EPA tiOG/2-75-068
Hovember 1975
Environmental Protection Techaoloa Series
ENVIRONMENTAL PROBLEM DEFINITION FOR
PETROLEUM REFINERIES,
SYNTHETIC NATURAL GAS PLANTS. AND
It JITUiAL GAS PLANTS
iesiarcii Laboratory
of Rssaarch and Development
U.S. EmriremRMrtxi Protection Agency
Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development,
U.S. Environmental Protection Agency, have been grouped into
five series. These five broad categories were established to
facilitate further development and application of environmental
technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in
related fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed
to develop and demonstrate instrumentation, equipment and
methodology to repair or prevent environmental degradation from
point and non-point sources of pollution. This work provides the
new or improved technology required for the Control and treatment
of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U. S. Environmental Protection
Agency, and approved for publication. Approval does not signify that
the contents necessarily reflect the views and policies of the Agency, nor
does mention of trade names or commercial products constitute endorse-
ment or recommendation for use.
This document is available to the public through the National
Technical Information Service, Springfield, Virginia 22161.
-------
EPAv600/2-75-068
ENVIRONMENTAL PROBLEM DEFINITION
FOR PETROLEUM REFINERIES,
SYNTHETIC NATURAL GAS PLANTS, AND
LIQUEFIED NATURAL GAS PLANTS
E.G. Cavanaugh, J.D. Colley, P.S. Dzierlenga,
V. M. Felix, D. C. Jones, and T. P. Nelson
Radian Corporation
8500 Shoal Creek Boulevard
Austin, Texas 78766
Contract No. 68-02-1319, Task 18
ROAPNo. 21ADD-042
Program Element No. 1AB013
EPA Task Officer: L. Lorenzi, Jr.
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
November 1975
-------
ABSTRACT
Information in this report was compiled for the
purpose of providing the Environmental Protection Agency with
technical support in the area of total environmental problem
definition for petroleum refineries, synthetic natural gas (SNG)
plants, and liquefied natural gas (LNG) plants.
Process descriptions are presented for each plant.
Where applicable, comparisions to other types of energy con-
version plants are made. Potential ambient air emissions,
liquid effluents, and solid wastes are identified and the
status of monitoring methods and control techniques for these
emissions and wastes are discussed.
The problems involved with the siting of new plants
because of the impact of these emissions and wastes are considered,
Areas where research and development can be usefully applied to
these environmental problems are identified.
111
-------
TABLE OF CONTENTS
1.0
1.1
1.2
1.3
1.3.1
1.3.2
1.3.3
1.3.4
1.4
1.5
2.0
2.1
2.1.1
2.2
2.2.1
2.2.1.1
2.2.1.2
2.2.1.3
2.2.1.4
2.2.2
2.2.2.1
2.2.2.2
2.2.2.3
2.2.2.4
2.2.2.5
2.2.2.6
2.2.3
2.3
2.3.1
INTRODUCTION
Project Objectives
Modular Concept
Summary of Environmental Problems
Fuel Oil Refinery
Gasoline Refinery Module
Liquefied Natural Gas (LNG) Plants: Peak-
Shaving and Base Load
Synthetic Natural Gas (SNG) Plant
Comparison of Module Emissions
Evaluation of Environmental Requirements . . .
PROCESS TECHNOLOGY DESCRIPTION
Petroleum Refining
Refinery Processes
LNG Process Technology Description
LNG Peak-Shaving Plant .
Natural Gas Feed Preparation
Liquefaction Cycles
Storage
Regasification Systems
Base Load Plant
Natural Gas Conditioning and Purification . .
Heavy Hydrocarbon Stripping
Liquefaction Cycles ...
Storage
Transportation
Regasification
Satellite LNG Facilities
SNG Production Technology
Processing Steps
Page
. . 1
. . 1
. . 2
. . 3
. . 3
. . 7
. . 8
. . 9
. . 11
11
. . 14
. . 15
. . 16
. . 52
. . 54
. . 54
. . 57
. . 62
. . 64
. . 68
. . 71
. : 81
. . 84
. . 84
. . 84
. . 87
. . 87
. . 88
. . 88
IV
-------
TABLE OF CONTENTS (Cont.)
Page
2.4 New Technologies 98
2.4.1 Coal Gasification 102
2.4.2 Coal Liquefaction 108
2.4.3 Shale Oil Production Ill
3.0 IDENTIFICATION OF EMISSIONS AND EFFLUENTS ... 116
3.1 Fuel Oil Refinery Module 118
3.1.1 Module Basis 118
3.1.2 Module Description 118
3.1.3 Module Emissions 126
3.1.3.1 Air Emissions 126
3.1.3.2 Water Effluents 133
3.1.3.3 Solid Wastes 133
3.2 Gasoline Refinery Module 136
3.2.1 Module Basis 136
3.2.2 Module Description 136
3.2.3 Module Emissions 146
3.2.3.1 Air Emissions 146
3.2.3.2 Water Effluents 154
3.2.3.3 Solid Wastes 155
3.3 LNG Module 158
3.3.1 Peak-Shaving Module 158
3.3.1.1 Peak-Shaving Module Description 158
3.3.1.2 Peak-Shaving Module Emissions 163
3.3.2 Base Load Module 168
3.3.2.1 Base Load Module Description 168
3.3.2.2 Base Load Module Emissions 173
3.4 SNG Module 177
3.4.1 Module Basis 177
3.4.2 Module Description 177
3.4.3 Module Emissions 183
3.5 Comparison of Module Emissions 188
v
-------
TABLE OF CONTENTS (Cont.)
Page
4.0 MONITORING TECHNOLOGY 190
4.1 Ambient Air Quality Monitoring 192
4.1.1 Background . 192
4.1.2 General Monitoring Considerations 193
4.2 Source Monitoring (Air) 198
4.2.1 Background . 198
4.2.2 General Monitoring Procedures. 201
4.3 Effluent Water Monitoring 202
4.3.1 Background 202
4.3.2 General Water Monitoring 204
4.4 Solid Waste 208
5.0 EMISSION CONTROL METHODS 210
5.1 Air Emission Control 211
5.1.1 Particulates Emission Control 211
5.1.1.1 Sludge Incineration Particulate Control. . . . 211
5.1.1.2 FCCU Particulate Control 218
5.1.2 SO Emission Control 225
X
5.1.2.1 Fuel Desulfurization 226
5.1.2.2 Flue Gas Treating '. 232
5.1.3 NO Emission Control 237
i X
5.1.3.1 Combustion Modifications 238
5.1.3.2 Fuel Modifications 240
5.1.3.3 Design Modifications 240
5.1.3.4 Flue Gas Treating 241
5.1.3.5 NO Emissions From-Incinerators 247
x
5.1.4 CO Emission Control 248
5.1.5 Hydrocarbon Emissions Control 253
5.1.5.1 Carbonate Vent Gas 253
5.1.5.2 Storage Control 254
5.1.5.3 Loading Rack Controls 262
VI
-------
TABLE OF CONTENTS (Cont.)
Page
5.1.5.4 Combustion Source Controls 267
5.1.5.5 Incinerators 267
5.1.5.6 Process Source Controls 269
5.1.5.7 Fugitive Source Controls 273
5.2 Wastewater Treatment 277
5.2.1 Optimization of Energy Use 278
5.2.2 Elimination of Wastewater Sources 280
5.2.3 Segregated Wastewater System 280
5.2.4 Inplant Wastewater Treatment 285
5.2.5 Process Wastewater Treatment 287
5.2.5.1 Pretreatment 291
5.2.5.2 Suspended Solids Removal 294
5.2.5.3 Dissolved Solids Removal 298
5.2.5.4 Tertiary Dissolved Solids Removal or Treatment. 303
5.2.5.5 Final Liquids Disposal 310
5.2.5.6 Sludge Handling 312
5.2.5.7 LNG and SNG Plant Wastewater Treating 321
5.2.6 Summary 324
5.3 Solids Emission Control 325
5.3.1 Sludge Disposal Methods 325
5.3.2 Catalyst Solid Disposal 329
6.0 PLANT IMPACT AND SITING PROBLEMS 331
6.1 Petroleum Refinery Impact 332
6.1.1 Refinery Effluents 333
6.1.1.1 Air Emission Impact 333
6.1.1.2 Water Quality Impact 354
6.1.1.3 Solid Wastes 361
6.1.2 Raw Material Availability 364
6.1.2.1 Feedstock Availability 364
6.1.2.2 Water Availability 370
6.1.3 Product Transportation 375
VII
-------
TABLE OF CONTENTS (Cont.)
Page
6.2 SNG Plant Impact 379
6.2.1 SNG Plant Effluents 379
6.2.1.1 Air Emissions Impact. 380
6.3 LNG Plant Impact 386
6.3.1 LNG Plant Effluents 386
6.3.1.1 Liquefaction Plant 386
6.3.1.2 Regasification Plant 389
6.3.2 Raw Material Availability 389
6.3.2.1 Feedstock Availability 390
6.3.2.2 Water Availability . 390
6.3.3 Product Shipping and Receiving 392
APPENDIX 6.1-1
PRIMARY (SECONDARY) AMBIENT AIR QUALITY STANDARDS
OF ARBITRARILY SELECTED STATES 394
APPENDIX 6.1-2
WATER QUALITY MODELS 402
APPENDIX 6.1-3
MAJOR WATER LAWS 411
APPENDIX 6.1-4
ATMOSPHERIC STABILITY CLASSES AND PLUME DISPERSION
CHARACTERISTICS . . . 422
APPENDIX 6.1-5
ATMOSPHERIC DISPERSION MODELS .... 431
7.0 AREAS FOR RESEARCH AND DEVELOPMENT 439
REFERENCES
viii
-------
LIST OF TABLES
TABLE 1.4-1 COMPARISON OF MODULE EMISSIONS
Page
, 12
TABLE 2.1-1 TYPICAL PRODUCTS FOR CRUDE DISTILLATION ... 19
TABLE 2.1-2 GAS TREATING PROCESSES 21
TABLE 2.1-3 NATURE OF PRODUCT STORAGE AT REFINERIES ... 29
TABLE 2.1-4
TYPICAL GLAUS PLANT SULFUR RECOVERY FOR
VARIOUS FEED COMPOSITIONS WITH AVERAGE
ORGANIC BY-PRODUCTS AND ENTRAINMENT
ALLOWANCE
33
TABLE 2.1-5 PROCESSES USED FOR SULFUR REMOVAL FROM
GLAUS TAIL GAS
34
TABLE 2.2-1 LNG PROJECTS
70
TABLE 2.2-2 PROPOSED LNG PROJECTS 70
TABLE 2.2-3 GAS SWEETENING PROCESSES 76
TABLE 3.1-1 SUMMARY OF ENVIRONMENTAL IMPACT -
FUEL OIL REFINERY MODEL
119
TABLE 3.1-2 CRUDE FEEDSTOCK CHARACTERISTICS 120
TABLE 3.1-3 FUEL OIL REFINERY MODULE HEAT REQUIREMENTS. . 125
TABLE 3.1-4 MODULE ATMOSPHERIC EMISSIONS-FUEL OIL
REFINERY MODULE
127
IX
-------
LIST OF TABLES (Cont.)
Page
TABLE 3.1-5 EMISSION FACTORS FOR FUEL OIL REFINERY FUEL
USE 128
TABLE 3.1-6 SLUDGE INCINERATION EMISSION FACTORS .... 131
TABLE 3.1-7 WASTEWATER EFFLUENT QUALITY 134
TABLE 3.2-1 SUMMARY OF ENVIRONMENTAL IMPACT -
GASOLINE REFINERY MODULE 137
TABLE 3.2-2 CRUDE FEEDSTOCK CHARACTERISTICS 138
TABLE 3.2-3 GASOLINE REFINERY MODULE HEAT REQUIREMENTS . 145
TABLE 3.2-4 MODULE ATMOSPHERIC EMISSIONS-GASOLINE
REFINERY MODULE 147
TABLE 3.2-5 EMISSION FACTORS FOR GASOLINE REFINERY FUEL
USE 149
TABLE 3.2-6 SLUDGE INCINERATION EMISSION FACTORS .... 153
TABLE 3.2-7 WASTEWATER EFFLUENT QUALITY 156
TABLE 3.3-1 SUMMARY OF ENVIRONMENTAL EMISSIONS-LNG
MODULE . . . 159
TABLE 3.3-2 MODULE HEAT REQUIREMENT 164
TABLE 3.3-3 SUMMARY OF LNG MODULE AIR EMISSIONS 165
TABLE 3.3-4 EMISSION FACTORS FOR LNG MODULE 167
-------
LIST OF TABLES (Cont.)
TABLE 3.3-5 SUMMARY OF ENVIRONMENTAL EMISSIONS.-BASE
LOAD LNG MODULE 169
TABLE 3.3-6 SUMMARY OF LNG BASE LOAD MODULE AIR EMIS-
SIONS
174
TABLE 3.3-7 EMISSION FACTORS FOR LNG MODULE 174
TABLE 3.4-1 SUMMARY OF ENVIRONMENTAL IMPACT- SNG PLANT
MODULE 178
TABLE 3.4-2 PLANT GAS COMPOSITIONS 181
TABLE 3.4-3 MODULE HEAT REQUIREMENT 184
TABLE 3.4-4 MODULE ATMOSPHERIC EMISSIONS 185
TABLE 3.4-5 MODULE FUEL COMBUSTION EMISSION FACTORS. . . 186
TABLE 3.5-1 COMPARISON OF MODULE EMISSIONS 189
TABLE 4.1-1 SUMMARY OF FEDERAL AMBIENT AIR STANDARDS . . 194
TABLE 4.3-1 WATER ANALYSIS COSTS 205
TABLE 4.3-2 TYPICAL INSTRUMENTAL METHODS 207
TABLE 5.1-1 RULE-OF-THUMB COSTS OF TYPICAL COLLECTORS OF
STANDARD MILD-STEEL CONSTRUCTION 217
TABLE 5.1-2 HYDRODESULFURIZATION PROCESSES 230
TABLE 5.1-3 LEADING COMMERCIALLY AVAILABLE PROCESSES FOR
S0? REMOVAL FROM FLUE GASES 233
XI
-------
LIST OF TABLES (Cont.)
Page
TABLE 5.1-4 POTENTIALLY FEASIBLE S02 SCRUBBING SYSTEMS
IN PILOT PLANT DEVELOPMENT STAGE 236
TABLE 5.1-5 CLASSIFICATION OF OIL BURNERS ACCORDING TO
APPLICATION AND LIST OF POSSIBLE DEFECTS- . 250
TABLE 5.1-6 NATURE OF PRODUCT STORAGE AT REFINERIES . . 255
TABLE 5.1-7 RELATIVE EFFECTIVENESS OF PAINTS IN KEEPING
TANKS FROM WARMING IN THE SUN 263
TABLE 5.2-1 METHODS FOR REDUCING REFINERY WASTEWATER. . 281
TABLE 5.2-2 SEGREGATED WASTEWATER STREAMS 283
TABLE 5.2-3 NEUTRALIZATION REQUIREMENTS 295
TABLE 5.2-4 OPERATIONAL CHARACTERISTICS OF ACTIVATED-
SLUDGE PROCESSES
300
TABLE 5.2-5 COMPARISON OF LOW-RATE AND HIGH-RATE
TRICKLING FILTERS
302
TABLE 5.2-6 COMPARISON OF BIOLOGICAL PROCESSES 304
TABLE 5.2-7 CHLORINATION APPLICATIONS IN WASTEWATER
COLLECTION, TREATMENT, AND DISPOSAL . . .
306
TABLE 5.2-8 EXPECTED POLLUTION REDUCTION FROM COOLING
TOWER TREATMENT . 313
Xll
-------
LIST OF TABLES (Cent.)
TABLE 6.1-1 REFINERY AIR EMISSIONS
Page
335
TABLE 6.1-2
EMISSIONS AND STACK PARAMETERS-300,000 BPCD
GASOLINE REFINERY (REFERENCE MODULE).... 336
TABLE 6.1-3 "WORST" CASE METEOROLOGICAL CONDITIONS
ASSOCIATED WITH THE OCCURRENCE OF HIGH 24-
HOUR AMBIENT CONCENTRATIONS: BRAZORIA,
TEXAS 341
TABLE 6.1-4 SCHEMES USED IN EVALUATING EXPECTED GROUND
LEVEL POLLUTANT CONCENTRATIONS 348
TABLE 6.1-5 SUMMARY OF FEDERAL AND STATE AMBIENT AIR
QUALITY STANDARDS AND PREDICTED MAXIMUM CON-
CENTRATIONS FOR 300,000 BPCD REFINERY
EMISSIONS 350
TABLE 6.1-6 SUMMARY OF FEDERAL AND STATE AMBIENT AIR
QUALITY STANDARDS AND PREDICTED MAXIMUM CON-
CENTRATIONS -300 , 000 BPCD REFINERY
EMISSIONS 351
TABLE 6.1-7 SUMMARY OF FEDERAL AND STATE AMBIENT AIR
QUALITY STANDARDS AND PREDICTED MAXIMUM CON-
CENTRATIONS FOR-300,000 BPCD REFINERY
EMISSIONS 352
TABLE 6.1-8 SUMMARY OF REFINERY WASTEWATER EFFLUENTS AND
APPLICABLE TREATMENTS 355
-------
LIST OF TABLES (Cont.)
Page
TABLE 6.1-9 APPROXIMATE COMPOSITION OF REFINERY EF-
FLUENT AT THE OUTFALL 356
TABLE 6.1-10 SELECTED U.S. PORTS HANDLING SIGNIFICANT
AMOUNTS OF BULK CARGO 366
TABLE 6.1-11 OIL SPILL STATISTICS (BARRELS) 369
TABLE 6.1-12 WATER USED FOR ENERGY 371
TABLE 6.2-1 AIR EMISSIONS - SNG MODULE COMPARISONS ... 381
TABLE 6.2-2 3-HOUR NON-METHANE HYDROCARBONS 383
TABLE 6.3-1 AIR EMISSIONS - LNG MODULE COMPARISONS ... 388
TABLE 6.3-2 APPROXIMATE GUIDE TO APPRAISING ENVIRON-
MENTAL FACTORS OF LNG FACILITIES 393
xiv
-------
FIGURE 2.1-1
FIGURE 2.1-2
FIGURE 2.1-3
FIGURE 2.1-4
FIGURE 2.1-5
FIGURE 2.1-6
FIGURE 2.2-1
FIGURE 2.2-2
FIGURE 2.2-3
FIGURE 2.2-4
FIGURE 2.2-5
FIGURE 2.2-6
FIGURE 2.2-7
LIST OF FIGURES
PROCESSING PLAN FOR TYPICAL TOPPING
REFINERY
Page
39
FUEL OIL PROCESSING SEQUENCE 41
GASOLINE REFINERY PROCESSING SEQUENCE 43
PROCESSING PLAN FOR TYPICAL COMPLETE
REFINERY
48
LUBE OIL PROCESSING SEQUENCE 49
PETROCHEMICAL PROCESSING SEQUENCE 51
GENERALIZED FLOW PLAN FOR LNG SCHEME,
53
MOLECULAR SIEVE SYSTEM FOR COMBINED NATURAL
GAS DEHYDRATION AND C02 REMOVAL 56
TYPICAL CASCADE CYCLE FOR NATURAL GAS
LIQUEFACTION
58
TYPICAL MIXED REFRIGERANT LIQUEFACTION
CYCLE 60
TYPICAL EXPANDER LIQUEFACTION CYCLE 61
SUBMERGED COMBUSTION VAPORIZER 66
INDIRECT-FIRED INTERMEDIATE FLUID
VAPORIZER
67
xv
-------
LIST OF FIGURES (Cont.)
FIGURE 2.2-8
FIGURE 2.2-9
FIGURE 2.2-10
FIGURE 2.2-11
FIGURE 2.2-12
FIGURE 2.3-1
FIGURE 2.3-2
FIGURE 2.3-3
FIGURE 2.3-4
FIGURE 2.4-1
FIGURE 2.4-2
FIGURE 2.4-3
FIGURE 2.4-4
FIGURE 3.1-1
FIGURE 3.2-1
TYPICAL AMINE TREATING UNIT
TYPICAL GLYCOL DEHYDRATION UNIT
GLAUS SULFUR RECOVERY UNIT
REFRIGERATED ABSORPTION FOR RECOVERING
HEAVY HYDROCARBONS
LOW TEMPERATURE FRACTIONATION.
SNG PLANT PROCESSING SEQUENCE
NAPHTHA HYDRODESULFURIZATION UNIT
SNG PLANT UTILIZING TWO- STAGE
METHANATION
BENFIELD C02 REMOVAL UNIT
LOW BTU GASIFICATION PROCESS
HIGH BTU GASIFICATION PROCESS
COAL LIQUEFACTION PROCESS
SHALE OIL PROCESS
FUEL OIL REFINERY MODULE
GASOLINE REFINERY MODULE
74
78
80
83
85
89
91
95
97
104
107
109
114
121
139
XVI
-------
LIST OF FIGURES (Cont.)
FIGURE 3.3-1 LNG PEAK-SHAVING PLANT - 10 MM SCFD
CAPACITY .161
FIGURE 3.3-2 LNG BASE LOAD PLANT 170
FIGURE 3.4-1 SNG MODULE PROCESSING SEQUENCE 179
FIGURE 5.1-1 INCINERATOR EMISSIONS CONTROLS 213
FIGURE 5.1-2 VENTURI SCRUBBER 214
FIGURE 5.1-3 IMPINGEMENT PLATE SCRUBBER 215
FIGURE 5.1-4 TYPICAL SIMPLE FABRIC FILTER BAGHOUSE
DESIGN
216
FIGURE 5.1-5 CYCLONE 219
FIGURE 5.1-6 PLATE-TYPE ELECTROSTATIC PRECIPITATOR 221
FIGURE 5.1-7 FCCU PARTICULATE CONTROL SYSTEMS 222
FIGURE 5.1-8 SEPARATOR - ENERGY RECOVERY SYSTEM 223
FIGURE 5.1-9 THE DUCON GRANULAR BED FILTER 224
FIGURE 5.1-10 TYPICAL GAS TREATING PROCESS UTILIZING
AN AMINE SORBENT
227
FIGURE 5.1-11 SIMPLIFIED FLEXICOKING FLOW PLAN 231
FIGURE 5.1-12 PROCESS FLOW SCHEME OF THE SHELL FLUE
GAS DESULFURIZATION SYSTEM
235
xvi i
-------
LIST OF FIGURES (Cont.)
FIGURE 5.1-13 CATALYTIC H2 OR NH3 REDUCTION PROCESS,
Page
244
FIGURE 5.1-14 TYCO'S ISOTHERMAL SULFURIC ACID SCRUBBING.
SYSTEM 246
FIGURE 5.1-15 WATER-COOLED, CARBON MONOXIDE WASTE-HEAT
BOILER 252
FIGURE 5.1-16 SEALING DEVICES FOR FLOATING-ROOF TANKS 258
FIGURE 5.1-17 COMPLETE VAPOR RECOVERY SYSTEM 260
FIGURE 5.1-18 HYDROCARBON VAPOR SCRUBBERS 261
FIGURE 5.1-19 TOP LOADING ARM EQUIPPED WITH A VAPOR
RECOVERY NOZZLE
265
FIGURE 5.1-20 BOTTOM LOADING VAPOR RECOVERY SYSTEM 266
FIGURE 5.1-21 CATALYTIC AFTERBURNER 268
FIGURE 5.1-22 TYPICAL FLARE INSTALLATION 272
FIGURE 5.2-1 ENERGY CONSUMPTION IN A REFINERY AS A
PERCENTAGE OF CRUDE OIL FOR VARIOUS YEARS., 279
FIGURE 5.2-2 TYPICAL SEGREGATED WASTEWATER SYSTEM 284
FIGURE 5.2-3 TYPICAL SOUR WATER STEAM STRIPPER 286
FIGURE 5.2-4 CHEVRON WWT PROCESS 288
xviii
-------
LIST OF FIGURES (Cont.)
Page
FIGURE 5.2-5 EXTRACTION OF PHENOL FROM REFINERY
WASTE 289
FIGURE 5.2-6 TYPICAL OXIDIZING UNIT 290
FIGURE 5.2-7 WASTEWATER PROCESSING ALTERNATIVES FOR A
REFINERY 292
FIGURE 5 . 2-8 CORRUGATED PLATE INTERCEPTOR 293
FIGURE 5.2-9 BATCH NEUTRALIZATION OF SPENT CAUSTIC
WITH ACID 296
FIGURE 5.2-10 SCHEMATIC OF DISSOLVED-AIR FLOATATION TANK
WITHOUT RECYCLE 297
FIGURE 5.2-11 CONVENTIONAL ACTIVATED SLUDGE PROCESS 299
FIGURE 5.2-12 USEFUL RANGES OF SEPARATION PROCESSES 307
FIGURE 5.2-13 SCHEMATIC OF A MECHANICAL THICKENER 315
FIGURE 5.2-14 FLOWSHEET OF CONTINUOUS VACUUM FILTRATION.. 319
FIGURE 5.2-15 PASSAVANT FILTER SLUDGE DEWATERING
SYSTEM 320
FIGURE 5.2-16 FLASH-DRYING SYSTEM WITH MIXED REFUSE
INCINERATOR 322
FIGURE 5 . 2-17 SPRAY DRYER WITH PARALLEL FLOW 323
xix
-------
LIST OF FIGURES (Cont.)
Page
FIGURE 5.3-1 FLUIDIZED BED INCINERATOR 328
FIGURE 5.3-2 GENERALIZED FLOW SCHEME FOR A ZIMPRO WET
AIR OXIDATION UNIT 33°
FIGURE 6.1-1 HYPOTHETICAL CONFIGURATION OF A 300,000
BPCD GASOLINE REFINERY (REFERENCE
MODULE) 339
FIGURE 6.1-2 ISOPLETHS (m x 102) OF MEAN ANNUAL MORNING
MIXING HEIGHTS 343
FIGURE 6.1-3 ISOPLETHS (m sec'1) OF MEAN ANNUAL WIND
SPEED AVERAGED THROUGH THE MORNING MIXING
LAYER 343
FIGURE 6.1-4 ISOPLETHS (m x 102) OF MEAN ANNUAL
AFTERNOON MIXING HEIGHTS 344
FIGURE 6.1-5 ISOPLETHS (m sec'1) OF MEAN ANNUAL WIND
SPEED AVERAGED THROUGH THE AFTERNOON MIXING
LAYER 344
FIGURE 6.1-6 ANNUAL WIND ROSE VICTORIA, TEXAS
1964-1973 • 346
FIGURE 6.1-7 PROPOSED WATER MANAGEMENT PLAN 358
FIGURE 6.1-8 DISSOLVED OXYGEN CONCENTRATION PROFILE 362
FIGURE 6.1-9 CRUDE OIL PIPELINES 367
xx
-------
LIST OF FIGURES (Cont.)
Page
FIGURE 6.1-10 RELATIVE WATER ABUNDANCE OR DEFICIT
ACROSS THE U. S 372
FIGURE 6.1-11 LARGE RIVERS OF THE UNITED STATES 373
FIGURE 6.1-12 WATERWAYS OF THE UNITED STATES 376
FIGURE 6 .1-13 PRODUCT PIPELINES 378
FIGURE 6.3-1 LNG PIPELINES, PLANTS, AND POTENTIAL ENTRY
PORTS 391
xxi
-------
1.0 INTRODUCTION
1.1 Project Objectives
The objective of this report is to provide technical
information on the total environmental problem definition with
respect to petroleum refineries, synthetic natural gas (SNG)
plants based on conventional liquid feedstocks, and liquefied
natural gas (LNG) plants. This report includes the following
maj or elements:
(1) A review of the present technological
position (state of the art) of petroleum
refineries, SNG plants, and LNG plants
and their relationship to synthetic fuel
processes
(2) Definition of modules representing
typical refineries (fuel oil producing
and gasoline producing), SNG plants,
and LNG plants.
(3) Identification of emissions and
effluents from these modules in terms
of the media impacted (air, water,
solid) and the quantity and composition
of effluent streams.
(4) A review of effluent monitoring methods
which could be applied to the projected
effluent streams.
(5) A review and comparision of control methods
which may be employed at refineries, SNG
plants, and LNG plants.
-------
(6) Identification of plant impact and siting
problems.
(7) Identification of priority work areas for
research and development activities.
The work performed in this study divides into essentially
two areas: (1) the identification of emissions and emission
sources for the three technologies and (2) the evaluation of
environmental requirements resulting from these emissions in
terms of monitoring methods, control techniques, plant impact
and siting problems, and areas for research and development.
Sections 2.0 and 3.0 of this report examine the refinery, SNG,
and LNG technologies and establish the emissions associated
with these industries, while Sections 4.0-7.0 address the various
environmental aspects and problems resulting from these emissions.
Only criteria pollutants such as particulates, SO , CO, NO , and
X X
HC are quantified in Section 3.0; however, Sections 4.0-7.0 are
more qualitative in nature, covering potential pollutants such
as trace elements and trace organics.
1.2 Modular Concept
A modular approach is utilized in this analysis. Since
the initial step in establishing a representative module for an
industry involves a review of the technological position of that
industry, state of the art descriptions for the petroleum
refining, SNG, and LNG industries are presented in Section 2.0.
Various classifications and services within each technology are
reviewed. Processing options and alternatives are considered, and
typical processes and processing sequences are identified for
each technology. Also included in this section is a comparision
of the refining, SNG, and LNG industries with new energy tech-
nologies (coal gasification, coal liquefaction, and shale oil
production). The purpose of this comparison is to identify
-2-
-------
areas of the new energy systems to which the information generated
in this study might apply.
Utilizing typical processing sequences developed in
Section 2.0, process modules are derived in Section 3.0. The
modules presented in this report are for a fuel oil refinery,
gasoline refinery, SNG plant, and LNG plant. Due to the many
processing alternatives possible in petroleum refinery operations,
two refinery modules are presented, each representing a different
product emphasis (gasoline and fuel oil). Module flow rates are de-
termined assuming typical size commercial plant operation and utili-
zing specific process yield data. After each module is established
in terms of processes, flow rates, and energy or fuel demand, the
emission sources and emissions for that module are presented in
terms of source and media impacted. Air emissions, water effluents
and solid wastes are considered in this study. Also included in
Section 3.0 is a comparision of the emissions resulting from the
different modules on a common Btu output basis.
1.3 Summary of Environmental Problems
Using the modules given in the previous sections,
environmental problem areas can be determined, for each of the
energy conversion plants. Once the problem areas have been
recognized, Section 4.0 and Section 5.0 can be used to define
specific monitoring and emission control techniques which can
be applied to each of the problem areas.
1.3.1 Fuel Oil Refinery
Air Emissions
The major air pollution sources for particulates,
SO , CO, and NO within the fuel oil refinery module are the prc
X . X
cess heaters and boilers. These sources include the following:
-3-
-------
crude distillation,
gas oil hydrotreater,
naphtha hydrotreater,
heavy naphtha reformer,
C5/C6 isomerization,
propane deasphalting unit,
tail gas treating plant, and
light ends recovery.
The heaters and boilers used for these units are fired by refinery
fuel gas, heavy fuel oil, or coke gas from the flexicoker.
The major source of hydrocarbon emissions are general
fugitive emissions throughout the refinery. The exact sources
are difficult to identify and even harder to quantify. The
number used as the overall fugitive emission factor in the
refinery module (0.1 wt% of the throughput) is at best a rough
estimation.
Another major source of hydrocarbons is the crude
and petroleum products storage tanks. With the increasing
value of hydrocarbons, more effective and costly systems for
controlling storage tank losses are becoming economically
feasible.
-4-
-------
Another primary pollution source is the tail gas from
the acid gas treating plant. New methods of controlling emissions
can remove up to 99.970 of the hydrogen sulfide originally intro-
duced to the acid gas plant. However, due to the relatively
large volume of the acid gas stream, the total emission of SO
A.
to the atmosphere after treating is still very substantial.
A final air emission source within the fuel oil refinery
module is the sludge incinerator. Other methods of handling
the sludge such as landfilling will reduce or eliminate the
air pollution from this source.
Water Effluents
The major sources of contaminated water within the
fuel oil refinery are the following:
• sour water stripper condensate,
contaminated process water,
cooling tower blowdown,
caustic wash water, and
desalter water.
Other potential contaminated .water sources are oily process
area storm water, oily cleaning water, and oily water from a
ship's ballast (if located near a docking facility). The com-
bined wastewater from these sources is treated in a wastewater
treating plant. Presently, many alternatives exist for treating
wastewater and a treating facility can be designed to handle
many specific wastewater problems.
-5-
-------
Uncontaminated wastewater is also generated within the
refinery. This water is handled separately from the contaminated
water by a segregated wastewater system. The uncontaminated
water along with treated process wastewater is retained in a
holding pond for a certain period of time before final discharge
to the local environment.
Solid Wastes
Sources of solid wastes within the fuel oil refinery
module include:
entrained solids in the crude,
silt from surface drainage,
silt from water supply,
corrosion products from process units and
sewer systems,
solids from maintenance and cleaning
operations,
sludge from water treatment facilities
(or ash from the sludge incinerator), and
spent catalyst.
Being inert and acceptable for landfill, the generated solid
wastes do not usually present an environmental problem.
-6-
-------
1.3.2 Gasoline Refinery Module
Air Emissions
The major air pollution sources for particulates,
SO , CO, and NO within the gasoline refinery module are the
X X
process heaters and boilers. These sources include all those
indicated for the fuel oil refinery in addition to a middle
distillate hydrotreater, a heavy hydrocrackate reformer, and alky-
lation plant, and a hydrogen plant. Once again the fuels which
are used include refinery fuel gas, heavy fuel oil, and coke
gas from the flexicoker along with naphtha which is combusted
in the hydrogen plant.
Major sources of hydrocarbon emissions within the
gasoline refinery are the same as the fuel oil refinery. The
majority of hydrocarbon emissions results from general fugitive
emissions and from crude and petroleum product storage tanks.
Also, as in the fuel oil refinery, a primary emitter of SO is
<&
the acid gas treating plant.
A unique air pollution source within the gasoline
refinery module is the fluidized catalytic cracking unit (FCCU).
Uncontrolled the FCCU would be a major source of particulates,
SO , and carbon monoxide. Controls as described in Section 5.1
X
have, however, greatly reduced this problem.
Water Effluents
The water problems involved with the gasoline refinery
module are the same as the fuel oil refinery module.
-7-
-------
Solid Wastes
The solid wastes generated in the gasoline refinery
module are the same as the fuel oil refinery module.
1.3.3 Liquefied Natural Gas (LNG) Plants: Peak-Shaving and
Base Load
Air Emissions
The major air pollution sources for particulates,
SO , CO, and NO within the LNG modules are natural gas-fired
X X
heaters in the boiler units and the regasifiers. A heater is
also needed to regenerate the molecular sieve bed but the heat
load is small relative to the boiler and regasifier, and thus
the emissions are considered negligible.
The major source of hydrocarbon emissions are fugitive
emissions from all of the processing units. The same emission
factor used in the refineries is used here to determine the
fugitive losses.
If a glycol unit is being used to dehydrate the natural
gas before liquefaction, then there will be a continuous discharge
of glycol vapor (as triethylene glycol) along with the water
vapor. About 0.05 gallons of the glycol is emitted per million
standard cubic feet of natural gas processed.
Water Effluents
The majority of the wastewater effluent generated
within the LNG module is from the acid and caustic wash water
streams used for the demineralizer regeneration. This stream is
very small and can be handled by a small holding pond or by
direct discharge into a municipal sewage system.
-8-
-------
Solid Wastes
There are no significant solid wastes generated within
the LNG module.
1.3.4 Synthetic Natural Gas (SNG) Plant
Air Emissions
The major sources of particulates, SO , carbon monoxide,
X
and NO within the SNG module are process heaters and boilers.
X
These sources include the following:
preheaters,
super heater,
steam boiler,
Benfield C02 removal system, and
glycol dehydration unit.
All of these units are fired with low sulfur naphtha.
The Benfield COa removal system is a major source
of hydrocarbon emissions from the LNG module. Although the
concentration of hydrocarbons (mainly methane) is small in the
vented COa stream, the large volume of vented gas makes the
total amount of hydrocarbons emitted very substantial. The
other sources of hydrocarbon emissions are fugitive emissions
from all the processing units and emissions from naphtha storage
tanks. Additional hydrocarbons are emitted in the form of
glycol from the glycol dehydration unit venting system.
-9-
-------
Water Effluents
The wastewater from the SNG module consists of the
acid and caustic wash water streams used for the demineralizer
regeneration, the cooling system and boiler blowdown streams,
and the waste solution from the Benfield system. The total
stream flow is small and can be handled by a small holding pond
or by direct discharge into a municipal sewage system.
Solid Wastes
There are no daily discharges of solid wastes from an
SNG plant. Disposal of spent catalysts occur periodically
but do not pose an environmental problem because the catalyst
is inert and acceptable for landfilling.
-10-
-------
1.4 Comparison of Module Emissions
In the following subsections the emission rates will be
related to the specific module charge capacity. This approach
is used in order to present the emission impact of a typical
size plant for the specific industry and hence facilitate its
environmental assessment for each technology. In this sub-
section all of the modules are adjusted to a 1012 Btu/day output
of primary product. This adjustment is made in order to present
the different module emissions on a common basis and provide a
convenient comparison of the emission impact of the various
technologies. This comparison is presented in Table 1.4-1.
The large hydrocarbon emissions that result from these modules
are primarily a result of fugitive losses (assumed 0.1 wt% of
throughput).
1.5 Evaluation of Environmental Requirements
After the emissions and emission sources have been
identified, various environmental control requirements and prob-
lems are addressed. Applicable emission and effluent monitoring
methods are presented in Section 4.0. Differences in ambient air
sampling and effluent sampling are discusse.d, and monitoring data
such as accuracy and costs per sample are presented. Problem
areas associated with monitoring technology are identified and
gaps in technology noted.
Following the discussion of monitoring methods,
emission control techniques are addressed in Section 5.0.
Potential control methods are described and alternative control
methods are compared. Problem areas such as fugitive and
toxic chemical emissions are discussed. Control methods
capable of the most pollutant reduction are identified and any
techniques having potential for near zero emissions discharge
are described.
-11-
-------
TABLE 1.4-1
COMPARISON OF MODULE EMISSIONS
Basis: 1012Btu/Day Output Primary Fuels
i
t—1
N3
EMISSIONS AND EFFLUENTS
Air Emissions (Ib/day)
Particulates
sox
CO
NOX
HC
Water Effluents (Ib/day)
Suspended Solids
Dissolved Solids
Organic Material
Solid Wastes (Ib/day)
FUEL OIL111
REFINERY
6,
17,
1,
12,
78,
720
000
280
600
700
GASOLINE lij
REFINERY
12,300
26,800
2,680
35,340
90,600
PEAK-SHAVING 13]
LNG PLANT
5
5
75
47
,900
790
,600
,800
,000
BASE
LNG
2,
11.
2,
82,
43,
LOAD[31
PLANT
350
280
540
460
200
SNG PLANT
3
1
2
41
130
.750
,620
,310
.800
.000
[2]
266
9,850
56
8,500
295
10,900
62
16,500
0
0
0
0
0
0
0
0
negligible
I Primary fuels for the refinery modules are considered to be the gasoline and middle distillate or
light fuel oil product streams. The total heating values of these product streams (gasoline: 5.248x
106 Btu/bbl, middle distillate: 5.7x106Btu/bbl, light fuel oil: 5.825x106Btu/bbl) are combined and
adjusted to a 1012Btu/day output basis.
'Pipeline quality (1000 Btu/SCF) synthesis gas is considered to be the primary fuel from the SNG plant.
'Primary fuel from the LNG facility is regasified liquefied natural gas (1000 Btu/SCF).
-------
Refinery, SNG, and LNG plant siting problems are re-
viewed in Section 6.0. The consequences of emissions to ambient
air, wastewater effluents to receiving waters, and solid waste
effluents from the plants are discussed. Development of sampling
and analytical strategies for hazardous emissions are indicated.
Criteria such as raw material supply, energy supply, product
transportation, and Federal, state and local laws are also con-
sidered.
On the basis of information covered in the preceding
sections, suggested priority work for research and development
activities are presented in Section 7.0. Among the priority
areas considered are studies of air monitoring methods for
determining fugitive emissions, and for tracing pollutants
in the atmosphere to their sources. Water monitoring and
effluent controls and examination of the compositions of solid
wastes are also suggested. Cost analyses of future plant
design alternatives are discussed.
-13-
-------
2.0 PROCESS TECHNOLOGY DESCRIPTION
Descriptions of the present technological positions
of petroleum refineries, SNG plants, and LNG plants are presented
in this section of the report. The processes and operations
associated with each of the three industries are identified and
the major processing alternatives discussed. Typical processing
sequences are characterized and presented for each industry.
The specific purpose of this state-of-the-art review is to
provide a basis for the selection of specific modules to repre-
sent the technologies for emission determinations.
Also included in this section is a comparison between
the operations involved with the refining, SNG, and LNG in-
dustries and the processes associated with the new energy
technologies of coal gasification, coal liquefaction, and shale
oil production. Areas of similarity between the technologies of
this report and the new energy systems are identified in order to
establish the areas of the new energy technologies to which the
emission and control information discussed in this report might
apply.
-14-
-------
2.1 Petroleum Refining
Petroleum refining is an established industry and, as
a result, the technology associated with crude oil refining is
well defined. In general, the processing steps involved with
refining depend upon the quality of the crude oil and the product
distribution required.
Crude petroleum is a mixture of many different ,
hydrocarbon compounds. These compounds are distinguished by
their hydrocarbon type and by their normal boiling temperatures.
The hydrocarbon types include paraffins, naphthenes, and aro-
matics, and the normal boiling temperatures encompass a range
that exceeds 1000°F for most crudes (NE-044). Effects of •""
crude quality may be minimized to some extent by varying process
parameters such as pressure, temperature, and residence time.
In addition, refineries receiving a variety of crudes will
normally try to mix these crudes in order to achieve a consistent,
medium range, feedstock and thus avoid major changes in feed
quality as well as extreme crude types.
The factor that impacts processing sequence the most
is the product slate required from the refinery. Petroleum
refineries are capable of producing a wide range of products
and any of these products may be emphasized depending on overall
marketing strategy. Major product streams include light hydro-
carbons, gasoline, diesel and jet fuels, a light (distillate) fuel
oil, and a heavy (residual) fuel oil. Considerable fuel gas is
also produced; however, this stream is normally consumed on-site
to satisfy process heat requirements. In addition, a portion of the
fuel oil make is usually allocated for internal consumption. Other
products which may be associated with a refinery include petro-
chemicals, middle distillates, lube oils, waxes, asphalts, greases,
coke, and miscellaneous specialty products.
-15-
-------
In general, crude oil is refined by initially separating
the crude into various hydrocarbon fractions of specific dis-
tillation ranges. Although this separation step is common to
all refineries, the processes utilized on the straight run crude
fractions following the separation step depend upon the specific
refinery requirements.
Possible processing objectives may include:
treating the straight run streams to
remove impurities and undesirable
components with the minimum amount
of upgrading
significantly altering the straight
run product slate and quality with
conversion processes
obtaining special cuts and utilizing
specific processes for lube oil or
petrochemical production
Many processes and options are available to a refinery for
meeting any of these specific goals.
2.1.1 Refinery Processes
Operations associated with refineries may be roughly
categorized into areas of separation, treating, conversion,
blending, storage, and auxiliary processes. Conversion processes
may be further classified into cracking processes and combina-
tion and rearrangement (octane upgrading) processes.
-16-
-------
Separation
Although many processes may utilize a distillation
train or flash system to separate a process effluent into
product streams, the major areas of separation associated with
a refinery are crude separation and light end separation.
Crude separation is the initial processing step in
refinery operations. The process involves the physical separa-
tion of hydrocarbon components in the crude into fractions or
intermediates of specified boiling temperature ranges. This
operation is well established as the initial processing pro-
cedure in a refinery. The main difference between refineries is
the type and relative amounts of product streams obtained in
the separation process. The degree of separation that is made
and the amount of equipment that is required is largely governed
by the crude petroleum characteristics (VA-064) and by the pro-
ducts required from the refinery (WH-019). The crude separation
may be accomplished in one to three fractionation stages. These
stages include one atmospheric plus one or two vacuum fraction-
ation stages.
A topping unit separates the crude in an atmospheric
stage only. Streams from a topping unit normally include fuel
gas, naphtha, middle distillates, distillate fuel oil, and the re-
duced crude (atmospheric tower bottoms). Depending upon the refinery
objectives the naphtha stream may be split into light and heavy
naphtha and the fuel oil into light, middle, and heavy distillate.
The differences between the topping unit and other separation
processes is that the atmospheric tower bottoms are not separated.
This reduced crude stream is normally either used as a heavy
fuel oil or routed off-site for further processing.
-17-
-------
Crude distillation units employed for separation of
the entire crude stream utilize one or two vacuum towers for the
heavier fractions, Besides obtaining the same product streams
as a topping unit these separation units may either recover
additional gas oil from the reduced crude while producing a
heavy vacuum resid or else separate the reduced crude into special
lube oil cuts along with a resid stream. Normally one vacuum
stage is sufficient if the objective is to receive additional gas
oil from the reduced crude, whereas, two stages may be utilized
if many lube oil cuts are desired. Typical products from a
crude distillation unit are shown in Table 2.1-1.
Light ends recovery (sometimes known as vapor recovery)
involves the separation of refinery gases from the crude distilla-
tion unit and other processing units into individual component
streams, The separation is accomplished by absorption and/or
distillation. The recovery process that is utilized primarily
depends upon the desired purity of the products. For example,
an ethane-methane split of refinery gas would require the use
of cryogenic fractionation. On the other hand, a reasonable
ethane-propane split (60 to 75% propane-propene recovery) can
be achieved in conventional fractionation equipment (NE-044).
Treating
Streams from the crude separation step contain sulfur
compounds and other undesirable components which must be removed
due to the effect on product quality, catalyst sensitivity,
odor, and corrosivity. Although refiners originially could
treat only select streams, all streams from the crude dis-
tillation unit can now be desulfurized except for residua.
Desulfurization processes utilized in a refinery include both
gas treating and hydrotreating processes.
-18-
-------
TABLE 2.1-1
TYPICAL PRODUCTS FOR CRUDE DISTILLATION
Crude Fraction Typical Boiling Range
Light Ends Ci» and lighter
Light Naphtha 30 - 300°F
Heavy Naphtha 300 - 400°F
Kerosine 400 - 500°F
Light Gas Oil 400 - 600°F
Heavy Gas Oil 600 - 800°F
Vacuum Gas Oils 800 - 1100°F
Residue >1100°F
Source: (BL-078)
-19-
-------
Refinery gases separated in the crude distillation
unit and produced in various processing units contain a variety
of acid gas species, of which the major contaminant is hydrogen
sulfide. These acid gases are normally removed from the light
ends in a gas treating unit by absorption with an aqueous
regenerative solvent, A number of gas treating processes are
available and they are distinguished primarily by the regenerative
sorbent employer. Examples of gas treating processes which may
be utilized are presented in Table 2.1-2. Amine-based sorbents
are most commonly used in refinery applications (NG-002).
Desulfurization of petroleum cuts by hydrotreating is
widely practiced in modern refineries because of: (1) environmental
protection laws limiting the sulfur levels in fuels; (2) the
decrease of available low sulfur crudes ; (3) the undesirable
properties of sulfur and sulfur compounds including corrosiveness,
odor, color, instability, and catalyst poisoning tendencies; and
(4) hydrotreating petroleum stocks catalytically converts organic
compounds of sulfur, nitrogen, and oxygen into hydrocarbons
and removable sulfide, ammonia, and water. Although a stream
encompassing several product cuts may be desulfurized at one
time (as in crude desulfurization), petroleum fractions are
normally hydrotreated separately due to the varying sulfur
limits on the various fuels and the wide range of catalysts
and reactor conditions required to hydrotreat the various petro-
leum fractions. Hydrotreaters for naphtha, middle distillate,
distillate fuel oil, and residual oil streams are utilized in
refineries. Desulfurization is also achieved in a hydrocracker,
but the main purpose of this process is to crack a straight
run gas oil cut or a cycle gas oil from a FCCU to gasoline and
jet fuel fractions. Likewise slight desulfurization of residual
or lube oil fractions is accomplished in a hy-finishing process
in which the lube oil cut is mildly hydrogenated over a fixed
bed catalyst. Although some desulfurization results, objective
of this process might be color improvement or oxidation stability.
-20-
-------
Name
Adip
Fluor Econamine
SNPA
Sulfinol
TABLE 2.1-2
GAS TREATING PROCESSES
Solvent
AMINE SYSTEMS
Alkanolamine
DGA
DBA
Tetrahydrothiolene
Dioxide & Alkanolamine
Licensor
Shell
Fluor
Parsons
Shell
Benfield
Catacarb
Giammarco Vetro-
coke
ALKALI-CARBONATE SYSTEMS
Potassium carbonate
solution with Benfield
additives
Potassium salt solution
with additives
Potassium carbonate
with arsenic trioxide
Benfield
Eickmeyer
Vetrocoke
Fluor Solvent
Purisol
Rectisol
Selexol
PHYSICAL ABSORPTION SYSTEMS
Propylene Carbonate Fluor
N-Methyl-Pyrrolidone Lurgi
Methanol Lurgi
Dimethyl Ether of Allied
Polyethylene Glycol
-21-
-------
Cracking
Cracking processes convert heavy oils into petroleum
fractions of lower boiling range and correspondingly lower molecular
weight. These processes are very important in respect to the
quantity of gasoline or other light products obtained from a
barrel of crude. General categories of cracking processes are
thermal cracking, fluid catalytic cracking, and hydrocracking.
The thermal cracking concept was the first cracking
process to be developed and employed in refineries due to the
simplistic approach involved with just heating the hydrocarbon
fractions. Different thermal cracking processes and applications
are as follows:
(1) Thermal cracking of gas oil for naphtha
or gasoline production - the original
service of the thermal cracking process, this
application has mostly been taken over by
more sophisticated processes such as fluid
catalytic crackers or hydrocrackers. The
disadvantage of thermal cracking is that the
process is not selective and consequently the
yield of desired product is relatively low.
An advantage of this operation is that
the simplicity of the process combined
with the fact that no catalyst is employed
allows a thermal cracker to handle almost
any process stream.
(2) Visbreaking - visbreaking is a mild variation
of thermal cracking applied to a residual
fuel oil or reduced crude. The mild con-
ditions (880°F) are to minimize coke formation.
-22-
-------
The specific purpose of visbreaking is to
reduce the viscosity of the feed so as to
lessen the amount of blending stock required
to upgrade the feed to fuel oil specifications.
(3) Delayed coking - delayed coking is applied
to a residual stream and uses severe con-
ditions (1800°F-2000°F) to crack the feed-
stock to a coke gas, distillates, and coke.
(4) Fluid coking - fluid coking is a newer and
more flexible coking process. Fluid coking
converts the residual stream to higher value
products and results in less coke than delayed
coking (HA-282),
The primary advantages of catalytic cracking over
thermal cracking are in the production of a maximum of light
hydrocarbons at Ci» rather than at Ca and a higher yield of light
gasoline compounds (Cs and Ce)- Gasoline from a catalytic
cracker is also higher in branched paraffins, eyeloparaffins, and
aromatics, all of which increase the quality of the gasoline.
Feedstocks to catalytic cracking include gas oils
from both atmospheric and vacuum distillation and cracked
fractions from such processes as delayed or fluid coking.
Hydrocracking is a very flexible process which cracks
the feed in the presence of a high hydrogen partial pressure
(1200-1700 psi). Hydrocracking may be applied to any stream;
however, a gas oil stream is the usual feed. Hydrocracking is
-23-
-------
normally used on a high sulfur straight run gas oil that would
be unsatisfactory for catalytic cracking or on a gas oil effluent
from another cracking process. Hydrocracking complements catalytic
cracking and provides flexibility in meeting various product
demands. Products from a hydrocracker may include light ends,
light gasoline, heavy gasoline, and middle distillate. Heavier
fractions are normally recycled to extinction. Hydrocrackates
are saturated and are higher in branched chains than catalytic
cracking processes.
Combination
Combination processes normally combine two light hydro-
carbons to produce a gasoline range hydrocarbon. Either a poly-
merization process or an alkylation process may be used for this
purpose. A polymerization process combines two or more gaseous
olefins into a liquid product. An alkylation process joins an
olefin and an isoparaffin (isobutane) in order to produce a gas-
oline range hydrocarbon. The olefin feed to either unit is
usually obtained from the catalytic cracker. Isoparaffins for
an alkylation unit may be supplied from a hydrocracker. Both
processes utilize catalyst such as phosphoric, sulfuric, or
hydrofluoric acid.
Alkylation has grown at the expense of polymerization
due to two distinct advantages. A polymerization unit yields 1.0
bbl of gasoline for every 1.4 bbl olefin feed, whereas, 1.4 bbl of
olefin combined with isobutane is an alkylation process yields
approximately 2.5 bbl of alkylate (HY-008). In addition, the
alkylate has a motor octane rating approximately 12 octane
numbers higher than the polymer product. As a result, alkylation
processes are primarily utilized to obtain gasoline components
from olefin. Polymerization processes in this service are
-24-
-------
mainly older units. Polymerization of olefins is employed for
the production of petrochemicals.
Rearrangement
The primary purpose of rearrangement processes is
to change (rearrange) the molecular structure of the feedstock
to produce a high quality stream for gasoline blending. The
two major rearrangement processes in the refinery are catalytic
reforming and isomerization.
Catalytic reformers convert low octane naphthas into
high octane naphthas by catalytically rearranging and dehydro-
genating naphthenes and paraffins, forming aromatics such as
benzene, toluene, and xylenes. Reformer feedstock is normally
a desulfurized straight run or cracked naphtha (100-400°F).
The high octane aromatic products may be used for gasoline
blending or used for petrochemical feedstocks. Heavy naphthas
are usually fed when making gasoline and light naphthas when
making aromatics for the petrochemical industry (NA-182).
If the refinery is emphasizing petrochemicals, a liquid-liquid
aromatic extraction unit may be incorporated within the catalytic
reformer. The aromatic extraction unit separates the reformate
stream into a raffinate stream containing the non-aromatics and
an extract stream containing 95% aromatics (DE-070). Hydrogen
is also produced (800-1500 scf/bbl of feed) as a part of the
reforming process (NA-182). This aspect of catalytic reformers
is important in supplying the total hydrogen demand of the
refinery. Hydrogen from the reformer does not require desulfuri-
zation since the reformer feed streams are previously hydro-
treated to protect the reformer catalyst. Total streams from
the reformer usually include the reformate, light ends, and
hydrogen.
-25-
-------
Isomerization units are used to increase the octane
rating of pentane and hexane fractions by rearranging the normal
paraffins into isoparaffins. The feed to an isomerization unit
is normally desulfurized straight chain pentane and hexane
fractions. The product is a low sensitivity gasoline blending
stock consisting of up to 7570 isomers and having a clear research
octane number of 80 to 85 (RI-044). The reaction takes place
over a chlorinated platinum-aluminum-oxide catalyst at a temper-
ature of 320°F and a pressure of 400 psig (LA-078). This process
may also produce isobutanes for alkylation if this compound is
not sufficiently generated by the other refinery processes.
Any ICk produced is routed to alkylation while the C5/C6 effluent
goes to gasoline blending.
Blending
Refinery blending operations involve the mixing of
various components to achieve a product of desired characteristics.
Blending operations are associated with the final products of
gasoline, aviation fuels, heating fuels, lubricating oils, greases,
and waxes. The relatively few base and intermediate stocks may
be blended to produce over 2000 finished products (NA-182).
Although small volume blending may be performed in a mixing.
vessel, bulk product blending is normally achieved as an in-line
operation prior to product tankage.
The most common blending operation in petroleum refineries
involves the final step in the preparation of gasoline. Various
gasoline components such as catalytically cracked gasoline,
hydrocrackate, reformate, isomerate, alkylate, and butane are
combined with additives such as dye and tetraethyl lead (TEL)
in the necessary proportion to meet gasoline marketing speci-
fications . This blending is normally accomplished in a mixing
manifold prior to storage.
-26-
-------
Storage
Storage capacity is required for liquid feedstocks,
intermediate products, and final products. There are five basic
types of storage tanks used by refineries. These include fixed
roof, floating roof, internal floating cover, variable space,
and pressure. The application of these tanks largely depends
on the volatility of the stored liquid.
The fixed roof tank is the least expensive and the
most common type of tank used. It is a cylindrical steel tank
with a conical steel roof, Today fixed roof tanks are normally
equipped with pressure/vacuum valves set at only a few inches
of HaO to contain minor vapor volume expansion.
Floating roof tanks are cylindrical steel tanks simi-
lar to fixed roof tanks. However, instead of a fixed roof, they
are equipped with a sliding roof, designed to float on the surface
of the product. A sliding seal attached to the roof seals the
annular space between the roof and vessel wall from product
evaporation. Floating roof tanks eliminate the vapor space of
fixed roof tanks.
Internal floating covers are a modification of floating
roofs, designed to deal with the buoyancy problems caused by
snow and rain. They are essentially fixed roof tanks equipped
with an internal floating coyer similar to a floating roof.
Internal floating covers contain sliding seals to seal the annular
space between the cover and the vessel wall from evaporation.
-27-
-------
There are two basic types of variable vapor space
tanks, lifter roof and diaphragm. The lifter roof tank has a
telescopic roof, free to travel up and down as the vapor space
expands and contracts. A second type is the diaphragm tank
equipped with an internal flexible diaphragm to cope with vapor
volume changes.
Pressure tanks are used to store highly volatile
products. These tanks come in a very wide range of shapes and
are designed to eliminate evaporation emissions by storing the
product under high pressures. Pressure tanks are commonly de-
signed for pressures up to 200 psig.
Fixed roof, floating roof, and internal floating
cover tanks are the most common tanks in refinery service.
These tanks range in size from 20,000 to 500,000 bbl and
average 70,000 bbl (MS-001).
Table 2.1-3 indicates the vapor pressures (EN-043),
volumes (MS-001), and types of storage tanks used for several
major refinery products. Federal emission regulations currently
require hydrocarbon products with true vapor pressures (under
storage temperatures) ranging from 1.5 to 11.1 psia be stored in
floating roof tanks or their equivalent. Normally internal
floating covers are considered equivalent to floating roof
tanks.
-28-
-------
TABLE 2«]>3
Product
NATURE OF PRODUCT STORAGE AT REFINERIES
True Vapor
Pressure
psia @ 60°F Types of Storage Tanks
Qty. Stored
1968
(106bbl)
Fuel Gas
Propane 105
Butane 26
Motor Gasoline 4-6
Aviation Gasoline 2.5-3
Jet Naphtha 1.1
Jet Kerosene <0.1
Kerosene <0.1
No. 2 Distillate <0.1
No. 6 Residual . <0.1
Crude Oil 2
Cryogenic - Pressurized
Pressurized
Pressurized
Vapor Saver, Fixed Roof,
Floating Roof
Vapor Saver, Fixed Roof,
Floating Roof
Vapor Saver, Fixed Roof,
Floating Roof
Fixed Roof
Fixed Roof
*
Fixed Roof
Fixed Roof .
Vapor Saver, Fixed Roof,
Floating Roof
204
14
18
31
46
346
137
-29-
-------
Auxiliary Operations
A number of important auxiliary operations are required
in petroleum refining. These auxiliary operations include such
processes as crude desalting, hydrogen production, sulfur recovery,
water treatment, and power generation.
Crude Desalting
Crude desalting is normally the first unit in an oil
refinery. The function of this unit is to remove the inorganic
salts and brines from the incoming crude in order to prevent
process fouling, corrosion, and catalyst poisoning. Crude may
be desalted by either electrical or chemical means. Electro-
static crude desalting is the more prevalent process (RA-119).
In this process incoming crude is heated and then mixed with
water and passed through an emulsifier. The water-oil emulsion
is then routed through a treating vessel where a high voltage
field demulsifies the oil and water. Impurities from the
crude oil are removed in the water effluent.
Chemical crude desalting utilizes coalescing agents
instead of a high voltage field to demulsify the aqueous and or-
ganic phases.
Hydrogen Generation
Hydrogen is consumed in many refinery processes including
hydrotreating, hydrocracking, and isomerization. Hydrogen must
be available in order for these units to operate efficiently;
therefore, a plant hydrogen balance must be maintained. Although
a large portion of the hydrogen demand of a refinery may be
supplied by catalytic reformers, these units cannot normally
supply a sufficient amount of hydrogen to meet the total refinery
demand. A hydrogen generation unit is utilized to provide the
balance of the refinery hydrogen demand.
, -30-
-------
Hydrogen may be produced by either steam-hydrocarbon
reforming or partial oxidation. In steam hydrocarbon reforming
the steam is catalytically reacted with a light hydrocarbon
such as methane or naphtha to produce hydrogen. Reaction temper-
ature is approximately 1700°F (HY-007) . The generalized reforming
reaction is
nH2° + nCO
In partial oxidation a hydrocarbon feedstock is partially com-
busted with oxygen and steam, producing hydrogen from the following
reactions .
C H + ^ 02 t nCO + E H2
n m ^ z
+ nHz° •*• nCO +
The water gas shift reaction
CO + H20 £ C02 + H2
establishes the final gas composition in both processes and then
the COa is removed. Options in hydrogen production result from
the fact that steam-hydrocarbon reforming was developed for
a gas or naphtha feed, while partial oxidation accommodates a
heavy residual oil feed,
Sulfur Recovery
Sulfur recovery involves conversion of the hydrogen
sulfide content of acid gases into elemental sulfur. The Glaus
process is the widely accepted process for sulfur recovery in
the refinery industry. The H2S is combusted with a sub-stoichio-
metric air supply producing sulfur, sulfur dioxide and water.
-31-
-------
H2S + %02 Z S + H20
H2S + 3/2 02 J S02 + H20
Additional sulfur recovery is obtained in a series of catalytic
reactors by reacting H2S with S02. The number of reactors utilized
in the'Glaus unit is the main refinery option since conversion
varies with the number of reactors employed. Potential sulfur
recovery efficiencies for the different number of conversion
stages and °/Q H2S in the feed are shown in Table 2.1-4.
Tail gas from the Glaus plant can be further treated
by a tail gas treating unit. Many tail gas treating processes
are available, utilizing many different reaction mechanisms.
Examples of tail gas processes are given in Table 2.1-5.
Utilizing a tail gas treating unit in conjunction with a Glaus
plant usually increases the total sulfur recovery to between
99-99. 97o depending upon the Glaus plant efficiency and the
specific tail gas process employed.
Water Treating
Many options are available to the refiner concerning
wastewater handling. These alternatives include options in
refinery equipment,
in-plant pretreatment versus a waste
treatment plant,
segregation of wastewater streams, and
degree of primary, secondary, and tertiary
treatment.
-32-
-------
TABLE 2.1-4
TYPICAL GLAUS PLANT SULFUR RECOVERY FOR VARIOUS FEED
COMPOSITIONS WITH AVERAGE ORGANIC BY-PRODUCTS
.AND ENTRAINMENT ALLOWANCE
Hydrogen sulfide
In sulfur plant
Calculated percentage recovery
%
20
30
40
50
60
70
80
90
Two reactors
92.7
93.1
93.5
93.9
94.4
94.7
95.0
95.3
Three reactors
93.8
94.4
94.8
95.3
95.7
96.1
96.4
96.6
Four reactors
95.0
95.7
96.1
96.5
96.7
96.8
97.0
97.1
Source: (BA-166)
-33-
-------
TABLE 2.1-5
PROCESSES USED FOR SULFUR REMOVAL FROM GLAUS TAIL -GAS
Name
Beavon Sullur
Removal Process
CleanAir Sullur
Process
IFP Sulfur
Recovery Process
Shell's Flue Gas
Desullurization
Process
SNPA-Sulfuric
Acid Process
Sullreen Process
Wellman-S02
Retovery Process
Developer
Ralph M. Parsons &
Union Oil Co. of Califor-
nia
J. F. Pritchard & Co. and
Texas Gulf Sulfur Co.
Institut Francais du
Petrole
Koninklijke/Shell Lab-
cratorium, the Neth-
erlands
SNPAand Haldor Topsoe
SNPA and Lurgi
Gesellschaften
Wellman Power Gas
Operation
Los Angeles refinery.
Union Oil Co. of Califor-
nia
Pilot plant work. Okotoks
plant, Texas Gulf Sulfur
Co.
Philadelphia refinery,
Gulf Oil Co.
Demonstration plant.
Lone Pine Creek plant,
Hudson's Bay Oil & Gas
Co.
Nippon Petroleum
Refining Co., Japan
Idemitsu Oil Co., Japan
Kyokutoh Oil Co., Japan
Showa Oil Co., Japan
Pilot plant work, Pernis,
the Netherlands
Yokkaichi refinery of
Showa-Hokkaichi Oil Co.
SNPA sulfur plant, Lacq
field
SNPA sulfur plant, Lacq
field
Aquilaine's Ram River
sullur plant. Rockv
Mountain House, Alberta
Olin Cliemic.il Co.,
Paulsboro, N. J.
Japanese Synthetic
Rubuei Co.. China, Japan
Toa t«enivo Hur.v'O
relmeiy. Ranacawa,
Japan
t.1 Scftmlo, C.ilit.
Alhcd Chemical Co.
r.iiliiiiii: .iciti plant,
CIliuKO
Olin ('. i|>. Bulimic anil
lil.iut, I'uilis II. iv, Mil.
Abstract
Tail gas from Claus sulfur recovery plant is
catalyticallyhydrotreated at atmospheric pres-
sure. All sulfur compounds are converted to
H2S which is then processed through a Stret-
ford unit.
