considers
both an unlined and a lined impoundment in order to cover the range of possi-
ble experience. The unlined impoundment represents the simplest set of con-
ditions (nonhazardous waste with favorable site conditions). The lined im-
poundment represents a much more complex situation (hazardous waste with un-
favorable site conditions). The two surface impoundments are presented in
Figure 4-11. For the purpose of this assessment, the lined impoundment design
assumes a liner consisting of 1.5 m of clay. Both designs assume the completed
site will be 10 m deep. When the impoundment is filled, it is capped with a
cover consisting of 0.3 m clay and 0.5 m sand to prevent infiltration of
precipitation.
The capital investment for a surface impoundment with an annual capacity
of 467,000 Mg/yr is estimated to be 4.3 and 12.5 million dollars, respective-
ly, for the unlined and lined impoundment design. These costs include site
preparation costs (site clearing and installation of the groundwater monitor-
ing system and liners where required), final cover, and revegetation costs.
It was assumed no excavation is required. If the whole impoundment has to
be excavated, the capital investment, which includes site preparation cost,
will be increased by 8 and 40 times, respectively, for the lined and non-lined
386
-------
a) SINGLE LINER
CO
CO
—I
LEAK DETECTION
SYSTEM
b) NO LINER
UNSATURATEDZONE
-GROUNDWATER:
Figure 4-11. Surface impoundment design
-------
Section 4
Solid Waste Control
Disposal
designs. To be consistent with cost estimates for the base plant and pollu-
tion control in other media, no land cost is included. Assuming a land cost
2
of $0.50/m , including this cost item will increase the capital investment
by 7% and 20% for the two designs.
The total annualized unit costs for the two impoundment designs are
estimated to be $2.1/Mg and $5.7/Mg. This includes labor and amortized
capital cost, but does not include pumping cost and other costs such as admin-
istrative, closure/post closure, and liability costs. The pumping (trans-
portation cost) would depend on the topography of the site and the distance
between the site and the plant. The other costs would depend on the classi-
fication of wastes under RCRA. EPA has estimated that for a 50,000 Mg per
year commercial hazardous waste landfill, administrative and other compliance
costs amount to $9/Mg.
Land Treatment
Land treatment refers to the use of land or soil as a medium to treat and
dispose of waste. Also known as landfarming, landspreading, and soil applica-
tion, land treatment has been practiced successfully for the treatment and
disposal of municipal wastewater treatment sludges and petroleum industry oily
wastes for many years. It relies on the ability of naturally occurring soil
microorganisms to decompose and utilize organic compounds under aerobic con-
ditions. The design and operation of land treatment systems would be affected
by whether or not the wastes were considered to be hazardous.
Wastes added to soil are subject to one or more of the following proces-
ses: (1) decomposition/degradation; (2) leaching; (3) volatilization; and
(4) incorporation into the soil matrix (e.g., through ion-exchange or adsorp-
tion). It is the degradation processes which treat the waste to reduce its
objectionable properties; these processes must be maximized during land treat-
ment, while the other processes must be minimized or eliminated. Applying
biodegradable wastes, maintaining proper (aerobic) conditions for microbial
388
-------
Section 4
Solid Waste Control
Disposal
action, and avoiding or pretreating wastes which are toxic to the microorgan-
isms will encourage degradation processes. Proper site selection and proper
site management will minimize leaching and subsequent contamination of ground-
water. If volatile wastes are to be land-treated, subsurface injection of
the wastes or immediate tilling after application will minimize air pollution.
Wastes with high concentrations of toxic substances such as arsenic, cad-
mium, lead, and mercury should not be land treated in sites where food chain
crops are grown, as they may be incorporated into the soil and taken up by .1
plants. Prior to land treating the biological oxidation sludge, long term
studies should be performed to confirm that the waste is degradable in the soil,
that there is no accumulation of non-degradable toxic substances in the soil,
and to establish the area required for the particular soil-waste combination
at hand.
Assuming biosludge is applied 10 times/yr, with an application rate of
o
0.015 Mg/m /application and a factor of 2 to account for land required for
roads, buffer zones, dikes, etc., it is estimated that 280,000 m2 of land is
required to land treat 21,000 Mg/yr of biosludge. The annualized unit cost
for this is estimated to be $7.6/Mg.
389
-------
Section 4
Inorganic Ash/Sludge
Gasifier Slag
4.3.2 Inorganic Ashes and Sludges
Table 4-48 summarizes the flow rates of the waste streams that consist
primarily of inorganic ashes or sludges. These streams are the largest vol-
ume solid waste streams from a K-T based indirect liquefaction plant. The
application of the available control techniques to these specific waste
streams from K-T plants are evaluated in this section.
TABLE 4-48. ESTIMATED FLOW RATES FOR THE INORGANIC ASH AND SLUDGE STREAMS
Stream Description Stream Flow, Kg/hr
Gasifier Slag (Stream 207) 10,371
Dewatered Gasifier Dust (Stream 209) 59,217
Boiler Bottom Ash (Stream 304) 745
Boiler Fly Ash (Stream 302) 2,980
FGD Sludge (Stream 424) 10,722
Raw Water Treatment Sludge (Stream 300) 432
4.3.2.1 Gasifier Slag (Stream 207)
The gasifier slag is a coarse pebble-sized material which is physically
stable and essentially chemically inert. As discussed in Section 3, this
stream is essentially coal ash with little or no carbon if the gasifier is
operated properly. Leachate may include low levels of trace elements (see
Table 3-7 for laboratory leachate characteristics). Tests of leachate for
organics have not been conducted, but organic levels would be expected to be
low. Potential may exist for very low level H~S evolution from the slag,
particularly under acidic conditions, as a result of sulfide present in the
interstitial water or due to the reaction of metallic sulfides in the slag.
However, due to the low sulfur content of the slag and interstitial water,
the potential for sulfide evolution appears to be low. The following tech-
niques are applicable to controlling this stream.
390
-------
Section 4
Inorganic Ash/Sludge
Gasifier Slag
Treatment
The techniques applicable to treatment of gasifier ash are fixation and
encapsulation. Treatment of gasifier slag may be appropriate if future data
indicate that significant concentrations of trace metals are in leachate from
the gasifier ash. Currently available leach data indicate the trace elements
in the leachate are low and should not be significantly different from con-
ventional coal boiler bottom ash.
The performance of treating this stream is dependent upon the specific
process (additive agent) used, and can only be established after thorough
treatability studies. The cost for treating this stream is also dependent
upon the process used. Assuming that the fixation processes which have been
applied to FGD sludge are applicable to treating this stream, the unit cost
will be about $10 to $17 per Mg (95-98).
Resource Recovery
Gasifier slag may be utilized in a number of commercial applications, just
as boiler bottom and fly ash from fossil-fueled power plants have been used,
The National Ash Association reported that 24.3 percent of the coal boiler ash
produced in 1977 was reused in commercial applications. Ash has been
used commercially as a partial replacement for cement in concrete, as fill
material for roads and other construction projects, and as blast grit and
roofing granules. The slag may need to be crushed and sized before it can
be used in such applications.
The major constraints on reuse of gasifier ash are those of the market
for the material. Market conditions will vary from site to site. Given the
fact that not all ashes from existing power plants are commercially utilized,
it may oe difficult to find markets where all or significant quantities of
the slag from the K-T facility can be reused. Users for the slag will likely
be limited to those who are located in the vicinity of the plant. The eco-
nomic viability of reuse decreases with increasing distance to market and
391
-------
Section 4
Inorganic Ash/Sludge
Gasifier Slag
hence increasing transportation costs. If market conditions change so that
commercial reuse ceases, the waste management techniques for the ash will
need to be altered. Long-term contracts with users may lessen the potential
for market interruptions.
Disposal
Gasifier slag can be disposed of in two ways: in landfills or in im-
poundments.
Landfill --
In landfill ing, gasifier slags are usually brought to the site by trucks,
spread on the surface of land or previously placed ashes, and compacted. As
the pile height increases, a working face with safety slope (assumed to be
3:1) is developed to ensure stability of the fill.
If the gasifier slag is determined to be nonhazardous, then in the most
favorable case (e.g., favorable site conditions), an unlined landfill might
be possible. On the other hand, if the wastes were considered to contain
hazardous components, a lined landfill would be necessary where hydrogeologic
or other site factors are unfavorable.
Based on current and proposed practices in the synfuel and other indus-
tries, this stream is likely to be co-disposed with some other solid waste
streams from the plant in one common landfill. Although more than one land-
fill/disposal facility (e.g., one landfill design for hazardous waste and
one landfill for nonhazardous waste) may be operated in a K-T facility, for
costing purposes in the PCTM, one landfill accepting the wastes from the K-T
plant is assumed. By considering the alternatives of all wastes being dis-
posed of in a nonhazardous waste landfill with no liner and in a hazardous
waste landfill with double liners, the range of landfill cost estimates in
the PCTM should bracket the costs that might be encountered in practice for
any split of the wastes between hazardous and nonhazardous categories.
392
-------
Section 4
Inorganic Ash/Sludge
Gasifier Slag
Assuming a landfill is designed to accept the gasifier slag, boiler
bottom and fly ash, FGD sludge, and raw water treatment sludges, the capacity
of the landfill would be 200,000 Mg/yr. As shown in Figure 4-10, the annual-
ized unit cost for the unlined and lined landfill design will be $3.2/Mg and
$5.9/Mg, respectively. The total annualized cost that would be attributable
to the gasifier slag stream alone would be $262,000/yr and $482,000/yr,
respectively, for the two landfill designs.
If the gasifier slag were disposed by itself in a separate, dedicated
landfill, annualized unit costs for this smaller landfill ($82,000 Mg/yr)
would be $4.5/Mg and $8.2/Mg for a nonlined and lined landfill, respecitvely.
The total annualized cost for this case would be $370,000/yr and $670,000/yr,
respectively.
Surface Impoundment --
One major difference between disposing the slag in a landfill and sur-
face impoundment is the means of transporting the slag to the disposal site.
Surface impoundment is usually used for storage or disposal of wet ashes
which are transported hydraulically to the impoundment in a fluid state.
For storage impoundments, the ashes are dredged periodically and disposed of
in landfills. For disposal impoundments, the ashes are left in place and are
covered to prevent erosion and infiltration of precipitation. The unit cost
for surface impoundment would be similar to the cost of landfill, assuming
no excavation is required. The total disposal cost may be higher because
surface impoundments generally are used for the disposal of wet, not yet de-
watered wastes; extra icapacity is needed for the water content in the waste.
4.3.2.2 Dewatered Gasifier Dust (Stream 209)
The gasifier dust is the largest volume solid waste stream from a K-T
methanol plant using Illinois No. 6 coal as feed. As presented in Section
3.3.1, this stream is estimated to be generated at a rate of 59.2 Mg/hr,
and is assumed to contain approximately 50% water, 30% ash, and 20% carbon.
393
-------
Section 4
Inorganic Ash/Sludge
Dewatered Gasifier Dust
The dust has a high affinity for water and is likely to have poor structural
stability at moisture contents exceeding 60% (99).
Treatment
Treatment with cement-based or other fixation techniques may be practiced
to improve the stability of the dust when deposited on land. Several proprie-
tary fixation techniques are commercially available. However the performance
of these processes when applied to this stream cannot be assessed without
thorough treatability studies. One potential problem in fixing this stream is
that it may be difficult to engage the dust in pozzolanic reactions with the
chemical additives. Although in principle the ash content of the stream is
capable of engaging in pozzolanic reactions with the chemical agents added,
the ash may not be available due to the coating effect of the carbon. The
net effect may be that much more chemical agent is needed, and considering
the size of the waste stream, associated costs may be prohibitive.
Resource Recovery
Due to its high carbon content, the dewatered gasifier dust can be used
as fuel in a boiler. However, this material presents two major problems with
respect to conventional pulverized coal-fired boilers. First, extensive dry-
ing, and thus special handling and drying equipment, is needed due to its high
water content. Second, the dust contains little or no volatiles and may re-
sult in flame stability problems unless a readily combustible supplemental
fuel is added or other countermeasures are taken. For these reasons, the use
of fluidized bed combustion (FBC) boiler is being considered for the North
Alabama Coal Gasification Consortium Project; this approach is used as an
example in this section to illustrate a resource recovery alternative.
An FBC boiler is comprised of a granular bed material which is suspended
or "fluidized" by a stream of air. The fuel is injected into this bed and
burned. Alkaline sorbents, typically limestone or dolomite, are also injected
into the bed to react with S02 formed during the combustion of high sulfur
394
-------
Section 4
Inorganic Ash/Sludge
Dewatered Gasifier Dust
fuels. The inert material of the fuel in part exits the top part of the bed
with the flue gas, and remainder is removed from the bottom of the bed with
the spent sorbent.
The secondary waste streams generated from the FBC boiler include a flue
gas stream and a spent sorbent stream. The estimated characteristics of these
two streams are presented in Table 4-49. The applicable control techniques
for the spent sorbent stream are discussed in Section 4.3.2.6 and the flue gas
control techniques are discussed in Section 4.1. Because of subsequent parti-
culate emission controls applied, essentially all of the ash in the flue gas
will be captured and will appear as a solid waste stream. The control for
this fly ash stream is discussed in Section 4.3.2.7.
The feasibility of burning the gasifier dust is highly dependent upon
the cost of displaced fuel (i.e., coal), cost and availability of land, and
other site-specific factors. Table 4-50 presents the estimated costs for
such an application assuming zero credit for the steam generated. The capi-
tal investment presented is factored from cost estimates developed for coal-
fired FBC boilers. As shown, burning the gasifier dust in the FBC boiler
would cost $8.4/Mg. However, it is estimated that about 6.3 x 10 Mg/yr
of steam will be generated and if a steam credit of $6.2/Mg were assumed,
the FBC boiler cost would break even, i.e., the steam credit would offset
the capitalized and operating cost. As discussed previously, a K-T facility
is expected to be self-sufficient with respect to steam and, as such, steam
from an FBC boiler would probably be used to generate electric power. How-
ever, the appropriate credit for power generated by dust combustion in a self-
sufficient K-T facility cannot be assessed without a detailed energy balance
around the plant and more detailed cost estimates for the boilers. Perform-
ing such analyses is outside the scope of this manual and, hence, no credit
has been taken for generated power.
395
-------
TABLE 4-49. CHARACTERISTICS OF FLUE GAS AND SPENT BED MEDIA FROM FBC BOILER
Spent Bed Media
Flue Gas (Stream 413)* (Stream 414)t
Component Kmol/hr Kg/hr Kg/hr
co2
N2
H20
so2
°2
Ar
Fly Ash
Total
CaS04
CaO
CaC03
Inert
Total
892
4,415
1,773
13
273
53
7,419
39,278
123,664
31,916
838
8,746
2,109
1,846
208,397
1,603
1,221
177
393
3,394
7T
NOx is approximately 150 ng/J which is equivalent to 56 kg/hr (as N02).
"''Characteristics of spent bed media are calculated based upon the model
presented on page 366 of Reference 87.
396
-------
Section 4
Inorganic Ash/Sludge
Dewatered Gasifier Dust
TABLE 4-50. ESTIMATED CAPITAL INVESTMENT AND TOTAL ANNUALIZED COST FOR
BURNING DEWATERED GASIFIER DUST IN FBC BOILER
Item ~~~~ FBC Boiler
Capital Investment, $106 15.6
Total Annualized Cost, $106 3.9
Total Annualized Unit Cost, $/Mg 8.4
% Base Plant Capital Investment 1.4
% Base Plant Total Annualized Cost 1.1
One other factor that affects the economics of this alternative is the
total amount of solid waste being generated. Burning the dust will reduce
the quantity of final waste by more than 60% (from 59.2 Mg/hr of dewatered
gasifier dust to 21.7 Mg/hr of dry FBC boiler ash and spent bed media). In
addition, power generated by FBC for in-plant use will result in less coal
being fed to the pulverized-coal-fired boiler (in self-sufficient facilities);
thus, less ash and FGD sludge will be generated from that part of the plant.
It is estimated that about 8.9 Mg/hr of coal would be displaced by the burning
of the gasifier dust. This amounts to a reduction of the coal-fired boiler
size and the associated ash and FGD sludge generated by 35% (see Table 3-30).
As presented in Table 4-48, a 35% reduction in boiler ash and FGD sludges
amounts to about 5 Mg/hr. Thus, in effect, the use of FBC boiler for burning
dewatered gasifier dust will reduce the total solid waste that is sent to
ultimate disposal by 32.5 Mg/hr.
Assuming the ash from the FBC boiler is co-disposed with other inorganic
streams (gasifier slag, boiler bottom and fly ash, FGD sludge, and raw water
treatment sludge), the landfill capacity would be 37,000 Mg/yr. The unit
annual ized cost for this size landfill would be $2.7/Mg and $5.4/Mg, depending
on whether the landfill is lined or not (see Figure 4-10). Assuming these
unit costs apply, a reduction of 32.5 Mg/hr of waste quantity would amount to
397
-------
Section 4
Inorganic Ash/Sludge
Dewatered Gasifier Dust
savings of 0.7 and 1.4 million dollars per year, respectively. The incre-
mental total annualized cost for disposing of the FBC boiler ash and spent
bed media (21.7 Mg/hr) would be $460,000/yr and $920,000/yr, respectively.
Disposal
The applicable disposal technique for controlling gasification dust is
surface impoundment. Landfill can also be used if the structural stability
of the dust is improved by fixation techniques. When disposing of the gasi-
fier dust in a surface impoundment, no mechanical dewatering pretreatment is
required. Quenched gasifier dust is transported to the site hydraulically.
The dust settles to the bottom of the site. Excess water is recycled and
reused as quench water. When the site is filled, it is capped with a cover
consisting of 0.3 m clay and 0.5 m sand to avoid infiltration. It is assumed
that the water content of the dust at the finished site is 50%. Mechanical
dewatering of the dust prior to impoundment can reduce the pond area needed
and may be employed where land is costly or of limited availability.
As discussed in Section 4.3.1.3, two surface impoundment designs are
considered in the PCTM to cover the range of possible experience. Table 4-51
summarizes the estimated costs for the two surface impoundment designs as
shown in Figure 4-11. The total capital investment for the non-lined and
lined designs are, respectively, 4.3 and 12.5 million dollars, and the total
annualized unit costs are $2.10 and $5.70/Mg.
398
-------
TABLE 4-51. SUMMARY OF CAPITAL INVESTMENT AND TOTAL ANNUALIZED
COST FOR DISPOSING OF GASIFIER DUST IN SURFACE
IMPOUNDMENT
Lined Surface Non-Lined
Item Impoundment Surface Impoundment
Capital Investment, 106$ 12.5 4.3
Total Annualized Cost, 105$ 2.5 0.9
Total Annualized Unit
Cost, $/Mg 2.1 5.7
% Base Plant Capital 1.1 0.4
Investment
% Base Plant Annualized 0.7 0.3
Cost
399
-------
Section 4
Inorganic Ash/Sludge
Boiler Ash
4.3.2.3 Boiler Bottom Ash (Stream 304)
The applicable control techniques for boiler bottom ash are similar to
those for the gasification slag. Where landfill is the technique selected,
the stream is likely co-disposed with other solid waste streams in one
common landfill.