Three stags process: Stage 1 converts essen-
tially all S02 to sullur with some conversion of
HjS to sulfur. Stage 2 converts remaining hy-
drogen sullide to'sulfur in a Siretlord unit.
Stage 3 is a polishing unit to i educe Ihe COS
and C$2 level in the tail gas which is normally
installed between Ihe Claus plant an,1 Stage I.
Tail gas from a Claus unit is fed into an ab-
sorber, where the Claus reaction occurs in a
solvent in the presence of a calalyst. Sullur is
produced in the molten state directly Irom the
base of the absorber. No conversion of COS
and CS2 is claimed.
Dry process for removing SOZ from flue gas
from the incinerator in a parallel pass?se solid
bed swing reactor. This is a cvclic process in
which a copper on alumina acceptor is used for
acceptance and regeneration of the S02 at
750° F.
A purge gas stream to separate the oxidizing
and reducing atmor.pheres is required lor both
the acceptance and regeneration sleps.
S02 concentration step is required.
Tail gas is incinerated transforming all sulfur
compounds to SO?. The gas is then passed
through a convnlcr containing a vanadium
oxide-bared cnlaly;.t. S02 is oxidized to S03
with a 9C°; yield. The hot convener gas ex-
changes he.it in the concentrator, and then
goes through an absorber. Dilute acid pro-
duced is tliiMi «enl to a concentrator in which
the heat con to :il ol kv.r. liuin the convene, eva-
porates p.nt ol Ihe v.p itei fiom the acid.
Activated caibon bed citalvzes the Claus reac-
tion behveen the Hjjj and S02 in t,nl r,,i$ and
adsorbs eleineiit.il sullur loimed. Ineit recen-
eralion IMS is u>cd at elevated iempoi.ilu.es
to desoib the sullur. Bed is then cooled and
placed back on leiclicm cycle. No conversion
of COS and CSi is claimed.
Sullur plant incmtn.tlw ellluenl is cooled to
IbO* F and cunt.icU-.l with a sodium sulhte
solution. S02 in the J;.TS le.u Is to lorm sodium
bitiilhtf. Tin- r !•. can be bl.ippcd to low con-
centrations ol SO..
Alte.natiV'1 i-iriM:i:i.i!ion schemes h,we hoen
ii-. -.'ii. In urn1 | l.iii. tii-- sO> nch M'lulion lioni
the abr.!!!;...! II, •.•.•,-. n.I'.i .in ev.iiici.iloi. civs-
l;:lh;i'i ftiii-u- Ilir lir ullilu {li'cVinpn-..". in S02
Siillili1 en .l.il'i .lie mil ."ilvi'ii tu ''0 i"fii-
cul.il"il. Ihe n0; 1 v..itei v.ipur (PCtilcd lu Cl.iub
(ilj.it.
^ndiiini hydiyxide clientic.il rrakeup is re-
gunrd.
Extraneous
procesi
feed streams
required
Fuel gas and air
Fuel gas and air
None
Reducing gas
H2, H2/CO mix-
lu.es, or light
parattinic hy-
drocarbons
Fuel cas and air
Inert gas lo.r
reseneiation
None
Sulfur
removal
Removal to
250 ppm SOz
or less
Removal to
250 ppm SO:
or less
S02 removal
to 1,000 ppm
90% SOZ
removal
90°; SOj
conversion
75"? 0( sulfur
in Ihe Claus
plant tail gas
SO; removal
to 100 ppm
Product
Sulfur
Sulfur
Sulfur
S02 formed
is recycled
through a
Claus unit
34% sulluric
acid
Sullur
60*; S02 and
•10-;; water
vapor
Source: (BA-166)
-34-
-------
Options in design will effect the amount of wastewater generated.
Alternative processes and equipment which can significantly
reduce the wastewater load in a refinery are shown in Table 5.2-1,
In-plant pretreatment processes are considered the
water treating facilities located upstream of the process waste-
water treating plant. The amount of in-plant pretreatment
utilized is a major alternative since this pretreatment may often
be more efficient and economical than the waste treatment plant.
In-plant pretreatment processes include sour water strippers,
spent caustic oxidizers, spent caustic neutralizers, and
chlorination of sanitary wastes.
Another area of difference in wastewater handling is
in the segregation of water streams. Good effluent segregation
systems can significantly reduce the cost of wastewater handling
and treatment. Effluent may be separated according to dissolved
solids content, oil content, phenol content, sulfide content,
toxic chemical content and sanitary sewage content.
Waste treatment plants differ in the amount of primary,
secondary and tertiary treatment utilized. Primary treatment
facilities are involved in the physical upgrading of aqueous
effluents prior to discharge or secondary treatment. Primary
treatment processes include API separators; settling chambers
and clarifiers; air flotation, coagulation, and flocculation
systems; and alkaline and acidic neutralization. Secondary
treatment is for the removal of BOD and COD from wastewater.
Secondary treatment methods include aerobic biological treatment,
anerobic biological treatment, and chemical oxidation using
chlorine or ozone. The most common secondary method is aerobic
biological treatment in aerated lagoons. Tertiary treatment
consists of more severe water processing. Tertiary treatment
may be considered to consist of processes such as activated
-35-
-------
carbon treatment, ion exchange, reverse osmosis, and evaporation.
Although these processes are not now commonly employed, as
refineries move toward zero wastewater discharge tertiary treat-
ment will be used more extensively.
Power Generation
Although all refineries require steam generation, the
specific refining processes employed and the overall plant steam
balance determine whether a steam generation facility in con-
tinuous operation is necessary. Refineries producing a large
amount of steam from processes such as the FCCU CO boiler may
only require a steam boiler as a backup system, whereas other
refineries require steam generation facilities in continuous
operation. Likewise many refineries rely on an outside supply
of electricity while others, utilize a power plant for electrical
generation. Major factors effecting the utilization of electrical
generation facility are the availability and cost of outside
electricity and the refinery electrical demands,
Petroleum Refinery Categories
Although no two refineries are exactly alike, petroleum
refineries may be classified in general groups according to either
(1) the general purpose of the refinery (portion of the product
slate emphasized) or (2) any specialty processing associated
with the refinery such as lube oil or petrochemical processing.
Topping, fuel oil, and gasoline refineries each produce a dif-
ferent product yield structure and consequently utilize signi-
ficantly different processing sequences. Refineries producing
lube oils or petrochemicals employ special processes on selected
process streams; however, the greater part of the refinery may
be essentially a fuel oil or gasoline processing sequence. For
-36-
-------
this review, petroleum refineries are classified into the fol-
lowing five basic categories:
(1) topping refineries,
(2) fuel oil refineries,
(3) gasoline refineries,
(4) lube oil refineries, and
(5) petrochemical refineries.
Topping Refinery
The topping refinery is the simplest of the basic
refinery types. The topping or skimming operation is a simple
atmospheric-pressure distillation. The purpose of this type of
refinery is to obtain the atmospheric straight run fractions such
as naphtha, middle distillate, and fuel oil from the crude
stream.
A topping refinery does not typically utilize many
upgrading operations. Middle distillate and fuel oil streams
are normally desulfurized and routed to product tankage. The
straight run naphtha stream is desulfurized and may be either
marketed as such or upgraded further with a catalytic reformer.
If the naphtha stream is not upgraded with a catalytic reformer,
a hydrogen generation unit such as a steam-naphtha reformer is
required in order to supply hydrogen to the hydrotreaters /
Other processing units which may be associated with a topping
refinery include a gas treating unit for light ends and a sulfur
recovery unit. The reduced crude from the topping unit (atmos-
pheric tower bottoms) is routed off-site for further processing.
-37-
-------
A processing sequence for a topping refinery is shown
in Figure 2.1-1- The crude oil is initially split into the
straight run fractions of light ends, naphtha, heavy naphtha,
distillate fuel oil, and reduced crude oil. The light ends are
normally treated for acid gas removal and either used for fuel
gas or separated for petrochemical feedstocks. Light naphtha
is stabilized, hydrotreated and routed to gasoline blending.
Straight run heavy naphtha is hydrotreated and catalytically
reformed prior to routing to gasoline blending. Distillates are
hydrotreated and routed to product storage as distillate fuel
oil. Reduced crude oil is either routed to tankage as a heavy
fuel oil or goes off-site for further processing.
The processing sequence utilized in a topping refinery
is essentially duplicated in the other major categories of
refineries since atmospheric distillation is normally the first
major operation in all refineries and all straight run streams
are usually sweetened in refineries prior to further processing.
Since conversion processes are not normally associated with top-
ping refineries, the product yield and distribution is directly
related to crude quality.
Fuel Oil Refinery
A fuel oil refinery utilizes a processing sequence
established to promote the yield of product fuel oil. The fuel
oil produced may be approximately 40-60 vol% of the liquid pro-
duct with the bulk of the remainder of the products being gasoline,
Major elements in this type of refinery include a
crude distillation unit to separate the feed into straight run
fractions, light end treating and recovery units, hydrotreaters
for the crude distillation cuts, naphtha upgrading processes,
and thermal cracking or coking units for the heavy resid.
-38-
-------
Crude oil
GOT
Straight fun nopnlnQ
Heovy nophtho
Row kerosne
Distillates
• Fuel gat
StobSzed straight
n*i gasoline
•-{Catalytic reformer
Reduced crude oft
Gosofine
stock
. Kerofiino
-*- Ught fuel ofl
and
diesel fud
fuel ofl
FIGURE 2.1-1 PROCESSING PLAN FOR TYPICAL
TOPPING REFINERY
Source: EL-065
-------
A flow diagram for a typical fuel oil refinery is
shown in Figure 2.1-2.
This type of refinery utilizes essentially the same
processes as the topping refinery, but also includes units for
processing the heavy crude fractions - crude distillation,
propane deasphalting, flexicoker, and deasphalted oil hydrotreater,
In addition units for octane improvement such as isomerization and
catalytic reforming are normally utilized in the fuel oil re-
finery. These two units upgrade the straight run naphtha stream
to motor gasoline quality.
Fuel oil is obtained from the heavy resid stream by
(1) extraction of the asphalt with a propane deasphalting unit
and (2) cracking of the asphalt in a flexicoker to yield light
ends, naphtha and fuel oil. Although a flexicoker is shown in
this processing sequence any of several cracking processes
such as delayed coking and visbreaking which are capable of
handling the heavy asphalt stream may be used. Flexicoking
was chosen as typical, however, because of its potential to
produce more of the lighter, more valuable fuel products such
as naphtha and refinery fuel gas. Although the fuel oil product
is increased by cracking heavier fractions, the fuel oil yield
is not normally increased at the expense of the gasoline avail-
able from straight run naphtha. Although flexicoking is not
currently being used extensively in the refining industry, the
process appears to offer a viable way for achieving greater
naphtha and fuel oil yields and as such should see greater
utilization in the future.
Gasoline Refinery
In a refinery emphasizing gasoline production, exten-
sive cracking and upgrading facilities are utilized in order to
produce and refine the gasoline. A flow diagram for a typical
-40-
-------
I
.p-
DE-ASPHALTED OIL
HYDROTREATER
TAIL GAS
FUEL GAS
Co
MOTOR
GASOLINE
. LT. FUEL
OIL
HEAVY
FUEL
OIL
COKE
FIGURE 2.1-2 FUEL OIL PROCESSING SEQUENCE
-------
gasoline refinery is shown in Figure 2.1-3. Cracking processes
utilized include a fluid catalytic cracker, hydrocracker, and
flexicoker. Upgrading or rearranging processes include two
reformers, an isomerization unit, and an alkylation process.
Raw crude is initially separated into light ends,
naphtha, middle distillates, gas oil, and vacuum resid. Light
ends are routed to gas treating for acid gas removal and then
separated in a light ends recovery unit. Straight run naphtha
is hydrotreated and then split into light and heavy naphtha. The
light naphtha is routed through an isomerization unit to gasoline
blending. The heavy naphtha is catalytically reformed and then
also goes to gasoline blending. Middle distillate from the
crude distillation unit is hydrotreated and routed to product
tankage. Straight run gas oil is split between a catalytic
cracker and a hydrocracker. Both cracking units are used
because the products from the FCCU and hydrocracker complement
each other as well as providing refinery flexibility. Gas oil
to the catalytic cracker is first hydrotreated to protect the
cracking catalyst. Products from the FCCU include light ends,
Ca/Ci*, gasoline, and fuel oil. Light ends from the FCCU are
routed to gas treating. The C3/C4 stream is treated by caustic
scrubbing in a Merox unit and routed to an alkylation unit. The
alkylation receives this olefin stream and isobutane from light
ends recovery. Alkylate and excess butane go to gasoline blend-
ing. Cat gasoline is also Merox treated and routed to gasoline
blending. Heavy and light cycle oil from the FCCU goes to product
tankage as heavy fuel oil.
The hydrocracker produces light ends, a light naphtha
(hydrocrackate), and a heavy naphtha (hydrocrackate). Light
ends are routed to gas treating and then to gas recovery.
These light ends are a major source of the isobutane required
for alkylation. The light hydrocrackate is routed to gasoline
blending. Heavy hydrocrackate is routed to a catalytic reformer
-42-
-------
LIGHT ENDS
LIGHT ENDS
STRAIGHT RUN
NAPHTHA
HYDROTREATER
TAIL GAS
FUEL GAS
ETHANE AND
ETBYLENE
PROPANE AND
PROPYLENE
I
•P-
CO
I
DESALTED CRUDE
MID-
DISTILLATES
HYDROTREATER
HYDROTREATED
DAP
HVY.
FUEL pIL
ASPHALT
FLEXICOKER
r
i r
LOW SULFUR FUEL GAS
NAPHTHA
DISTILLATE i '
GAS OIL
HYDROGEN
PLANT
.H,
FIGURE 2.1-3 GASOLINE REFINERY PROCESSING SEQUENCE
-------
and then to gasoline blending. Vacuum resid from the crude
distillation unit is routed to a propane deasphalting unit for
additional recovery of gas oil. The extracted oil is routed to
a deasphalted oil hydrotreater. The deasphalted oil hydrotreater
produces light ends, naphtha, and a heavy fuel oil. Light ends
are routed to acid gas treating, naphtha to gasoline blending,
and fuel oil to product tankage.
Asphalt from the deasphalting unit goes to a flexicoker
for cracking. The flexicoker produces fuel gas, light ends,
naphtha, and coke. The fuel gas is normally consumed at the unit;
however, this stream can be routed into the refinery fuel
gas mix drum. Light ends go to acid gas removal and then to
light end recovery. Naphtha may be routed to gasoline blending
or to a hydrogen generation plant. The hydrogen plant is required
to provide the hydrogen that cannot be supplied from the catalytic
reformer. Coke produced from the flexicoker is taken to product
storage.
Lube Oil Refinery
A lube oil producing refinery may essentially be either
a gasoline or fuel oil refinery with the modification of the lube
oil processing sequence. The operations involved with lube oil
processing are as follows:
vacuum distillation,
solvent deasphalting,
lube stock treating,
dewaxing,
finishing, and
blending and compounding.
-44-
-------
Individual raw lube oil fractions are cut at the
vacuum distillation tower of the crude distillation unit.
Preferred lube oil stocks are high-boiling paraffins with several
side chains. Normally three to five lube stocks are separated
on the vacuum tower.
Heavy oil from the vacuum tower is routed to a solvent
deasphalting unit. Propane is normally the solvent utilized.
Asphaltenes entrained in the lube oil streams must be removed
since they have an adverse effect in processing in other lube
oil units - poor color and lower yield from treating, slower
filtration in solvent dewaxing, and more coking in hy-finishing.
The deasphalting unit produces an asphalt and a heavy lube oil
known as "brightstock."
Deasphalted oil is routed to treating processes to
improve the viscosity characteristics, color, and carbon residue
content. Treating processes which may be utilized include sulfur-
ic acid treating, phenol extraction and furfural treating. In
sulfuric acid treatment the oil is contacted with sulfuric acid
and then the resulting acid sludge is separated from the oil.
This process results in an acid sludge disposal problem. Phenol
extraction and furfural treating are the more prominent forms of
lube oil treating. Both processes use the respective solvents
to separate armoatics and naphthenes form the charge oil. Other
treating processes which are in use are as follows:
Edeleanu Process - liquid S02/benzene
solvent
-45-
-------
Duo-sol - propane and cresylic acid
solvent
Following treatment the oils are routed to a solvent
dewaxing process for removal of wax and improvement of pour
point. The process involves mixing the oil with a solvent,
cooling the mixture and filtering out the precipitated wax, and
separating the oil from the solvent. The most prominent solvent
is a mixture of methylethylketone (MEK) and benzene and/or
toluene. Other solvents which are in use include propane,
acetone, ethylene dichloride and benzene, and dichloroethane
and methyl chloride.
The product from the solvent dewaxing unit, known as
slack wax, has too high of oil content to meet marketing speci-
fications. In order to reduce the oil content, the slack wax
may be heated to "sweat" the oil out. Another method of lowering
the oil content involves mixing the wax with dewaxing solvent
and filtering again. This operation is known as repulping.
Slack wax must usually go through the repulping process twice
for the oil to be sufficiently reduced.
Oil from the dewaxing unit goes to finishing for color
improvement and oxidation stability. Lube oils are finished by
removal of traces of resinous materials and compounds which can
potentially form organic acids. Oil is finished by either clay
treating or hydrofinishing. Clay treating involves percolating
the oil through a packed clay column. This process is not
used on a large scale due to problems with spent clay disposal
(SO-076). Hy-finishing involves hydrotreating the lube in
order to remove organic nitrogen and oxygen compounds.
-46-
-------
The finished lube oil stocks are blended in various
proportions to produce the finished lube oil products. The many
different lube products result from blending the several lube
oil stocks and the addition of special additive packages. Bulk
products may be mixed in-line prior to storage, while small
volume or specialty products may be prepared in compounding
vessels.
The relationship of the lube processing facilities to
the other refinery processes is illustrated in Figure 2.1-4.
A typical lube oil processing sequence is shown in Figure 2.1-5.
Petrochemical Refinery
Petrochemical processing operations may be a part of
any basic refinery configuration. The units normally associated
with refineries are an olefins plant and an aromatics plant.
The olefins plant may receive a feed of light hydrocarbons
(C2-CO from the light ends recovery unit producing ethylene,
propylene, and butadiene. Other feedstocks which may be routed
to an olefins plant include naphtha and gas oil. Modern olefin
plants are frequently designed for total feedstock flexibility
and are therefore able to meet production demands with any
rates of feedstocks (ST-221). By-products from the olefins plant
are a hydrogen stream, a pyrolysis gasoline, and a pyrolysis oil.
The hydrogen is normally routed to the refinery hydrotreating
or hydrocracking facilities while the pyrolysis oil is sent to
heavy fuel oil product tankage. Pyrolysis gasoline is normally
routed to the aromatics plant.
The aromatics plant receives the pyrolysis gasoline
from the olefin unit as well as any aromatic rich naphtha streams
from the refinery. A major source of aromatics in the refinery is
the catalytic reforming units. When an aromatics plant
-47-
-------
DRY GAS
00
LIGHT
NAPHTHA
STRAIGHT RUN GASOLINE
HYDROCRACKED GASOLINE
CATALYTIC
REFORMING
UNIT '
HEAVY
NAPHTHA
RAW
KEROSINE
MIDDLE
DITILLATES
HYDROGEN
PLANT
HYDROCRACKED
GASOLINE
SULFUR
PLANT
HEAVY GAS OIL
HYDROGEN
SULFIDE
CATALYTIC
GASOLINE
CATALYTIC
CRACKING
UNIT
LIGHT FUEL OIL
COKER
GASOLINE
REDUCED
CRUDE
LUBE DISTILLATES
*- FUEL GAS
*- LP GAS
MOTOR
GASOLINE
AVIATION
GASOLINE
^. OLEFINS TO
*^ CHEMICAL
KEROSINE
*. LIGHT FUEL
OIL
DIESEL
FUEL
SULFUR
LUBES
WAXES
*" GREASES
HEAVY FUEL
OIL
*- ASPHALT
COKE
FIGURE 2.1-4 PROCESSING PLAN FOR TYPICAL COMPLETE REFINERY
Source: NA-032
-------
Propane
Deasphalting
Unit
Furfural
Treating
Unit
Solvent
Dewaxing
Unit
Hy-finishing
Unit
Lube Cuts
From Vacuum
Distillation
Unit
f
Asphalt
Repulping
Process
Lube
Oil
Blending
Lube
Oil
Product
Wax
FIGURE 2.1-5 LUBE OIL PROCESSING SEQUENCE
-------
is included in the refinery a liquid-liquid aromatic extraction
unit is used in conjunction with the catalytic reformers. The
aromatic extraction unit separates the reformate stream into
a raffinate stream containing the non-aromatics and an extract
stream containing 9570 aromatics (DE-070) . Specialty cuts from
other processes may be routed to the aromatics plant if the amount
of aromatics makes the separation and handling of these streams
economically feasible. Product streams from an aromatics plant
are benzenes, toluenes, and xylenes. A non-aromatic naphtha
stream is routed to the refinery gasoline blending facilities.
A simple block flow diagram of the petrochemical
facilities is presented in Figure 2.1-6. These facilities
may be associated with any basic refinery processing sequence
(topping, fuel oil, gasoline) regardless of complexity.
-50-
-------
Hydrogen
Light Ends
Naphtha
Gas Oil
Fuel Gas to
Refinery
Olefins
Plant
Ul
Reformate From
Catalytic Re-
formers
Solvent
Extract
Unit
Ethylene
Propylene
Butadiene
Aromatics
Plant
To Gasoline
Blending
To Fuel Oil
Storage
FIGURE 2.1-6 PETROCHEMICAL PROCESSING SEQUENCE
-------
2.2 LNG Process Technology Description
Liquefied natural gas (LNG) has enjoyed a widespread
and growing use in the world in the last ten years. This growth
has primarily resulted from the increasing demand for and attrac-
tiveness of natural gas as a fuel as well as the added convenience
that transportation and storage of liquefied natural gas has over
natural gas. One cubic foot of LNG is equivalent to over 600
standard cubic feet of natural gas.
There are numerous LNG facilities in service or under
construction in the U.S. and throughout the world. Basically,
two types of plants exist, the base load plant and the peak-
shaving plant. Large base load liquefaction installations are
located in areas such as the Middle East, Indonesia, and Alaska
where significant amounts of natural gas are produced, yet very
little is consumed. The gas is piped to a plant located on the
coast, liquefied, and transported via LNG tanker to a regasifi-
cation facility located in an area with a large natural gas demand.
The peak-shaving gas liquefaction plant is used for a
different purpose than the base load facility. Its function is
to liquefy natural gas during surplus periods and store it for
peak demand periods. At peak demand, LNG is withdrawn from storage,
regasified, and sent into the pipeline distribution grid. These
plants are generally much smaller than base load operations in
liquefaction capacity (LE-156).
However, regardless of the size, the basic liquefaction
processing steps are similar for base load and peak-shaving plants.
Figure 2.2-1 presents a generalized flow sketch of an LNG scheme.
If the liquefaction facilities are near an area of large gas
consumption and remote from gas production fields, as is likely
-52-
-------
Transportation
r~
to Raw
I Field
Gas
an
LI
Acid-Gas Removal
d Sulfur Recovery
Dehydration
Tvoical Natural Gas Proc
Heavy Hydrocar
Recovery
Bssine Plant
_. . t
Pipeline """
Quality
son Natural Gas
_J
4
i
Liquefaction Liquefaction
Pretreatment i
1
t
\
\
\
\
^ Pipeline
y Gas*'
/ Regasif ication
/
/
Storage
FIGURE 2.2-1 GENERALIZED FLOW PLAN FOR LNG SCHEME
-------
for peak-shaving installations, the natural gas feed to the LNG
plant is usually,of pipeline quality. That is to say, it has
already been processed through a gas plant for the removal of
impurities such as water, hydrogen sulfide, and carbon dioxide
and for the recovery of ethane and heavier hydrocarbon compounds.
For this reason, a section on conventional gas processing opera-
tions will not be included in the technology description of
peak-shaving liquefaction plants presented in Section 2.2.1.
This processing sequence is not the usual situation
encountered in base load LNG operations. Typically, the base
load plant will receive a raw natural gas which requires ex-
tensive processing prior to liquefaction. This gas purification
section may resemble in sequence the flow pattern presented in
Figure 2.2-1 or, as is more often the case, the gas processing
steps may be integrated providing for a more efficient operation.
This case is dicussed in Section 2.2.2.
2.2.1 LNG Peak-Shaving Plant
There has been an increased tempo in peak-shaving plant
development in recent years. By the end of 1975 there will be about
fifty peak-shaving installations operating in the U.S. with a
combined liquefaction capacity of over 300 million cubic feet
per day. The majority of these plants are located in the north
and northeastern areas of the country, with Massachusetts having
the most liquefaction facilities at six. The individual plant
capacities range from a low of 0.5 MM scfd to a high of 25.0 MM
scfd. The average plant size is in the 5-10 MM scfd range (US-191).
2.2.1.1 Natural Gas Feed Preparation
Since the gas fed to peak-shaving liquefaction installa-
tions has been normally processed in a natural gas plant, the
-54-
-------
degree of clean up required is minimal (LO-102). Liquefaction
of natural gas requires process temperatures as low as -260 F.
Therefore, any constituents of the inlet gas stream that may
become solid at these temperatures must be removed to the extent
that they will remain in solution in the LNG to avoid significant
fouling or plugging problems. The two constituents found in the
gas feed to peak-shaving plants that must be so reduced are water
and carbon dioxide. In addition, process requirements necessitate
removal of hydrogen sulfide, should it be present (AM-127).
Permissible concentrations of impurities in natural
gas feeds depend on the choice of the subsequent liquefaction
process and particularly on the susceptibility to fouling and
blockage of heat exchangers and expansion engines used for re-
frigeration. Generally it is desirable that the water content
of the gas should be less than 1 ppm. Carbon dioxide concentra-
tions should be in the range of 50 to 150 ppm. Hydrogen sulfide,
as far as potential fouling is concerned, could probably be as
high as 30 to 50 ppm; but in fact other considerations such as
odor, corrosion and toxicity restrict it to a maximum of 3 ppm
or less.
This clean up job is most often accomplished with the
use of molecular sieve synthetic zeolite adsorbents (LO-102).
They offer an economical and very thorough one-step C02 , H20,
and H2S removal ability. Should the quantity of H2S removed be
large enough, it is recovered as sulfur in a Glaus plant. A
flow sheet of a typical molecular sieve process for application
to a peak-shaving plant is shown in Figure 2.2-2. After physical
separation of entrained solids and liquids, the incoming gas
flows downward through a tower filled with molecular sieves.
Water is removed in the upper section of the adsorbent bed
and carbon dioxide is removed in the lower section. Effluent
natural gas typically contains less than 20 ppm C02 and less than
1 ppm HaO. When the tower approaches saturation, the inlet stream
-55-
-------
RETURN TO
PIPELINE
Ol
COOLER l>
V
WATER
KNOCK-OUT
H20 _
SECTION
NATURAL GAS
FEED
C02
SECTION'
V
INLET
SEPARATOR
I
TOWER "A"
PURIFICATION
TOWER "B"
REGENERATION
HEATER
REGENERATION
GAS FROM
COLD BOX
PURIFIED GAS TO
LIQUEFACTION
FIGURE 2.2-2 MOLECULAR SIEVE SYSTEM FOR COMBINED
NATURAL GAS DEHYDRATION AND C02
REMOVAL
SOURCE: (AM-127)
-------
is switched to a second tower, while the adsorbent in the first
is regenerated by flowing heated, dry gas counterflow to the
direction of the stream that was being cleaned. After leaving
the tower, the warm, moist regeneration gas is cooled and much
of the water is condensed, separated, and removed from the
system. The regeneration gas is then passed back to the pipeline
main, mixed with the incoming gas to the adsorbing tower, or
used as fuel for a boiler or prime mover (AM-127).
2.2.1.2 Liquefaction Cycles
Although there are many refrigeration cycles which can
liquefy natural gas, the three types most commonly used in LNG
plants are: the cascade, the mixed refrigerant, and the expander,
Whether they are labeled as such, most present-day natural gas
liquefaction cycles are variations or modifications of these
three basic types.
Cascade Cycle
A typical cascade cycle diagramed in Figure 2.2-3
(IN-029) is a combination of vapor-compression refrigeration
stages which may normally utilize three refrigerants: propane,
ethylerie, and methane.
Propane is compressed from about 15 psia
to a pressure sufficient for condensation
by air or water.
Ethylene is compressed in two stages
and condensed in the low-pressure propane
evaporator. The propane and ethylene
precool and condense the natural gas stream.
-57-
-------
FIGURE 2.2-3 TYPICAL CASCADE CYCLE FOR NATURAL GAS
LIQUEFACTION
SOURCE: (IN-029)
-58-
-------
' The third refrigerant, methane, is
sometimes used in a closed cycle for
subcooling the LNG, the methane be-
ing condensed with ethylene.
Sometimes an open cycle system is used, in which a small
side stream of product LNG is recycled to subcool the liquefaction
stream. A cascade cycle is usually the most thermodynamically
efficient, requiring the lowest horsepower (WH-032).
Mixed-Refrigerant Cycle
The mixed-refrigerant cycle is also a vapor-compression
type of cycle using air or water to condense the refrigerant.
However, in this cycle, pictured in Figure 2.2-4, it is possible
to obtain a low temperature with one mixed refrigerant. The
need for many compressors or a multiservice compressor and many
evaporators is thereby eliminated.
The simplicity of this cycle decreases the amount of
control, piping, and mechanical equipment required; however, many
actual mixed-refrigerant cycles are less thermodynamically effi-
cient than the cascade cycle.
Expander Cycle
\
The expander cycle,.shown in Figure 2.2-5, is most
popular when used in parallel with an existing regulator station.
An expander system uses the refrigeration available from expand-
ing gas. This refrigeration capacity normally occurs in a dis-
tribution system where gas pressure is dropped between a cross-
country pipeline and a low-pressure line. For this reason the
expander cycle has found specialized application in peak-shaving
liquefaction plants (IN-029).
-59-
-------
LNG to Storage
Start
Heat Exchanger J
Staged Propane
-si-
Mixed Refrigerant
Treated Natural Gas Fee]
C.W.
FIGURE 2.2-4 TYPICAL MIXED REFRIGERANT
LIQUEFACTION CYCLE
SOURCE: (HY-014)
-60-
-------
To Plant Fuel
i
o\
Treated
Feed Gas
Inlet
Recompressor
Residue
Heat
Dehydration
To Sales
i
2
: Inlet Gas ij TQ\
Separator j \^s
',er
•\
r ' rl
Expander
i
^
___
Supplemental
Compression
if Required
Condensate
Stripper (or
Fractionator)
Condensate
Stripper
Reboiler
Raw Product
Cooler
LNG
to
Storage
FIGURE 2.2-5 TYPICAL EXPANDER LIQUEFACTION CYCLE
SOURCE: (HY-014)
-------
In order to provide sufficient refrigeration, these
facilities use a flow through the plant of 8 to 15 times the
desired liquefaction rate. Therefore, a large quantity of
gas must be handled, both in the gas clean up system and the
cold box.
An advantage of this type of an expander cycle is that
the system is relatively simple since the refrigerant can be
handled in one stream. Also, little external horsepower is re-
quired. However, if there is insufficient flow from a high-
pressure line to a low-pressure line, it is necessary to provide
compression equipment for recycling gas in a closed loop, making
this system less attractive. Other disadvantages are that large
quantities of gas must be handled and areas for application
(adjacent to a regulator station) are potentially limited.
2.2.1.3 Storage
Storage facilities for LNG are required whether the
liquid is to be used to meet winter shortages of gas or to supply
base load gas by long distance shipment. In the latter case com-
plete ships' cargoes have to be loaded into and unloaded from
LNG tankers, i.e., storage capacity must be at least equal to
the maximum volume of LNG expected in any one shipment. Storage
for peak shaving, on the other hand, depends on the number of
days per year during which gas is to be liquefied - 200 to 220
in a temperate climate - and on the daily capacity of the lique-
faction plant (LO-102).
LNG can be contained on shore in three basic types of
storage facilities: the above-ground double-walled metal tanks,
prestressed concrete tanks, and in-ground or cavern spaces.
-62-
-------
Metal Tanks
All but one of the United States' peak-shaving plants
uses the above-ground metal tank for storage. The popularity
of this type of tankage is due to the improved control of heat
leakage, easier access for repairs, and lack of geological
constraints.
The embrittlement of mild steel at temperatures below
-50°C makes it necessary to provide aluminum or 97o nickel-
stainless steel to contain the liquid. The outer shell of these
containers is carbon steel. Sandwiched between the two vessels
at the bottom of the tank is a load-bearing insulation and between
the walls of the tank is an insulation system such as loose-fill
perlite. The roof should be covered with glass fiber or a
similar lightweight insulating material. This method of con-
struction results in a container similar to the ones used on-
board LNG ships.
Other Types of Storage
There are available to the LNG industry various alter-
natives to the metal tank. However, to date there are few facil-
ities which use a different storage technique.
Concrete has passed low-temperature and LNG-immersion
tests and as a result can be classified a suitable construction
material if correctly prepared. When used to form a large
container, it must be reinforced with prestressed or pqststressed
rods or wire to prevent cracks resulting from thermal stresses.
These storage tanks may be insulated on the outside, or inside
of the concrete wall, and can be located either above or below
ground (IN-029).
-63-
-------
Cryogenic in-ground storage has been the subject of a
great deal of research and study in recent years. As a result
of these efforts, this storage system has been brought to the
point where accurate construction estimates can be made. The
container itself consists of an excavated, earthen storage
cavity with no insulation and no liner. The roof may be made
of either 97o nickel steel or aluminum on the inner lining, with
some insulation separating this inner wall from the outside
shell (IN-029).
Cavern storage presents yet another method of storing
LNG. The Institute of Gas Technology has developed a large
volume room and pillow storage technique which could be applicable
to peak-shaving facilities. The room would be close to the sur-
face and insulated to reduce boil off. Gaz de France has designed
a large volume storage which could be utilized in base load opera-
tions. This concept consists of a vertical shaft used as an
access to a large horizontal storage gallery, which is excavated
in impermeable strata. The gallery is operated at a pressure
corresponding to the hydrostatic pressure of the overburden
(IN-029).
2.2.1.4 Regasification Systems
It has been traditional to vaporize LNG before burning
it. Whether this will be necessary or desirable for future LNG-
fueled power plants is open to question. Existing LNG base load,
peak-shaving and peak-shaving satellite plants serve gas pipe-
lines and, therefore, vaporization of LNG is mandatory. Fortu-
nately, there are a variety of reliable vaporizers available and
these do not represent an inordinate part of the invested capital
of the LNG system. Fired vaporizers absorb approximately 2% of
the heating value that they transmit to the pipeline (IN-029).
-64-
-------
Direct-Fired Vaporizers
The direct-fired vaporizer consists of stainless steel
finned tubing stacked on a rectangular chamber. LNG comes in at
the bottom and leaves as a gas out the top. A hot inert gas at
about 1000°F flows in a dual elliptical path and maintains a
heat flux of about 20,000 Btu/hr/ft2 of pipe surface. The gas
circulates an average of 5 times around this bank before going
out the stack. The efficiency of this unit is about 70-75% and
can be improved to 877» if heated air is returned to the blower
(IN-029).
Submerged Combustion Vaporizers
Figure 2.2-6 illustrates the principles of the submerged
combustion heat exchanger. The LNG is circulated in stainless
steel tubes that are immersed in a hot water bath. High pres-
sure fuel and air are burned in the downcomer and the products
of combustion flow through the water bubbling up inside the weir.
The efficiency of this system is 90-9570 based on the
higher heating value of the fuel. Also, the submerged combustion
system has a longer thermal reservoir than the direct-fired ex-
changer. This tends to provide a cushion for sudden fluctua-
tions in demand.
Indirect-Fired Intermediate Fluid Vaporizers
Figure 2.2-7 shows a typical indirect-fired system.
In this system, the LNG is vaporized and superheated in a heat
exchanger with pentane. The pentane is iri turn heated by a
water-glycol mixture which is heated by a natural gas-fired
furnace.
-65-
-------
NG Ouf/ef
Beth Level
Dov/ncomer
Sfac/<
LNG Inlet
FIGURE 2.2-6 SUBMERGED COMBUSTION
VAPORIZER
SOURCE: (IN-029)
Ton ft
IVafer Overflow
V/eir
Tube Coil
-66-
-------
I.NC STORAGE TANK
LNG
VAPORIZER
PENFANE WATER
HEAT EXCHANGER
r-£T
PENTANE
LOOP
NATURAL GAS
TO DISTKIOUriON 1 r
< t_l
300 Psig MAX
~60°F
V/ATER HEATER
jl
J
x^\ •
'
>
»
HOT
V.'ATER
LOOP
-•
NA
TURAL GAS
FIRED
) k
|
FIGURE 2.2-7 INDIRECT-FIRED INTERMEDIATE
FLUID VAPORIZER
SOURCE: (IN-029)
-67-
-------
Ambient Air Vaporizers
This type of vaporizer has been used in small appli-
cations, but would not be suited for the large demands of base
load plants due to the low rate of heat transfer per unit area
of heat exchanger. A very large surface area would be required
to vaporize the LNG. Ambient air temperature, winter to
summer, would have a significant effect on output. This type
of exchanger uses aluminum-plate fins and special cross-section
tubing to provide as much area for heat exchange as possible.
Unfired Water Vaporizers
In these vaporizers LNG is pumped into manifolds in
the bottom of banks of vertical tubing. The tubing is internally
finned and water flows down the outside of the tubes in a thin
film. The formation of ice on the tube banks is controlled by
regulating the amount of water flow. Such installations compare
well with other gasification systems when utilized in connection
with a coastal LNG terminal receiving tank shipments from foreign
sources.
2.2.2 Base Load Plant
Base load LNG plants are based upon the economy of
tanker shipment of methane in the liquid phase. Large lique-
faction facilities are located in areas of abundant natural gas
supply, and the corresponding regasification units are sited to
serve large consumption centers.
Only a little over 10 years have passed since LNG
first reached commercial status. Now LNG is the principal means
of transporting excess gas from world areas where there is plenty
to areas where energy is short. So far there are more than 5,000
-68-
-------
MM scfd of capacity in operation or under construction around
the world.
Table 2.2-1 illustrates that this is equivalent to
nearly 10% total consumption of the United States or about 5%
of world consumption. In addition to these projects, a number
of others are under various stages of consideration (Table
2.2-2). In all it has been estimated that about 140,000 to
230,000 MM scfd of natural gas is available in the world for
LNG movement (WA-168).
Major LNG projects can be described in terms of
a number of basic component parts. One can differentiate
between the components of a base load LNG project as those
which are located at or near the gas field and those which are
located at the receiving terminals. The components at or near
the gas field are:
gas production facilities, i.e.,
gas wells, field lines, measurement
and control equipment, pressure re-
duction and initial purification
facilities, and well-servicing
equipment;
pipelines from the gas field to the
liquefaction plant, generally from
inland or submarine fields to a
suitable deep water port for ships
in the 100-150,000 dwt class;
liquefaction plant which normally
includes gas purification, since
frequently gas is piped impure;
and
-69-
-------
TABLE 2.2-1
LNG PROJECTS
Project
Algeria-U.K
Algeria-France
(Arzew to Le Havre)
Alaska-Japan
Libya-Spain
Libya-Italy.
Algeria- France
(Skikda to Fos)
Abu Dhabi- Japan
Algeria
(Bethiova)
(Kalimantan)
(Lho Seumawe)
Algeria
(Skikda)
Total
Principal*
CAMEL
British Methane
CAMEL
Gaz de France
Phillips
Marathon
Gaz Natural
EN I
Shell
Brunei Govt.
Mitsubishi
Gaz de France
ADNOC
BP
Mitsui
CFP
Bridgestone
Sonatrach
El Paso Natural Gas
Huffco
Mobil
Berlin
1964
Compl.
1966
Compl.
1969
Compl.
1971
Compl.
1972
Compl.
1972
Compl.
1973
Compl.
Under
constr. (1976)
Under
constr. (1976)
Under
constr. (1977)
Under
constr. (1977)
Under
constr.
Approi.
capacity,
MMcfd
100
SO
160
120
330
770
380
330
1,000
650
1200
175
5.165
TABLE 2.2-2
PROPOSED LNG PROJECTS
Project
(Bethiova)
Indonesia-U.S
Iran-U.S
Alaska/U.S. Mainland.
Participants
Sagape
Pertamina
Pacific Lighting
NIGC
El Paso
Sopex
Distrigas
Pacific Lighting
Status
Engineering
Reported under
negotiation
Reported under
negotiation
negotiation
Mid-1979
Early 1980s
Proposed
Reported under
consideration
Under consideration
Approx.
capacity,
MMcfd
1.550
550
2.000
then
3.000
500
3.600
-70-
-------
liquid .storage of sufficient
capacity to load ships without
causing delay.
Those components located at the receiving terminal
include:
a suitable harbor for handling the
LNG tankers,
storage tanks for LNG large
enough to receive entire ships'
cargo without causing delay,
regasification facilities for the
LNG, and
connecting pipelines with pres-
sure regulators, measuring equip-
ment, odorizers, etc., to connect
the terminal with existing gas
systems.
The following sections discuss those components of
base load liquefaction plants which were not covered in the
peak-shaving LNG sections. The technology review will include
complete gas clean-up operations, heavy hydrocarbon recovery,
and LNG transportation via tanker.
2.2.2.1 Natural Gas Conditioning and Purification
Natural gas supplied by pipeline to a base load
liquefaction plant from fields situated within about 100
miles of the plant is as a rule only purified at the well-
head to a minimal extent. This means that water, acidic
-71-
-------
gas, i.e., carbon dioxide and hydrogen sulfide, higher hydro^
carbons and other impurities such as dirt or entrained oil
droplets may be present in the liquefaction feed.
Before natural gas can be liquefied, these contami-
nants must be removed, since they would solidify on cooling
and plug the piping or foul the heat exchangers.
Liquid Separation
The first stage of any gas conditioning operation
will consist of a trap for the collection of liquid products
present in the feed to the liquefaction plant (LO-102). Depend-
ing on ambient temperature, water content of the gas, and pressure
drop, glycol or methanol may be injected as an antifreeze. If
this is done, glycol or methanol recovery from the aqueous layer
on the gas/liquid separator will be required. This usually
involves fractionation in a small atmospheric distillation plant.
After reduction of liquid water, glycol and heavier
hydrocarbons by simple gas/liquid separation, the gas is cooled
by heat exchange to a temperature near freezing. At pipeline
pressure, this results in further condensation, and additional
water and heavy hydrocarbons separate out in a knock-out drum.
Acid Gas Removal
The process which follows next in gas processing
is generally referred to as gas sweetening and serves to remove
both H2S and COa. The acid gases present in the natural gas
have a limited solubility in LNG. Their concentrations have
to be reduced to avoid freezing-out in the liquefaction unit,
and consequently plugging or fouling of the heat exchangers
and piping. Carbon dioxide removal to levels of 50 ppm and
-72-
-------
less is essential, and HaS removal to even lower concentrations
is required of the pretreatment facilities for an LNG installa-
tion.
There exist two basic methods for combined C02/H2S
removal - dry or wet systems. The dr.y system that is used by
the LNG industry is the molecular sieve. Its major advantage
over the wet sweetening systems is the degree of clean-up
achievable as well as simultaneous dehydration. For a descrip-
tion of this system see Section 2.2.1.1 on the peak-shaving
clean-up discussion.
The wet sweetening systems can operate by two basical-
ly different mechanisms, i.e., a reversible chemical reaction
may take place between the acid gas and the solvent, or alter-
natively the acid gas may merely dissolve in the absorber liquid,
in preference to and generally at a faster rate than the other
gas components.
Typical chemically reactive solvents include aqueous
solutions of most alkanolamines such as monoethanolamine (MEA),
diethanolamine (DEA), diglycolamine (DGA), di-isopropanolamine
(Adip), triethanolamine (TEA), and anthraquinone disulphonic
acid (Stretford solution). In all these extractions, with the
exception of the Stretford process, acid gases are absorbed at
near ambient temperature by the alkaline compound and are re-
leased by heating to near its boiling point. Figure 2.2-8 shows
a typical amine treating unit." The Stretford solution, on the
other hand, also contains sodium vanadate, sodium carbonate and
a trace of chelated iron. When blown with air, HaS is oxidized
to elemental sulfur, which can be removed by filtration.
A series of absorption solvents based on potassium
carbonate act in similar fashion to the alkanolamines. In the
-73-
-------
ABSORBER
REACTIVATOR
Star
Lean Solution
Rich Solution
Acid Gas
FIGURE 2.2-8 TYPICAL AMINE TREATING UNIT
SOURCE: (HY-014)
-------
Benfield, Vetrocoke, and Catacarb processes, carbon dioxide
reacts with potassium carbonate to form bicarbonate, which
decomposes at elevated temperatures. A similar reaction
takes place with H2S. Various additives, frequently arsenates,
accelerate H2S removal by forming thioarsenates, which decom-
pose into arsenates and elemental sulfur (Giammarco Vetrocoke
process). Catacarb and Benfield additives assist the rate of
gas absorption by accelerating hydration of C02 gas.
Physical absorbents for acidic gases include anhydrous
propylene carbonate (Fluor solvent), N-methyl-pyrrolidone
(Purisol), and the dimethyl ether of polyethylene glycol (Sele-
xol). In certain instances physical absorbents need not be
heated but can be flashed at reduced pressure to release the
absorbed acidic gases. Their main disadvantage, compared with
chemical absorbents, is their tendency to remove higher hydro-
carbons from the gas, which is particularly undesirable where
sulfur is to be recovered from the acid gas in a Glaus plant.
Another disadvantage of chemical absorption is the
highly corrosive nature of both absorbents and, particularly,
absorbent-acid compounds. In an attempt to find an acceptable
compromise, hybrid processes have been developed such as the
Sulfinol extraction process which uses a mixture of the physical
solvent sulpholane and chemical absorbents of the alkanolamine
type.
Typical operating conditions for a number of reactive
solvents are listed in Table 2.2-3. A number of other solvents,
both chemically reactive and physically absorbent, have been
proposed and used commercially to sweeten natural gases from
various sources. However, the above-mentioned processes account
for the bulk of modern gas purification plants. It should be
noted from Table 2.2-3 that while most processes adequately
-75-
-------
TABLE 2.2-3
Gas Sweetening Processes
Process
Adip
Benfield
Catacarb
Econavrrine
Fluor
Girbotol
Purisol
Seiexol
Stretford
SNPA
Su'ifinol
Vetrocokc (CO2)
(H2S)
Solvent
Di-isopropylarnine
Activated K2CO3
Activated K2CO3
Digiycolarnine
Propyiene carbonate
MEA
DEA or TEA
Ar-methylpyrrolidone
DimethoxypoiygSycol
Amhraquincr.edisulphonic acid,
Na-CO.,, AsiOj, Na2VO4> chel.
Fs compound
Modified DEA
Sulpho'ane, alkanol-amine
K2CO3, As2O3
NazCO3, As2O3
Type Typical concentrations
Chem. P'nys. Initial Residual (ppm)
H2S COi H2S CO2
X
X
X
X
X
X
X
Hybrid
X
X
Bulk*
H2S/CO2>1
H2S/CO2>1
3-20 vol %
X Bulk
1-3 vcl %
3-10 vol %
X Bulk
X Bulk
<0-5%
Bulk
Bulk
—
>1-0%
—
Bulk
Bulk
Bulk
Bulk
Buik
Any
Bulk
Buik
Bulk
—
0-5
30
30
0-5
0-5
0-5
1-2
0-5
0-5
0-5
0-5
0-5
—
5
150
150
1 000
1 000
1 000
1 000
3 000
3000
—
1000
3000
10-30
—
*Predominant component in the gas
SOURCE: (IN-029)
-------
remove tUS, none accomplish a thorough enough job of C02 removal
prior to liquefaction. Therefore, it would be necessary to have
a molecular sieve as a final pretreatment of the natural gas
before it is liquefied.
Dehydration
After removal of acidic impurities by means of a
chemically reactive solvent, the gases are generally saturated
with water (LO-102). Water vapor is probably the most common
undesirable impurity in natural gas streams. It is not the
water vapor itself that is objectionable, but rather the liquid
or solid phase that may precipitate from the gas when it is
compressed or cooled. Liquid water almost always accelerates
corrosion, arid ice or solid hydrates can plug valves, fittings,
and even gas lines. To avoid these problems and those related
to actual ice formation in the cold box of an LNG plant, it
has been determined that water concentrations in the incoming
gas should be reduced to 1 ppm (IN-029).
This dehydration is usually accomplished in one of
two ways at a base load LNG plant. The wet gas may first be
passed to a glycol unit which is followed by a small molecular
sieve system or it may be dehydrated in one step by passage
through a large molecular sieve system.
In the flow diagram of a typical glycol dehydration
plant pictured in Figure 2.2-9, water vapor is continuously
absorbed from the process gas stream by countercurrent contact
with a high concentration glycol solution in a packed or bubble
tray column. The dried gas passes out the top of this column,
with the dilute glycol passing to a regenerator section where
the glycol is concentrated to levels as high as 99.8 percent.
-77-
-------
Stripping Gas arid Water Vapor Out
oo
i
Dry Gas
jlycol-Gas
.Contactor
I • \
Wet Gas | / \ j Rich Glycol
FIGURE 2.2-9 TYPICAL GLYCOL DEHYDRATION UNIT
SOURCE: (HY-014)
-------
There are other dry dessicants available besides the
molecular sieve for dehydration of the natural gas. However, for
such reasons as a high sensitivity to poisoning, decline in rated
capacity with pressure, and relatively short life, these systems
do not appear attractive for LNG application (IN-029).
Sulfur Recovery
A means of disposal of the sulfur compounds separated
in the acid gas removal units is needed in a base load LNG
plant. A Glaus sulfur recovery plant can provide an efficient
means of converting the removed sulfur compounds to elemental
sulfur for disposal.
The original technique for conversion of H2S to sulfur
was the Glaus-Chance Process. It has been modified considerably
in recent years. In fact, the Mathieson Chemical Company has
developed a considerably improved process which has been used
successfully in quite a few modern installations. A flow dia-
gram of the modified Glaus-Chance used in the Mathieson Process
is shown in Figure 2.2-10.
The first step consists of burning the feed gas in a
specially designed reactor furnace. Flue gases from this furnace
are partially cooled in a waste heat boiler and then run to a
catalyst converter. After passing through the first stage of
the converter the gases are run to a boiler feed water economizer
and then back through the second stage of the converter.
The final step is to pass the gases into a wash tower
where they are cooled and the sulfur is condensed by direct
contact with a recirculated stream of liquid sulfur. Exit
gases from the tower are primarily N2, C02, and water vapor with
some S02, H2S, COS, and CS2.
-79-
-------
i
c»
o
Sulfur
Sulfa
FIGURE 2.2-10 GLAUS SULFUR RECOVERY UNIT
SOURCE: (IN-029)
-------
Sulfur removal efficiencies are dependent on the
hydrogen sulfide concentration in the acid gas fed to the unit,
the number of catalytic stages and the quality of the catalyst
used. They can range from 90 to 98 percent.
Under the conditions prevailing in the reaction furnace,
formation of some carbonyl sulfide (COS) and carbon disulfide
(CSa) is inevitable if the acid gas contains COa and hydrocarbons.
Although the amounts of COS and CS2 formed are relatively small,
especially if the hydrocarbon content of the acid gas is low,
they are significant as potential air pollutants. A special
catalyst may be placed in one or several of the catalyst con-
verters to largely hydrolyze COS and CS2 to HaS and COa, and
thus prevent CSa and COS from escaping into the atmosphere
(IN-029).
2.2.2.2 Heavy Hydrocarbon Stripping
Heavy hydrocarbons are recovered from natural gas
streams for both economic and operational reasons. The heavy
hydrocarbon components of a gas may be worth considerably more
when condensed and sold as a liquid than when sold as a gas.
Another reason for recovery is that the presence of even small
amounts of liquids in a pipeline can easily reduce the efficiency
of gas flow by 10%, since liquids increase the pressure drop
required for a given flow rate. Also, the presence of heavy
hydrocarbons in the natural gas entering a liquefaction unit
can result in freeze-ups in the heat exchangers or require
inclusion of additional liquid separators and special piping in
the cold box to remove these materials from the process gas
stream.
Molecular sieves and other solid adsorbents offer one
method for the removal of heavier hydrocarbons from natural gas
-81-
-------
streams. However, they have a much smaller capacity for adsorbing
hydrocarbons than they have for water vapor. Therefore, these
types of heavy hydrocarbon recovery units are seldom used in
LNG plants.
Refrigerated Absorption
Refrigerated absorption offers an economical means
of recovering ethane, propane, and higher components in a natural
gas stream. This type of plant can theoretically achieve desired
recoveries at any practical temperature and pressure by circu-
lating the required amount of absorption oil.
As can be seen in Figure 2.2-11, high pressure natural
gas, after drying and acid-gas removal, flows to a demethanizing
absorber operated at essentially feed-pressure. In the absorber,
uhc feed gas is contacted with refrigerated absorption oil which
can be composed of natural gasoline components recovered from the
gas itself or some other hydrocarbon oil.
Rich oil from the bottom of the absorber is sent through
heat exchangers to a stripper for regeneration. Light components
are removed from the rich oil in the stripper by having the oil
count:ercurrently contact open steam. The overhead product from
the"still is condensed and separated from water. This condensate
is then fractionated in a distillation column to recover the
individual components ethane, propane, butane, and the stabilized
natural gasoline.
High recoveries of ethane using this process are
uneconomical, due to the large steam requirement and amount of
oil that must be circulated. Yet it is a. favorable process •
for LNG rcpjote locations since the refrigerant (propane) and
the absorption oil (natural, gasoline.) can be recovered, from the
feed gas itself.
-82-
-------
DEMETHANIZING DEETHANIZING
DRYER ABSORBER ABSORBER DEBUTANIZER DEPROPANIZER
oo
Natural
Gas Feed
Start-
Natural Gas Product
cw
Propane Refrigerant
*» Propane Product
Butane Product
Gasoline Product
FIGURE 2.2-11 REFRIGERATED ABSORPTION FOR RECOVERING
HEAVY HYDROCARBONS
SOURCE: (HY-014)
-------
Low Temperature Fractionation
This method of heavy hydrocarbon recovery can be
efficiently utilized in an LNG plant. Figure 2.2-12 shows the
flow diagram for this technique of recovery. The natural gas is
cooled to a point where all the Ca and higher components have
liquefied and then fractionated in a tower designed for separat-
ing the methane from these liquefied heavier hydrocarbons. The
final natural gas liquefaction process used has little or no
effect on the hydrocarbon recovery.
The recovered heavy hydrocarbons are usually used as
make-up refrigerant for the cold box and for plant fuel.
2.2.2.3 Liquefaction Cycles
Those liquefaction cycles presented in Section 2.2.1
of the peak-shaving technology discussion, with the exception
of the expander cycle, are used in LNG base load plants as
well.
2.2.2.4 Storage
The storage facilities used at the base load plant,
as well as at the receiving terminal, are usually double-walled
metal tanks. See the section on storage in the peak-shaving
discussion for a description of the storage alternatives available,
I
2.2.2.5 Transportation
In base load LNG operations where natural gas is lique-
fied for export from countries with a surplus to areas deficient
in gas, the LNG is pumped from storage, through deep-water load-
ing facilities, to ocean-going vessels suitable for the long
-84-
-------
NATURAL
GAS
FEED
610 PSIA
595 PSIA
PIPELINE
GAS
H2OB C02
REMOVAL
TAIL
GAS
18 PSIA
WARM END
EXCHANGERS
TT
COLD END
LIOUEFIERS
REFRIGERATION
, 600 PSIA
I -I20CF
165 PSIA
-133 CF
•~T"
DEMETHANI2ER
+ 75°F
STORAGE
ETHANE+
PRODUCT
FIGURE 2.2-12 LOW TEMPERATURE FRACTIONATION
SOURCE: (HY-014)
-85-
-------
distance transport of such a specialized cargo. A typical
capacity for ships of current design is the equivalent of
about 2,000 MM SCF.
The LNG tankers differ from other ships in the design
and construction of their cargo tanks, accommodation for a vapor
reliquefaction unit, and use of vaporized cargo as fuel (LO-102).
Very careful design of the cargo tanks and their surroundings is
essential. Two basically different methods which have been used
are the construction of self-supporting or free-standing LNG
tanks or the use of the hull as support for insulating layers
and gas-impermeable membranes. Either type effectively provides
the safety of a double hull.
.The principles involved in reliquefaction of LNG
vapor on board a ship are the same as those of the full-scale
shore liquefaction plant, but there exist minor differences.
High liquefaction performance and plant efficiency are normally
sacrificed in shipboard equipment in favor of low weight and
plant size. Reciprocating machinery is used, partly because of
the smaller gas volumes to be compressed and also for reasons
of flexibility (LO-102).
While it is possible to reliquefy all the gas vaporized
by heat leakage into the LNG tanks, and this is in fact essential
when the ship is loaded with LNG and stationary, there are al-
ternative means of vapor disposal. In particular, the ship's
propulsion engines and auxiliary boilers can be run on methane,
whether they are of the steam turbine or diesel type. This is
the case provided that the boil-off gas is pressurized and pre-
heated before combustion and that provision has been made for
dual combustion in the equipment (LO-102).
-86-
-------
2.2.2.6 Regasification
After the LNG has been unloaded from the tanker into
the receiving terminals storage facilities, it is regasified
and injected into a pipeline distribution main. This operation
will be a continuous operation unlike the regasification at
peak-shaving plants where it occurs only to meet high gas demand.
Section 2.2,1 describes those alternatives available to base
load receiving terminals.
2.2.3 Satellite LNG Facilities
A particular form of peak-shaving plant is the so-
called satellite LNG facility. This generally consists of a
storage tank, a vaporizer, and odorization equipment. The tank
is filled by truck, rail, or barge transportation of LNG from
a peak-shaving liquefaction plant. Several satellites can be
supplied from one central liquefaction plant (LO-102). Satellite
LNG plants usually operate unattended, the flow of LNG to the
vaporizer being regulated by the gas pressure in the distribu-
tion grid.
In addition to their function as peak gas producers,
satellites can also be used to distribute gas in new areas which
are not connected to the main supply system. Under these cir-
cumstances, a local LNG tank is filled regularly from the central
tankage throughout the year.
-87-
-------
2.3 SNG Production Technology
Synthetic natural gas . (SNG) from liquids may be produced
using either of two basic feedstocks, crude oil or a light
petroleum such as naphtha. The official policy of the Federal
Energy Administration, however, is to discourage manufacture of
SNG from oil as an inefficient (8-10% energy loss), uneconomic
use of resources (FE-085). The production of SNG from light
petroleum derivatives such as LPG and NGL is also discouraged.
As a result, the 13 SNG plants that are currently operating or
under construction in the U.S. plan to use light naphtha as a
feedstock or switch from a light petroleum derivative to naphtha.
It is estimated that plants producing SNG from crude oil will
not be utilized in the United States in the near future. For
this reason the SNG technology addressed in this section applies
to processes employing a light naphtha feedstock (360-370°F
boiling end point).
2.3.1 Processing Steps
The preferred method of producing SNG from naphtha in
large, quantities is catalytic gasification followed by methana-
tion. The basic process is available from the British Gas
Council (CRG - Catalytic Rich Gas Process), Japan Gasoline Co.
(MRG - Methane Rich Gas Process), or BASF/Lurgi (Gasynthan
Process), either directly or through their licenses. Each of
these companies is currently operating gasification processes
in a number of commercial plants. The four basic steps in the
process are shown in Figure 2.3-1. These processing steps are
desulfurization, gasification, methanation, and purification
(carbon dioxide removal and dehydration).
Desulfurization
The sulfur content in the naphtha feedstock to an SNG
plant can range anywhere from one ppm to 1000 ppm (BR-103).
-88-
-------
I
00
VO
I
Nahtha
Feed
^_ SNG
Product
Sulfur Removal
Gasification
Methanation
CO
Removal
FIGURE 2.3-1 SNG PLANT PROCESSING SEQUENCE
-------
Since sulfur compounds represent a poison to the gasificatipn
catalyst, the sulfur content must be lowered to 0.2 ppm to avoid
any significant catalyst poisoning. Desulfurization is accomplished
in two stages: the first consists of catalytic conversion of
sulfur compounds to H2S; the second involves removal of the H2S
from the naphtha.
After vaporizing the naphtha feed and mixing it with a
recycled hydrogen stream, the mixture of vaporized feedstock and
recycled hydrogen containing gas is heated to about 650-700°F
and passed over a desulfurization catalyst. The active ingredi-
ent is generally nickel-molybdenum. Organic sulfur compounds
in the feedstock, mercaptans and thiophenes, are converted to
H2S. Olefin saturation and minor cracking occur. Some methana-
tion of carbon oxides may also occur, depending upon the charac-
teristics of the catalyst.
After the organic sulfurs have been converted to H2S,
two process alternatives for the removal of the H2S exist. In
one case, the treated gases are passed over a bed of zinc oxide
which absorbs H2S and forms zinc sulfide. Besides zinc oxide,
less costly iron oxide can be used. However, during upset con-
ditions, when the concentration of hydrogen is unusually high,
iron oxide and iron sulfide may be reduced to metallic iron,
with release of hydrogen sulfide. Therefore, a bed of zinc
oxide is generally used as a final step.
The second alternative available for the removal of the
H2S is to route the reactor effluent to a stripper or frac-
tionator where the sour gas may be separated. The naphtha
product containing 0.2 ppm sulfur is taken from the bottom of
the stripper. Figure 2.3-2 shows the steps involved in this
type of desulfurization (RA-119).
-90-
-------
Feed,-
Make-Up—
Hydrogen
Hydrogen
Recycle
Reactor
Preheater
Separator
Sour
Gas
Stripper
or
Fractionator
«•—J— Steam
Naphtha
Product
FIGURE 2.3-2 NAPHTHA HYDRODESULFURIZATION UNIT
-91-
-------
Once the H2S has been removed from the naphtha stream,
the sour gas may be flared, burned in a preheater or boiler, or
routed to an amine treating unit. A choice between these al-
ternatives is based on the economics and local air pollution
regulations for each SNG plant.
Catalytic Gasification
After desulfurization, steam is added to the purified
feedstock, the temperature of the mixture is increased to about
750 to 850°F, and the mixture is sent to an adiabatic catalytic
reactor. Operating pressures range from 200 to 600 psi. Sub-
stantially higher pressures have been tested in pilot-plant
operations but not yet in commercial installations. The reactor
essentially consists of a catalyst filled vessel. The gasifica-
tion catalyst used is nickel based, but may contain promoters to
enhance performance characteristics.
The preferred feedstock to the gasifier is LPG or a
light naphtha. The British Gas Council, Japan Gasoline Co.,
and BASF/Lurgi processes can all use hydrocarbon feeds as heavy
as straight-run naphtha having a distillation end-point of about
356-365°F. The Lurgi process utilizes a catalyst that can
gasify naphtha having an end-point as high as 400°F. Naphthas
with higher end-points have been tested successfully in pilot-
plant work, and it has been recently reported that the British
Gas Council CRG process can handle naphtha feedstocks with a
final boiling point up to 465°F (BE-246). However, catalyst
life generally decreases as the average molecular weight and
aromatic content increase. In addition, olefins tend to crack
and form carbon on the catalyst, so only limited heavy materials
amounts can be tolerated in the feedstock.
-92-
-------
One feature of the Japan Gasoline Co. process is a
catalyst that tolerates feed containing up to 3 ppm sulfur. It
is reported that this catalyst is promoted with copper and
chromium oxides (BR-103).
A portion of the gasified effluent from the catalytic
gasifier can be recycled back to the entrance of the gasifier.
This recycle is particularly desirable when gasifying heavy
naphtha, but less advantageous when feeding light naphtha.
Range of composition for the gas leaving the reactor
is typically as follows:
Analyses (dry)
Component Volume %
CH,, 60-75
C02 20-22
CO 0.5-1.0
H2 10-18
An average Btu content for this stream in the CRG process is
680 Btu/scf (HY-014). The component analysis and Btu content
of this stream is a function of the type process used as well
as the operating conditions of the unit.
Methanation
Methanation is essentially a continuation of the gasi-
fication stage, but it occurs at a lower temperature to promote
formation of methane. In some designs, the same catalyst may
be employed for methanation as for gasification, or the cata-
lyst may have the same basic composition but with different
promoters.
-93-
-------
One or two stages of methanation may be used, although
it is more common to use two stages when gasifying naphtha. The
number of stages, the operating temperature, and amount of
steam left in the reacting gas are dependent on the product-gas
specification (carbon monoxide and hydrogen contents, as well as
heating value) and catalyst performance (i.e., ability to pro-
mote methanation in the presence cf significant quantities of
steam, and ability to avoid carbon formation in the presence
of minuimum quantities of steam).
One process alternative possible in the methanation
step, particularly when treating a light naphtha, is to intorduce
part of the naphtha feed directly from the vaporizer into the
first-stage methanator (which is the hydrogasification step in
the British Gas Council process). This reduces total energy
requirements but may increase catalyst investment costs and
makeup requirements. The optimum design appears to be largely
a function of naphtha composition.
The methanation step uses adiabatic catalyst beds.
Representative temperatures may be 625-725°F in the first bed
and 570-620°F in the second. A representative flow diagram
for the gasification of naphtha using two methanators is shown
in Figure 2.3-3.
Purification
The gas exiting from the methanation section contains
significant amounts of carbon dioxide and some water vapor.
Both of these components of the gas stream must be removed. The
C02 must be taken out for improvement of the product-gas quality
and heating value, and the H20 removed to prevent condensation
and corrosion in the transmission system. A typical analysis
(dry basis) of the gas exiting from the final methanator is (HY-014)
-94-
-------
Sulfur Removal
CRG
A
Naphtha
i
.0;
Fired
Heaters
A
Fired
Heater.
HP
Steam
Recycle Gas
Drying
Methanators
.SNG
CO2 Removal
FIGURE 2.3-3 SNG PLANT UTILIZING TWO-STAGE METHANATION
-------
CH., 79.0 mole 7»
H2 1.1 mole %
CO < 0.1 mole 7.
C02 19.9 mole 7,
Any conventional C02- removal system may be used to
improve product gas quality. Various activated hot carbonate
processes, or other proprietary designs may be employed. Some
of the more popular COz- removal systems are: Benfield,
Diglycolamine or Econamine, Fluor Solvent, Girbitol, Selexol,
Catacarb, and Sulfinol. A flow diagram of the Benfield process
is shown in Figure 2.3-4.
The Benfield process pictured is similar in many
respects to the other popular C02 removal methods mentioned. The
raw gas is contacted with potassium carbonate solution containing
Benfield additives in an absorber column. The C02 is absorbed
here under pressures which range from 100 to 2000 psig in dif-
ferent units. The rich solution from the absorber is let down
to about atmospheric pressure and stripped in a regenerator tower
to drive off the absorbed C02. The regenerated solution is then
recycled to the absorber and the C02 is vented to the atmosphere.
A small amount of methane is lost through this C02 vent. The
methane loss is estimated at about .074 Ibs per 1000 scf of SNG
produced (LO-095).
After C02 removal, the final step in the manufacture
of SNG consists of gas dehydration. Numerous processes are
available for the drying of the gas to meet pipeline specifica-
tions. Basically, the two alternatives are the dry-bed systems
and the wet-scrubber systems. Either type can adequately de-
hydrate the SNG. The water which is recovered can be used as
boiler make-up feed water or it can be used in the plant cooling
water system.
-96-
-------
230
I
VO
-vl
I
GAS OUT
FIGURE 2.3-4 BENFIELD C02 REMOVAL UNIT
SOURCE: (IN-029)
-------
2.4 New Technologies
Operations and processing sequences associated with
advanced energy systems are examined in this section in order
to identify any processing similarities which exist between these
new technologies and the petroleum refining, SNG, and LNG indus-
tries. The specific purpose of this new technology survey is
to identify areas which are sufficiently similar to operations
in the subject industries of this study as to have the same
emission sources and the same type of emissions. To the extent
that these areas of similarity are established, the monitoring
and emission control techniques associated with the refinery,
SNG, and LNG industries may be related to the new energy tech-
nologies . The technologies which are specifically considered in
this section are coal gasification, coal liquefaction, and
shale oil production.
"~ These new energy technologies may basically be divided
into the major processing steps of raw material preparation,
/ conversion, and product upgrading. The unique operations
f~" associated with these processes are primarily located in the
conversion step. Raw material preparation techniques, while
not duplicated in the refining, SNG, or LNG industries, are
similar to solids handling operations in industries such as
coal mining or rock quarrying. Product upgrading is normally
performed by conventional refining processes and consequently is
an area where emissions should be very similar to the industries
in this study.
Raw Material Preparation: Raw material preparation
involves crushing and sizing to the particular requirements of the
conversion process being employed. In general, conventional
preparation and solids handling techniques are used. The opera-
tions in this processing step do not resemble any procedures used
-98-
-------
in the refining, SNG, or LNG industries; however, the preparation
techniques are similar to operations in established industries
such as coal mining. Potential emissions and emission sources
which may result from this processing step include the following:
particulates from crushing and sizing
operations
particulates and combustion products
from thermal dryers
combustion products from internal
combustion sources
fugitive particulate emissions from
solids handling, transportation, and
ore stockpiles
ore stockpile run-off (weathering of
organics, leaching of water-soluble
components)
solid wastes from crushing and sizing
operations
solids and/or water from particulate
control systems.
Conversion: The converions step is normally the ~
^unique part of the process — the operation which not only
distinguishes the process from existing industries but also
characterizes the process within the specific energy technology.
This step is the heart of the process and, as a result, impacts
the amount of ore preparation and product upgrading required.
-99-
-------
The new technological developments associated with the new energy
systems are primarily related to this processing step. The
conversion step is not similar to any operations in the refinery,
SNG, or LNG industries. Conventional emission control techniques
may, however, be applied to the extent that common emission
sources resulting in fuel combustion and fugitive emissions
exist within this processing step.
Product Upgrading: All new energy technologies require
product upgrading since a marketable product is not produced
directly from the conversion step. In general this processing
step consists of product separation, gas recovery and purifica-
tion, liquid product upgrading, and by-product recovery. This
processing step utilizes conventional upgrading procedures and
consequently involves many operations which are similar or
identical to petroleum refining, SNG, or LNG processes. Processes
which may be involved in this upgrading step are as follows:
distillation
gas treating
sulfur recovery
hydrotreating or hydrocracking
thermal cracking or coking
ammonia separation
shift conversion
methanation
Emission sources in this step include fuel combustion emissions
from the various process heaters, sulfur compound emissions
(SO , COS, CS2) from the sulfur recovery unit, fugitive hydro-
X
carbon emissions, ammonia emissions from ammonia handling
facilities and hydrocarbon emissions from liquid product
^storage. Due to the metallic components in the coal and shale
feed, trace metallic emissions may also potentially result from
-100-
-------
these processes. Although some metals are primarily retained
in the ash and others may be detected in condensate streams
(AT-042, FO-026) little work has been published to date concern-
ing the specific fate of metallic components in the process
streams.
In addition to the operations related to the three
major processing steps, the new energy technologies utilize
auxiliary processes such as power generation and water treating
facilities. The auxiliary processes employ existing technology
and represent another area where similarities exist between new
energy systems and the industries of this study.
Therefore, the areas of new energy technologies which
are similar to the refinery, SNG, or LNG industries, include the
product upgrading processes and the auxiliary operations. In-
formation presented in this study for the established industries
should be applicable to these areas of the advanced energy sys-
tems. All of those areas of similarity represent air emission
sources. Assuming that all solid wastes are combined into one
waste stream, the impact of the coal ash or spent shale precludes
any similarities with waste from the refinery, SNG, or LNG in-
dustries. Likewise the trace elements and trace organics associ-
ated with the coal or shale make the water treating problems
involved with new energy technologies much more complex than
processes utilizing an oil or gas feed. Although certain as-
pects of water treating such as pH, temperature, and suspended
solids may be handled by conventional techniques, the overall
water management problem associated with new energy technologies
is not comparable to refinery, SNG, or LNG systems.
Despite the processing differences that occur in the
new energy systems, many of the emissions are still the same as
those encountered in established industries. Consequently,
such emissions as the criteria air pollutants (particulates,
SO NO , 00, HC) will at least initially be controlled in the
•v -*x
-101-
-------
same manner as discussed in this report. The more pressing
environmental problems associated with the new energy systems
involve the distribution and form of trace elements and trace
organics, water management (make-up and discharge), and solids
handling and disposal.
2.4.1 Coal Gasification
Coal gasification involves the production of fuel gas
by the reaction of the carbon'.dn the coal with steam and oxygen,
The processes of this energy technology may be divided into two
groups depending upon the heating value of the product gas.
Low Btu gasification processes produce a CO and H2 rich gas
which may have a heating value between 150-450 Btu/scf. High
Btu gasification processes utilize more extensive upgrading
operations to produce a pipeline quality gas of approximately
1000 Btu/scf.
Low Btu Gasification
The following processes are typically involved in
a low Btu gasification system:
coal preparation
oxygen plant (optional)
power and steam generation plant
gasifier
gas cooling
gas liquor separation
gas liquor and effluent water
treatment (ammonia separation)
gas purification
sulfur recovery
cooling water system
-102-
-------
The processing sequence is illustrated in Figure 2.4-1. The
oxygen plant is optional. If air is used as the oxygen source
a low Btu gas (150-300 Btu/scf) is produced, whereas, if pure
oxygen is fed to the gasifier, a medium Btu gas (300-450 Btu/scf)
results.
Emission sources associated with low Btu gasification
are as follows:
Air emissions result from coal preparation,
gasifier vents, process heaters, steam and
power generation, sulfur recovery, fugitive
particulate and hydrocarbon emission sources,
ammonia storage, and hydrocarbon storage.
Water effluents are projected to be con-
trolled for zero discharge (US-112, US-164).
Streams which may potentially contribute
to a wastewater stream are coal pile run-off
ash quench water, process wastewater, cooling
tower blowdown, gas purification blowdown, and
any water used in emission control systems.
Solid waste is generated as discard from
coal preparation, ash from gasifier, sus-
pended solids in the make-up water, particu-
lates from contr.ol systems, spent catalyst
(periodic), and miscellaneous solids
generated during cleaning and maintenance.
The gasifier is the unique part of this technology.
The type of gasifier used characterizes the specific process.
Although certain gasification procedures may in special situ-
ations be used in refineries, the gasification step is not
-103-
-------
VENT or
A
i
M
O
COAL ^
H,0
AIR
COAL
PREPARATION
\
/
POWER AND
STEAM PLANT
STEAM
i
1
OXYGEN PLANT
L
*
v^
NITROGEN
SULFUR
RECOVERY
UNIT
*
OXYGEN /AIR
f
GASIFIER
I
<
\
r
>- ASH
CONDENSATE
GAS
COOLING
i
f
. GAS LIQUOR
SEPARATION
fer
w. TA
i
GAS
PURIFICATION
(COS /H S
REMOVAL)
Y
R AND TAR OILS
FUEL CAS fc
NAPHTHA
V
POWER TO
PLANT USERS
RECLAIMED WATER
TO IN-PLANT USERS
GAS LIQUOR
AND
EFFLUENT WATER
TREATMENT
_&. AMMONIA
-*- PHENOLS
FIGURE 2.4-1 LOW BTU GASIFICATION PROCESS
-------
considered to be similar to the normal processing operations of
the industries of this study.
Areas such as coal preparation and oxygen generation
utilize conventional technology, but the refining, SNG, and LNG
industries do not normally utilize similar processes. Effluent
water treatment and ammonia separation operations may also
utilize existing technology; however, ammonia production is
not always economically feasible in petroleum refining and
water treating should not be considered similar due to the poten-
tial impact of trace elements in the coal. Processes which are
similar to operations of industries in this study include power
and steam generation, gas cooling, gas liquor separation, gas
purification, sulfur recovery, and cooling water systems. There-
fore, emission sources which are similar to the refining, SNG,
or LNG industries are:
fuel combustion emissions from power and
steam generation
sulfur compound emissions from sulfur
recovery (SO , COS, CS2)
X
fugitive hydrocarbon sources
hydrocarbon emissions from liquid by-
products storage
High Btu Gasification
High Btu gasification processes produce essentially
pure methane from the coal by adding hydrogen derived from
steam and discarding carbon in the form of C02 and/or char.
The main difference from low Btu processing is the inclusion
of shift conversion and methanation processes in the processing
-105-
-------
sequence. The processing sequence for a typical high Btu
gasification process is shown in Figure 2.4-2. Emission sources
are the same as for low Btu gasification with the addition of
the following sources:
fuel combustion emissions from process
heaters associated with shift conversion
and methanation
fugitive emission sources associated with
shift conversion, methanation, and com-
pression
water effluent from dehydration of
pipeline gas.
The same processing similarities also exist between high Btu
processes and the refining, SNG and LNG industries.
Emission sources which are similar to these industries
are
fuel combustion emissions from process
heaters
fuel combustion emissions from power and
steam generation
SO emissions from sulfur recovery
X
fugitive hydrocarbon sources
hydrocarbon emissions from liquid by-
products storage.
-106-
-------
o
«xj
I
AIR
COAL -»
H20 -»
POWER TO
PLANT USERS
RECLAIMED WATER TO
IN-P.LANT USERS
AMMONIA
PHENOLS
FIGURE 2.4-2 HIGH BTU GASIFICATION PROCESS
-------
2.4.2 Coal Liquefaction
The basis of coal liquefaction is the cracking of
the coal molecule and the addition of hydrogen or removal of
carbon to produce a liquid product. Coal liquefaction processes
fall into one of two categories - (1) processes that utilize
hydrogen to assist in cracking the coal molecule and increasing
the H:C ratio, and (2) processes that rely on thermal cracking
and the removal of carbon (carbonization processes) to increase
the H:C ratio. Although these two types of liquefaction processes
have different approaches and consequently, different technical
problems, these differences are primarily confined to the reac-
tor section of the process. For the purpose of this analysis
the two categories of liquefaction processes may be considered
at the same time.
Processes utilized in a coal liquefaction plant are
as follows:
^ • coal preparation
/ • hydrogen production (gasifier train)
v/ • coal conversion
1^ product separation
^ ^ gas treating and recovery
\--
^ • sulfur recovery
\/ d- liquid fuels hydrotreating
i. • power and steam generation
water treating
-- ' ammonia separation
v • cooling water systems
A processing sequence for the major processes is shown in
Figure 2.4-3. If a gasifier is employed for char utilization
and hydrogen production the operations associated with low
Btu gasification (Figure 2.4-1) may be included as part of the
gasifier train. Depending upon the complexity of the liquefaction
-108-
-------
a
FlMl
CM!
Coil ^
rr.par.tlo.
Ck>r
t
CtiUUr
(Hrdro|«a
^reduction)
• i i '
"~<«
-»
j
Prodw
1
•y4n>tr»*t*r* C«»
CM Tr
1 RMO
e.tlng _^ it|lfut _^
«.ry »«eo«rr
1 f
Baphth.
HDS
t
CM Oil
•93
^ Hcpbch.
^ CM- Oil
Sulfur
FIGURE 2.4-3 COAL LIQUEFACTION PROCESS
-109-
-------
plant and the flexibility desired, a shift converter and
methanation may also be included in the gas processing sequence
Emission sources associated with coal liquefaction
are as follows:
Air emissions result from coal preparation,
process heaters, steam and power generation,
sulfur recovery, fugitive particulate and
hydrocarbon emission sources, ammonia
storage, and hydrocarbon storage.
Water blowdown and waste streams are expected
to be sufficiently minimized to allow contain-
ment in evaporation ponds and, therefore,
result in a zero water discharge (BA-230,
HI-083). Streams which may potentially
contribute to a wastewater stream are coal
pile run-off, ash quench water, process
wastewater, cooling tower blowdown, gas
treating blowdown, and any water used in
emission control systems.
Solid waste is generated as discard from
coal preparation, ash from gasifier, sus-
pended solids in the make-up water, partic-
ulates from control systems, spent catalyst
(periodic), and miscellaneous solids
generated during cleaning and maintenance.
Coal liquefaction processing areas which are not
similar to the industries of this study are coal preparation,
coal conversion, and char gasification (hydrogen production).
Although conventional water treating techniques may be used,
water treating cannot be considered an area of similarity
due to the potential for trace elements and trace organics
-110-
-------
from the coal. The ammonia separation is accomplished with
conventional techniques; however, this by-product recovery
operation is essentially an optional process which may not be
economically attractive in many refining operations. | Processes
which may definitely be considered to have corresponding opera-
tions in the refinery, SNG, or LNG industries include product
separation, gas treating and recovery, sulfur recovery, liquid
fuels hydrotreating, power generation, and cooling water systems.
-—"~~
The emission sources which these processes represent are the
common sources between the new energy systems and the industries
of this study. These sources are as follows:
fuel combustion emissions from process
heaters
fuel combustion emissions from power
' plants
sulfur compound emissions from the sulfur
recovery stack (SO , COS, CS2)
-
-------
must be heated to approximately 900°F. This heating (retorting)
step is the base requirement of all oil shale processes. Oil
shale processes may be divided into two major classes, depending
upon whether the retorting occurs above or below ground (ex situ
or in situ).
In situ processing involves fracturing the shale, in-
jection of retorting fluids, retorting of the shale in-place,
recovery of the product, and shale oil upgrading processes.
The intriguing advantage of in situ processing is that the
massive solids handling and disposal problems associated with
ex situ processes may be avoided. In situ oil shale processing
is, however, still in the conceptual stage, whereas ex situ
processing, which relies on more developed technology, is much
more advanced. Since the shale oil upgrading procedures are
essentially the same for both in situ and ex situ processes, the
comparison of similarities between oil shale processing and
the refining, SNG, and LNG is made by examining ex situ processes
Processes involved with ex situ shale oil processing
are as follows:
raw shale preparation
retorting
spent shale moisturizing and disposal
product separation
gas treating and recovery
sulfur recovery
hydrogen production
delayed coking
liquid product hydrotreating
power generation
water treating
• . ammonia separation
cooling water systems
-112-
-------
A processing sequence for the major processes is shown in
Figure 2.4-4.
Emission sources associated with shale oil production
are as follows:
Air emissions result from raw shale
preparation, retort preheaters, spent
shale moisturizers, spent shale handling,
process heaters, steam and power genera-
tion, sulfur recovery, fugitive particulate
and hydrocarbon emission sources, ammonia
storage, and hydrocarbon storage.
Water effluents are projected to be con-
trolled for zero discharge (US-093, CO-175).
Streams which may potentially contribute to
a wastewater stream are raw shale pile run-
off, process wastewater, cooling tower blow-
down, gas purification blowdown, spent shale
disposal run-off, and any water from emission
control systems.
Solid waste is generated as discard from
raw shale preparation, spent shale,
suspended solids in the make-up water,
particulates from control systems, spent
catalyst (periodic), and miscellaneous
solids generated in cleaning and maintenance.
Areas of shale oil production which are not similar
to the industries of this study are raw shale preparation,
retorting, and spent shale moisturizing and disposal. Effluent
water treatment and ammonia separation operations may utilize
-113-
-------
r»ti
t
r. k> °" T'««*lng
•nd
Recover;
H,S
t
*»_., R" Sh«U n-r^rr -a. I'roduct
Shale Preparation ^ R.tort » Separation
1
Spent Sh«l«
Koltturizing
L
Sul
Recc
fur
)»ery
Hydrogen
Production
i
k
HDS
1
. .,«. C«V«r
I
Cat Oil
BOS
*• Sulfur
Rjdrog**
— -•» To HDS
Unit*
— *>N(phcha
— t»Cae Oil
Sbal*
Cofca
FIGURE 2.4-4 SHALE OIL PROCESS
-114-
-------
existing technology; however, ammonia production is not always
economically feasible in refining operations and water treat-
ing should not be considered similar due to the potential
impact of trace elements in the shale. Processes which are
similar to industries of this study are product separation,
gas treating and recovery, sulfur recovery, hydrogen production,
delayed coking, liquid product hydrotreating, and power genera-
tion and cooling water systems. Emission sources which are similar
to the refinery, SNG, or LNG industries are:
fuel combustion emissions from process
heaters
fuel combustion emissions from power
generation
sulfur compound emissions from sulfur
recovery (SO , COS, CS2)
X
fugitive hydrocarbon emissions
hydrocarbon emissions from liquid
product storage.
-115-
-------
3.0 . IDENTIFICATION OF EMISSIONS AND EFFLUENTS
Utilizing the typical industry processing sequences
developed in Section 2.0, representative process modules for
the following are presented in this section:
fuel oil refinery
gasoline refinery
SNG plant
• LNG plant
There are many processing alternatives possible in petroleum
refinery operations; however this discussion is limited to two
refinery modules. While fuel oil and gasoline production
represent two different areas of processing, together they
account for the bulk of refinery output. In addition these two
types are representative of a large number of existing re-
fineries as well as major demand areas for new refinery appli-
cations .
Module flow rates are determined assuming typical size
commercial plant operation and utilizing specific process yield
data. After each module is established in terms of processes,
flow rates, and energy.or fuel demand, the emission sources and
emissions are presented. Emissions are related to specific
sources and organized according to the media impacted (air
emission, water effluent, and solid waste). Only criteria
pollutants such as particulates, SO , CO, NO , and HC are
X X •
quantified. The water pollutants which are quantified include
BOD, COD, ammonia sulfides, total phosphorous, phenol, oil,
suspended solids, and dissolved solids. The solid wastes are just
considered as the total weight of solids produced.
-116-
-------
All of the emissions are related to the module basis,
allowing for convenient assessment of a typical plant impact.
Emissions from all modules are also adjusted to a common Btu
output basis for comparison of emission impacts among the various
technologies.
-117-
-------
3.1 Fuel Oil Refinery Module
3.1.1 Module Basis
The fuel oil refinery module is based on a typical
commercial size operation of 200,000 barrels per day crude
capacity.* The emission values calculated in the fuel oil
refinery section are, therefore, presented on this 200,000 BPD
basis. A summary of total calculated emissions from the fuel
oil refinery module is shown in Table 3.1-1.
3.1.2 Module Description
The crude feedstock for this module is assumed to have
a 31°API gravity and a sulfur content of 1.5 wt7o. The heating
value of the crude is assumed to be 5.8xl06Btu/bbl (BA-230).
Characteristics of the crude charge are summarized in Table 3.1-2
The processing sequences utilized in this module are
shown in Figure 3.1-1 along with the major process flow rates.
The liquid product yield resulting from this module per barrel
of crude is as follows:
motor gasoline 0.43 bbl
• light fuel oil 0.42 bbl
• heavy fuel oil 0.08 bbl
'"'All flow rates for this module are based on calendar days
-118-
-------
TABLE 3.1-1
SUMMARY OF ENVIRONMENTAL IMPACT
Fuel Oil Refinery Module
Basis: 200,000 bbl/day Crude Feed
Air (Ib/day)
Particulates 6,320
SOX 16,000
NOX 11,830
CO 1,200
Hydrocarbons 73,970
Water (Ib/day)
Suspended Solids 250
Dissolved Solids 9,260
Organic Material 52 „ 5
Solid Wastes (tons/day) 4.0
-119-
-------
TABLE 3.1-2
CRUDE FEEDSTOCK CHARACTERISTICS
Fuel Oil Refinery Module
Crude Characteristics
API0 Gravity
Sulfur Content
Heating Value
True Boiling Point Range
68/375°F
375/600°F
600/1050°F
1050+°F
31
1.5 wt%
5.8xl06 Btu/bbl
Wt% of Crude
20.4
19.1
36.0
24.5
-120-
-------
FUEL OIL REFINERY MODULE
LIGHT ENDS
12 20x10° Ib/day
Hn
18.500 Ib/dayJ
DESALTED
CRUDE
200,000 bbl/day
60.91xl05 Ib/day
1.5 wt% S
SULFUR
RECOVERY
TAIL GAS
TREATING
ELEMENTAL SULFUR
Ml.000 Ib/dav
EUMNTAL SULFUR
3,090 Ib/day
GAS TREATING
PLANT
NAPHTHA
HYDROTREATER
48,000 Ib/day
9.30xl06 Ib/day
2 216,000 Ib/dav
DE-ASPHALTED OIL
HYDROTREATER
T
734,000 Ib/day
COKE
128.000 Ib/dav
COKE GAS
(from gasifier)
1.49xl06 lb/HHy
FIGURE 3.1-1
-* TAIL GAS
1.797x10° Ib/day
FUEL GAS
0.1 Rr/scf 313,400 Ib/day
^2 404.700 Ib/day
c3
2.280x10° Ib/day
10.58x10° Ib/day
MOTOR
GASOLINE
>2.39xl05 Ib/day
32.950 hbl/day
^ LT. FUEL
OIL
26.86 xlO6 Ib/day
Si.OsO bbl/da-.'
O.D vt'. S
HEAVY
• FUEL
OIL
5.257x10 Ib/d»v
15.470 bbl/day
0.3 n't S
-------
The crude feed to the refinery is first desalted and
then routed to a crude distillation unit where the crude is dis-
tilled both at atmospheric pressure and in a vacuum column to pro-
duce four product cuts: (1) sour light ends (Ci-Ci* hydrocarbons),
(2) naphtha, (3) gas oil, and (4) vacuum resid. The resid pro-
duct is routed to a propane deasphalting unit for extraction of
gas oil. The deasphalted oil from this unit is then hydro-
treated for sulfur removal. The three product streams resulting
from hydrotreating are sour light ends, naphtha, and heavy fuel
oil. The heavy fuel oil is recovered and either routed to pro-
duct tankage or used to fire process heaters within the refinery.
The naphtha stream is routed to the gasoline blending area.
The light ends formed are routed to the gas treating plant for
acid gas removal.
The asphalt produced at the propane deasphalting unit
is fed to a flexicoker. The flexicoker acts as a fluid bed
coking unit and a coke gasifier. The fluid bed section produces
a sour naphtha stream and a sour light ends stream. The gasifier
produces the final coke product and a low Btu coke gas which is
sweetened and fired in the propane deasphalting unit process
heater. The light ends from the flexicoker go the the gas
treating plant. The naphtha product is combined with the
straight run naphtha cut from the crude distillation unit and
routed to a naphtha hydrotreater.
The crude naphtha and naphtha from the flexicoker are
hydrotreated for removal of sulfur. The sour light ends produced
go to the gas treating plant. The sweetened naphtha is split
into a light naphtha stream (True Boiling Point 68-180°F) and
a heavy naphtha stream (True Boiling Point 180-375°F). This
.-122-
-------
split is defined, assuming that the incoming stream is 20 wt
percent light naphtha and 80 wt percent heavy naphtha. The
light naphtha from the hydrotreater is fed to a C5/C6 isomeriza-
tion unit. The isomerization unit is used to increase the
octane rating of pentane and hexane fractions by catalytically
rearranging normal paraffins into isoparaffins. The heavy
naphtha from the hydrotreater is fed to a catalytic reformer.
The catalytic reforming process converts low octane naphtha
into high octane naphtha by catalytically rearranging and de-
hydrogenating naphthenes and paraffins, forming benzenes,
toluenes, and xylenes. The products of these two naphtha streams
are blended along with other petroleum components for motor
gasoline.
All of the collected sour light ends are amine treated
in the gas treating plant for removal of the H2S. The sweetened
light ends are then recovered as fuel gas, ethane (C2), propane
(C3), and butane (CO products. The fuel gas is burned in
process heaters, while the butanes are blended with the motor
gasoline. The ethane and propane rich product streams are
routed to pressurized storage vessels or product lines.
The acid gas from the gas treating plant goes to a
sulfur recovery facility for hydrogen sulfide removal. A Glaus
plant in conjunction with a tail gas treating unit is utilized
for sulfur recovery. In the Glaus plant the H2S is partially
combusted with oxygen and stoichiometrically reacted to form a
solid elemental sulfur product and water. Hydrogen sulfide re-
moval ranges from 95 percent to 98 percent in this unit (HY-014).
The tail gas from the Glaus plant is routed to a Tail Gas Treat-
ing Unit for additional sulfur removal. After conversion of
all sulfur species in the gas to H2S, the tail gas is contacted
with an alkanolamine solution for H2S removal. This final tail
gas treating unit results in a total equivalent sulfur removal
from the acid gas stream of greater than 99.8 percent (HY-014).
1 -123-
-------
The liquid wastes accumulated from the fuel oil re-
finery module are treated in both primary and secondary waste
water treatment facilities. The secondary treating can be either
activated sludge or aerated lagoons. The sludge from this treat-
ment is incinerated. Auxiliary units such as waste water treat-
ing facilities and incinerators are not shown in Figure 3.1-1.
The module heat requirements are calculated utilizing
the module flow rates and the specific process utility demand
information. The total module heat demand is 4.47 x 1010 Btu
per day. After specific module heat demands are established,
the allocations of the refinery fuels to meet these demands are
determined.
All of the fuel gas produced from the refinery module
is allocated and consumed within the refinery. Fuel gas is pre-
ferentially used in the smaller process heaters. The fuel gas
is capable of supply 7.47 x 109 Btu per day with a calculated
heating value of about 900 Btu per scf. The remainder of the
heat is essentially supplied by 0.3 wt percent sulfur fuel oil.
The fuel oil has a heating value of 6.3 x 106 Btu per barrel
(EN-071). Low Btu coke gas from the gasifier will supply a
minor portion of the heat demand. The coke gas has a heating
value of 1,598 Btu per pound and supplies 2.38 x 109 Btu per day
(FL-047). The specific fuels used in each refinery module unit
are shown in Table 3.1-3.
-124-
-------
TABLE 3.1-3
FUEL OIL REFINERY
MODULE HEAT REQUIREMENTS
Unit
Crude Unit
Gas Oil Hydrotreater
Naphtha Hydrotreater
Heavy Naphtha
Reformer
Propane Deasphalting
Deasphalted Oil
Hydrotreater
C5/CS Isomerization
Tail Gas Treating
Light Ends Recovery
Unit Heat
Requirement (Btu/day)
2.0 x
4.67 x 109
1.01 x 109
1.272 x 1010
3.65 x 109
1.57 x 109
7.78 x 108
1.80 x 108
1.40 x 108
Fuel Used
Fuel Oil
Fuel Oil & Fuel Gas
Fuel Gas
Fuel Gas
Coke Gas & Fuel Gas
Fuel Gas
Fuel Oil
Fuel Gas
Fuel Gas
-125-
-------
3.1.3 Module Emissions
3.1.3.1 Air Emissions
Air emissions from the fuel oil refinery module result
from fuel combustion, sulfur recovery, sludge incineration,
petroleum storage, and miscellaneous hydrocarbon emissions
throughout the refinery units. Module air emissions from the
specific sources are given in Table 3.1-4.
Fuel Combustion Emissions
Utilizing fuel demand data for the various processes,
fuel combustion emission sources are determined to be the
following (HY-013, HY-014):
crude distillation
gas oil hydrotreater
naphtha hydrotreater
heavy naphtha reformer
Cs/Ce isomerization
propane deasphalting unit
deasphalted oil hydrotreater
tail gas treating plant
light ends recovery
Although each unit may contain several fuel combustion emission
sources, all flue gas streams within one unit are assumed to be
combined and routed to one stack. Therefore, each unit requir-
ing fuel combustion represents one emission source. The emis-
sions from each unit are based on the type of fuel used, such as
fuel gas, fuel oil, or coke gas, and the EPA emission factors. .
These factors are shown in Table 3.1-5. The S0x emissions from
-126-
-------
TABLE 3.1-4
MODULE ATMOSPHERIC EMISSIONS
FUEL OIL REFINERY MODULE
(Ib/day)
Basis: 200,000 bbl/day Crude Feed
Particulates SOX C0_ Hydrocarbons NOX
Crude Distillation 3,080 6,400 533 533 5,330
Gas Oil Hydrotreater 280 539 90 128 1,084
Naphtha Hydrotreater 19 27 16.2 27.7 219
Hvy. Naphtha Reformer 1,960 4,073 340 340 3,390
Propane Deasphalting Unit 474 673 49.6 84.8 680
Deasphalted Oil Hydrotreater 30 42 25.2 42.9 340
,L Tail Gas Treating 3.4 3,410* 3.0 4.9 39
N5
71 Light Ends Recovery 2.6 3.7 2.2 3.3 30.4
CB/CS Isomerization 120 250 21 20.8 208
Storage
1) Crude - _ _ 7,550
2) Motor Gasoline - - - 3,730
3) Light Fuel Oil ... 564
4) Heavy Fuel Oil - Neg
Sludge Incineration 354 597 125 41 510
Miscellaneous Emissions ... 60,900
TOTAL 6,320 16,000 1,200 73,970 11,830
* Mainly Due to the Tail Gas Itself
-------
TABLE 3.1-5
EMISSION FACTORS FOR
Air Pollutant
Particulates
Sulfur Oxides (SOX )
CO
Hydrocarbons
Nitrogen Oxides (NOV )
REFINERY FUEL
Fuel Gas
lb/1000 SCF
0,02
2 x SG*
0.017
0.029
Oo23
FUEL OIL
USE
Fuel Oil
Ib/barrel
Oo97
6.72 x SQ**
0.168
0.168
1,68
Coke Gas
lb/1000 SCF
0.02
O V C *^
0.0013
0.0022
0,018
Sn is Equal to the Sulfur Concentration of the Gas.
Lr
** S0 is Equal to the Weight Percent Sulfur in the Fuel Oil.
^ Calculated Using EPA Emission Factors and the Ratio of Coke
Gas to Fuel Gas Heating Values.
Source: (EN-071)
-128-
-------
the fuel oil are calculated using a sulfur content of 0.3 wt
percent. The SOX emissions from combustion of fuel gas are
calculated assuming the H2S concentration in the fuel gas is
in compliance with the Federal regulation of 0.10 grain per dscf
(ST-124).
Due to the different composition of the coke gas,
estimated values for the coke gas emission factors had to be
determined. The particulates emission factor is assumed to be
the same as the fuel gas. The S0x emission factor is also as-
sumed the same and based on a hydrogen sulfide concentration in
compliance with the Federal standard of 0.10 grain/dscf (ST-124),
The emission factors for CO, hydrocarbons, and nitrogen oxides
are determined by multiplying the EPA fuel gas emission factors
for each of these constituents by the ratio of the heating value
of the coke gas to the heating value of the fuel gas.
Sulfur Recovery Emissions
The efficiency of sulfur removal by the sulfur re-
covery plant and the tail gas treating plant is approximately
99.8 percent (HY-014). The 0.2 percent not recovered is ex-
hausted as S02 at a rate of 3,410 Ib/day. The tail gas is
routed to a stack within the refinery.
Sludge Incineration Emissions
The oily sludge from the API separator and the bio-
logical sludge from the waste treatment facilities are both in-
cinerated. The quantity of oil incinerated in the oily sludge
is based on the following (MA-226).
-129-
-------
(1) 0.0015 bbl of oily sludge/bbl crude
throughput is produced,
(2) Oily sludge is 36.6 wt percent oil, and
(3) Weight of the sludge is 340 Ib/bbl.
The emissions from burning the oily sludge are based on the as-
sumption that the oil in the sludge has the same characterisitcs
of fuel oil and thus the same emissions factors (EN-071).
The biological sludge produced in the refinery is
calculated to be 4280 Ib/day. This value is based on the
following:
(1) 9,000 Ib BOD removed/day (US-056),
(2) 0.5 Ib volatile solids formed/lb of
BOD removed (BE-047), and
(3) The BOD removal efficiency is 95
percent (BE-047).
The emission factors used for biological sludge are the EPA
emission factors for municipal wastes (EN-071). The emission
factors are given in Table 3.1-6.
Petroleum Storage Emissions
In order to calculate the hydrocarbon emissions from
petroleum storage, the following assumptions are used:
(1) Storage capacity is one month for both
feed and products.
-130-
-------
TABLE 3.1-6
SLUDGE INCINERATION EMISSION FACTORS
Oily Sludge Biological Sludge
Pollutant Ib Emission/1,000 gal Sludge Ib Emission/Ton
Particulates 23 30
S03 47 2,5
CO 4 35
Hydrocarbons 3 1.5
NO, 40 3
-131-
-------
(2) Only crude, gasoline, and light fuel
oil storage will result in hydrocarbon
emissions.
(3) Heavy fuel oil storage and pressurized
storage of high volatility products
will result in negligible hydrocarbon
emissions.
(4) Crude, gasoline, and fuel oils will be
stored in floating roof tanks.
EPA emission factors for storage in floating roof tanks are used
to calculate petroleum storage emissions. These factors are as
follows (EN-071):
crude - 0.029 lb/day-103 gal
gasoline - 0.033 lb/day-103 gal
light fuel oil - 0.0052 lb/day-103 gal
Miscellaneous Hydrocarbon Emissions
There are numerous miscellaneous hydrocarbon emissions
in petroleum refineries which escape from sources such as valve
stems, flanges, loading racks, equipment leaks, pump seals, sumps,
drains, sewers, rupture discs, and API separators. Based on
literature data, these miscellaneous hydrocarbon emissions amount
to about 0.1 percent of the refinery capacity for a new, well-
designed, well-maintained refinery (RA-119, DA-069, MS-001,
AM-055). The composition of these hydrocarbons can be expected
to be a composite of all volatile intermediate and refined pro-
ducts .
-132-
-------
3.1.3.2 Water Effluents
Module water effluents have been estimated from published
information (RA-119). The wastewater generation rate is taken
as 15 gallons per barrel of crude feed. This value is believed
to be reasonable considering modern water conservation techni-
ques, segregation of wastewater streams, air cooling, and
recycle. The concentrations of the pollutants are based on
the efficiencies of primary and secondary wastewater treatment
facilities. These concentrations are given in Table 3.1-7.
3.1.3.3 Solid Wastes
The solid wastes from a refinery are highly variable.
Possible sources of solid waste in a refinery are the following:
(1) entrained solids in the crude,
(2) silt from surface drainage,
(3) silt from water supply,
(4) corrosion products from process units
and sewer systems,
(5) solids from maintenance and cleaning
operations,
(6) water treatment facilities, including
ash from the -sludge incinerator, and
(7) spent catalyst.
With the exception of spent catalyst, all the solids collect in
the API separator and the waste water treating facilities. The
-133-
-------
TABLE 3.1-7
WASTEWATER EFFLUENT QUALITY
Fuel Oil Refinery
Basis: 200,000 bbl/day Crude Feed
Flow Rate - 3.0 x 106 gal/day
Concentration
BOD 15 ppm
COD 80 ppm
Ammonia 2 ppm
Hydrogen Sulfide 0.1 ppm
Total Phosphorous 2 ppm
Phenol 0.1 ppm
Oil 2 ppm
Suspended Solids 10 ppm
Dissolved Solids 370 ppm
Source: (RA-119)
-134-
-------
solid waste is estimated at four tons per day. Three of the
four tons per day are from the solid wastes from the API separa-
tor and the wastewater treatment facility. The other ton is
from spent catalyst and is only an average of the intermittent
catalyst regenerations. The solid wastes are suitable for land-
fill.
-135-
-------
3.2 Gasoline Refinery Module
3.2.1 Module Basis
The gasoline refinery basis is the same as the basis
used for the fuel oil refinery, a crude feed rate of 200,000
Vc
bbl/day. The emissions given in this section for a gasoline
refinery are presented on this 200,000 barrel feed basis. A
summary of total emissions from the gasoline refinery module is
shown in Table 3.2-1. :
3.2.2 Module Description
The crude feedstock for this module is assumed to have
a 31° API gravity and a sulfur content of 1.5 wt percent. The
heating value of the crude is assumed to be 5.8 x 106 Btu/bbl
(BA-230). Characteristics of the crude charge are summarized
in Table 3.2-2.
The processing sequences utilized in the gasoline re-
finery module are given in Figure 3.2-1 along with the major
process flow rates. The gasoline refinery, has the same basic
processing steps as the fuel oil refinery with the addition of
the fluidized cat cracking unit and the hydrocracker. These units
are employed to achieve the additional cracking capacity neces-
sary for increasing the gasoline product yield. The product
yield based upon one barrel of feed is as follows:
light ends (C2-C3), 9.02 Ib
gasoline, 0.625 bbl
middle distillates, 0.207 bbl
fuel oil, 0.105 bbl
coke, 0.875 Ib
*A11 flow rates for this module are based on calendar days
-136-
-------
TABLE 3.2-1
SUMMARY OF ENVIRONMENTAL IMPACT
Gasoline Refinery Module
Basis: 200,000 bbl/day Crude Feed
Air (Ib/day)
Particulates 10,380
SOV 22,750
X
NOX 29,990
CO 2,270
Hydrocarbons 76,800
Water (Ib/day)
Suspended Solids 250
Dissolved Solids 9,260
Organic Material 52.5
Solid Wastes (tons/day) 7.0
-137-
-------
TABLE 3.2-2
CRUDE FEEDSTOCK CHARACTERISTICS
Gasoline Oil Refinery Module
Crude Characteristics
API0 Gravity 31
Sulfur Content 1.5 wt %
Heating Value 5.8 x 106 Btu/bbl
True Boiling Point Range wt % of Crude
68/375°F 20.4
375/600°F 19.1
600/1050°F 36.0
1050+°F 24.5
-138-
-------
FIGURE 3.2-1 GASOLINE REFINERY MODULE
LIGHT ENDS
» TAIL GAS
'38.600 Ib/day
218.000 Ib/day
DE-ASPHALTED OIL
HYDROTREATER
ASPHALT
NAPHTHA ~
HYDROTREATED
5.11x10° Ib/day
0
FLEXICOKER
_J t 1 L.
993x10* lb/d«y
LOW SULFUR FUEL W
2.155xlO£ Ib/day *
NAPHTHA
793.000 Ib/day
DISTILLATE .
1.74X106 Ib/day
GAS OIL
S
HYDROGEN
PLANT
7
2
0
t
319.00
8.87*10° lb/d«y
COKE STEAK 3.01x10' lb/d.y
175.000 Ib/day
HVY.
FUEL OIL
7.19X102 lb/d«y
21.060 bbl/diy
0.3 HEX S
Co
VO
I
-------
In this module the crude is desalted and introduced to
a crude distillation unit which consists of both an atmospheric
distillation column and a vacuum distillation column along with
various other flash tanks which are needed to make the distillate
cuts. Five cuts are taken from the crudes: (1) sour light ends
(Ci-C.* hydrocarbons), (2) straight run naphtha, (3) middle distil-
lates, (4) gas oil, and (5) residual product.
Sour light ends from the crude distillation unit are
combined with other sour light ends from various refinery
processes and routed to a gas treating plant. The sour light
ends are contacted with an amine solution at the gas treating
plant for removal of H2S. The H2S is subsequently stripped from
the amine and routed to a sulfur recovery unit. The sweetened
light ends go to a light ends recovery unit for separation into
specific product streams.
The straight run naphtha from the crude distillation
unit is hydrotreated for sulfur removal. The naphtha stream is
then split into a light naphtha stream (TBP 68-180°F) and a
heavy naphtha (TBP 180-375°F). The split is assumed to be 20
wt percent light naphtha and 80 wt percent heavy naphtha. The
light naphtha is run through a C5/C6 isomerization unit. The
isomerization unit is used to increase the octane rating of
pentane and hexane fractions by catalytically rearranging the
normal paraffins into isoparaffins. The heavy naphtha from the
hydrotreater is fed into a catalytic reformer. The catalytic
reforming process converts low octane naphtha into high octane
naphtha by catalytically rearranging and dehydrogenating
naphthenes and paraffins to form benezene, toluene, and xylene.
-140-
-------
The products of these two naphtha streams are blended for motor
gasoline. The light ends from the catalytic reformer are routed
to the gas treating plant, while the hydrogen from the reformer
is separated and used in other refinery processes.
The middle distillate cut from the crude distillation
unit is also hydrotreated for sulfur removal. The light ends
which are produced are sent to the gas treating plant and the
product is sent to storage as a hydrotreated middle distillate.
The straight run gas oil cut from the crude distilla-
tion unit and the gas oil produced in the flexicoker (discussed
later) are combined and then split equally between a hydrocracker
and a fluid catalytic cracking unit (FCCU). The gas oil to the
FCCU is first hydrotreated to protect the catalyst from
poisoning and to reduce S02 emissions during catalyst regenera-
tion. The main difference between the products from the two
cracking processes is the fact that the hydrocracker products
are much more saturated than the products from the FCCU, due to
the large amounts of hydrogen utilized in the hydrocracking
process.
The FCCU will produce four different product streams.
A light ends stream is produced and routed to the gas treating
plant for H2S removal. A C3/Ci» olefinic cut which is produced
is sent to a Merox treating unit for additional sulfur removal
by caustic scrubbing and then routed to an alkylation unit. An
FCCU gasoline product stream is also treated in a Merox system
prior to being routed to the gasoline blending facilities. The
combination of heavy and light cycle oil which is produced
is routed to storage for a heavy fuel oil product. During the
regeneration of the catalyst in the FCCU, a large quantity of
off-gas is produced. This off-gas is a major source of air
emissions and must be carefully controlled.
-141-
-------
The gas oil and hydrogen going into the hydrocracker
are converted into essentially two product streams; light ends
and a reformer feedstock. The light ends are fed to the gas
treating plant for H2S removal. The hydrocrackate (naphtha) is
fed to a catalytic reformer where hydrogen, light ends, and re-
formate are produced. The hydrogen from the reformer is separated
and recycled to the hydrogen consuming refinery processes. The
reformer light ends are sent to the gas treating plant, while the
reformate is routed to gasoline blending.
The final crude distillation unit product is the vacuum
resid. The resid is fed to a propane deasphalting unit where a
gas oil is produced by extraction. The deasphalted oil is routed
to a hydrotreater for removal of sulfur. The light ends produced
in the hydrotreater are routed to the gas treating plant. The
remaining product is split into a naphtha stream which goes to the
gasoline blending facilities and a hydrotreated, deasphalted oil
which is routed to storage as a heavy fuel oil product.
The asphalt from the deasphalting unit is fed to a
flexicoker. The flexicoker is a fluid coking process in which
the coke product is gasified to produce a usable low Btu fuel
gas. The fuel gas is hydrotreated and used as fuel in the pro-
pane deasphalting unit. The gas oil produced in the flexicoker
is combined with the straight run gas oil from the crude distil-
lation unit and fed to the FCCU and the hydrocracker. The light
ends produced are sent to the gas treating plant. The naphtha
produced is combined with a portion of the reformate from the
straight run naphtha reformer and routed to a hydrogen plant
where hydrogen is produced to balance the hydrogen demand within
the refinery. The hydrogen plant is a steam-naphtha reforming
process. The hydrogen is formed by a multiple-step shift con-
version of naphtha and steam into carbon dioxide and hydrogen.
-142-
-------
Of the naphtha fed to the hydrogen plant, 37 percent is used for
process heater fuel (VO-025).
All the light ends produced by the refinery processes
are treated in the gas treating plant and then fed to a light
ends recovery unit which splits the stream into: (1) fuel gas,
(2) ethane and ethylene, (3) propane and propylene, (4) isobutane,
and (5) mixed, butanes. The isobutane from the light ends is
fed along with the C3/Cn olefins from the FCCU to a HF alkylation
unit. The olefins and isobutane are catalytically reacted to
produce a high octane component for gasoline blending. The
mixed butanes from the light ends recovery unit are also blended
with the gasoline.
The H2S from the amine gas treating plant is sent to
a sulfur recovery plant for conversion to recoverable sulfur. A
Glaus plant in conjunction with a tail gas treating unit is used
for sulfur recovery. The Glaus plant catalytically reacts stoi-
chiometric amounts of H2S and S02 to form sulfur and water. The
overall conversion is in the range of 95 percent to 98 percent
(HY-014). The tail gas from the Glaus plant is treated for
additional sulfur removal by washing it with an alkanolamine
solution in an absorption column. The final result is greater
than 99.8 percent removal of the equivalent sulfur in the origi-
nal sour acid gas (HY-014).
The liquid wastes resulting from this module are
handled in both primary and secondary treatment facilities.
The sludge from the water treatment facilities is incinerated.
Auxiliary units such as waste water treating facilities and the
incinerator are not shown in Figure 3.2-1.
-143-
-------
The heat requirements for each unit are calculated
from the specific module demands and the calculated flows through
each unit. The specific heat requirement for each unit is given
on Table 3.2-3. The total module heat requirement is 7.02 x 1010
Btu/day.
After the total module heat requirements are estab-
lished, allocations of the refinery fuels are made. All the
fuel gas produced within the refinery is consumed within the
refinery. The fuel gas is preferentially used in smaller heaters.
However, after the fuel gas has been allocated, 0.3 wt percent
fuel oil must be used in the remaining process heaters. Coke
gas from the flexicoker is also used as a low Btu fuel gas and
fired along with fuel gas in the deasphalting unit process
heater. The hydrogen plant heater is fired with naphtha. The
heating valves of the various fuels are as follows:
Fuel gas 969 Btu/scf (calculated)
Fuel oil 6.3 x 106 Btu/bbl (EN-071)
Coke gas 1,598 Btu/lb (FL-047)
Naphtha 5.248 x 106 Btu/bbl (EN-071)
The fuels used in the specific unit process heaters are given
in Table 3.2-3.
-144-
-------
TABLE 3.2-3
GASOLINE REFINERY
MODULE HEAT REQUIREMENTS
Unit
Crude Unit
Mid-Distillate
Hydrotreater
St. Run Naphtha
Hydrotreater
Heavy Naphtha Reformer
Gas Oil Hydrotreater
FCCU
Hydrocracker
Hvy. Hydrocrackate
Reformer
Propane Deasphalting
Deasphalted Oil
Hydrotreater
HF Alkylation
CS/C8 Isomerization
Light Ends Recovery
Tail Gas Treating
Hydrogen Plant
Unit Heat
Requirement (Btu/day)
2.0 x 1010
2.14 x 109
8.53 x 10"
1.078 x 1010
2.13 x 109
5.89 x 109
6.20 x 109
1.07 x 1010
3.68 x 109
1.39 x 109
2.1 x 107
6.81 x 108
2.01 x 107
1.50 x 108
5.72 x 10*
Fuel Used
Fuel Oil
Fuel Gas
Fuel Gas
Fuel Oil
Fuel Oil
Fuel Gas
Fuel Gas
Fuel Oil
Coke Gas & Fuel Gas
Fuel Gas
Fuel Gas
Fuel Oil
Fuel Gas
Fuel Gas
Naphtha
-145-
-------
3.2.3 Module Emissions
3.2.3.1 Air Emissions
Air emissions from the gasoline refinery module result
from fuel combustion, CO boiler, sulfur recovery, steam-hydro-
carbon reforming (hydrogen plant), sludge incineration, petro-
leum storage, and miscellaneous hydrocarbon emissions. Module
air emissions from the specific sources are given in Table 3.2-4,
Fuel Combustion Emissions
Utilizing fuel demand data for the various processes,
fuel combustion emission sources are determined to be the
following (HY-013, HY-014, VO-025):
crude distillation
middle distillate hydrotreater
straight run naphtha hydrotreater
heavy naphtha reformer
gas oil hydrotreater
fluidized catalytic cracking unit
CO boiler
hydrocracker
heavy hydrocrackate reformer
propane deasphalting unit
deasphalted oil hydrotreater
HF alkylation
Cs/Ce isomerization
light ends recovery
tail gas treating plant
hydrogen plant
-146-
-------
TABLE 3.2-4
Crude Distillation
Mid-Distillate 11. T.
St. Run Naphtha H.T.
Hvy. Naphtha Reformer
Gas Oil Hydrotreater
FCCU
CO Boiler
Hydrocracker
Hvy. Hydrocrackate Reformer
Propane Deasphalting
Deasphalted Oil H.T.
HF Alkylation
Light Ends Recovery
C8/Ce Isomerization
Hydrogen Plant
Storage
1) Crude
2) Motor Gasoline
3) Mid-Distillates
4) Hvy. Fuel Oil
Sludge Incineration
Tail Gas Treating
Miscellaneous
TOTAL
MODULE ATMOSPHERIC EMISSIONS
GASOLINE REFINERY MODULE
(Ib/day)
Basis: 200,000 bbl/day Crude Feed
Particulates
3,080
45
18
1,660
326
124
.491
131
irmer 1,650
622
29
0,4
11
105
1,720
.
-
-
-
361
3.1
-
10,380
SO,
6,400
64
26
3,560
703
176
3,020
185
3,540
884
42
0.6
15
225
345
-
-
-
-
600
Tail Gas
3,560
-
22,750
CO
533
38
15
287
57
106
256
110
285
48
25
0.4
9.2
18.5
344
-
-
-
-
133
2.6
-
2,270
Hydrocarbons
533
65.4
25.7
288
56
181
160
1,077
286
81
42
0.7
15.7
18.2
338
7,550
5.050
97
Neg.
42
4.6
60,900
76,800
NO,
5,330
518
207
2,870
567
1,425
6,370
1,110
2,854
656
336
5
125
181
6,890
-
-
-
-
510
36
-
29,990
-147-
-------
Although each unit may contain several fuel combustion emission
sources, all flue gas streams within one unit are assumed to be
combined and routed to one stack. Therefore, each unit requir-
ing fuel combustion represents one emission source. The emis-
sions from each unit are based on the type of fuel used, such as
fuel gas, fuel oil, or coke gas, and the EPA emission factors.
All of the fuel combustion emission factors are shown in Table
3.2-5. The SOX emissions from the fuel oil are calculated using
a sulfur content of 0.3 wt percent. The SOX emissions from com-
bustion of fuel gas are calculated assuming the H2S concentra-
tion in the fuel gas is in compliance with the Federal regula-
tion of 0.10 grains per dscf (ST-124).
The steam-hydrocarbon reforming has special emission
factors due to the high operating temperature (1700°F) of the
process heater which tents to enhance the formation of NOX.
Special emission factors must also be used because naphtha in-
stead of fuel oil or fuel gas is used in the process heaters
(AT-040).
Due to the different composition of the coke gas,
estimated values for the coke gas emission factors had to be
determined. The particulates emission factor is assumed to be
the same as for fuel gas. The S0x emission factor is also as-
sumed the same and based on a hydrogen sulfide concentration in
compliance with the Federal standard of 0.10 grain HaS per dscf
(ST-124). The emission factor for CO, hydrocarbons, and nitro-
gen oxides are determined by multiplying the EPA emission factors
for each of these constituents for fuel gas by the ratio of the
heating value of the coke gas to the heating value of the fuel
gas (EN-071).
-148-
-------
TABLE 3.2-5
Air Pollutant
Particulates
Sulfur Oxides (SCf)
CO
Hydrocarbons
)N FACTORS FOR GASOLINE REFINERY FUEL USE
Fuel Gas
lb/1000 scf
0.02
2 x g 1
0.017
0.029 .
Fuel Oil
Ib/barrel
0.97
6.72 x SQ2
0.168
0,168
4
Coke Gas
lb/1000 scf
0.02
2 x S
0.0013
0.0022
Steam-Hydrocarbon
Reforming
(lb/1,000 scf)
0.845
6.72 x S/
0.168
0.1655
Nitrogen Oxides (N0x)
0.23
1.68
0.018
3.36
SG is Equal to the Sulfur Concentration of the Gas.
o
SQ is Equal to the Weight Percent Sulfur in the Fuel Oil.
3 SN is Equal to the Weight Percent Sulfur in the Naphtha Fuel.
^ Calculated using EPA Emission Factors and the Ratio of Coke Gas to Fuel Gas Heating
Values.
5 Reference (AT-040)
Source: (EN-071)
-------
CO Boiler
The CO boiler flue gas rate is estimated at 64,000 scfm
(CU-016) . Emissions from the CO boiler are calculated as follows:
(1) Particulates are calculated to be the maximum
allowed by Federal emission laws, 0.027 gr/dscf
(EN-196) .
(2) SOz emission is calculated assuming the sulfur
in the coke (on the FCCU catalyst) is 0.21 wt
percent of the coke, and all the sulfur in the
coke is converted to SOz •
(3) The hydrocarbon emission factor used was based
on the assumption that the concentration of
the hydrocarbons in the flue gas is equal to
the hydrocarbon concentration in the flue gas
from the combustion of residual oil. This con-
centration is 1.65 x 10" 6 Ib/scf of flue gas,
and includes aldehyde emissions.
(4) The regenerator flue gas entering the CO boiler
contains 71 Ib NO /I, 000 bbl of cat cracker feed
and 54 Ib NH3/ 1,000 bbl of cat cracker feed
(EN-071) . In a CO boiler, it is assumed that the
only NO formed in the CO boiler is from the
X
combustion of NH3 to NO . With these premises,
X
a NO emission factor for CO boilers of 166
X
Ib NO 71,000 bbl cat cracker capacity, based
X
upon total combustion of NH3 to NO, was used.
(5) The emission factor used for calculating
the CO emission from the module CO boiler is
-150-
-------
20 ppm of the boiler flue gas. This factor is
based on a survey for EPA which reported 20 ppm
to be the average CO concentration in the CO
boiler flue gas (EN-072).
Sulfur Recovery
The efficiency of sulfur removal by the sulfur re-
covery plant and the tail gas treating plant is approximately
99.8 percent (HY-014). The 0.2 percent not recovered is ex-
hausted as S02 at a rate of 3,390 Ib/day. The tail gas is
routed to a stack within the refinery.
Steam-Hydrocarbon Reforming (Hydrogen Plant)
Due to the high1 operating temperature (approximately
1700°F) of the steam-hydrocarbon reforming plant, special emis-
sion factors must be used. The high temperatures that must be
achieved within the heater place a limit on the degree of NOV
X
emission control that can be practiced through modification of
combustion techniques. The emission factors for steam-hydro-
carbon reforming are listed in Table 3.2-5.
Sludge Incineration
The oily sludge from the API separator and the bio-
logical sludge from the waste treatment facilities are both in-
cinerated. The quantity of oil incinerated in the oily sludge
is based on the following (MA-226).
(1) 0.0015 bbl of oily sludge/bbl crude throughput
is produced,
(2) the oily sludge is 36.6 wt percent oil, and
(3) the weight of the sludge is 340 Ib/bbl.
-151-
-------
The emissions from burning the oily sludge are based on the as-
sumption that the oil in the sludge has the same characteristics
of fuel oil and thus the same emission factors. These emission
factors are shown in Table 3.2-6.
The biological sludge produced within the refinery is
calculated to be 4,750 Ib/day. This value is based on the
following:
(1) 10,000 Ib BOD removed/day (US-056),
(2) 0.5 Ib volatile solids formed/Ib of BOD
removed (BE-047), and
(3) The BOD removal efficiency is 95 percent
(BE-047).
The emission factors used for biological sludge are the EPA
emission factors for municipal wastes incineration (EN-071).
These factors are also given in Table 3.2-6.
Petroleum Storage
In order to calculate the hydrocarbon emissions from
petroleum storage, the following assumptions are used:
(1) Storage capacity is one month for feed and
products,
(2) Only crude and gasoline storage will result
in significant hydrocarbon emissions. Light
fuel oil storage will result in a small hydro-
carbon emission,
-152-
-------
TABLE 3.2-6
SLUDGE INCINERATION EMISSION FACTORS
Oily Sludge Biological Sludge
Pollutant Ib Emission/1.000 gal Sludge Ib Emission/Ton
Particulates 23 30
S0a 47 2.5
CO 4 35
Hydrocarbons 3 1.5
NOX 40 3
Source: (EN-071)
-153-
-------
(3) Heavy fuel oil storage and pressurized storage
of high volatility products will result in
negligible emission, and
(4) Crude, gasoline, and fuel oils will be stored
in floating-roof tanks.
Hydrocarbon emission factors for floating roof tanks are the
following (EN-071):
crude oil 0.029 lb/day-103 gal
gasoline 0.033 lb/day-103 gal
light fuel oil 0.0052 lb/day-103 gal
Miscellaneous Hydrocarbon Emissions
There are numerous miscellaneous hydrocarbon emissions
in petroleum refineries which escape from sources such as valve
stems, flanges, loading racks, equipment leaks, pump seals,
sumps, drains, sewers, ruptured discs, and API separators. Based
on literature data, these mescellaneous hydrocarbon emissions
amount to about 0.1 percent of the refinery capacity for a new,
well-designed, well-maintained refinery (RA.-119) . The composition
of these hydrocarbons can be expected to be a composite of all
volatile intermediate and refined products.
3.2.3.2 Water Effluents
Module water effluents have been estimated from published
information (RA-119). The wastewater generation rate is taken
as 15 gallons per barrel of crude feed. This value is believed
-154-
-------
to be reasonable considering modern water conservation techniques,
segregation of wastewater streams, air cooling, and recycle. The
concentrations of the pollutants are based on the efficiencies
of primary and secondary wastewater treatment facilities. The
concentrations of pollutants in the effluent are given in
Table 3.2-7.
3.2.3.3 Solid Wastes
The solid wastes from a refinery are highly variable.
Possible sources of solid wastes in a refinery are the following:
(1) entrained solids in the crude,
(2) silt from surface drainage,
(3) silt from water supply,
(4) corrosion products from process units and
sewer systems,
(5) solids from maintenance and cleaning operations,
(6) water treatment facilities, including ash from
the sludge incinerator, and
(7) spent catalyst.
With the exception of spent catalyst, the solids collect in the
API separator and the waste water treating facilities. The
solid wastes are estimated at seven tons per day. Three of the
-155-
-------
TABLE 3.2-7
WASTEWATER EFFLUENT QUALITY
Gasoline Refinery Module
Basis: 200,000 bbl/day Crude Feed
Flow Rate - 3.0 x 106 gal/day
Concentration
BOD 15 ppm
COD 80 ppm
Ammonia 2 ppm
Hydrogen Sulfide 0.1 ppm
Total Phosphorous 2 ppm
Phenol 0.1 ppm
Oil 2 ppm
Suspended Solids 10 ppm
Dissolved Solids 370 ppm
Source: (RA-119)
-156-
-------
seven tons are from the solid wastes from the API separator and
the waste water treatment facility. THe other four tons are
from spent catalyst from the cat cracker and the hydrodesulfuri-
zation units and are only averages of the intermittent catalyst
regenerations.
-157-
-------
3.3 LNG Module
LNG facilities in service or under construction
throughout the world are designed for either peak-shaving or
base load applications. Most plants in the U.S., though,
are designed for peak-shaving operations . These peak-shaving
plants were developed to satisfy an area's peak gas demand at
the times when the natural gas supply would be insufficient.
At such times, liquefied natural gas from a peak-shaving plant
could be withdrawn from storage, regasified, and fed into the
distribution lines. During periods when the potential supply
exceeds the demand, surplus natural gas may be liquefied and
stored. With this type of approach, a significant natural
gas storage capability can be provided. Currently, there are
fifty-five such peak-shaving plants in operation in the U.S.
They range in size from 5 x 105scfd to 25.0 x 106scfd of
liquefaction capacity.
3.3.1 Peak-Shaving Module
The basis for the LNG module is a typical size peak-
shaving plant operating in the U.S. The liquefaction capacity
of the module is 10 x 106 scfd* of pipeline natural gas. All
emissions determined for this module are based on this lique-
faction capacity. A summary of emissions from the plant is
presented in Table 3.3-1.
3.3.1.1 Peak-Shaving Module Description
In this section, the processing steps utilized in the
LNG module are briefly discussed. These processes include acid
* All flow rates in this module are based on calendar days.
-158-
-------
TABLE 3.3-1
SUMMARY OF ENVIRONMENTAL EMISSIONS
LNG MODULE
Basis: 10 x 106 scfd of Natural Gas Liquefied
Air (Ib/day)
Particulates 59
S02 7.9
N0x 758
HC 470
CO 56
Water (Ib/day)
Suspended Solids 0
Dissolved Solids 0
Total 0
Solids (tons/day) 0
-159-
-------
gas removal, dehydration, liquefaction, storage, and regasifi-
cation. A flow diagram is shown in Figure 3.3-1 illustrating
the processing sequence.
Processing Steps
To insure that no solids form in the cold box during
methane liquefaction, the incoming stream to the plant must
first be treated for the removal of C02 and H20. In industry
today, molecular sieves are by far the most popular unit for
handling this clean-up job. They easily purify the natural gas
feedstock to less than 1 ppm H20 and less than 50 ppm C02- At
the same time, they also remove most of the sulfur compounds
present in the feed stream (H2S, COS, CS2, and mercaptans).
A two-bed molecular sieve unit is used in this module
for acid gas and H20 removal. One bed is on-line absorbing
while the other is being regenerated. A gas-fired heater is
required to heat the gas for molecular sieve regeneration,
This heater consumes 5,000 Btu per hour per million scfd of gas
treated (IN-029). The quantity of gas needed for regeneration
is typically around 2% of the daily throughput of the sieves.
The gas resulting from molecular sieve regeneration contains
higher concentrations of sulfur, C02, and H20. This gas is
combined with fuel to the boilers in order to supply module
heat requirements. The molecular sieve operates at 1 atm pressure
and at the temperature of the incoming gas, usually ambient (HA-274)
After passing through the molecular sieve bed, the
natural gas composition is:
Methane - 98%
Ethane - 1%
Propane - 0.5%
Nitrogen - 0.5%
-160-
-------
Regeneration
Gas to Fuel
Line
.204 MM scfd
^-
Pipeline
Gas 10.204
MM scfd
Acid Gas Re-
moval & Dehy-
dration
^
10.0
MM sc
Liquefaction
fd
fc—
10.0
MM scfd
Storage
Regasif i-
cation
^—
Variable
Variable
FIGURE 3.3-1 LNG PEAK-SHAVING PLANT - 10 MM SCFD CAPACITY
-------
Liquefaction of the methane stream is the next step
in the LNG plant. Liquefaction is accomplished with a single
mixed refrigerant liquefaction cycle. The mixed refrigerant,
composed of nitrogen and light hydrocarbons from methane through
pentane, is circulated in a closed refrigeration loop. This
loop contains a compressor, a partial condenser, a refrigerant
heat exchanger, and a Joule-Thompson expansion valve. The
power to drive the compressor requires approximately 13% of
the gas charged to the liquefaction unit (IN-029).
LNG storage is accommodated in an above-ground double
walled metallic tank with a storage capacity of 2.0 x 109 scf.
Normal boiloff due to heat leaks is approximately 0.040% of
the tank capacity per day when full. The boiloff gas is com-
pressed and routed to the distribution system. The storage tank
is operated at atmospheric pressure.
The regasification system is a submerged combustion
type with a maximum sendout capacity of 200 MM scfd. Approxi-
mately 2% of the gas vaporized must be combusted to supply the
necessary heat (IN-029). For this module it is assumed that
pipeline natural gas will be used to supply this heat requirement
Module Flow Rates
For this module, natural gas is fed to the plant at
a rate of 13.502 MM scfd. Before this gas enters the molecular
sieves for cleanup, a stream is taken off at the rate of 3.298
MM scfd to supply fuel for the plant.
This leaves a stream of 10.204 MM scfd entering the
molecular sieve unit. A small stream amounting to 2% of this
unit's throughput, 204,000 scfd, is diverted from the downstream
side of the molecular sieve beds and heated for use as a re-
generation gas. This gas is then fed into the fuel line for
use as plant fuel.