As discussed in Section 4.3.2.1, annualized unit costs for landfilling
the solid waste streams in a common landfill are estimated to be $3.2/Mg and
$5.9/Mg, respectively, for a non-lined and lined landfill. Based on these
unit costs, the incremental annualized disposal cost attributable to this
stream would be $18,000/yr and $34,000/yr, respectively.
4.3.2.4 Boiler Fly Ash (Stream 302)
Applicable control techniques for this stream are similar to those dis-
cussed for the gasification slag. Available technology requires that gasi-
fier slag and boiler bottom ash be quenched before any subsequent handling
of disposal. Boiler fly ash, however, may be collected and handled dry (via
a dry ESP or baghouse) or wet (via a wet ESP or scrubber). The choice of
collection technology depends in part on site-specific disposal factors and
also on factors specific to coal type. Some fly ashes tend to undergo fixa-
tion reactions when wetted, much as Portland cement does. Recently, more
power plants are converting to dry collection systems for boiler fly ash.
When boiler fly ash is collected and handled entirely in the dry form, it
can be potentially recovered as a resource.
Where landfill is the disposal technique selected, this stream is likely
co-disposed with other solid waste streams in one common landfill. As dis-
cussed in Section 4.3.2.1, annualized unit costs for landfilling the solid
waste streams in a common landfill are estimated to be $3.2/Mr and $5.9/Mg,
respectively, for a non-lined and lined landfill. Based on these unit costs,
the incremental disposal cost attributable to this stream would be $74,000/
yr and $136,000/yr, respectively.
400
-------
Section 4
Inorganic Ash/Sludge
FGD Sludge
4.3.2.5 Boiler FGD Sludge (Stream 424)
Applicable techniques for controlling FGD sludge include fixation, sur-
face impoundment and landfills. These techniques have been widely used in
disposing of FGD sludges from existing coal-fired power plants.
Treatment - Fixation
FGD sludge typically contains 30 to 50 percent solids after thickening
or filtration. It is a poor landfill material in this form because it is
thixotropic. To rectify this problem, treatment by fixation may be practiced.
Several patented processes are available commercially for fixing FGD sludges.
One mixes the sludge with boiler fly ash and lime while another adds a pro-
prietary chemical (basically a cementitious agent) as the hardening material.
Typically these proprietary processes involve dewatering the sludge and com-
bining the sludge with proprietary additives which promote pozzolanic reac-
tions, resulting in a material less Teachable, less permeable, and struc-
turally more suitable for landfill. Proprietary methods which have been
successfully applied to fixing FGD sludges include Chemfix (addition of
Portland cement and sodium silicate), Calcilox (calcined blast furnace slag
and lime), IUCS - Poz - 0 - Tec (fly ash and lime under controlled tempera-
ture and moisture conditions), ICT (lime, betonite, and cement), and Research-
Cottrell (sludge dewatering prior to fly ash admixing).
Unit cost for these treatments ranges from $10 to $17 per Mg of sludge
fixed. Another treatment alternative practiced in many coal-fired power
plants is mixing the FGD sludge with -boiler bottom and fly ash before dis-
posal . For coals that generate ashes that are alkaline, mixing the ash with
the sludge will also initiate pozzolanic reactions.
401
-------
Section 4
Inorganic Ash/Sludge
FGD Sludge
Disposal
Disposal techniques applicable to FGD sludge include landfill and surface
impoundment. Because of its low solids content and structural instability,
FGD sludge may be treated by fixation or mixing with boiler fly ash prior to
disposal. The use of surface impoundments will reduce the liquid content of
the sludge, but the dried solids are readily soluble when exposed to moisture
after disposal, so proper surface impoundment closure is essential.
Assuming this stream is mixed with the gasifier slag, boiler bottom ash
and fly ash, and raw water treatment sludge before disposal in a common land-
fill, the unit disposal cost for a non-lined and a lined landfill is esti-
mated to be $3.2/Mg and $5.9/Mg, respectively (see Section 4.3.2.1). The
incremental annualized cost attributable to this stream would be $270,000/yr
and $500,000/yr, respectively.
402
-------
Section 4
Inorganic Ash/Sludge
Spent Sorbent/Fly Ash,
Raw Water Treatment Sludge
4.3.2.6 Spent Sorbent from FBC Boiler (Stream 414)
The characteristics of this material are similar to the solids in the
FGD sludge stream. Thus the applicable control techniques are similar to those
for the FGD sludge with the exception that since this is a dry material, it
can be landfilled directly without fixation. When landfill is the technique
selected this stream is likely combined with other solid waste streams and
disposed of in one common landfill.
4.3.2.7 Fly Ash from FBC Boiler (Stream 413)
The applicable control techniques for this stream are identical to those
for Boiler Fly Ash.
4.3.2.8 Raw Water Treatment Sludge (Stream 300)
The applicable treatment and disposal techniques for this stream would
be similar to those for the FGD sludge, except that the optimum fixation pro-
cess and hence the treatment cost may differ. Where landfill is the technique
selected, this stream is likely co-disposed with other solid waste streams in
one common landfill. Assuming unit costs of $3.2/Mg and $5.9/Mg respectively,
for a non-lined and lined landfill, disposing this stream will cost $11,000
and $20,000/yr.
403
-------
Section 4
Recovered By-Products
Elemental Sulfur/Coal Dust
4.3.3 Recovered By-Products
This source type includes elemental sulfur recovered from the bulk sul-
fur removal processes and collected dust from particulate control, if either
of these materials proves to be unsalable. The flow rate of these two
streams is estimated to be, respectively, 8.3 Mg/hr and 0.14 to 1.9 Mg/hr.
4.3.3.1 Recovered Elemental Sulfur (Streams 403 and 408)
Recovered elemental sulfur can be sold as by-product. However, the sul-
fur may be contaminated with carbonaceous impurities (from Claus plant) or
vanadates, thiosulfates and thiocyanate salts (from Beavon Stretford tail gas
treatment unit), making it non-marketable without further in-plant processing.
If the sulfur cannot be sold, it may be disposed of in landfills.
There is a potential for elemental sulfur to be oxidized in a landfill
environment, and such oxidation results in acid generation,. Acidic leachate
could solublize trace elements from other wastes in the landfill. Hence,
it may be desirable to co-dispose waste elemental sulfur with alkaline wastes
such as FGD sludges or spent FBC sorbent.
4•3•3•2 Collected Coal Dust from Particulate Control (Stream 400)
This secondary waste stream consists primarily of coal dust collected
throughout the coal preparation operations. This dust can be reused as feed
to the gasifier or boiler, or can be disposed of in landfills. When land-
filled, spraying the dust with wdier ;nay be required to reduce dust emission.
404
-------
Section 4
Organic Sludges
Biosludge
4.3.4 Organic Sludges
This waste category includes one stream, namely, the biosludge from the
biological treatment process (Stream 415). This is a secondary waste stream;
the flow rate of this stream has been estimated to be 2.65 Mg/hr, assuming
water is used in the cyanide wash operation in the base plant. Impact of
cyanide wash design on the wastewater characteristics, and thus the biosludge
characteristics have been presented in Sections 3.4 and 4.2.
Treatment
Although no data are available on the composition of this waste at pre-
sent, it is highly probable that some of the nonbiodegradable toxic organics
that might have been present in the raw process liquor (such as polycyclic or-
ganics and aromatic amines) will end up in the sludge through sorption. These
organics can be destroyed by incineration.
Incineration — Incineration of municipal and industrial biological oxi-
dation sludges has been practiced for many years. The application of this
treatment technique to this organic sludge could be expected to destroy
greater than 99% of most organics and reduce the quantity of waste that re-
quires ultimate disposal. Assuming the biosludge is 20% solids and 70% of
the solids are volatile, the total waste quantity will be reduced by 94%
through incineration.
Table 4-52 presents the estimated costs for incinerating the biosludge in
a rotary kiln incinerator (63 GJ/hr energy input) with energy recovery. The
heating value of the biosludge was assumed to be 23 MJ/kg. As shown in Table
4-52, the capital investment for incineration is estimated to be 14 to 16
million dollars; the total annualized cost is 5.8 million dollars per year,
or about $270/MG.
405
-------
Section 4
Organic Sludges
Biosludge
TABLE 4-52. ESTIMATED TREATMENT/DISPOSAL COST FOR
BIOSLUDGE
Item
Total Capital Investment, $10
Total Annual i zed Cost, $106
Annual ized Unit Cost, $/Mg
% Base Plant Capital
Incineration
14 to 16
5.8
270
1 2 to 1 4
Land Treatment
0.5
0.2
7.60
0.05
Investment
% Base Plant Annualized 1.6 0.06
Cost
Two secondary waste streams are generated by this process, namely, a flue
gas stream and a residue stream. It is not possible to estimate the charac-
teristics of the flue gas, but the cost for controlling this is included in
the incinerator cost estimates presented in Table 4-52. The incinerator is
assumed to be equipped with a scrubber for particulate control.
The flow rate of the residue stream is estimated to be 0.16 Mg/hr.
Assuming a 99.9% destruction of organics, the residue is expected to contain
about 0.33% organics and other inert materials. Most of the trace metals
originally present in the biosludge will accumulate in the residue. Appli-
cable treatment/disposal techniques include fixation/encapsulation and landfill
Disposal
Biological treatment sludges may be disposed of in landfills, surface
impoundment or by land treatment. Landfill and surface impoundment have been
discussed in the previous sections. The following is a brief description of
land treatment of biosludge.
406
-------
Section 4
Organic Sludges
Biosludge
Land Treatment —
In land treatment, biological treatment sludge may or may not require
dewatering prior to applying to the land. Depending on the physical state,
or the degree of dewatering performed, the sludges are transported to the
land treatment site either by truck or hydraulic means. The sludges on land
are spread with bulldozers, loaders, graders, or box spreaders. The site is
generally subdivided into several plots which are treated in sequence. After
waste application and evaporation of any associated water, the plot is plowed
periodically until the waste has been decomposed. Chemicals such as nitrogen,
phosphorus, and potassium may be added periodically as nutrients, and neutra-
lizing agents (e.g., lime) may be added to maintain the proper pH level (7 to
9).
The estimated costs for land treating the biosludges are summarized in
Table 4-52. The capital investment presented in Table 4-52 includes land
preparation costs ($0.52/m2), waste spreading equipment costs ($160,000), and
monitoring well costs ($25,000). The annualized costs include labor cost,
fuel cost, monitoring cost, maintenance cost, and amortized capital costs.
No land nor transportation cost was included in the cost presented. It is
2
estimated that about 32,000 m of land is required. Assuming a unit land
2
cost of $0.50/m , this would increase the total capital investment by more
than 25%. Depending on the distance, including the transportation cost may
more than double the annualized unit costs presented.
407
-------
Section 4
Spent Catalyst/Sulfur Guard
4.3.5 Spent Catalyst and Sulfur Guard
Nine types of catalysts may potentially be used in a K-T based indirect
liquefaction plant. These materials eventually become deactivated and require
decommissioning and disposal. Spent sulfur guard, which is not a catalyst,
is also included in this discussion because (1) this is also a small volume,
intermittant stream, and (2) applicable controls are similar. Table 4-53
summarizes the estimated spent catalyst generation rates. It should be
pointed out that although the flow rates are presented in Mg/yr, these streams
only occur interim'ttantly, about once every three to five years.
TABLE 4-53. ESTIMATED GENERATION RATES FOR SPENT CATALYSTS AND SULFUR GUARD
Catalyst/Guard Material Generation Rate, Mg/yr
Spent Shift Catalyst (Stream 217) 14 " 24
Spent Sulfur Guard (Stream 218) 80
Methanol Synthesis Catalyst (Stream 227) 60 -100
Spent Methanation Catalyst (Stream 238) 40
Spent Claus Catalyst (Stream 402) 10
Spent SCOT Catalyst (Stream 410) 3
Spent Beavon Catalyst (Stream 407) 5
Spent Mobil Synthesis Catalyst (Stream 232) 300
Spent F-T Synthesis Catalyst (Stream 222) 3500
Spent NO Reduction Catalyst (Stream 212) 16-27
X
408
-------
Section 4
Spent Catalysts/Sulfur Guard
Due to the proprietary nature of most catalysts, there is little informa-
tion available on the reuse and disposal techniques applicable to specific
catalysts. Because of this, spent catalyst reuse, treatment, and disposal are
discussed in general terms in the following sections, with only brief mention
of specific techniques and their applicability to individual catalysts.
Treatment
Spent catalysts may be chemically fixed or encapsulated before final
disposal to prevent leaching of undesirable substances. When fixing these
with cement-based techniques, the weight of the fixed material may be twice
its original weight; i.e., a 1:1 chemical/spent catalyst ratio may he needed.
As discussed before, the performance and cost for this alternative can only
be established after thorough treatability studies.
Resource Recovery and Reuse
Spent catalysts may be reused after reactivation by a contractor or the
original vendor. Also, the metal components of the catalyst may be recovered
for other uses. The economics of the required regeneration processes and the
market value of the metals will determine whether recovery and reuse are pos-
sible. In practice, return of the spent catalyst to the vendor for processing
will likely be the approach of choice in many cases.
Because of the current tight cobalt supply and the high demand for this
metal, it appears that the cobalt-based shift catalyst could be economically
recovered. Increasing cobalt prices have fostered interest by catalyst
manufacturers to develop improved methods to regenerate the catalyst, to re-
cover the metal, and to search for other catalysts (mainly nickel-based) which
can be used in place of the cobalt-based shift catalyst.
Regeneration of spent sulfur guard, Claus catalyst, Mobil M synthesis
catalyst, and Fischer-Tropsch synthesis catalyst is expected to be economi-
cally unattractive because of the low market values of the base materials of
409
-------
Section 4
Spent Catalysts/Sulfur Guard
these catalysts (zinc for sulfur guard, bauxite for Claus catalyst, zeolite
for Mobil M synthesis catalyst, and iron for Fischer-Tropsch synthesis cata-
lyst). Spent methanation catalyst (nickel-based) and Fischer-Tropsch synthe-
sis catalyst (iron oxide-based), although deactivated as far as catalyst
activity is concerned, still have a large capacity for adsorption of sulfur
compounds and can be used as sulfur guard bed material,
DisposaJ
Spent catalysts and sulfur guards may be chemically fixed or encapsulated
before final disposal to minimize leaching of toxic substances or they may be
disposed of once they are decommissioned. When disposed of, these materials
are likely to be placed in landfills (i.e., assuming that these wastes are
disposed of in a common landfill along with other plant solid waste). As
indicated in Table 4-53, the overall spent catalyst generation rate is largely
dependent upon the synthesis process incorporated in the plant. As discussed
before, the unit costs would be $3.2/Mg and $5.9/Mg, respectively, for a
non-lined and lined landfill; the incremental annual disposal cost for these
materials would range from $720 to $22,000, depending on the landfill design
and the synthesis process used.
410
-------
Section 5
Data Gaps/Limitations
SECTION 5
DATA GAPS AND LIMITATIONS
Because of the inherent gaps and limitations which exist in the data base
used to support this document, it is important for readers to understand the
extent to which the performance and cost estimates presented here are supported
by actual operating data, extrapolations from closely related applications, or
engineering calculations and/or judgements. The purpose of this section,
therefore, is to convey to the reader a sense of the applicability and
completeness of the data base. This information should contribute to a
better understanding of how this document should be used by indicating the
confidence which can be placed in the uncontrolled discharge rates and the
effectiveness of specific controls.
Since the early 1970s the EPA has sponsored a significant environmental
assessment program addressing synthetic fuels from coal technologies. This
work has involved a combination of theoretical studies and plant data
acquisition programs. These efforts have contributed both data and background
knowledge used in the development of this manual. The major data acquisition
programs sponsored or cosponsored by the EPA which have provided background
data used in the development of this PCTM are listed in Table 5-1, at the
end of this section. As indicated, the data encompass specific research pro-
jects, pilot-level sampling and analysis projects, and source sampling of
foreign and domestic commercial production facilities.
Waste streams which are unique to K-T based indirect liquefaction facili-
ties have been emphasized in this section. These streams differ from waste
streams generated in other industries because of their composition and/or the
approaches applicable to their control. Waste streams which do not differ in
composition from wastes generated in other industries and which do not require
special consideration with respect to control approaches (e.g., boiler flue
gas, coal pile runoff, raw water treatment sludges, and boiler ash) are not
411
-------
Section 5
Data Gaps/Limitations
considered in this section. These non-unique waste streams have not been
considered because any significant data limitations or gaps related to their
characterization have been a concern in other industries, and programs to fill
identified data gaps are already underway or are currently being planned.
Key data sources and the bases for characterization estimates, data gaps and
limitations, and research needs relating to unique waste streams are summar-
ized by waste medium in Tables 5-2 through 5-4 at the end of this section.
Similar information relating to applicable pollution control technologies and
associated secondary waste streams are presented in these tables following
each waste stream or group of waste streams combined for common treatment.
In general, estimates of characteristics of the uncontrolled unique
waste streams were based upon data from commercial scale K-T gasification
facilities (e.g., AECI Limited, Modderfontein, South Africa and the Nitrogeneous
Ferterlizer Inudstry, S.A., Ptolemais, Greece). Although these facilities pro-
duce hydrogen for ammonia manufacture, many of the individual process oper-
ations upstream of product synthesis are similar to those proposed for in-
direct liquefaction facilities. Data from these operations have in some in-
stances been used directly or have been extrapolated, consistent with proposed
designs. Waste characterization data for F-T, methanol, and Mobil M-gasoline
synthesis processes have been based upon published designs for commercial scale
facilities.
Since none of the existing K-T gasification facilities employ the range
of pollution controls which are likely to be utilized in the U.S., little
direct operating experience is available to accurately predict the performance
or costs of applicable controls. For certain control systems (e.g., Glaus
bulk sulfur recovery and Wellman-Lord tail gas treatment) the existing data
base from related applications is sufficient to indicate gross pollutant re-
moval efficiencies and associated costs reasonably well. However, for control
systems for which performance and cost are highly sensitive to individual com-
ponents present in the waste stream (e.g., activated sludge and chemical oxi-
412
-------
Section 5
Data Gaps/Limitations
dation), only limited data or data from somewhat different applications are
generally available. The limitations in the data of this type are twofold.
First, the characteristics of the treated waste streams in related applica-
tions of the subject controls are often known only in terms of major consti-
tuents, gross parameters, or classes of substances. Little may be known about
specific organics, trace elements, or general toxic properties. Secondly, the
performance characteristics of many controls are uncertain for some of the
specific waste streams addressed in this PCTM even for the major constituent
and gross properties. Key data sources and bases for performance and cost
estimates for example controls discussed in conjunction with unique waste
streams in Section 4 are summarized in Tables 5-2 through 5-4.
A data gap or limitation which exists for essentially all pollution con-
trol technologies relates to reliability. Because most of the potentially
applicable pollution control technologies have not been employed in coal
gasification/indirect liquefaction facilities, few directly related reliability
data are available. Further, the overall characteristics and variability of
waste streams in coal conversion facilities are often sufficiently different
from those encountered in other industries that reliability data accumulated
in other industries may not be applicable to coal conversion processes. These
are particularly significant considerations with respect to wastewater treat-
ment technologies and, to a lesser extent, hold for gaseous and solid waste
control technologies also. It should be noted, however, that such data gaps
cannot be addressed for specific controls prior to the application of these
controls to coal conversion waste streams.