-162-
-------
After cleanup, the gas stream passes into the lique-
faction unit at the rate of 10 MM scfd. After liquefaction,
the natural gas is pumped to a 2 billion scf capacity storage
tank.
The regasifier is designed for a maximum sendout
capacity of 200 MM scfd. Combining with this amount the daily
boiloff rate of 800,000 scfd from the LNG tank, the maximum
output from this module is 200.8 MM scfd. However, this large
discharge rate is used only in times of peak gas demand. For
the purpose of this module, the environmental effect of re-
gasification will be based upon a sendout rate of 100 MM scfd .
Module Heat Requirements
Overall module heat requirements are determined from
process unit utility requirements and flow rates. The heat
requirements for the various process units are presented in
Table 3.3-2. The total module heat requirement is 3.30 x 109
Btu/day. This heat requirement is supplied by pipeline gas
(9470) and molecular sieve regeneration gas (6%) .
3.3.1.2 Peak-Shaving Module Emissions
Air Emissions
Atmospheric emission sources within the natural gas
liquefaction module include the flue gases from: (1) the boil-
er which supplies power to the liquefaction train compressors,
(2) the molecular sieve regeneration gas heater, and (3) the
regasifier, as well as miscellaneous fugitive hydrocarbon
emission sources. A summary of the air emissions is presented
in Table 3.3-3.
-163-
-------
TABLE 3.3-2
MODULE HEAT REQUIREMENT
Basis.- 10 MM scfd of natural gas liquefied.
Heat Requirement per MM scf Unit Heat Requirement
Unit Charge (M Btu/MM scf) Flow Rate (MM scfd) (Btu/day)
M Molecular Sieves 120 10.204 .01 x 108
Liquefaction
Compressor 130,000 10.000 13.00 x 108
Regasifier 20,000 100.000 19.97 x 108
32.98 x 108
-------
TABLE 3.3-3
SUMMARY OF LNG MODULE AIR EMISSIONS
Module Basis: 10 MM scfd Natural Gas Liquefied
POLLUTANTS EMITTED (Ibs/day)
PROCESS PARTICULATES S02 NO EC CO
~ — ' X ' - —
Liquefaction Unit Boiler 23.4 7.8 299.0 3.9 22.1
Regasifier 35.9 0.1 459.3 6.0 34.0
Fugitive Hydrocarbon Losses - - 460
TOTAL 59.3 7.9 758.3 469.9 56.1
-165-
-------
Stack gas emission rates for the liquefaction unit
boiler are calculated using EPA emission factors for natural
gas-fired industrial boilers (EN-071). These factors are used
since the boiler required by the LNG module is equivalent in
size to industrial boilers. The emission factors used are
presented in Table 3.3-4. The S02 factor includes the extra
7.8 Ibs/day of S02 formed by the combustion of the higher sulfur
content molecular sieve regeneration gas in the boiler. The
EPA emission factors for a small industrial boiler are used to
determine emissions from the regasifier. These emission fac-
tors are also presented in Table 3.3-4. The molecular sieve
regeneration gas heater does require natural gas to be burned
as its heat source. However, as can be seen from Table 3.3-2,
the amount of gas needed per day is very small, and so its
emissions are negligible.
Miscellaneous fugitive hydrocarbon emission sources
result from process leaks at pump seals, valve stems, flanges,
etc. The quantity of these emissions is dependent upon the
amount of attention given to plant maintenance. Thus, it is
difficult to estimate these hydrocarbon losses; however, it was
assumed that 0.1 weight percent of the plant throughput is a
reasonable estimate. The daily amount emitted for this module
is shown in Table 3.3-3.
Liquid Effluents
The boiler make-up feed water for the plant is first
passed through an ion exchange resin unit for demineralization.
Regeneration of this unit with acid and caustic wash water streams
results in this module's only significant liquid effluent stream.
This stream is discharged into holding ponds on the plant site
where the water is evaporated. Since no liquid leaves the plant
boundaries, the module wastewater effluent is considered to be
zero.
-166-
-------
TABLE 3.3-4
EMISSION FACTORS FOR LNG MODULE
EMISSION FACTOR
(lb/10b scf)z
PROCESS P ARTICULATES S02 ~TR5 HC CO
™ "" '—" •- — -. - „—.^£
Liquefaction Unit Boiler 18 61 230 . 3 17
Regasifier 18 .06 230 3 17
Includes the sulfur from the molecular sieve regeneration gases
2SOURCE: (EN-071)
-167-
-------
Solid Wastes
An LNG plant has no solid wastes generated by any of
its processes.
3.3.2 Base Load Module
The base load liquefaction scheme was designed as a
means for correcting a supply/demand imbalance through the lique-
faction and transportation of large amounts of natural gas to
centers of consumption. These operations strongly resemble
those of a peak-shaving liquefaction scheme, though usually on
a larger scale.
• The basis for the base load LNG module is a plant
which has a liquefaction capacity of 750 MM scfd* of natural
gas. All emissions calculated for this module are based on
this liquefaction rate. A summary of the module emissions is
shown in Table 3.3-5.
3.3.2.1 Base Load Module Description
In this section, the processing steps utilized in
the base load module are briefly discussed. These processes
include, in order, liquid knock-out, acid gas removal, dehydra-
tion, final purification, heavy hydrocarbon recovery, lique-
faction, storage/transportation, and revaporization. A flow
diagram is presented in Figure 3.3-2 illustrating the process
sequence.
Processing Steps
The natural gas feed to the plant has had little
processing, therefore, extensive treatment of the gas is needed
* All flow rates in this module are based on calendar days
-168-
-------
TABLE 3.3-5
SUMMARY OF ENVIRONMENTAL EMISSIONS
BASE LOAD LNG MODULE
Basis: 750 MM scfd Liquefaction Capacity
Air (Ib/day)
Particulates 1,770
S02 8,460
N0v 62,000
X
HC (including TEG*) 32,560
CO 1,910
Water (Ib/day) 0
Solids (tons/day) 0
* Triethylene Glycol
-169-
-------
Acid Gas
Light
1—>• Plant Fuel
Feed Gas
815 MM SCFD
•-J
O
to CLaus Plant
I
-
Knock-
t
I
Acid Gas Glycol Molecular
Removal Dehydration Sieve Final
Treatment
i
1
Refrigerated
Absorption
..iquefaction
1
i
1
i
i
i
1
i
i
'
T^LK
Hyd
rocarbons
i
•*
k
Boil-Off Gas
Transportation
via Tanker '.
LNG Storage
Boil-Off Gas
SCKD
To Pipeline
LNG Storage Regasification
FIGURE 3.3-2 LNG BASE LOAD PLANT
-------
prior to its liquefaction. The first unit encountered by the
incoming gas is a liquid knock-out to remove any water or heavy
hydrocarbons which may have condensed in the pipeline during
transmission. After removal of liquids, the gas enters a
Girbitol unit where a lean monoethanolamine solution removes
the bulk of the acid gases in a packed tower. The HaS contain-
ing gases removed from the top of amine regeneration are sent
to a Glaus sulfur recovery plant. After this treatment the
natural gas is saturated with water. A glycol unit utilizing
triethylene glycol contacts the wet gas with the hygroscopic
liquid in a bubble tray column. Here the gas gives up the
bulk of its water vapor to the glycol and passes out the top
of the tower to further processing.
To insure that no solids form in the cold box during
liquefaction, the gas needs further conditioning for the removal
of COa, H20, and H2S to acceptable levels. A molecular sieve
unit is used to remove to trace levels the quantities of these
contaminants which were present after amine and glycol treating.
Still present in the clean natural gas, though, are
significant fractions of heavier hydrocarbons which may freeze
out during liquefaction. These are removed in a refrigerated
absorption system using heavier hydrocarbons as the solvent.
The bottoms from the benzene column which still contain
lighter fractions are fed to the refrigerant-makeup units.
These units, each consisting of a demethanizer, deethanizer,
depropanizer, and debutanizer column, yield hydrocarbon fractions
suitable for makeup of refrigerant losses and a light gasoline.
Production in excess of refrigerant losses is injected into
LNG as far as quality permits. The balance of the light hydro-
carbons are used as plant fuel gas. The light gasoline is
transferred to an oil company via pipeline.
-171-
-------
Liquefaction of the methane stream exiting from the
refrigerated absorption unit is the next processing step. The
mixed refrigerant cycle (MKC.) process is used for liquefying
the natural gas. The coolant used consists of a mixture of
hydrocarbons extracted from the natural gas and nitrogen.
Following liquefaction, the LNG is pumped to either
of two, 2 billion scf capacity above-ground metal tanks. These
tanks hold the LNG prior to the arrival of an LNG tanker for
shipment to a designated receiving terminal.
The tankers are used to move large volumes of LNG
from the liquefaction plant to a purchaser. Upon arrival at
the receiving terminal, the LNG is unloaded from the tankers
and stored in either of two, 2 billion scf capacity metal
above-ground tanks. These storage tanks hold the LNG until
it is regasified for distribution to the gas mains.
Module Flow Rates
Natural gas from the field is supplied to the plant
at the rate of 815 MM scfd. Approximately 91 percent of the
feed is methane, with the remainder being heavier hydrocarbons,
carbon dioxide, water vapor, sulfur compounds, and nitrogen.
After processing and conditioning the gas is fed to the lique-
faction units at the rate of 750 MM scfd. Of this gas, roughly
98 percent is methane with the remainder being mostly ethane
and some propane. The LNG is pumped to storage from the lique-
faction unit at the rate of 750 MM scfd. From there it is
loaded onto LNG tankers for transport to the receiving terminal.
At the receiving terminal, the LNG is withdrawn from storage
and regasified at the rate of 750 MM scfd.
-172-
-------
Module Heat Requirement
Based on a liquefaction and regasification rate of
750 MM scfd and assuming that 13% of the gas throughput to the
liquefaction unit is required as fuel to the plant and 2% of
the throughput to the regasifiers is needed as heat input to
vaporize the LNG, the module heat requirement is 1.125 x 1011
Btu/day. The heat requirement at the plant is supplied by boil-
off gas from storage, gas from process units (mainly separated
heavy hydrocarbons), and fresh gas from the incoming feed.
Roughly 40% of this plant fuel is supplied by the unprocessed
fresh gas. Assuming a sulfur content in the field gas of 0.5
volume percent, the boilers will be burning a gas with a 0.20
volume percent sulfur composition.
3.3.2.2 Base Load Module Emissions
Air Emissions
Atmospheric emission sources within the natural gas
base load liquefaction module include the flue gases from:
(1) the boiler plant which supplies steam for the various
clean-up, hydrocarbon recovery, and liquefaction processes,
as well as providing steam for driving turbines of compressors,
electric-power generators and pumps, (2) the regasifier,
(3) the Glaus plant, and (4) the glycol system. Also, there
are miscellaneous fugitive hydrocarbon emissions from within
the liquefaction plant. A summary of the air emissions is
presented in Table 3.3-6.
Stack gas emissions rates for the plant boilers are
calculated using EPA emission factors for natural gas-fired
utility size boilers (EN-071). These factors are used since the
boilers used in this module are similar in size to a utility
boiler. These emission factors are presented in Table 3.3-7.
-173-
-------
TABLE 3.3-6
SUMMARY OF LNG BASE LOAD MODULE AIR EMISSIONS
Module Basis: 750 MM scfd of Natural Gas
Liquefied
Pollutants Emitted Clbs/dav')
PROCESS P ARTICULATES S02 NO.. HC CO TEG
Plant Boilers 1,500 8,290 58,500 98 1,658
Glycol Unit 370
Regasifier 270 0 3,450 45 255
Glaus Plant - 172 - -
Fugitive Hydro-
carbon Losses - - - 32,000
TOTAL 1,770 8,462 61,950 32,193 1,913 370
TABLE 3.3-7
PROCESS
Plant Boilers
Regasifier
EMISSION FACTORS
Emission Factor
PARTICULATES SO 2
15 851
18 O2
FOR LNG MODULE
(lb/106 scf)
NO
600
230
HC
1
3
CO
17
17
1 Factor includes sulfur present in feed gas used for fuel
which has not been treated.
2 Sulfur-free fuel used.
-174-
-------
The regasifier emissions were determined using factors for a
large industrial boiler (EN-071). These factors are given in
Table 3.3-7. Since the regasifier is fired from sulfur-free
storage boil-off and vaporized LNG, there are no sulfur dioxide
emissions.
The tail gas from the Glaus plant containing uncon-
verted H2S and other sulfur compounds is treated by the Beavon
tail gas treatment process. It removes the sulfur in the tail
gas to a concentration of 250 ppm (S02) or lower before discharge
to the atmosphere (RA-119). The daily emission rate of S02 from
the Glaus plant is presented in Table 3.3-6.
In the operation of the glycol dehydration unit,
water vapor is continuously vented from the triethylene glycol
(TEG) regenerator column. Through this vent stream it is
reported that 0.05 gallons of TEG per MM scf of gas processed
is lost to the atmosphere (PR-052). The daily discharge rate
of TEG is given in Table 3.3-6.
Miscellaneous fugitive hydrocarbon emissions result
from process leaks at pump seals, valve stems, flanges, etc.
The quantity of these emissions is dependent upon the amount
of attention given to plant maintenance. Thus, it is difficult
to estimate these hydrocarbon losses; however, it was assumed
that 0.2 weight percent of the plant throughtout is a reasonable
estimate. Table 3.3-6 gives the estimated amount lost daily.
Liquid Effluents
The majority of the liquid effluent generated by the
module is from the acid and caustic wash water streams used for
the demineralizer regeneration. It is assumed that these
streams, as well as other liquid waste streams, are processed
-175-
-------
at a central water treatment plant. The liquid from this plant
is then discharged to one of three containment/evaporation
ponds. Since the ponds are located within plant boundaries,
the module wastewater effluent is reported to be zero.
Solid Wastes
There are no solid wastes generated within this
module which will cause a solids handling problem.
-176-
-------
3.4 SNG Module
Increasing demand, combined with declining reserves
and exploration, have resulted in significant natural gas
shortages in the United States. These trends have encouraged
gas producers to seek a reliable and efficient means to produce
substitute natural gas to augment the slowly dwindling gas supplies
Current technology is available to gasify liquid feedstocks
ranging from LPG to crude oil on a commercial scale. However,
government allocation programs and official policy is to limit
feedstocks to naphtha only (FE-085). Currently there are 13
SNG plants built or under construction in the U.S. Plant size
ranges from an SNG output of 20 MM scf per day to 250 MM scf
per day.
3.4.1 Module Basis
The basis for the SNG module is a typical medium-
sized plant operating in the U.S. The gasification potential is
125 MM scf per day* of pipeline quality natural gas. All
emissions for this module are determined based on this gasifica-
tion capacity. Table 3.4-1 summarizes the emissions resulting
from this module.
3.4.2 Module Description
*
In this section, the processing steps utilized in the
SNG module are briefly discussed. Figure 3.4-1 presents a flow
disgram of the processing sequence chosen for this module.
These processes include naphtha hydrodesulfurization, gasifica-
tion in a Catalytic Rich Gas (CRG) reactor, methanation (two
stages), C02 removal, dehydration, and hydrogen production.
*A11 flow rates in this module are based on calendar days.
-177-
-------
TABLE 3.4-1
SUMMARY OF ENVIRONMENTAL IMPACT
SNG PLANT MODULE
Basis: 125 MM scfd of SNG produced
Air (Ib/day)
Particulates 469.4
S02 202.7
NO 5,216.1
A.
HC (including TEG*) 16,241.8
CO 229.1
Water (Ib/day) 0
Solid (Ib/day) negligible
* Triethylene Glycol
-178-
-------
CO2 and
I
120 vent
Hydro-
desulfurization
Two Stage
Methanation
Naphtha
Feed 25,065'
bbl/day A
Hydrogen Plant
SNG
^ Product
125 x 106 scfd
FIGURE 3.4-1 SNG MODULE PROCESSING SEQUENCE
-------
Processing Steps
The first step involved in gasifying naphtha is the
reduction of sulfur in the charge to a 0.2 ppm concentration.
This reduction is accomplished in a naphtha hydrodesulfurization
unit, which hydrogenates the sulfur compounds present in the
vaporized feed to hydrogen sulfide over a nickel molybdenum
catalyst. Following the hydrotreating the vaporized naphtha is
routed through a bed of zinc oxide to remove the H2S. The H2S
reacts with the zinc oxide to form zinc sulfide.
After this purification step, the naphtha is mixed
with steam (2 Ibs of steam per Ib of naphtha) and heated in a
superheater to a temperature of1 850°F prior to entering the
CRG reactor. In this vessel naphtha is converted to a mixture
of methane, carbon dioxide, hydrogen, and a small amount of
carbon monoxide in the presence of a catalyst.
The gases leaving this reactor are at a temperature
of about 900°F. From the reactor they enter the methanation
section, of the plant where the methane content in the gas
is increased. Methanation is accomplished in two stages. After
the first stage, a slipstream is diverted as the feed for the
hydrogen generation unit. The main stream is routed to the second
stage methanation, preheated, and reacted to convert the remaining
hydrogen and carbon monoxide to methane. Gas leaving this final
converter contains for the most part a mixture of methane and
carbon dioxide. Table 3.4-2 shows the composition of the gas
as it leaves the CRG unit and the two methanators.
-180-
-------
TABLE 3.4-2
PLANT GAS COMPOSITIONS
Process Stage:
Gas Composition
mole
percent
CHi,
H2
CO
CO 2
Out
CRG
61.
17.
1.
20.
Out 1st
Methanator
2
0
0
8
74.
4.
0.
20.
8
6
2
4
Out 2nd
Methanator
79.0
1.0
<0.1
19.9
The remaining processes in the module are concerned
with upgrading the SNG by the removal of C02 and water vapor.
The C02 is removed in a Benfield unit which utilizes a hot
potassium carbonate aqueous wash in a packed tower. This
solution is regenerated through the use of steam in a C02 stripper,
The carbon dioxide released in the stripper is vented to the
atmosphere.
Before the gas is delivered to the battery limits the
water content of the gas is reduced to the required value of
about 6 pounds per million scf of gas in a drying unit. The de-
hydration system employed in this module is a glycol unit.
The wet gas is contacted countercurrently with a 99.970
triethylene glycol solution in a packed column. The glycol
absorbs the water present in the SNG providing a dry product
meeting sales specifications.
After passing through the C02 and water removal units,
the following composition for the SNG is obtained:
CH,, - 98.05% (mole)
H2 - 1.45
CO - <0.10
C02 - .50
-181-
-------
Module Flow Rates
Naphtha is fed to this module at the rate of 27,245
bbls per day. This amount satisfies both the feed and fuel
requirements of the plant. Assuming the module is 92% thermally
efficient, 2,180 bbls per day of naphtha is consumed as fuel
in the plant (BA-230).
Thus, 25,065 bbls per day of naphtha enter the hydro-
treater for sulfur removal. From the HDS unit the vaporized
naphtha is gasified and methanated. From the second stage
methanator, 155 MM scfd of gas is produced. Of this gas, 19.9
percent is C02 which must be reduced to a final concentration
of 0.5,percent. Thus, the Benfield C02 removal unit vents
approximately 30 MM scfd of C02. The final product of 125 million
standard cubic feet per day of SNG enters the dehydration unit
for reduction of water vapor to pipeline specifications. The
small amount of water removed from the gas stream is vented to
the atmosphere, while the SNG is transmitted off-site to sales.
Module Heat Requirements
Overall module heat requirements are determined from
the required amount of steam, the utility requirements of the
C02 and H20' removal units, and the fuel demand of the process
heaters. Public Service Electric and Gas Company (LO-095)
reports that exothermic reactions in the gasifier and methanators
produce enough heat to supply 60% of the plant's process steam
requirements. The remaining steam requirement is supplied by
combusting naphtha in steam boilers. The fuel burned in the
two preheaters and one superheater is also naphtha. Assuming
that the fuel requirements of the module are supplied entirely
by naphtha and that 40% of the steam requirement for the process
is generated by naphtha combustion in the steam boilers, the
-182-
-------
SNG module fuel consumption is 2,180 bpd. The heat requirements
for the process units are presented in Table 3.4-3.
3.4.3 Module Emissions
Air emission sources within the synthetic natural gas
module include (1) the flue gases from the naphtha fuel combustion
in the two preheaters, the superheater, the steam boiler, the
Benfield C02 removal system, and the glycol dehydration unit,
(2) the methane released with the C02 that is vented from the
Benfield system, (3) hydrocarbon emissions from the naphtha
storage tanks, and (4) miscellaneous fugitive hydrocarbon
emission sources. A summary of module air emissions is pre-
sented in Table 3.4-4.
The emissions from the fuel combustion sources are
calculated using factors derived from the 1973 edition of the
EPA emission factor book (EN-071). Since no factors are
presented for naphtha combustion, the average between the natural
gas emission factors and the distillate fuel oil emission factors
(compared on a Btu basis) are used. These factors are presented
in Table 3.4-5.
Besides fuel combustion emissions from the glycol de-
hydration unit, some glycol (considered triethylene glycol) is
emitted at the water vapor vent on the glycol regenerator.
These losses are estimated to be 0.1 gallons of glycol per
MM scf of gas dehydrated (PR-052)f of which half is vented to
the atmosphere. For this module, about 30 Ibs per day of tri-
ethylene glycol is vented to the atmosphere. This emission rate
is reported as a hydrocarbon loss from the module.
-183-
-------
oo
-p-
TABLE 3.4-3
MODULE HEAT REQUIREMENT
Basis: 125 MM scfd of SNG produced
Unit
HDS Preheater
Superheater & Methanator
Preheater
Boiler
Benfield C02 Removal
Glycol Dehydration
Hydrogen Generation
Unit Heater
Heat Requirement
Per MM scfd
Output (MBtu/MM scfd)
4.00x103
37.87xl03
36.68x103
24.00xl03
l.OOxlO3
1.22xl03
Unit Heat Requirement
(Btu/day)
5.0xl08
47.3xl08
45.8xl08
30.0xl08
1.3xl08
1.6xl08
130.6xl08
-------
TABLE 3.4-4
00
Ul
MODULE ATMOSPHERIC EMISSIONS
(Ibs/day)
Process Particulates S02
HDS
Hydi
Preheater
rogen Generation
17.9
5.7
7.7
2.5
NOX
197
100
.5
.5
HC
6
2
.1
.0
CO
11
3
.0
.6
Unit Heater
Superheater and 2nd 169.5
Stage Methanator
Preheater'
Steam Boiler
Glycol Unit
Benfield System
Naphtha Storage
Fugitive Hydrocarbon
Losses
TOTAL 469.4
73.2 1,870.0
57.8 104.0
164.
4.
107.
-
-
1
7
5
70.
2.
46.
-
-
9
0
4
1,810
51
1,186
-
•
.7
.4
.0
56
1
9,340
403
6,375
.1
.6
*
.2
.0
101.
2.
6.
-
-
0
9
6
202.7 5,216.1 16,241.8 229.1
'includes 9300 Ib/day methane loss from C02 vent.
-------
TABLE 3.4-5
MODULE FUEL COMBUSTION EMISSION FACTORS
Emission Factors
Pollutant (lb/103 gal)
Particulates 4.4
S02 1.9
NO 48'5*
CO 2'7
XA NO factor of 72.8 lb/103 gal is used for
the hydrogen generation heater because of
the higher combustion temperatures (RA-119)
-186-
-------
Public Service Electric and Gas Company (LO-095) reports
that their Harrison SNG plant with a gasification capacity of
20 mm scfd releases approximately 270 tons per year of methane
from the Benfield C02 removal system. The specific point of
emission is the C02 vent from the top of the potassium carbonate
regenerator. Therefore, for the 125 mm scfd capacity of this
module, approximately 1,700 tons per year or 9,300 pounds per
day of methane is emitted.
Another source of hydrocarbon emissions is the naphtha
storage tanks located on the plant site. These floating roof
tanks will have a total capacity of 800,000 barrels of naphtha
(one-month plant requirement). The emissions are estimated using
the supplement to the 1973 EPA emission factor book on hydro-
carbon losses from floating roof tanks (EN-071). The factor for
naphtha jet fuel is used since its Reid vapor pressure is close
to that of naphtha. This factor is 0.012 lb/day-103 gal.
The final source of air emissions from this module is
from fugitive hydrocarbon losses. These miscellaneous leaks
result from process leaks at pump seals, valve stems, and flanges.
Although these emissions may become significant, particularly
in a facility where plant maintenance is not given sufficient
attention, it is difficult to quantify these types of emissions.
As a rough estimate, 0.1 weight percent of the incoming feed
is considered lost as a fugitive emission (RA-119).
Liquid Effluents
The boiler make-up feedwater for the plant is first
passed through an ion exchange resin unit for demineralization.
Regeneration of this demineralization unit with acid and caustic
wash water streams is one of the plant's liquid effluent streams.
-187-
-------
Other sources of liquid wastes are the plant cooling system and
boiler blowdown streams and waste solution from the Benfield
system. These streams are discharged into holding ponds on the
plant site where the water is evaporated. Since no liquid
leaves the plant boundaries, the module wastewater effluent is
considered to be zero.
Solid Wastes
There are no daily discharges of solid wastes from an
SNG plant gasifying naphtha. Disposal of spent, inert catalysts
occur periodically but are believed to pose no environmental
problems.
3.5 Comparison of Module Emissions
In the previous subsections the emission rates are
related to the specific module charge capacity. This approach
is used in order to present the emission impact of a typical
size plant for the specific industry and hence facilitate its
environmental assessment for each technology. In this sub-
section all of the modules are adjusted to a 1012 Btu/day output
of primary product. This adjustment is made in order to present
the different module emissions on a common basis and provide a
convenient comparison of the emission impact of the various
technologies. This comparison is presented in Table 3.5-1,
The large hydrocarbon emissions that result from these modules
are primarily a result of fugitive losses (assumed 0.1 wt% of
throughput).
-188-
-------
TABLE 3.5-1
COMPARISON OF MODULE EMISSIONS
Basis: 1012Btu/Day Output Primary Fuels
EMISSIONS AND EFFLUENTS
Air Emissions (Ib/day)
Particulates
SOX
CO
NOX
HC
FUEL OIL[1]
REFINERY
6,720
17,000
1,280
12,600
78,700
GASOLINE lij
REFINERY
12,300
26.800
2,680
35,340
90,600
PEAK-SHAVING ^
LNG PLANT
5,900
790
5,600
75,800
4,700
BASE LOAD[31
LNG PLANT
2,350
11,200
2,540
82,460
43,200
SNG PLANT
3,750
1,620
2,310
41,800
130,000
[2]
00
VO
Water Effluents (Ib/day)
Suspended Solids
Dissolved Solids
Organic Material
Solid Wastes (Ib/day)
266
9,850
56
295
10,900
62
0
0
0
8,500
16,500
0
0
0
0
0
negligible
[^Primary fuels for the refinery-modules are considered to be the gasoline and middle distillate or
light fuel oil product streams. The total heating values of these product streams (gasoline: 5.248x
106 Btu/bbl, middle distillate: 5.7xl06Btu/bbl, light fuel oil: 5.825x106Btu/bbl) are combined and
adjusted to a 1012 Btu/day output basis.
[2]Pipeline quality (1000 Btu/SCF) synthesis gas is considered to be the primary fuel from the SNG plant.
13]Primary fuel from the LNG facility is regasified liquefied natural gas (1000 Btu/SCF).
-------
4.0 MONITORING TECHNOLOGY
Monitoring is normally divided into two categories, source
and ambient. Source monitoring typically involves measuring both
the concentration and flow rate in an exit stream to determine
the total amount of a particular species emitted. Ambient moni-
toring generally involves measurement of the concentration of a
species at a remote point where it has been diluted by mixing.
Monitoring may be either continuous or intermittent.
Ambient monitoring is generally performed on a continuous basis
since the regulations are written in terms of the maximum allow-
able average concentration over a particular time interval and
since the dilution of a species depends on mixing conditions
which are difficult to predict with any certainty. Without
continuous monitoring, there is little assurance that the
maximum levels actually are detected.
In source monitoring, the concentrations do not vary so
widely or so rapidly, and maximums may sometimes be predicted
as a consequence of changes in operating parameters (such as
fuels). Regulations often are expressed in terms of allowable
emissions as a function of weight or heat content of the feed-
stock. Because of this, source monitoring has tended to be of
the intermittent type in the past, although continuous source
monitoring is now required for some pollutant species in some
industries.
Whether ambient or source, monitoring methods can be
divided into several categories. These include manual labora-
tory methods, automated laboratory methods, manual field methods,
and automated field methods.
-190-
-------
The impetus for monitoring is normally provided through
a requirement to demonstrate compliance with federal and/or state
regulations. For many species these regulations include both
source and ambient limits.
Most regulations designate a particular analysis proce-
dure as the standard or "reference" method for a given species.
These reference procedures have tended to be of the manual labora-
tory type since this type of analysis has historically provided
the greatest confidence level in the results. This is probably
because the basic standards in this type work are normally chemi-
cals of reliable purity whereas the standards used in manual or
automatic field methods are often derived from a reference gas
which may be questionable because its concentration is dependent
on certain temperatures or flow rates in a calibration unit.
Field methods do have the advantage that they provide real time
analysis whereas laboratory methods require the collection of
a sample which must then be stabilized or preserved in some
manner to keep it from changing in composition during transport
to the laboratory.
Refineries, LNG plants, and SNG plants will all produce
certain emissions as has been discussed in earlier sections.
The spectrum of emissions is greatest from refineries because
of the raw nature of the feedstocks and because of the wide
variety of processes which may occur. The natural gas feed-
stock to LNG liquefaction plants is normally quite clean com-
pared to crude petroleum. It is purified even more prior to
liquefaction (thus producing some emissions), so that the feed-
stock to the regasification plant is very clean. SNG plants
using crude oil as a feedstock could conceivable have as broad
a spectrum of emissions as refineries; however, as was discussed
in Section 2.3, it is unlikely that any SNG plant in this country
will use crude oil as a feedstock. This discussion will thus assume
-191-
-------
a light naphtha feedstock. Since light naphtha is derived from
refineries, it can be seen that emissions from SNG plants will
be more limited than those from refineries.
Since the emissions from refineries should include those
from SNG plants and LNG plants, refineries will be taken as a
representative case for the purposes of monitoring requirements.
As discussed earlier most monitoring requirements are the result
of state or Federal regulations, although in some cases source
monitoring may be used as an input to control of operating param-
eters. Monitoring requirements will be divided into ambient air
monitoring, source air monitoring, and source water monitoring,
all as applicable to petroleum refineries. For each category
a brief review of the Federal regulatory framework will be pre-
sented, along with a discussion of monitoring methods. A dis-
cussion of solid waste disposal site monitoring is also provided.
4.1 Ambient Air Quality Monitoring
4.1.1 Background
For ambient air quality, Federal regulations have estab-
listed certain criteria pollutants, namely, nitrogen dioxide
(N02), sulfur dioxide (S02), carbon monoxide (CO), photochemical
oxidants (03), particulates, and non-methane hydrocarbons (NMHC).
Both primary standards (to safeguard human health) and secondary
standards (to prevent damage to clothes, buildings, plants, ani-
mals, etc.) have been established. The averaging time varies
for different species, with short term averages generally ex-
pressed as "not to be exceeded more than once per year". The
actual regulations are quite lengthy, and are described in the
Federal Register, Volume 36, No. 228, page 22384 and later modi-
fied slightly in Federal Register, Volume 38, No. 178, page 25678.
-192-
-------
A summary is provided in Table 4.1-1. It should be noted that all
measurements are to be corrected to reference conditions of 25°C
and 760 mm Hg.
The various states have established their own ambient
air quality regulations, and in many cases they have different
averaging periods and/or more stringent limits than Federal regu-
lations. Due to their wide variety these will not be considered
here, but it is noteworthy for the purposes of monitoring around
a refinery that many include regulations on hydrogen sulfide (H2S).
In addition to specifying limits on the criteria pollu-
tants, the Federal government has also established reference
methods for their analysis. These procedures for particulates,
total oxidants, nitrogen dioxide, and sulfur dioxide are outlined
in Federal Register, Volume 36, No. 84, Part II, April 30, 1971..
For non-methane hydrocarbons, the procedure is defined in Federal
Register, Volume 36, No. 228, page 22394. Because some of these
procedures are inconvenient for continuous field monitoring, and
because many other types of analyzers are presently being used
for field monitoring, certain mechanisms have been defined whereby
other analysis procedures may be designated as reference or equi-
valent methods. These mechanisms are outlined in Federal Register,
Volume 40, No. 33, page 7042, 1975. Because many types of in-
struments are currently under evaluation as reference or equiva-
lent methods, it is not possible at this time to specify which
monitoring methods will be acceptable.
4-1-2 General Monitoring Considerations
One of the most difficult areas in ambient air monitor-
ing is the siting of the various monitors. This is especially
true when the purpose of the monitoring is to evaluate the impact
of a particular source. This is typically done based on dispersion
-193-
-------
TABLE 4.1-1
SJMMARY OF FEDERAL AMBIENT AIR STANDARDS
All numbers in mi crograms/ cubic meter (yg/m3)
Pollutant
SO 2
Particulate
CO
Total Oxidant
Non-Methane HC
Averaging
Time
Annual
24 hour*
3 hour--
Annual
24 hour*
8 hour*
1 hour*
1 hour*
3 hour*
Primary
Standard
80
365
-
75
260
10,000
40,000
160
160
Secondary
Standard
-
1,300
60
150
10,000
40,000
160
160
(6-9 AM)
NO 2
Annual
100
100
*Not to be exceeded more than once per year.
-194-
-------
modeling which utilizes historical meteorological data plus data
regarding the source to predict the location of maximum pollutant
concentrations for annual and short term averages. The Environ-
mental Protection Agency has recently undertaken several studies
on optimum siting criteria for several pollutants.
For the reference procedures defined in the Federal
Register, the accuracy of the analysis is given, and it is ex-
pected that any new reference or equivalent methods will be of
equal or greater accuracy. For SO2 analysis by the reference
method, the relative standard deviation at the 95 percent con-
fidence level is 4.6 percent. For total oxidants the accuracy
is given as ±7 percent. For carbon monoxide an accuracy of ±1
percent of full scale is given (full scale normally is 58 milli-
grams per cubic meter). The accuracy of particulate analysis
is given as ±50 percent, and for non-methane hydrocarbons the
accuracy is 2 percent of full scale.
Costs are difficult to assess on a per sample basis
for continuous analysis. Instrument costs are in the following
ranges; however, multicomponent analyzers may be even higher.
Pollutant Thousands of Dollars
S02 4-8
N02 4-8
Ozone 3-5
NMHC 5-10
CO 3-10
Particulates 0.5-15
In some cases a calibration unit costing several thousand dollars
may be required to accomplish a multi-point calibration. In
-195-
-------
addition, some type of temperature-controlled shelter is required,
as is a data recording system. Finally, processing the data will
require manpower and/or additional hardware. It can be seen that
actual monitoring costs will depend on availability of facilities
and manpower, and will be different in almost every case. Many
companies now offer an ambient air monitoring service in which
they assume total responsibility for all instrument operation
and data processing, and provide a summary report to the client
on a regular basis.
The most commonly used nitrogen oxide monitors employ
the chemiluminescent technique. This method is specific for
nitric oxide, so nitrogen dioxide must be converted to nitric
oxide prior to its analysis. By obtaining a nitric oxide measure-
ment with and without conversion of nitrogen dioxide, the nitro-
gen dioxide concentration can be obtained by difference. Other
popular methods for measuring nitrogen oxides include electro-
chemical analyzers, second derivative spectroscopy, bubblers and
colorimetric analyzers, and membrane-electrochemical analyzers
(IN-056).
Sulfur dioxide and hydrogen sulfide are most commonly
measured with flame photometric analyzers. These detectors act-
ually respond to the total sulfur content of a molecule, there-
fore, selectivity scrubbers are installed on the inlet to re-
move all sulfur species but the one of interest. Other analysis
methods for S02 include electrochemical analyzers (membrane and
non-membrane), pulsed fluorescent analyzers, second derivative
spectroscopy analyzers, and colorimetry (IN-056). S02, HaS, COS,
CSz, and other sulfur species can be measured with gas chromato-
graphic analyzers using flame photometric detectors.
The most commonly used method for continuous ozone
analysis is the chemiluminescent technique, in which ozone is
reacted with ethylene or in some cases with Rhodamine-B. Other
-196-
-------
methods include ultraviolet absorption, electrochemical analyzers,
second derivative spectroscopic analyzers, and wet chemistry bubblers
Carbon monoxide is normally measured with either infra-
red analyzers or gas chromatographs which convert the CO to CHi»
and measure it via a flame ionization detector. Electrochemical
analyzers for CO also are available.
Hydrocarbons are most commonly measured using chromato-
graphic separation of the methane, detection via a flame ioniza-
tion detector, and determination of non-methane hydrocarbons
as the difference between total hydrocarbons and methane (IN-056)•
Continuous analyzers are now available which separate the hydro-
carbons into methane, ethylene, acetylene, and total hydrocarbons.
To obtain a more detailed analysis, a manually operated gas chroma-
tographic system or gas chromatograph - mass spectroscopy combina-
tion is required. This may be achieved by collecting bag samples
and taking them to the laboratory, or by installing an instrument
in a field site. Continuous analysis of this sort is difficult
since some concentration of the sample is normally required.
Aldehydes and organic acids are detected by the flame
ionization detectors in regular environmental chromatographs. In
the results, however, they are lumped into the non-methane hydro-
carbons. These species can be detected using bag samples and
laboratory chromatographs or with pulse polarographs.
Continuous ammonia analyzers are available based on an
electrochemical principle.
Particulates are most commonly determined using the
EPA High Volume sampler. A weighed filter is exposed to a
measured flow of air for 24 hours, then reweighed. Average
-197-
-------
particulate mass per unit volume for the 24 hour period is obtained.
Other samplers based on mass measurement using beta particles
have become available in recent years. These systems can be
programmed to measure for much shorter time periods than 24 hours.
Samples collected with particulate analyzers can be
subjected to a detailed analysis for content of various trace
elements. The most popular methods for trace element analysis
are x-ray fluorescence, atomic absorption, spark source mass
spectrometer, and neutron activation analysis.
4.2 Source Monitoring (Air)
4.2.1 Background
As was the case with ambient monitoring, the Federal
government has established regulations for source monitoring;
however, these have been on an industry by industry basis, and
are for new sources. Many states have also established emission
regulations, but the variety of these is too great and changes
too rapid to be covered here.
For the purpose of source monitoring at a petroleum
refinery the relevant regulations cover emissions from fluid
catalytic cracking unit catalyst regenerators, fluid catalytic
cracking unit incinerator-waste heat boilers, fuel gas combus-
tion devices, and storage vessels for petroleum liquids (see
Standards of Performance for New Stationary Sources, Federal
Register, Volume 39, No. 47, page 9308, 1974). It is noteworthy
that these regulations define petroleum as the crude oil removed
from the earth and the oils derived from tar sands, shale, and
coal. The regulations establish emission limits for particu-
lates (both mass and opacity limits), carbon monoxide (as con-
centration by volume in the exhaust gas), and sulfur dioxide
(as H2S concentration in the fuel gas).
-198-
-------
Monitoring, calibration, and reporting procedures are
also specified in the same article of the Federal Register.
For the purposes of determining compliance with the emission
limits, certain EPA methods are specified:
Method 1 - Sample arid Velocity Traverses for
Stationary Sources
Method 2 - Determination of Stack Gas Velocity and
Volumetric Flow Rate
Method 3 - Gas Analysis for Carbon Dioxide, Excess
Air, and Dry Molecular Weight
Method 4 - Determination of Moisture in Stack Gases
Method 5 - Determination of Particulate Emissions from
Stationary Sources
Method 6 - Determination of Sulfur Dioxide Emissions
from Stationary Sources
Method 7 - Determination of Nitrogen Oxides Emissions
from Stationary Sources
Method 8 - Determination of Sulfuric Acid Mist and
Sulfur Dioxide Emissions from Stationary
Sources
Method 9 - Visual Determination of the Opacity of
Emissions from Stationary Sources
Method 10- Determination of Carbon Monoxide Emissions
from Stationary Sources
-199-
-------
Method 11- Determination of Hydrogen Sulfide Emissions
from Stationary Sources
Methods 1 through 9 are described in Federal Register,
Volume 36, No. 247, pages 24882-24895. Methods 10 and 11 are
described in Federal Register, Volume 39, No. 47, pages 9319-9323
These reference methods for determining compliance
are generally non-continuous, manual methods. However, contin-
uous monitors are also required for opacity and carbon monoxide
measurements on the fluid catalytic cracking unit catalyst re-
generator, hydrogen sulfide in fuel gases, and sulfur dioxide in
exhaust streams from fuel gas combustors. The specific form of
the continuous analyzers is not defined in the Standards of Per-
formance for Petroleum Refineries.
This problem is addressed in a more recent publication,
Stationary Sources, Proposed Emission Monitoring and Performance
Testing Requirements, Federal Register, Volume 39, No. 177, page
32852. It should be noted that these are proposed rules, not
promulgated standards. Probably the most significant aspect of
these proposed rules is that no "product line certification" is
given whereby certain instrument specifications (measurement
method, response time, etc.) would be provided for each pollu-
tant, and any instrument meeting these specifications could be
installed on a specific source in fulfillment of the monitoring
requirements for that source. Instead, certain performance speci-
fications are defined, and it is left to the owner or operator
of each source to demonstrate that the continuous monitoring
system he has selected meets those performance specifications
on his source. Detailed procedures and forms are provided for
demonstrating agreement with the performance specifications for
opacity, nitrogen oxides, sulfur dioxide, and oxygen. Perfor-
mance specifications for systems which monitor hydrogen sulfide
and carbon monoxide are to be proposed at a later date.
-200-
-------
These same proposed rules include a modification to
EPA Method 9 for the visual determination of opacity of emissions
from stationary sources. Special exceptions are provided for
those who have already installed continuous analyzers since the
first standards were promulgated on December 23, 1971.
4.2.2 General Monitoring Procedures
Generally speaking, accurate source monitoring is more
difficult than ambient monitoring. The gas streams typically
are hot, difficult to access, and the pollutant concentrations
may vary as a function of position in the exit stack. Introduc-
tion of standard gases in a meaningful manner often is difficult.
As with ambient monitoring, source monitoring may be
intermittent or continuous, manual or automated. In'addition,
continuous monitors can be divided into in-situ and external
systems. The external monitors normally require a sample condi-
tioning system.
As discussed earlier, the reference methods tend to
be of the intermittent manual type. Because of sampling diffi-
culties these tend to be somewhat more expensive than ambient
monitoring. As an example, it has been estimated by EPA that
one particulate analysis using Method 5 will cost from $3,000 to
$10,000 depending on the source, including some 300 man-hours
of effort (Federal Register, Volume 39, No. 47, page 9309). A
single ambient particulate measurement using a high-volume sampler
costs only a small fraction of this amount. The Federal Register
does not list measurement accuracies for Methods 1-11.
The same general instrument types are used for contin-
uous source monitoring as for ambient monitoring. Pollutant
concentrations in stack gases are normally several orders of
-201-
-------
magnitude higher than ambient levels so that the instruments
must either be designed for higher levels or dilution systems
must be used. For external monitors, a sample conditioning sys-
tem often is required to provide an air sample suitable for
analysis. This involves filtering out particulates (unless par-
ticulates are being monitored), removing excess water, and pro-
viding for introduction of calibration gases.
Because sample conditioning systems increase cost and
complexity, and may decrease reliability, several in-situ moni-
tors have been developed. In these systems the detection or
measurement portion is mounted directly in the exit stack. As
with the external monitors, however, calibration is often diffi-
cult. An excellent discussion of external analyzers, in-situ
analyzers, and remote sensors has been provided by Nader (NA-113).
Nitrogen oxides are commonly measured with chemilumi-
nescent analyzers, infrared analyzers, or ultraviolet analyzers.
Sulfur dioxide is measured with ultraviolet analyzers, flame
photometric analyzers, infrared analyzers, or pulsed-fluores-
cence analyzers. Carbon monoxide is most commonly measured with
infrared analyzers. Continuous particulate analyzers utilize
the beta particle detector.
4.3 Effluent Water Monitoring
4.3.1 Background
Based on the Federal Water Pollution Control Act, the
Federal government has established guidelines and standards for
the Petroleum Refining Point Source Category (see Federal Register,
Volume 39, No. 91, page 16560, 1974). Standards are established
for the following subcategories:
-202-
-------
Topping Subcategory
Cracking Subcategory
Petrochemical Subcategory
Lube Subcategory
Integrated Subcategory
These standards include both existing and new sources,
and are sub-divided according to the following groups:
1) Effluent limitation guidelines representing the
degree of effluent reduction attainable by the
application of the best practicable control tech-
nology currently available (BPCTCA)
2) Effluent limitations guidelines representing
the degree of effluent reduction attainable by
the application of the best available technology
economically achievable (BATEA)
3) Standards of performance for new sources
4) Pretreatment standards for new sources
Exceptions to the limits prescribed for the BPCTCA
group may be obtainable from the EPA Regional Administrator in
certain cases.
Limits are listed in terms of maximum for any one day
and average of daily values for thirty consecutive days, and
the units are expressed as kilograms per thousand cubic meters
of feedstock. The specified limits in each group are scaled
according to size factors and process configurations for indivi-
dual plants. (Proposed"amendments to the guidelines are provided
in Federal Register, Vol. 39, No. 202, page 37069, 1974.) The
species for which limits have been set include the following:
-203-
-------
BOD5 Ammonia as N
TSS Sulfide
COD Total Chromium
Oil and grease Hexavalent Chromium
Phenolic compounds pH
Total organic carbon (TOG) limits are set for runoff
water and once through cooling water. Also, in some cases,
TOG may be substituted for COD if the effluent contains more
than 1000 milligrams per liter of chloride.
EPA has established test procedures for the analysis
of many pollutants, including those listed above (see Federal
Register, Volume 38, No. 199, page 28758, 1973). These proce-
dures are contained in the document "Methods of Chemical Analysis
of Water and Wastes", U.S. Environmental Protection Agency, EPA
report 625-/6-74-003, 1974. For each species, an analysis method
is described including sampling and preservation techniques,
the required apparatus, and the precision and accuracy.
Analysis costs are dependent on the required number of
analyses of each type. Assuming only one analysis of each
type, however, commercial water analysis laboratories will nor-
mally charge prices in the range of those in Table 4.3-1.
It should be noted that the Petroleum Refining Point
Source Category briefly addresses solid waste control. No defi-
nite regulations or limits are established; however, recommen-
dations for choice of landfill sites and for record keeping re-
garding these sites are presented.
4.3.2 General Water Monitoring
Automated analyzers for water analysis have been
available for many years for both field and laboratory applications
-204-
-------
TABLE 4.3-1
WATER ANALYSIS COSTS
BOD5 $10 - 15
TSS 5-7
COD 10 - 15
Oil and Grease 10 - 15
Phenolics 15 - 20
Ammonia 10 - 15
Sulfide 8-12
Total Chromium 3-5
Hexavalent Chromium 5 - 8
pH . 1
-205-
-------
The species normally measured in the field are temperature, pH,
conductivity, and dissolved oxygen; however, recent developments
in membrane electrodes have greatly expanded the parameters which
can be detected.
Automated laboratory analyzers are very useful when
large numbers of similar samples are to be analyzed; however,
for a few samples the set-up time largely negates the automatic
analysis feature.
The general situation regarding laboratory versus field
and manual versus automated samplers has been well summarized
by Phillips and Mark (PH-019). Table 4.3-2 provides a summary
of their results.
It can be seen that automated field analyzers are
available for many of the species for which effluent limitations
have been established for petroleum refineries. These include
total organic carbon, chemical oxygen demand, ammonia, and
chromium. Of course pH monitors are also available. Automated
laboratory analyzers can be used for analysis of phenolics and
sulfide.
Other metals besides chromium could be analyzed for
if desired. Manual or automated laboratory analyzers using
the atomic absorption method or neutron activation method can
detect many metals down to the sub-parts per billion level.
Both soluble species and solid species (present in the suspended
solids) can be examined.
Analysis for specific hydrocarbon species normally
requires a gas chromatograph for separation plus some form of
detector. Combining a mass spectrograph with the gas chromato-
-206-
-------
TABLE 4.3-2
TYPICAL INSTRUMENTAL METHODS
Instrument Method
Pollutant Measured
Manual Laboratory Analyzers
Atomic absorption
Colorimetric
Emission spectrometry
Gas chromatography
Gas membrane electrodes
Ion selective electrode
Activation analysis
X-ray fluorescence
Gas chromatography/
Mass spectrometry
Thin-layer chromatography
Infrared
spectrophotometry
Metals
Metals; nutrients (ammonia,
nitrate, nitrite, phosphate)
chemical oxygen demand; total
organic carbon
Metals; phosphorus
Pesticides
Dissolved oxygen; ammonia,
nitrite; BOD
Nitrate
Metals; nitrogen; phosphorus
Metals
Pesticides
Pesticides
Total organic carbon
Automated Laboratory Analyzers
Atomic absorption Metals
Colorimetric Metals; nutrients
Gas chromatography Pesticides
Colorimetric
Electrode
Volumetric titration
Manual Field Monitors
Metals; nitrite; phosphate
DO; metals
DO; nitrate
Atomic absorption
Colorimetric
Automated Field Monitors
Hg
Electrometric
Flame ionization; infrared
Cr; MnO,,-; P0,,= ; Fe; Cu; NH3;
N03-; N02-; total phosphorus;
chemical oxygen demand
Cu; dissolved oxygen; NH3; N02-
Total organic carbon
-207-
-------
graph provides a particularly powerful tool for hydrocarbon
studies, although analysis of this sort becomes quite expensive.
For the measurement of aldehydes, manual polarographic
and gas chromatographic techniques are available.
4.4 Solid Waste
Refinery solid wastes are normally either incinerated
or disposed of in a landfill. In the case of incineration, the
problem reverts to the air monitoring situation. Landfill dis-
posal is normally regulated on a state or local basis, if at all.
The principles set forth in EPA's "Land Disposal of Solid Waste
Guidelines" (40 CFR, Part 241) may be used as guidance for
acceptable land disposal techniques, however.
Solid waste disposal from refineries is briefly addres-
sed in Petroleum Refining Point Source Category Effluent Guide-
lines and Standards, Federal Register, Volume 39, No. 91, page
16563, 1974. This section points out that best practical control
technology and best available control technology as they are
known today require disposal of the pollutants removed from waste
waters in the form of solid wastes and liquid concentrates.
Whether thes'e wastes contain hazardous metals or organic species
is not well established. Caution thus dictates that landfill
sites be selected to prevent horizontal or vertical transport
of species from the landfill into groundwater or surface waters.
When this is not accomplished by the natural geologic conditions,
adequate liners should be provided. It is recommended that per-
manent records be kept as to the location and nature of the
disposal sites.
-208-
-------
Monitoring requirements for landfills in general and
those for refinery wastes specifically have not been well
established. The problem is basically viewed as one where
soluble species dissolve in water percolating through the
landfill, and are transported into nearby groundwater aquifers.
The identity and solubility of the species in the landfill thus
are important, but both of these can change with time in the
complex chemical and bacterial environment of the landfill.
To further complicate matters, soluble species transported
out of the landfill area may subsequently be removed via
interaction with soil particles.
A series of monitoring wells can be installed around
a landfill to monitor any changes in groundwater quality. Such
a network must be very carefully designed, however, as ground-
water movement often is quite slow. Considerable contamination
can occur before it is detected, and it is almost impossible to
remove all the contamination once it reaches the aquifer.
-209-
-------
5.0 EMISSION CONTROL METHODS
In this section emission control techniques are ex-
amined for the air, water, and solids emissions resulting from
the refinery, LNG, and SNG modules. Both currently available
control methods and potential control techniques are considered.
Due to the similarity in emissions and emission sources and the
fact that LNG and SNG processes may be considered as one process
or subset of petroleum processing, the refinery, LNG, and SNG
technologies are considered together rather than individually.
Control technologies are considered in relation to the specific
emission, i.e., particulate, hydrocarbon, CO, etc., and the specific
source or application. Controls are discussed for gaseous,
liquid, and solid wastes.
-210-
-------
5.1 Air Emission Control
As established in Section 3.0 of this study, the
major air emissions from the refinery, SNG, and LNG industries
contain particulates, sulfur oxides, carbon monoxide, nitrogen
oxides, and hydrocarbons. The current technology and practices
used in the control of these emissions are addressed in this
section.
5.1.1 Particulates Emission Control
The major source of particulates from the LNG, SNG,
and refinery modules are process heaters, boilers, incinerators,
and the fluidized catalytic cracking unit (FCCU) catalyst re-
generator in the gasoline refinery operation. The particulates
from the process heaters and the boilers result from ash within the
fuel oil, naphtha, or fuel gas combusted. Presently, there is
no incentive for particulate controls on these types of process
heaters and boilers, due to the low concentration of particu-
lates in the flue gas. There is presently no proposed Federal
Standard for particulate concentrations in flue gases from
process heaters used in refining, LNG, or SNG plants.
5.1.1.1 Sludge Incineration Particulate Control
The proposed particulate emission standard from the
EPA for sludge incineration is 0.031 gr/dscf (EN-072). These
particulates primarily result from fly ash in the incinerated
sludge. Average uncontrolled particulate emissions are 0.9
grain/dscf for a multiple-hearth incinerator and 8.0 grain/dscf
for a fluidized bed incinerator (EN-072). In order to meet the
proposed standard, particulate removal efficiencies of 96.6%
-211-
-------
for a multiple-hearth incinerator and 99.6% for a fluidized
bed are required. Flow diagrams for the controlled multiple-
hearth furnace and the controlled fluidized bed reactor are given
in Figure 5.1-1.
Typical particulate control for sludge incinerators
is by a venturi scrubber or an impingment-type scrubber (EN-072).
With a venturi scrubber (Figure 5.1-2), water which is fed
through jets in a venturi section is suspended as water drop-
lets. The fly ash collects on the suspended water droplets
and is removed along with the water in a cyclone. An impinge-
ment scrubber (Figure 5.1-3) works on much the same principle.
Fly ash laden gas is blown upward through a series of perforated
plates which are covered with a stream of water. The gas atomi-
zes the water and the water spray droplets collect the fly
ash. The fly ash and water are again removed by a cyclone.
Overall efficiencies for a single plate range from 90 to 9870
for 1 micron particles or larger (NA-029). The advantage of
the impingement method is a lower pressure drop through the
scrubber. An alternate method for particulate removal would
be bag or fabric filters (Figure 5.1-4). Bag filters will give
removal efficiencies of greater than 9970; however, the bag
filters are generally more expensive to install and operate
(NA-029). Electrostatic precipitators are another alterna-
tive; however, they too are more expensive than wet scrubbing
(NA-029). An approximate monetary comparison is given in Table
5.1-1.
Disposal of incinerable sludge by methods that do not
involve the use of incineration are most desirable from the
air pollution standpoint. In many cases these alternative
-212-
-------
SLUDGE
CAKE
SHAFT COOLINGi
AIR FAN
Controlled multiple-hearth furnace, scrubber.
EMISSIONS
FAN
ASH PIT
AIR BLOWER AIR BLOWER
Controlled fluidized bed reactor, scrubber.
FIGURE 5.1-1 INCINERATOR EMISSIONS CONTROLS
SOURCE: (EN-072)
-213-
-------
liquid
in
gas & liquid
to entrain-
ment separation
FIGURE 5.1-2 VENTURI SCRUBBER
SOURCE: (NA-029)
-214-
-------
IMPINGEMENT
BAFFLE STAGE
AGGLOMERATING
SLOT STAGE
•. IMPINGEMENT SCRUBBER
TARGET,
PLATE
a
D
a
n
D
o
o
o
WATER
LEVEL
ORIFICE
PLATE
GAS FLOW
ARRANGEMENT OF "TARGET PLATES"
IN IMPINGEMENT SCRUBBER •
WATER DROPLETS ATOMIZED
AT EDGES OF ORIFICES
DOWNSPOUT TO
LOWER STAGE
k. IMPINGEMENT PLATE DETAILS
FIGURE 5.1-3 IMPINGEMENT PLATE SCRUBBER
SOURCE: (NA-029)
-215-
-------
CLEAN AIR
OUTLET
DIRTY AIR1
INLET
CLEAN AIR
SIDE
FILTER
BAGS
CELL PLATE
FIGURE 5.1-4 TYPICAL SIMPLE FABRIC FILTER BAGHOUSE DESIGN
SOURCE: (NA-029)
-216-
-------
TABLE 5.1-1
RULE-OF-THUMB COSTS OF TYPICAL COLLECTORS OF
STANDARD MILD-STEEL CONSTRUCTION
(MAR 1973)
Dollars Per Cubic Feet Per Minute
Type of Collector
Equipment Erection Yearly Maintenance
Cost Cost and Repair Cost
Mechanical Collector
0.07-0.25 0.03-0.12 0.005-0,02
Electrostatic Precipitator 0.25-1.00 0.12-0.50 0.01-0.025
Fabric Filter
0.35-1.25 0.25-0.50
0.02-0.08
Wet Scrubber
0.10-0.40 0.04-0.16
0.02-0.05
SOURCE: (RE-070)
-217-
-------
methods may also prove more economical. One method for
sludge removal is landfill. Air emissions from landfill
are limited to diesel combustion of the hauling and compacting
equipment and miscellaneous particulate emissions entrained in
the air by earth-moving equipment. The miscellaneous particu-
lates can be controlled by proper waterspraying techniques.
Other alternatives are described in Section 5.3 on Solids
Emission Control.
5.1.1.2 FCCU Particulate Control
The particulates from the FCCU catalyst regenerator
result from catalyst fines. The proposed standard particulate
level from the FCCU regenerator is 0.022 grain per scf of flue
gas. Particulate emissions from the catalyst regeneration process
are on the order of 0.1 to 0.2 pounds per ton of catalyst re-
circulated (NA-029). Based on the estimated module air flow
rate of 64,000 cfm and a catalyst regeneration rate of 2,500
tons per hour, it can be calculated that the catalyst regenerator
will require 98% removal of fines to meet this requirement (CU-016)
Modern-day techniques of wet scrubbing or baghouse
filters are quite adequate for this type of removal, except for
the fact that the flue gas exits the regenerator at a temperature
too high for these control methods. Normally, a multiple
cyclone type arrangement is employed for particulate control.
Basically, a cyclone is a settling device in which a strong
centrifugal force, acting radially, is used in place of a rela-
tively weak gravitational force acting vertically (Figure 5.1-5).
The multiple cyclone setup removes nearly all of the particles
greater than 40 microns (WI-073). The gases from the cyclones
are introduced into a carbon monoxide boiler where the CO is
combusted to COS . The sensible heat of the flue gas stream and
the heat of combustion of the CO are used to produce steam.
-218-
-------
Dust
and
gas
FIGURE 5.1-5 CYCLONE
-219-
-------
From the CO boiler the gas is routed through an electro-
static precipitator where the remaining particulates are further
reduced. The electrostatic precipitator removes particulates
by charging the particles with a high-voltage direct-current
corona. The charged particles are collected on a large plate-
like collection electrode. The dust is removed from the electrodes
by rapping or washing. An electrostatic precipitator is shown
in Figure 5.1-6. Typical fluid catalytic cracker particulate
control systems are shown in Figure 5.1-7.
An alternate method for removal of fines coming from
the multiple cyclones would be a process developed by Shell
which uses a multiple tube, swirl vane type of centrifugal
separator (WI-073). In conjunction with the separator a
turbo-expander is used to recover some of the power in the
flue gas stream. The main separator has an efficiency of 99.5%
for the 10 micron size particle, which compares to 85% for the
highest efficiency large cyclone (WI-073). The centrifugal
separator concentrates the greatest part of the particulate
matter in a small underflow stream. This underflow stream
is then cleaned by use of baghouses or wet scrubbers. A flow
scheme of the Shell-type process is given in Figure 5.1-8.
A very new method for recovering catalyst fines is
by the use of a granular bed filter. The granular bed filter
was designed to reduce fine catalyst losses and is used in place
of both the multiple cyclones and the electrostatic precipitator.
The catalyst dust-laden gas is blown into a chamber containing
granular sand bed filter elements. The catalyst fines are
collected on the surface and in the interstices of the bed.
When a predetermined resistance pressure is reached across the
bed, a short pulse of high pressure air is shot in reverse flow
through the filter elements. The air fluidizes the sand bed
and releases the catalyst fines which are captured within the
chamber. A granular bed filter is shown in Figure 5.1-9.
-220-
-------
FLUE GAS
ELECTRODE
WEIGHT
UUU A UUDD
FLY ASH DISPOSAL
•DRY OR VET SLUICING
COLLECTING
PLATES
STACK GAS
HIGH TENSION
WIRE ELECTRODE
FIGURE 5.1-6 PLATE-TYPE ELECTROSTATIC PRECIPITATOR
-221-
-------
POWER RECOVERY (OPTIONAL) ' r~~
11
EMISSIONS
1
h
ELECTROSTATIC
PRECIPITATOR
CARBON MONOXIDE
BOILER
STACK
Fluid catalytic cracking unit regenerator with carbon monoxide boiler and electro-
static precipitator.
STEAM-*
LIGHT CYCLE OIL -*-
HEAVY CYCLE OIL -*-
BOTTOMS -*-
FEED —
FRACTIONATOR
ELECTROSTATIC STACK
PRECIPITATOR
onii CB ' >
™^»™^ DUIL>bl» I I
S^~~\ REGENERATOR>jl U
m /,—, 7\
X
REGENERATOR
REACTOR
•FUEL] |
DUST
AIR
-AIR
Petroleum refinery fluid catalytic cracking unit with control system.
FIGURE 5.1-7 FCCU PARTIOTJLATF, CONTROL SYSTEMS
SOURCE: (NA-029) -222-
-------
98%
FCCY REGENERATOR
FLUE GAS
ro
CENTRIFUGAL
SEPARATOR
2%
TURBO-EXPANDER
PARTICULATE
SEPARATOR
TO CO BOILER
FIGURE 5.1-8 SEPARATOR - ENERGY RECOVERY SYSTEM
-------
Catalyst t>?
Laden
Dust
*1
•H
H
^ts4
<=»
^
:-.i%-t£(
«.
i
<3>
c :
c>
I Clean Gas
l|
uv^a
v
^
spS
"*' '
^.
•
•*
- Fixed
Sand Bed
*
Collected <*ij
Catalyst
E
E
*l
'?•:•«
•
•
<2>
*
,
Blow Back
Gas
• •
U"-.
•' •.
v'
......
.*''
•>
*•
— Fluidized
Sand Bed
*"
Operating cycle.
Cleaning cycle.
Filter
Element
Regenerator
Flue Gas
Inlet
\
Blowluck
Collttctcd Gas Ports
Catiilyst
Fini-s
Outer
Screen
Granular
Sand
Bed
Filter element internals.
(Courtesy of Ducon Co., Inc.,
Mineola, N. Y.)
FIGURE 5.1-9 THE DUCON GRANULAR BED FILTER
SOURCE: (KA-149)
-224-
-------
5.1.2 SO.. Emission Control
^MB^^^H^B^^^B^B^^^^BO^-^^HW^^V^B^^^M^^^^^
Major sulfur oxide emission sources from the refinery,
LNG, and SNG modules include process heaters, boilers, the tail
gas treating plant, the CO boiler, the incinerator, and catalyst
regeneration processes. The Federal government has established
stringent ambient SO air quality levels (EL-062). Control of
X
SO is of vital importance in the design of a modern refinery,
A.
LNG plant, or SNG plant.
Control of sulfur oxide emissions can be accomplished
in the following ways:
design of processes to conserve energy,
use low sulfur fuels,
fuel desulfurization, and
removal sulfur products after combustion of fuel.
A combination of any or all of these techniques will reduce the
amount of SO emissions.
J\.
Conserving energy is an obvious solution, but until
the recent energy shortages had not been examined very closely.
Design of new energy-saving plants and processes in the
petroleum as well as all other industries is of prime importance
Use of low sulfur fuel is also an obvious solution
for reducing SO emissions. Natural gas is the cleanest fuel,
A.
but it is also in the shortest supply. Refineries, SNG plants,
and LNG plants are in the unique position, however, where they
either produce or process natural gas and thus have easy access
-225-
-------
to it. Although the price of natural gas is climbing, refineries
are tending to use natural or fuel gas produced from the refining
operations to fire their process heaters and thus reduce emissions.
5.1.2.1 Fuel Desulfurization
Sulfur or hydrogen sulfide are removed from fuels
before firing to reduce eventual SO emissions to the atmosphere.
5C
The two main fuels in the refinery, SNG plant, or LNG plant are
natural ga.s and heavy fuel oil. Natural gas as well as
refinery fuel gases such as coke-gas are usually treated by an
absorption process involving an aqueous, regenerative sorbent.
A number of gas treating processes are available, and they are
distinguished primarily by the regenerative sorbent employed.
Popular sorbents are amine-based solvents, hot carbonate solutions.
and various organic liquids such as N-methyl-pyrrolidone and
dimethyl ether of polyethylene gycol (HY-014). Sulfur compounds
may also be removed through adsorption processes by the use of
molecular sieves. These absorption and adsorption processes
have the additional advantage of removing C02 and thus increasing
the heating value of the fuel gas. A typical treated natural
gas stream has a H2S concentration of 0.25 grain per scf (HY-014).
Amine-based solvents are most commonly used (NG-002).
The solutions most often employed contain either monoethanolamine
(MEA), diethanolamine (DEA), or triethanolamine (TEA). The amine
sorbent used will depend on the properties of the sour gas.
The most common amine solution is a 10 to 2070 DEA solution.
A typical amine treating plant is shown in Figure 5.1-10.
The "sour" gas is contacted with the amine solution in an absorber
to remove H2S. The rich amine solution is pumped to a regenerator
where "acid"1 gas is removed from the amine sorbent.
-226-
-------
Purified Gas
Impure
Gas
Absorber
Solution
Cooler
Lean Solution
X
Acid Gas
Cooler
Steam
Rich Solution
Reactivate!- Reboiler
FIGURE 5.1-10 TYPICAL GAS TREATING PROCESS
UTILIZING AN AMINE SORBENT
-227-
-------
The "acid!1 gas is processed through a Glaus plant for
recovery of sulfur compounds as elemental sulfur. Conversion
efficiencies in the Glaus plant up to 987o can be attained but
will depend on the hydrogen sulfide concentrations in the acid
gas fed to the unit, the number of catalytic stages and the
quality of the catalyst used. When processing large volumes
of acid gas, however, the total SO emission from the Glaus unit
X
is large and requires further treatment by a tail gas treating
unit. Some of the tail gas units available are the following:
Beavon (Union Oil of California),
Cleanair (J. F. Pritchard and Co.),
IFF Process (Institut Francais du
Pe'trole),
' Shell Glaus Off-gas Treating - SCOT
(Shell Development Co.),
Sulfreen (SNPA/Lurgi; the R. M. Parsons Co.),
and
W-L SO., recovery (Wellman - Power Gas,
Inc.)
All of these units will increase tne Glaus recovery of equiva-
lent sulfur in the tail gas to greater than 99.5% (HY-014). The
use of different units is determined by the characteristics of
the tail gas, the operating conditions, and the economics of the
situation.
The other major fuel used in the refinery is heavy fuel
oil. Treating heavy fuel oil for sulfur removal is done by
hydrodesulfurization. Processes available for hydrodesulfurization
-228-
-------
are listed in Table 5.1-2 (HY-013). Residuals treating processes
which are listed usually produce a heavy fuel oil as one of the
products from hydrodesulfurization. Major problems arise
from inability to economically convert over 90% of the sulfur in
the residual to H S and from metals buildup in the reactor.
Heavy metals tend to poison or plug catalysts used in hydro-
desulfurizing residuals which means added expense due to frequent
replenishing of catalysts.
A new and rather novel method for treating residual
material in a refinery is flexicoking. The flexicoking process
was designed by Exxon Research and Development and is shown in
Figure 5.1-11. The advantage of flexicoking is that its major.
product streams are lighter than the fuel oil products from
resid hydrodesulfurization excluding a relatively small amount
of coke product. The product streams being lighter are more
efficiently treated for sulfur removal. Also, the heavy metals
in the feedstock are conviently concentrated in the coke material.
Lighter fuels such as gas oil or naphtha which are also
combusted in the refinery and SNG plant are readily hydrode-
sulfurized to a low sulfur concentration. The majority of
existing processes involve reaction of sulfur in the fuel with
hydrogen gas over a catalyst bed. Typical catalysts are cobalt-
molybdenum, nickel-molybdenum, and nickel-tungsten (HY-013).
The gas oil and naphtha sulfur content can be economically reduced
to 5 ppm or less (HY-013). Sulfur oxide emissions from the
fluidized catalytic cracker (and eventually the CO boiler) can
be reduced by desulfurizing the gas oil feed entering the unit.
Fixed bed catalysts such as those used for hydrode-
sulfurization in the refinery, the SNG plant, or the LNG plant
are usually replaced on an annual or biannual basis. In the past,
hydrogen sulfide and sulfur/dioxide emissions from regeneration
processes have been either uncontrolled or controlled by routing
-229-
-------
TABLE 5.1-2
HYDRODESULFURIZATION PROCESSES
Process Desulfurized
(Source) Petroleum Products
Description
% Sulfur
Reduction
Development
Status
Developer
H-oil
(HY-013, JI-008)
RCD Isomax
(WA-073)
GO-fining
(HY-013)
Resid-f ining
(HY-013)
Gulf-HDS
(HY-013)
KOS and VRDS
(HY-013)
Resid Hydro-,-
processing
(HY-013)
Residue De-
sulfurization
(GO-051)
IFP Hydrofin-
ing (AU-015)
Heavy gas oils
and residuals
Heavy fuel oils
Heavy fuel oils
Residuals
Residuals
Residuals
Residuals
Residuals
Residuals
Embullated bed reactor-vapor/
liquid system where the cata-
lyst is fluidized by an up-
ward flow of liquid
Fixed bed catalyst reactor
Fixed bed catalyst reactor
Fixed bed catalyst reactor
Fixed bed catalyst reactor
operated at high pressure
Fixed bed catalyst reactor
Fixed bed reactor with
highly selective catalyst
Fixed bed reactor with
highly selective catalyst
Fixed bed catalyst reactor
88-90%
80-93%
90%
60-90%
90%
'• 80-90%
80-90%
80-90%
NA
Full-scale
Full-scale
Full-scale
Full-scale
Full-scale
Full-scale
Pilot Plant
Full-scale
Pilot Plant
Cities Service Research
and Development Co.
and HRI
UOP Process Div. of
Universal Oil Products Co.
Exxon Research and
Engineering Co. and
Union Oil Co. of California
Exxon Research and
Engineering Co. and
Union Oil Co. of California
Gulf Research and
Development Co.
Chevron Research Co.
Standard Oil Co.
(Indiana)
BP Trading Ltd.
Institut Francais
du Petrole,'
U>
O
-------
SIMPLIFIED FLEXICOKING
FLOW PLAN
U>
REACTOR PRODUCTS
TO FRACTIONATOR
RECYCLE FEED
RESID
FEED
STEAM
GENERATION COOLING
FINES
REMOVAL
STEAM
COKE
GAS
SULFUR
REMOVAL
T
SULFUR
AIR BLOWER
Source: KE-128
FIGURE 5.1-11
-------
the regeneration gas to a furnace for incineration. With new
standards on ambient SO quality, however, additional treating
X
of the regeneration gas may be required before atmospheric emission,
5.1.2.2 Flue Gas Treating
Another potential technique for controlling sulfur
oxide emissions is by flue gas treating. Present research is
directed towards application of these methods to large coal or
oil-fired utility boilers, because they are a large emission
source of SO • The refinery environment, however, is a different
X
situation in that the heaters are either oil or natural gas-
fired and have a much smaller gas volume. The crude unit heater,
which is the largest process heater in the 200,000 bbl per day
refinery modules, is approximately equivalent to an 85 Mw power
plant while the smallest heater is about 1,000 times smaller.
A large number of processes are under development.
Several that are at an advanced development stage are listed in
Table 5.1-3 (EL-062). The processes are listed as throwaway or
recovery, "wet" or "dry". Throwaway processes are where the sulfur
is "thrown away" in some unmarketable sulfur form such as CaSO^,
while recovery processes recover the sulfur in a marketable form
such as sulfuric acid or elemental sulfur. The"wet" processes
are those in which an aqueous or liquid sorbent is used, while a
"dry" process uses a solid sorbent or activated carbon (char) bed.
Lime-limestone type scrubbing is the most advanced of
the sulfur oxide removal methods; however, it is doubtful that
this S0p removal process is the best choice for flue gas treatment
in a refinery. Waste disposal of the waste sludge is a difficult
.problem and, depending on the application, may be a significant
economic penalty. The most economical flue gas treating methods
for a refinery are likely to be the recovery methods. There are
-232-
-------
TABLES.1-3
LEADING COMMERCIALLY AVAILABLE PROCESSES FOR S02 REMOVAL FROM FLUE GASES
U>
Uo
I
Process Name
Wee Scrubbing
With Limestone
Wet Scrubbing
With Lime
Double Alkali
Magnesia Scrub-
bing (MGO)
Catalytic
Oxidation
Wellman-Lord/
SOz Reduction
Shell - CUO
Catalyst Re-
duction
Type of
Process
Wet
Throwaway
Wet
Throwaway
Wet
Throwaway
Wet
Recovery
Dry
Recovery
Wet
Recovery
Dry
Recovery
.
: Comments
Probably the least expensive process
to install.
Higher sorbent costs but increased
efficiency
Smaller scrubbers and liquor flows
possible.
Sulfuric acid produced at
central plant from MgSOs s.-l- ••
shipped from power plant. Re-
generated MgO xs returned 'to
the scrubbing system.
Catalytic oxidation occurs at 850°F.
producing 80% sulfuric acid
Wellman-Lord process produces con-
centrated S02 by thermal stripping
of NaHSOs. S02 may be reduced to S
with natural gas.
Advantages of dry scrubbing, re-
generated by hydrogen, low pressure
drop across absorber.
Sulfur
Reduction
>8.5%
>90%
>90%
90%
85%
90%
90%
Developer
Several Developers
Several Developers
General Motors,
Combustion Equip.
Assoc. , Inc. and
A. D. Little, Inc.
Chemico-Basics
Monsanto Enviro-
Chera Systems, Inc.
Wellman-Lord
Shell Development
Co.
Sources
EL-062
EP-009
SL-053
EL-062
SL-053
LA- 14 2
EL-062
SL-053
EL-C62
KO-133
MI -13 7
EL-062
PO-091
EL-062
PO-1C9
-------
definite reasons for this. First, many recovery processes require
HP or Hp S as reducing agents to regenerate the sorbent and pro-
duce sulfur. Availability of these gases within the refinery
(or SNG plant) makes recovery process economics much more favorable
Secondly, many recovery processes need Glaus plant capacity
and the refinery is normally already equipped with a Glaus plant.
This would mean design of a new refinery would only require a
larger Glaus unit to handle the acid gases from the flue gas
treaters. The added sulfur production can be credited as a
marketable product. Finally, if sulfuric acid is the final
product such as from the catalytic oxidation process, the
produced acid can be used in the alkylation process. If sulfuric
acid-type alkylation is used, however, one must then contend with
the problem of waste acid disposal which is another source of
SOV emissions.
2*.
One of the commercial SO removal processes being used
as a flue gas treater in a refinery is the Shell flue gas de-
sulfurization process (PO-109). This process uses cupric oxide
as a dry, selective adsorbent to remove the S0p . The regenera-
tion processes releases the sulfur in the form of SO^ which is
recovered as elemental sulfur in a Glaus unit. A flow diagram
of the Shell unit is shown in Figure 5.1-12.
Currently, much effort is being put into development
of "second generation" S0p scrubbing systems. Potentially
applicable S0o scrubbers for refinery heaters and for boilers
are shown in Table 5.1-4. All of these processes are currently
in the pilot-plant stage of development.
Incineration of oily sludge and biological waste is
also a source of SO in the refinery. Once again as mentioned
X
in the section of particulates control, the best solution is to
landfill the generated solids. However, if incineration is
deemed necessary, stack gas cleaning processes such as just dis-
cussed could be used effectively in reducing SO emissions.
X
-234-
-------
N>
Ui
I
TREATED
FLUE GAS
BYPASS
FLUE GAS
TO REACTORS
115,000 Nm3/hr
400*0
ABSORBER
OFF-GAS
EXCESS
STRIPPED
WATER
REGENERATION GAS
ACCEPTANCE TIME: 120 MIN.
SO, TO
GLAUS UNIT
REGENERATION
OFF-GAS. 400°C
FIGURE 5.1-12 PROCESS FLOW SCHEME OF THE SHELL FLUE GAS DESULFURIZATION
SYSTEM
-------
TABLE 5.1-4
POTENTIALLY FEASIBLE S0c SCRUBBING SYSTEMS
IN PILOT PLAHT DEVELOPMENT STA.JE
A. Dry Processes
1. Metal Oxides
B&W/Esso
Processes
Final Product
concentrated H^ SO , Sulfur
2. Carbon
a. Bergbau Forschung
b. Westvaco
concentrated H2S04
sulfur
B. Wet Processes
Alkali Absorbents (recovery)
a. Stone and Webster/Ionics - (Sodium
absorption-electrolytic regeneration)
b. Bureau of Mines Citrate (Sodium
citrate absoption-H2 regeneration)
c. Stauffer Chemical Co. Powerclaus
(Sodium phosphate absorption)
d. TVA Ammonium Bisulfate
e. Catalytic/Institute Francais
du Petrole (IFF) Ammonia
f. Consolidation Coal Potassium
Formate
weak
H2S04,
Sulfur
Sulfur
Sulfur
(NH4)2 S04
Sulfur
Sulfur
Sulfur
-236-
-------
5.1.3 NO.. Emission Control
X ""™ ""~
NO emissions in the petroleum refineries, the LNG
A\
processing plant, and the SNG processing plant modules result
from process heaters, boilers, the steam-naphtha reformer,
incinerators, and the CO boiler which is fueled by the fluidized
catalyst regenerator flue gas. The major factors determining
the amount of NO produced are the adiabatic flame temperature
X
within the process heater, the amount of oxygen available to
combine with the nitrogen, and the length of time which N2 and
Oz remain in high concentrations in high temperature regions
(BA-230).
Nitrogen oxide and dioxide can potentially be controlled
by various process techniques or modifications. The four gen-
eral categories describing these techniques or modifications are
the following:
1) combustion modifications to alleviate
conditions favorable to NO formation,
A.
2) fuel modifications, by denitrification,
use of additives or the substitution of
low NO forming fuels,
X
3) new or alternate designs of low NO
X
forming processes, and
4) treatment of flue gases for NO removal.
X
All four of these control techniques will be reviewed in the
following sections. One must, however, be careful in applying
the developing technology to process heaters or small boilers
since most current developments are for large utility boilers.
The applicability to process heaters and small industrial boilers
is largely undefined (BR-199).
-2.37-
-------
5.1.3.1 Cqmbu s ti oji Modifica tjLon s
One combustion modification for reducing NO involves
X
reducing the excess air used in firing the heater (JA-056).
This method, however, presents a problem for small process
heaters in that regulation of air flow may be quite difficult
due to the varying composition and heat content of the fuel
gas and fuel oil. Generally, process heaters are fired with a
large excess of air to compensate for these fluctuations. This
method actually does help reduce NO emissions in that the
X
temperature within the heater is reduced due- to the "cooling"
effect of the excess air. The reduced temperature reduces
NO formations.
A second method for NO reduction is two-stage com-
X
bustion. In two-stage combustion, some burners are operated
"fuel-rich". This is to say that the fuel is combusted with an
air flow supplying only part of the stoichiometric amount of
oxygen required for complete combustion. Other burners are run
"air-rich" or on air only. The NO emissions reduction may be
explained by the following factors:
1) there is a lack of oxygen available
for NO formation in the "fuel-rich"
burners,
2) the flame temperature is lower in the
"fuel-rich" burner,
3) heat removed between the two stages
will cause a decrease in the maximum
flame temperature, and
-238-
-------
4) the effective residence time for
NO formation at the
X
is reduced (JA-056).
NO formation at the peak temperature
X
Two-stage combustion may, however, not be applicable to small
process heaters. Excellent combustion conditions are needed
for two-stage combustion to work adequately. If conditions are
not favorable, flames will impinge on furnace tubes causing hot
spots and caking (NA-005).
Another combustion technique for reducing NO emissions
J\,
involves reducing the load on the heater and thus reducing the fire box
temperature in the heater (JA-056). With a decreased fire box
temperature, the oxygen and nitrogen will have less residence
time in a high temperature environment. A disadvantage of this
technique is the requirement for greater amount of fuel for an
equivalent heat load.
A final combustion modification for reducing NO emis-
X
sions is flue gas recirculation (JA-056). A portion of the
gases are recycled back into the combustion chamber in front of
the flame. The flue gas can be recycled either by free or forced
draft systems. The recycled flue gases have two effects:
1) the temperature of the flame zone is
reduced by the cool gases, and
2) the concentration of oxygen available
for NO formation is reduced (JA-056).
X
-239-.
-------
A water or steam injection technique has been described
for large gas-fired burners for the control of NO (JA-056). The
X
injection of water or steam is probably unfeasible for process
heaters due to problems involved with increased corrosion and
decreased efficiency.
5.1.3.2 Fuel Modifications
The formation of NO can result from either the fixa-
A.
tion of atmospheric N2 or the converson of fuel-bound nitrogen
or both. Removal of nitrogen from air before combustion is
impractical. Nitrogen which is held in liquid fuels, however,
can be removed to various degrees by hydrogenation, which is
usually done concurrently with hydrodesulfurization.
Additives such as metal oxides have some possibility
to catalytically reduce or decompose NO to Np. The emission
reduction from this strategy is quit*3, limited and the cost
effectiveness is likely to be poor (BR-199).
An obvious solution to the reduction of NO emission
X
would be to fire fuels containing smaller amounts of nitrogen.
Natural gas contains less nitrogen per equivalent of energy
than heavy fuel oil. The use of natural gas will, however, be
determined by its availability. Low BTU gases such as the coke-
gas are also expected to lower NO emissions due to reduced
X
flame temperatures characteristics of the lower heating value
fuels (BR-199).
3.1.3.3 Design Modifications
Design modifications of burners are also used to
reduce NO emissions. Narrower spray angles which produce a
X
low degree of atomization of the fuel have been found to give
-240-
-------
lower NO emissions (BA-003). Turbulence has also been found
A.
to affect NO production through entrainment of cooler gases.
X
A burner which produces a long "lazy" flame has been found to
produce less NO than an intense, short flame (BA-003).
X
Proper burner location and spacing also helps to re-
duce NO emissions. Burner arrangements which lower the flame
X
temperature and radiate heat more easily are the most advanta-
geous. Modifications such as tangential firing of burners do
not allow flames to interact and thus lower the flame tempera-
ture. Tangential firing also enhances the radiation of heat
as compared with front or opposite "fired furnaces (BA-QQ3).
5.1.3.4 Flue Gas Treating
Flue gas treating is still very much in the develop-
ment stage. Combustion flue gas treating can be evaluated in
the following general categories:
1) Catalytic decomposition,
2) Catalytic reduction,
(a) Non-selective
(b) Selective
3) Adsorption/reaction by solids, and
4) Absorption/reaction by liquids.
Denitrification of flue gas of oil-fired boilers or
heaters is different from those burning fuel gas because:
-241-
-------
Catalysts which are used must be
resistant to greater sulfur oxide
and heavy metals concentrations for
oil-fired heaters (OK-012),
The particulates in flue gases from
oil-fired units must be removed
before entering and clogging the
catalyst beds (OK-012), and
Nitrogen oxide concentrations are
generally higher for oil-fired heaters
as compared to fuel gas-fired heaters
(NA-005).
The four categories of flue gas treating processes mentioned
above will be briefly described in the following sections.
Catalytic Decomposition
The decomposition of nitric oxide is thermodynamically
favored. However, catalysts have been ineffective in enhancing
the reaction rate. Little research has been conducted with
concentrations of NO as low as those found in the flue gases
(BA-003).
Catalytic Reduction
Nonselective reduction of NO occurs when the reducing
X
agent such as H2 or CH4 reacts with other easily reduced com-
pounds before reacting with the NO . A noble metal catalyst
X
-242-
-------
has been used to enhance the reaction rate of reduction. Noble
metal catalysts would not be adaptable to flue gases because of
sulfur poisoning. Experiments by Ryason and Harkins,
however, show the reduction of both NO and SOa will occur in
the presence of a copper on aluminum catalyst (BA-003) . The
major problem with this method is that the excess air must be
finely controlled to allow little or no free 02 in the flue gas.
Environics presently have a noble metal catalytic reduction
process on a gas-fired boiler in Los Angeles.
Selective reduction is much preferred to nonselective
reduction in that it will allow the normal excess air used in
firing a heater or boiler. In such a process, the reducing agent
would be added to the flue gas in controlled amounts to selectively
react with the NO to form N and Hp 0 over a catalyst. Hydrogen,
ammonia, carbon monoxide, and hydrogen sulfide have been selected
as possible selective reductants. A possible flow scheme for
Ha or NH selective reduction is shown in Figure 5.1-13. Ammonia
3
reduction and HaS or Ha reduction have the added option or
being also an SO removal system. The 90% removal of NO by
X X
ammonia reduction from an exhaust gas from a methanol process
reformer with an NO concentration 200 ppm has been successfully
X
accomplished on a commercial level by Japan's Sumitomo Chemical
Company and Japan Gasoline Company (OK-012). The removal of NO
by reaction with hydrogen sulfide and Hp is presently in the
experimental stages.
Adsorption/Reaction by Solids
A number of solid materials have been cited as adsor-
bents for NO . These materials include such common adsorbents
H
as silica gel, alumina, char, and molecular sieves where ad-
sorption is due primarily to physical forces, and metal oxides,
-243-
-------
B
0
I
L
E
R
REDUCTANT
(H OR
PRECIPITATOR
FLY ASH
CATALYTIC
REDUCER
AIR
PREHEATER
AIR
ID FAN
FIGURE 5.1-13 CATALYTIC H OR NH REDUCTION PROCESS
Source: (BA-003)
-244-
-------
ion exchange resins, and hydroxides where attraction is prin-
cipally chemical in nature (BA-003). An adsorbent containing
5 percent NaC&02 and 10 to 15 percent Na2C03 supported on acti-
vated alumina can remove 100 percent of NO in gas streams
X
containing low concentrations of NO . However, NO emission
x x
control by this method has been described as impractical since
make-up costs for the oxidant are high (BA-003). Of the metal
oxides, manganese and alkalized ferric oxides show the greatest
technical potential. Problems that occur with solid systems
are agglomeration of the fluidized bed due to molten nitrate
salts in the regeneration process, attrition losses which
require flue gas clean-up, and catalyst losses (BA-003).
Absorption/Reaction by Liquids
Aqueous absorption system appears to offer potential
for combined NO and SO emission control but few results of
X X
development work have been published. Sulfuric acid scrubbing
of NO has been investigated for large utility boilers by Tyco
A.
Laboratories (BA-003). The sulfuric acid process, as described
by Tyco, is shown in Figure 5.1-14. One significant disadvantage of
the sulfuric acid scrubber is that in order to remove a large
percentage of NO from the original flue gas, the process must
run at very high efficiency. The efficiency, however, is
decreased by an increase in S02 concentration in the flue gas
and by the presence of NO in the NO2 recycle stream. Chiyoda
Chemical Engineering Company has further developed the sulfuric
acid scrubber to include a second absorption step where the NO
from an oxidation step which also oxidizes SO to sulfuric acid
3
is absorbed in an alkaline solution. Removal efficiencies for
NOX are claimed to range up to 95% (OG-010).
-245-
-------
Cleaned
Flue Gas
To Stack
ID Fan
so2
Oxidizer
Scrubber
Flue
Gas
NO.
\\x\\\
HN03
Reactor
HN03
Product
V
Fly
Ash
Cyclone
Nil I
NOHSO.
in
Sulfuric
Acid
Catalytic
Decomposer/
Oxidizer
H2S04
Product
FIGURE 5.1-14 TYCO'S ISOTHERMAL SULFURIC ACID SCRUBBING SYSTEM
Source: (BA-003)
-246-
-------
5.1.3.5 NO Emissions From Incinerators
NO emissions from incinerators can be reduced by not
X
incinerating at all. Landfill is the best alternative and has
been described in the section on particulates removal. If the
material is to be incinerated, a low firebox temperature and a
low percentage of excess air are required to reduce NO . Some
4\.
incinerators are fired with methane or fuel oil along with the
incinerable sludge. This combination of combustion materials
causes a substantial increase in the NO emission and should
X
be avoided. NO emissions may also be decreased by recovering
X
heat from the fire box by methods such as generating steam. By
controlling the heat removed, the NO emissions can be controlled.
X
Incinerators that reduce NO by changing combustion conditions
1\.
are not available except on a developmental basis.
-247-
-------
5.1.4 CO Emission Control
Atmospheric CO emissions from the refinery, LNG plant,
and SNG plant come from process heaters, boilers, the CO
boiler and the incinerator. CO emissions from process heaters
and boilers can be controlled by (NA-004):
Complete combustion
Energy conservation
Energy source substitution
Gas cleaning
Collection and flaring of miscellaneous
CO emissions
Use of all of these methods will reduce CO emissions greatly
in a refinery, LNG plant, or SNG plant.
Complete combustion is the best method for controlling CO
emissions. Good practice includes proper design, application,
installation, operation, and maintenance of process heaters
and burners and auxiliary systems. Guides for good practices
have been published by the fuel industry, equipment manufacturers,
engineering associations, and government agencies (NA-004).
Probably the greatest factor in CO formation is the
amount of excess air used in combusting fuels. A low excess
air rate will result in incomplete oxidation of the carbon
and the formation of CO. One must, however, be aware that a
high excess air rate can have an effect of increasing CO produc-
tion, also. With high excess air rates, the air cools the flame
temperature which results in incomplete combustion.
-248-
-------
Carbon monoxide emissions can be minimized by design-
ing for (1) a high combustion temperature, (2) intimate contact
among fuel, oxygen, and combustion gases, (3) sufficient re-
action time, and (4) low effluent temperature (NA-004). Firing
in excess of the design conditions is perhaps the greatest cause
of excessive CO emissions from stationary sources. Proper choice
of burners and good housekeeping practices as to cleaning
burners will result in low CO emissions and better control of
process heaters. Various types of burners and possible defects
which will produce CO are shown in Table 5.1-5. Proper design
of flue gas vent area is also highly desirable in that it allows
for smooth flow of gases through the furnace with no back pressure
to impede the air flow.
Energy conservation is an obvious method for reducing
CO emissions, along with every other emission involved with
combustion. Techniques such as preheating the furnace air and
preheating the process side material result in lower loads on
the heaters .or furnaces and thusly, lower fuels combustion.
Combustion control systems are also considered a source of
energy conservation in that the most complete combustion will
result in the greatest amount of energy released per unit mass
of fuel.
Energy source substitution is also used to reduce CO
emissions. The burning of natural gas will give about 35 times
less CO per an equivalent of energy released as compared to fuel
oil (NA-004). The availability of natural gas will, however,
determine if it can be substituted for fuel oil. Also, if
available, hydroelectric energy or nuclear energy as substitute
for fossil fuels will reduce the total amount of CO emitted to
the atmosphere.
-249-
-------
TABLE 5.1-5
CLASSIFICATION OF OIL BURNERS ACCORDING TO APPLICATION AND
LIST OF POSSIBLE DEFECTS
Burner type
Commercial, Industrial
Pressure atomizing
Horizontal rotary
cup
Steam atomizing
Air atomizing
Applications
Steam boilers,
process furnaces
Steam boilers,
process furnaces
Steam boilers,
process furnaces
Steam boilers,
process furnaces
Oil type
usually used
No. 4, 5
No. 4,5,6
No. 5, 6
No. 5
Defects that cause excessive CO emissions
Oil preheat too low or too high, nozzle wear,
nozzle partly clogged, impaired air supply,
clogged flue gas passages, poor draft, overload-
ing
Oil preheat too low or too high, burner partly
clogged or dirty, impaired air supply, clogged
flue gas passages, poor draft, overloading
Oil preheat too low or too high, burner partly
clogged or dirty, impaired air supply, clogged
flue gas passages, poor draft, overloading, insuf-
ficient atomizing pressure
Oil preheat too low or too high, burner partly
clogged or dirty, impaired air supply, clogged
flue gas passages, poor draft, overloading, insuf-
ficient atomizing pressure
SOURCE: (NA-004)
-250-
-------
Gas cleaning is also a method for reducing CO emissions.
For the refinery, LNG plant, and SNG plant, the only gas cleaning
units are the shift converters on the hydrogen plant in the
gasoline refinery and the SNG plant, and the CO boiler in the
fluidized catalytic cracking unit.
The hydrogen plant is a steam-naphtha reforming process.
The hydrogen synthesis reaction in the reforming process is the
following (VO-025):
CnHm + nH20 •*• nCO +
The CO produced is eliminated from the synthesis gas by a
multiple shift conversion followed by a methanation of the
residual carbon monoxide. The reactions for the carbon mon-
oxide shift (1) and the methanation (2) are the following:
(1) CO + H20 -> 002 + H2
(2) CO + 3H2 + CHi, + H20
The final carbon dioxide produced is removed by absorption with
MEA. Other C0a removal processes available include Giammarco-
Vetrocoke, Benfield, Catacarb and Sulfinol.
The CO in the off-gas from the FCCU regenerator is
removed in a CO boiler. The CO is burned to recover its heating
value. In most cases supplementary fuel such as fuel oil is
burned in the boiler to keep the combustion temperature at approxi-
mately 1800 F and to supply sufficient energy to produce steam.
The average emission of CO from a FCCU regenerator is 13,700
pounds of CO per 1,000 barrels of fresh feed (NA-004). The CO
boiler will reduce the carbon monoxide down to 100 to 150 ppm
in the final effluent regenerator gas. A typical carbon monoxide
waste-heat boiler is shown in Figure 5.1-15.
-251-
-------
FIGURE 5.1-15 WATER-COOLED, CARBON MONOXIDE WASTE-HEAT BOILER
SOJRCE: (DA-069)
-252-
-------
5.1.5 Hydrocarbon Emissions Control
Hydrocarbon emission sources in the refinery, the LNG
plant, and the SNG plant are many and quite varied. Some of the
major refinery sources are storage tanks, fluidized catalytic
cracking units, boilers and process heaters, the blowdown system,
process drains, vacuum jets, and cooling towers. Hydrocarbon
emissions from the LNG module are quite small and mainly due to
gas-fired process heaters and boilers. Any storage of LNG on-
site is in pressurized vessels which emit negligible amounts of
hydrocarbons. The SNG module has hydrocarbon emissions from
boilers and process heaters, storage, and the regenerator for the
hot carbonate solution used for CO absorption. As compared
to the storage and regeneration process, the boiler and heaters
give off negligible hydrocarbons. The following sections
describe methods for controlling hydrocarbon emissions from
these sources.
5.1.5.1 Carbonate Vent Gas
The hot carbonate system used in the SNG module
(Benfield process) is used to reduce the CO content in the final
methane-fuel gas in order to increase the heating value of the
gas. The regeneration of the potassium carbonate absorbent
produces a large volume of C0a gas which has a typical hydro-
carbon content of 0.027 wt percent (LO-095). Although the con-
centrations are very small, the total weight of hydrocarbon emitted
is large because of the large volume of CO vented to the atmo-
sphere. The other C0g removal processes for upgrading synthesis
gas use basically the same principle of high pressure (approxi-
mately 1,000 psig) absorption of the COs (and Hp S) and removal
from the absorbent by a depressurized regeneration process and
thus are expected to emit equivalent amounts of hydrocarbons.
-253-
-------
A possible control method for hydrocarbon emissions is
to route the off-gas to a boiler or process heater to combust
the hydrocarbons. One would gain the heat of combustion of the
hydrocarbons, but would lose the energy which is required in
sensible heating of the inert off-gas (mainly C0p). Another
possible control method is by first partially flashing off-gas
from the rich absorbent solution and recycling this gas back
through the absorber. The flashed gas is expected to contain
an appreciably higher concentration of hydrocarbons than the
directly vented regenerator gas, because hydrocarbons have a
much lower solubility in the absorbent solution than the CO,,
or HpS.
5.1.5.2 Storage Control
A major hydrocarbon emission source in the petroleum
industry is tankage. Storage emissions depend on diurnal temp-
erature and pressure changes, filling operations, volatilization,
solar radiation, and mechanical condition of the tanks.
Proper design of storage tanks will control hydro-
carbon emissions greatly. There are five basic types of stor-
age tanks used in the petroleum industry. These are: fixed
roof, floating roof, internal floating cover, variable space,
and pressure. The applicability of these tanks largely depends
on the volatility of the stored liquid. Table 5,1-6 shows the
type of tank generally used for storing certain volatile petro-
leum products (EN-043, MS-001).
-254-
-------
TABLE 5.1-6
NATURE OF PRODUCT STORAGE AT REFINERIES
Product
Fuel Gas
Propane
Butane
Motor Gasoline
True Vapor
Pressure
psia @ 60°F
105
26
4-6
Aviation Gasoline 2.5-3
Jet Naphtha 1.1
Jet Kerosene <0.1
Kerosene <0.1
No. 2 Distillate <0.1
No. 6 Residual <0.1
Crude Oil 2
Types of
Storage Tanks
Cryogenic - Pressurized
Pressurized
Pressurized
Variable Space, Fixed
Roof, Floating Roof
Variable Space, Fixed
Roof, Floating Roof
Variable Space, Fixed
Roof, Floating Roof
Fixed Roof
Fixed Roof
Fixed Roof
Fixed Roof
Variable Space, Fixed
Roof, Floating Roof
Qty. Stored
1968
(106 bbl)
204
14
18
31
46
346
137
-255-
-------
From a hydrocarbon emissions standpoint, any storage
facility which is flexible enough to retain all vapors emitted
from the stored hydrocarbon at an economically feasible level
is the system most desirable. The fixed roof tank is the only
tank mentioned above which is not designed for containing vapors
emitted. Usually, in a fixed roof tank, the vapors are vented
to the atmosphere through a pressure-vacuum vent. The variable
space-type storage tank is designed to control normal diurnal
breathing losses and small filling operations. Large changes
in the vapor content can not be handled by the variable space
tank alone and thus vapors will have to be vented to the atmosphere,
Pressure vessels are used to hold highly volatile petroleum
products under pressure. The shape of the pressure vessel
depends on the pressure required. Spheres can be operated at
pressures up to 217 psi; spheroids, up to 50 psi; noded spheroids,
up to 20 psi; and plain or noded hemispheroids, up to 75 psi
and 1\ psi, respectively (DA-069).
Floating Roof Tanks
The floating roof tank is the most commonly used tank
for controlling hydrocarbon emissions. Modern designs include
pontoon deck floating roofs, double-deck floating roofs, and
trussed-pan floating roofs. The major concerns in design of
the roof are structural support and reduction of heat conduction
due to solar radiation. The floating roof is constructed about
eight inches shorter in diameter than the inside tank diameter.
-256-
-------
The space between is usually sealed by vertical shoes (metal
plates) connected by braces to the floating roof. A fabric
seal is also included to reduce hydrocarbon emissions. The
space between the roof and the wall is the greatest source of
emissions from a floating roof tank. Frequently an additional
secondary seal is added to act as a wiper reducing the wicking
action associated with floating roof tanks (NA-032). Another
sealing device is a flexible tube which floats on the liquid
surface and keeps contact between the roof and the tank wall.
Different tube seals are shown in Figure 5.1-16. Floating roof
tanks are about 91 percent efficient in controlling hydrocarbon
emissions from gasoline storage (EN-071).
Fixed Roof Tanks
In order to limit emissions from fixed roof tanks,
floating covers have been designed. One type is a floating
plastic blanket. The blanket is constructed with plastic
floats underneath and custom manufactured so that only a one-
inch gap remains around the periphery (DA-069). A skirt is
placed above this gap to further eliminate fugitive vapors.
Another new technique in sealing fixed roof tanks is a floating
microsphere blanket. The microspheres are made of plastic resins
less dense than the liquid petroleum product. The microspheres
entirely cover the liquid surface and their fluidity gives the
added advantage of being able to flow around internal tank parts.
Vapor Recovery System
With present day prices of petroleum products increasing,
more sophisticated systems for recovering hydrocarbon vapors are
becoming economically feasible. One system employed is an in-
tegrated vapor recovery system. The vapor recovery system is a
closed system which is set up to recover hydrocarbon vapors
-257-
-------
-R!M BAND
NORMAL
HROCUCT LEVEL
TANK SHELL.
ADAPTABLE
SEAL SUPPORT
TOP DECK
BOTTOM DECX
WEATHER SHIELD-
CURTAIN-
INFLATED
SEAL TUBE
TANK SHEll-
itfj—
Ik
-HANGER BAR
-SUPPORT RING
TO? OEO;
EOTTC.v.
V/EATHER SHIELD —
CUKtAIN SEAL r
SEAL ENVHOI'E
SEAI SUPPORT ,
RING
RESILIENT
UfrETHANt: tOAM
TOP DECK
FIGURE 5.1-16
SEALING DEVICES FOR
FLOATING-ROOF TANKS:
(UPPER LEFT) LIQUID-
FILLED TUBE SEAL,
(UPPER RIGHT) IN-
FLATED TUBE SEAL,
(LOWER LEFT) FOAM-
FILLED TUBE SEAL
(CHICAGO BRIDGE AND
IRON CO., CHICAGO,
ILL.)
Source: (DA-069)
-258-
-------
emitted from storage facilities and also the loading facilities.
A typical vapor recovery system including a vapor saver is shown
in Figure 5.1-17. The variable space tank included is designed
to control breathing losses and small vapor changes within the
system. The vapor recovery unit which handles large vapor volume
changes such as during loading operations or periods of drastic
ambient temperature or pressure changes includes a compressor-
refrigeration system. Vapor recovery units can liquefy hydro-
carbon vapors by several principles which include compression-
refrigeration, absorption, and adsorption. They also can employ
a combination of these principles. The efficiency of vapor
recovery units typically ranges from 907o to 957o, depending upon
the composition and concentration of the hydrocarbon vapors
processed (EN-071). Vapor recovery units are manifolded into
the vapor collection systems of tankage and loading operations
for the reliquefaction of hydrocarbon vapors into product.
Vapor recovery systems are quite expensive as compared to
floating roof tanks, and only give a little greater efficiency
in recovering vapors.
Another possible control technique is maintaining
wet scrubbers or condensers on the vents of fixed roof tanks.
The wet scrubbers can be bubble-cap tray towers, packed towers,
spray towers, or Venturi scrubbers. These types of scrubbers
are shown in Figure 5.1-18. The common absorbents for organic
vapors are water, mineral oil, nonvolatile hydrocarbon oils,
and aqueous solutions (e.g., solutions of oxidizing agents,
sodium carbonate, or sodium hydroxide). If water is used, the
hydrocarbon rich water is sent to a closed waste water stream
to be treated at the waste water treatment facility. The other
absorbents can be regenerated, with the collected hydrocarbons
flared or recycled, depending on the amounts recovered.
-259-
-------
VAPOR RECOVERY
PROCESSING UNIT
-------
SHELL"
TRAY
DOWNSPOUT
TRAY
SUPPORT
RING
TRAY —f
STIFFENER
VAPOR
RISER
FROTH
^•LIQUID IN
BUBBLE CAP
.. SIDESTREAM
^"WITHDRAWAL
^.INTERMEDIATE
FEED
-GAS IN
LIQUID OUT
Schematic diagram of a bubble-cap tray tower.
LIQUID— fr=n-N
nlw
PACKING
RE STRAINER
LIQUID
RE-DISTRIBUTOR
PACKING
SUPPORT
GAS IN
-LIQUID
OUT
Packed tower.
EXHAUST
GAS
INLET
CLEAN
GAS
OUTLET
_Ji
LIQUID
SPRAY
MOISTURE
•ELIMINATORS
=> LIQUID
ABSORBENT
INLET
ABSORBENT.
CONTAMINANT
SOLUTION
OUTLET
CLEAN GAS
OUTLET
LIQUID
ABSORBENT INLET
EXHAUST
GAS •
INLET
ENTRAINMENT-*
SEPARATOR
ABSORBENT.
CONTAMINANT
SOLUTION
OUTLET
Venturi scrubber.
Spray tower.
FIGURE 5.1-18 HYDROCARBON VAPOR SCRUBBERS
SOURCE: (NA-032)
-261-
-------
Activated carbon is an adsorbent that can be used for
hydrocarbon emissions. The adsorbed hydrocarbons are removed
from the carbon by steam stripping and then recovered by decanta-
tion or distillation. Costs of activated carbon adsorbers are
high, but the recovery of valuable hydrocarbons enhances the
feasibility of the operation.
Storage Tank Maintenance
Heat from solar radiation causes problems by increas-
ing hydrocarbon boil-off. Painting tanks and proper tank design
reduce the radiation effect. Paints are chosen as to those
that best reflect solar radiation. Table 5.1-7 lists the ef-
fectiveness of various paints on reflecting heat. Proper tank
design includes a double-deck pontoon-type floating roof or
trussed floating roof to avoid direct warm metal-liquid contact.
Tank diameter also effects the amount of hydrocarbon emitted.
A smaller tank diameter will have less emissions (DA-069).
Proper maintenance practices help to eliminate hydro-
carbon emissions. Particular trouble spots are leaky and poorly
regulated vents on fixed roof tanks and leaky seals on floating
roof tanks. Maintaining properly painted tanks helps in elimi-
nating emissions. Proper scheduling such as pumping liquids into
storage tanks during cool hours and withdrawing liquids at hotter
times and maintaining short periods between pumping operations
should be followed.
"•---'•• - Loading Rack_Con_tr^)lj3
Hydrocarbon emissions from transport loading operations
are generally controlled by the use of a vapor collection device
manifolded into a vapor recovery unit (see Figure 5.1-17). The
transport vehicle may be a tank truck, rail car, barge, or
marine vessel.
-------
TABLE 5.1-7
RELATIVE EFFECTIVENESS OF
PAINTS IN KEEPING TANKS FROM WARMING IN
THE SUN (Nelson, 1953)
Color
Black
No paint
Red (bright)
Red (dark)
Green (dark)
Red"
Aluminum (weathered)
Green (dark chrome)
Green
Blue
Gray
Blue (dark Prussian)
Yellow
Gray (light)
Aluminum
Tan
Aluminum (new)
Red iroji oxide
Creamier pale blue
Green (light)
Gray (glossy)
Blue (light)
Pink (light)
Cream (light)
White
Tin plate
Mirror or sun shaded
Relative effectiveness
as reflector or
rejector of heat, %
0
10.0
17.2
21.3
21.3
27.6
35.5
40.4
40.8
45.5
47.0
49.5
56.5
57.0
59.2
64.5
67.0
69.5
72.8
78.5
81.0
85.0
86.5
88.5
90.0
97.5
100.0
-263-
-------
The type of vapor collection system installed depends
on how the transport vehicle is loaded. If the unit is top
loaded, vapors are recovered through a top loading arm (Figure
5.1-19). Product is loaded through a central channel in the
nozzle. Displaced vapors from the compartment being loaded flow
into an annular vapor space surrounding the central channel and
in turn flow into a hose leading to a vapor recovery system.
If the transport is bottom loaded, the equipment
needed to recover the vapor is considerably less complicated.
Vapor and liquid lines are independent of each other with
resultant simplification of design. Figure 5.1-20 shows a typical
installation. Product is dispensed into the bottom of the trans-
port and displaced vapors are collected from the tank vents and
returned to a vapor recovery unit.
Bottom loading vapor recovery has many advantages over
top loading vapor recovery. Bottom loading generates much less
vapor, generates almost no mist and is safer from a static elec-
tricity point of view.
The vapor collection efficiency of loading controls is
in excess of 95 percent. However, the overall emission reduc-
tion is also dependent on the efficiency of the vapor recovery
unit. A 90 percent efficient vapor recovery unit would make a
loading control system 85 percent efficient.
-264-
-------
MISCELLANEOUS PARTS
ITEM
1
2
3
4
5
6
7
B
9
10
11
PART NO.
3A20-F-30
2775*
3420-F-40
H-5936
0-S37-M
H-5898-RP
H-S906-M
H-i905-M
H-S318-
C-16G7-A
C-2479^
DESCRIPTION
Swivel Joint, 3"
Boom
Swivel Joint, 4"
Swivel Joint 3"
Handle
Hose
Elbow
Cofd Grip
Collat Sub-Assembly
Link
Gasket
QTY.
2
2
2
1
ITEM
12
13
14
IS
16
17
18
PART NO.
H-4190-M
0-836-M
3630-30
H-4189-M
H-5952
3840-FO-40
710
C-5S5-A
417-FKA-4"
3476-F-40
DESCRIPTION
Gasket, 4"
Upper hbndle& Pip«
Swivel Joint. 3"
Gasket, 3"
Swivel Joint Sub-Assembly, 4"
Swivel Joint Only
4x2 7,6 Nipple Only
4" Flange Only
Loading Valve
Swivel Joint, 4"
QTY.
6
FIGURE 5.1-19 TOP LOADING ARM EQUIPPED WITH A VAPOR RECOVERY NOZZLE
-265-
-------
FIGURE 5.1-20 BOTTOM LOADING VAPOR RECOVERY SYSTEM
-266-
-------
5.1.5.4 Combustion Source Controls
Hydrocarbon emissions from process heaters and steam
boilers can be minimized by adjusting the fuel to air ratio for
optimum fuel combustion. To insure optimum combustion conditions
are maintained, some refineries have installed oxygen analyzers
and smoke alarms on heater and boiler stacks (WA-086). For
process heaters that do emit large quantities of hydrocarbons,
catalytic afterburners can be used. The use of the afterburners,
however, requires additional fuel for complete combustion of
the hydrocarbons. The catalyst allows the combustion to take
place at a lower temperature (NA-032). A catalytic afterburner
is shown in Figure 5.1-21.
Internal combustion engines used to drive older com-
pressors have inherently high hydrocarbon emissions. The major
means of controlling hydrocarbon emissions from this source is
by carburetion adjustments similar to those applied to automobile
engines for emission control. Economic considerations coupled
with increased concern for emission reductions is inducing
refineries to phase out the use of internal combustion engines.
5.1.5.5 Incinerators
The best possible way to eliminate hydrocarbon emis-
sions from sludge incineration is by disposing of the material
by landfill. This operation is described in the section on
Solids Emissions Control.
If the sludge is incinerated, auxiliary burners are
used for secondary combustion of the wastes. The auxiliary
burners will increase the incineration temperature in certain
locations to promote the reduction of hydrocarbons. Good con-
tact between the combustible material and air will also reduce
hydrocarbon emission. Good contact is promoted by baffling of the
physical mixing apparatus, introducing air at strategic locations,
and locating auxiliary burners to promote mixing (NA-032).
-267-
-------
OXIDIZED GAS OUTLET
CATALYST
ELEMENTS
BLOWER
FUME ENTRY
FROM OVEN
RS ^-PREHEAT
BURNER
FIGURE 5.1-21 CATALYTIC AFTERBURNER
-268-
-------
As an indirect method of reducing hydrocarbons, the
heat from incineration can be recovered and used to produce
steam. The heat load recovered would mean a reduction in fuel
needed to produce steam within the refinery where the sludge
incinerator is located.
5.1.5.6 Process Source Controls
Catalytic Cracker Regenerators
There are two major control measures applicable to the
reduction of hydrocarbon emissions in the flue gas of catalytic
cracker regenerators. The first of these is incineration in a
carbon monoxide waste-heat boiler. By incinerating regenerator
flue gas in CO waste-heat boilers, the hydrocarbon emissions
are reduced to a neglibible amount and valuable thermal energy
is recovered from the flue gas.
A second control measure applicable to the flue gas from
moving bed catalytic cracker regenerators as well as the flue gas
from regenerating operations for other catalysts is incinera-
tion in a heater fire box or smoke plume burner. These regen-
erators produce significantly less flue gas than FCC regenera-
tors and may not justify a CO boiler. Catalysts in this category
may include reformer, isomerization, and hydrocracking catalysts.
Hydrocarbon emissions in regenerator flue gas are reduced to
negligible quantities by incineration in heater fire-boxes and
smoke plume burners.
Although neither CO boilers nor other forms of regen-
erator flue gas incineration are extensively used today, they
are becoming standard equipment in new refineries and expansions
of existing units. This is a result of both energy conservation
and increased concern for air quality.
-269-
-------
Vacuum Jets - Barometric Condensers
Hydrocarbon emissions from barometric condensers on
vacuum jets are attributable to both the venting of non-conden-
sable hydrocarbons as well as to the evaporation of hydrocarbons
from the oily barometric condensates.
Three measures for minimizing oily condensate genera-
tion are mechanical vacuum pumps, lean oil absorption, and
surface condensers. While mechanical vacuum pumps have little
effect on the quantity of non-condensable hydrocarbons generated,
they do eliminate the generation of oily steam condensate. The
insertion of a lean oil absorption unit between the vacuum
tower and the first stage vacuum jet helps to minimize the
quantities of both non-condensables and oily condensate (AM-055).
The rich oil effluent is reused as charge stock and not regen-
erated. Surface condensers in place of barometric condensers
minimize oily condensates but have little effect on the quantity
of non-condensables (AT-040) .
Because there are no means to completely eliminate the
generation of non-condensable vapors from vacuum pumps or steam
ejectors, these emissions must be controlled by either vapor
incinerators or vapor recovery units. Vapor incinerators combust
the vapors by catalytic or direct flame methods. Vapor recovery
units on the other hand recover the hydrocarbon vapors and return
them to processing streams.
The maximum degree of control attainable for the hydro-
carbon vapors from vacuum jets equipped with barometric condensers
is effectively 100 percent (AT-040). Currently however, controls
for vacuum units are not widely applied in the petroleum industry.
-270-
-------
Slowdown Systems
Blowdown emissions can be effectively controlled by
venting into an integrated vapor-liquid recovery system. All
units and equipment subject to shutdowns, upsets, emergency
venting, and purging are manifolded into a multi-pressure col-
lection system. Discharges into the collection system are
segregated according to their operating pressures. A series of
flash drums and condensers arranged in descending pressures
separates the blowdown into vapor pressure cuts. These recovered
gaseous and liquid cuts can be either flared and/or re-refined.
A typical flaring system is shown in Figure 5.1-22.
Fully integrated blowdown recovery systems can reduce
refinery blowdown emissions to 5 Ibs of hydrocarbon/103bbl of
refinery feed (AT-040). Because most refineries are currently
applying some degree of blowdown system control the average
refinery emissions from blowdown systems range from 120 Ibs to
200 Ibs of hydrocarbons/103bbl of refinery feed (MS-001, AT-040).
Process Drains and Waste Water Separators
Control measures for reducing the evaporative hydro-
carbon emissions from process drains and waste water separators
center around 1) reducing the quantity of hydrocarbons evaporated,
and 2) enclosing the waste water systems.
The quantity of hydrocarbons evaporated can first be
reduced by minimizing through good housekeeping the volume of
oil leaked to the waste water systems. Lowering the temperature
of the waste water will also reduce hydrocarbon evaporation
(AM-055).
Measures for enclosing waste water systems include
manhole covers, catch basin liquid seals, and fixed or floating
roofs for API separators. The potential also exists for some
form of vapor disposal or vapor recovery device in conjunction
with fixed roofs on API separators (EL-033).
-271-
-------
• ATER
STEAK
^ TO FLARE
STACK
OILY WATER SEWER
(TO SOUR WATER SYSTEM
IF LARGE QUANTITIES
OF H2S ARE FLARED
CONTINUOUSLY)
ALTERNATE SEALING METHOD
CXATEP SEAL)
K3
K5
I
FROM RELIEF SLOPE TOWARD
0» VENT DRUM -«.
HEAOER SYSTEM >—•»
SLOPE TOWARD
DRUM
STEAUS
MvENT
~ T i
K1OCK-OUT DRUM
TO OIL RECOVERY
FACILITIES OR SLOP *-
^
3=
S
PANEL fa
MOUNTED V RATIO
,ICNITEH LINE
STEAM TO NOZZLE MANIFOLD
FOR SMOKELESS BURNING
POWER SUP»LY
SPARK IGNITER
xS
»«
STEAM DRIVEN PUMP
4 ELECTRICALLY
DRIVEN SPARE
AIB Si/PPLY
• ' I-'L'EI. CAS
TO PILOTS
—« STEAM FOR
SMOKELESS
BURKING
,\i>ie: This represents an operable system arrangement and
its components. Arrangement of the system will vary with the
performance required. Correspondingly, the selection of types
and quantities of components, as well as their application!,
muM match the needs of the particular plant and iU speci-
fications.
FIGURE 5.1-22
^SOURCE:
TYPICAL FLARE INSTALLATION
-------
Studies of Los Angeles County refineries indicate that
hydrocarbon emissions from controlled waste water systems are as
low as 10 lbs/103 bbl of refinery feed (AT-040). On a nation-
wide basis and accounting for the existing degree of control, it
is estimated that hydrocarbon emissions from waste water systems
in 1972 averaged 105 lbs/103 bbl refinery feed (MS-001).
Cooling Towers
The control of hydrocarbon emissions from cooling
towers is best effected at the point where hydrocarbon con-
taminants enter the cooling water. Hence, systems for detection
of contamination in water, proper maintenance, speedy repair
of leaks, and good housekeeping programs in general are necessary
to minimize the air pollution occurring at the cooling tower.
In addition, water that has been used in direct contact condensers
should be eliminated from cooling towers. Greater use of air
cooling will also control hydrocarbon emissions by reducing the
size of the cooling water system (DA-069).
Refineries practicing good housekeeping in Los Angeles
County have succeeded in reducing their cooling tower emissions
to approximately 10 lbs/103 bbl refinery feed (AT-040, AM-055).
5.1.5.7 Fugitive Source Controls
Although inconspicuous, fugitive hydrocarbon emission
sources are generally significant because of their abundance.
Regular maintenance and good housekeeping are the major control
measures for minimizing fugitive hydrocarbon emissions.
Pumps and Compressor Seals
Pump and compressor seals inherently leak and there
are no practical means for eliminating hydrocarbon emissions
-273-
-------
from these sources. As brought out in the section on fugitive
emissions, the emissions from centrifugal pumps with mechanical
seals average 3.2 Ibs/day-seal and from centrifugal pumps with
packed seals average 4.8 Ibs/day-seal. Therefore, a 33 percent
reduction in hydrocarbon emissions from centrifugal pumps may
be effected by installing mechanical seals in place of packed
seals. There are no alternatives to using packed seals on re-
ciprocating pumps. Dual sets of seals may also be installed
with provisions to vent the volatile vapors that leak past the
first seal into a vapor recovery system.
Frequent inspection and maintenance of seals are very
important measures for the minimization of pump and compressor
leaks.
Pressure Relief Valves
Hydrocarbon emissions from pressure relief valves are
sometimes controlled by manifolding to a vapor control device
or a blowdown system (DA-069). For valves where it is not de-
sirable, because of convenience or safety aspects, to discharge
into a closed system frangible blanks called rupture discs can
be installed before the valve. Rupture discs serve to prevent
the pressure relief valve from leaking as well as protect the
valve seat from corrosive environments (WA-086).
The hydrocarbon emissions from relief valves controlled
by rupture discs or blowdown systems are negligible.
Pipeline Valves and Flanges
Hydrocarbon emissions originating from product leaks
at valves and flanges can be controlled by regular inspec-
tion and prompt maintenance of valve packing boxes and flange
-274-
-------
gaskets. Because of its dependence on the nature of the products
handled, the degree of maintenance, and the characteristics of
the equipment, the emissions reduction from controlling valves
and flanges is undefinable.
Pipeline Blind Changing
Emissions from the changing of blinds can be minimized
by pumping out the pipeline and then flushing the line with
water before breaking the flange. Slight vacuums can be main-
tained in the pipeline for the case of highly volatile hydro-
carbons. Spillage can also be minimized by the use of special
"line" blinds in place of the common "slip" blinds. A survey of
Los Angeles County refineries indicated that spillage from line
blinds was 40 percent of the spillage for slip blinds. In
addition, combinations of line blinds in conjunction with gate
valves allow changing of line blinds while the pipeline is under
pressure (DA-069).
Purging Sampling Lines
One means for controlling the hydrocarbon emissions
generated by purging sampling lines is the installation of
drains and flushing facilities at each sample point. Conscious
efforts to avoid excessive sampling in addition to flushing
sample purges into the drain have a significant impact on the
hydrocarbon emissions from sampling operations.
Miscellaneous Emissions
There are several other fugitive emission sources
which are collectively significant but not common to all re-
fineries and not easily identifiable. The control of these
sources is basically centered around regular inspection, proper
-275-
-------
maintenance, and good housekeeping. The efficiency of these
control measures is dependent on the degree to which they are
performed and the nature of the emission sources.
-276-
-------
5.2 Wastewater Treatment
Various forms of.wastewater treatment are involved in
LNG and SNG plants and refineries. The refinery wastewater
treatment procedures will be discussed first because they
are the largest and most complete. Many of the techniques
and methods used in the refineries will be applicable in the
SNG and LNG plant treatment facilities.
The wastewater treatment procedures for the fuel oil
refinery and the gasoline refinery are considered together and
assumed basically the same. The difference between the two in
wastewater sources is the inclusion in the gasoline refinery
of sour water from the fluid catalytic cracking unit and the
hydrocracker, caustic solution from the Merox treating units,
and possible acid wastewater from the alkylation unit. The
sour water from the crackers is steam stripped to remove sulfur
compounds, ammonia, and phenols. The stripped sour water is
then recycled for extraction of the remaining phenols or added
directly to the process wastewater stream. Caustic Merox solution
poses a special problem as far as pollution. The methods for
control of the caustic are discussed later in the report in
the section on neutralization of acids and bases. Acid waste-
water from the HF alkylation is believed to be no major problem
and is merely routed to the process water sewer system.
Methods of treatment and reduction of refinery waste-
water are many and quite varied. These methods can be, however,
classified into five general categories and they are the following
1) design of processes providing for
lower energy use,
-277-
-------
2) elimination of sources of wastewater,
3) design of a segregated wastewater treat-
ment techniques,
4) inplant treatment of wastewater streams,
and
5) "tailor-making" of the final process
wastewater treatment facilities.
Use of all these methods will produce a high purity final waste-
water effluent.
5.2.1 Optimization of Energy Use
Optimizing energy use has always been of concern in re-
finery design; however, in the last few years the threat of
energy shortages has enhanced this concern. Generally speaking
a reduction of energy use can be correlated to the reduction of
cooling water used within the refinery. Approximately 8170
of the water used in a refinery is for cooling and condensing,
and about 20% of which is returned as wastewater (TH-038).
Figure 5.2-1 shows a continual decline of energy consumption as
a percentage of crude oil during the last 25 years which would
mean an ultimate decline in wastewater (NE-088). More efficient
heat exchange is an excellent method for reducing the total
energy consumption. An example of this would be heating a cold
process stream with a hot process stream. Discussion of specific
areas of heat reduction is beyond the scope of this report.
-278-
-------
Energy consumed
f. .
14
I ''"
3
2. 10
o
* '••
c
• f »
3
| 4
>. ' .
o
1 2
' **^
" ' . 0
•/•
V
(By /-tomplexity refinsftss, 1924-1973)
•X)'50
1926\^
-*4oit V
1955
r- 1«
N
965
19°70
-. savi
0
X
1
19 re
s1;
973
lalio
771
s
nsh
.
ne
v
P-
w
1
re
>7
\
iner
_
Lim
lee
es
ted
rnir
r
by
1
--
H
-.
-»-
\
U 2 2.5 3 4 5 4 .7 8 9 10
'••';." . .• Refinery fuel pries, equivolenf S/bbl :.•:'. . .. .:•;,• '•:
:'•'•• ' ' . .-.'-. - • '.• •-.••-•. -. ..„-. -06i-i/
FIGURE 5.2-1 ENERGY CONSUMPTION IN A REFINERY
AS A PERCENTAGE OF CRUDE OIL FOR
VARIOUS YEARS
-279-
-------
5.2.2 Elimination of Wastewater Sources
Elimination of wastewater sources can eliminate many
unnecessary pollutants. Many of these methods or procedures
are inexpensive and easily adapted. Proper equipment choice
will also result in less wastewater produced. Table 5.2-1
gives examples of the correct methods to use to reduce waste-
water produced within a refinery (TH-038).
5.2.3 Segregated Wastewater System
A new and very popular idea in refinery wastewater
design is a segregated wastewater system. A segregated system
has advantages of decreasing the amount of wastewater to be
treated and concentrating the pollutants in the wastewater.
Typical constituents of a segregated system are given in
Table 5.2-2. A flow diagram of the segregated system is
shown in Figure 5.2-2 (RA-117).
The clean water system consists of an observation
channel or light duty API separator and a large surge basin.
Any small amount of oil along with debris is removed in the ob-
servation channel. The water in the surge pond can either be
routed through the process water treating facility or drained
directly to the final holding pond depending on whether it is
contaminated or not. Sanitary wastewater can be treated by an
inplant bio-treating facility design the same as a municipal
wastetreating plant (e.g. activated sludge, chlorination), or
if available, may be disposed of directly to a municipal sewer
system. If the refinery has docking for tankers, treatment
facilities must be provided for the contaminated ballast water.
The oily water resulting from ballast is held in a holding tank
equipped with floating oil skimmers. Recovered oil is recycled
-280
-------
TABLE 5.2-1
METHODS FOR REDUCING REFINERY WASTEWATER
WRONG METHOD CORRECT METHOD
1) Use once through cooling 1) Recycle cooling water
water. through cooling towers.
2) Use barometric condensers. 2) Use surface condensers.
3) Direct separation of oil- 3) Addition of light oil to
and-water mixtures. enhance the separation of
oil-and-water separations.
4) Use once through scrubbing. 4) Recycle scrubbing water
to scrubber to concentrate
pollutants.
5) Use steam vacuum jets. 5) Use a vacuum pump.
6) Use water for cooling. 6) Use air cooling.
7) Use processes producing 7) Use minimal wastewater
large amounts of wastewater. producing processes.
8) Desulfurize with water. 8) Hydrodesulfurize.
9) Surface cleanup with water. 9) Surface cleanup with dry
or mechanical cleaning
devices.
-281-
-------
TABLE 5.2-1 METHODS FOR REDUCING REFINERY WASTEWATER (Cont.)
WRONG METHOD
CORRECT METHOD
10) Use a water seal on a
flare stack.
10) Use a molecular seal on
a flare stack.
11) Direct steamout or wash-
out of vessels which hold
heavy oils.
12) Dump small amounts of oil
to the API separator.
13) Route pump jacket cool-
ing water to the sewer.
11) Flush heavy oil vessels
with light oil before
steamout or washout.
12) Recover any small amount
of oil.
13) Recycle pump jacket cool-
ing water to the cooling
tower.
-282-
-------
00
Process Water
1) Stripped sour
condensate
2) Contaminated pro-
cess water
3) Cooling tower
blowdown
4) Oily process
area storm
water
5) Caustic wash
water
6) Oily cleaning
water
TABLE 5.2-2
SEGREGATED WASTEWATER STREAMS
Clean Water
1) Non-oily storm
water
2) Once through un-
contaminated
water
Sanitary Water
1) Water and refuse
from plant sinks
and bathroom
fixtures
Ship1s Ballast
Water
1) Oily water from
the ship's
ballast
7) Desalter water
-------
FIGURE 5.2-2 TYPICAL SEGREGATED WASTEWATER SYSTEM
SANITARY WATER
SECONDARY
BIG-TREATING
SYSTEM
CLEAN WATER
OBSERVATION
CHANNEL AND
SURGE POND
00
-p-
I
PROCESS WATER
PROCESS
WATER
TREATMENT
T
SHIPS
BALLAST WATER
STORAGE TANK
WITH
OIL SKIMMERS
FINAL
HOLDING POND
FINAL EFFLUENT WATER
-------
back to the refinery whereas the skimmed water, is either intro-
duced to the process water treatment facility or routed to the
final holding pond, depending on whether it is or is not con-
taminated. Treatment of the process water is described in de-
tail later in the report.
Also included in a segregated flow plan is an exten-
sive system known as the "drip system" (GL-027). This system
is designed to capture any fugitive oil leaks from pumps, sample
lines, and overflows of vessels. The captured oil is returned
to the process. The "drip system" will reduce excess oil
flowing into the wastewater treatment facility.
5.2.4 Inplant Wastewater Treatment
Water pollution control of wastewater from refinery
units can be achieved1 at the source within the refinery itself.
One inplant technique commonly used is steam stripping of sour
water or condensate from process units such as the distillation
column, the hydrocracker, the fluidized catalytic cracker, and
the reforming units. A stripper is operated by blowing steam
countercurrently with sour water or condensate in a tray or
packed column to remove H S, NH , and phenol. With efficient
steam stripping contaminants removal of 99-100% for H,,S, 95-99%
for NH , and 50-70% for phenols has been achieved (WA-082). A
3
typical sour water stripper is shown in Figure 5.2-3.
The sour gas must be further processed to separate
the Hg S and NH . The reasons for the separation are to prevent
corrosion and to allow the Hg S to be recovered in the Glaus plant.
Problems have arisen with corrosion due to ammonium hydrosulfide
(NH4 SH). The problems are a result of high concentrations
of NH3 in the HpS feed to the Glaus plant (BR-140). A solution
to this problem is use of a "stepped" stripping process where
the H S is steam stripped in one column while the ammonia is
«
stripped in a second column. This is the basis for the Chevron WWT
-------
FltSH
. US .
STEAM
Typical sour water steam stripper.
FIGURE 5.2-3
-286-
-------
process which is shown in Figure 5.2-4. The separation can also
be achieved by a DEA absorption process. The I^S can be re-
covered as elemental sulfur in the Glaus unit. The ammonia can
be recovered as pure anhydrous ammonia which is sold or combusted
in process heaters.
The stripped water is further treated by another in-
plant technique, solvent extraction. Foul water containing in
excess of 300 ppm of phenols is extracted with either raw or
desalted crude oil or light catalytic cycle oil. The effluent
water has about 9070 of the phenol removed and contains 20 to
30 ppb of oil (WI-142). Extraction with raw crude has the
added advantage of desalting the crude with the phenol-laden
wastewater. A typical extraction process is shown in Figure
5.2-5 (WI-142).
Oxidization of high sulfide content caustics is an-
other inplant wastewater treatment technique. Sulfidic spent
caustics may contain as much as 50,000 ppm of sulfide which is
equivalent to a theoretical oxygen demand of 100,000 (BE-147).
Spent caustics cannot, however, be steam stripped since sodium
sulfide does not hydrolyze to any extent. The alkaline sulfides
can be economically oxidized with air to form thiosulfates and
sulfates and thus reduce, the oxygen demand in the final process
wastewater treatment facility. A typical spent caustic oxidizer
is shown in Figure 5.2-6 (BE-147).
5.2.5 Process Wastewater Treatment
After having completed all reductions and treatments
within the plant the final process wastewater must be treated
to meet Federal and local standards of wastewater effluent
quality. Refinery process wastewaters are quite varied and some-
times unique, and thus require specific types of treatment.