With regard to the products and by-products produced in K-T based lique-
faction facilities, most of the physical and chemical properties reported are
estimates relating to conceptual designs of commercial scale facilities.
Despite the fact that the Fischer-Tropsch (F-T) process is applied commercially
in South Africa, detailed chemical analysis data are not publicly available
for F-T products. Detailed chemical analyses are not available for Mobil M-
413
-------
Section 5
Data Gaps/Limitations
gasoline or coal-derived methanol, which are not currently produced on a commer-
cial scale. Also, there are no chemical characterization data available for
by-product sulfur from indirect liquefaction facilities. The data sources and
bases for characterization estimates, data gaps and limitations, and research
needs relating to products and by-products produced in K-T indirect liquefaction
facilities are summarized in Table 5-5, at the end of this section.
414
-------
TABLE 5-1. COMPLETED AND ONGOING DATA ACQUISITION PROGRAMS AT COAL GASIFICATION
FACILITIES SPONSORED OR CO-SPONSORED BY THE EPA
Facility
Information Classification
Coal Used
Products
Medium/High Btu Gasification
and Indirect Liquefaction
Facilities (Foreign)
• Lurgi Gasification
- Kosovo, Yugoslavia
- SASOL, S.A.
- Westfield, Scotland
t Koppers-Totzek Gasification
- Modderfontein, S.A.
- Ptolemais, Greece
- Kutahya, Turkey
• Winkler Gasification
- Kutakya, Turkey
* Texaco Gasification
- Federal Republic of Germany
Low-Btu Gasification Facilities
Data acquisition
Plant visit and discussions
Plant visit and discussions
Data acquisition
Data acquisition (TVA & EPA)
Plant visit and discussions
Plant visit and discussions
Data acquisition (EPRI, TVA
& EPA)
Lignite
Low rank bituminous
Various
High volatile "B"
bituminous
111. No. 6 bituminous
Lignite
Lignite
111. No. 6 bituminous
Medium Btu gas
Various via indirect
1iquefaction
Test center
Ammonia, methanol
Ammonia
Ammonia
Ammonia
Test center
* Wellman GalusHa
- Site No. 1
- Site No. 2
• Chapman/Wilputte
• Riley
• Stoic (Foster Wheeler)
Control Research Facilities
• Raw/Acid Gas Cleanup (Fluidized
Bed Gasifier)
t Wastewater Treatatnlity Studies
» Pollutant Identification (Bench
Scale Gasifier)
• Ash Leaching Evaluations
Other Domestic Facilities
• Texaco Gasification
- Ammonia from coal plant, TVA
• Rectisol Acid Gas Cleanup
Data acquisition
Data acquisition
Data acquisition
Data acquisition
Data acquisition (DOE & EPA)
North Carolina State Univ.
Univ. of North Carolina
Research Triangle Institute
University of Illinois
Data acquisition (TVA & EPA)
Texaco, Wilmington, CA
Anthracite
Lignite
Low sulfur bituminous
Ligm te
Western bituminous
Various
Various
Various
Various
111. No. 6 bitutrnnous
(in shakedown)
Oil fired partial
oxidation
Fuel gas •
Test center
Fuel gas
Test
Fuel gas
Test center
Test center
Test center
Test center
Ammonia
Process hydrogen
-------
TABLE 5-2. DATA GAPS AND RESEARCH NEEDS - GASEOUS MEDIUM
Key for Technology Status and Data Sources
Technology Status
Data Source/Location
A. Commercial application 1n a K-T gasification facility
B. Bench scale or pilot testing
C. Technology transfer from another industry - similar
but not identical streams
D. Conceptual
1. AECI Limited, Modderfonteln, South Africa
2. Nitrogenous Fertilizer Industry, S.A., Ptolemais, Greece
3. Technology transfer from related industries
3A. Petroleum refining/petrochemical production
3B. Coke production
3C. Electric power generation
3D. Natural gas processing
3E. Municipal waste treatment
4. Conceptual or proposed design/engineering studies
5. Vendor supplied information
Data Source and Basis
Data Gaps/Limitations
Research Needs
en Uncontrolled Primary Waste Streams
H;S-Rich Offgas (Stream 216)
Offgas composition is based upon selective Rectisol
performance data from commercial K-T coal gasifica-
tion at AECI Limited, Modderfonteln, South Africa
and from several commercial oil partial oxidation
units. H?S to COS ratios in raw gas are also sub-
stantiated by commercial scale gasification tests
with an Illinois No. 6 coal at the Nitrogenous
Fertilizers Industry, S.A., Ptolemais, Greece.
Status:
Data Sources:
PCTM References.
A, C
1, 2, 3A, 5
Sections 3.3.6, 4.1.1.1,
4.1.1.5, and Appendix D.
The selectivity of acid gas removal is dependent
upon several design factors; however, available
data indicate that selective Rectisol can eco-
nomically produce an HoS-nch offgas containing
at least 25% total sulfur for a wide range of
feed coal sulfur contents. Relative quantities
of sulfur species in raw gas, and therefore in
the H2S-rich offgas, may differ somewhat for
different rank coals.
Verification of sulfur and minor constituent
characterization data in U.S. facilities is
desirable.
Sour Gas from Cyanide wash Flasn (Stream 214)
Flash gas compositions have been estimated from
gas solubility data.
Status: D
Data Sources: 4
PCTM References: Sections 3.3.4, 4.1.1.2,
and 4.1.1.5
No waste gas generation or characterization data
are available.
Generation rate and characterization data are
needed to define applicable control alternatives.
(Continued)
-------
TABLE 5-2. (Continued)
Data Source and Basis
Data Gaps/Limitations
Research Needs
Control Techniques
Claus Bulk Sulfur Removal
Performance and cost estimates are based upon data
from applications In coke plants, oil refineries,
and gats processing plants. Although no data are
available for coal gasification applications, data
from other applications cover the range of all
constituents encountered in K-T coal gasification.
Status: C
Data Sources: 3A, 3B, 3D, 5
PCTM References: Sections 4.1.1, 4.1.1.1,
and 4.1.1.5
Secondary Waste Streams
Spent catalyst is the principal secondary waste
stream from Claus.
Beavon/Stretford Tail Gas Treatment
Performance and cost estimates are based upon
data from treatment of Claus plant tail gases
with low CO^ feed composition from petroleum
refinery applications.
Status: C
Data Sources: 3A
PCTM References: Sections 4.1.1, 4.1.1.1.
and 4.1.1.5
Secondary Waste Streams
t Sour Condensate
Stretford Solution Purge
• Stretford Oxidizer Vent Gas
Performance, cost, and reliability data in coal
gasification applications are not available.
Spent catalyst generation rates and characteristics
are not accurately known.
Uncertainties relate to performance of the cata-
lytic hydrogenation section and the operability
of the Stretford sulfur recovery section in high
C02 applications. Process costs as a function
of both volumetric flow and sulfur loadings are
not accurately known.
Condensate characteristics, particularly with
respect to S=, CN", and NHJ are not available.
Limited characterization data are available.
Purge rates in high CO? applications are unknown.
Reductive incineration of solution purge has not
been demonstrated at commercial scale.
Characterization data are not available for oxi-
dizer vent gas. However, in K-T coal gasifica-
tion applications there are no gaseous components
of environmental concern having potential for
being present in the vent gas.
Performance, cost, and reliability data should be
obtained for K-T coal gasification applications.
Generation rate and characterization data are needed
to define applicable control alternatives. Such data
may be obtainable from Claus plant operators 1n related
industries.
The operability, performance, and cost of Beavon/
Stretford tail gas treatment in high C0.2 applications
should be determined.
Condensate characteristics should be determined to
define pollution control alternatives. Appropriate
data may be obtainable from licensors or operators
of refinery units.
Purge generation rates and characteristics should be
determined in coal gasification facilities when the
technology is applied commercially. Characterization,
performance, and cost data for reductive incineration
should be obtained when the process is applied
commercially.
Oxidizer vent gas should be characterized in K-T coal
gasification facilities when the technology is applied
commercially.
(Continued)
-------
TABLE 5-2. (Continued)
Data Source and Basis
Data Gaps/Limitations
Research Needs
CO
Secondary Haste Streams (Continued)
I Spent Beavon Catalyst
SCOT Tail Gas Treatment
Performance and cost estimates are based upon
data from treatment of Claus plant tail gases
with low CO^ feed compositions from petroleum
refinery applications.
Status: C
Data Sources: 3A, 5
PCTM References: Sections 4.1.1, 4.1.1.1,
and 4.1.1.5
Secondary Haste Streams
• Sour Condensate
t Spent SCOT Catalyst
Mellman-Lord (M-L) Tall Gas Treatment
Performance and cost estimates are based upon
data fron refinery and power plant applications.
Status: C, D
Data Sources: 3A, 3C. 4. 5
PCTH References: Sections 4.1.1. 4.1.1.1,
and 4.1.1.5
Secondary Haste Streams
o Sour Condensate
o Thlosulfate/Sulfate By-Product Purge
Spent catalyst generation rates and characteristics
are not accurately known.
Uncertainties relate to performance and cost in
high COj applications. SCOT units are used in
high CO? applications in natural gas processing
although associated performance and cost data
are not currently available.
Condensate characteristics, particularly with
respect to S=, CN~, and NH^ are not available.
Spent catalyst generation rates and characteristics
are not accurately known.
Since W-L feed gas is incinerated, and the
absorption solution is not affected by CO? con-
centration, data from existing units should be
directly transferable to K-T coal gasification
applications. Effects of volumetric flow rate
and sulfur loading on cost are uncertain.
Limited characterization data are available
Limited generation rate and characterization
data are available
Characterization data are needed to define applicable
control/resource recovery alternatives. Appropriate
data may be obtainable through licensors or operators
of refinery units.
SCOT performance in high CO? applications should be
determined. Units in gas plant applications are best
suited for demonstrating applicability to K-T coal
gasification facilities.
Condensate characteristics should be determined to
define pollution control alternatives. Appropriate
data may be obtainable through licensors or operators
of refinery units.
Characterization data are needed to define applicable
control/resource recovery alternatives. Appropriate
data may be obtainable through licensors or operators
of refinery units.
None
Additional characterization data are
desirable
Additional generation rate and characteriza-
tion data are desirable.
(Continued)
-------
TABLE 5-2. (Continued)
Data Source and Basis
Data Gaps/Limitations
Research Needs
Thermal Incineration
Performance and cost estimates are based upon
data from waste gas and liquid waste Incin-
eration applications.
Status: C
Data Sources: 3A, 3B, 3C, 30
PCTM References: Sections 4.1.1, 4.1.1.1,
and 4.1.1.5
Uncertainties relate primarily to the operabillty
of and emissions from Incinerators with feed gases
having high C02 levels and low heating valves.
Applicable data on operabllity of and emissions from
incinerators may be obtainable from operators of
refinery units or gas plants.
i-D
Uncontrolled Primary Waste Stream
C0?-Rich Offgas (Stream 219)
Offgas total sulfur and CO concentrations are based
upon selective Rectisol performance data from com-
mercial K-T coal gasification at AEC1 Limited,
Modderfontein, South Africa and from several com-
mercial oil partial oxidation units. Design modi-
fications discussed in conjuntion with control of
CO emissions are based on licensor inputs.
Status:
Data Resources:
PCTM References:
A, C
1, 3A, 5
Sections 3.3.6 and 4.1.1.2
Control Technology
Catalytic Incineration
Performance and cost estimates are based upon
data from industrial waste gas incineration
applications.
Baseline offgas characterization data, and spe-
cifics of CO moderating Rectisol design modifi-
cations and associated performance in indirect
coal liquefaction facilities are not available.
Available data on performance and cost are
1imited.
Offgas characteristics in indirect coal liquefaction
facilities should be determined. It appears likely
that two streams would require characterization:
(1) a COj-rich vent gas, and (2) a CO-rich waste gas.
Performance and cost data for waste streams similar
to those produced in K-T coal gasification facilities
should be obtained.
Status:
Data Sources:
PCTM References:
C
3A
Sections 4.1.1 and 4.1.1.3
Uncontrolled Primary Waste St,ream
Fugitive Organic Emissions from
Process Equipment (Stream 241)
Emission estimates for fugutive organic emissions
from leaking process equipment were based on
emission factors for these components for con-
ventional petroleum refining process equipment.
Component counts were based on plot plans for
potential K-T based synthetic fuel plants.
Status:
Data Sources:
PCTM References.
C
3A
Sections 3.7.1 and 4.1.5
Characterization data are not available for syn-
thetic liquids or their vapors.
Exact composition of vapors for synthetic fuel plant
process equipment may differ from petroleum refinery
process equipment, however, the control technology
that is applicable is not likely to be affected by
these characteristics. Since these emissions directly
enter the atmosphere, additional information on spe-
cific constituents of the vapor may be desirable.
(Continued)
-------
TABLE 5-2. (Continued)
Data Source and Basis
Data Gaps/Limitations
Research Needs
Control Technologies
Leak Detection and Repair Methods
Leak detection and repair methods have been
successfully used to control fugitive organic
emissions from petroleum refining equipment.
This technique was assumed to be applicable
to indirect liquefaction process equipment.
Emission estimates were based upon emission
factors developed from test data for
petroleum refining process equipment
The performance of leak detection and repair
methods are not expected to be different for
K-T based synthetic fuel plants.
Research needs relate to characterization of the
emissions and not to the controls.
Status:
Data Sources:
PCTM References:
C
3A
Sections 4.1.5 and 4.1.5.2
ro
O
Equipment Specification
Replacement of leaking equipment with leakless
components has been successfully used for the
control of fugitive organic emissions from
petroleum refining equipment. This technique
was assumed to be applicable to indirect
liquefaction process equipment. Emission
estimates were based upon emission factors
developed from test data for petroleum
refining process equipment.
The performance of equipment specification is not
expected to be different for K-T based synthetic
fuel plants.
Research needs relate to characterization of the
emissions and not to the controls.
Status.
Data Sources:
PCTM References:
C
3A
Sections 4.1.5 and 4.1.5.2
Uncontolled Primary Waste Stream
Storage Emissions (Streams 308 to 313)
Emission estimates were based upon emission factors
developed from test data for conventional petroleum
liquid storage. The amount and type of liquids
stored were determined from the synthesis process
used Liquids with vapor pressure greater than
10 kPa were assumed to be stored in floating roof
tanks. Low vapor pressure liquids were assumed
to be stored in fixed roof tanks.
Characterization data are not available for
synthetic liquids or their vapors.
Exact composition of vapors from synthetic fuel storage
may differ from those of conventional petroleum liquids;
however, the control technology that is applicable is
not likely to be affected by these characteristics.
Since these emissions directly enter the atmosphere,
additional information on specific constituents of the
vapor insy be desirable.
Status:
Data Sources:
PCTM References:
C
3A
Sections 3.6.5 and 4.1.5
(Continued)
-------
TABLE 5-2. (Continued)
Data Source and Basis
Data Gaps/Limitations
Research Needs
Control Technology
Secondary Seals on Floating Roof Tanks
Since secondary seals are used to control
evaporative emissions from- floating roof tanks
storing petroleum liquids, they were assumed to
be applicable to storage tanks containing syn-
thetic liquids. Emission estimates were based
upon emission factors developed from test data
for petroleum liquids storage.
Status:
Data Resources:
PCTM References:
C
3A
Section 4.1.5
The performance of secondary seals is not expected
to be different for synthetic liquid storage tanks.
Research needs relate to characterization of the
emissions and not to the controls.
Internal Floater* or fixed Roof Tanks
Since Internal floaters are used to control
evaporative emissions from fixed roof tanks
storing petroleum liquids, they were assumed
to be applicable to storage tanks containing
synthetic liquids. Emission estimates were
based upon emission factors developed from
test data for petroleum liquids storage.
Status: C
Data Sources: 3A
PCTM References: Section 4.1.5
The performance of internal floaters is not
expected to be different for synthetic liquid
storage tanks.
Research needs relate to characterization of the
emissions and not to the controls.
-------
TABLE 5-3. DATA GAPS AND RESEARCH NEEDS - AQUEOUS MEDIUM
Technology Status
A. Commercial application In a K-T gasification facility
B. Bench or pilot scale testing
C. Technology transfer from another Industry - similar
but not identical streams
D. Conceptual
Key For Technology Status and Data Sources
Data Source/Location
1. AECI Limited, Modderfontein. South Africa
2. Nitrogenous Fertilizer Industry, S.A., Ptolemals, Greece
3. Technology transfer from related industries
3A. Petroleum refining/petrochemical production
3B. Coke production
3C. Electric power generation
3D. Natural gas processing
3E. Municipal waste treatment
4. Conceptual or proposed design/engineering studies
5. Vendor supplied information
Data Source and Basis
Data Gaps/Limitations
Research Needs
ro
ro
Uncontrolled Primary Waste Streams
Cooling and Dust Removal Slowdown (Stream 210)
Slowdown composition is based upon data from com-
mercial scale gasification tests with an Illinois
No. 6 coal at the Nitrogenous Fertilizers Industry,
S.A. Ptolemals, Greece. Halide concentrations
were adjusted consistent with coal Cl" and F"
contents.
Status:
Data Sources:
PCTM References:
A, D
2, 4
Sections 3.3.1 and 4.2.3.3
Available characterization data are limited
to gasification of one U.S. bituminous coal
and several foreign sub-bituminous coals and
lignites. However, the quantity and quality
of blowdown are highly dependent upon coal
characteristics and design specifics. Data
from gasification of foreign coals reflects
designs which are not being proposed for use
in the U.S.
The existing data base adequately defines ranges of
the primary parameters of concern (eg CN", SCN", NH|
and TOS). However, these data should be verified 1n
U.S. facilities.
Primary Compression and Cooling Condensate (Stream 211)
Condensate flow rate and NH^, Cl", and F" concentra-
tions have been estimated oased upon wasner cooler
performance during commercial scale gasification
tests with an Illinois No. 6 coal at the Nitrogenous
Fertilizer Industry, S.A., Ptolemais, Greece. Other
parameters are based upon condensate characterization
data from commercial K-T gasification at AECI Limited
which gasifies South African sub-bituminous coal.
Condensate characterization data are avail-
able only from AECI Limited, and reflect
one coal and one washer cooler design.
Further characterization of this stream may be of
limited value due to its low flow rate relative to
similar process streams (e.g., cooling and dust
removal blowdown).
Status:
Data Sources:
PCTM References:
A, D
1, 2, 4
Sections 3.3.2 and 4.2.3.2
(Continued)
-------
TABLE 5-3. (Continued)
Data Source and Basis
Data Gaps/Limitations
Research Needs
Cyanide Mash Water (Stream 215)
The flow rate of cyanide wash water 1s based upon
nominal flow rate data from AECI Limited at
Modderfonteln, South Africa. Wash water composi-
tion has been estimated from gas solubility data.
No characterization data are available.
Characterization with respect to CN". SCN" and S 1s
desirable for defining control processes and costs.