-287-
-------
HYDROGEN SULFIOE PRODUCT 50 PPM (WT) AMMONIA MAXIMUM
RECYCLE
DEGASSED
SOUR WATER
FROM STORAGE
TANK
DEAERATED
CONOENSATE
HYDROGEN
SULFIOE
STRIPPER
AMMONIA PRODUCT 5 PPM (WT)
HYDROGEN SULFIDE MAXIMUM
STRIPPED WATER PRODUCT TO PROCESS UNITS
50 PPM (WT.) AMMONIA, 5 PPM (WT.) HYDROGEN SULFIDE MAXIMUM
FIGURE 5.2-4 CHEVRON WWt PROCESS
Source: KL-032
-288-
-------
31J2L
FIGURE 5.2-5
EXTRACTION OF PHENOL
FROM REFINERY WASTE
-289-
-------
Foulwoler
(containing
sulfides)
Steam
Air >
F,.-'4URE 5.2-6 TYPICAL OXIDIZING UNIT
-290-
-------
Not only the quality of the wastewater to be treated, but also
the economical and physical limitations will determine the type
of final wastewater treatment facility. The design of a complete
"tailor-made" treatment plant has many alternatives. Figure
5.2-7 shows many of the various alternatives open for treatment
of refinery wastes.
5.2.5.1 Pretreatment
Generally, the first step in wastewater treatment is
referred to as pretreatment. Pretreatment processes will "pre-
pare" the wastewater for further treatment by making the waste-
water easier to treat. Pretreatment by aeration and grit removal
is completed in an aerated grit chamber. The aeration helps
improve the settling characteristics of the solids in the wastes
and also improves the odor of the wastewater. This method is not
especially designed to handle oily wastewater and, therefore,
should probably be used only when the wastewater has a low
oil content (< LOOppm).
For high oil content, wastewater pretreatment can be
performed by an API separator, a corrugated plate interceptor,
or flocculation. An API separator and a corrugated plate inter-
ceptor (CPI) are specifically designed for oil removal. The
CPI is based on the theory of oil-water separation which states
that the controlling parameter for separation is the surface area
per unit flow. The well-designed CPI unit will produce an
effluent with a lower oil content than the API separator. The
CPI is shown in Figure 5.2-8 (TH-076).
Flocculation is a technique where oil as well as
other organic particles within the wastewater are agglomerated
by flocculating agents in order to improve their settling
characteristics. Two common flocculants are alum and polyelectro-
lytes which are polar, synthetic, water soluble, organic polymers
-291-
-------
I
ro
ho
PRETREATUENT
SUSPENDED 3 L
GRIT REUOVAl.
-
"*" -
CORRUGATED
~~^ >LATE INTERCEPTOR
COAGULATION —
FLOCCULATION —
-
—
-
_ SEDIMENTATION
AND SKIUUING
FLOTATION
SCREENING AND
FILTRATION
MECHANICAL
THICKENING
FLOTATION
CLARIFICATION!
JDS
Y
TERTIARY DISSOLVED
SOLIDS REMOVAL
LIQUIDS DISPOSAL
HEAT REMOVAL
"
-
M
— «• TRICKLE IILTER
_ ANAEROBIC
TREATMENT
__
_
—
I ,
CHI OTINATION
ION EXCHANGE
UEU8RANE
SEPARATION
PROCESS
FILTRATION
COAGULATION
AND FLOCCULATION
1
SLUDCE DIGESTION
AEROBIC
DIGESTION
ANAEROBIC
DIGESTION
SLUDGE
SLUDGE CONDITIONING
^ CHEMICAL
CONDITIONING
*
SLUG
"
PRESSURE
FILTRATION
— *• LACOONING
..^ CONTROLLED
DISCHARGE
— »• EVAPORATION
"1
^
-*.
COOLING TOWER
(OXIDATION)
SPHAY PONDS
REUS
SLUDGE AND ASH DISPOSAL
1
FERTILIZER AND
SOIL CONDITIONER
r--
FIGURE 5.2-7 WASTEWATER PROCESSING ALTERNATIVES FOR A REFINERY
-------
M.VM0I
fUuimuit
/MUT WEIR
8T co-crcre
%/
PUTE MStKBLT CWSISTW
Of 24 OK 41 COmtlKUTEO.
MIUU.CL funs.
FIGURE 5.2-8 CORRUGATED PLATE
INTERCEPTOR
-293-
-------
of high molecular weight (FR-119). While flocculation gives
excellent removal of oil and the added advantage of removal of
other particulates, it also has higher capital and operating
costs than the CPI unit or API unit (BE-147).
If the process wastewater is acidic or basic, it must
be neutralized to allow for optimum biological treatment. Various
alkali and acid requirements for neutralizing acidic or basic
wastewaters is shown in Table 5.2-3 (TE-111). A typical schematic
of a batch neutralization is shown in Figure 5.2-9 (BE-147).
This would be the treatment used for neutralizing spent Merox
treating caustic solutions.
5.2.5.2 Suspended Solids Removal
After pretreatment the wastewater is treated for sus-
pended solids removal. Three common methods are sedimentation,
air floatation, and screening and filtration. Sedimentation
or gravity settling is no doubt the oldest technique of waste-
water treatment. Sedimentation design is based on the settling
properties of the wastewater particulates.
Air floatation is a wastewater process where air
under pressure (approx. 40 psig) is saturated within the waste-
water. When the pressure is released, millions of fine air
bubbles less than 100 micron in diameter attach themselves to the
particulates in the wastewater and float them to the surface.
Where a waste can be treated by either sedimentation or floata-
tion, dissolved-air floatation gives higher separation rates
and solids concentration (TE-111). The dissolved air will also
enhance the biological oxidation of the waste and the odor
characteristics. Air floatation, however, has higher operating
costs than sedimentation (BE-147). A dissolved air floatation
unit is shown in Figure 5.2-10 (ME-095).
-294-
-------
TABLE 5.2-3
NEUTRALIZATION REQUIREMENTS
Alkali Requirements for
Acid Neutralization
Approx. Neutralization
dosage Approx. cost
Ib./ib. cost cents/lb.
H2SO4 cents/lb.* H2SO4
Dolomitic
limestone
High calcium
limestone
Dolomitic lime,
unslaked
High calcium lime,
unslaked
Dolomitic lime, •
hydra ted
High calcium lime.
hydra ted
Anhydrous
ammonia
Soda ash
Caustic soda
• 1968 basis
0.95
1.06
0.53
0.60
0.65
0.80
0.35
1.10
0.80
0.2
0.2
0.5
0.5
0.6
0.6
4.0
1.5
2.5
0.2
0.2
0.3
0.3
0.4
0.5
. 1.4
1.6
2.0
Acid Requirements for
Alkali Neutralization
Acid
Approx.
dosage Approx.
Ib./lb. cost
CaCO3 cents/lb.*
Neutralization
cost
cents/lb.
CaC03
H2SO«, 66°Be
HCI, 20°Be
Flue gas,
15% C02
Sulfur (2)
1.0
2.0
3.0
0.3
1.5
1.5
(1)
2.0
1.5
3.0
0.6
(1) Cost would be based on blower design and equipment
amortization.
(2) The use of sulfur would produce a reducing condition
which might require additional treatment to produce
an oxygen-containing effluent.
° 1968 basis
-295-
-------
Neulro
iring
.ai
Liquid level
Educlor
Cooler
C.W. ^
:=
C.W.
M
H\
Acid from pump *" .
or blowcose : '
\
Vj 1
i
\
)
mixer
\
/ Steam
f Mixer valve1
i
i
I
\
i
r
1
i
1..
m
Gas to disposal
/" ^
xxxx
9
•— •
(L
N/
--
rr
G)
->
•7:
„*. Demister pod
^
^ 1
-tx}-, -D Vent
V/ ° ' '
rtxj-y. o Interface sampler
\ / (j X
_F -_ ^ — *X fe _
-U+j t M >
\____ j (Locale top of drum below
' f J^~ ^__^ botloin Irycock)
1 .'---(L(u_ \
77 drum _ T-—^ i Steam
4 4 T
J' 1 j Condensate
4 "" Sludge '
Acid oils'
(to storage)
Sprung woler
?A (lo stripper or sewer)
' M t
l~ Spent phenolic caustic
FIGURE 5.2-9 BATCH NEUTRALIZATION OF SPENT CAUSTIC WITH ACID
-296-
-------
Baffle
Effluent
Chemicals
Thickened
sludge
Influent
Chemical
mix tank
Pressurizing /
pump
-. '-u-' •'*?•./.-"^y-*:-' toni^. M . » .
_Air \
Retention |
tank
Bottom sludge collector
often included
Pressure
reducing
valve
FIGURE 5.2-10 SCHEMATIC OF DISSOLVED-AIR FLOATATION TANK
WITHOUT RECYCLE
-297-
-------
Generally screening and filtration should be used for
suspended solids removal when the wastewater has a small solids
loading and when one desires a high degree of separation. Under
high solids loadings the economics will suggest a sedimentation
unit or floatation unit. The wastewater effluent from the
suspended solids removal unit is on the order of 5-20 ppm of
oil and 25-60 ppm of suspended solids (BE-156).
5.2.5.3 Dissolved Solids Removal
Once the suspended solids have been removed, the
process wastewater is treated for dissolved solids removal by
secondary methods. Options which are available as shown in
Figure 5.2-7 include activated sludge, trickle filter, aerated
lagoon, and anaerobic treatment. The correct choice of process
will be determined by the land available, the characteristics
of the wastewater, and the economics.
Activated sludge is a very popular bio-treating process
due to its flexibility in handling varying dissolved solids
loading and wastewater flow rates. A conventional activated
sludge process consists of an aeration tank, a secondary clarifier,
and a sludge recycle system. Wastewater enters the aeration tank
along with recycle sludge and contacts with dissolved air which
promotes biological oxidation of the dissolved organics. The
oxidized sludge is settled out in a clarifier. Part of the sludge
is recycled back to the aeration tank and the remaining sludge
is wasted to the sludge treating facility. A conventional
activated sludge system is shown in Figure 5.2-11 (KE-095).
The BOD removal efficiencies and the applications of the con-
ventional activated sludge process and various process modifi-
cations are shown in Table 5.2-4 (ME-095). The activated sludge
process has the advantages over trickle filter or aerated
lagoon of a smaller land requirement and a more convenient
sludge handling system.
-298-
-------
INFLUENT
I
S3
PLUG FLOW AERATION TANK
•If- EFFLUENT
SLUDGE RETURN
WASTE SLUDGE
FIGURE 5.2-11 CONVENTIONAL ACTIVATED SLUDGE PROCESS
-------
TABLE 5.2-4
o
o
OPERATIONAL CHARACTERISTICS OF ACTIVATED- SLUDGE PROCESSES
BOD removal
Process modification
Conventional
Complete-mix
Step-aeration
Modi'ied-aeration
Contact-stabilization
Extended-aeration
Kraus process
High-rate aeration
Pure-oxygen systems
Flow model
Plug-How
Complete-mix
Plug-flow
Plug-flow
Plug-flow
Complete-mix
Plug-flow
Complete-mix
Complete-mix
reactors in
series
Aeration system
Diffused-air,
mechanical aerators
Diffused-air,
mechanical aerators
Diffused-air
Diffused-air
Diffused-air,
mechanical aerators
Diffused-air,
mechanical aerators
Diffused-air
Mechanical aerators
Mechanical aerators
efficiency, °/
85-95
85-95
85-95
60-75
80-90
75-95
85-95
75-90
85-95
b Application
Low-strength domestic wastes, susceptible to shock loads
General application, resistant to shock loads, surface aerators
General application to wide range of wastes
Intermediate degree of treatment where cell tissue in
effluent is not objectionable
Expansion of existing systems, package plants, flexible
the
Small communities, package plants, flexible, surface aerators
Low-nitrogen, high-strength wastes
Use with turbine aerators to transfer oxygen and control
floe size, general application
the
General application, use where limited volume is available,
use near economical source of oxygen, turbine or surface
aerators
-------
The aerated lagoon is a very popular technique for
biological treatment of refinery process wastewater (PR-046).
The aerated lagoon is a basin where wastewater is biologically
treated by oxidation. Oxygen is supplied by means of surface
aerators or diffused aeration units (ME-095). Aerated lagoons
can either be run on a once through basis or with a recycle.
Disadvantages with the process are a. large land requirement
and difficulty in sludge handling.
The trickling filter is another alternative for second-
ary removal of dissolved solids. Trickling filters consist
of a bed of rock or other packing material which can support a
biological growth. The wastewater is distributed over the bed
and allowed to "trickle" through the voids contacting with the
supported biomass. Trickling filters are classified by hy-
draulic or organic loading as high-rate or low-rate. A compari-
son of the two is shown in Table 5.2-5 (ME-095). An improve-
ment in the trickling filter is the use of a fabricated poly-
vinyl chloride packing instead of rock for greater BOD loadings
(TE-111). An alternative trickling filter process is Allis-
Chalmers Bio-Disc process in which micro-organisms grow on ro-
tating discs that are partially submerged in the wastewater.
The company reports BOD loadings of 600 Ib per 1,000 cubic feet
and BOD removal of 9070 for raw wastes containing 1,000 mg/liter
of dissolved organics (TE-111). Disadvantages of the trickling
filter are large land use, difficulty in sludge handling, and
production of offensive odors- (ME-095).
All aeration processes can experience a size (or land
area) reduction when pure oxygen is used instead of air. Also,
most industrial effluents are considerably stronger than sanitary
wastes, and thus, demand a high oxygen uptake rate during bio-
logical treatment. Further, variations in theoretical oxygen
-301-
-------
TABLE 5.2-5
COMPARISON OF LOW-RATE AND
HIGH-RATE TRICKLING FILTERS
Factor
Low-rate filter
High-rate filter
Hydraulic loading, mgad
Organic loading,
Ib BODj/acre-ft-day
Depth, ft
Recirculatian
Rock volume
Power requirements
Filter flies
Sloughing
Operation
Dosing interval
Effluent
1 to 4
300 to 1,000
6 to 10
Nona
5 to 10 times
None
Many
Intermittent
Simple
Not more than 5 min
(generally intermittent)
Fully nitrified
10 to 40
1,000 to 5,000
3 to 8
1:1 to 4:1
1
10 to 50 hp/mg
Few, larvae are washed away
Continuous
Some skill
Not more than 15 sec
(continuous)
Nitrification at low loadings
-302-
-------
demand fluctuate greater in industrial wastewater than municipal
wastewater due to the fact that spills or upsets are immediately
felt at the treatment facility and are not generally diluted.
The economics will, however, determine the feasibility of the
use of pure oxygen in bio-treating.
Anaerobic treatment processes such as stabilization
ponds can be used to treat refinery wastewater but are not
used frequently because concentrations of organic material are
generally low and because of the types of compounds present.
Also, the anaerobic process gives off HaS as a result of de-
composition of sulfurous material in the wastewater, which
results in an odor problem (AM-062, ME-095).
A comparison of all the above four biological treat-
ment process as far as area requirement, BOD loading, and BOD
removal is given in Table 5.2-6 (TE-111).
5.2.5.4 Tertiary Dissolved Solids Removal or Treatment
To further improve the quality of the wastewater
effluent, processes defined as tertiary processes can be used for
removal of special troublesome pollutants. Presently, tertiary
processes are not a vital part of a treatment facility; however,
in the future with stricter wastewater effluent standards
tertiary treatment will probably become a basic unit of the
wastewater treatment facility. As shown in Figure 5.2-7, the
tertiary processes considered for refinery wastewater are
chlorination, ion exchange, membrane separation processes,
activated carbon, filtration, and coagulation and floatation.
-303-
-------
TABLE 5.2-6
Comparison of Biological Processes—
Requirements to treat 1,700 Ib. BOD/day
Biological
loading BOD
Ib. BOD/ removal,
Area, acres 1,000 cu.ft %
Stabilization pond
Aerated lagoon
Activated sludge
Extended
Conventional
High rate
Trickling filter
Rock
57'
5.75-
0.23
0.08
0.046
0.2 to 0.5
0.09 to 0.23
1.15 to 1.60
11.0 to 30.0
33.0 to 400
57.0 to 150
0.7 to 50
70 to 90
80 to 90
95+
90
70
40 to 70
Plastic media
0.02 to 0.08 20 to 200 50 to 70
(1) 5-ft. deep (2) 10-ft. deep
-304-
-------
Chlorination is probably the most commonly used process
today for final chemical treatment of wastewater. Chlorine
compounds used include calcium hypochlorite, sodium hypo-
chlorite, and pure chlorine. The former two are generally
used in smaller wastewater treatment facility or for safety
reasons. Common dosages of chlorine for various wastewaters
are shown in Table 5.2-7 (ME-095).
Ion exchange as a tertiary technique is used for the
removal of inorganic ions or nutrients from the wastewater.
Three common nutrients removed by ion exchange are nitrates,
phosphates, and ammonia. Pilot plant operations have shown
70 percent removal of phosphates, 90 percent removel of nitrates,
and 93-97 percent removal of ammonia (CU-008). Zeolite is the
main ion exchange resin employed. Ion exchange is also used
in the demineralization of boiler water feeds. Ion exchange
has also been mentioned as a possible secondary method for the
removal of organics from the wastewater. Resin regeneration
and plugging have, however, been major problems in this type of
removal (CU-008). Another possible use of ion exchange is in the
recovery of chromates (Cr04) from cooling tower blowdowns.
The chromates are used as corrosion inhibitors and can be eco-
nomically recovered by ion exchange (BE-156).
New and diversified methods of tertiary treatment of
refinery process wastewater for removal of very small particulates
are membrane separation processes. The typical membrane process
uses a semipermeable-type membrane to concentrate various com-
ponents of the wastewater stream. Variations in the driving forces
used and the degree of separation define the specific membrane
process. Examples of membrane processes are electrodialysis
reverse osmosis, ultrafiltration, and microfiltration. The use-
ful size ranges of particulates which can be removed are illus-
trated in Figure 5.2-12 (LA-150). Reverse osmosis is the only
membrane process which has been used commercially in the treatment
of municipal wastewater.
-305-
-------
TABLE 5.2-7
CHLORINATION APPLICATIONS IN WASTEWATER
COLLECTION, TREATMENT, AND DISPOSAL
Application
Dosage range,
mg/liter
Remarks
Collection:
Slime-growth control
Corrosion control (HjS)
Odor control
Treatment:
Grease removal
BOD reduction
Ferrous sulfate oxidation
Filter-ponding control
Filter-fly control
Sludge-bulking control
Digester supernatant oxidation
Digester and Imhoff tank
foaming control
Nitrate reduction
Disposal:
Bacterial reduction
Disinfection
1-10
2-9*
2-9*
2-10
0.5-2f
1-10
0.1-0.5
1-10
20-140
2-15
Control of fungi and slime-
producing bacteria
Control brought about by
destruction of H2S in sewers
Especially in pump stations
and long flat sewers
Added before preaeration
Oxidation of organic substances
Production of ferric sulfate
and ferric chloride
Residual at filter nozzles
Residual at filter nozzles,
used during fly season
Temporary control measure
See Chap. 14 Conversion of nitrate to ammonia
2-20 Plant overflows, storm water
See Table 11-8 Depends on nature of wastewater
• Per mg/liter of H»S.
t Per mg/liter of BODi destroyed.
t GFeSO«-7H»O + 3CI, - ZFeCI, + 2Fei(SO,), + «H,O.
-306-
-------
o
Angstroms
Microns 1
It
Hllfl
1
lllllL
II
...
1
I
Microfilters J |
Ultrafiltraiion
J 1 1 im ' „ 1 _ 1
Reverse osmosis
Ilcctrociidlysis
10.
0-" 10'3
•* Ionic range
UUr;icentfi(ii<|t'
JlJUii
i
!
Ui ^
T'Ti
Filter
GortUifuijC
nun
102 103
102 . 10'1
Macromolecular
range
i
104
1.0
1 Mi
1 " "x*
r
I
cron -
rticle —
ange
!ii]'
105
10
Fine
••» « — panic
rang
1
1
10s
102
In ! «
1C - <
5 1
i
IU
1C
K
Coarse -
particle — «-
range
FIGURE 5.2-12 USEFUL RANGES OF SEPARATION PROCESSES
-307-
-------
In reverse osmosis, membranes allow passage of water
and/or hydrogen bonding solvents but impede the passage of salts
and small molecules. Pressure is applied to the "polluted"
wastewater stream to overcome the osmotic pressure and force
water through the membrane. Possible uses in the refinery are
treatment of cooling tower blowdown, boiler water blowdown,
wastewater treatment plant rinses, paved utility area drain
water, clean storm water, desalter water, API separator effluent,
and selected tank bottom water draw-offs (NE-087). Major dis-
advantages of reverse osmosis are: (1) the present availability
of only a few membranes such as cellulose acetate and aromatic
polyamide which will limit operating conditions and treatable
wastewater, (2) fouling of membranes due to high solids loading,
and (3) concentrated solutions which exert an osmotic pressure
so great that it would be uneconomical to treat (NU-009, LE-148).
Activated carbon is a very promising and flexible
method of tertiary treatment. The activated carbon acts as a
molecular trap where molecules can diffuse in but are then slowed
down within the carbon ma.trix. Large surface to volume ratios
(450 to 1,800 square meters/gin) indicate the complex structure
of the carbon (HU-094). The trapped molecules are either thrown
away with the activated carbon or regenerated by oxidation or
desorption. As a. tertiary treatment process, activated carbon
can remove as much as 95 percent of the dissolved organics from
a typical industrial wastewater (BE-156).
Filtration can also be used as a tertiary treatment.
Types of filters currently used in wastewater treatment include
slow and rapid sand filters, multi-media filters, and moving
bed filters (MBF).
-308-
-------
Microfiltration or microstrainers can be used to
remove particles in the 10 micron to 10 micron range (Figure
5.2-12). Filtration of this type is performed on a variable,
low-speed rotating drum. The degree of separation will depend
on the solids loading and the filter media.
Slow sand filters consist of a 6 to 15 inch layer of
sand which is placed over a layer of coarser material of similar
thickness, and also a drainage system. When in service,-the
filters give about 60 percent removal of suspended solids and
40 percent removal of BOD at hydraulic loading rates of 1.5-
2.5 gal/ft"/hr (CU-008). Disadvantages of the system are moderate
performance at best, large area requirement, and high mainte-
nance costs. Rapid sand filters are constructed much the
same as the slow sand filters, but are designed for loading in
the range of 2-6 gpm/ft . Under these loadings and with the
use of coagulants removal of suspended solids is approximately
70 percent and of BOD 80 percent (CU-008). Slow or rapid sand
filtration can be used for tertiary treatment of wastewater
but are generally not recommended due to the fact that multi-
media filtration gives better separation and is approximately
equal in costs.
Multi-media is the recommended tertiary sand filter
treatment in that it gives "in-depth" filtration of the sludge.
"In-depth" filtration is the filtering of wastewater within the
sand filter and not just filtration on the surface as with slow
or rapid single component sand filters. Some dual medium filter
beds which have been used in treating wastewater are (1) anthra-
cite and sand, (2) activated carbon and sand, (3) resin beds
and sand, and (4) resin beds and anthracite (ME-095). Multi-
medium filters that show promise are composed of (1) anthracite,
sand, and garnet; (2) activated carbon, anthracite, and sand;
(3) weighted spherical resin beads (charged and uncharged),
anthracite, and sand; and (4) activated carbon, sand, and garnet
(ME-095). Multi-media filtration can tolerate higher suspended
solids loadings than the single component sand filter.
-309-
-------
The moving bed filter is 'a new method of applying
sand filtration that is being applied by Johns-Manville Products
Corporation (CU-008). A sand filter bed is moved countercurrently
to incoming wastewater by means of a hydraulically activated
diaphram. The sand is cleaned in a wash tower and recycled back
to the sand filter. The continuous operation means no stopping
for backwashing as in a conventional unit. In pilot plant
operation and using coagulation prior to filtering BOD was
reduced from 40-64 mg/1 to 8.8-10.0 mg/1, COD from 111-172 mg/1
to 39-43 mg/1, and total phosphate from 30-40 mg/1 to 1.5-2.5
mg/1 (CU-008).
Coagulation and flocculation have also been used a
tertiary removal technique. The American Oil Company has adapted
this method whereas previously other installations in the petro-
leum industry have used chemical coagulation plus air floatation
as a secondary treatment after primary gravity settling and be-
fore bio-oxidation pond (FR-119). Typical coagulants used are
alum (A12 (S04 ) • 141^0), sodium aluminate (tfaA10p ) , ferrous
sulphate (FeS04), ferric chloride (FeCl^), lime (CaO), and poly-
electrolytes . The flocculation-air floatation tertiary treatment
results in removals of 70-85 percent of the oil, 30-50 percent
of suspended solids, 45-55 percent of the 5-day BOD, and 70-85
percent of phosphates (FR-119).
5.2.5.5 Final Liquids Disposal
The final liquids disposal should be of no real problem
if the quality of the effluent is within the standards. Methods
of disposal include direct discharge into the receiving waters
or into the ocean, controlled discharge from a holding pond,
or eventual discharge from a series of lagoons (lagooning).
If for some reason discharging is not feasible, zero discharge
can be accomplished through evaporation ponds. This method
is very valuable in arid climates.
-310-
-------
To cut back the wastewater discharge to the environ-
ment, the treated wastewater can be further purified to allow
for its eventual reuse in the refinery. These processes are
oxidation-type processes where the wastewater effluent is saturated
with oxygen from air. This final oxidation is for removal of
hard to oxidize components such as phenol, nitrates, and ammonia.
Further treating will also include heat removal from the waste-
water stream. Examples of the cooling-oxidation processes
are spray ponds, air stripping, autoxidation, and cooling towers.
Spray ponds or lagoons are holding ponds with some
type of aeration system (the same design as aerated lagoons).
The spray pond will have a lower oxygen demand but require a long
retention time for oxidation of nitrates (ME-095).
Air stripping is a process which has been successful
in the removal of ammonia from wastewater streams. The theory
is based on the equilibrium as shown in the following equation:
NH3 + HgO ;» NH.*"1" + OH"
As the wastewater is made more basic, the equilibrium is shifted
to more NH which is stripped by air (ME-095). The stripping is
3
performed in a packed column by countercurrent contact of the
air and gas. Air stripping has also been performed on raw
sewage. Stripping will remove lighter hydrocarbons but heavier
organics will stay in the water.
Autoxidation uses an oxidation-aeration tower and
takes advantage of stripping and cooling actions as well as
oxidation. The process removes residual hydrocarbons catalytically
by radical addition to form a hydroperoxide and subsequent de-
composition of the peroxide (PR-046). Contacting is done in an
-311-
-------
aeration tower containing the radical forming catalyst. Cobalt,
nickel, chromium, and iron have been reported as catalysts
(PR-046).
A final oxidation and cooling method is by means of a
cooling tower. Organic wastes and other pollutants such as
heavy metals are removed through a combination of biodegrada-
tion, precipitation, sorption, and volatilization (GL-027).
Normally one would expect fouling of heat exchange equipment by
biological or inorganic materials from the cooling tower water.
This, however, does not occur and is believed due to the ab-
normally large amount of heavy metals within the cooling water.
If slime growths are a problem increased chlorination and the
occasional use of dispersants is recommended (HA-132). The
biological sludge is removed from the bottom of the cooling tower
and disposed of through incineration or landfill. Expected re-
duction in pollutants from a cooling tower are shown in Table
5.2-8 (HA-132).
5.2.5.6 Sludge Handling
Sludge from a refinery wastewater are of two kinds;
oily and biological. The oily sludges results from the API
separator or the CPI unit treatment of the wastewater, stable
emulsions from tank bottoms, and also from skimming operations
in the ship's ballast water holding tank, primary sedimentation,
and air floatation, and also minor amounts from the clean water
observation channel. For several years refiners have disposed of
eily sludge through a special landfilling technique. This tech-
nique, however, requires on the order of seven acres of land
and is not effective if soil conditions allow contamination of
underground water supplies (CH-196). Incineration of oily
wastes has many advantages and is sometimes the only solution.
Incineration is discussed in the following section on solid
waste disposal. Another alternative would be to recycle the
oil back into the refinery. This method requires additional
-312-
-------
TABLE 5.2-8
EXPECTED POLLUTION REDUCTION
FROM COOLING TOWER TREATMENT
Pollutant Percent Reduction
Oil 60
Phenols 80
Ammonia-nitrogen 60
BOD 60
Suspended Solids 50
-313-
-------
oil-water separation equipment. As the value of hydrocarbons
increases, this method of removal may become economically feasible
Biological sludges resulting from sedimentation and
bio-treating are collected in sedimentation tanks, flocculation
units, screens, filters, and clarifiers. The steps involved
in sludge handling before sludge combustion and final disposal
are shown in Figure 5.2-7 (Pg. 292) and include concentration,
digestion, conditioning, and dewatering and drying. Total
solids in the raw sludge range from 2 to 770 with a typical sludge
being 4% (ME-095).
Sludge concentration processes include gravity or
mechanical thickening and dissolved air floatation. A mechanical
thickener is designed on the same principles as sedimentation.
A solids concentration of 5 to 6% can result from mechanical
thickening. A typical mechanical thickener is shown in Figure
5.2-13 (ME-095). Floatation thickeners are the same design as
the thickener described earlier for suspended solids removal
(Figure 5.2-10, Pg. 297). Concentration of the sludge ranges
from 4 to 8%. Concentration will be aided by the addition of
polyelectrolytes (ME-095).
Sludge digestion can be completed by either aerobic
or anaerobic methods. Anaerobic digestion of biological sludges
from refineries is rarely done even though there may be some
value in it. Anaerobic treatment can be only justified economi-
cally for large installations (AM-062). Design of an anaerobic
digester for a refinery would be the same as for a municipa.1
wastewater treating facility.
Aerobic digestion of sludge usually results from ex-
tended aeration, contact stabilization, or aeration in lagoons
with large retention times. The advantage of aerobic digestion
over anaerobic are:
-314-
-------
Conduit to motor
Influent pipe
Conduit to
overload alarm
.X /
' '• • • . • '--Overload alarm' \'
' ' ''
r. Direction of rotation . ...--.. ...-
Effluent channel
PLAM
Turntable base
Handrail
Weir
Influent pipe
•—'Center column
: Center cage
Center scraper
pipe
SECTION A-A
FIGURE 5.2-13 SCHEMATIC OF A MECHANICAL THICKENER
(FROM DORR-OLIVER)
-315-
-------
(1) volatile-solids reduction approxi-
mately equal to that obtained
anaerobically,
(2) lower BOD concentrations in super-
natant liquor,
(3) production of an odorless, humus-
like, biologically stable end product
that can be disposed of easily,
(4) production of a sludge with excellent
dewatering characteristics,
(5) fewer operational problems, and
(6) lower capital costs.
The major disadvantage is in operating costs of the aeration
eqvtipment.
Sludge conditioning is preformed for the sole purpose
of improving the dewatering characteristics of the sludge. The
two types most commonly used are chemical treatment and heat
treatment. Chemical treatment is in essence coagulation and
employs the same coagulants as previously mentioned such as
ferric chloride, lime, alum, and polyelectrolytes.
Heating the sludge for short periods of time under
pressure will result in coagulation of solids, breakdown of the
gel structure, and a reduction in the affinity of the solids for
water. Additional advantages are near sterilization and deodor-
ization of the sludge.. The supernatant from heat treated sludge
is high in concentration of low molecular weight, highly soluble
-316-
-------
organic compounds. These compounds are easy to biologically
treat and should be returned to the biological treating area of
waste disposal plant.
Other techniques investigated for sludge conditioning
are freezing and irradiation (ME-095). Much research needs to
be done, however, to make these processes feasible or economical.
Sludge drying and dewatering is made easier once the
sludge has been conditioned. Dewatering and drying processes in-
clude drying beds, vacuum filtration, centrifugation, pressure
filtration, and heat drying. The choice among this method
will depend on the characteristics of the sludge, the land
available, the method of final sludge disposal, and the econom-
ics of the situation.
Drying beds are used to dewater digested sludge. The
total number of beds will be determined by the digested sludge
production rate and the moisture content desired in the sludge.
Each bed is designed to hold approximately one-half to one
load from a digester. The beds are sand and include an under-
ground drainage system to drain water. If odor is a problem,
the beds can be covered with green-house types of enclosures.
Under favorable conditions a 10 to 15 day retention time will
result in a sludge containing 60% water (ME-095). Drying beds
will require a large land area.
Vacuum filtration is probably the most widely employed
mechanical means of dewatering sludges. There are many vacuum
filters which can be applied to dewatering sludge. In select-
ing a filter, important factors to consider are the following:
(1) sludge slurry character, (2) sludge production level,
(3) required results, and (4) materials of construction. The
most common type of vacuum filter is the rotary drum. There
are variations in the rotary drum such as multicompartment,
-317-
-------
single-compartment, belt, precoat, Dorrco, hopper dewaters,
and top feed. A typical continious rotary drum filtration pro-
cess is shown in Figure 5.2-.14 (CH-196). Other types of vacuum
filters include scroll-discharge, tilting-pan, disk, and batch
leaf. Moisture content of a vacuum filtered sludge usually is
on the order of 70 to 80 percent; however, filters may be op-
erated to produce a cake of 60 to 70 percent if desired (ME-095).
Centrifugation can also be used to dewater refinery
process wastewater sludges. The commonly used sewage treatment
centrifuge is the solid-bowl type. The moisture content of the
sludge cake produced is 75 to 80 percent (ME-095).. One major draw-
back of centrifuging is disposal of the centrate which is relatively
high in suspended, nonsettling solids. Recycle of the centrate
back into the wastewater treating facility could result in a
large solids loading. Two methods to help eliminate this problem
are (1) design of a longer liquid retention time in the centri-
fuge and (2) use of coagulating agents (ME-095). The scroll-
discharge centrifuge is another type centrifuge that could poss-
ibly be employed in dewatering wastewater sludges.
Pressure filtration is preformed by a filter press
consisting of a series of filter cloth-fitted, rectangular
plates supported face to face. The plates are held together
by a pressure sufficient to seal them to withstand the pressure
applied during the filtration processes. The moisture content
of the cake produced is 55 to 70 percent (ME-095). The filter
press is capable of handling most any type of sludge and has
the advantage of producing a filtrate which contains 10-20 mg/1
of suspended solids and less than 200 mg/1 BOD (CU-008). Condi-
tioning of the sludge may or may not be needed. A typical filter
press operation is shown in Figure 5.2-15 (CU-008).
-318-
-------
- VVasli sprays
Air connection
• Continuous rotary filter
Moisture
trap
30 ft
Dry
vacuum
Air out pump
Barometric St.-ol
FIGURE 5.2-14 FLOWSHEET OF CONTINOUS VACUUM FILTRATION
-319-
-------
FIGURE 5.2-15 PASSAVANT FILTER SLUDGE
DEWATERING SYSTEM (COURTESY OF
PASSAVANT CO.)
-320-
-------
Heat drying may also be used to dewater sludges and
has been used to dry sludges for more efficient incineration
or processing into fertilizer. The drying is most commonly
preformed in the C. E. Raymond Flash Drying System (Figure
5.2-16) (ME-095). Alternative dryers are multiple hearth
incinerators and rotary kilns. The dewatered sludge has a
moisture content of less than 10 percent (ME-095). Pretreat-
ment of the sludge by filtration or centrifugation may be de-
sired to lower the heat requirement for drying the sludge.
Spray drying may also be used for drying of sludge but appli-
cation has been extremely limited. A spray drying system is
shown in Figure 5.2-17 (ME-095). Heat drying processes gener-
ally have high capital and operating costs.
5.2.5.7 LNG and SNG Plant Wastewater Treating
The LNG and SNG plants have very small wastewater
streams. In the LNG plant boiler blowdown is recycled through
detnineralizers and thus the only major process wastewater source
is regenerant wastes from the blowdown. The major constituent
of the effluent stream is dissolved salts. Water effluents
from the process area will be slightly oily due mostly from
lubricants. A dike system around the processing area will con-
tain oily process effluent water. The oily wastewater is pro-
cessed by a oily water separator of the CPI (corrugated plate
inteceptor) type. The separator effluent can be treated with
a small coagulation, sedimentation, or filtration system. The
final effluent can be wasted to an evaporation pond if there
is enough land area. To protect the environment against poten-
tially polluting fluids the evaporation pond should be constructed
of concrete or asphalt or lined with a flexible membrane liner.
If land is not available the effluent can be drained directly
into the municipal sewer system. A holding tank of some sort
should be included to contain possible spills in the processing
area.
-321-
-------
Relief valve
j Refractory
j Hot gases to drying system
I Drying system
| Spent gases and vapors
j Pneumatic fertilizer
handling system
Mixed refuse incinerator ;.
Ash discharge
Storage vent ian
Storage bin
Slide gate
i
j Bagging scali:
Cooling and convoying
FIGURE 5.2-16 FLASH-DRYING SYSTEM WITH MIXED REFUSE INCINERATOR
[FROM COMBUSTION ENGINEERING]
-------
Stack
^~ Li'-juor or sluny
Fe^H lank-, / supply from
Feed pt;mp-.
i »TT »
H-:it-a:r ". . Spray
inl~t • disk
Instant drying chamber
w
Fan exhaust
10 atmosphere
Cos
"m r
Exhaust fan
Fan platform
J,
\
urnace
t
1
[ \\ /y
Otiilcl
... i
F^v^
D*-x
Ef:ST Cvclons
RTY NX X^Motvri^rl ^
p,o <•''-' N\ S/ c|u-,
valve ~-\ v — r/
\ Va»
vaive..
"^
^9
Vy Hopper
"C?
Motor— ^_ ^^ f/3) Spiral conveyor |_|
r~ i r-~^ -ffl —
*
level
^•Finished product to stor
age
FIGURE 5.2-17 SPRAY DRYER WITH PARALLEL FLOW
[FROM INSTANT DRYING COMPANY]
-323-
-------
The SNG plant will have a greater wastewater effluent
rate per Btu fuel produced than the LNG plant. The increased
effluent flow is mainly due to a large steam load and waste
solution from the Benfield system. The same treatment facilities
used in the LNG plant can be used in the SNG plant.
Other techniques for reducing the wastewater effluent
rate for LNG or SNG plants include use of air cooling, design of
a system using a minimal amount of energy, recycle of water
wherever possible, and use of proper cleaning and housekeeping
methods.
5.2.6 Summary
The wastewater treatment units and procedures described
in this section have been or potentially can be applied to the
wastewater from refineries, SNG plants, and LNG plants. It
should be noted that the information given for processes for
which no refinery, SNG plant or LNG plant application has been
cited is from municipal wastewater treatment facilities. Con-
sideration must therefore be given to the differences in industrial
wastewater and municipal wastewater in order to apply these
processes.
Sludge combustion and sludge and ash disposal are
discussed in the following section on Solids Emission Control.
-324-
-------
5.3 Solids Emission Control
The LNG plant has no appreciable solid wastes asso-
ciated with the production site. The SNG plants have solid spent
catalyst wastes which are generated on an intermittent basis.
The amount of catalyst is small and is suitable for landfill
and thus poses no real pollution problem. Solid wastes from a
refinery consist of dirt, grit, oily sludges, and settled out
sludges removed in the primary treatment processes, bacterial
sludges removed in the secondary treatment clarifiers, chemical
sludges resulting from chemical treatment of the wastewater, and
finally, intermittent spent catalysts.
5.3.1 Sludge Disposal Methods
The dirt and grit obtained from the grit chamber are
disposed of in landfills. The oily sludges, primary clarifier
sludges, bacterial sludges, and chemical sludges which have
been dewatered or dried can be handled in various methods.
The most popular methods for sludge disposal in the
past have been ocean dumping or drying in open beds followed
by landfill (SO-080). Problems have arisen with both methods
which make them unfeasible for disposal methods. The ocean dump-
ing method is being strictly regulated and eliminated where
possible by Federal and local government. Drying in open beds
or evaporation ponds can create major odor problem. Another
problem is that open beds also require a large land area which
may not be available to modern refineries located in metropolitan
areas. Suitable landfill areas required for the dried sludge
may also become scarce within populated areas and thus present
a problem of transportation of the sludge to the proper landfill
site.
-325-
-------
Chemflx Process
A recently developed landfill method for handling of
sludge waste is the Chemfix process (WI-144). In this process,
a three phase reaction forms a chemical matrix which traps the
sludge in a pseudomineral material which is suitable for landfill.
Advantages of the process are relatively low cost, controlled
rate of solidification, high continous throughput rate, mo-
bility, small volume increase due to chemical additives, ability
to react with complex waste mixtures, ability to process low-
solids wastes without discharge from the process, and nontoxicity
of the solid material (WI-144).
Incineration
One alternative to landfill is sludge incineration.
The type of incinerator used is dependent on the moisture
content of the sludge. Rotary kilns can operate over a wide
range of 5 to 70 percent solids. Lower solid concentrations,
2 to 10 percent, will requite a special fluidized sand bed
incinerator, while higher concentrations, 40 to 70 percent can
be combusted in a simpler stationary multiple hearth incinerator
(RA^OSl). In order to obtain these concentrations some type
of thickener must be employed. Thickeners are discussed in the
wastewater treatment section of this report (Section 5.2).
The economics of sludge incineration show that for small
lean sludge quantities concentration is not economically feasible,
while large lean streams prove economical to concentrate in,
order to reduce the size of the incinerator required (RA-081).
-326-
-------
American Oil Company's Mandan, North Dakota, refinery
presently uses a fluidized hot sand incinerator (MA-226). The
incinerator has the added advantage of being able to incinerate
spent caustic from various refinery units. The solids from the
incinerator can be landfilled or used as an excellent substitute
for a mixture of sand and rock salt used on icy roads (MA-226).
The Mandan incinerator is shown in Figure 5.3-1.
Sludge Spraying on Land
Another sludge disposal method is spraying dilute
sludge on poor soils to increase fertility (SO-080). Sludge
injection into the soil has also been suggested to help eliminate
odor problems involved with spraying. Methods such as these,
however, require a market for the sludge and a means of trans-
porting the sludge to the market.
Lagooning
Lagooning may be used as a simple and economical
method for handling ultimate sludge disposal if the refinery is
located in a remote area. In lagooning the sludge organic solids
are stabilized by aerobic and anaerobic decompostion which
may give rise to objectionable odors (ME-095). To avoid material
buildup the stabilized sludge will have to be removed from the
basins intermittently.
Fertilizer Production
Sludge may be also handled by heat drying and treat-
ing to produce a fertilizer. Technology for fertilizer production
exists, but the major problem arises from finding a market for
selling the fertilizer. Heat treatment costs and transportation
costs also raise the price of the final fertilizer product.
-327-
-------
ho
00
I
SPENT
CAUSTIC
WATER
(TEMP. CONTROL)
LIQUID OILY
SLUDGE
SOLID OILY
SLUDGE
TORCH OIL
PREHEATER
AIR ~
OVHD
GASES J
INCINERATOR
CYCLONE
RETURN
WATER
SCRUBBER
I
WATER
TO SEWER
SCREW.
CONVEYOR [ ' /DUMPSTER
FIGURE 5.3-1 FLUIDIZED BED INCINERATOR
Source: AM-020
-------
WET OXIDATION
Wet oxidation is a process for destruction of dissolved
or suspended organic matter by oxidizing with air at tempera-
tures above the normal boiling point of water (212°F, 100°C),
and under pressure. The process has been used traditionally
to treat sewage sludge but is finding new applications in
industrial wastewater treatment. Zimpro Inc., Rothschild, WI,
has over 130 units in operation or under construction, several
operating on coke oven gas liquors and ethylene cracking
wastewater.
Performance is typically in the range of 90-95% COD
reduction, and better than 9970 reductions in phenolics, cyanide,
and sulfur compounds.
5.3.2 Catalyst Solid Disposal
The disposable catalysts in the refinery are
those from the cat cracker and the hydrodesulfurization
units. The cat cracker uses either a synthetic silica alumina
or natural silica alumina catalyst which are highly inert and
applicable .to landfilling (NE-044). Hydrodesulfurization
catalysts are generally inert, non-noble catalysts which are
suitable for landfill (RA-119). Noble catalyst will not be
discarded, but will be recycled for metal values. The spent
resid hydrodesulfurization catalyst will contain vanadium and
nickel and could possibly be disposed of to a recovery opera-
tion, thus reducing the amount of catalyst requiring disposal.
-329-
-------
FIGURE 5.3-2
GENERALIZED FLOW SCHEME FOR A
ZIMPRCf WET AIR OXIDATION UNIT
co
LO
o
RAW WASTE
HIGH
PRESSURE
PUMP
HOT
OXIDIZED
LIQUID
1
I OFF-GAS
ISO PSIG I
STEAM
'FEED
HEAT
EXCHANGER
COOL
OXIDIZED
LIQUID
(BRINE)
REACTOR
GENERATOR
COMPRESSOR
BOILER
FEED
WATER
AIR
SEPARATOR
-CONDENSATE
(From Zimpro, Inc.)
-------
6.0 PLANT IMPACT AND SITING PROBLEMS
The purpose of this section is to identify the environ-
mental impact and siting problems of new refineries, SNG and LNG
plants, and their associated support facilities. To accomplish
these tasks, basic information on the type, size, and the specific
location of the plant or refinery is required. Appropriate
premises have been set for purposes of this report. Modules of
the following type and size are considered:
Refinery
Oil: 200,000 bbl/day
Gasoline: 200,000 bbl/day
• SNG Plant: 125 MM scf/day of SNG produced
• LNG Plant
Peak Shaving: 100 MM scf/day
Base Load: 750 MM scf/day
The environmental consequences investigated are primarily
the gaseous effluents to the ambient air, the wastewater effluents
to the receiving waters, and the solid wastes for disposal.
The effects of these effluents on the air and water quality in
the area where an oil refinery, an SNG plant, or an LNG plant
is intended to be located can be best described through use of
available computational schemes, such as ambient air and water
impact models. The results of these computations can be com-
pared with the existing local, state, and Federal regulations.
This study does not call for a specific site; however,
impacts are site specific. For this report, a comparative analysis
is used. The approach is to compare the effluents of the modules
in this study with the effluents from a reference module on
-331-
-------
which extensive environmental impact investigations have been
performed. To achieve a meaningful comparison, it is assumed
that the stack parameters, meteorological conditions, receiving
water body conditions, and process control technology for the
reference modules and for those presented in this report are the
same.
Siting problems related to feedstock availability
and its transportation, energy, water requirements (circulating
and makeup water), product transportation, and status of Federal,
state, and local laws are identified and discussed in general
terms.
The environmental impact and siting problems of the
modules are studied in a conservative manner. This implies
that the major support facilities (power plant, storage farm,
and marine terminal) are integral parts of the oil refinery,
SNG, and LNG plant sites. The environmental consequences and
siting problems of the modules are presented and discussed in
the following pages.
6.1 Petroleum Refinery Impact
This study considers fuel oil and gasoline refineries.
Each of the two refinery types is characterized as a module
with a feed capacity of 200,000 bbl/calendar day (bpcd) of raw
crude. Detailed descriptions of the fuel oil and gasoline
refinery modules are presented in Sections 3.1 and 3.2.
This section deals with refinery effluents and regu-
lations, raw material availability, water requirements, and
product transportation.
-332-
-------
6.1.1 Refinery Effluents
Refinery effluents of primary concern in this report
are air emissions, water effluents, and solid wastes. Their
impact can be assessed by estimating the impacts of each on
ambient air and water quality.
6.1.1.1 Air Emissions Impact
Reasonable estimates of the impact on ambient air
quality which might result from the location of a new petroleum
refinery at a given site can be obtained by using air dispersion
models. There are three types of models; short-term average,
long-term average, and 24-hour (hybrid). The models are based
on the Gaussian dispersion approximation originally formulated
by Button (SU-044) and modified by Pasquill (PA-095) and by
Gifford (GI-035).
The long-term average model uses historial meteor-
logical data to estimate annual average pollutant concentrations.
These estimated annual averages and certain statistical assumptions
(LA-100) are then used to estimate maximum concentrations for
averaging times less than one year. The short-term model computes
estimated concentrations corresponding to a 10-minute averaging
time. A statistical assumption (TU-020), different from that
employed in the annual average model, is used to transform the
10-minute average estimates to estimates corresponding to other
averaging times. The 24-hour model is, in essence, a hybrid of
the two models, in that it incorporates some of the averaging
features of the long-term model with the statistical assumptions
of the short term model.
-333-
-------
The atmospheric dispersion models require specific
information for input data. This information consists of est-
mated emissions, the refinery configuration, and meteorological
data. The computed air emissions of the 200,000 bpcd refinery
modules, given in Sections 3.1 and 3.2, are summarized in Table
6.1-1. In addition, Table 6.1-1 shows a summary of the calculated
air emissions of a reference refinery model based on a gasoline
refinery with a feed capacity of 300,000 bpcd of raw crude (RA-181)
In Table 6.1-1, the particulates released by the 200,000
bpcd gasoline refinery module are 440 Ib/hr compared to 353
Ib/hr from the 300,000 bpcd reference module. Some of the units
in the gasoline refinery module, such as crude distillation,
heavy naphtha reformer, heavy hydrocrackate reformer and hydrogen
plant are fuel oil fired units; thus, they produce higher particu-
late emissions than the predominantly gas fired units in the
reference modules as indicated in Table 6.1-2.
Refinery Configuration
The basic process elements of a refinery complex which
give rise to gaseous pollutant emissions include distillation,
hydro-desulfurization, catalytic cracking, reforming, and
isomerization. Power plant combustion processes also contribute
to pollutant emissions. Finally, fugitive losses of hydrocarbons
from process and storage areas must also be considered as
emission sources.
Besides the pollutant emissions from the major sources,
ambient air quality in the plant vicinity depends on the spatial
distribution and physical characteristics of the emission sources.
Consequently, to make definite predictions concerning pollutant
dispersion, it is necessary to specify the locations and physical
stack heights for each of the major sources in the refinery complex.
-334-
-------
TABLE 6.1-1
REFINERY AIR EMISSIONS
Pollutant
Particulates, Ib/hr.
200,000 BPCD Fuel Oil 200,000 BPCD Gasoline 300,000 BPCD Gasoline
Refinery Module Refinery Module Refinery (Reference)
263
432'
353
Sulfur Dioxide (S02), Ib/hr.
Hydrocarbons (HC), Ib/hr.
Carbon Monoxide (CO), Ib/hr,
Nitrogen Oxides (NO ), Ib/hr,
X
667
3,082
50
493
948
3,200
95
1,250
1,918
6,418
138'
1,846'
Excessive particulate emissions are due to the combustion of some fuel oil
in the gasoline refinery module whereas only fuel gas is combusted in the
_ reference refinery.
The larger SOz emission rate is due to a larger volume of tail gas being
emitted from the reference refinery which is a result of a higher percentage
_ of sulfur in the crude.
The larger hydrocarbon rate is due to the greater crude and petroleum storage
, capacity in the reference refinery.
These values are comparable on a size of refinery basis.
US
w
Ul
-------
TABLE 6..1-2
u>
co
1)
2)
3)
4)
5)
•6)
7)
£)
9)
10)
11)
12)
13)
14)
15)
EMISSIONS
Source
No. 1 Crude
Unit ATM.
Disc. Her.
No. 1 Crude
Unit V.-ic.
Disc. Her.
No. 2 Crude
Ur.it AT>:.
Dist. Kcr.
So. 2 Crude
Ur.it V.-.c.
Dist. Kcr.
Xo. 1 Do-
or, it
No. 2 Be-
asphalcing
Gas Oil
!1DS Unit
ACS id KDS
S.R. Naphtha
KDS Ur.ic
S.R. Naphtha
Sofonr.inj
Unit
Isoncrisa-
ticn Unit
HyJrocrack-
ir.£ Unit
H.C. N'cphcha
KDS Ur.it
ii.C. Ncpiicha
Reforair.g
Unit
Alkylation
Unit
AND STACK PARAMETERS
Heat f2)
Input Process
KM Gas
BtuVHr Fired
500
125
500
125
317
317
117
108
70.3
533
20.3
425
41.7
442
108
765x10'
191x10'
765x10'
191x10'
434x10'
48 4x10 >
173x10'
165x10'
108x10'
896x10'
31.8x10'
650x10'
63.8x10'
675x10'
165x10'
300,000 BPCD GASOLINE REFINERY
Emissions Ibs/hr [3j
Particulates
3.6
2.15
8.50
2.15
5.40
5.40
1.99
1.86
1.22
10.1
0.357
7.31
0.716
7.57
1.86
_SO;
20.5
5.14
20.5
5.14
13.0
13.0.
4.78
4.43
2.90
24.1
0.35
17.4
1.71
1S.1
4.43
Total
Ornanics
1.43
0.358
1.43
0.353
0.905
0.905
0.332
0.310
0.202
1.67
C.0595
1.22
0.119
1.27.
0.310
CO NOx
8.1 110
2.03 27.4
3.1 110
2.03 27.4
5.10 69.5
5.10 69.5
1.88 25.5
1.76 23.7
1.15 15.6
9.50 128
0.337 4.56
6.90 93.3
0.676 9.17
7.16 96.7
1.76 23.7
Mass
Flow
Ibs/hc
510x10'
123x10'
510x10'
123x10"
324x10'
324x10'
119x10'
111x10'
72,5x10'
599x10'
21. 2x10 '
435x10'
42.7x10'
452x10'
111x10*
(REFERENCE MODULE) [1]
Ftach Parameters [:
196x10'
49x10'
196x10'
45x10'
124x10'
124x10'
•45.7x10'
42.4x10'
27.8x10'
230x10 '
8.2x10'
157x10'
16.4x10'
173x10'
42.4x10*
Velocity
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
Height
200
200
200
200
200
200
200
200
200
200
200
200
200
200
2CO
Temp .
450
450
450
450
450
450
450
450
450
450
450
450
450
450
450
Diameter
ft.
3.33
4.16
8.33
4.16
6.62
6.62
4.02
3.87
3.14
9.02
1.70
7. 68
2.41
7.33
3.37
-------
TABLE 6.1-2
EMISSIONS AND STACK PARAMETERS 300,000 BPCD GASOLINE
Page Two
Heat Emissions Ibs/hr 131
REFINERY (REFERENCE MODULE)
Stack Parcr.nters
Input
KM Fuel
Source Btu/Hr SCFH Pnrticulates
16) Mia. Dist.
HDS Unit 138 210x10' 2.36
17) FCCU Unit 333 508x10' 5.72
18) FCCU Regene-
rator [5] . . 34.4
19) Partial Oxi-
dation No.
1 [6! 10.9 16.6x10' 0.186
20) Partial Oxi-
da t ion Kos .
2 and 3 (6) 21.7 33.2x10' 0.372
' 21) Partial Oxi-
dation N'os.
4 and 5 16] 54.9 84.2x10' 0.942
' 22) A-Stean
l*> Generation 1108 [7J 137
CO
-J 22) B-Electrical
1 Generation 933 [7] 98
23) Sulfur Re-
covery [8]
24) Sludge In-
cineration 8.40
25) Miscellaneous
[9J
26) Petroleum
Storaec
Total 353
111 One ealnlcn lo-jrec per proem or pr ceistn; aquaro
(plo: plan) wa» used with Che followl £ exceptional
A ATM and ac. Here. - K». 1 Crude U 1C
C S.R. S.i? tl-a HDS and RVferalng Unit
D H.C. Nap tha Has end aefoi-jinj Unt a
I fCUU Pro eta Htr, and Rcjcn.
(HY'OIJ) >
Velocity - 60 F?S
T«vtrature - 450T
kel(.:>t • 200 ft. (Initial eatlnate)
. *
It) fertlel Baleatlea eaUaalone end racea vere apllc
PI Beet reoulred fro. fuel oil (22. SX of total Beat
Total
S0> Orr.anica CO
5.64 0.392 2.22
13.6 0.95 5.40
395 15.4 13.8
• 0.44 0.031 0.176
0.89 0.062 0.352
2.26 0.157 0.892
629 22.2 26.3
J31 18.7 22.2
150 '
34.4 1.35 5.49
3810
2538
1918 6418 138
19} Sulfur recovery facilities aaau
of 99.«: recovery (U-119). Su
[t] Kiscellaneoue HC emleilona were
vtl of. charge (CM-CO).
Kass
Flow
NOx Ibs/hr
30.1 141x10'
73.0 340x10'
359 724x10'
2.38 11.1x10'
4.76 22.1x10'
12.1 56.1x10'
231 1060x10'
238 890x10'
. 1950x10'
12.0 40x10'
1846
n»d CO be capable
ICur recovery call
1)
eeeuraed to be 0.1
Velocity Hoip.ht Tea?.
ACFM fp" ^. «F
54x10' 60 200 450
131x10' 60 200 450
243x10' 60 200 350
4.24x10' 60 200 450
8. -48x10' 60 200 450
21.5x10' 60 200 450
404x10* 60 200 450
342x10' 60 200 . 450
727x10' 60 200 450
15.3x10' • 60 200 450
17
50
Dic-.ctcr
ft.
4.37
680
9.27
1.22
1.73
2.76
11.96
11.0
16.04
2.33
SOURCE: (RA-181)
-------
Stack parameters associated with process heaters are
selected on the basis of current trends in the refining industry.
Heater flue gas outlet temperatures were set at 450°F, a relatively
low temperature. This is because future high costs for fuels
will make maximum heat recovery profitable. The stack heights were
set at 200 feet, somewhat taller than current practice, in order
to demonstrate maximum dispersion effects. Typical stack exit
velocities were set at 60 ft/sec. Emissions from petroleum
Storage tanks are assumed to have a stack height of 50 feet, while
fugitive losses in the processing area have a stack height of
17 feet.
The physical configuration of the sources within the
site area for the 300,000 bpcd reference module is shown in Figure
6.1-1. The two refinery modules are assumed to be built in
approximately the same configuration and in such a way that
geometrical symmetry is preserved. The orientation of the site
is such that the line of the stacks is at right angles to the
prevailing wind.
Meteorological Data
Two sets of meteorological data are necessary for
estimating the short-term and annual average maximum concen-
trations. These data consist of 24-hour and annual meteoro-
logical conditions. The 24-hour data are specifically employed
when computing short-term maximum concentrations. The annual
meterological information is used when calculating annual
average maximum concentrations.
Climatological conditions vary from one area to another.
An arbitrary representative site was chosen as a basis for
selecting meteorological data. In estimating maximum concentra-
tions at the selected location, questions of the following nature
were considered:
-338-
-------
WIND DIRECTION
.
LO
u>
VO
1
i
i
i
i
i
i
i
i
i
fr
i
1
i
i
i
i
i
i
I
I
(
Waste Wat<
i
-/
\
-*T
rir
,f
Plant
Sulfur
Storage
Dry Tankage Area
25-MlSCELLANEOUS HC THROUGHOUT PLANT
26-HC EMISSIONS FROM PETROLEUM STORAGE
1
i
i
API Separator &
Waste Water
Treating
No. 1 - Partial
Oxidation Unic
[Til
Fuel
Oil
System
No. 1 1
Pcnt:ane
Deapphalt |
Unit fi|
No. 2 & 3 \]^
Partial Oxidation "Jnit
& Shift Conversion
— •*
No. 364 l2l|
Partial Oxidation Unic
& Shift Conversion
=j i ;
No. 1. 2. 3
Sulfur Plai
No. 2 [_§_
Deat.; halting
Unit
&
ts
4- |2c)
FCt: Merox
Unit
Fluid Cat Cracking
T7]
•
— >
^^
Unit
{Ti]
122 A 1 teec-.aier
Steam Treating
Generation <">d Air
! Compression
No. 1 |j_
Crude &
Vacuun
HeatersFp
Li
CO HDS Unit
No. 1
Crude & Vacuua Unit
No. 1 [Z
Crude &1—
Vncuuia
Unic _
Heaters] 4
Crude
S0. i Li
DAO KOE Unit
•
No. 2
& Vacuua Units
lisj
Hydrocracking Unit
Future Procacs Unic
R
t
|22-:i '«•« r:ectri-!
"a::e:: cal
Cooling Storage c..w.
Tower Arni £• ?".-ps s-a^ian
No. 1 & 2 [jjjj ;:o. 1 6 2
-., j ^. /-ire i
i!id Dist
Caroline
HD3 Unit Trcit.
l£ji-'a?h HTU «, li£J:
u e-l n
r ~ ," Refonacr Ur.tcs
"3<
i <
No. 1 6, 2 j:;°- ! & 2
?t-.C3.-.t
SAT CAS Plant jsc-rr< rr.--;cn
i- m
L£J Alky Kerrx
Alkylatlon Unit
Unit
1
I
^ — — — -- .— x
i JIJ5J H. C. L±j
Kaph K7U & Reformer 1
c>"2 P--
J 2^ Ln"1
r
'»
I
Fucure Proccoa Unit
FIGURE 6.1-1' HYPOTHETICAL CONFIGURATION OF A 300,000 BPCD GASOLINE
REFINERY (REFERENCE MODULE)
-------
. What is the probability that a "worst"
case for 24-hour data from a selected
area can occur at any location at least
once within a year?
Is there a general behavior pattern
for pollutant dispersions on an annual
basis?
The first question considers whether a set of 24-hour
meteorological data from a given area at a certain time of the
year (for example that of Brazoria, Texas, given in Table 6.1-3),
that is considered "worst" from the standpoint of maximum con-
centrations, can also occur elsewhere at other times of the year.
The second question pertains to a situation where the annual
pollutant dispersion can be characterized in general terms.
Concentrations of air pollutants are functions of
dilution and diffusion processes. The absence of either process
results an increase of ground level concentration. In the
free atmosphere, both of these processes depend almost entirely
on familiar meteorological parameters, viz, atmospheric stability,
wind speed and direction, mixing depth, temperature, turbulence,
precipitation, inversions, etc. These parameters, one way or
another, are interrelated.
The first four parameters are the meteorological
input conditions required by the atmospheric dispersion models;
an example is the 24-hour climatological conditions for Brazoria,
Texas in a typical July as shown in Table 6.1-3. This particular
location, during a typical July, might exhibit six types of
atmospheric stability classes. These are identified as types
A,' B, C, D, E and F. "D" stability, a neutral type, occur nine
hours during the 24-hour example period. When "D" stability
-340-
-------
TABLE 6.1-3
"WORST" CASE METEOROLOGICAL CONDITIONS
WITH THE
OCCURRENCE OF HIGH
24-HOUR AMBIENT
ASSOCIATED
CONCENTRATIONS :
BRAZORIA, TEXAS
TIME
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
STABILITY
F
F
F
F
F
F
D
D
C
B
A
A
A
A
B
B
C
D
D
D
D
D
D
D
(day)
(day)
(day)
(night)
(night)
(night)
(night)
(night)
(night)
TEMP
65°F
65°F
65°F
65°F
65°F
65°F
70°F
70°F
73°F
74°F
80°F
80°F
80°F
80°F
84°F
84°F
85°F
80°F
80°F
80°F
80°F
80°F
80°F
80°F
MIXING
DEPTH
400
400
400
400
400
400
500
500
600
900
1200
1200
1200
1200
1500
1500
1500
1200
1000
1000
1000
1000
1000
1000
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
WIND
SPEED
3 knots
3 knots
3 knots
3 knots
3 knots
3 knots
5 knots
5 knots
7 knots
7 knots
5 knots
5 knots
5 knots
5 knots
7 knots
7 knots
10 knots
15 knots
15 knots
15 knots
15 knots
15 knots
15 knots
15 knots
-341-
-------
occurs, pollutants tend to travel farther from the source
before reaching the ground. For convenience, descriptions of
the stability classes based on EPA's Climatological Dispersion
Model (CDM) are listed as follows.
Stability Class
A Extremely unstable
B Moderately unstable
C Slightly unstable
D (day) Neutral (daytime)
D (night) Neutral to slightly
stable (nighttime)
E + F Stable to extremely
stable
There is a high probability that the stability classes in Table
6.1-3 can occur at other places in the U.S. at different times
of the year.
The 24-hour mixing depth (or height), wind speed, and
direction parameters are difficult to quantify in general terms;
for example, during summer, mean morning mixing depth decreases
from south to north, then from the central regions it decreases
past the mountain regions and then increases to the west coast.
It increases eastward and exhibits the highest value along the
Gulf Coast, as shown in Figures 6.1-2 and 6.1-3. The wind speed
decreases from south to north, then from the central regions
the wind speed alternately increases and decreases either west-
ward or eastward. In the afternoon, the mixing depth, wind
speed, and direction parameters change from the above behavior
to that of the isopleths in Figures 6.1-4 and 6.1-5.
-342-
-------
FIGURE 61-2 ISOPLETHS (m x 102) OF MEAN ANNUAL MORNING MIXING
HEIGHTS
FIGURE 6.1-3' ISOPLETHS (m sec'1) OF MEAN ANNUAL WIND SPEED AVERAGED
THROUGH THE MORNING MIXING LAYER
Source: (HO-049)
-343-
-------
14 12
FIGURE 61-4 ISOPLETHS (m x 102) OF MEAN ANNUAL AFTERNOON MIXING
HEIGHTS
9 9
FIGURE 6.1-5 ISOPLETHS (m sec-1) OF MEAN ANNUAL WIND SPEED
AVERAGED THROUGH THE AFTERNOON MIXING LAYER
.Source: (HO-049)
-344-
-------
Annual meteorological data consist of the relative
frequency of occurrence of the atmospheric stability classes
and wind direction and speed during the year. Figure 6.1-6
is an example of the annual meteorological data that might be
used for estimating the annual maximum concentrations of pollut-
ants from a petroleum refinery if it were located in Brazoria,
Texas. However, because the variations of the mixing height,
wind speed, and wind direction parameters from the west coast
towards the east coast, and from the northern regions towards
the southern areas of the U.S., do not follow a definite pattern,
it is difficult to characterize pollutant dispersion for all
areas in a specific manner. Appendix 6.1-4 gives a general
description of plume characteristics during various stability
conditions and meterological parameter changes.
In other words, both short-term and annual average
maximum concentrations estimated for a given site cannot be
described in general terms simply because of the complex nature
of the meteorological conditions which exist from region to
region. This also implies that the atmospheric dispersion models
must be exercised for a specific site with the appropriate
meteorological data.
Example Cases
Dispersion of Refinery Pollutant Emissions
Recent studies have involved the use of dispersion models
to predict maximum pollutant concentrations from petroleum refin-
eries. These emissions were then compared to primary and
secondary Federal standards and state and local ambient standards.
Comparison to standards of other states, where instructive, were
included, also. A description of the various atmospheric
dispersion models used are given in Appendix 6.1-5.
-345-
-------
w
ANNUAL PERCENTAGE OCCURENCE
OF STABILITY CLASSES
"A" STABILITY - .5%
"B" STABILITY - 3.9%
"C" STABILITY - 9.0%
"D(DAY)" STABILITY - 28.8%
"D(NIGHT)" STABILITY - 24.7%
"E & F" STABILITY - 33.1%
L=l
17.7%
S
11.455
FIGURE 6.1-6 ANNUAL WIND ROSE -
VICTORIA, TEXAS 1964-1973
-346-
-------
Basis for Examples
For the example case, it was assumed that the 300,000
bpcd reference refinery module described earlier in this report
is sited in Brazoria, Texas. Meteorological conditions in Table
6.1-3 and Figure 6.1-6 along with the source emissions inventory
shown in Table 6.1-2, were used as inputs to the previously
described atmospheric dispersion models to compute the expected
ground level pollutant concentrations.
Three schemes were considered in the analysis. For
Scheme 1, the annual, short-term, and 24-hour maximum concentra-
tions are computed using the data available for the 300,000
bpcd refinery module, and the results were compared with the
existing ambient air quality standards. For Scheme 2, the
annual, short-term and 24-hour maximum concentrations were
computed in a manner similar to Scheme 1, but with the assumptions
that the refinery module particulate emissions were increased by
50% and the resulting annual, short term, and 24-hour maximum
concentrations were increased by 50%, 25%, and 25%, respectively.
For Scheme 3, the annual, short-term, and 24-hour maximum concen-
trations were increased by 75%, 50% and 50% respectively. Schemes
2 and 3 were intended to represent gross estimates of maximum
concentrations that can occur at any site in the U.S., and to
indicate how these maximum concentrations compare with the
existing ambient air quality standards. A summary of the bases
for Schemes 1, 2, and 3 is given in Table 6.1-4. The following
assumptions apply to the three schemes:
A constant mean wind direction of 180
was assumed in making the 24-hour com-
putations. This is a "worst" case con-
sideration, meaning it will give rise to
the highest maximum concentration for any
set of meteorological conditions.
-347-
-------
TABLE 6.1-4
SCHEMES USED IN EVALUATING EXPECTED GROUND
LEVEL POLLUTANT CONCENTRATIONS
Scheme 1 Scheme 2 •Scheme 3
Particulate Emissions No Increase Increase by 5070 Increase by 50%
Annual Term Max. Cone, No Increase Increase by 50% Increase by 75%
Shprt; Term Max. Cone. No Increase Increase by 2570 Increase by 5070
24-Hour Term Max. Cone. No Increase Increase by 25% Increase by 50%
-348-
-------
For shorter averaging times, two sets of
conditions were used. The first corresponded
to an unstable atmosphere, giving rise to
maximum ground level concentrations due to
emissions from tall stacks. (Stability Class
A and a 5 knot wind speed were used for the
unstable condition). The second set corre-
sponded to a very stable atmosphere, and
giving rise to maximum ground level con-
centrations due to emissions of non-buoyant
material near the ground (fugitive hydro-
carbon losses). Stability class E and a
2 knot wind speed are used for the stable
condition.
The probability of the unstable condition
occurring is low. The second set of
conditions, however, represented a typical
nighttime condition. For comparison
purposes, a more typical daytime condition
(represented by stability class D and a
wind speed of 9 knots) was also considered.
Tables 6.1-5, 6.1-6, and 6.1-7 contain summaries of
the Federal and state ambient air quality standards and the
predicted maximum concentrations for the 300,000 bpcd reference
module emissions for Schemes ,1, 2, and 3, respectively, at full
production capacity. These predictions indicate that levels
of sulfur dioxide, particulate matter, nitrogen dioxide, and
carbon dioxide arising from the refinery emissions are well
below state (also see Appendix 6.1-1) and Federal ambient air
quality standards. The predicted three-hour hydrocarbon maximum
on all three schemes exceeds the Federal guidelines for hydro-
carbon levels under all meteorological conditions. These results
are typical for large refinery complexes (RA-119).
-349-
-------
TABLE 6.1-5
SCHEME
SUMMARY'OF FEDERAL AND STATE AMBIENT AIR QUALITY STANDARDS
AND PREDICTED MAXIMUM CONCENTRATIONS FOR-300.OOP BPCD REFINERY EMISSIONS
(Units are micrograms per cubic meter with ppm in parentheses)
Sulfur Oxides (5)
Annual Average (A.M.)
24-Hr. Maximum*
3-Hr. Maximum*
30-Minute Maximum
Particulate (c)
Annual Average (G.M.)V
24-Hr. Maxirnura*
5-Hr. Maxima
3-Hr. Maxixum
1-Kr. Maximum
N'itrogen Dioxide
! Ar-.r.ual Average (A.M.)
!-"on-Ke thane Hydrocarbons
3-Hr. Maximum* (6-9 a.m.)
Carbon Monoxide
6-Hr. Kaxirrum*
1-Kr. Maximum*
Federal
Primary
Standard
80(0.03)
365(0.14)
75
260
100(0.05)
160(0. 24)+
10000(9.)
40000(35.)
Federal
Secondary
Standard
1300(0.5)
60
150
Texas
S tandard**
1061(0.4)***
100
200
400
Computed
Maxiraum+4-
Annual Average
4.2(0.002)
0.7
5.3(0.003)
Computed
MaximumH-
24-Hour Average
26.5(0.01)
5.1
Computed Short-
Term MaximunrH-
Unstable Condition
135(0.051)
168(0.063)
23.0
25.5
31.8
1500(2.26)
9.6(0.01)
14.6(0.01)
Computed Short-
Terra Maximum*-*-
Stable Condition
t||
>36++*
>45"f^+
>64++
>7'f++
>84-f+
40000(60.26)
Compute,! Short-
Term Kaxio-.ucrr-h
Typical Cor.ditions
36.1(0.014)
44.9(0.017)
8.1
605(0. ?1)
4.4(0.004)
* Not to be exceeded more than once per year.
** Texas standards apply to individual sources and are not to be exceeded at any point at any time.
*** The rescission of this standard for new sources in Brazoria County (and certain other counties) is being considered by the Texas Air Control Board.
+ This standard is interpreted by the Texas Air Control Board and by EPA Co be a guideline and not a regulation.
Maximum values are those that occur on or outside the plant boundary.
"*~'~l"These maxima are beyond the computational range used.
(a)Scheme 1 is for the annual, short-term, and 24-hour computed maximum .concentrations.
(b)Arithmatic mean.
(c)Geometric mean.
-------
TABLE 6.1-6
SCHEME 2 ta)
SUMMARY OF FEDERAL AND STATE AMBIENT AIR QUALITY STANDARDS
AND PREDICTED MAXIMUM CONCENTRATIONS-300,OOP BPCD REFINERY EMISSIONS
(Units are micrograms per cubic meter with ppm in parentheses)
Sulfur Oxides «•> •
Annual Average (A.H.)<>D'
24-Hr. Maxir.um*
3-Hr. Maximum*
30-Xir.ute Maximum
Particulate .,„•.
Annual Average (C.M.)W
24-Hr. Maxir^jtr*
5-Hr. Max.i^uni
3-Hr. XaxiTum
1-Hr. Maxir.un
Nitrogen Dioxide
1 Annual Average (A.M.)
Hon-Mc-thane Hydrocarbons
3-Hr. Maxi-ur* (6-9 a.m.)
Ctrbon Monoxide
6-Hr. Xaxi^ur-.*
1-Hr. Maxirru-,*
Federal
Primary
S tandard
80(0.03)
365(0.14)
75
260
100(0.05)
160(0.24)"'"
10000(9.)
40000(35.)
Federal '
Secondary
Standard
1300(0.5)
60
150
Texas
Standard**
1061(0.4)***
100
200
400
Computed
Maximum-H-
Annual Average
6.3(0.003)
1.6
8.0(0.0045)
Computed
Maximum 1 1
24-Hour Average
33.1(0.012)
9.6
Computed Short-
Term Maximunri-f
Unstable Condition
>-,
169(0.064)
210(0.079)
43.1
47.8
59.6
1875(2.85)
12.0(0.012)
18.2(0.012)
Computed Short-
Term Kaxinrucrt-t-
Stable Condition
>«ttt
>56+++
>11.2ttt
>13.lttT
>15.0^
50000(75.3)
Cor.puteJ S!iort-
Terai KaxixurrH-
Typical Corditions
45.1(0.018)
56.1(0.021)
15.2
756.2(1.14)
6.9(0.005)
* Not co be exceeded more than once per year.
** Texas standards apply to individual sources and are not to b'e exceeded at any point at'any time.
*** The rescission of this standard for new sources in Brazoria County (and certain other counties) is being considered by the Texas Air Control Board.
This standard la interpreted by the Texas Air Control Board and by EPA to be a guideline and not a regulation.
Kaxieum values are those that occur on or outside the plant boundary.
11 a
These maxima are beyond the computational range used.
(a)Scheme 2 is the same as Scheme 1 except that particulate emissions were increased by 50% and the resulting annual, short-term, and 24-hour
maximum concentrations were increased by 50%, 25%, and 25% respectively.
(b)Arithmetic mean.
(c)Geometric mean.
GO
-------
TABLE 6.1-7
SCHEME
SUMMARY OF FEDERAL AilD STATE AMKTF.NT AIR QUALITY STANDARDS
AKD PREDICTED MAXIMUM CONCENTRATIONS FOR 300, OOP BPCD REFINERY EMISSIONS
(Units are micrograms per cubic meter with ppm in parentheses)
Sulfur Oxides f} \
Annual Average (A.M.)*- ;
2-i-Hr. y.axi:r.u!n*
3-Kr. Maximum*
20-Minute Maximum
Particulate /ni
Annual Average (C.M.) V'
24-Hr. Naxir-uir.*
5-Hr. Kaxiirura
3-Hr. Maxi-rum
1 1-Hr. Maxi-u^
Nitrogen Dioxicle
A-r.ual Average (A.M.)
Kon-Ke thane Hydrocarbons
3-Hr. >!axi--uTi* (6-9 a.m.)
Carbon Monoxide
8-Hr. Xaxir^urr.*
1-Hr. Xaxi=u=i*
Federal
Primary
Standard
80(0.03)
365(0.14)
75
260
100(0.05)
160(0. 24)+
10000(9.)
40000(35.)
Federal
Secondary
Standard
1300(0.5)
60
150
Texas
Standard**
1061(0.4)***
100
200
400
Computed
KaxiniutrrH-
Annual Average
7.35(0.0035)
1.84
9.3(0.005)
Computed
Maximum-H-
24-Hour Average
39.8(0.015)
11.5
Computed Short-
Term Maximum-H-
Unstable Condition
203(0.077)
252(0.094)
51.8
57.4
71.6
2250(3.39)
14.4(0.015)
21.9(0.015)
Computed Short-
Term Maxi^U!Tri-t-
Stable Condition
_t_ i i
>54+++
>68+4H"
IP
60000(90.4)
ComputeJ S'.iort-
Tenr. Kaxiiruei-i-
Tvp-ical Conditions
54.2(0.021)
67.4(0.026)
18.2
.
907.5(1.37)
6.6(0.006)
* Not to be exceeded more than once per year.
** Texas standards apply to individual sources and are not to be exceeded at any point at'any time.
*** The rescission of this standard for new sources in Brazoria County (and certain other counties) is being considered by the Texas Air Control Board.
"*" This standard is interpreted by the Texas Air Control Board and by EPA to be a guideline and not a regulation.
"*""*" Kaxiaua values ara those that occur on or outside the plant boundary.
1 ' 'These maxima beyonrf the computational ran.c.o used.
(a)Schenw 3 is the same as SHio.me 1, except, that parti culafe emissions wore increased by 50% and the resulting annual, short-term, and 24-hour maximum
concentrations were increased by 75%, 50%, and 50%.
(b)Arithmatic mean.
(c)Geometric mean.
(.A)
Ul
K>
I
-------
The results of the computations demonstrate that the
300,000 bpcd gasoline refinery has little impact on SOa, particu-
late, NO , and CO ambient concentrations. Therefore, it should
J\.
be expected that both the 200,000 bpcd fuel oil and gasoline
refinery modules that were considered in this study will have a
similar minimal impact. The estimated hydrocarbon concentrations
exceed the Federal guideline significantly. A similar result
is expected for the subject fuel oil and gasoline refineries.
-353-
-------
6.1.1.2 Water Quality Impact
Aqueous Effluents
Wastewater will be generated at multiple sources in
the refinery module. Table 6.1-8 identifies the major sources
in the plant and the species of pollutants present in each.
As noted on the table, the various effluents are categorized
as process wastes, cooling tower blowdown, or auxiliary refinery
systems and wastes. The primary contaminants present in the
refinery's wastes include sulfides, ammonia, phenols, oil,
dissolved and suspended solids, biochemical oxygen demand (BOD),
and chemical oxygen demand (COD). Under existing Federal
limitations, discharge of the above wastes will not be permitted,
Water management can exercise a number of strategies
through direct implementation of the wastewater treatment
processes. There are four types of applicable wastewater
treatments: in-plant; primary; secondary; and tertiary. The
degree to which each of these processes is utilized depends
on the local area discharge regulations, the quality of waste-
water effluents prior to treatment, and the degree of recycle
or reuse of water. A summary of pollutant types contained in
the refinery module wastewater streams is presented in Table
6.1-9.
-354-
-------
TABLE 6.1-8
i
CO
Ol
Ui
I
SUMMARY OF REFINERY WASTEWATER EFFLUENTS AND APPLICABLE TREATMENTS
Malor Pollutant Present
A.
B
C.
Waste Source H»S NHi Phenols
Process Wastes
1. Crude Desalting . • . X
2. Atmospheric' Distillation XX X
3. Pentane Deasphaltir.g
4. Eeasphalted Oil KDS X X
5. Partial Oxidation X X
6. Kydrocracking X X
7. Fluid Catalytic Cracking XX X
8. K? Alkylation X . X
9. Sulfur Recovery ' X
(Tail Gcs Treating Unit)
Refinery Cooling Systett*
(coolir-s tower)
Auxiliary Refinery Systems
ar.d Wastes
1. Ste£~ Generation
3. Sir.itary V.'istas13
4. Precipitation Runoffc
5. Water Treatment Wastes
6. Miscellaneous Operations
Dissolved
Oil Solids
X X
X
X
X
X X
X
X
X
X
X
x
X
X
X X
Suspended
Solids BOD
X
X
•X
X
X
X •
X
X
X X
X X
X
X X
Aoolicable
Primary
COD In-Plantd Treatr.ent
X X
X XX
X X
XX X
XX X
XX X
XX X
X ' X X
X
x x
X
x x
Treatments
Secondary
Treatr.ent
X
X
X
X
X
X
X
X
X
X
Tertiary
Treatment
x
x
X
x
X
X
aCcolir.^ tcwer aperstir.s at 20 cycles of concentration.
~3.epreser.cs refinery sanitary waste, only.
clncluic* runoff frcrr. process end tank fina areas, only.
Rafer* to sour wacar »cripp«r.
SOURCE: (BE-147, AM-041, RA-181)
-------
TABLE 6.1-9
APPROXIMATE COMPOSITION
AQUEOUS EFFLUENT 200
CONSTITUENT
,000 BPCD FUEL OIL
REFINERY MODULE
OF REFINERY EFFLUENT AT THE OUTFALL
200,000 BPCD GASOLINE
REFINERY MODULE
300,000 BPCD GASOLINE
REFINERY
BOD , ppm
COD, ppm
Ammonia , ppm
H2S,ppm
Total Phosphorous
(as P0i») ,ppm
Phenols ,ppm
Oil and Grease, ppm
Suspended Solids, ppm
Dissolved Solids, ppm
Total Nitrogen
(as NO 3) ,ppm
Mer cap tans , ppm
Flow, million gallons/day
*Cooling tower blowdown.
SOURCE: (RA-181)
Ln
1
15
80
2
0.1
2
0.1
2
10
370
-
-
3
15
80
2
0.1
2
0.1
2
10
370
-
-
3
<20
<80
< 2
0.
190
0.
2
10
15,000
230
None
0.
1
1
25*
-------
The wastewater treatment processes assumed for the
200,000 bpcd fuel oil and gasoline refinery modules consist
of in-plant waste treatment, primary waste treatment, and a
high-efficiency secondary waste treatment (such as activated
sludge). The 300,000 bpcd gasoline refinery reference module
has the same water treating facilities plus a tertiary treatment
unit. Also, the aqueous effluents of the reference module are
treated and circulated to the cooling tower, as shown in Figure
6.1-7. In-plant, primary, and secondary treatments are utilized
to meet existing effluent limitations. Tertiary treatment, in
the form of water desalination processes, is typical of facilities
that will be required to meet the minimal effluent limitations
proposed for 1985.
The comparatively large concentrations of total phos-
phorus, dissolved solids, and total nitrogen in the 300,000
bpcd reference refinery as shown in Table 6.1-9 are due to the
recycle of treated wastewater back into the refinery cooling
tower. The resulting effluent stream consists entirely of the
cooling tower blowdown and has much less volumetric flow than
the effluent from the 200,000 bpcd gasoline refinery module which
does not practice the recirculation technique.
Effluents from the three refinery modules are assumed
to have the composition shown in Table 6.1-9; the actual com-
position of wastewater discharged to the receiving stream may
differ somewhat from that shown, depending on final process
configuration and design. This difference is not anticipated
to be great and no significant change in 'the water impact
assessment is expected.
-357-
-------
Ul
00
OILY WASTES <
OILY WASTES
CONTAMINATED <
WITH K S
SEWAGE J
1
SOUR j
COXDEN'SATE
NON-OILY
W4STFS 1
WASTE WATER
SOURCES
"PRECIPITATION RUNOFF
CRUDE DESALTER ,
PEKTANE DEASPHALT ,
MISC. OPERATIONS
-
["ATM DISTILLATION
DEASFHALTED OIL HDS •
PARTIAL OXIDATION ,
HYDROCRACKING ,
FLUID CAT. CRACKING ,
HF ALKYLATION
'SANITARY WASTES
i
r SULFUR RECOVERY
S^EfW GEN^RA^IO^
POWER GENERATION ,
PRECIPITATION RUNOFF »
LIKE SOFTENER
SODIUM ZEOLITE UNIT
ION EXCHANGE UNIT
'
IN- PLANT
TREATMENT
SOUR WATER
STRIPPER
SOUR WATER
.
PRIMARY
TREATMENT
FLOTATION
SOLIDS
DEWATERING
NEUTRALIZATION
*
J
SECONDARY • TERTIARY I REFINERY
TREATMENT TREATMENT 1 EFFLUENT
BIOLOGICAL
OXIDATION ~"
COOLING
TOWER
BLOkUOirJN
TOWER 1
1
1
t.
TD,?^TED ^_ rDtSALI^ATION !
(FOR REUSE) T_^UUKS_5 j
t
1 TREATMENT
1 WASTES
FIGURE 6.1-7 PROPOSED WATER MANAGEMENT PLAN
-------
By comparison, the "reference" module effluent has
a much higher concentration of total phosphorous (POO and
dissolved solids, though its flow rate is just 1/12 of the
discharge rate of either of the other two refinery modules in this
study. Furthermore, the "reference" module effluent contains
detectable amounts of nitrogen (N03). The main differences
can be attributed to the fact that the "reference" module is
much larger and that it circulates the treated water to the
cooling tower. This recycle and reuse of water gives rise to
the higer content of dissolved solids in these effluent. There-
fore, to promote water use efficiency, the EPA has placed no
restriction on total dissolved solids in either its effluent
limitation guidelines or new source performance standards.
The oxygen-demanding material in the effluent (repre-
sented by COD, BOD, ammonia, and sulfide) are probable maximum
values, in that an indeterminant amount of oxidation will occur
upon aeration in the cooling tower. The concentrations shown
are conservative-case estimates for the blowdown. In the case
of the two refinery modules, the effluent discharge rate is
estimated at 15 gallons per barrel of oil throughput.
Water Quality Modeling
The water quality impact of industrial pollutants
is usually directly related to the concentrations in which they
are present in receiving waters. It is, therefore, necessary
to determine the pollutant concentrations in waters receiving
refinery effluent discharges. Once the effluent concentrations
are known, they can be compared to pre-existing ambient levels, to
levels known to be harmful, or to government regulations to
assess their overall water quality impact..
-359-
-------
Mathematical models are useful tools in computing the
pollutant concentrations in receiving waters which result from
predicted refinery effluent discharges. Two levels of complexity
are exhibited by the models. A very simple model is used to
predict pollutant concentrations. This model gives background
concentrations that are some distance from the effluent outfall
and applies to well mixed streams. The other model, in addition
to giving background concentrations, computes constant-concen-
tration contours - extending from effluent outfall to the region
of background concentrations. This model is especially useful
in showing pollutant distributions in tidal waters during con-
ditions of flood, slack, and ebb tides. A detailed discussion
of the two types of models is presented in Appendix 6.1-2.
The data required for modeling includes the effluent
pollutant concentrations and effluent flow rates predicted for
a refinery module; and flow rates, tidal conditions, and config-
uration for the receiving water body. The behavior of water
bodies varies widely throughout the U.S. In order to obtain
meaningful computed pollutant concentrations in the receiving
water, it is important to define as accurately as possible the
point of condition of the waters receiving the refinery effluent
at the release.
Dispersion of Refinery Effluents
The dispersion of the effluents from the "reference"
refinery are considered for example purposes. In this case,
the effluents are discharged into a river at a point some twenty
miles from its mouth on the Texas Gulf Coast. The river is in-
fluenced by diurnal tides. The bottom of the river is below
mean sea level, and is subject to saline water incursion. Both
the tidal effect and the incursion are more significant at low
stream flow conditions.
-360-
-------
The discharge of the effluent increases the total
dissolved solids content of the river by about 70 ppm at median
flow conditions. This is a nominal increase of one percent. The
dispersion of non-conserved pollutants can be illustrated by the
calculate effect on dissolved oxygen (DO) concentration shown
in Figure 6.1-8. The calculation is made on the seven day -
ten year low flow condition. The localized DO depression is
five percent of the saturation value.
These are the results of one example. Another study
(RA-119) considered the siting of large new refineries at five
U.S. Coastal sites. In all cases, very little impact was pre-
dicted. With full implementation of presently available tech-
nology, refineries can be designed for minimal impact on re-
ceiving waters.
6.1.1.3 Solid Wastes
The solid wastes associated with the refinery processing
have been identified in Sections 3.1.3 and 3.2.3. The disposal
of these solid wastes is the primary concern in this section.
Refinery-generated solid wastes have been disposed of in the
past by either the landfill method or the sea disposal technique.
Disposal by sanitary landfill is applicable for all
general plant waste as received, with the exception perhaps of
water treatment plant sludges1. Land disposal of oily sludges
and emulsions, by mixing at depths up to 6 inches, has been
practiced for several years (BE-228). The nature and the
sizeable volume of solid wastes for land disposal is of great
concern. Sea disposal of solid wastes having high specific
gravities is being practiced (BE-228). There are several approved
disposal sites, but all of them forbid the disposal of oil and
other floatable materials. Although disposal at sea is being
used, it has become less acceptable on a long range basis.
-361-
-------
F IC U R F. 6.1-8
DISSOLVED OXYGEN CONCENTRATION
PROFILE.
RIVER FLOW - 4.09 CFS
T - 27° C
CSAT " 7-
z:
o
t-
•z.
U
o
•z.
o
o
o
Q
7.70
7.60
7.66
7.64
7.62
7-40.
7.46
7.44.
7.42.
7.40
40.4
18.C
0.2
-362-
(OUTFALL)
RIVER MILES
-------
Incineration of solid wastes, while acceptable, is
not an ultimate disposal method. The disposal of the incinerated
solid waste and their effects on petroleum refinery siting are
discussed in this section.
An anticipated on-site incineration unit for oily
solids and other organic sludges, as well as for much of the
office, shop, and other non-processing refuse, will reduce
considerably the volume of material to be disposed of. The
resulting incinerator ash, in addition to discarded catalyst,
will require some type of ultimate disposal.
The hydrology and geology of the site are basic con-
siderations for solid waste disposal sites. Ideally, geological
materials at the site must possess the necessary impermeability
characteristics, with hydraulic conductivities below 10 cm/sec.
Also, water-transmitting material and groundwater levels must
not be encountered at any depth.
Another prerequisite in selection of a solid waste
disposal site is assurance that the site will not be flooded.
(This denies a potential source of water to form leachate, and
also avoids possible erosion of the cover material.) The site
should be located within the flood-protection levee of the re-
finery if within general proximity of a flood zone.
In general, all problems pertaining to hydrology,
geology, and potential flooding nature of a dump site, covering
periods both during and after the operational life of the refinery ,
must be studied as part of any refinery siting program.
-363-
-------
6.1.2 Raw Material Availability
Raw material availability presents a unique set of
environmental impact problems and problems associated with
petroleum refinery siting. Raw material refers to both feed-
stock and water. Their implications in this study are discussed
in the following sections.
6.1.2.1 Feedstock Availability
Feedstock for petroleum refineries may come from
either foreign of domestic sources. In 1973, the United States
crude oil and lease condensation production averaged 9.2 million
barrels per day, and 3.2 million barrels of crude per day were
imported (AM-099). These crude oil supplies served as feedstocks
to the 247 operational petroleum refineries in the U.S. (01-008).
The feedstocks reached the refineries through various means of
transportation. Thus, feedstock availability is related to the
mode of transportation, and its relationship to refinery siting
is discussed below.
Foreign Source
Feedstock from abroad creates some environmental
impact problems at U.S. port facilities. The chance of major
oil spills is one problem. Tanker sizes are in some measure
related to the problem.
Proposed petroleum refineries that will receive their
crude oil supply from abroad are likely to be sited near coastal
waters. As for the economics of feedstock, previous studies
indicate that larger tankers carrying vast amounts of crude oil
offer cost advantages (US-124). Tankers being considered are in
the very large crude carrier (VLCC) class, i.e., 250,000 DWT
-364-
-------
and ultra large crude carrier (ULCC) class, i.e., about 400,000
DWT. These tankers cannot be received by the existing U.S. ports
as indicated in Table 6.1-10. Figure 6.1-9 shows the geographical
location of selected U.S. ports. Thus, a problem exists for
imported feedstock. However, this problem can be resolved
through the development of strategic superport facilities,
such as offshore deepwater-terminals or the dredging of harbor
channels. Figure 6.1-9 also indicates the regional location of
potential ports for supertankers. This development of super-
port facilities will have a potential environmental impact.
Of the two superport facilities mentioned above, the construction
of offshore deepwater-terminals would have a less adverse
environmental impact than dredging or the use of existing port
facilities (PR-074). The President's Energy Message on April
18, 1973, stated (PR-074).
"The environmental advantage of offshore
deepwater ports is that they reduce the
risks of collision and grounding and
minimize the probability that spilled
oil will reach beaches or estuaries.
The most valid environmental concern
involves the impact of primary and
secondary economic development, such
as refineries and petrochemical plants,
associated with the port. These risks
are recognized and can be controlled
through land use planning and adequate
local zoning. Dispersion of facilities
versus concentration with only a few
ports would probably significantly
reduce the environmental impact on any
particular region".'
The 1972 Council on Environmental Quality (CEQ) made preliminary
assessments on the probable environmental impact of the opera-
tion of ports for supertankers and they are discussed in
Reference US-124.
-365-
-------
TABLE 6.1-10
SELECTED U.S. PORTS HANDLING SIGNIFICANT
AMOUNTS OF BULK CARGO
EAST COAST
Delaware River ports
Hampton Roads, Va.
New York, -N.Y.
Portland, Me.
Baltimore, Md.
Boston, Mass.
GULF COAST
New Orleans, La.
Tampa, Fla.
Baton Rouge, La.
Mobile, Ala.
Corpus Christi, Tex.
Houston, Tex.
Brownsville, Tex.
Pascagoula, Miss.
PACIFIC COAST
Long Beach, Calif.
Los Angeles, Calif.
San Fran. Bay ports, Calif.
Seattle, Wash.
GREAT LAKES
Chicago, 111.
Indiana Harbor, Inc.
Detroit, Mich.
Duluth/Superior, Minn./Wis.
Buffalo, N.Y.
Ashtabula, Ohio
Cleveland, Ohio
Conneaut, Ohio
Toledo, Ohio
Source: (US-124)
Controlling
Depth
(feet)
40
45
35
45
42
40
40
34
40
40
45
40
36
38
52
51
35
73
28
29
29
32
29
29
29
27
28
Est. Maximum Permissible Vessel
Size When Fully Loaded (dwt.)
53,000
80,000
40,000
80,000
53,000
40,000
50,000
35,000
50,000
45,000
50,000
50,000
30,000
35,000
150,000
150,000
40,000
250,000
10,000
12,000
12,000
12,000
12,000
12,000
15,000
13,000
12,000
-366-
-------
CRUDE OIL PIPELINES
u>
LEGEND
CRUDE OR PRODUCING AREA
REFINING AREA
PLANNED OR UNDER CONSTRUCTION
BULK CARGO TERMINAL
-fr REGIONAL LOCATION OF THE POTENTIAL PORTS FOR SUPERTANKERS (US-124)
FREEPORT. TEXES (2 SITES)
MISSISSIPPI DELTA, LOUISIANA (2 SITES) NEW YORK BIGHT. NEW YORK-NEW JERSEY (2 SITES)
'OUTSIDE DELAWARE BAY (2 SITES) RARITAN BAY. NEW JERSEY
DELAWARE BAY (INSIDE) MACHIAS BAY. MAINE
FIGURE 6.1-9
-------
The environmental impact associated with imported
feedstock is oil spills. Sources of oil spills are ports
and tankers. Oil spill statistics, given in Table 6.1-11,
explicitly indicate that terminals and ships are the two major
contributors to water pollution. A study on the probability
of large oil spills, estimated by vessel class, provides
some insight into the environmental merit of small versus
large tankers (US-124). The result of the study points out
that as the tanker size increases, the probability of large
spill decreases, mainly because large tankers have a lesser
number of port calls per year .
Domestic Source
Current domestic crude production is essentially all
designated as part of the feedstock to existing refineries.
New refineries depending upon domestic crude would necessarily
be related to new crude production. If this new production is
from new producing areas, new transportation facilities will be
needed. The problems and environmental impact related to
domestic crude sources are mainly charged to the mode of trans-
portation elected to move the feedstock to the refineries.
Movement of crude can be accomplished by pipelines, tank barges,
or rail tank cars.
The major environmental concern in crude oil trans-
portation is oil spills. In 1972, the Department of Transporta-
tion and U.S. Coast Guard reported that of the 18.8 million
gallons of oil spilled on waters, 19.970 came from tank barges
and 6.6% came from pipelines (US-159). Reports on rail tank
cars oil spills are sparse. Oil spills on land will eventually
reach water bodies in the form of nondegradable organics (HI-090),
which amount to about 570 of the total quantity of oil spilled.
-368-
-------
TABLE 6.1-11
OIL SPILL STATISTICS (BARRELS)
Type of Spill
1971
1972
Petroleum Industry Related Spills
Terminal
. Number
Volume
Ships (offshore)
Number
Volume
Offshore Production Facilities
Number
Volume
Onshore Pipeline
Number
Volume
Total
Number
Volume
All Spills
Number
Volume
1,475
125,800
22
400
2,452
15,600
74
8,700
4,023
150,500
7,461
205,000
1,632
54,700
32
51,600
2,252
5,700
. 162
29,300
4,078
141,300
8,287
518,000
Source: (FE-076) and
The Massachusetts Institute of Technology Department of
Ocean Engineering, 1974, "Analysis of Oil Spill Statistics",
prepared for the Council on Environmental Quality under
Contract No. EQC330, using U.S. Coast Guard data.
-369-
-------
6.1.2.2 Water Availability
Water is essential to almost every facet of energy
conversion processes. The extraction of fuel resources, fuel
preparation, transportation of fuels, utilization of fuels to
generate energy, and disposal of waste products in an environ-
mentally acceptable manner involve water. The need for water
varies with the source of energy, region of development, and
degree of complexity of environmental control.
Table 6.1-12 gives the major uses of water for various
forms of energy processes. Note in the table that refineries'
water requirement is 43 gallons per barrel or 7.58 gallons per
million Btu's, mainly as process and cooling water. For example,
a 300,000 bpcd gasoline refinery will require as much as 13
million gallons of water per day for its operation. This is a
substantial quantity of water. It is a critical factor in
selecting the refinery site.
In general, the Mississippi River divides the relatively
humid east coast from the more arid western U.S. Some western
areas (notably the northwest) have heavy rainfall, but by and
large, the western central states are a relatively arid region.
Figure 6.1-10 shows the relative water abundance or deficit
across the U.S. (US-083). Water abundance means that annual
rainfall exceeds evaporation losses. For deficits, evaporation
losses exceed rainfall.
Figure 6.1-11 gives a general impression of the sizes
and locations of major rivers in the U.S. The western central
states do not have an abundant, large river water supply.
Production facilities in these areas are presently being con-
sidered that would require water at rates exceeding reliable
natural water supply at the facility site. Therefore, elaborate
schemes are being studied for bringing the water to the facilities.
-370-
-------
TABLE 6.1-12
WATER USED FOR ENERGY
(Source: FE-076)
Energy Source
Western coal
mining
Eastern surface
mining
Eastern surface
.11 in ing
Oil shale
Coal gasification
' Coal liquefaction
i Nuclear
Oil and gas
production
Refineries
Fossil fuel
power plants
Gas processing
Standard
Unit
ton
ton
ton
barrel
MSCF
barrel
Kwh
barrel
barrel
Kwh
MSCF
Consumption Demand
For Water.
6-14.7 gal/ton
15. 8-18.0 gal/ton
145.4 gal/bbl
72-158 gal/MSCF
175 - 1,134 gal/bbl
0.80 gal/Kwh
17.3 gal/bbl
43 gal/bbl.
0.41 gal/Kwh
1.67 gal/MSCF
Water Needed
Gal/106 BTU
0.25 - 0.61
0.66 - 0.75
30.1
72 - 158
31 - 200
234.46
3.05
7.58
120.16
1.67
Major Uses
Of Water
Dust Control
Coal Washing
Dust Control
Coal Washing
Dust Control
Coal Washing
Mining, cooling,
oil shale disposal
preparation
Process use
Cooking use
Process use
Cooking use
Cooling, uranium minir;g
Well drilling, secondary
and tetiary recovery
Process H20
Cooling H20
Cooling H20
Cooling HjO
plants
-------
ATER SURPLUS OR DEFICIENCY
INCHES
E3SJ PO TO >80
0 TO 20
0 TO -20
T04-40
FIGURE 6.1-10
RELATIVE WATER ABUNDANCE OR
DEFICIT ACROSS THE U.S.
-------
u>
Fiovvs of Large Rivers
AVERAGE ROW !«. li./sec.)
I 20.000
1 50.000
j 100.000
_•_ 250.000
' •• 500.0CO
FIGURE 6.1-11 LARGE RIVERS OF THE UNITED STATES
-------
Water laws and regulations are also important factors
in refinery siting. The Federal Energy Administration's (FEA)
Project Independence Report (FE-076) indicates that the avail-
ability of water in any area is governed partly by Federal actions,
but more importantly by physical conditions and by state and
local prerogatives. Also, the report stresses four factors
that determine availability:
(1) Runoff - Some regions have inadequate
rainfall and runoff to meet the demands
of all water users.
(2) Institutional Factors - Federal and
state laws, Indian water rights,
interstate compacts, and international
treaties govern the allocation of
water to the different users.
(3) Environmental Considerations - The
Federal Water Pollution Control Act
Amendments, regulating thermal pol-
lution, sedimentation and acid run-
off from strip mining, increases in
salinity, salt water intrusion, and
coastal water quality affect water
availability.
(4) Capital Investment and Repayment -
Construction of water supply projects for
energy activities may be impeded by debt
limitations and failures in authorizing
bond issues.
The second factor is of direct consequence in this report.
-374-
-------
The major constraints in water use in some regions of
the U.S. are primarily political. There are legal agreements
between the Federal government, the individual states, and the
river basin authorities which interplay throughout the whole
water availability question. A brief summary of the major laws,
namely the Law of River Compacts, State Water Laws, and Federal
Water Laws are discussed in Appendix 6.1-3.
6.1.3 Product Transportation
In siting a petroleum refinery, it is important to
consider the facilities that are available for reaching the
market. These facilities fall in the area of product trans-
portation. There are three major modes of product transporta-
tion; railroad, waterway, and pipeline.
Railroad
The environmental impact of rail transportation for
petroleum products can be charged to air emissions (mainly
hydrocarbon) and liquid product spills. These constitute
relative small factors in refinery, SNG, or LNG siting con-
siderations.
Waterways
In this country, domestic waterway systems include
barge movements on the inland and intercoastal waterways and
ship movements on the Great Lakes and on the oceans. Figure
6.1-12 shows the U.S. waterway system. Nearly half of all
petroleum and petroleum products are transported by means of this
waterway system. Environmental consequences from water trans-
portation system are attributed to spills. Statistics for 1972 on
polluting incidents in and around U.S. waters indicate that
-375-
-------
WATERWAYS OF THE UNITED STATES
(SOURCE: FE-076)
--J
cr\
i
KAVIWBU LENCTH5 AND DEPTHS
OF UNITED STATES WATERWAY ROUTES
Reprinted wifri pemiission of
American Waterways Operators, he.
FIGURE 6.1-12
-------
light petroleum products were responsible for 3570 of the total
spills while heavy oils accounted for 970 (US-159) .
Pipelines
The U.S. oil pipeline network consisted of 170,000
miles by the end of 1973 (IN-047), with 38% dedicated to product
transport and the remaining 62% to crude gathering and trunk
lines. The locations of the product lines in the U.S. are shown
in Figure 6.1-13. These pipelines interconnect existing petroleum
refineries and the major markets. New refineries will require
new feedstock and product pipelines. The main environmental
concern with petroleum product transportation by pipeline is
spills. The impacts of these spills on soil are not currently
well defined.
-377-
-------
PRODUCT PIPELINES
(SOURCE: FE-076)
(jO
^J
00
LEGEND
PLANNED OR UNDER CONSTRUCTION
EXISTING PIPELINES
FIGURE 6.1-13
-------
6.2 SNG Plant Impact
There is limited production of SNG in the U.S. In the
U.S. today a principal technique of producing SNG is through
gasification of naphtha and lighter petroleum fractions. Gasi-
fication of other feestocks such as middle distillates, gas
oil, crude oil and coal, can also produce SNG.
The design, construction, and operation of SNG plants
must comply with Federal environmental legislation and regulations
concerning air quality (Air Quality Act, 42 U.S.C.A. 1857, 40
C.F.R. 50, et seq.), and with the Federal Water Pollution Act
of 1972, the Solid Waste Disposal Act, ,*ud the Rivers and Harbors
Act of 1899 (The Refuse Act).
This section deals with the example of siting naphtha-
based SNG plants of the 125 MM scfd capacity class, operating with
the Catalytic Rich Gas (CRG) process. The CRG process is dis-
cussed in Section 2.3. Environmental impacts and problems asso-
ciated with SNG plant siting are described in terms of effluents
(air and water emissions), raw materials (feedstock and water
requirements), product transportation, and the existing laws
(Federal, state, and local) that interact with these parameters.
The approach to the analysis is similar to the scheme that was
used in the discussion of petroleum refinery siting presented
in the preceding section.
6.2.1 SNG Plant Effluents
Effluents of main concern from an SNG plant employing
the CRG process are air emissions and liquid effluents. There
are essentially no solid wastes from an SNG plant with this type
of process other than spent catalysts (FE-084). The spent cata-
lysts can be returned to the catalyst vendors for reclaiming of
valuable metals.
-379-
-------
Environmental effects of air emissions and liquid
effluents from an SNG plant are assessed from the standpoint of
ambient air quality and water quality.
6.2.1.1 Air Emissions Impact
The major air emissions from SNG plants are particulates,
sulfur dioxide (SO ), hydrocarbons (HC), carbon monoxide (CO),
S ,
and nitrogen oxides (NO ). Estimated air emissions from the 125
X
MM scfd SNG plant module described in Section 3.4 are summarized
in Table 6.2-1 along with the air emissions of the 300,000 bpcd .
refinery "reference" module. Table 6 . 2-1 explicitly indicates
that SNG plant air emissions are by far smaller than those of the
refinery "reference" module, by factors ranging from 9 to 240.
In order to quantify the impact of the air emissions
on the ambient air quality of a site, it is necessary to compute
the downwind maximum concentrations of the pollutants for a
given set of data. This data would include stack parameters,
plant configuration, and meteorological data of the site. If
this information is available, the maximum concentrations can
easily be computed by employing air dispersion models. The impact
of the SNG plant on the ambient air quality of a site can be
described through logical comparative analysis with the impact
of the refinery "reference" module which was discussed in Section
6.1.1.1.
To establish a meaningful comparison, it must be assumed
that the stack parameters, basic plant configuration, and meteoro-
logical data that were used for the refinery "reference" module
analysis also apply to the SNG plant module. In addition,
Schemes 1-3 that were discussed and summarized in Table 6.1-4
-380-
-------
TABLE 6.2-1
AIR EMISSIONS - SNG MODULE COMPARISONS
oo
Pollutant
Particulates, Ib/hr
Sulfur Dioxide (S0a ) , Ib/hr
Hydrocarbons (HC) , Ib/hr
Carbon Monoxide (CO) , Ib/hr
Nitrogen Oxides (NO ), Ib/hr
X
125 MM scfd SNG
Plant Module
20
8
677
12
217
300,000 bpcd Gasoline
Refinery (Reference)
353
1918
6418
138
1846
-------
must also hold for the present case. Summaries of Federal and
state ambient air quality standards and predicted maximum con-
centrations for the 300,000 bpcd refinery "reference" module
shown on Tables 6.1-5, 6, and 7 are applicable in this comparison.
The trend of the scheme is towards the "worst" and most
conservative estimate, i.e. Scheme 3. As seen in these tables,
the computed maximum concentrations of the refinery "reference"
module, which has much higher air emissions than the SNG plant,
are below the Federal ambient air quality standards for particu-
lates, S02, N0x, and CO that will be released by a 125 MM scfd
SNG plant. No attempt has been made to characterize hydrocarbon
emissions. ~~
The SNG plant HC emission given in Table 6.2-1 is
smaller than that of the refinery'"reference" module by a factor
of 9. Therefore, if all the conditions explained in the previous
discussions hold, and if.a decrease of HC emission by a factor of
9 would reduce the computed maximum concentration of the pollutant
by the same factor, then the results for the three schemes would
be those given in Table 6.2-2. The computed short-term HC maximum
concentration for typical meteorological conditions would be
below the EPA 3-hour non-methane hydrocarbon guideline. The
computed short term HC maximum concentrations for unstable and
stable meteorological conditions exceed the EPA guideline.
Water Quality Impact
SNG plant modules based on naphtha feed have minimal
liquid effluents. M. W. Kellogg's report for EPA on a 150 MM
scfd SNG plant (KE-129), addresses the sources of liquid ef-
fluents, viz., cooling tower blowdown, boiler blowdown, and
backwash water from water treating process. Liquid effluents
of this type consist mainly of totally dissolved solids (TDS)
-382-
-------
TABLE 6.2-2
3-HOUR NON-METHANE HYDROCARBONS
(UNITS ARE ug/m3 WITH ppm IN PARENTHESES)
EPA Guideline
Computed Short-
Term Maximum"""
Unstable Condition
Computed Short-
Term Maximum
Stable Condition
Computed Short-
Term Maximum"""
Typical Conditions
Scheme 1
67(0.10)
Scheme 2
84(0.13)
Scheme 3
160(0.24)
167(0.25)
160(0.24)
208(0.32)
160(0.24)
250(0.38)
4444(6.70) 5556(8.37) 6667(10.04)
101(0.15)
Maximum values are those that occur on or outside the plant
boundary.
-383-
-------
of about 37,000 ppm. The boiler blowdown, cooling tower blow-
down, and backwash water from water treating process contributed
4%, 5%, and 91% of the TDS, respectively.
EPA has placed no restriction on total dissolved solids
in its effluent limitation guidelines and new source performance
standards. Therefore liquid effluents from an SNG plant cannot
be categorized as pollutants.
Raw Material Availability
Raw material pertains to both feedstock and water.
Their implications for the current study are presented separately.
Feedstock Availability
Naphtha is the basic feedstock being considered for
the SNG plants. Other feedstocks such as ethane, LPG, middle
distillates, gas oil, and crude oil can be used by SNG plants,
but their availability under the FEA's mandatory oil-allocation
program (FE-085) is not certain.
Naphtha is chiefly a domestic product and can be trans-
ported via product pipelines, waterways, or railroads. Typically
a 125 MM scfd SNG plant will need about 25,000 barrels of naphtha
per day (LI-095). Thus, it is rather important for a proposed
SNG plant of this capacity to be sited along the feedstock routes.
As mentioned in the previous section, the existing waterways and
railroad facilities can be used for the transportation of naphtha.
Movements of naphtha through both modes of transportation generally
belong to the private sector with little Federal regulation other
than for safety and environmental reasons. The main environmental
concern with these modes of transportation, especially tank barges
through waterways is the potential hazard of naphtha spills.
-384-
-------
Proposed SNG plants may require new feedstock pipeline
routes. The ownership and operation of the oil product pipeline
system is generally private and has little Federal regulation
other than for safety and environmental reasons. New pipeline
routes will create some environmental effects during the phase of
construction, but these effects are beyond the scope of the
present work and are not dealt with here. The major concern
with the operation of a pipeline system is the potential hazard
of spill. In the event of spill, some naphtha could possibly
reach water bodies in the form of non-degradable organics as
discussed in Section 6.1.2.1. Therefore, in siting SNG plants,
potential environmental effects of feedstock pipeline construction
and operation must be given due consideration.
Water Availability
SNG plants' makeup water requirements will be about
4.9 gallons per thousand scf of SNG produced (WO-046). A
typical 125 MM scfd SNG plant will need as much as 612,500
gallons of water per day. This amount of water may or may not
be critical to SNG plant siting. Constraints on water avail-
ability are discussed in Section 6.1.2.2.
-385-
-------
6.3 LNG Plant Impact
Liquefied natural gas (LNG) is one of the most promising
sources of supplemental energy. The usual LNG "life-cycle"
embraces: natural gas supply and delivery, liquefaction and
storage, then shipping, followed by receiving storage and re-
gasification. Peak shaving is the predominant LNG application in
the U.S. Peak shaving consists chiefly of gas delivery, liquefac-
tion and storage, and regasification. This operation is described
in Section 2.2. Development of the large Alaskan natural gas
reserve could lead to a domestic base load LNG industry.
Major areas under consideration are effluents, raw
material (feedstock and water) availability, and product trans-
portation which might create constraints on the siting of an
LNG plant. To provide a perspective on LNG technology, two types
of plants are considered: one with 10 MM scfd liquefaction
capacity and a 100 MM scfd regasification rate for peak shaving,
the other a 750 MM scfd plant for base load purposes.
6.3.1 LNG Plant Effluents
Liquefaction and regasification plants are the main
sources of effluents. Their effects on siting are addressed
in this section.
6.3.1.1 Liquefaction Plant
Potential primary effluent sources found within the
liquefaction plant include air emissions from boiler stacks
and miscellaneous fugitive losses, and liquid effluents from
water treatment facilities.
-386-
-------
Air Emissions Impact
Potential air emissions from a peak shaving plant module
and a 750 MM scfd base load LNG liquefaction plant are described
in Section 3.3 and are summarized in Table 6.3-1. The air emissions
from a 300,000 bpcd gasoline refinery, which serves as a reference
for ambient air quality analysis, are also included. Observe
in the table that the estimated particulates, sulfur dioxide,
NO , and carbon monoxide emissions from both 10 MM scfd and
X
750 MM scfd LNG liquefaction plants are less than or of the same
order as the calculated air emissions of the refinery "reference"
module. These observations, along with the results of the air
emissions impact of a 300,000 bpcd "reference" module in Section
6.1.1.1, indicate that both peak shaving and base load plants
will have minimal impact on the ambient air quality with regard
to these potential pollutants. With regard to hydrocarbon
pollution, the LNG hydrocarbon emissions are primarily methane,
and are not regulated within the non-methane hydrocarbon guide-
line.
Water Quality Impact
The major liquid effluent streams of LNG liquefaction
plants include acid and caustic wash water streams used for
demineralizer regeneration. The LNG liquefaction plant described
in Section 3.3 includes a holding pond for the effluent streams.
This method is a common practice in the process and utility
industries. Discharging liquid effluents into a holding pond
within the property of the liquefaction plant is not considered
a menace to the environment. Of course, this assumes that the
liquid effluents will be in an immobile condition, i.e., there
will be no seepage into the groundwater bodies. To achieve this
condition, proper siting is required. The hydraulic and geological
-387-
-------
TABLE 6.3-1
AIR EMISSIONS - LNG MODULE COMPARISONS
Pollutant
Particulates , Ib/hr
Sulfur Dioxide (S02) , Ib/hr.
Hydrocarbons (HC) , Ib/hr.
Carbon Monoxide (CO), Ib/hr.
Nitrogen Oxides (NO ) , Ib/hr.
X
10 MM scfd LNG
Plant Module
2
<1
20
2
30
750 MM scfd LNG
Plant Case Study
73
2,410
1,330
80
2,570
300,000 bpcd Gasoline
Refinery (Reference)
353
1,918
6,418
138
L846
CO
00
-------
conditions of a selected site must be known to allow prediction
of its potential to hold the discharged liquid effluents.
6.3.1.2 Regasification Plant
Potential emission sources found within the regasifica-
tion plant include possible "negative" thermal discharge of
circulating water.
A potential environmental issue that will stem from
the process operation of the regasification plant is the so-
called "negative" thermal pollution. It results from the use
of water for vaporizing and warming the gas. In the process
the discharge water will be several degrees colder than the
incoming natural water stream. Existing water-quality standards
for temperature generally specify limits as the temperature
rises above some specified level, ranging from 0° to 5°F.
However, some state jurisdictions set standards for both posi-
tive and negative temperature changes. Although "negative"
thermal discharge is beyond the scope of this study, it is
mentioned as a unique form of pollution that could pose a siting
problem for a regasification plant using a once-through heating
system.
6.3.2 Raw Material Availability
There are some environmental impacts and problems
in siting of an LNG plant that are associated with raw material
availability. Raw material in the discussions refers to both
feedstock'and water.
-389-
-------
6.3.2.1 Feedstock Availability
Natural gas is the feedstock for an LNG liquefaction
plant. It is transported by gas pipeline. An LNG peak shaving
plant must be sited along the main arteries of the natural gas
pipeline network. The main air emissions will come from the
compressor stations. Compressor stations can be driven by
gas engines, gas turbines, or by electricity. Emissions from
the first two prime movers are predominantly'NO . Since the
X
frequency of compressor stations is one every 50 to 75 miles
(BA-234), the NO emissions are unlikely to have a significant ad-
J\.
verse effect on the environment. Electrically driven compressor
stations must be energized through an electrical system, thus
siting must consider this factor. Major U.S. gas pipelines and
ports of entry for LNG are shown in Figure 6.3-1.
6.3.2.2 Water Availability
A major factor that needs to be considered in siting an
LNG liquefaction plant is water availability. Cooling water usage
for a 750 MM scfd LNG liquefaction plant may range from 500 to
1,000 million gallons per day (FE-084). Furthermore, total steam
power for a 750 MM scfd LNG liquefaction plant will add to the
fresh water make-up requirement (FE-084). The water flow rate
for a regasification plant that will regasify 750 MM scfd of LNG
may range from 300 to 500 million gallons per day. The water
requirements for 10 MM scfd peak shaving plants will be much
less.
The need for such substantial quantities of water will
present a siting problem at some U.S. sites.
-390-
-------
VO
I—1
I
{^CC^^^. ..'• •••'''''JMAJOR NATURAL GAS PIF-HLINES. LNG .PEAK SHAVING PLANTS,jV~;>*';;/;'• .'V ; ;-•'"•-';:.
j ,£|§J'« /• XIT^^^.>;..;,JAND POTENTIAL. LNG MARINE TERMINALS;^ ::':•" ;;v-: ..'•... " • =' ' '
©•' "'/•-.• xx^ '* <^i '"-^ ,-s/' n"f""' -^ •/ \\ '" " • " "' '•->'" '•••'•''• "'•• ''• •• '- •• '
/ / <^\ IS.* -! y'V \ { Y\ .r-f , '?-•••;.".• r;;.. . - s;
KEDKKA'I. I'OWER co.>rwissio\
2. PorcldnJ, K:HriO 40 ').0 S-.i--:: .in; : .. •.: ..i..::.-.• :'-.-.: ,'t.!j,.'. ,•;,:
U,':iilM.:;:.vii':. 1 :-!.•::.
3. Dyal-aii, Masu.:chuSGCCS 40 y.fi D.;is:^-v |i..;;j i-.t. .1.
v',; Ccnar.icut lalafld, RhoUs Inland 40 3.2 G:;.-i.I i.1-.-::':!;!:! f.-r r.-.rciv i.:,; I.i'G.
1. Net,- York, Xuw York 35-3; 5.0 !!'..;:-. .i.rir..: tiv.liU- ;:,M--if..
:iro.i - ^oar in.-ij.-r ;;as~ transir.i^:.;on
1 i.-.jr..
7. Cliasapcakc Bay, Virginia -o 3.0 SL.vt-r.il j>o.i:;iblti yive.i. Hj-.ivy K.-tvv
3. Savannah, Georgia 40 7.0 ;;Vilr -^jy:- .;a;j r.r.;ns:riiisior. line.
9. Mobile, Alaba=a «>0 6.0 Ka.-.r =ijar ;js -rjr.sr..isai3n lir.-js.
1. :--1;o Chai-ics, Louisiana 40
i6. Sen Fran.-isco - Or.kland, Calif. ;;
40 3.2-n.O -
17. Huc-r.er.i:, California 40 ' ti-^ .'. !..;-.ir T. -*.- 5,-ts r.j'JrT1-;sT.;nn ii;-._-
vi'.,il..-. !.^i^- •-,-.-:• .••;....
13. Los Ar.g=los - I-ong Bea=ii 49-50 2.a-5.4 •;•«;; ^v;™^ a:';'•'." •~'""'r "J1'" 8^
11. San Dic;;o, California 40 3.0-5.7 jjor -nai ^ K,.^trjn!:5i»sion lin-
FIGURE 6.3-1 LNG PIPELINES, PLANTS, AND POTENTIAL ENTRY PORTS
-------
6.3.3 Product Shipping and Receiving
Since LNG is predominantly transported by ship, pro-
duct shipping and receiving must be given some considerations
when siting LNG plants. During in-port operations, loading
and unloading of LNG will result in some atmospheric emissions,
primarily from boil-off of gas and from the use of other fuels.
Considerable quantities of boil-off gas have been observed at the
site during and after LNG unloading. There are also some water
effluents, as well as solid wastes, resulting from the disposal
of shipboard debris and sanitary wastes.
The foremost concern during product shipping and
receiving operations is human safety. Spillage of LNG,
because of its cryogenic temperatures, can cause brittle
fracture. Also spillage is a potential hazard, arising from
the LNG fuel properties. Methane-air mixtures are flammable
in the concentration range from 5.3 to 13.9 volume percent
methane (WA-157). Other environmental effects that may arise
from product shipping and receiving operations are given in
Table 6.3-2.
Another siting consideration is the fact that current
U.S. ports are not capable of handling the new generation of LNG
ships. Potential LNG marine terminals are shown in Figure
6.3-1. The development of these terminals will have some
secondary environmental effects. LNG shipping, receiving, and
storage, must abide by the National, State, and Local LNG Codes
and Standards. An excerpt of the current status of these LNG
codes and standards appears in reference BA-276.
-392-
-------
TABLE 6.3-2
LO
to
I
APPROXIMATE GUIDE TO APPRAISING ENVIRONMENTAL
FACTORS OF LNG FACILITIES
(Source: WA-157)
0 - NONE
1 - SOME
2 - MAJOR
I. LIQUEFACTION PLANT
A. GAS CLEAN-UP
8. 'COMPRESSION
C. COOLING
D. STORAGE
E. AUXILIARIES
II. MARINE TRANSP.
A. LIQUID LOADING
B. DEEP WATER OP.
C. IN-PORT OPER.
D. LIQUID UNLOAD
III. RECEIVING TERM.
A. TRANSFER '
B. STORAGE
C. REGASIF.
D. ODORIZATION
E. AUXILIARIES
IV. LAND TRANSP.
A. LOAD/UNLOAD
B. TRANSPORT
C. MAINT.
V. PEAK SHAVING
A. GAS CLEAN-UP
B. COMPRESSION
C. COOLING
D. STORAGE
E. REGASIF.
F. ODORI2ATION
EFFLUENTS
THERMAL
0
1
2
0
2
0
0
1
0
0
0
2
0
0
0
0
0
0
1
2
0
2
0
CHEMICAL
AIRBORNE
^
2
0
0
1
2
1
2
2
1
2
1
1
2
1
1
2
0
1
0
0
1
1
2
LIQUID
1
0
1
0
1
1
1
1
1
0
0
1
0
1
0
0
1
. 0
0
1
0
1
0
SOLID
1
0
1
0
0
0
1
1
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
NOISE
0
2
0
0
1
0
0
1
0
1
0
0
0
0
0
2
1
0
2
0
0
0
0
HYDRAULIC
0
0
1
0
0
0
0
2
0
1
0
0
0
0
0
0
0
0
0
1
0
0
0
IMPACTS
ECOLOGICAL
TERREST.
0
. 0
1
1
1
1
0
0
1
1
1
0
0
0
1
2
0
0
0
1
1
0
0
AQUATIC
1
0
2
0
1
1
1
2
1
1
0
1
0
0
1
2
1
0
0
2
0
1
0
HUMAN
HEALTH
1
1
1
0
1
1
0
1
1
0
0
1
2
1
1
0
0
1
1
1
0
1
2
HUMAN
SAFETY
0
1
0
2
0
2
1
1
2
1
2
1
0
0
2
2
1
0
1
0
2
1
0
HUMAN
ACTIVITIES
1
1
1
1
0
1
1
2
1
1
1
0
1
1
0
2
0
1
1
1
1
0
1
0 - Suggests that there may be no need to evaluate or assess.
1 - Indicates that the effluent or impact may have to be
considered but detailed evaluation or assessment is
probably unnecessary.
2 - Indicates that the effluents and impacts which will
probably deserve quantitative evaluation and assessment.
-------
APPENDIX 6.1-1
PRIMARY (SECONDARY)
AMBIENT AIR QUALITY STANDARDS OF ARBITRARILY
SELECTED STATES
-394-
-------
u>
VO
PRIMARY (SECONDARY)
AMBIENT AIR QUALITY STANDARDS OF ARBITRARILY SELECTED STATES
(Units are micrograms per cubic meter with ppm in brackets)
Pollutant
California
Colorado
Delaware
1. Arithmatic mean.
2. Geometric mean.
3. Not to be exceeded more than once per year
Florida
Sulfur Oxides -, See Follow -
Annual Averages (A.M.) ing Pages
24- hr. Maximum 3
3-hr. Maximum 3
30-minute Maximum
Particulate
Annual Average (G.M.) 2
24-hr. Maximum-^
5-hr. Maximum .
3-hr. Maximum
1 hr. Maximum
Nitrogen Dioxide
Annual Average (A0M.)
Non-Methane Hydrocarbons
3-hr. Maximum 3 (6-9 A.M0)
Carbon Monoxide
8-hr. Maximum 3
1-hr. Maximum 3
25 80(60)
150 370(260)
(1,300)
55 70(60)
180 200(150)
500
100(100)
131(131)
9,200(9,200)
4tiyOOO (40,000)
60
260
1^00
60
150
100
160
1QOOO
4QOOO
-------
CALIFORNIA; SAN FRANCISCO
Bay Area Air Pollution Control District Regulations
Applicable to Refineries
1. No person shall cause, let, permit, suffer, or allow the
emission for more than three minutes in any one hour of
a gas stream containing air contaminants which, at the
emission point or within a reasonable distance of the
emission point, in as dark or darker in shade as that
designated as No. 1 on the Ringleman Chart as published
in the United States Bureau of Mines Information
Circular 7718, or of such capacity as to obscure an
observer's view to a degree equal to or greater than
does smoke described above.
2. The emission does not contain more than "n" grains of
particulate matter for standard cubic feet, where
n = 0.06/L
and L is the significant dimension of the emission point
in feet.
3. No person shall cause, let, permit, suffer, or allow any
emission from a heat transfer operation of particulate
matter in excess of 0.15 grain per standard dry cubic
foot of exhaust gas.
-396-
-------
California: San Francisco (cont.)
Bay Area Air Pollution Control District Regulations
Applicable to Refineries
4. No person shall cause, let, permit, suffer, or allow any
emission of SOa xvhidh results in ground level concentra-
tions of S02 at any given point in excess of 1.5 ppm
(Vol) for 3 consecutive minutes or 0.5 ppm (Vol) averaged
over 60 consecutive minutes, or 0.04 ppm (Vol) averaged
over 24 hours, or any of the following limits:
S0a Concentration Total Cumulative Exposure
ppm (Vol) Between Midnight and the
Next Succeeding Midnight
in Hours
1.5 0.05
0.5 1.0
0.3 3.2
0.1 9.6
0.04 24.0
5. No person shall cause, let, permit, suffer, or allow the
emissions of gas containing S02 in excess of 300 ppm (Vol)
6. No person shall cause, let, permit, suffer, or allow an
emission of an effluent containing a concentration of more
than 50 ppm of organic compounds calculated as hexane (or
300 ppm total "carbon").
-397-
-------
PRIMARY (SECONDARY)
oo
VO
00
AMBIENT AIR
(Units are
Pollutants
Sulfur Oxides
Annual Averages (A.M.)
24-hr. Maximum 3
3 - h r . Ma x imum 3
30 -minute Maximum
Particulate
Annual Average (G0M0)2
24-hr. Max imum 3
5 - hr , Maximum
3-hr. Maximum
1-hr, Maximum
Nitrogen Dioxide
Annual Average (A0M.)
Non-Methane Hydrocarbons
3-hr. Maximum 3 (6-9 A
Carbon Monoxide
8-hr. Maximum ;?
1 - h r . Max imum -
QUALITY STANDARDS OF
micrograms per cubic
Georgia
1 43
229
60
150
100
.Mo) 98
10,400
40,000
ARBITRARILY
meter with
Illinois
75(60)
260(150)
100
160
10,000
40,000
SELECTED STATES
ppm in bracke
Indiana
80(60)
365(260)
75(60)
260(150)
100
160
10,000
40,000
ts)
New Mexico
[0.02]
[0.10]
60
150
[0.05]
[0.19]
[8.7]
[13.1]
1. Arithmatic mean.
2. Geometric mean.
3. Not to be exceeded more than once per year.
-------
u>
PRIMARY (SECONDARY)
AMBIENT AIR QUALITY STANDARDS OF ARBITRARILY SELECTED STATES
(Units are micrograms per cubic meter with ppm in brackets)
Pollutants
Maryland-
Massachusetts
Montana
5-hr. Maximum
3-hr. Maximum
1-hr. Maximum
Nitrogen Dioxide
Annual Average (A0M.) 100
Non-Methane.Hydrocarbons
'3-hr. Maximum 3 (6-9 A. M.)
Carbon Monoxide
8-hr c Maximum 3
1-hr. Maximum
120
9,200
New Jersey
Sulfur Oxides
Annual Averages (A.M.) 1
24 -hr. Maximum 3
3-hr. Maximum 3
30-minute Maximum
Particulate
Annual Average (G.M.) 2
24 -hr. Maximum 3
79(39)
262(131)
75(65)
160(140) .
73
300
75
180
[0.02]
[0.10]
75
200
80(60)
365(260)
(3,300)
75(60)
260(150)
100
160
10,000
40,000
±. Aritnmatic mean.
2. Geometric mean.
3. Not to be exceeded more than once per year
-------
PRIMARY (SECONDARY)
AMBIENT AIR QUALITY STANDARDS OF ARBITRARILY SELECTED STATES
(Units are micrograms per cubic meter with ppm in brackets)
Pollutant New York North Carolina North Dakota
Sulfur Oxides
Annual Averages., (A.M0) 80 60 60
24-hr. Maximum ^ 365(260) 260 260
3-hr. Maximum J 1,300
30-minute Maximum
Particulate
Annual Average (G0M.)^ . 60 60
24-hr. Maximum-3 250 150 150
5-hr. Maximum
3-hr. Maximum
1-hr. Maximum
Nitrogen Dioxide
Annual Average (A0M0) 100 100 100
Non-Methane Hydrocarbons
3-hr. Maximum 3 (6-9 A.M.) 160 160 160
Carbon Monoxide
8-hr. Maximum ^ 10,000 10,000 10,000
1-hr. Maximum 3 40,000 40,000 40,000
1. Arithmatic mean.
2. Geometric mean.
3. Not to be exceeded more than once per year.
-------
PRIMARY (SECONDARY)
AMBIENT AIR QUALITY STANDARDS OF ARBITRARILY SELECTED STATES
(Units are micrograms per cubic meter with ppm in brackets)
Pollutant
Texas
Utah
I
o
M
I
Sulfur Oxides -,
Annual Averageso(A.M0)
24-hr. Maximum ~
3-hr. Maximum
30-minute Maximum
Particulate „
Annual Average-(G.M.)