Status:
Data Sources:
PCTM References:
A, D
1. 4
Sections 3.3.4 and 4.2.3.1
Synthesis Mastewaters - Mobil M (Stream 233),
F-T (Stream 223), and Methanol Synthesis (Stream 236)
Condensate
Characterization and flow estimates are from con-
ceptual designs and from data on product/compound
production rates. Fischer-Tropsch estimates
reflect commercial scale operating data, while
Mobil M-gasoline estimates reflect pilot scale
data. Methanol estimates are from engineering/
cost studies rather than direct test data.
Uncertainties relate to both the extent
to which by-product organics would be
reclaimed within the upgrading operation
and the exact species which are present
in the waste. The biodegradabilities
of these wastes are also not established.
Actual characterization data may be obtained from
SASOL (F-T) and from existing methanol plants. Mobil M
data will have to await the construction of the first
commercial plant.
Status:
Data Sources:
PCTM References:
B. D
4
Sections 3.4 and 4.2.2
Control Technologies
Filtration
Removal of suspended solids is based on perfor-
mances realized in parallel applications, par-
ticularly the petroleum refining industry.
Actual performance will depend on the character-
istics of the filter media and the characteristics
of the suspended solids, including particle size
distribution and tendency to agglomerate. Costs
are based on vendor quotes.
Performance has not been determined for the K-T
wastewaters.
Existing experience from application of filtration to
parallel industries is sufficient to estimate perfor-
mance to within reasonable limits of accuracy. More
refined'estimates would have little bearing on the
applicability of this control technology.
Status:
Data Sources:
PCTM References:
3A, 3E, 5
Sections 4.2.1.1, 4.2.2.1,
4.2.3.1, 4.2.3.2 and 4.2.4
(Continued)
-------
TABLE 5-3. (Continued)
Data Source and Basis
Data Gaps/Limitations
Research Needs
Polysulfide Addition
Conversion of cyanide to thiocyanate is based on
the results of EPA-sponsored preliminary kinetic
studies. There exists no known precedent for
effecting the conversion as the basis of a water
pollution control process, although polysulfide
is commonly added at key process points in petro-
leum refineries to control cyanide-induced corro-
sion. Capital costs are based on those require-
ments for chemical addition processes such as
chemical oxidation and disinfection. Operating
costs are dominated by chemical requirements,
estimated from both stoichiometric and kinetic
characteristics of the polysulfide reaction.
Status:
Data Sources:
PCTM References:
B, C, D
3A, 4
Section 4.2.1.4, 4.2.3.1
and 4.2.4
Activated Sludge - Removal of Organics
Removal of dissolved organics are expected to
exceed that typically realized in parallel appli-
cations since only simple, highly biodegradable
organics are involved. Actual performance will
be specific to the exact characteristics of the
wastewater. Costs are based on values reported
in the literature, appropriately scaled on the
basis of system loading.
Status:
Date Sources:
PCTM References:
B, C, D
3A, 3B, 3E, 4
Section 4.2.1.5, 4.2.2.1,
and 4.2.4
(1) Optimal conditions for the polysulfide (1)
reaction are not known, particularly the pH
dependence and the influence of chemical
species other than cyanide and thiocyanate.
(2) Residual polysulfide is expected to precipitate
at near neutral pH, posing possible problems for
downstream control processes. Characteristics
of this precipitate including its settleability
and filterability are not known. (2)
(3) The feasibility of identifiable methods of
adding polysulfide to the wastewater is not
known. In particular, it may prove feasible
to add polysulfide directly to the cyanide
wash on the gas cooling/dust removal circuit.
(3)
(1) Performance has not been determined for the
K-T wastewaters. Only by direct testing of
the wastewater to be treated can the perfor-
mance of an activated sludge system be deter-
mined with certainty.
(2) The partitioning of cyanide between that bio-
degraded, that stripped from solution, and
that escaping in the effluent is not known.
Several studies in the open literature report
cyanide as being biodegradable when it occurs
as part of a large matrix of organic species,
but removals vary widely. Similar unknowns
exist for sulfide.
The affect of residual polysulfide on downstream
processes and Its amenability to removal by sedi-
mentation or filtration needs to be determined.
In particular, since a biological treatment process
may well follow polysulfide addition, research is
needed to determine the fate of residual poly-
sulfide in an activated sludge system and its
affect on the microorganisms.
Laboratory tests are needed to more fully define
the reaction for varying conditions of pH, temper-
ature, and concentrations of cyanide and other
species. Experience in the petroleum refining
industry combined with the results of preliminary
studies Indicate that the reaction occurs at a
greatly increased rate when the polysulfide is
added as ammonium polysulfide rather than sodium
polysulfide. New research efforts should consider
this.
Research is needed to determine the feasibility
of accomplishing the cyanide conversion 1n con-
junction with some other process, whether that
be some other water pollution control process or
a part of the base plant. Likely possibilities
Include addition of polysulfide to the cyanide
wash or the gas cooling/dust removal circuit, to
an ion exchange column, or to an activated sludge
system.
The fate of cyanide and sulfide in the activated sludge
system needs to be determined. The biodegradabllity of
cyanide in particular is highly dependent on the partic-
ular microorganisms that are held in the activated
sludge reactor. Since the characteristics of these
microorganisms are specific to the characteristics of
the wastewater being treated, it is necessary to eval-
uate tne cyanide and suiflde species in a system treat-
Ing either an exact sample of the wastewater or one
very carefully synthesized.
(Continued)
-------
TABLE 5-3. (Continued)
Data Source and Basis
Data Gaps/Limitations
Research Needs
ro
en
Activated Sludge - Removal of Thiocyanate and Amnonia
Removal of thlocyanate and ammonia from waste-
waters containing essentially no organics is
based on performance recorded in the literature
and the results of a laboratory scale study.
Costs are based on values reported in the
literature, appropriately scaled on the basis
of system loading.
Performance has not been determined for the K-T
wastewaters.
Status:
Data Sources:
PCTM References:
B, C, D
3E, 4
Sections 4.2.1.4, 4.2.3.1,
4.2.3.2. and 4.2.4
Biological Denitrification
Removals of nitrate are based on performances
typically realized in parallel industries and
municipal applications. Costs are based on values
reported in the literature, appropriately scaled
on the basis of system loading.
Performance has not been determined for the K-T
wastewaters.
Status:
Data Sources:
PCTM References:
B, C, D
3E, 4
Sections 4.2.1.4, 4.2.3.1,
4.2.3.2, and 4.2.4
Clinoptilolite-based Ion Exchange
Removal of ammonia is based on the results of
EPA-sponsored, preliminary laboratory scale
studies using a synthetic wastewater having an
ammonia concentration representative of that
expected in the K-T wastewater. Costs are based
on values reported in the literature for two
plants treating municipal-strength waste-
waters: the Upper Occoquan plant (Virginia)
and the South Lake Tahoe plant (California).
Costs are extrapolated on the basis of
ammonia loading.
Status:
Data Sources:
PCTM References:
B, C, D
3E, 4
Sections 4.2.4, 4.2.3.1,
4.2.3.2, and 4.2.4
Performance has not been determined for the K-T
wastewaters. Since all existing clinoptilollte-
based ion exchange systems are designed to treat
municipal-strength wastewaters, little is known
about the cost characteristics of a system
designed to treat the higher ammonia loadings
associated wtih the K-T wastewaters.
Bench-scale studies using actual or synthetic waste-
waters could better assess the performance and require-
ments of this system. Commercial scale performance
should be determined when U.S. facilities become fully
operational.
Bench-scale studies using actual or synthetic waste-
waters could better assess the performance and require-
ments of this system. Commercial scale performance
should be determined when U.S. facilities become fully
operational.
Preliminary studies need to be supported by a more com-
prehensive effort. Research should be directed toward
determining feasibility of the process by concentrating
on the following areas of study: maximum period of
operating cycle, ammonia leakage during loading as a
function of residence time, regeneration requirements
of the cl1noptilol1te resin (including attrition),
affects of reduced sulfur and other reduced nitrogen
species on the resin, and possible changes in species
other than ammonia passing through the bed.
Information accumulated by operating systems needs to
be assembled and then applied to K-T-related systems.
Areas of information would include more detailed cost
data and operating experience with regeneration
equipment.
(Continued)
-------
TABLE 5-3. (Continued)
Data Source and Basis
Data Gaps/Limitations
Research Needs
ro
cr>
Activated Carbon Adsorption
Removals of all pollutants are based on data from
the refining and by-product coking industries.
Actual removal efficiencies will depend on pH,
molecule size, and structure of the organics
present in the wastewater. Costs are based on
data from the refining and by-product coking
industries, extrapolated on the basis of COD
loading.
Status:
Data Sources:
PCTM References:
B, C
3A
Sections 4.2,1.6, 4..2.2.}
and 4.2.4
Chemical Oxidation
Removals are based on the chemical oxidation
amenability of individual species in the waste-
water. Costs are based on data available in the
open literature.
Status. B, C, D
Data Sources: 3A, 3E, 4
PCTM References: Sections 4.2
4.2.3.2, and
Thermal Oxidation
.1.6, 4.2.3.1,
4.2.4
Destruction of organics is based on that realized
in other industries where organics, sometimes
much more refractory than those in the K-T
wastewaters, are removed. Costs are based on
vendor quotes.
Status: C
Data Sources: 3A, 3B, 5
PCTM References: Sections 4.2.1.6, 4.2.2.1
and 4.2.4
Performance has not been determined for the K-T
wastewaters.
No experience exists for testing the K-T waste-
water. Overall pH dependence and general
reactor requirements are not known. For waste-
waters where organics are present, removals are
uncertain, and unknown chemical species not
present in the influent are likely to be found
in the effluent.
Performance has not been determined for the K-T
wastewaters.
Laboratory or pilot scale tests are needed on the K-T
wastewaters. While many of the individual chemical
species determining COD are known, the performance can-
not be synthesized from performances typical for indi-
vidual species. Performance is generally specific to
the wastewater being treated and therefore must be deter-
mined on an individual case basis.
Testing of the K-T wastewaters on a scale larger than
laboratory scale is needed to better assess treatability.
Destruction of chemical species by chemical oxidation
under carefully controlled conditions are not necessarily
realized in a full scale process.
Additional research is not likely to improve the present
data base. Considerable research has been completed in
recent years, particularly In the area of destroying
organics regarded as hazardous. The capabilities and
limitations of the process are thereby well established.
Cooling Tower Concentration
All chemical species with exception of ammonia
and sulfide are assumed to be concentrated into
the blowdown stream without losses due to
volatilization or drift.
Status:
Data Sources:
PCTM References:
D
4
Sectons 4.2.1.7, 4.2.2.1,
4.2.3.1, 4.2.3.2 and 4.2.4
Performance has not been determined for the K-T
wastewaters. Loss of volatile species and poten-
tial for corrosion are not known.
Regardless of research, the feasibility of cooling tower
concentration would be determined on an individual case
basis. Commercial scale performance should be evaluated
in fully operational U.S. facilities.
(Continued)
-------
TABLE 5-3. (Continued)
Data Source and Basis Data Gaps/Limitations Research Needs
Forced Evaporation
All chemical species with the exception of ammonia Performance has not been determined for the K-T Unknowns are not expected to have a significant impact
and sulfide are assumed to be concentrated into wastewaters. The concentration of individual on the applicability of this process. However, perform-
the blowdown stream without losses due to volatil- chemical species that will be carried over into ance should be evaluated in fully operational U.S.
Ization. Costs are based on vendor quotes. the evaporator overhead is not known. Extent of facilities.
corrosion potential is not known.
Status: D
Data Sources: 4, 5
PCTM References: Sections 4.2.1.7, 4.2.2.1,
4.2.3.1, 4.2.3.2 and 4.2.4
-------
TABLE 5-4. DATA GAPS AND RESEARCH NEEDS - SOLID MEIDUM
Key For Technology Status and Data Sources
Technology Status
Data Source/Location
A. Commercial application 1n a K-T gasification facility
B. Bench or pilot scale testing
C. Technology transfer from another industry - similar
but not identical streams
D. Conceptual
1. AECI Limited, Modderfonteln, South Africa
2. Nitrogenous Fertilizer Industry, S.A., Ptolemais, Greece
3. Technology transfer from related industries
3A. Petroleum refining/petrochemical production
3B. Coke production
3C. Electric power generation
3D. Natural gas processing
3E. Municipal waste treatment
4. Conceptual or proposed design/engineering studies
5. Vendor supplied information
Data Source and Basis
Data Gaps/Limitations
Research Needs
1^3 Uncontrolled Primary Waste Streams
00 Quenched Slag (Stream 207)
Slag generation rates, Teachability data and resi-
dual moisture are based upon commercial scale gasi-
fication of an Illinois No. 6 coal at the Nitro-
genous Fertilizers Industry, S.A. in Ptolemais,
Greece. Contaminants in the residual moisture have
been estimated assuming a common slag quench/washer
cooler water circuit, without slag rinsing.
Slag/dust ash partitioning and slag leachability
data are not available for other bituminous
coals or lower rank coals.
Status:
Data Sources.
PCTM References:
A, D
2, 4
Sections 3.2.1 and 4.3.2.1
Slag/dust partitioning and slag Teachability data
for other bituminous coals and lower rank coals should
be determined. Partitioning data for various ranks
of foreign coals may be obtainable through licensors/
operators.
(Continued)
-------
TABLE 5-4. (Continued)
Data Source and Basis
Data Gaps/Limitations
Research Needs
Dewatered K-T Dust (Stream 209)
The composition of dry K-T dust and the associ-
ated liquids are based upon data from commercial
scale gasification of Illinois No. 6 coal at the
Nitrogenous Fertilizers Industry, S.A. in Ptolemais,
Greece. A nominal dewatered dust moisture content
has been assumed consistent with that indicated
in a conceptual design for a sub-bituminous coal
based hydrogen production facility.
Slag/dust ash partitioning data, dust carbon and
sulfur contents, and dust Teachability data are
not available for other bituminous coals or lower
rank coals. Dust dewatering data and structural
stability data are not available.
Status:
Data Sources:
PCTM References:
A. D
2, 4
Sections 3.2.1 and 4.3.2.1
Control Technologies
Resource Recovery
Reuse as Construction Material
Potential alternatives for recycling the gasi-
fier slag are based upon current practices for
ash from coal-fired power plants.
Status:
Data Sources:
PCTM Reference:
C
3C
Section 4.3.2.1
Fluidized Bed Combustion of Dust
The burning of K-T dust in a fluidized bed
combustion (FBC) boiler is based upon an
approach under consideration for a proposed
K-T plant in Alabama and an existing K-T
plant in Modderfontein, South Africa.
Status: B, D
Data Sources: 1, 4
PCTM Reference: Section 4.3.2.1
The available market for recycling the slag is
uncertain.
(1) The viability of commercial scale FBC boiler
using dewatered K-T dust as fed has not been
proven.
(2) Although there are cost data for coal-fired
FBC units, these data may not apply to com-
bustion of high moisture and ash K-T dust.
Slag/dust ash partitioning data, dust carbon and sulfur
contents, and dust Teachability data for other bitumi-
nous coals and lower rank coals should be determined.
Partitioning and dust characterization data for various
ranks of foreign coals may be obtainable through
licensors/operators. Dust dewatering data should be
obtained for a range of coal types. Engineering and
geotechnical data are needed to establish design data
for land-based disposal techniques.
The availability of market is highly dependent upon
local conditions which can only be assessed on a site-
by-site basis.
The viability of FBC with K-T dust should be determined.
Data might be obtainable through AECI Limited which is
currently evaluating this alternative.
Secondary Waste Streams
o FBC Boiler Ash
o Spent FBC Boiler Bed Sorbcnt
No Teachability data are available.
The flow rate and the Teachability of this mate-
rial is not known.
The Teachability data should be determined for range
of coal types.
The flow rate and the Teachability of this material
might be obtained from AECI.
(Continued)
-------
TABLE 5-4. (Continued)
Data Source and Basis
Data Gaps/Limitations
Research Needs
Landfill
Landfilling of the waste is based upon current
practices in the electric utility industry and
other industries.
Status:
Data Sources:
PCTM Reference:
C
3C
Section 4.3.2.1
The long term leachate generation rates and char-
acteristics are not known, and the compatibility
and long term performance of landfill liners have
not been established.
Surface Impoundment
Disposal of waste in surface impoundments is
based upon current practices in the electric
utility industry and other industries.
Same as landfill.
Status:
Data Sources:
PCTM Reference:
Fixation
C
3C
Section 4.3.2.1
Fixation of the K-T dust to Improve Us struc-
tural stability and leachablllty characteristic
1s based on current practices for FGD sludge
and other industrial wastes.
Performance and cost for the application are not
known.
Status:
Data Sources:
PCTM References:
C
3C
Section 4.3.2.1
Long term column or landfill cell studies are needed
to characterize the leachate and to determine liner
performance. Appropriate materials for these tests
may not be available until the first K-T plant is con-
structed in the U.S. Also, since these are long term
studies the data may not be available in time to
influence facility designs that are currently in
advanced stages.
Same as landfill.
Treatability studies with various chemical additives
should be performed to determine the technical and
economic feasibility of this approach.
-------
TABLE 5-5. DATA GAPS AND RESEARCH NEEDS - PRODUCTS/BY-PRODUCTS
Key for Technology Status and Data Sources
Technology Status
Data Sources
A. Commercial application In a K-T gasification facility
B. Bench scale or pilot testing
C. Technology transfer front another Industry - similar but not
identical streams
D. Conceptual
!. AECI Limited, Modderfontein, South Africa
2. «ftrogeneous ferttlizer Industry, S.A., PtoJawls.
3 Technology transfer from related Industries
3A. Petroleum refining/petrochemical production
38. Coke production
3C. Electric power generation
3D Natural gas processing
3E. Municipal waste treatement
3F. Methanol production
4 Conceptual or proposed design/engineering studies
5. Vendor supplied Information
Data Source and Basis
Data Gaps/Limitations
Research Needs
Flscher-Tropsch Synthesis Products
baseline. Diesel Oil. Heavy Oil. Alcohols.
LPG and SBG (Stream tOO through 1Q7)
Production rates and composition estimates are
based upon conceptual designs for coimercial
scale F-r faculties.
Status:
Data Sources
PCTH References
D
4
Sections 3 4.2 and 3 5 2
Characterization, storage stabi1ity,and «*•
data for F-T products are lirnted The influence
of synthesis process variables on product cfcjrac-
medlates processdbility, and consumption/utilization
characteristics are similarly largely unknown.
Detailed characterization data are
desirable. Environment*! effects
associated with fugitive and evapora-
tive emissions and product utilization
should b* determined Such effects
should be evaluated relative to environ
mental effects associated with analog
petroleum products.
Methanol Synthesis Product
Fue _1_G r_a de Jteth anp_j_ iS_tream__i 08J
Production rates are based upon conceptual
designs for commercial scale methanol
facilities. Composition estimates are
based upon conceptual designs and analyses
of metha.no! produced from petroleum feedstocks
Source:
3ata Sources:
?CTM References.
C, D
3F, 4
Sections 3 4.1 and 3 S 1
lobll M-gasoltne Synthesis Products
gasoline. HI »fd Butanes . and Propane
.'Streams 109. HO. and 1HJ
^reduction rates and composition estimates are
used upon conceptual designs of commercial
-,cale Hob II H-gasollne facilities
Status.
Jata Sources:
References:
D
4
Sections 3 4 4 and 3 5.3
Characterization data for methanol from indirect
coal liquefaction are limited The Influence of
synthesis process variables on product wthanol
characteristics 1s unknown
Characterization data for Mobil M-gasollne
products are limited The Influence of synthesis
process variables on product characteristics
is unknown.
The EPA Office of Pesticides and Toxic
Substances has declared that methanol,
does not require Premanufacturlng Notices
for the production of coal-based methanol.
On this basis, no research effort )s
required.
Detailed characterization data are desirable.
Environmental effects asslciated with fugitive
and evaporative emissions and product
utilization should be determined. Such
effects shoutd be evaluated1 relative to
environmental effects associated with analog
petroleum products.
3y-Products
Sulfur from Claus .
Stream 403 and 41
I Stretford Process
^reduction rate estimates are based upon pub-
Msned control performance data and engineering
estimates of waste gas sulfur contents
.tatus
iata Sources -
Deferences:
C, D
3A. 3B. 3D, 4
Sections 4.1.1. 4 1.1 1
and 4.1.1 5
Essentially, no characterization data are available
for Claus and Stretford sulfur
Characterization of sulfur is desirable,
particularly with respect to contaminants
possible downstream processing for various
purposes
431
-------
Section 6
References
SECTION 6
REFERENCES
1. Schreiner, Max. Research Guidance Studies to Assess Gasoline from Coal
by Methanol-to-Gasoline and Sasol-Type Fischer-Tropsch Technologies,
Mobil Research and Development Corporation, FE-2447-13, August 1978.
2. Conceptual Design of a Coal-to-Methanol-to-Gasoline Commercial Plant.
Volume I, Badger Plants Incorporated, Cambridge, Mass., FE-2416-43.
3. R.M. Parsons Co., Screening Evaluation for Synthetic Liquid Fuels
Manufacture, EPRI AF-523, August 1977.
4. Billings, Roger E. Hydrogen from Coal Cost Estimation Guidebooks.
NASA-CR-164692, 1981.
5. Fischer-Tropsch Design Project Capital Cost Validation, U.S. Army
Engineer Division, Huntsville, Alabama, U.S. Department of Energy
Report No. FE-1759-2, October 1977.
6. Preliminary Economic Analysis of Lurgi Plant Producing 250 Million SCFD
Gas from New Mexico Coal, U.S. Department of the Interior, Bureau of
Mines, Morgantown, West Virginia, ERDA Document No. 76-5, March 1976.
7. Gesellschaft ftir Kohle-Technologie mbH. Large Scale Gasification Tests
with U.S. Coal in Ptolemais, Greece, for Tennessee Valley Authority, USA.
Essen, Federal Republic of Germany. Final Report, Vol. Ill, October 1981
8. Firnhaber, B. and R. Wetzel. Status of Entrained Coal Gasification
According to Koppers-Totzek and Shell-Koppers. The Institute of Chemical
Engineers, Symposium Series No. 62, 1980.
9. Trials of American Coals in a Lurgi Gasifier at Westfield, Scotland.
Woodall-Duckham, Ltd., Sussex, England. ERDA R&D Report No. 105, 1974.
10. Souther, N.F., et al. Potential Trace Element Emissions from the Gasi-
fication of Illinois Coal, Illinois Institute of Environmental Quality
No. 75-08, February 1975.
11. Axetell, K., Jr. Survey of Fugitive Dust from Coal Mines, (PEDCo
Environmental, Inc.), EPA-908/1-78-003, February 1978.
12. Blackwood, T.R. and,R.A. Wachter. Source Assessment: Coal Storage
Piles, Monsanto Research Corporation, Dayton, Ohio, May 1978.
432
-------
Section 6
References
13. Jutze, G.A., et al. Technical Guidance for Control of Industrial Process
Fugitive Emissions. PEDCo Environmental, Cincinnati, Ohio, March 1977,
NTIS: PB 272-288.
14. Buroff, J., J. Strauss, A. Jung and L. McGilvary. Environmental Assess-
ment: Source Test and Evaluation Report, Coal Preparation Plant No. 1
(Versar Inc.) U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, July 1980.
15. Buroff, J., J. Strauss, A. Jung and L. McGilvray. Environmental Assess-
ment: Source Test and Evaluation Report, Coal Preparation Plant No. 2.
Draft Report (Versar Inc.) U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, March 1981.
16. Wewerka, E.M., et al. Trace Element Characterization of Coal Wastes:
Second Annual Progress Report, Los Alamos Scientific Laboratories, EPA-
600/7-78-028, July 1978.
17. Cox, D.P., T.Y.J. Chu and R.J. Ruane. Characterization of Coal Pile
Drainage, TVA, EPA-600/7-79-051, February 1979.
18. Ferraro, F.A. Treatment of Precipitation Runoff from Coal Storage Piles,
Presented at Third Symposium on Coal Preparation, NCA/BCR, Louisville,
Kentucky, October 1977.
19. Wewerka, E.M., J.M. Williams, P.L. Wanek, and J.D. Olsen. Environmental
Contamination from Trace Elements in Coal Preparation Wastes, Los Alamos
Scientific Laboratories, EPA-600/7-76-007.
20. PEDCo-Environmental, Inc., Assessment of Fugitive Particulate Emission
Factors for Industrial Processes, Cincinnati, Ohio, EPA-450/3-78-107,
September 1978. NTIS:PB-288-859.
21. PEDCo-Environmental, Inc., Environmental Assessment of Coal Transporta-
tion, Cincinnati, Ohio, EPA-600/7-78-08, May 1978.
22. Rittenhouse, R.C. Fuel: Handling and Storage at Power Plants, Power
Engineering, December 1979. pp. 42-50.
23. U.S. EPA, Compilation of Air Pollution Emission Factors, Office of Air
Quality Planning and Standards, Research Triangle Park, N.C., AP-42.
24. Zee, C.A., J. Clausen and K.W. Crawford. Environmental Assessment Source
Test and Evaluation Report, Koppers-Totzek Process, EPA-600/7-81-009,
January 1981.
433
-------
Section 6
References
25. FMC Corporation, Gasification of COED Chars in a Koppers-Totzek Gasifier,
EPRI AF-615, July 1978.
26. The Dravo Corporation. Handbook of Gasifiers and Gas Treatment Systems.
Report prepared for the United States Energy Research and Development
Administration, FE-1772-11, February 1976.
27. GKT Gesellschaft fur Kohle-Technologie mbH. GKT's Coal Gasification
Process Facts and Data.
28. Chaurey, K.H. and K.C. Sharma. Coal Based Ammonia Plants - Preliminary
Operating Experience of Coal Gasification at Talcher and Ramagundam
Fertilizer Plants of the Fertilizer Corporation of India. Symposium
on Ammonia from Coal, Tennessee Valley Authority, Muscle Shoals,
Alabama, May 8-10, 1979.
29. Engelbrecht, A.D. and L.J. Partridge. Operating Experience with a 1000
ton/day Ammonia Plant at Modderfontein. Symposium on Ammonia from Coal,
Tennessee Valley Authority, Muscle Shoals, Alabama, May 8-10, 1979.
30. Farnsworth, J.F., D.M. Mitsak, and J.F. Kamody. Clean Environment
with K-T Process. Presented at EPA Meeting: Environmental Aspects of
Fuel Conversion Technology, St. Louis, Missouri. May 13-16, 1974.
31. Hunter, C.A. and K.Y. Yu. Characterization of Solid Wastes from In-
direct Liquefaction Facilities. Environmental Aspects of Fuel Conver-
sion Technology - VI, A Symposium on Coal-Based Synfuels, Denver,
Colorado, October 26-30, 1981
32. Tennessee Valley Authority, Office of Natural Resources. Wastewater
and Solid Waste Characteristics, Process and Nonprocess Units. Pre-
pared for Office of Coal Gasification, October 1982.
33. Unpublished TRW non-proprietary data obtained during testing at the
Nitrogenous Fertilizer Industry (NFI), Ptolemais, Greece.
34. Wetzel, R.E., K.W. Crawford, and W.C.Yee. Environmental Aspects of the
GKT Coal Gasification Process. EPA Symposium on Environmental Aspects
of Fuel Conversion Technology-IV, Denver, Colorado, October 26-30, 1981.
35. Ranke, G. Acid Gas Separation by Rectisol in SNG Processes, Linde AG,
Munich, Germany, copy of presentation obtained through Lotepro Corpora-
tion, New York, N.Y.
434
-------
Section 6
References
36. Seidell, A., and W.F. Linke. Solubilities, Inorganic and Metal-Organic
Compounds. D. Von Nostrand Company, Inc., Princeton, New Jersey, 1958.
37. Scholz, W.H. Rectisol: A Low-Temperature Scrubbing Process for Gas
Purification, Advances in Cryogenic Engineering, Vol. 15, 1969.
38. Lotepro Corporation. Capabilities brochure by Lotepro Corporation.
39 Information supplied by South African Coal, Oil, and Gas Corp. Ltd.,
to EPA's Industrial Environmental Research Laboratory, Research Triangle
Park, November 1974.
40. Satterfield, C.N. Heterogeneous Catalysts in Practice, McGraw-Hill
Book Co., New York, N.Y., 1980.
41. Catalyst Handbook, Imperial Chemical Industries, Ltd., Springer-Verlag,
New York, N.Y., 1970.
42. Katalco 71-7 High-Temperature Shift Catalyst Data Sheet, Katalco Corp.,
Oak Brook, Illinois, 1981.
43. Dybkjaer, Ib. Carbon Monoxide Shift Catalysts and Carbonyl Sulfide
Hydrolysis Catalysts. Ammonia from Coal Symposium, Tennessee Valley
Authority, May 8-10, 1979.
44. Katalco 49-2 Cobalt-Molybdenum Shift Catalyst Data Sheet, Katalco Corp.,
Oak Brook, Illinois.
45. Katalco 52-2 Low Temperature Shift Catalyst Data Sheet, Katalco Corp.,
Oak Brook, Illinois, 1976.
46. Telephone Memorandum, Blythe, B., Davy McKee, Lakeland, Florida,
November 5, 1981.
47. Telephone Memorandum, Sandlin, A.D., Harshaw Chemical Co., Houston,
Texas, August 14, 1981.
48. Telephone Memorandum, Rounthwaite, D., Katalco Corp., Oak Brook,
Illinois, November 5, 1981.
49. Kohl, A. and F. Reisenfeld. Gas Purification, Gulf Publishing Col,
Houston, Texas, 1974.
50. Maddox, R.N. Gas and liquid Sweetening, Campbell Petroleum Series, 1974.
435
-------
Section 6
References
51. U.S. EPA, Control of Emissions from Lurgi Coal Gasification Plants,
Office of Air Quality Planning and Standards, Research Triangle Park,
N.C. EPA-450/2-78-012, OAQPS 1.2-073, March 1978, 178 p.
52. Information provided to TRW by Linde AG, November 1982.
53. Allen, D.W. Final Sulfur Removal in Ammonia from Coal Plants, Symposium
on Ammonia from Coal, Tennessee Valley Authority, Muscle Shoals,
Alabama, May 8-10, 1979.
54. Ghassemi, M., K.W. Crawford, and S. Quinlivan. Environmental Assessment
Data Base for High-Btu Gasification Technology. Volumes I-III, TRW
Environmental Engineering Division, Redondo Beach, California, EPA-600/
7-78-186a, b, and c, September 1978.
55. Ghassemi, M. et al. Applicability of Petroleum Refinery Control
Technologies to Coal Conversion. (TRW Inc.), EPA 606/7-78-190, Oct.
1978.
56. Manual on Disposal of Refinery Wastes - Volume on Liquid Wastes.
American Petroleum Institute, First Edition 1969.
57. Manual on Disposal of Refinery Wastes - Volume on Atmospheric Emissions,
American Petroleum Institute, Publication No. 931.
58. Sinor, J.W. Evaluation of Background Data Relating to New Source Per-
formance Standards for Lurgi Gasification, Cameron Engineers Inc ,
Denver, Colorado, EPA 600/7-77-057, June 1977, 233 p.
59. Mehta, D.D. and W.W. Pan. Purify Methanol This Way. Hydrocarbon
Processing, February 1971. pp. 115-120.
60. C.F. Braun and Company, Carbonyl Formation in Coal Gasification Plants,
Prepared for Energy Research and Development Administration and American
Gas Association, FE-2240-16, December 1974.
61. Storch, H.H., et al., Synthetic Liquid Fuels from Hydrogenation of Carbon
Monoxide, U.S. Bureau of Mines, 1948.
62. Hueper, W.C. Experimental Carcinogenic Studies of Hydrogenated Coal Oils,
II, Fischer-Tropsch Oils, Industrial Medicine and Surgery, October, 1956,
p. 459-462.
436
-------
Section 6
References
63. Hoogendorn, J.C. Experience with Fischer-Tropsch Synthesis at SASOL,
paper presented at Institute for Gas Technology, Chicago, Illinois,
1973, pp. 353-365.
64. Schreiner, E. Motor Gasolines, Summer 1979, U.S. Department of Energy,
Bartlesville Energy Technology Center, Bartlesville, Oklahoma, February
1980.
65. Mullowney, J.F. and P.F. Mako. Coal to Transport Fuels and Chemicals:
SASOL Two/SASOL Three, paper presented at American Chemical Society
National Meeting, Division of Petroleum Chemistry, Las Vegas, Nevada,
1980, 20 pp.
66. Bowden, J.M. and D.W. Brinkman. Stability of Alternate Fuels. Hydrocar-
bon Processing, July 1980, pp. 77-82.
67. Enviro Control Inc. Relative Health Effects of Gasoline and Heating Fuel
Derived from Petroleum or Synthetic Crudes, prepared for U.S. Department
of Energy under Contract No. DE-AC01-79PE-70021, Washington, D.C., 1980,
79 pp.
68. Smith, R.P. Toxic Responses of the Blood, Chapter 14. In: Casarett and
Doull's Toxicology: The Basic Science of Poisons, Second Edition,
J. Doull, C.D. Klaasen and M.O. Amdur, eds., Macmillan Publishing Co.,
Inc., New York, New York, 1980, pp. 311-331.
69. Enviro Control Inc., Trip Report, SASOL I, Sasolburg, South Africa,
December 5-7, 1977, Prepared for National Institute for Occupational
Safety and Health, Rockville, Maryland.
70. Seames, Wayne S. The Kosovo Lurgi Gasification Plant Phase I and Phase
II Data Compilation, Draft Report prepared by Radian Corp. for EPA,
February 18, 1980.
71. U.S. Geological Survey, Quality of Surface Waters of the U.S., Water
Supply Paper 2143, 1969.
72. Hart, F.C., et al. The Impact of RCRA (PL 94-580) on Utility Solid Wastes,
EPRI Report FP-878, TPS 78-799, August 1978.
73. Goldstein, D.J. and D. Yung. Water Conservation and Pollution Control in
Coal Conversion Processes, EPA-600/7-77-065, June 1977.
437
-------
Section 6
References
74. Dickey, J.B. and D.N. Dwyer. Managing Waste Heat with the Water Cooling
Tower, Missouri Valley Electric Association, 1979 Engineering Conference,
3rd Edition.
75. Scharle, W.J. Large Oxygen Plant Economics and Reliability, TVA
Symposium on Ammonia from Coal, Muscle Shoals, Alabama, May 8-10, 1979.
76. Op. Cit., Reference 23.
77. Pacific Environmental Services. Compliance Analysis of Small Bulk Plants,
Report prepared under EPA Contract 68-01-3156, Task No. 17, October 1976.
78. Wetherold, R. and L. Provost. Emission Factors and Frequency of Leak
Occurrence for Fittings in Refinery Process Units. EPA 600/2-79-044,
February 1979.
79. Wetherold, R.G., L.P. Provost, and C.D. Smith. Assessment of Atmos-
pheric Emissions from Petroleum Refining, Appendix F. EPA-600/2-80-075q,
U.S. Environmental Protection Agency, April 1980.
80. Water Reuse Studies, API Publication 549, August 1977.
81. Climatic Atlas of the United States, DOC-NOAA, 1974.
82. Coal Gasification Project, Draft Environmental Impact Statement.
Tennessee Valley Authority, 1980.
83. Lim, K.J., H. Lips and R.J. Milligan. Technology Assessment Report for
Industrial Boiler Applications: NO Combustion Modification, EPA
600/7-79-178f, December 1979.
84. Acurex Corp., Control Techniques for Nitrogen Oxides Emissions from
Stationary Sources - Second Edition, EPA-450/1-78-001, January 1978.
85. Telephone Memorandum, Joe Barrows, Babcock and Wilcox, Ohio, June 30,
1982.
86. TRW In-house information obtained from Mike Boughton, Redondo Beach,
California, September 1981.
87. Young, C.W., et al. Technology Assessment Report for Industrial Boiler
Applications: Fluidized Bed-Combustion, EPA 600/7-79-178e, November
1979.
88. Information provided to TRW from Black & Veatch, April 1981.
438
-------
Section 6
References
89. Castelini, J. Fugitive Coal Dust Control. Power Engineering, July
1979. pp. 86-87.
90. Op. Cit., Reference 13.
91. Bulk Gasoline Terminals - Background Information for Proposed Standard.
Draft report, Emission Standards and Engineering Division, U.S. EPA,
May 1980.
92. VOC Emissions from Volatile Organic Liquid Storage Tanks - Background
Information for Proposed Standards, Preliminary Draft, Emission
Standards and Engineering Division, U.S. EPA, November 1980.
93. Control of Volatile Organic Compound Leaks from Petroleum Refinery
Equipment, Guideline Series, Emission Standards and Engineering Division,
U.S. EPA, EPA 450/2-78-036, June 1978.
94. Industrial Ventilation. A manual of recommended practice by the Committee
on Industrial Ventilation, Michigan, U.S.A. Twelfth Edition, 1972.
95. Pojasek, R.B., ed. Toxic and Hazardous Waste Disposal, Vol. I. Ann
Arbor Science Publication, Inc., Ann Arbor, Michigan, 1979.
96. Conner, J.R. Disposal of Liquid Wastes by Chemical Fixation, Waste Age,
September 1974, pp. 26-45.
97. Michael Baker, Jr., Inc. Electric Power Research Institute Report No.
EPRI FR-671, January 1978.
98. Survey of Solidification/Stabilization Technology for Hazardous Industrial
Wastes. EPA Report No. EPA 600/2-79-056, July 1979.
99. Couch, A.T. and W.L.E. Davey. The Use of Fluidized Combustion to Burn
the Fly Ash from Koppers-Totzek Gasifiers. The International Coal Con-
version Conference, Pretoria, South Africa, August 16-20, 1982.
439
-------
Appendix A
Costing
APPENDIX A
COSTING METHODOLOGY, BASES, AND ASSUMPTIONS
Capital and total annualized cost estimates were developed for uncon-
trolled Koppers-Totzek (K-T) based synfuels facilities and for pollution
control processes in order to provide an indication of the economic impact
of pollution controls. These estimates are based primarily on factored
estimates of costs contained in non-proprietary published literature. As
such, they should be viewed only as general indicators of expected costs and
should not be construed as definitive cost estimates for a specific plant.
All costs have been adjusted to a 1980 dollar basis using generally accepted
cost indexes such as the Chemical Engineering (CE) plant cost annual index.
To the extent possible, the same methodology was used to develop capital
and total annualized cost estimates for both the base plants and pollution
controls. Details of these methodologies are presented in Sections A.I and
A.2, respectively. The bases for the base plant cost estimates are pre-
sented in Section A.3. Bases for the pollution control cost estimates are
presented in the Pollution Control Appendices for the PCTM series.
A.I CAPITAL COST ESTIMATING METHODOLOGY
Costs presented as capital costs or investments in the K-T PCTM are
total depreciable investments (TDI). TDI includes:
1) purchased and delivered equipment costs;
2) labor and materials costs to install equipment;
3) indirect installation charges, such as
• engineering and construction costs,
• contractor fees, and
t project and process contingency reserves, and;
4) interest expenses on capital spent prior to start of production
(interest during construction).
A-l
-------
Appendix A
Costing
A variety of methods can be used to estimate the above cost items, although
most methodologies utilize a factored approach. In factored cost estimates,
the cost of purchased and delivered equipment is obtained from vendor quotes
or estimated from previous projects using similar equipment. The remaining
cost items are then estimated as a "factor" times the purchased and delivered
equipment costs or other subsequently derived subtotal.
For costs estimates developed for this PCTM, the major source of cost
information was the open literature, although some vendor quotes were used.
In general, literature cost information is not reported as delivered equip-
ment costs: some data are published as installed equipment costs (purchased
equipment plus direct installation costs), some include one or more of the
indirect installation charges (listed previously), some are TDI estimates, and
others are total capital investment requirements (TDI plus working capital and
land costs).
In order to provide consistency in the various capital cost estimates
required for the PCTM, a capital cost methodology was developed. The method-
ology (and cost factors) used are summarized in Table A-l. Most cost data
obtained from the literature were installed equipment costs (IEC) which often
included components such as piping, instrumentation, and substructures. As
indicated in the table, indirect installation charges were estimated as 48%
of the IEC. Adding the indirects to the IEC gave the total plant (or process)
costs. Interest during construction (IDC) was estimated as 22.6% of the total
plant costs. The total depreciable investment (TDI) is the sum of these two
terms. Working capital (WC) for the base plant cost estimates has been
assumed to be the value of a 60 day coal inventory. No provision for working
capital related to pollution control equipment has been included. Summing
the TDI and WC gives the total capital investment (TCI). It should be noted
that the PCTM does not address a specific K-T synfuels facility or facility
location, and thus the amount and value of associated land is unknown. There-
fore, for cost estimating purposes, the cost of land has not been included in
the total capital investment estimates presented.
A-2
-------
Appendix A
Costing
TABLE A-l. CAPITAL COST ESTIMATING METHOD
Installed Equipment Costs (IEC)
Indirect Installation Costs (IIC)
Engineering and Construction (25% of IEC)
Fees (3% of IEC)
Contingency (20% of IEC)
Total Plant Costs (TPC = IEC + IIC)
Interest During Construction (IDC = 22.6% of TPC)
Total Depreciable Investment (TDI = TPC + IDC)
Working Capital (WC = value of 60 day coal inventory for base plant cost
estimates)
*Total Capital,Investment (TCI = TDI + WC)
*The PCTM does not address a specific K-T synfuels facility or facility
location, and thus the amount and value of associated land is not unknown.
Therefore, for cost estimating purposes the cost of land has not been in-
cluded in total capital investment estimates presented in this PCTM.
A.2 ANNUALIZED COST ESTIMATING METHODOLOGY
Annualized costs consist of annual operating expenses plus annualized
capital-related charges. Annual operating expenses include costs for labor
(operating, supervision, and maintenance), raw materials, chemicals, cata-
lysts, utilities (steam, electricity, cooling water, etc.), and overhead.
Capital-related charges include interest on working capital, local taxes,
insurance, depreciation, income taxes, and return on investment. The unit
costs or factors used to estimate total annualized costs in this PCTM are
listed in Table A-2. All of the terms listed in Table A-2, except capital
recovery, are expressed in first year costs (i.e., in constant 1980 dollars).
The capital recovery, however, is a levelized value calculated using standard
present worth and levelized cost procedures and the economic assumptions
listed in Table A-3.
A-3
-------
TABLE A-2. UNIT COSTS AND FACTORS FOR ANNUALIZED COST ESTIMATES
Operating Labor (Sll/hr)
Supervision (15% of operating labor)
Maintenance (2% of total depreciable investment)
Maintenance Supervision (5% of maintenance)
Raw Illinois No. 6 Coal ($35.44/Mg)
Raw Water ($0.031/m3)
Utilities
Steam ($6.30 to $9.10/Mg depending on quality)
Electricity ($0.033/kW-hr)
Fuel Gas ($1.79/GJ) ?
Cooling Water ($0.08/nr) ?
Boiler feed water ($0.264/m )
Chemicals and Catalysts (representative early 1980 costs)
Overhead Charges
Plant overhead (50% of operating labor)
General and administrative overhead (15% of operating labor and maintenance)
Laboratory Charges (5% of operating labor)
Capital Related Charges
Interest on working capital (12% of working capital)
Local taxes and insurance (3.5% of total depreciable investment)
Capital recovery, including income taxes, depreciation, and profit (13.7% of total depreciable
investment)
TOTAL ANNUALIZED COSTS (summation of above items)
-------
Appendix A
Costing
TABLE A-3. ASSUMPTIONS USED TO CALCULATE CAPITAL RECOVERY FACTOR
Financing basis: 100% equity
Desired after tax return on investment: 12% of total depreciable invest-
ment
Income tax rate: 48% of taxable income
Economic facility life: 20 years
Facility life for depreciation purposes: 16 years
Depreciation method: sum-of-the-years-digit
Investment tax credit: 20% of total depreciable invest-
ment
A.3 BASE PLANT COSTS
Capital and total annualized cost estimates were developed for base
plants examined. ("Base plant" in this PCTM refers to a K-T based synfuel
plant with fuel production capabilities but without pollution control devices.)
The cost estimates were developed from information found in the open literature
and adjusted to the bases used in this PCTM. Specifically, the literature
data were 1) adjusted to delete the cost of pollution controls (to the extent
those costs could be identified), 2) scaled to the plant capacities examined
in the PCTM, and 3) escalated to a 1980 dollar basis.
A.3.1 Base Plant Capital Costs
The major source of cost data for the methanol synthesis base plant is
an engineering study performed by the Ralph M. Parsons Company (6). The
plant capacity examined in this manual is approximately 36 percent of the
capacity examined in the Parsons study. After the identifiable pollution
control costs were subtracted from the installed equipment costs (IEC) report-
ed in Reference 6, the IEC was adjusted to the PCTM capacity by multiplying
A-5
-------
Appendix A
Costing
by 0.437. This factor equals 0.36 raised to the 0.8 power, which was con-
sidered appropriate because many parts of the plant in the Parsons study
consisted of multiple (4 to 7) trains. Thus, the majority, but not all of
the desired capacity reduction (and hence cost reduction) could be accomplish-
ed by eliminating one or more process trains. In the cases of the Fischer-
Tropsch (F-T) and Mobil M-gasoline syntheses base plants, the same approach
was used although capital costs associated with methanol synthesis in the
Parsons study were adjusted to reflect differences in costs among the three
synthesis operations based on other cost data sources (7, 8, 9, 10).
The resulting adjusted installed equipment costs (IEC) were then escala-
ted to 1980 dollars using the Chemical Engineering (CE) plant cost annual
index. Finally, the escalated IEC was used to compute the other elements of
the total capital investment as outlined earlier in Section A.I. The result-
ing base plant capital cost estimates are summarized in Table A-4.
TABLE A-4. CAPITAL COSTS FOR UNCONTROLLED K-T BASED
INDIRECT LIQUEFACTION PLANTS
Capital Costs, 106 Dollars (1980 basis)
Item
Installed cost
Contractors Overhead and Fee
Engineering and Construction
Contingency
Subtotal, Facility Cost
Interest During Construction
Working Capital
Total Capital Investment
Methanol
603
18
151
121
893
201
17
1111
Fischer-
Tropsch
714
21
178
143
1056
238
20
1314
Mobi 1
M- Gasoline
657
20
164
131
972
220
17
1209
A-6
-------
Appendix A
Costing
A.3.2 Base Plant Annualized Costs
Total annualized costs for the base plants have been estimated as the
sum of the total operating costs and annualized capital changes. The total
operating costs are based upon the annual coal cost ($35.44/Mg) and cost
estimates for "water, catalyst, and chemicals" and "other operating costs"
presented for the K-T based methanol production facility examined in the
Parsons study (6). Published operating cost estimates were scaled directly
on the basis of plant capacity and escalated to 1980 dollars. Published cost
estimates for "water, catalyst, and chemicals" and "other operating costs"
relate specifically to a methanol production facility; however, no adjustment
has been made for F-T and Mobil M-gasoline synthesis cases. Furthermore,
it is not known if any adjustment is required since insufficient details of
these estimates are available. Similarly, the annual operating costs for
pollution control equipment have not been deducted from cost estimates for
"water, catalyst, and chemicals" and "other operating costs" since insuffi-
cient details of these estimates are available to do so. Capital-related
charges have been estimated as outlined earlier in Section A.2.
The resulting total annualized cost estimates for K-T base plants are
summarized in Table A-5. It should be noted that annual coal costs and
capital-related charges comprise about 84% of the total annualized cost.
Therefore, uncertainties associated with estimated costs for "water, catalyst,
and chemicals" and "other operating costs" are not expected to have a major
impact on the estimated total annualized cost.
A-7
-------
Appendix A
Costing
TABLE A-5. ANNUALIZED COSTS FOR K-T BASED INDIRECT LIQUEFACTION PLANTS*
Item
Coal
Water, Catalyst, and Chemicals*
Other Operating Costs*
Total Operating Cost
Capital Charges
Total Annuali zed Cost
Annual
Methanol
98
4
53
155
191
346
ized Cost,
Fischer-
Tropsch
114
4
63
181
226
407
106Dollars
Mobil
M-Gasoline
95
4
51
150
208
358
*Annual operating costs relating to "water, catalyst, and chemicals" and
"other operating costs" are based upon published cost estimates for a K-T
based methanol production facility (6). Insufficient details were available
to enable adjustment, if any is required, for F-T and Mobil M-gasoline
synthesis cases or to deduct the annual operating costs for pollution
controls.
A.4 REFERENCES
1. Guthrie, K.M. Process Plant Estimating, Evaluation and Control. Crafts-
man Book Company of America, Solana Beach, CA. 1974.
2. Costs and Analysis Manual for Utility Boiler Standards Support
Document, Pedco Environmental, 1979.
3. Peters, M.S., and K.D. Timmerhaus. Plant Design and Economics for
Chemical Engineers. 2e. McGraw-Hill Book Company, New York, NY. 1968.
4. Uhl, V.W. A Standard Procedure of Economic Evaluation for Pollution
Control Operations. EPA-600/8-79-018a, June 1979.
5. Synthetic Gas-Coal Task Force. The Supply-Technical Advisory Task Force
Synthetic Coal-Gas. April 1973.
A-8
-------
Appendix A
Costing
6. Ralph M. Parsons Company. Screening Evaluation for Synthetic Liquid
Fuels Manufacture, EPRI AF-523, August 1977.
7. Schreiner, Max. Research Guidance Studies to Access Gasoline from Coal
by Methanol-to-Gasoline and Sasol-Type Fischer-Tropsch Technologies,
Mobil Research and Development Corporation, FE-2447-13, August 1973.
8. Badger Plants Incorporated. Conceptual Design of a Coal-to-Methanol-to-
Gasoline Commercial Plant. Volume I, FE-2416-43, March 1979.
9. Fischer-Tropsch Design Project Capital Cost Validation, U.S. Army
Engineer Division, Huntsville, Alabama. FE-1759-2, October 1977.
10. Preliminary Economic Analysis of Lurgi Plant Producing 250 Million SCFD
Gas from New Mexico Coal, U.S. Department of the Interior, Bureau of
Mines, Morgantown, West Virginia, ERDA Dcoument No. 76-5, March 1976.
A-9
-------
Section B
Rectisol AGR
APPENDIX B
RECTISOL ACID GAS REMOVAL PROCESS
B.I PROCESS DESCRIPTION
Rectisol is an acid gas removal process which removes carbon dioxide,
hydrogen sulfide, carbonyl sulfide, organic sulfur compounds, hydrogen cyanide,
ammonia, benzene, and gum-forming hydrocarbons from synthesis gases by means
of physical absorption in an organic solvent (especially cold methanol) at
temperatures below 273K. Operation is based upon the fact that these com-
pounds, particularly the reduced sulfur species and carbon dioxide, are very
soluble at high pressure in cold methanol and are readily recoverable by
flash desorption. This is demonstrated in Figure B-l, which presents carbon
dioxide solubility as a function of partial pressure (1). Consider, for
example, the absorption of carbon dioxide at a partial pressure of 1.0 MPa.
From Figure B-l it is evident that at least 90 percent of the dissolved
carbon dioxide may be desorbed by isothermal flashing at methanol tempera-
tures of 258K or lower.
Solubility data for compounds at a partial pressure of 0.1 MPa over
methanol are presented in Figure B-2 (2). It should be noted that gas solu-
bilities generally increase with increasing partial pressure but that solu-
bility coefficients (the ratio of solubility to partial pressure) do not
increase appreciably with pressure until partial pressures exceed 0.1 to 0.2
MPa. Solubility coefficients of hydrogen sulfide and carbon dioxide are
seen to increase substantially with decreasing temperature while those of
major product gases such as hydrogen, carbon monoxide, and methane are rela-
tively temperature independent. For this reason, Rectisol absorption columns
operate at low temperatures, typically in the range of 253 to 213K (1,3,4).
Low temperature operation also reduces solvent losses by reducing the partial
pressure of methanol in the product streams.
B-l
-------
50 100 150
SOLUBILITY OF CO2, VOL/VOL
Figure B-l. Effect of partial pressure on solubility of carbon dioxide in
methanol (1)
B-2
-------
SOLUBILITY COEFFICIENT (X) AT ONE ATMOSPHERE PARTIAL PRESSURE
[kmol OF DISSOLVED GAS/(Mg OF SOLVENT x MPa PARTIAL PRESSURE OF GAS)]
in
c
CD
CD
I
ro
co
o
GO
I
CO
o
-h
&>
CO
(15
CU
3
O
ro
-------
Appendix B
Rectisol AGR
Because the solubilities of reduced sulfur species (e.g., hydrogen sul-
fide and carbonyl sulfide) in methanol are substantially greater than that of
carbon dioxide at the same partial pressure, the Rectisol process is capable
of selective recovery of reduced sulfur species versus carbon dioxide; to some
degree, this holds for all physical absorption solvents capable of absorbing
reduced sulfur species and carbon dioxide almost independently.
The Rectisol process was jointly developed by Linde Aktiengesellschaft
(Munich, Germany) and Lurgi Mineral61technik (Frankfurt, Germany) and is
currently licensed by both companies. It is also available through their U.S.
subsidiaries, Lotepro Corp. (New York, NY) and Lurgi Corp. (River Edge, NJ),
respectively. The Gelleschaft fur Kohle Technologie (GKT, E'.ssen, Germany)
also has a limited Rectisol license applying to Koppers-Totzek (K-T) gasifica-
tion facilities.
Selective Rectisol Process Configurations
A variety of selective Rectisol units are currently being used in appli-
cations such as ammonia and methanol synthesis, medium-Btu gas synthesis,
natural gas purification, and refinery hydrogen production. Although selec-
tive Rectisol designs are site- and process-specific, common key features in-
clude low temperature operation, sequential hydrogen sulfide-carbon dioxide
absorption, discrete methanol regeneration columns for hydrogen sulfide and
carbon dioxide recovery, and separation of methanol and water by distillation.
However, there are significant differences among the designs in use which
relate to both the feed gas composition and the product specifications.
Examples of selective Rectisol process configurations used in coal
gasification applications are presented in Figures B-3 and B-4. The process
presented in Figure B-3 is used by AECI limited in Modderforitein, Republic of
South Africa, and desulfurizes an essentially hydrocarbon-free quenched K-T
gas prior to carbon monoxide shift conversion and subsequent carbon dioxide
removal (5,6). Methanol is added to the feed gas prior to cooling and hydrogen
B-4
-------
GO
I
C.1
HjS-RCCH GAS
REGENERATION
COLUMN
(C02I
CO2 RICH GAS
Figure B-3. Process flow diagram of the Modderfontein selective Rectisol system (5,6)
-------
CRUDE
PRODUCT
GAS FROM
GAS
PRODUCTION
SECTION
PRODUCT GAS
CLEAN PRODUCT
* GAS TO GAS
DISTRIBUTION
PRODUCT GAS
GAS
CONDENSATE
TO MEDIUM
OIL SEPARATOR
IN TAR OIL
SEPARATION
SECTION
«Q
)LER
I
METHANOL
LIQUOR
«-|HO
PURE PRODUCT
GAS TO AMMONIA
PLANT
WASTE '
GAS
HjS RICH WASTE
GAS TO BUHNER
CO, RICH WASTE GAS
. < A I * CO2RICH WASTE
GAS VENT
CYANIC WATER
TO TAR Oil
SEPARATION
SECTION V-5J
Figure B-4. Process flow diagram of the Kosovo selective Rectisol section (7,8)
-------
Appendix B
Rectisol AGR
sulfide absorption to prevent icing. Moisture in the feed gas is removed
from the hydrogen sulfide absorber in solution with methanol, which is re-
covered by distillation. Hydrogen sulfide and carbonyl sulfide are absorbed
from the feed gas using sulfur-free methanol from the carbon dioxide regenera-
tion column. Rich methanol from the hydrogen sulfide absorber is partially
flashed to liberate absorbed hydrogen and carbon monoxide which is compressed
and combined with the cold feed gas. Additional flashing and stripping in
the concentration column, with re-absorption of reduced sulfur species in
sulfur-free methanol, produces a sulfur-rich methanol stream for hot regen-
eration and a carbon dioxide off-gas. Hydrogen sulfide is recovered by
stripping with methanol vapor in the regeneration column.
Carbon dioxide is removed from shifted process gas by absorption in
regenerated methanol. Methanol is added to the shift gas prior to cooling
and carbon dioxide absoprtion to prevent icing, and moisture in the shift gas
is removed from the carbon dioxide absorber in solution with methanol. Rich
methanol from the carbon dioxide absorber is partially flashed to recover
absorbed hydrogen which is compressed and combined with the cold feed gas to
the hydrogen sulfide absorber. Carbon dioxide is recovered by flashing and
stripping with nitrogen in the carbon dioxide regeneration column.
It should be noted that desulfurization prior to shift conversion enables
the use of conventional shift catalysts (e.g., iron-chromium and copper-zinc)
and facilitates process selectivity by absorbing hydrogen sulfide in the or?-,
ence of a minimum of carbon dioxide (approximately 10 to 12% by volume for
K-T coal gasification, 18 to 20% for Texaco coal gasification, and 5 to 6%
for gas produced by partial oxidation of oil). However, in conjunction with
partial oxidation of liquid hydrocarbons for hydrogen or ammonia production,
shift conversion employing sulfur tolerant cobalt-molybdate shift catalysts
often precedes acid gas removal. Selective Rectisol configurations for such
systems are similar to that presented in Figure B-3 except that no gas pro-
cessing occurs between hydrogen sulfide absorption and carbon dioxide
B-7
-------
Appendix B
Rectisol AGR
absorption. Shift conversion prior to acid gas removal results in an in-
creased concentration of carbon dioxide in the hydrogen sulfide absorber feed
gas (up to about 42 percent by volume). Owing to the less favorable carbon
dioxide to hydrogen sulfide ratio after shift conversion, a greater degree
of methanol enrichment is required to achieve the same selectivity attainable
with an unshifted feed gas.
The process presented in Figure B-4 is used at the Kosovo Gasification
Plant near Pristina, Yugoslavia for the production of medium Btu fuel gas and
hydrogen for ammonia synthesis (7,8). Feed gas to the Rectisol unit is generated
by gasification of lignite in oxygen-blown Lurgi gasifiers. Cooled crude
product gas from gasification is further cooled by sequential washing with
cold water and methanol in the two stage cooler. Condensed gas liquor from
the water wash section is flashed to liberate dissolved sour gases, and the
organic phase is recovered from wash water in the naphtha separator. Con-
densed gas liquor from the cold methanol wash section is flashed, and
methanol and condensed moisture are recovered from the naphtha phase by
extraction with water. Dissolved organics in the aqueous phase are recovered
by distillation. Naphtha from the naphtha separator and the naphtha/methanol/
water extractor is sent to by-product storage via the naphtha surge tank.
Cyanic water from naphtha separation and methanol/water distillation is sent
to tar/oil separation.
Product gas from the two stage cooler is scrubbed with carbon dioxide-rich
methanol in the hydrogen sulfide absorber for bulk removal of reduced sulfur
species. Carbon dioxide is removed from the first absorber top gas in two
carbon dioxide absorbers. Bulk carbon dioxide removal is achieved in the
first absorber by washing with carbon dioxide-lean methanol and regenerated
methanol. Overhead gas from the first carbon dioxide absorber is fed directly
into the fuel gas distribution system. When a higher purity gas is required
for feed to the cryogenic hydrogen separation unit, additional carbon dioxide
removal is achieved in the second carbon dioxide absorber using regenerated
methanol.
B-8
-------
Appendix B
Rectisol AGR
Hydrogen sulfide-rich methanol is regenerated by multi-stage flashing
in the hydrogen sulfide flash tower, and steam stripping in the methanol
regeneration column. Hydrogen sulfide-rich waste gas from methanol regenera-
tion is combined with flash gas from the naphtha separator and the methanol
pre-wash flash tank prior to disposal. Carbon dioxide-rich methanol is regen-
erated by multi-stage flashing and nitrogen stripping in the carbon dioxide
flash tower.
Based upon publicly available data, it is not known how the Kosovo
Rectisol design compares with other selective Rectisol units currently pro-
cessing Lurgi crude gas. Several selective Rectisol designs have been pre-
pared for proposed Lurgi gasification facilities in the United States (e.g.,
facilities for Wesco, El Paso Natural Gas Co., Hampshire Energy Co., and Nakota
Co.) (9). However, data with respect to process configuration are generally
proprietary.
Configurations of the two units presented in Figures B-3 and B-4 differ
in several respects. Principal differences result from (1) the fact that
Lurgi crude gas contains significant levels of condensible hydrocarbons
(approximately 0.01 kg C5+ aliphatics, benzene, toluene, and other aromatics
per kg MAP coal) which must be removed prior to acid gas removal (10), (2) the
need for two-stage acid gas removal if sulfur intolerant catalysts are used
for shift conversion, and (3) the fact that at Kosovo all hydrogen sulfide
containing offgases are simply burned so that high sulfide concentrations are
not necessary as would be the case for Claus processing. Differences in the
performance of these two units are detailed in Section B.3.
Non-Selective Rectisol Process Configurations
Non-selective Rectisol processes differ from selective processes in that
all acid gas constituents are absorbed simultaneously and no carbon dioxide
regenerator or reabsorber is used to produce high pruity carbon dioxide vent
gas. An example of a commercial non-selective Rectisol unit is presented in
B-9
-------
Appendix B
Rectisol AGR
Figure B-5 which is a simplified schematic of the South African Oil, Coal and
Gas Corporation's Sasol I acid gas removal system (1). Feed gas to acid gas
removal is crude or partially shifted Lurgi gas from Fischer-Tropsch synthe-
sis. The feed gas is split into three streams which are cooled in each of
two stages by refrigeration, heat exchange with cold high pressure flash gas
(including carbon dioxide-rich flash gases above 100 kPa), and heat exchange
with cold product gas. Condensed moisture and hydrocarbons are recovered
from the combined feed gas following the first cooling stage, and methanol
is added to prevent icing in the second gas cooling stage. Following the
second gas cooling stage, the condensed gas liquor is recovered from the coal
gas and sent to the naphtha separator for by-product and methanol recovery.
Cooled gas is washed with cold methanol in three consecutive stages.
In the first absorption or prewash stage, the cooled gas is washed with
flashed methanol from the expansion tower to remove the final traces of
condensible organics along with some hydrogen sulfide, carbon dioxide, and
organic sulfur compounds. Rich methanol from the first stage absorber is com-
bined with gas liquor from the second gas cooling stage and sent to the naphtha
separator. Separator feed is flashed and extracted with water to yield an
aqueous methanol phase and a by-product naphtha phase containing organic sul-
ful compounds. Methanol is recovered from the aqueous phase by distillation.
Bulk acid gas removal is achieved in the second or main wash stage or
absorption by washing with flashed methanol from the expansion tower. Rich
methanol from the second stage absorber is regenerated along with the methanol/
water still overhead in an expansion tower. Regeneration is by pressure
reduction in six stages to a final pressure of about 30 KPa,. High pressure
flash gas consisting primarily of carbon dioxide, carbon monoxide, and hydro-
gen is used to cool the Rectisol feed gas and then used as on-site fuel gas.
Low pressure flash gas is compressed and flared.
The third or finewash stage absorber effects final gas purification by
washing the second stage absorber effluent gas with completely stripped
B-10
-------
METHANOL
NH
FEED GAS
HIGH PRESSURE
FLASH GAS
PRODUCT GAS
CD
I
ONO
GE
1ST
AGE
r^
THIRD
ABSORBER
HOT
REGENER-
ATION
ATMOSPHERIC
' FLASH GAS
HOT
REGENERATOR
CYANIC WATER
LOW PRESSURE
FLASH GAS
Figure B-5. Process flow diagram of the Sasol I non-selective Rectisol section (1)
-------
Appendix B
Rectisol AGR
methanol from the hot regenerator. Rich methanol from the third stage
absorber is partly regenerated by flashing to atmospheric pressure and then
completely stripped of acid gas in a distillation column. Atmospheric flash
gas from the hot regenerator is released for incineration. Cold product gas
is used to precool the Rectisol feed gas and then sent to liquid synthesis.
Based upon publicly available data, it is not known how the Sasol I
Rectisol design compares with other commercial non-selective Rectisol processes,
although a similar design has been used in the Sasol II facility which was
commissioned in 1980 (11). Non-selective Rectisol designs have been prepared
for several proposed Lurgi gasification facilities in the United States in-
cluding those proposed by Great Plains Gasification Associates (currently under
construction), Wycoalgas Inc., Tenneco Coal Gasification, and El Paso Natural
Gas Co. (9,12). A schematic of the Great Plains non-selective Rectisol section
is presented in Figure B-6 (13). This schematic indicates a similar config-
uration to that of the Sasol I facility but includes details such as the pre-
wash flash vessel and the azeotrope distillation column which are not included
in Figure B-5.
B.2 PROCESS APPLICABILITY
The Rectisol process is used in three typical applications: 1) Removal
of carbon dioxide, hydrogen sulfide, carbonyl sulfide, organic sulfur compounds,
hydrogen cyanide, ammonia, benzene, and gum-forming hydrocarbons from crude
gas produced by coal gasification for syngas and SNG production; 2) Removal
of hydrogen sulfide, carbonyl sulfide, and carbon dioxide from gas produced by
partial oxidation for syngas or hydrogen production; 3) used in conjunction
with low temperature liquefaction and fractionation plants for removal of
acidic components present at moderate levels. Process limitations in these
applications primarily relate to requirements for high pressure, low tempera-
ture operation and methanol contamination by minor constituents present in
the feed gas.
B-12
-------
DO
I
CO
Figure B-6. Process flow diagram of the Great Plains non-selective Rectisol section (13)
-------
Appendix B
Rectisol AGR
As with any other physical absorption process, the minimum circulation
rate of solvent required for complete removal of a gaseous constituent is
inversely proportional to the partial pressure of the constituent in the feed
gas and to the solubility coefficient for the constituent in the solvent used.
Process economics depend mainly upon the solvent circulation rate because the
circulation rate influences the size of all equipment and, therefore, the
capital costs. Solvent circulation rate also affects the operating costs
since pumping costs are proportional to circulation rate and regeneration costs
are nearly proportional to the circulation rate (14). Therefore, the economics
of physical absorption processes improves with increasing acid gas partial
pressures. Physical solvent type acid gas removal processes are typically
selected when acid gas partial pressures are greater than about 1.0 to 1.4 MPa
(1, 15). Feed acid gas partial pressures at existing Rectisol units in coal
gasification and partial oxidation applications are in the range of 0.4 to
2.6 MPa (3,5,6).
As indicated in Figure B-2, the solubilities of most gases of interest
increase with decreasing methanol temperature. Thus, for reasons mentioned
above, Rectisol economics improve with decreasing methanol temperature.
Rectisol absorption columns operate at low temperatures, typically in the range
of 253 to 213K (1,3,4 ). An additional benefit of low temperature operation
is the attendant reduction of methanol losses. Vapor pressure data for
methanol are presented in Figure B-7 (1). These data indicate that methanol
losses can be decreased by a factor of about three to four for each 20K
temperature reduction down to 253K and by about one order of magnitude for
each 20K temperature reduction below 253K.
Minor constituents such as ammonia, hydrogen cyanide, and nitrogen oxides
which may be present in the Rectisol feed gas can complicate operation or
result in fouling. Ammonia and hydrogen cyanide, which are very soluble in
methanol, make the regeneration process more complicated and result in
additional steam requirements (2). Further, the presence of ammonia and
B-14
-------
1000
100
oc
o
flc
ui
2
a. 10
O
E
LU
QC
I
LU
IT
Q.
QC
O
_1
o
<
I
1.0
0.10
0.01
193 213 233 253 273
TEMPERATURE(K)
293
313
333
Figure B-7. Vapor pressure of methanol (1)
B-15
-------
Appendix B
Rectisol AGR
hydrogen cyanide in the hydrogen sulfide fraction is not desirable due to the
potential for adverse reactions during subsequent sulfur recovery. These
contaminants may be removed from the feed gas by employing a prewash of either
cold water or methanol . This prewash also provides feed gas drying (partic-
ularly the methanol prewash) and, in low temperature gasification applica-
tions, removes condensible hydrocarbons.
One coal gasification facility has reported Rectisol fouling which is
attributed to the presence of oxygen and nitrogen oxides in the Rectisol feed
gas (16,21). Oxygen in the Rectisol feed gas results in oxidation of a por-
tion of the hydrogen sulfide to elemental sulfur. The presence of nitric
oxide in addition to oxygen accelerates the rate of sulfide oxidation.
Deposits of sulfur in columns resulted in reduced solvent circulation rates,
and fouling of heat exchangers resulted in insufficient cooling capability
to achieve the required degree of gas purification.
It has been determined that this fouling can be reduced by allowing low
levels of hydrogen cyanide and ammonia to enter the Rectisol unit to solubi-
lize sulfur by formation of ammonium thiocyanate which is ultimately removed
with the methanol/water distillation bottoms. When insufficient hydrogen
cyanide is present in the feed gas, sodium cyanide solution is injected into
the methanol. A more fundamental solution which has been implemented is the
hydrogenation of oxygen and nitrogen oxides over a cobalt molybdate catalyst
upstream of the Rectisol unit. Formation of elemental sulfur and the associ-
ated fouling of the Rectisol unit have not occurred since installation of the
catalytic hydrogenation unit (21).
B.3 PROCESS PERFORMANCE
Depending upon the product requirements and other site specific con-
straints, the Rectisol process can be designed to yield a product gas contain-
ing less than 0.1 ppmv total sulfur and less than 10 ppmv carbon dioxide. The
carbon dioxide content achievable in the purified gas is independent of the
B-16
-------
Appendix B
Rectisol AGR
type of Rectisol process employed (e.g., selective or non-selective Rectisol).
However, in the case of a non-selective Rectisol process, the utilities (steam,
cooling water, and refrigerant) would increase to obtain a product gas with
ppmv levels of carbon dioxide. Publicly available data indicate that in
gasification applications involving an essentially hydrocarbon-free feed gas,
selective Rectisol processes can produce a sulfur-rich offgas containing 25-
75% hydrogen sulfide and a carbon dioxide-rich offgas containing less than
10 ppmv total sulfur. The presence of moderate quantities of hydrocarbons
in the feed gas (9 to 16 percent) has no influence on the selectivity of
hydrogen sulfide recovery; hydrogen sulfide concentrations of 25 to 35 per-
cent in the hydrogen sulfide-rich offgas can be achieved along with a carbon
dioxide-rich offgas containing 10 ppmv total sulfur. However, C3 and C4
hydrocarbons present in the feed gas will tend to concentrate in the hydrogen
sulfide-rich offgas.
Performance data for selective Rectisol units treating essentially hydro-
carbon free feed gases are summarized in Table B-l. Plants 1 and 2 produce
hydrogen and ammonia synthesis gas, respectively, by partial oxidation of oil.
These plants utilize sulfur tolerant shift conversion catalysts which enable
shift conversion prior to acid gas removal. Therefore, the feed gases to
Plants 1 and 2 contain 31 to 35% carbon dioxide, 62-64% hydrogen, and less
than about 5% carbon monoxide (2,3,6,17). Plant 3 is a coal gasification
facility producing ammonia synthesis gas. This plant employs a two-stage
Rectisol system which removes sulfur species prior to shift conversion and
removes carbon dioxide subsequent to shift conversion (refer to Figure B-3 for
example process flow diagram). Feed gas to the Plant 3 sulfur absorber there-
fore contains only 12-13% carbon dioxide, 27-29% hydrogen and about 57% carbon
monoxide (2,6). Feed gas to the carbon dioxide absorber in Plant 3, which is
not included in Table B-l, contains 42-43% carbon dioxide, 53-54% hydrogen,
and about 3% carbon monoxide.
B-17
-------
TABLE B-l. SELECTIVE RECTISOL PERFORMANCE DATA FOR HIGH TEMPERATURE GASIFICATION APPLICATIONS*
CD
1
00
Gas Feed
jComponent Plant 1
H? 62.35-
63.74
N,+Ar 0.12 -
0.52
CO 3.24-
4.13
CH. 0.13-
0.17
CO,, 31.62-
33.23
H?S 0.26-
0.49
COS 10-63ppm
Flow Rate, 3562-
kmol/hr 3992
Temperature, K 303-313
Pressure, HPa 3.2-3.3
Gas, Mole I
Plant 2 Plant 3
61.59 27.5-
29.3
0.41 1.52
2.60 56.62
0.33 0.10
34.55 11.8-
13.3
0.52 0.59-
0.75
0.10
6112 4691-
4801
311
7.3 3.0-3.1
Puri
Plant 1
93.58-
94.08
0.17 -
0.82
4.86
0.19
0.24
slOppm
-------
Appendix B
Rectisol AGR
These selective Rectisol units are seen to perform similarly in most
respects over a wide range of operating pressures although there is a sub-
stantial range in the concentration of hydrogen sulfide, 25-72%, in the
sulfur-rich waste gas from Plant 3. Lotepro Corporation has indicated that
the higher hydrogen sulfide concentration is attainable at the expense of
higher refrigeration and stripping gas requirements (6). The amount of strip-
ping gas is a function of the hydrogen sulfide concentration desired in the
hydrogen sulfide-rich offgas, the type of Rectisol process, the feed gas
pressure, and the carbon dioxide, hydrogen sulfide, and carbonyl sulfide con-
centrations in the feed gas. Under given conditions an increase in stripping
gas of about 60% is necessary to increase the hydrogen sulfide concentration
from 25-70%.
Performance data for the Kosovo selective Rectisol unit (taken at partial
load and not fully representative of normal performance), which treats crude
gas from Lurgi gasification, are summarized in Table B-2 (refer to Figure B-4
for the process flow diagram). Data were obtained during three sampling
campaigns in the period of September 1977 to November 1978. Tabulated data
represent the best overall data obtained during these tests, and the ranges of
the available data. As indicated previously, the presence of moderate levels
of hydrocarbons in the feed gas has no influence on process selectivity.
Differences in process selectivities indicated in Tables B-l and B-2 primarily
reflect differences in process requirements. At the Kosovo facility, the
sulfur-containing gases are burned, and, therefore, high sulfur concentrations
in these offgases are not necessary. Thus, unlike the facilities cited in
Table B-l, the Kosovo facility does not utilize an enrichment stage. Also,
Kosovo's hydrogen and methane rich flash gases from the carbon dioxide and
hydrogen sulfide loaded methanol streams are added to the hydrogen sulfide
fraction rather than being recycled to the feed gas.
Available performance data for the Sasol I non-selective Rectisol unit,
which also treats crude gas from Lurgi gasification, are presented in Table
B-19
-------
TABLE B-2. SELECTIVE RECTISOL PERFORMANCE DATA FOR LURGI (KOSOVO) MEDIUM-BTU GASIFICATION (7,18)*
TO
I
ro
o
Gas Component
H2
°2
N2
CO
co2
CH4
C2H6
C2H4
C3
C4
C5
C6
Benzene
Tol uene
Xylene and
Ethylbenzene
Phenols
H2S
COS
CH3SH
C2H5SH
NH,
HCN
pH
Total Solids, mg/L
Total Nonvolatile
Solids, mg/L
Total Suspended
Solids, mg/L
Total Dissolved
Solids, mg/L
COD (as mg02/L)
Permanganate
(as mg02/L)
Total Sulfur, mg/L
Flow Rate,
kmol/hr
Flow Rate, m/hr
Temperature, K
Crude Product Gas
(Stream 73), mole
yalue+ Range
38 1
0.36
0 64
15
32
11.5
0 47
0.04
0 19
0.074
0.044
0.064
750ppmv
230ppmv
lOOppmv
-Ippmv
0.60
97ppmv
590ppmv
200ppmv
3.3ppnv
320ppmv
703
295
36-46
0.09-2.6
0.04-1.6
9 6-17
21-40
8 9-14.5
-1 ppmv-0. 76
-lppmv-0 11
0.07-0 40
0.02-0.24
0.01-0 06
0.02-0.20
660-840ppmv
200-260ppmv
16-1 TOppmv
--
0.44-0.78
63-1 20ppmv
460-700ppmv
98-270ppmv
• l-3.3ppmv
60-320ppmv
293-295
Clean Product Gas
(Stream 7 4) , mole
Value+ Ranqe
60 59-67
0.44 0 1-1.7
0.38 0.32-6 8
22 13-23
0.02 ..0.01-2 4
16 12-18
0 15 ~lppmv-0.18
—Ippmv
-Ippmv -lppmv-0. 09
—Ippmv
— Ippinv
0.03 • lppmv-0. 03
--
..
-Ippmv
•-Ippmv
0 1 7ppmv • 1 -0 2ppmv
1 1 ppmv -1- 1 . 9ppmv
l.Oppmv — l-1.7ppmv
-Ippmv < 1 - 30ppmv
--
491
._
C02-Rich Waste Gas
(Stream 7.2), mole
Value^ Ranqe
-0 01
-0 01
-0.01
-0.01
94
1.2
1 6
—1 ppmv
0 28
—1 ppmv
—1 ppmv
Ippmv
1 ppmv
-Ippmv
-Ippmv
1 ppmv
39ppmv
62ppnv
8. 5ppmv
4 4ppmv
4 6ppmv
1 3ppmv
90
292
••0 01-0.8
-0 01-1.8
-0.01-48
•0.01- 0.01
91-95
0.6-1 8
0 29-1 6
--
0 17-0 55
•lppmv-0 23
lppn,v-0 17
--
--
--
_.
--
20-90ppmv
-1 -62ppmv
8.2-9.7ppmv
3 4-6. Ippmv
1 -4.6ppmv
H^S-Rich Waste Gas Cyanic Water
(Stream 7.1), mole (Stream 7.5)
Value+ Ranqe ValueJ Ranqe
0.11
-0 01
-0.01
1 1
88
4 3
0.82
—1 ppmv
0.63
0 32
0.04
0.21
1 lOppmv
8ppmv
< Ippmv
-1 pnmv
4.54
420ppmv
0.21
780ppmv
0.22
200ppmv
102
285
0.02-0.11
-0.01-0 51
-0.01-3.2
1 1-3 5
85-92
4.1-4.7
0 34-0.97
--
0.22-1.1
0.14-0.58
0.03-0.21
0.01-0.22
40-110ppmv
4-8ppmv
__
--
1.6-5.0
360-540ppmv
0.19-0.48
670-850ppmv
• lppmv-0 22
83-200ppmv
11.9 11 4-12 1
7.30
450
140
590
205
570
60 52-68
0.8
353
Refer to Figure 8-4 for process flow diagram.
Values are best overall values from available data .
-------
Appendix B
Rectisol AGR
B-3 (refer to Figure B-5 for the process flow diagram). As initially designed,
the high pressure flash gas is used as an on-site fuel gas, the low pressure
flash gas is flared, and the atmospheric flash gas is vented to the atmosphere
through the power stack (19). More recently, a Stretford unit was designed
to treat the atmospheric flash gas which contains about 90% of the sulfur
species absorbed (17,19). Proposed designs for U.S. facilities indicate
that at least a portion of the high pressure flash gas stream is recycled to
the gasification plant for recovery of carbon monoxide, hydrogen, and methane;
some fraction of the high pressure flash gas may be combined with the other
waste gases for sulfur recovery (12,20). Therefore, the performance indicated
in Table B-3 may require some adjustment.
B.4 RECTISOL WASTE STREAMS
Secondary waste streams produced by the Rectisol acid gas removal pro-
cess are: (1) hydrogen sulfide-rich offgases, (2) carbon dioxide-rich off-
gases (selective Rectisol processes only), and (3) methanol/water distilla-
tion bottoms. Available characterization data for the offgas streams have
been summarized in Section B.3 for each of the three basic Rectisol process
configurations. The sulfur-rich offgas is typically sent to the sulfur re-
covery unit, either Claus or Stretford, or flared. When the Rectisol process
is used in conjunction with low temperature coal gasification systems (e.g.,
Lurgi gasifiers) the Rectisol feed gas contains significant concentrations
of C~ hydrocarbons relative to the concentration of hydrogen sulfide. The
naphtha fraction is recovered from the feed gas by washing prior to acid gas
removal. Lighter hydrocarbons largely pass through the prewash and are, to
some extent, absorbed with the acid gases. These light hydrocarbons, partic-
ularly the 035 and 045, tend to concentrate in the hydrogen sulfide-rich
offgas and may also be present in the carbon dioxide-rich offgas. Therefore,
unless special precautions are taken, high levels of these C, and C, hydro-
carbons in the Rectisol feed gas may result in off-color sulfur if Claus
sulfur recovery is employed or excessive tail gas hydrocarbon emissions if
B-21
-------
TABLE B-3. PERFORMANCE DATA FOR THE NON-SELECTIVE RECTISOL AT SASOL I (LURGI GASIFICATION)(17)*
CO
ro
ro
Gas Rectisol Feed
Component Gas, Mole %
H
CO
CH4
co2
N2+Ar
COS
cs2
RSH
Thiophene
Total sulfur
c+
L2
Flow Rate, NM3/hr
Temperature, K
Pressure, MPa
40.05
20.20
8.84
28.78
1.59
0.30
lOppmv
NA
20ppmv
NA
NA
0.54
381 ,000
303
2.6
Product
Gas, Mole %
57.30
28.40
11.38
0.93
1.77
ND
NA
NA
NA
NA
0.04ppmv
263,000
288
2.4
~F*> r> : — —
High Pressure
Flash Gas
21.4
18.2
11.4
46.7
1.5
0.32
NA
NA
NA
NA
NA
0.7
4,600
273
1.3
Off -Gases, Mole
Low Pressure
Flash Gas
2.6
4.8
7.2
83.4
0.8
0.49
NA
NA
NA
NA
NA
1.1
15,000
273
0.48
Of
h
Atmospheric
Flash Gas
0.14
0.0
0.9
97.2
0.03
0.88
30ppmv
2ppmv
280ppmv
2ppmv
NA
0.7
98,000
268
0.11
-------
Appendix B
Rectisol AGR
Stretford sulfur recovery is employed. An approach proposed in conjunction
with Wesco and Hampshire Energy Co. selective Rectisol units involves the use
of an amine unit (ADIP) to separate hydrocarbons from the Claus feed gas (9,20)
Carbon dioxide-rich offgas from selective Rectisol units is either sold
as by-product or vented to the atmosphere at existing facilities. As dis-
cussed above, light hydrocarbons present in the Rectisol feed gas are co-
absorbed to some extent with the acid gases and may be present in the carbon
dioxide-rich offgas. Further, steps taken within the Rectisol process to
minimize hydrocarbon levels in the hydrogen sulfide fraction will likely
result in increased hydrocarbon levels in the carbon dioxide offgases.
Similarly, carbon monoxide is coabsorbed and will be present in the carbon
dioxide-rich offgas due to its low solubility in methanol. Of course the
extent of carbon monoxide coabsorption, and therefore its potential concen-
tration in the carbon dioxide-rich offgas, depends upon its partial pressure.
Thus, for similar acid gas removal systems, processes requiring only partial
shift conversion (e.g., SNG, methanol, or acetic acid syntheses) would be
prone to higher concentrations of carbon monoxide in the carbon dioxide-rich
offgas. Therefore, proposed designs in Lurgi-based coal gasification appli-
cations indicate either incineration of the carbon dioxide-rich offgas for
control of hydrocarbon and carbon monoxide emissions, or sale of the offgas
as by-product; direct discharge to the atmosphere is not proposed. Also, at
least one non-Lurgi coal gasification plant currently under construction,
the Tennessee Eastman Kingport, Tennessee Texaco gasification project, pro-
poses catalytic incineration of a carbon monoxide enriched portion of the
carbon dioxide offgas for control of carbon monoxide emissions (21).
Publicly available characterization data for the methanol/water distilla-
tion bottoms are extremely limited. This is apparently due to the fact that
the size of the still bottoms stream is generally quite small relative to
other wastewater streams requiring similar wastewater treatment (e.g., gas
liquor and synthesis condensates). Thus, from an operational standpoint,
B-23
-------
Appendix B
Rectlsol AGR
the still bottoms are likely to be of minor significance other than for
checking still operation and methanol losses. One set of data, provided by
Sasol personnel (19), are presented in Table B-4. At the Sasol facility,
this waste stream is sent directly to biological treatment where it comprises
less than 2% of the feed to this system.
TABLE B-4. CHARACTERIZATION DATA FOR METHANOL/WATER DISTILLATION BOTTOMS AT
SASOL (19)
Parameter/Component Value
PH 9.7
Phenol, mg/L 18
Cyanides (as CN), mg/L 10.4
(includes thiocyanate)
Ammonia (as N), mg/L 42
Sulfides (as S) Trace
COD, mg/L 1,686
B.5 PROCESS RELIABILITY
The original Lurgi non-selective Rectisol unit built at Sasol in 1955 has
operated with an on-stream factor of about 97% (17). Normal maintenance in-
cludes partial shutdown about once per year for cleaning of critical equipment
and complete shutdown every two years during the normal plant downtime. Major
upsets in the Rectisol unit requiring process adjustments rarely occur (19).
As discussed in Section B.2, plugging problems in a two-stage selective
Rectisol unit at a coal gasification facility have been reported (16). This
problem has been attributed to deposition of elemental sulfur resulting from
the presence of nitrogen oxides in the Rectisol feed gas. Fouling was at
least partially controlled by allowing low levels of hydrogen cyanide and
ammonia to enter the Rectisol unit to solubilize sulfur by formation of
B-24
-------
Appendix B
Rectisol AGR
ammonium thiocyanate. A more fundamental solution is the hydrolysis of
nitrogen oxides and oxygen over a cobalt molybdate catalyst ahead of the
Rectisol unit. Detailed operating data from this facility are not available.
B.6 PROCESS ECONOMICS
Available capital costs and utility requirements for the Rectisol process
are summarized in Tables B-5 and B-6, respectively. Tabulated capital costs
are primarily conceptual design cost estimates while tabulated utility require-
ments are published data for existing units. It should be noted that the cost
of a Rectisol unit is influenced by a variety of considerations including the
feed gas flow rate and pressure, acid gas content, and heavy hydrocarbon
content, and the desired levels of selectivity and product purity. Due to
the number of variables and associated interdependences of these variables
which influence cost, costs of Rectisol systems tend to be highly case
specific.
TABLE B-5. CAPITAL COSTS FOR RECTISOL ACID GAS REMOVAL UNITS
Selective
Non-Selective
Dry
Feed Gas,
kmol/hr
6,100
96,384
52,786
57,574
Total
Pressure,
MPa
7.8
2.9
2.8
2.8
co2,
vol %
35
28.9
31.4
34.2
H2S,
vol %
0.25
0.250
0.135
0.130
Capital
Cost, $106
(adj. to
1980 basis)
13.5*
150. 6T
91.8*
81.9*
Reference
24
25
23
22
*The feed gas to this unit does not contain heavy hydrocarbons. Cost includes
refrigeration unit, erection, and plant startup. This is the same unit which
is identified as plant 5 in Table B-6.
''"Data are based upon a conceptual design cost estimate. Details of the cost
estimate are not available. The feed gas to this unit does not contain
heavy hydrocarbons.
*Data are based upon a conceptual design cost estimate. The feed gas to this
unit contains heavy hydrocarbons. Reported cost includes naphtha and methanol
recovery and erection. It is not specified whether the costs for a refrig-
eration unit and unit startup are included.
B-25
-------
TABLE B-6. UTILITY REQUIREMENTS FOR RECTISOL ACID GAS REMOVAL UNITS
CD
I
no
CTl
Flow Rate, kmol/hr
Pressure, MPa
Electric Power, kW hr/kmol
Low Pressure Steam, MJ/kmol
Cooling Water, MJ/kmol
Stripping Nitrogen, kmol/kmol
Makeup Methanol , kg/kmol
Refrigeration, MJ/kmol
(at 227 to 235K)
• -- -
Plant 1
3692-3992
3.2-3.3
0.14-0.15
5.14-5.54
1.20-1.42
0.067-0.072
Selective Rectisol*
Plant 2
6112
7.3
0.31
3.44
6.43
0.031
0.0085-0.0092 0.0057
2.09-2.29
Included
above in power
and cooling
water
Plant 3
7112
3.0
0.18
4.16
1.92
0.048
0.012
1.90
Plant 4
6350
3.3
0.57
5.09
9.52
0.067
0.0079
Included
above in power
and cool ing
water
Plant 5
eioo
7.8
0.168
2.77
1.91
0.043
0.007
1.24
Non-Selective
Rectisolt
Plant 6
16993
2.6
No data
3.27
0.682
No data
0.013
No data
*Plant 1 is a refinery producing hydrogen by partial oxidation of oil; shift conversion occurs prior to acid
gas removal (2,6).' Refer to Table B-l for performance data.
Plant 2 produces ammonia synthesis gas from crude hydrogen generated by partial oxidation; shift conversion
occurs prior to acid gas removal (3). Operating costs reflect the use of compression refrigeration. Refer
to Table B-l for performance data.
Plant 3 produces ammonia synthesis gas by coal gasification; shift conversion follows hydrogen sulfide
removal but precedes carbon dioxide removal (2,6). After sulfur removal the gas is increased from 3 MPa to
5 MPa by compression; the additional power required for compression is not included in the tabulated electric
power requirement. Tabulated data are based on gas flow rate after shift conversion. Refer to Figure B-3
for process flow diagram, and to Table B-l for performance data.
Plant 4 uses a Rectisol unit for purification of hydrogen from partial oxidation of heavy crude oil; shift
conversion occurs prior to acid gas removal (3).
Plant 5 produces ammonia synthesis gas by partial oxidation of oil; shift conversion occurs prior to acid
gas removal. Approximately 62% of the incoming carbon dioxide is provided as a carbon dioxide fraction con-
taining less than 1.5 ppmv sulfur for urea production (24).
tPlant 6 is the Sasol coal gasification facility (17). Refer to Figure B-5 for process flow diagram, and
Table B-3 for performance data.
-------
Appendix B
Rectisol AGR
B.7 REFERENCES
1. Kohl, A. and F. Reisenfeld. Gas Purification. Gulf Publishing Co.,
Houston, Texas, 1974.
2. Ranke, G. Acid Gas Separation by Rectisol in SNG Processes. Linde AG,
Munich, Germany. Copy of presentation obtained through Lotepro Corpora-
tion, New York, N.Y.
3. Scholz, W.H. Rectisol: A Low-Temperature Scrubbing Process for Gas
Purification, Advances in Cryogenic Engineering, Vol. 15, 1969.
4. Maddox, R.S. Gas and Liquid Sweetening, Campbell Petroleum Series, 1974.
5. Zee, C.A., J. Clausen, and K.W. Crawford. Environmental Assessment:
Source Test and Evaluation Report, Koppers-Totzek Process. EPA-600/7-81
009. January 1981.
6. Lotepro Corporation brochure.
7. Lee, K.W., W.S. Seames, R.V. Collings, K.J. Bombaugh, and G.C. Page.
Environmental Assessment: Source Test and Evaluation Report - Lurgi
(Kosovo) Medium-Btu Gasification, Final Report EPA-600/7-81-142.
August 1981.
8. Salja, B. and M. Mitrovic. Environmental and Engineering Evaluation
of the Kosovo Coal Gasification Plant, Yugoslavia. Symposium Proceedings:
Environmental Aspects of Fuel Conversion Technology, III, September 1977,
Hollywood, Florida. EPA-600/7-78-063. April 1978.
9. Beychok, M.R. and W.J. Rhodes. Comparison of Environmental Design Aspects
of Some Lurgi-Based Synfuels Plants. Symposium on Environmental Aspects
of Fuel Conversion Technology, Denver, Colorado, October 26-30, 1981.
EPA 600/9-82-017, August 1982.
10. Trials of American Coals in a Lurgi Gasifier at Westfield, Scotland.
Woodall-Duckham, Ltd., Sussex, England. ERDA R&D Report No. 105, 1974.
11. Cameron Synthetic Fuels Report. Rocky Mountain Division, The Pace
Company Consultants & Engineers, Inc. Volume 18-Number 4, December 1981.
12. Sinor, J.E. Evaluation of Background Data Relative to New Source Per-
formance Standards for Lurgi Gasification. Cameron Engineers, Inc.
EPA-600/7-77-057, June 1977.
1-27
-------
Appendix B
Rectisol AGR
13. Final Environmental Impact Statement: Great Plains Gasification Project,
Mercer County, North Dakota. Vol I. U.S. Department of Energy, Washing-
ton, D.C. August 1980.
14. Hochgesand, G. Rectisol and Purisol. Industrial and Engineering Chem-
istry, Vol 62, No. 7. July 1970.
15. Fleming, O.K. Acid Gas Removal Systems in Coal Gasification. Ammonia
from Coal Symposium. Tennessee Valley Authority. May 8-10, 1979.
16. Engelbrecht, A.D. and L.J. Partridge. Operating Experience on a 1000-
ton/day Ammonia Plant at Modderfontein. Ammonia from Coal Symposium.
Tennessee Valley Authority. May 8-10, 1979.
17. Control of Emissions from Lurgi Coal Gasification Plants. U.S. Environ-
mental Protection Agency, Emission Standards and Engineering Division.
EPA-450/2-78-012 (OAQPS No. 1.2-093). March 1978.
18. Bombaugh, K.J. and W. E. Corbett. Kosovo Gasification Test Program
Results - Part II, Data Analysis and Interpretation. Symposium on
Environmental Aspects of Fuel Conversion Technology, IV,, Hollywood,
Florida. April 17-20, 1979.
19. Data provided to EPA's Industrial Environmental Research Laboratory,
Research Triangle Park, N.C., by South African Coal, Oil and Gas Ltd.
(Sasol). November 1976.
20. Final Environmental Impact Statement. Western Gasification Company (WESCO)
Coal Gasification Project and Expansion of Navajo Mine by Utah International
Inc., New Mexico. U.S. Department of the Interior-Bureau of Reclamation,
Vol. I, II. January 14, 1976.
21. Review comments provided to TRW by Linde AG, April 1982.
22. Wham, R.M., J.F. Fisher, R.C. Forrester III, A.R.Irvine, R. Salmon,
S.P.N. Singh and W.C. Ulrich. Liquefaction Technology Assessment - Phase
I: Indirect Liquefaction of Coal to Methanol and Gasoline Using Available
Technology. ORNL-5664. Oak Ridge National Laboratory, Oak Ridge, Tenn.
February 1981.
23. Schreiner, Max. Research Guidance Studies to Assess Gasoline from Coal
by Methanol-to-Gasoline and Sasol-Type Fischer-Tropsch Technologies.
Mobil Research and Development Corporation, FE-2447-13. August 1978.
B-28
-------
Appendix B
Rectisol AGR
24. Information provided to TRW by Lotepro Corporation, January 1983.
25. Conceptual Design of a Coal to Methanol Commercial Plant. Volume IVA,
Badger Plants Incorporated, Cambridge, Mass. FE-2416-35 (Vol. 4A).
B-29
------- |