24-hr. Maximum
5-hr. Maximum
3-hr0 Maximum
1-hr. Maximum
Nitrogen Dioxide
Annual Average (A.M0)
Non-Methane Hydrocarbons
3-hr. Maximum 3 (6-9 A.M.)
Carbon Monoxide
8-hr. Maximum 3
1-hr. Maximum 3
See Tables 6.1-5,
-6, and -7
[0.03] ([0.02])
[0.14] ([0.10])
90
[0.05]
[0.24]
[9]
[35]
Virginia
80
365
1,300
75(60)
260(150)
100
160
10,000
40,000
Wyoming
60
260
1,300
60
150
100
160
10,000
40,000
1. Arithmetic mean.
2. Geometric mean.
3. Not to be exceeded more than once per year.
-------
APPENDIX 6.1-2
WATER QUALITY MODELS
-402-
-------
Mathematical models are useful tools in computing the
pollutant concentrations in receiving waters which result from
predicted refinery effluent discharges. Two levels of complexity
are exhibited by the models. A very simple model gives back-
ground concentrations some distance from the effluent outfall and
applies to well mixed streams. The other model, in addition to
giving -background concentrations, computes constant concentration
contours extending from effluent outfall to the region of back-
ground concentrations. This model, referred to as the Water
Quality Display Model (WQDM), is especially useful in showing
pollutant distributions in tidal waters during conditions of
flood, slack, and ebb tides. The two types of models will
be discussed separately.
(a) Well Mixed Stream Model
In either a fast flowing stream or in a tidal water-
way, pollutants introduced into the main flow channel are rapidly
well mixed with the receiving waters by turbulent mixing. This
means that at some relatively short distance from effluent out-
fall, a background concentration of the effluents in the receiv-
ing waters will exist. Higher concentrations naturally occur
at outfall, but they decrease to the background level over a
distance required for the well mixed condition to prevail. The
steady-state or mean background level is determined by the com-
bination of the effluent concentration of pollutants, the ef-
fluent flow rate, and the net outflow of the stream or waterway.
-403-
-------
The number which is of interest in determining the
water quality impact of the effluent is the mean number of parts
per million (or mg/.O of a particular pollutant which exists in
the receiving waters which would not have been there had the
refinery effluent not been discharged. For well mixed pollutants
this number is easily obtained from the formula:
Mean Additional
Pollutant Concentration
Dilution Factor
x
[Effluent Pollutant
Concentration
where the dilution factor is simply the ratio of the effluent
volumetric flow rate to the sum of the net stream outflow rate
and the effluent volumetric flow rate.
When this model is applied to a fast flowing, non-
tidal stream, the concentration gradients near outfall mentioned
previously are essentially constant and generally tend to extend
downstream only. When this simple model is applied to tidal
waters, however, it should be recognized that tidal f.lows will
cause temporal variations of concentrations in the vicinity of
effluent outfall during each day which extend both up and
downstream. Also, as mentioned above, concentrations near out-
fall will naturally exceed the mean concentrations computed
above. Ebb tide concentrations in the vicinity of effluent out-
fall will be less than flood tide concentrations there because
of greater ebb currents. Slack water outfall concentrations
will be substantially greater than either ebb or flood tide
concentrations. Slack water conditions fortunately exist for
relatively short periods, however. Gradients in concentrations
of varying magnitude exist at all times between initial effluent
concentrations at outfall and the mean concentrations computed
-404-
-------
according to the simple model described above. The mean levels
will exist at some distance from outfall, hoxvever, and they
are perhaps the best measure of the impact on area water quality
of additional waste water discharges.
It should be noted that the mean additional pollutant
concentration computed by the model assumes the effluent is
introduced into pure ambient water. When a pollutant concen-
tration, x, preexists in the receiving waters, the total resultant
concentration in the waters is given by:
Total Pollutant
Concentration
. = x +
Dilution Factor,
j
Effluent Pollutant
Concentration
-x
This equation follows from conservation of mass when two volumes
of fluid are added together. It should also be noted that the
mean, steady-state levels computed by the model depend on net
stream outflow, whereas the localized concentration gradients
near outfall discussed above for tidal waters depend temporally
on tidal flows.
(b) Water Quality Display Model (WQDM)
To quantify the distribution of pollutants near
effluent outfall which the simple model described above does
not do, the WQDM can be used. This model predicts relative
pollutant concentrations in the vicinity of outfall, and its
outputs can be scaled to predict absolute pollution levels
from outfall to the region of background levels. Discussions
of both the WQDM and how its results are scaled are presented
in this section.
-405-
-------
The WQDM is a computer model which computes stream
functions and diffusion equation solutions. From these results
the streamlines of flow in a stream can be plotted, and contours
of constant pollutant concentration surrounding effluent outfall
can also be plotted. The model utilizes Monte Carlo techniques
to obtain its numerical results.
Application of the WQDM to the water quality problem
under consideration here involves two main assumptions. The
first is that flow in the waterways investigated can be described
substantially correctly by potential flow. This means that
the stream function, $(x,y), from which the water velocity field
can be constructed can be obtained as the solution of Laplace's
equation, vp $(x,y) = 0. Also, the problem is assumed to be
two dimensional. Except for 'small regions near certain boundaries
of minor importance to the overall results, potential flow is
a valid assumption for the streams studied. Also, the depth
dimension of the waterways is used in the construction of boundary
conditions for each problem so that two dimensional solutions
are essentially valid. The second assumption is that periods of
steady-state flow exist, corresponding to ebb, flood, and slack
tide conditions, for time intervals long enough for roughly
steady-state concentrations to be established. This is valid
for roughly two hour periods spanning peak ebb flows, peak flood
flows, and slack water conditions.
The basis of obtaining Monte Carlo solutions to
Laplace's equation and to the diffusion equation stems from the
similarity of the equations obeyed by a particle performing a
random walk and the finite difference equations corresponding
to Laplace's equation and the diffusion equation. Brown dis-
cusses the application of Monte Carlo techniques to these two
equations (BR-132).
-406-
-------
Application of the WQDM to a problem involves the
following steps. First, the stream geometry is gridded and
boundary values are assigned to each point on the closed boundaries
of the stream. The stream shores comprise two boundaries and
cuts across the stream both up and downstream from the region
near outfall are used to close the boundaries. Since each shore
is assumed to be a limiting streamline, arbitrary boundary values
can be assigned to each. Usually one shore is set to zero and
the other is set at unity. Boundary values along the cross-
stream cuts are set to monotonically increasing values from zero-
to one. The exact variation along these segments of the boundaries
is tied to the channel depth along each cut. The values reflect
more flow at deeper parts of the channel. For the thusly estab-
lished conditions, the Monte Carlo solution of Laplace's equation
is obtained. Unlike finite difference techniques, the Monte
Carlo approach can be used to find the solution at a single
point within the modeled region. With the stream function,
$ (x,y), computed for all points in the modeled region, the
streamlines of the flow can be obtained as lines of constant
$ (x,y).
Next, solutions of the diffusion equation must be
obtained. To do this the results of the stream function computa-
tions must be available as is explained as follows. The basic
operation in the Monte Carlo,models is the random movement of
a particle from one grid point to another. For a particle
at an arbitrary internal grid point, it may move up, down, left,
or right. In solving Laplace's equation, the probabilities
associated with moving in each direction.are equal. But solution
of the diffusion equation in the presence of advective flow
-407-
-------
requires that movement with the current be favored. It is,
therefore, necessary to know the velocity field at each point of
the stream so that the probabilities governing the random walk
can be properly adjusted at each point in the stream. Since the
velocity at any point is proportional to the gradient of the
stream function, the stream function previously computed can be
used to set up the four required probabilities as given by:
> = * [:
x- L
u
11 y
py_ = < L1 - —
, 2K !" , /
where r = — 5- 1 + ( 1 +
L v
2 2 2
2K !" , / 6 c
2 2,2
c = u + u
x y
6 = a scale factor,(ft/grid division)5
9
K = the dispersion coefficient, (ft /min)
(A§)y
uy = 3 (A$)x
P = a proportionality constant; positive
for downstream flow and negative for
upstream flow, and
-408-
-------
and (A$) = the derivatives of $ (x,y) in the y and
x directions, respectively.
With the stream function input, the WQDM computes
relative pollutant concentrations at each point in the modeled
region by simulating the turbulent dispersion of pollutants
entrained in advective flow for roughly two hours. From these
results contours of constant concentration surrounding outfall
can be drawn. Peak or 100% concentration is at outfall. The
region beyond the zero percent concentration contour corresponds
to the mean background level region.
Scaling the contours to absolute levels is done
individually for each pollutant as follows. The zero percent
contour level is computed by the well mixed stream model.
The net stream outflow is used to compute the dilution factor
for this case. The 100% contour for ebb and flood flox^s is also
computed by the model of Section (a); however, calculation of
the dilution factor is modified for these cases. In defining
the dilution factor for these cases, it is important to define
the significance of the 100% contour. Pollutant concentrations
precisely at effluent outfall are not immediately diluted,
therefore, the raw effluent pollutant concentration could be
taken to correspond to the 1007o contour. This was not done,
however. Instead, it was deemed reasonable to associate the
100% contour concentration with the mean concentration occurring
over the smallest resolvable region of the modeled area. This
region is defined by a square with sides the length of a single
grid division. For this interpretation of the 100% contour,
it follows that the dilution factor is computed as before,
-409-
-------
except the stream flow is taken to be only the amount that flows
through the unit square centered at effluent outfall. Also,
it is part of either ebb or flood flows that are used in the
calculation, not net stream outflow.
Computation of the 10070 concentration for slack
water is based on the same interpretation of the 10070 contour
as above; however, the absence of stream flow means that a
dilution factor cannot be used. Instead, the mean concentration
over the minimum square is computed assuming diffusion from the
center of the square for a two hour period. The concentration
for such a problem has a Gaussian distribution from the center
of the square, and its sigma is a function of time. The sigma
at two hours can be compared to the dimensions of the minimum
square to allow scaling the minimum square in terms of the
Gaussian form of the pollutant concentration. It is then easy
to compute a scale factor for the raw effluent pollutant con-
centration which gives the mean concentration over the area
of the minimum square. This factor is simply the average of
a Gaussian density over the scaled dimensions of the minimum
square.
As for the simple model of Section (a), the concen-
trations discussed above are for discharges into pure water.
Total background concentrations resulting from discharges into
preexisting pollutants are computed as discussed in Section (a) .
-410-
-------
APPENDIX 6.1-3
MAJOR WATER LAWS
-411-
-------
Law of River Compacts
These compacts are generally made within a river basin
agency and have a direct effect on the member states of that agency.
The compacts exist to mediate the problems associated with
allocation of interstate water and the administration of water
rights. Because of these problems and existing or impending
litigation, several affected States have entered into interstate
compacts or requested court apportionment of the affected waters
for the river systems.
The Federal Constitution provides that no State shall
enter into any agreement or compact with another State, or with
a foreign power, without the consent of Congress. Approval by
Congress is required once the compact is ratified by the several
States and usually provides for a Federal representative serving
and reporting on the negotiations.
Upper Missouri River Basin
The interstate compacts which are applicable to the Fort
Union region are the Belle Fourche River Compact and Yellowstone
River Compact.
"The Belle Fourche River Compact between Wyoming and
South Dakota was approved by the Act of February 26, 1944. Under
this compact water right priorities theretofore established in
one State were to be recognized in the other. Of the remaining
unappropriated water, 90 percent is to be allocated to South Dakota
and 10 percent to Wyoming. Diversions and impoundments of water
in one State for use in the other State are authorized where
State appropriation laws are observed.
The Yellowstone River Compact among Wyoming, Montana,
and North Dakota was approved by the Act of October 30, 1951.
It divides Yelloxvstone River Basin Surplus waters (1) in the
Clarks Fork, 60 percent to Wyoming and 40 percent to Montana;
-412-
-------
(2) in the Big Horn River, 80 percent to Wyoming and 20 percent
to Montana; (3) in the Tongue River, 40 percent to Wyoming and 60
percent to Montana; and (4) in the Powder River, 42 percent to
Wyoming and 58 percent to Montana. The Compact provides that the
three signature States will not singly or jointly take actions
which adversely affect Indian water rights to those waters of
the Yellowstone River or its tributaries. Diversions and impound-
ments in one State for use in another State are authorized where
State appropriation laws are observed. Diversions out of the
Yellowstone Basin require the unanimous consent of all of the
compacting States" (NO-055).
Upper and Lower Colorado River Basins
"The Colorado River is perhaps the most regulated river
in the United States, and its utilization is such that very
little usable water now discharges from its mouth into the Gulf
of California. The cornerstone is the Colorado River Compact
of 1922, which the seven Basin states negotiated pursuant to the
Act of August 19, 1921 (42 Stat. 171). This Compact divides
the Colorado River Basin into two parts; i.e., the Upper Basin
and the Lower Basin, separated at a point on the river near the
Utah/Arizona border known as Lee Ferry. Article III(a) apportions
to each basin in perpetuity 7.5 m.a.f. of water per year. Article
III(c) provides that any future Mexican water rights, recognized
by the United States, are to be supplied as provided in the Compact
Article IH(d) obligates the Upper Basin not to deplete the flow
at Lee Ferry below an aggregate of 75 m.a.f. for any period of 10
consecutive years reckoned in continuing progressive series. In
1948 the Upper Basin States entered into a compact to divide the
water of the Upper Basin as described in Article III(a) apportions
among the States of Arizona, Colorado, New Mexico, Utah, and
Wyoming the Colorado River Compact water in the following manner:
-413-
-------
(1) Arizona, 50,000 a.f. •
(2) Colorado, New Mexico, Utah, and Wyoming, after
deduction of Arizona's 50,000 acre-feet: Colorado, 51.75 percent,
New Mexico, 11.25 percent, Utah, 23 percent, and Wyoming, 14
percent.
Article III(b)3 provides that no state shall exceed
its apportioned use in any year when such use deprives another
state of its water during that year. Curtailment in use of
water apportioned is to be determined by the Commission. The
Commission is to determine and allocate losses of water as a
result of reservoir storage. The Upper Colorado River Commission
'is created as an interstate administrative agency and its duties
are defined by Article VIII of the Compact. The Compact is not
to interfere with the right or power of any state to regulate
within its boundaries the appropriation, use, and control of
water apportioned to such state. The failure of any state to
use water shall not constitute a relinquishment or a forfeiture
of the right to use that water. Article XIX provides that the
obligation of the United States to the Indian tribes, the Mexican
Treaty or any rights of the United States to acquire waters in
the Upper Colorado River System are not to be affected" (US-168).
State Water Laws
In all parts of the country, water laws are largely
based on the Riparian Doctrine., the Appropriation Doctrine, or
a combination of both. A riparian right to withdraw water is
based on the ownership of land next to a surface-water body.
The right is independent of the use or non-use of the water. An
appropriation right is based upon the beneficial use of the
water. In other words, the first to appropriate and use the water
-414-
-------
has a priority over others who come along and appropriate at a
later time. An appropriation right is independent of the location
of the land with respect to the water.
• Figure B-l shows that all states roughly east of the
95th meridian follow the Riparian Doctrine exclusively, with the
exception of Mississippi and Florida. This eastern half of the
country coincides with the area of water surplus,shown on Figure
o.l.lO. In this humid region, the Riparian Doc-trine requires a
land owner to allow the stream to flow by or through his land
in its natural state. Thus, in its strict sense, the right does
not allow for the consumptive use of the water except for small
domestic needs. When irrigation becomes necessary in a riparian
state, the courts modify the doctrine to allow reasonable use in
relation to neighboring users. No riparian user can take all
the water of a stream and allow none to flow down to his neighbor.
This contrasts sharply with the appropriator's right in other
parts of the country to consume all that he needs. The Mountain
States follow the Appropriation Doctrine exclusively; other states
recognize both doctrines. In those states that recognize both,
the relative importance of each doctrine varies considerably.
In addition to riparian and appropriation rights, there
is a third kind of right based on need. Some examples of this
type are Indian water rights, Federal Reserve rights, and, in
some cases, municipal rights. These rights will be discussed
later.
"In all states, water laws relating to ground waters
generally are based on either the Riparian Doctrine or the
Appropriation Doctrine. The Eastern States generally use the
-415-
-------
I
JS
Y^K,
* \k
«a
^"SBSB
FIGURE B-l
SURFACE-WATER LAWS
"m
r
DOCTRINES RECOGNIZED
:••.-:•:.•;;;:: APPROPRIATION
BOTH APPROPRIATION
AND RIPARIAN
-------
English common-law version of Che Riparian Doctrine, which
gives absolute ownership of ground water to the land owner.
Fifteen states modify this version and apply the American rule
of 'reasonable use1, which restricts the landowner's rights in
relation to others. California goes one step further in the
modification of the Riparian Doctrine with its doctrine of
'correlative rights'. Here, the landowner's use must not only
be reasonable, but must be correlated with the uses of others
during times of shortage. When the supply is limited, use is
restricted to the lands directly overlying the common supply."
The applicable ground water laws in the United States are shown
on Figure B-2.
"While the Appropriation Doctrine seems to function
easily for surface-water supplies in the arid states, it runs
into some difficulty x^hen it is applied to ground water. The
main reason for this is that ground water is a hidden resource,
whose occurrence and movement are poorly understood by most
people.
Many states, in recent years, have begun to expand their
control and regulation of ground-water use, and this has resulted
in the creation of many special rules and regulations. Some of
these apply to methods of well construction, monitoring of changes
in ground-water levels and ground-water quality, periodic sub-
mission of data on ground-water use, and preventive measures to
minimize contamination and pollution" (GE-058).
Federal Water Laws
The Federal Government was given limited powers relating
to water resource development which are either expressly delegated
or can be reasonably implied from tho Constitution. There are a
-417-
-------
00
I
GROUND-WATER LAWS
DOCTRINES RECOGNIZED
r, '///ax APPROPRIATION
W:::?: COMMON LAV/ RIPARIAN
rI REASONABLE USE
j-Hft CORRELATIVE RIGHTS
-------
number of Federal laws that specifically effect the basins under
consideration. Some of the more important laws are:
Indian Water Rights
Under a variety of treaties, acts, and executive orders
enacted around the turn of the century, numerous Indian reservations
were created in the central western states.
"Responsibility for the administration of Indian lands
and waters on these reservations rests with the Bureau of Indian
Affairs; however, the rights to the lands and water actually
are vested with the Indians or tribes. The Indians of each
reservation appear to have some legal claim to the use of the
waters located on or flowing through or along its boundaries.
Such rights are read from the treaties and agreements between
the Indian tribes and the United States which have been approved
by acts of Congress or formalized by Executive Orders. The Indian
people claim a right to these waters free from State regulation
and with a priority at least as early as the date the reservation
was recognized or established. The Indian water right priority
is not conditioned on use and may be exercised at any time. The
Indian right can be quantified by fixing the amounts of water needed
to serve the purpose or purposes for which the reservation was
established. (See Winters v. United States, 207 U.S. 564 (1903);
Conrad Investment Co, v. U.S. 161 Fed 829 (9th cir. 1908);
U.S. v. Walker River Irrigation District 104 Fed 334 (9th cir.
1939); U.S. v. Ahtanum Irrigation District 235 Fed 321 (9th cir.
1956); and in closing, Arizona v. California 373; 546-600 (1963).
Thus if the purpose were to promote an agricultural economy, as
has been the case generally, the quantity of water reserved would
be the amount needed to serve the practically irrigable acreage
on the reservation. It also has been urged on behalf of the
Indians that since the purpose of the Indian reservation is to
-419-
-------
provide an economic base for the Indian people residing thereon,
it must follow that the Indian water right is a right to use
the available reservation waters for any beneficial use including
irrigation, livestock, domestic, power, recreation, industrial
and municipal purposes. Nevertheless, several State water
administrators continue to urge that the Indians are entitled
only to that water for which proper application under State
procedures has been made.
The irrigation of Indian lands was authorized by the
General Allotment Act of February 8, 1887, which also provided
that the Secretary of the Interior should make a just and equal
distribution of the available water among the Indians. Later
the Act of April 4, 1910, made specific provision for irrigation
developments on Indian reservations, and special authorizations
have been provided by Congress for many individual projects.
The right to use Indian water for nonirrigation purposes has
not been litigated or judically determined" (NO-055).
Mexican Water Treaty
"In the Mexican Water Treaty of 1944, Mexico is guaran-
teed an annual quantity of 1,500,000 a.f. of water from any and
all sources. The water is to be delivered to that section of the
river, away from any shores or bottom, near the international
boundary" (NO-055).
Boulder Canyon Acts
The Boulder Canyon Project Act of December 21, 1928
approved the Colorado River compact of 1922 and provided for the
construction of Hoover Dam and the All American Canal in the
Lower Colorado Basin.
The Boulder Canyon Project Adjustment Act of July 19,
1940 (54 Stat. 774), among other things, provided funds i:or
-420-
-------
planning .for the use of water in the states of the Upper
Colorado Basin.
Other Acts of Interest
An excellent review of other Federal laws that effect
the central western energy states is found in the Water Work Group
Report of the Northern Great Plains Resource Program, pgs 29-42
(NO-055). These laws are broken down into the basic areas of
interest in water resource development, including:
(1) Irrigation,
(2) Power,
(3) Navigation,
(4) Municipal and Industrial Water Supply,
(5) Flood Control,
(6) Watershed Protection and Flood Prevention,
(7) Outdoor Recreation, Fish and Wildlife,
(8) Environment,
(9) Water. Quality, and
(10) Planning.
More detailed discussions of legal water-use constraints are also
available in several of the other references cited in the
bibliography (US-168, NA-190, NA-176).
-421-
-------
APPENDIX 6.1-4
ATMOSPHERIC STABILITY CLASSES AND PLUME DISPERSION CHARACTERISTICS
-422-
-------
The dispersive characteristics of the planetary boundary
layer (the lowest 3000 to 4000 feet of the troposphere) are
primarily a function of the mean motion of the atmosphere and
fluctuations or turbulence about the mean motion. The turbulence
is to a great extent a function of the stability or temper-
ature lapse rate of the boundary layer. Stability can be divided
into six classes for purposes of the dispersion modeling of
pollutants. These classes, based on EPA's Climatological Dis-
persion Model (CDM) are:
Stability Class Description
A Extremely unstable
B Moderately unstable
C Slightly unstable
D (day) Neutral (daytime)
D (night) Neutral to slightly
stable (nighttime)
E + F Stable to extremely
stable
This classification scheme, developed by Pasquill and Gifford,
has associated with it a set of coefficients of diffusion, both
in the horizontal and the vertical. That is, each stability
class has a diffusivity associated with it. As the air becomes
more stable its diffusive capabilities decrease.
A consideration of the individual stability classes
as defined by Pasquill and their effects on pollutant dispersion
can now be conducted. Under the CDM scheme, the unstable
classes are A, B, and C. D (day) is the classification when
the lapse rate is neutral during the day. The four stability
classes listed above can occur only during the daylight hours.
-423-
-------
The occurrence of Classes D (night) and E 4- F is confined to the
nighttime hours. D (night) stability occurs during cloudy and/or
very windy nights, when vertical mixing is sufficient to prevent
the formation of very stable lapse rates or ground-based inver-
sions. E + F stability is confined to those nights with surface
winds of 9 knots or less and clear or scattered sky/cloud con-
ditions.
Stability Class A occurs very rarely, usually with a
relative frequency of occurrence of 2% or less on an annual
basis. For such a stability regime to occur, skies must be
clear, the sun must be nearly overhead, and winds must be 5
knots or less. In such a situation, the lapse rate is very
unstable, much greater than the dry adiabatic rate. Thermals
are numerous and vertical mixing and turbulence are extreme.
In such a case, a plume emitted from a stack will have a large
plume rise and, therefore, a large effective stack height.
However, the plume will not be dispersed rapidly because of the
light winds. In addition, the large coefficient of.vertical
diffusion due to strong vertical mixing causes the plume to
expand rapidly in the vertical so that high pollutant concentra-
tions reach the ground within a very short distance from the
stack. Since the plume trajectory in such a situation does
not follow a long path, there is not sufficient air flow to allow
for dispersion and dilution of the plume. Therefore, high ground-
level pollutant concentrations can occur during a short-term
sampling period under these meteorological conditions.
When a plume from a stack is dispersing under very un-
stable conditions, such as A or B stability, it exhibits a
phenomenon known as looping (Figure 1). In addition to diffus-
ing rapidly in the vertical because of forced mixing from the
-424-
-------
instability of the air, the plume exhibits a variety of shapes
and sizes because of the thermal currents. As a result, the
plume "loops" (from plume classification scheme by Church) caus-
ing alternately high and low ground-level pollutant concentra-
tions as the pollutants travel along the plume trajectory. In
such a case, the pollutant concentration measurements would have
a high standard deviation. Even though the pollutant concentra-
tion varies rapidly under looping conditions, a time averaging
process can be used to obtain an average concentration over a
short time interval of about 30 minutes to 1 hour.
B and C stabilities cause the plume to behave in much
the same manner as it does with A stability, except that the
plume does not diffuse in the vertical as rapidly as it does
with A stability. With B and C stabilities, therefore, the plume
touchdown is further downwind than with A stability, thus afford-
ing more dispersion and diffusion along the plume trajectory to
touchdown. Lower ground-level concentrations are the result.
Under unstable lapse rate conditions, the stronger the
winds, the lower the plume rise and the higher the ground-level
pollutant concentrations.
When the lapse rate is neutral such as with D (day) and
D (night) stabilities, the plume does not diffuse appreciably,
resulting in "coning" (Figure 2). If E + F stability prevails,
diffusion is minimal in both a lateral and vertical sense.
Therefore, the plume remains rather concentrated and advects
for great distances before reaching the ground (Figure 3).
Two other plume configurations are shown in Figures 4
and 5. Figure 4 is characteristic of nocturnal radiation con-
ditions, in which a shallow ground-based inversion is formed.
-425-
-------
The air above is. neutral. Therefore, the plume only disperses in
an upward direction since the air below is very stable and will
not mix. The result is "lofting".
Figure 5 illustrates the phenomenon known as "fumigation"
or inversion breakup. Such a condition occurs when a vertical
temperature distribution such as Figure 3 or Figure 4 is modified
by heating of the ground during the morning. . As a result, the
lowest layer of air becomes very unstable causing pollutants
which had been suspended aloft to diffuse rapidly to the ground.
*
A troublesome aspect of stability assessment is the fact
that the lower portions of the planetary boundary layer often
have multiple stratifications of stability. That is, the stabil-
ity of the air varies drastically with small increments of in-
creasing height above the surface. On many occasions, superadia-
batic ("A" stability) layers or strong inversion layers ("E + F"
stability) are confined to approximately the lowest 100 feet of
the atmosphere. Above this layer the air may stabilize or
destabilize rapidly with increasing height. If stability varies
considerably with height, attempts to classify the stability of
the planetary boundary layer into one classification independent
of height based on low-level stability determinations yield sta-
bility classifications that misrepresent the true state of the
atmosphere.
Such stability classifications independent of height can
also cause erroneous results in dispersion modeling efforts.
Distributions of stability measurements taken in the lowest
100 feet of the boundary layer will be skewed toward the un-
stable end of the classification spectrum because of the intro-
duction of components of surface heating and mechanically-induced
turbulence into the raw data. Such a stability distribution, if
used in dispersion modeling of tall stacks, could result in the
-426-
-------
prediction of mvTch higher ground-level concentrations of pol-
lutants than are actually experienced.
The effect that the stratification of the stability
of the atmosphere will have on a plume of emissions will be
quite variable. Factors which will influence the general dis-
persion and diffusion of the plume are: surface roughness
parameters, depth and gradient of the lapse rate stratifications,
wind speed, wind shear, and net radiation. A "looping" plume
may "cone" in some locations and vice versa.
In summary, the ground-level concentrations of pollutants
which are experienced during a short period of time are a function
of the stability of the air and the wind speed. Stability classifi-
cations simply indicate the change of temperature with increasing
height. If air temperatures decrease rapidly with height, the
air is unstable. If the air cools only moderately with height,
it is neutral. If the air warms with height, it is stable.
-427-
-------
M
Dry Adl»bitlc
Actual Laose
Teopecacur*
Figure 1
Uoscable: looping A,. B, C Stability
This condition is associated with unstable
conditions, since the air cools rapidly
with height. The plume exhibits this
"looping" behavior.when "A", "B11, or
"C" stability conditions exist. Plume
touchdown is relatively close to the
source.
•5
Temperature
Figure 2
Stable: coning, D & £ Stability
This plume behavior, called "coning", occurs
when neutral or slightly stable conditions
such as D (day) or D (night) stability exist.
The plume generally travels farther before
it reaches the ground than in the unstable cases
-428-
-------
Figure 3
u
f.
to
Dry AdUbaclc
Actual Lapse
Teaperature
Inversion: stable, F Stability
This plume behavior, called "fanning",
occurs during "E + F" stability, when the
air, is very stable, such as when an inver-
sion is present. The plume travels a .
great distance in an essentially undiluted
state-before it reaches the ground.
Dry Adlabactc
Actual Lapse
Teaperacure
Figure 4
Inversion belou-ncucral aloft:lofting
D Stability Aloft, F Stability Below
This plume behavior, called "lofting", occurs
when a shallow ground-based nocturnal radiation
inversion exists, with neutral conditions above
the inversion. When relatively tall stacks
are involved, the plume disperses above the
inversion, but cannot disperse through the
inversion to re.ach the ground.
-429-
-------
Dry Adlabatlc
Figure 5
u
M
«4
(I
B
Actual Lapse
Teoperature Unstable below - Inversion alofc: fuaigaclon
This condition, called "fumigation", often
results in very high short-term ground-level
pollutant concentrations. It is associated
with the break-up of a ground-based inver-
sion, such as is present in Figure 3 or
Figure 4, during the morning. The undiluted
pollutants aloft are rapidly mixed down
to the ground.
-430-
-------
APPENDIX 6.1-5
ATMOSPHERIC DISPERSION MODELS
-431-
-------
Radian uses three different models to calculate air
quality levels depending on the averaging time associated with
a particular Federal or state standard. Each of the three models
uses the same plant emissions data and standard Gaussian disper-
sion relations, but the detailed treatment of the meteorological
variables differs among the various models.
The three models used are a long-term average model,
a short-term model, and a 24-hour average model.
The long-term average model uses historical meteoro-
logical data to estimate annual average pollutant concentrations.
These estimated annual averages can also be used in conjunction
with certain statistical assumptions to estimate maximum concen-
trations for averaging times less than one year (e.g. monthly).
The short-term model computes estimated concentrations correspond-
ing to a 10-minute averaging time. A statistical assumption,
different from that employed in the annual average model, is used
to transform the 10-minute average estimates to estimates corre-
sponding to other averaging times (usually 3 hours or less).
The long-term model is used to estimate concentrations
for averaging times greater than three hours. The short-term
model is used to estimate concentrations for averaging times
less than three hours.
The long-term model is similar to the Climatological
Dispersion Model (CDM) recently developed at the National
Environmental Research Center , Research Triangle Park, North
Carolina. The CDM is a long-term average model which utilizes
long-term meteorological data in conjunction with Gaussian dis-
persion using Pasquill-Gifford dispersion coefficients. This
-432-
-------
model is essentially an updated version of the x^ell-known AQDM
(Air Quality Display Model) developed under EPA auspices and
both models share a common conceptual approach. The primary
differences between the two models relate to calculation of
plume rise for point sources, specification of mixing heights
and wind profiles, and the treatment of the effects of area
sources.
The average concentration CA due to area sources at
a particular receptor is given by
16
16
i [I 1k(P) I I
2rr
o k=l
S(P, z, u£, pm)J dp
where
k = index identifying wind direction sector
qk(p) = J Q(p,cp)dcp (k sector)
Q(p,cp) = emission rate of the area source per unit
area and unit time
-433-
-------
p = distance from the receptor to an infinitesimal
area source
cp = angle relative to polar coordinates centered
on the receptor
j& = index identifying the wind speed class
m = index identifying the class of the Pasquill
stability category
§(k,£,m) = joint frequency function (generally for an
annual period
S(p,z;u ,P ) = dispersion function defined in Equation 1
ft Ml '
z = height of receptor above ground level
u = representative wind speed
P «= Pasquill stability category
For point sources, 'the average concentration Cp due to
N point sources is given by
*<\>*>m) Gn S^nz'^,^
16 V V V
~ T^ L L L
n-1 1=1 m=l Pn (2)
-434-
-------
The-total .concentration for the averaging period is
the sum of concentrations of the point and area sources for that
averaging period.
For point sources, the effective stack height, h, is
the sum of the physical stack, h , and the plume rise, Ah:
h = hQ + Ah (3)
The plume rise, Ah, is computed v;ith formulas developed b}'
Briggs. For unstable and neutral conditions:
Ah = 1.6F1/3 .trV/3 P * 3.5X* (4)
and
Ah = 1.6F1/3 U~1(3.5X.v)P/fl p>3.5X* (5)
/° if F 55
F = g V R°f(T -T )/T 1
h . s svs a" s
where
g = acceleration due to gravity
V, = average exit velocity of gases of plume
S
II - inner radius of stack
s
-435-
-------
T = average temperature of gases of plume
s
T = ambient air temperature
CL
U = wind speed
p = distance from source to receptor
For stable conditions
Ah = 2.9 (F/Us) P > 2.4 Us~% (6)
(i.e., Equation (6) rather than Equation (4) or (5) is used
for stable conditions)
where
fi = ambient potential temperature
z = height.
The short-term model employed is similar to the long-
term model in that the Briggs Plume Rise Formulas and the
Pasquill-Gifford dispersion formula are employed. Specifically
the ten-minute average concentration due to N point sources is
computed from
N:
— 1 ^ (8)
CP=^ I Gn P(*,y,o;VPm)
n=l
-436-
-------
where
x = downwind distance
y = crosswind distance
P
-------
The joint probability function <}>(k>£>m) is obtained
from historic meteorological data collected at meteorological
stations near the site in question. Average ambient tempera-
ture and average daytime and nighttime mixing height for each
site are obtained from an analysis of the meteorological data
for the site.
Because of the uncertainty associated with the log-
normal statistical extrapolation procedure previously referred
to in connection with scaling annual values back to shorter
averaging times, which becomes larger as the procedure is applied
to shorter averaging times, Radian has felt it desirable to aug-
ment the two models described by a 24-hour average model. This
model might be considered as a hybrid of the two models previously
described in the sense that it incorporates some of the averag-
ing features of the long-term model with the statistical assump-
tions of the short-term model.
In essence, the model is based on a recognition of
the fact that during a 24-hour period meteorological conditions
and plant conditions will change with time sufficiently to in-
validate the two-tenths power scaling rule applied over a 24-
hour period. On the other hand, over such a period of time the
assumption that conditions will change according to an annual
frequency distribution is also not realistic. Consequently
the 24-hour period is assumed to be divided into an integral
number of shorter time intervals with specified plant emissions
and meteorological conditions which are assumed constant within
a time interval, but which can change from interval to interval.
For a given interval, the short-term model using the two-tenths
scaling rule is used to compute the concentration at a particular
receptor for the interval, and the final 24-hour concentration is
computed as a uniformly weighted average of the contributions
from the individual time intervals.
-438-
-------
7.0 AREAS FOR RESEARCH AND DEVELOPMENT
The evaluation of the environmental aspects of the oil
refining, LNG, and liquid-based SNG industries leads to the
identification of several areas of research and development
needs. These are itemized below.
Ambient Air Monitoring
There is a real need for an automated field hydrocarbon
monitor which will provide a detailed breakdown of the hydrocar-
bon species. This would provide an excellent clue to their sources,
and thus greatly simplify the control of such emissions. Fugitive
hydrocarbon emissions are a serious problem as has been discussed
earlier. Given information on species and ratios between various
species, along with compositions of the various process streams,
"fingerprinting" techniques could be developed to greatly local-
ize the source of fugitive emissions.
Pollutant Dispersion
Better methods of tracing pollutants in the atmosphere
are needed. With a wide number of easily detectable tracers,
the contributions of many local sources to the air quality in an
area could be determined and allow the optimum control" strategy
to be developed.
There also is a real need for an air monitoring ration-
ale, i.e., how many stations should be used, how should they be
sited, should they be moved seasonally, etc. It is presently
difficult to compare results from various networks due to un-
certainties resulting from the network designs.
-439-
-------
These needs are inter-related to the need for better
prediction from diffusion models. The data obtained identifying
sources with ambient concentrations is needed to "tune" developing
sophisticated models. The "tuned" models in turn would be of
great value in establishing air monitoring network designs.
Effects
Additional data is also needed regarding the fate of
pollutants in the environment and their long-term effects on
plant and animal life. In particular, the effects of hydrocarbons
and NO , separately and jointly, need to be better understood.
X
If additional information could be obtained regarding the role
of various hydrocarbon species or classes of hydrocarbon species
in the^formation of photochemical smog, more realistic ambient
air hydrocarbon standards could be developed. This would allow
the expenditures for controls to be channeled to the most
effective areas.
Water Monitoring
There is a need for automated field analyzers to detect
all the parameters for which emission limits have been set for
petroleum refineries. Reliable automated analyzers would greatly
reduce man-power requirements in demonstrating compliance with
the regulations. Also automated data acquisition systems to
provide summary reports and trigger alarms are needed.
-440-
-------
Water Effluent Control
The provisions of the Clean Water Act require zero
,y~
discharge or zero impact in future years. Development work is
needed in determination of the applicability of such technologies
as ion exchange, membrane separation processes, and forced
evaporation processes to refinery effluent streams.
Cost Impacts
Studies are needed to indicate trends in plant costs
as functions of increased emissions controls and improved fuel
economy. Because of increasing energy costs, the trends in
new plant designs are moving toward greater fuel utilization.
This has the secondary effects of reducing total emissions and
total cooling water requirements.
While energy conservation is in itself a desirable
feature, concurrent cost savings can in some cases cover the
expense of added control equipment. Engineering studies of the
various cost alternatives are needed.
"Negative" Thermal Pollution Control
LNG regasification plants' water discharge is a few
degrees colder than the temperature of the receiving water.
This is a form of a "negative" thermal pollution. This pollu-
tion will affect the life cycle of the marine organisms near
the outfall. The subject of "negative" thermal pollution has
been overshadowed by the "positive" thermal pollution. As a
result, less information has been collected regarding its
environmental impact. Some effort should be made to develop
schemes and design criteria for control of "negative" thermal
pollution. The study must investigate its adverse effects on
the environment and the cost of controlling it.
-441-
-------
Solid Wastes
The chemical composition of typical solid wastes from
refineries should be examined in detail to determine the content
of hazardous or potentially hazardous species. The solubility
of these species should be determined, and the long term chemical
stability-solubility of these materials in a landfill environ-
ment should be studied.
The movement and attenuation of any soluble hazardous
species in various types of soils should be examined. Also,
the long term stability in the landfill environment of liners
for disposal sites should be examined.
Hazardous Chemicals
A comprehensive and cost effective sampling and analytical
strategy is needed for plant effluent streams. The objective of
this effort would be to provide a means of detecting potentially
hazardous materials in these effluents. Field testing to verify
these strategies will also be needed.
-442-
-------
REFERENCES
-443-
-------
REFERENCES
AM-020 American Oil Company, Mandan Refinery, Fluid Bed Inciner-
ation of Petroleum Refinery Wastes, Washington, GPO (1971)
AM-041 American Petroleum Inst., Div. of Refining, Recommended
Rules for Design and Construction of Large, Welded, Low-
Pressure Storage Tanks, API Standard 6207 Washington,
D.C., 1970.
AM-055 American Petroleum Institute, Committee on Refinery
Environmental Control, Hydrocarbon Emissions from Refin-
eries, API Publication No. 928, Washington, D.C. (1973).
AM-062 American Petroleum Institute, Division of Refining,
Manual on Disposal of Refinery Wastes, Volume on Liquid
Wastes, First edition, Washington, D.C.(1969).
AM-099 American Petroleum Institute, Annual Statistical Review,
Petroleum Industry Statistics 1964-1973, Washington,
D.C. (1974).
AM-127 American Gas Association, Operating Section, LNG In-
formation Book. . . 1973, Arlington, VA..
AT-040 Atmospheric Emissions From Petroleum Refineries, A
Guide For Measurement and Control, PHS No.763, Wash-
ington, D.C., Public Health Service (1960).
AT-042 Attari, A., Fate of Trace Constituents of Coal During
Gasification, Final Report," PB "223-001, EPA 650/2-73-004,
Chicago, Illinois, Inst. of Gas Technology (1973).
AU-015 Audibert, F. and J. C. Lavergne, "Upgrading Residues
By the IFP Process," CEP 67(8), 71 (1971).
BA-003 Bartok, W., et al., Systems Study of Nitrogen Oxide
Control Methods for Stationary Sources,Final Rept.
Vol. I, PH-22-68-55.Linden, N.J..Esso Research and
Engineering Company (1969).
BA-166 Barry, Charles B., "Reduce Glaus Sulfur Emission,"
Hydrocarbon Proc. 51(4), 102-6 (1972).
BA-230 Battelle - Columbus and Pacific Northwest Labs.,
Environmental Considerations in Future Energy Growth,
Contract No. 68-01-0470, Columbus, Ohio (1973).
BA-234 Battelle-Columbus and Pacific Northwest Labs., Environ-
mental Considerations in Future Energy Growth, Contract
No. 68-01-0470, Columbus, Ohio (1973).
-444-
-------
REFERENCES (Cont.)
BA-276 Ball, W. Leonard, "Current Status of National., State, and
Local LNG Codes and Standards," A.G.A. 1972 Operating
Section Proc., D-211.
BE-047 Bennett, A. R., "A New Technique for the Rapid and Con-
tinuous Estimation of Carbon Dioxide," Inst. Gas Engrs.,
Copywright Publ. No. 370, 16-19 (1950).
BE-147 Beychok, Milton R., "Wastewater Treatment," Hydrocarbon
Proc. 50(12), 109 (1971).
BE-228 Bennett, Gary F., ed., Water - 1973, AICHE Symposium
Series 136, Vol. 70 (1974), N.Y., AICHE (1974).
BE-246 Bery, R. N. and R. J. Burcin, "Modern SNG Plants -
Guidelines for Economic Design and Execution," Paper 73-
D-76, A.G.A. 1973 Operating Section Proc., D-172.
BL-078 Bland, William F. and Robert L. Davidson, eds., Petro-
leum Processing Handbook, N.Y., McGraw-Hill, 19677
BR-103 Bresler, Sidney A. and John D. Ireland, "Substitute
Natural Gas: Processes, Equipment, Costs," Chem. Eng.
79(23), 94 (1972).
BR-132 Brown, George W., "Monte Carlo Methods," in Modern
Mathematics for the Engineer, Edwin F. Beckenbach, ed.,
N.Y., McGraw-Hill (1956).
BR-140 Brunet, M.J. and R.H. Parsons, "How Mobil Reduced Sour
Water Stripping Problems," Hydrocarbon Proc. October 1972,
107. -
BR-199 Brown, R.A., H.B. Mason, and R.J. Schrieber, Systems
Analysis Requirements for Nitrogen Oxide Control of
Stationary Sources, Final report, EPA 650/2-74-091.
PB 237 367, Contract No. 68-02-1318, Task 3, Mountain
View, CA, Aerotherm/Acurex Corporation (1974).
CH-196 Cheremisinoff, Paul N. and Richard A. Young, eds.,
Pollution Engineering Practice Handbook, Ann Arbor, MI,
Ann Arbor Science (1975).
CO-175 Colony Development Operation, Atlantic Richfield Co.,
Operator, An Environmental Impact Analysis for a Shale
Oil Complex at Parachute Creek, Colorado, Vol. 1, Pt. 1,
Plant Complex and Service Corridor (1974).
-445-
-------
REFERENCES (Cont.)
CU-008 Gulp, Russell L. and Gordon L. Gulp, Advanced Wastewater
Treatment, New York, Van Nostrand Reinhold (1971).
CU-016 Cuffe, Stanley T., "Catalyst Regeneration," Air
Pollution Engineering Manual, John A. Danielson, comp.
and ed., AP-40 Cincinnati, Ohio PHS (1967).
DA-069 Danielson, John A., comp. and ed., Air Pollution
Engineering Manual, 2nd ed., AP-40, Research Triangle
Park, N.C., EPA, Office of Air and Water Problems (1973).
DE-070 Deal, G. H., Jr., et al., "A Better Way to Extract Aro-
matics," Petrol. Refiner. September 1959, 195.
EL-033 Elkin, H.F. and R.A. Constable, "Source/Control of
Air Emissions," Hydrocarbon Proc. 51(10), 113 (1972).
EL-062 Elliot, Thomas C., "S02 Removal from Stack Gases "
Power 1974 (September), S1-S24.
EL-065 Elkin, Harold F., "Petroleum Refinery Emissions,"
Philadelphia, PA, Sun Oil Company.
EN-043 Environmental Conservation, Washington, D.C., National
Petroleum Council (1972).
EN-071 Environmental Protection Agency, Compilation of Air
Pollutant Emission Factors, 2nd ed. with supplements,
AP-42, Research Triangle Park, N.C. (1973).
EN-072 Environmental Protection Agency, Office of Air and Water
Programs, Office of Air Quality Planning and Standards,
Background Information for Proposed New Source Standards:
Asphalt Concrete Plants, Petroleum Refineries, Storage
Vessels, Secondary Lead Smelters and Refineries, Brass
or Bronze Ingot Production Plants, Iron and Steel Plants,
Sewage Treatment Plants, Vols. 1 and 2, Research Triangle
ParkT N.C. (1973).
EN-196 Environmental Protection Agency, "Air Programs; Standards
of Performance for New Stationary Sources. Additions and
Miscellaneous Amendments," Fed. Reg. 39(47), 9308 (1974).
EP-009 Epstein, M., et al., "Limestone Wet Scrubbing of S02,"
CEP 70(6), 53 (1974).
-446-
-------
REFERENCES (Cont.)
FE-076 Federal Energy Administration, Project Independence
Report. Washington, B.C. (1974).
FE-084 Federal Power Commission, National Gas Survey, 5 vols.
Vol. 1, FPC Report; Vol. 2_, Supply Task Force Reports;
Vol. 3, Transmission Task Force Reports; Vol.4,
Distribution Task Force Reports; Vol. 5, Special Reports,
Washington, D.C. (1973).
FE-085 "Feedstocks for SNG Plants to Dwindle " Oil Gas J.
12 August 1974. 54.
FL-047 "Felxicoking: An Advanced Fluid Coking Process," API
Proc. Refining 1972, 666.
FO-026 Forney, Albert J., e_t al. , Analysis of Tars, Chars,
Gases, and Water Found in Effluents From the Synthane
Process, Technical Progress Rept. 76, Pittsburgh,
PA, Pittsburgh Energy Research Center (1974).
FR-119 Franzen, A.E., V.G. Skogan, and J.F. Grutsch, "Tertiary
Treatment of Process Water," CEP 68(8), 65 (1972).
GE-058 Geraghty, James J., et al., Water Atlas of the United
States, Port Washington, N. Y., Water Information
Center (1973).
GI-035 Gifford, F. A., "Use of Routine Meteorological
Observations for Estimating Atmospheric Dispersion,"
Nuclear Safety 2. 47-51 (1961).
GL-027 Gloyna, E.F., D.L. Ford, and J. Eller, "Water Reuse in
Industry," J. WPCF 42(12, pt. 1), 237 (1970).
GO-051 Godsave, L.A. and M.K. Eyles, "New BP (British Petroleum)
Process Desulfurizes Residue,'" Oil Gas J. 69(41), 75-7 (1971)
HA-132 Haase, R., K-H. Ducker, and H.A. Kuppers,
"Aktivitatskoeffizienten Und Dissoziationskonstam.en
WaBriger Salpetersaur'e Unduberchlorsaure," Berichte
Bunsengesellschaft Phys. Chem. 69 (2) (1965JT
HA-282 Hatch, Lewis F., "A Chemical View of Refining", Hydro-
carbon Processing 48, 77 (1969).
HA-286 Hale, Dean, ed., "Peakshaving Capability Increases to
Meet 1975-75 Winter Demands," Pipeline Gas J. (December,
1974).
-447-
-------
REFERENCES (Cont.)
HI-083 Hittman Associates, Inc., Environmental Impacts, Effi-
ciency, and Cost of Energy Supplied by Emerging Technol-
ogies, Draft Report on Task 5, low-Btu gasification
or coal and Task 6, high-Btu gasification of coal
Columbia, Md. Feb. 1974.
HI-090 Hittman Associates, Inc., Environmental Impacts,
Efficiency, and Cost of Energy Supply and End Use,
Phase 1, Draft Final Report, HIT-561, Columbia, MD (1973).
HO-049 Holzworth, George C., Mixing Heights, Wind Speeds, and
Potential for Urban Air Pollution Throughout the
Contiguous United States, Research Triangle Park, N. C.,
EPA, Office of Air Program (1972).
HU-094 Hutchins, Roy A., "New Method Simplifies Design of
...Activated-Carbon Systems", Chemical Engineering 1973,
133 (August).
HY-007 "Hydrocarbon Processing 1969 Petrochemical Handbook
Issue," Hydrocarbon Proc. 48(11), 135-54 (1969).
HY-008 "Hydrocarbon Processing 1970 Refining Processes Handbook,"
Hydrocarbon Proc. 49(9), 163-274 (1970).
HY-013 "Hydrocarbon Processing Refining Process Handbook,"
Hydrocarbon Proc. 53(9) (1974).
HY-014 "Hydrocarbon Processing SNG/LNG Handbook," Hydrocarbon
Proc. 52(4) (1973).
IN-029 Institute of Gas Technology, Illinois Institute of
Technology, LNG: A Sulfur-Free Fuel for Power Generation,
Final Report, Contract No. PH 22-68-58, Chicago, IL (1969).
IN-056 R. C. Barras, "Instrumentation for Monitoring Air
Quality", ASTM Special Technical Publication 555 (1974).
JA-056 Jain; L.K., E.L. Calvin, and R.L. Looper, State of the
Art for Controlling NOV Emissions, Pt. 1, Utility Boilers,
PB 213 297, Charlotte, N.C. Catalytic, Inc. (1972).
JI-008 Jimeson, Robert M. and Roderick S. Spindt, eds., Pollution
Control and Energy Needs, Advances in Chemistry Series
127, Washington, D.C., ACS (1973).
KA-149 Kalen, B. and F. A. Zenz, "Filtering Effluent from a Cat
Cracker," CEP 69(5), 67 (1973).
-448-
-------
REFERENCES (Cent.)
KE-128 Kett, Terrence K., Gerard C. Lahn, and William L. Schriette,
"Flexicoking, A Versatile Residuum Conversion Process,"
Paper 4C, Presented at the AICHE 67th Annual Meeting,
Washington, B.C., (December 1974).
KE-129 Kellogg Co., 250 MMSCFD SNG From Naphtha British Gas
Council (CRG) Process, Houston, TX (1972).
KL-032 Klett, Robert J., "Treat Sour Water for Profit,"
Hydrocarbon Proc. 51(10), 97-9, Gulf Pub. (1972)
KO-133 Koehler, G.R., "Alkaline Scrubbing Removes S02,"
CEP 70(6), 63 (1974).
LA-078 Lafferty, W.L., Jr., and R.W. Stokeld, "Alkylation
and Isomerization,',' Adv. Chem. Ser. 103. 130-49 (1971).
'.*' :,-,.
LA-100 Larsen, R.J., A Mathematical Model For Relating Air Quality .
Measurements to Air Quality Standards, Washington, D.C.,
EPA, Office of Air Programs (1971).
LA-142 LaMantia, C.R., R.R. Lunt, and I.S. Shah, "Dual Alkali
Process for S02 Removal," CEP 70(6), 66 (1974).
LA-150 Lacey, Robert E., "Membrane Separation Processes,"
Chem. Eng. 79(19). 56 (1972).
LE-148 Leitner, G. F., "Reverse Osmosis for Water Recovery and
Reuse," CEP 69(6). 83 (1973).
LE-156 Lewis, James P., Private Communication, Transco Terminal
Co. (1 May 1975).
LI-095 Linden, Henry R., "The Role of SNG in the U.S. Energy
Balance," AGA, 1973 Operating Section, Proceedings,
pp. D-151 IT.
LO-095 Logan, E. C., "Harrison SNG Plant-- A Viable, Reliable
Operation," Pipeline Gas J. 1974 (May).
LO-102 Lorn, W.L., Liquefied Natural Gas. N. Y., Wiley (1974).
MA-226 Mallatt, R.C., J.F. Grutsch, and H.E. Simons, "Incinerate
Sludge and Caustic," Hydrocarbon Proc. 49. 121 (1970).
ME-095 Metcalf and Eddy, Inc., Wastewater Engineering--Collection,
Treatment, Disposal, McGraw-Hill Series in Water Resources
and Environmental Engineering, N. Y., McGraw-Hill (1972).
MI-137 Miller, W.E., "The Cat-Ox Process at Illinois Power "
CEP 70(6). 49 (1974).
-449-
-------
REFERENCES (Cont.)
MS-001 MSA Research Corporation, Hydrocarbon Pollutant Systems
Study, Vol. 1^, Stationary Sources, Effects, and Control,
APTD-1499, PB 219 073, Evans City, PA, MSA Research
Corporation (1972).
t
NA-004 National Air Pollution Control Techniques Advisory
Committee, -Control Techniques for Carbon Monoxide Emissions
from Stationary Sources, NAPCA Pub. No. AP-65 (March 1970).
NA-005 National Air Pollution Control Techniques Advisory
Committee, Control Techniques for Nitrogen Oxide Emissions
from Stationary" Sources, NAPCA Pub. No. AP-67 (March 1970).
NA-029 National Air Ppllution Control Administration, Control
Techniques for Particulate Air Pollutants. NAPCA Pub.
No. AP-51 (January 1969).'
NA-032 National Air Pollution Control Administration, Control
Techniques for Hydrocarbon and Organic Solvent Emissions
from Stationary Sources, AP-68, Washington, D.C.(1970).
NA-113 J. S. Nader, "Development in Sampling and Analysis,
Instrumentation for Stationary Sources", Journal of
the Air Pollution Control Association, Vol. 23, No. 7,
p. 587 (1973).
NA-176 National Petroleum Council, U.S. Energy Outlook, A
Report of the National Petroleum Councils Committee
on U.S. Energy Outlook, Washington, D.C. (1972).
NA-182 Nack, H., et al., Development of an Approach to Identi-
fication of Emerging Technology and Demonstration
Opportunities, EPA 650/2-74-048, Columbus, Ohio,
Battelle - Columbus Labs. (1974).
NA-190 National Petroleum Council, U.S. Energy Outlook, Water
Availability, Washington, D.C. (1973).
NE-044 Nelson, W.L., Petroleum Refinery Engineering, 4th ed. ,
McGraw-Hill Series in Chemical Engineering, N. Y. ,
McGraw-Hill (1958).
NE-087 Newkirk, Robert W. and Paul J. Schroeder, "R.O. Can
Help with Wastes," Hydrocarbon Proc. 51 (October),
103 (1972).
NE-088 Nelson, W.L., "Energy Consumption Declines," Oil Gas J.
17 March 1975. 132.
NG-002 "NG/SNG Handbook," Hydrocarbon Processing 50(4),
93-122 (1971).
-450-
-------
REFERENCES (Cont.)
NO-055 Northern Great Plains, Water Work Group, Report,
Helena, Montana, U.S. Bureau of Reclamation (1974).
NU-009 Nusbaum, I., R.E. Cruver, and J.H. Sleigh, Jr.,
"Reverse Osmosis - New Solutions and New Problems/'
CEP 68(1). 69 (1972).
OG-010 Ogawa, Yoshio, et al., "Simultaneous Treatment of Sulfurous
Oxide and Nitrogen Oxide in Exhaust Gas/' Japanese
L.P.P. 43895/74 (1974), Translated by Trans-Chem., Inc.,
Knoxville.TN.
01-008 Oil and Gas Journal 1 April 1974.
OK-012 Oka, Hideo, Eiichi Ichiki, and Tatsuo Shiraishi, "Process
Removes NOV Efficiently/' Hydrocarbon Proc. 53 (October),
113 (1974)
PA-095 Pasquill, F., "The Estimation of the Dispersion of
Windborne Material," Meteorological Mag. 90 (1063) (1961).
PH-019 S. C. Phillips and D. A. Mack, "Instruments for Water
Quality Measurements", Environmental Science and
Technology, Vol. 9, No. 3 (March 1975).
PO-091 Potter, B.H. and T.L. Craig, "Commercial Experience with
an S02 Recovery Process," CEP 68(8), 53 (1972).
PO-109 Pohlenz, J.B., "The ShelM'Flue Gas Desulfurization Process,"
Presented at the Flue Gas Desulfurization Symposium,
Atlanta, GA (November 1974).
PR-046 Prather, B. Vail, "Advanced Treatment of Petroleum Refinery
Wastewater by Autoxidation," J. W.P.C.F. 42(4), 596 (1970).
PR-052 Processes Research, Inc., Industrial Planning and
Research, Screening Report. Crude Oil and Natural Gas
Production Processes., Final Report, Contract No. 68-02-0242,
Cincinnati, Ohio (1972).
PR-074 Pross, T.W., "Marine Transportation-State-of-the
Art," Presented at the-Intersociety Conf. on Trans-
portation, Denver, CO, September 1973, N. Y., ASME (1973).
RA^OSl Rabb,' A. , "Sludge Disposal: A Growing Problem,"
Hydrocarbon Proc. 44(4). 149 (1965).
RA-117 Racine, W. J., "Plant Designed to Protect the Environment/'
Hydrocarbon Proc. 1972, 115 (March).
RA-119 Radian Corporation, A Program to Investigate Various
Factors in Refinery Siting, Final Report with Map
Inserts, Austin TX (1974).
-451-
-------
REFERENCES (Cont.)
RA-181 Radian Corp., Environmental Feasibility Studies for
Operation of A Petroleum Refinery Operation in Texas,
7 Vols., Austin, TX (Sept., 1975).r
RE-.070 Reigel, S.A., R.P. Bundy, and C.D. Doyle, "Baghouses -
What to Know Before You Buy," Pollution Eng. May 1973, 32.
RI-044 Richardson, A.H. and J.H.D. Hooper, "How BP's Isomeri-
zation Process is Performing," Petroleum and Petro-
chemical Int'l. 2(12), 52-5 (197T77
SL-053 Slack, A.V., "S02 Scrubbing: Why Not Recovery?," Elec.
World 1 November 1974, 33.
SO-076 Soudek, M., "What Lube Oil Processes to Use," Hydrocarbon
Proc. 53(12) (1974).
SO-080 "Solid Waste Disposal," Chem. Eng. 78(14). 155 (1971).
ST-124 "Standards of Performance for New Stationary Sources.
Proposed Standards for Seven Source Categories,"
Fed. Reg. 38(111), PT. 2 (1973).
ST-221 Stork, Karl, M.A. Abrahams, and A. Rhoe, "More Petro-
chemicals from Crude," Hydrocarbon Proc. 53 (November).
157 (1974).
SU-044 Sutton, O.G., "A Theory of Eddy Diffusion in the
Atmosphere," Proc. Royal Soc. A 135, 143-65 (1932).
TE-111 "Technology. Water Pollution Control," Chem Eng. 78(14),
65 (1971).'
TH-038 Thomson, Sidney J., "Techniques for Reducing Refinery
Wastewater," Oil Gas J. 68(40), 93-8 (1972).
TH-076 Thomson, S.J., "Data Improves Separator Design,"
Hydrocarbon Proc. 52(October), 81 (1973).
TU-020 Turner, D. Bruce, Workbook of Atmospheric Dispersion
Estimates, PHS Pub. No. 999-AP-26 (Revised ed.), Air
Resources Field Research Office, Environmental Science
Services Admin. (1969).
US-056 U.S. Department of the Interior, Federal Water Pollution
Control Admin., The Cost of Clean Water. Vol. Ill,
Industrial Wastes Profiles, No. 5, Petroleum Refining.
-452-
-------
REFERENCES (Cont.)
US-083 U.S. Dept. of the Interior, River of Life, Water:
The Environmental Challenge, Conservation Yearbook
No. 6, Washington, D.C., GPO (1970).
US-093 U.S. Dept. of the Interior, Final Environmental Statement
For the Prototype Oil Shale Leasing Program, 6 Vols.,
Washington, D.C. , GPO, (1973).
US-112 U.S. Dept. of the Interior, Bureau of Reclamation,
Upper Colorado Region, El Paso Gasification Project,
San Juan County, N.M., Draft Environmental Statement,
DES-74-77 (1974).
US-124 U.S. Dept. of Commerce, Maritime Administration,
Maritime Administration Tanker Construction Program,
Final Environmental Impact Statement, EIS-AA-73-0725-F,
Washington,D.C.~
US-159 U.S. Coast Guard, Polluting .Incidents In and Around U.S.
Waters, Calendar Year 1972, Washington, D.C.
US-164 U.S. Bureau of Reclamation, Upper Colorado Region,
WESCO Gasification Project and Expansion of Navajo
Mine by Utah International Inc., San Juan County,
New Mexico, Draft Environmental Statement, DBS 74-107,
Washington, D.C. (1974).
US-168 U.S. Dept. of the Interior, Water for Energy
Management Team, Report on Water for Energy in the
Upper Colorado River Basin (1974).
US-191 "U.S. and Canadian LNG Peakshaving Facilities,"
Pipeline Gas J. 1974 (December), 26.
VA-064 Van Winkle, Matthew, Distillation, N. Y., McGraw-Hill
(1967).
VO-025 Voogd, J. and Jack Tielrooy, "Improvements in Making
Hydrogen," Hydrocarbon Proc. 46(9). 115 (1967).
WA-073 Watkins, Charles H. and Frank Stolfa, "Recent Operating
Results with RCD Isomax," American Petroleum Inst. Div.
of Refining, Proceedings 1971, 713 (1971).~~
WA-082 Walker, G. J., "Design Sour Water Strippers Quickly,"
Hydrocarbon Proc. 48(6). 121 (1969).
-453-
-------
REFERENCES (Cont.)
WA-082 Walker, G.J., "Design Sour Water Strippers Quickly,"
Hydrocarbon Proc. 48(6), 121 (1969).
WA^086 Walters, R.M., "How An Urban Refinery Meets Air Pollution
Requirements," CEP 68(11), 85 (1972).
WA-157 Fourth International Conference on Liquefied Natural
Gas, June 24-27 (1975T
WA-168 Wall, James D., "LNG Goes World-Wide", Hydrocarbon
Proc. 54(4). 141 (1975).
WH-019 Wharton, G.W. and E.P. Hardin, "Three Stage Unit Improves
Crude Split," Petroleum Refiner 37(10), 105-8 (1958).
WH-032 "What's New in LNG," Reprints from Oil and Gas J. 1967-
1971, Tulsa, OK, Petroleum Publishing.
WI-073 Wilson, J.G. and D.W. Miller, "The Removal of Particulate
Matter from Fluid Bed Catalytic Cracking Unit Stack Gas,"
JAPCA 17(10), 682 (1967).
WI-142 Witt, P.A., Jr. andM.C. Forbes, "By-Product Recovery
via Solvent Extraction," CEP 67(10^. 90 (1971).
WIr-144 Wisniewski, Ralph, "Process Converts Sludge to Landfill,"
Oil Gas J. 17 March 1975, 133.
WO-046 American Gas Association 1973 Operating Section
Proceedings.
-454-
-------
TECHNICAL REPORT D/„
(Please read Instructions on the reverse be forefoot
1. REPORT NO.
EPA-600/2-75-068
2.
3. RECIPIENT'S ACCESSION-NO.
a. TITLE AND SUBTITLE Environmental Problem Definition for
Petroleum Refineries, Synthetic Natural Gas Plants,
and Liquefied Natural Gas Plants
5. REPORT DATE
November 1975
6. PERFORMING ORGANIZATION CODE
7-AUTHOR(sE.C.Cavanaugh, J.D.Colley, P.S.Dzierlenga,
V.M.Felix, D.C.Jones, and T. P. Nelson
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Radian Corporation
8500 Shoal Creek Boulevard
P.O. Box 9948
Austin, Texas 78766
10. PROGRAM ELEMENT NO.
1AB013- ROAP 21ADD-042
11. CONTRACT/GRANT NO.
68-02-1319, Task 18
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND P
Final; 1/75-10/75
ERIOD COVERED
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16.
rep0r£ gives process des criptions for petroleum refineries, for
synthetic natural gas plants based on liquid hydrocarbon feedstock, and for liquefied
natural gas plants. It compares these process descriptions with those for other
types of energy conversion plants, such as coal gasification and coal liquefaction.
It identifies potential ambient air emissions, liquid effluents, and solid wastes, and
discusses monitoring methods and control techniques for these emissions and wastes.
It identifies plant siting problems.
7.
KEY WORDS AND DOCUMENT ANALYSIS
• DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air Pollution
Liquefied Natural Gas
Petroleum Industry
Refineries
Wastes
Water Pollution
Monitors
Air Pollution Control
Stationary Sources
Synthetic Natural Gas
Solid Waste
13B
21D
05C
14 B
3. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report}
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |