vxEPA
United States Industrial Environmental Research EPA-600/7-79-178b
Environmental Protection Laboratory November 1979
Agency Research Triangle Park NC 27711
Technology Assessment
Report for Industrial
Boiler Applications:
Oil Cleaning
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. “Special” Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA s mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic,
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems: and integrated assessments of a wide’range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-79-178b
November 1979
Technology Assessment Report
for Industrial Boiler Applications:
Oil Cleaning
by
E.A. Comley, R.T. Keen, and M.F. Tyndall
Catalytic, Inc.
P.O. Box 240232
Charlotte, North Carolina 28224
Contract No. 68-02-2604
Task No. 2
Program Element No. INE825
EPA Project Officer: Samuel L. Rakes
Industrial Environmental Research Laboratory
Office of Environmental Engineering and Technology
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
FOREWORD
In the ensuing discussion of emission control technologies,
candidate technologies were compared using three emission control
levels labelled “moderate, intermediate, and stringent.” These control
levels were chosen only to encompass all candidate technologies and form
bases for comparison of technologies for control of specific pollutants
considering performance, costs, energy, and non-air environmental
effects.
From these comparisons, candidate “best” technologies for control
of individual pollutants are reconmiended by the contractor for
consideration in subsequent industrial boiler studies. These “best
technology” reconinendations do not consider combinations of technologies
to remove more than one pollutant and have not undergone the detailed
enviromuental, cost, and energy impact assessments necessary for
regulatory action. Therefore, the levels of “moderate, intermediate,
and stringent” and the reconinendation of “best technology” for
individual pollutants are not to be construed as indicative of the
regulations that will be developed for industrial boilers. EPA will
perform rigorous examination of several comprehensive regulatory options
before any decisions are made regarding the standards for emissions from
industrial boilers.
11
-------
PREFACE
The 1977 Amendments to the Clean Air Act required that emission standards
be developed for fossil-fuel-fired steam generators. Accordingly, the U.S.
Environmental Protection Agency (EPA) recently promulgated revisions to the 1971
new source performance standard (NSPS) for electric utility steam generating
units. Further, EPA has undertaken a study of industrial boilers with the
intent of proposing aNSPS for this category of sources. The study is being
directed by EPA’s Office of Air Quality Planning and Standards, and technical
support is being provided by EPA’s Office of Research and Development. As part
of this support, the Industrial Environmental Research Laboratory at Research
Triangle Park, N.C., prepared a series of technology assessment reports to aid
in determining the technological basis for the NSPS for industrial boilers.
This report is part of that series. The complete report series is listed below:
Title Report No .
The Population and Characteristics of Industrial! EPA-600/7-79-178a
Conunercial Boilers
Technology Assessment Report for Industrial Boiler EPA-600/7-79-l78b
Applications: Oil Cleaning
Technology Assessment Report for Industrial Boiler EPA-600/7-79-l78c
Applications: Coal Cleaning and Low Sulfur Coal
Technology Assessment Report for Industrial Boiler EPA-600/7-79-178d
Applications: Synthetic Fuels
Technology Assessment Report for Industrial Boiler EPA-600/7-79-178e
Applications: Fl uldized-Bed Combustion
Technology Assessment Report for Industrial Boiler EPA-600/7-79-l78f
Applications: NO Combustion Modification
Technology Assessment Report for Industrial Boiler EPA-600/7-79-178g
Applications: NO Flue Gas Treatment
Technology Assessment Report for Industrial Boiler EPA-600/7-79-178h
Applications: Particulate Collection
Technology Assessment Report for Industrial Boiler EPA-600/7-79-178i
Applications: Flue Gas Desul furl zati on
These reports will be integrated along with other information in the
document, “Industrial Boilers - Background Information for Proposed Standards,”
which will be Issued by the Office of Air Quality Planning and Standards.
111
-------
ABSTRACT
This study assesses the applicability of oil cleaning technology to
industrial boilers and is one of a series of technology assessment
reports to aid in determining the technological basis for a New Source
Performance Standard for Industrial Boilers. The status of development
and performance of alternative oil cleaning techniques were assessed and
the cost, energy, and environmental impacts of the most promising processes
were identified.
Hydrotreating processes (HDS) which produce cleaned liquid fuels
are considered the best system of emission reduction applicable to oil-
fired industrial boilers. The processes which clean oil by gasification
are either not generally suited to the small scale of industrial boilers
(POX) or are not conmiercially demonstrated (CAFB). The average capital
investment, as well as the overall energy requirements, increase with
increasing degree of desulfurization.
Sulfur oxide emissions and particulate emissions are highly dependent
on the sulfur and ash contents of the oil, respectively. Nitrogen oxide
emissions are dependent on fuel nitrogen content, as well as excess O ,
level and boiler size. In general, NO emissions decrease with decreasing
excess 09 and with increasing boiler s ze. The naturally low-sulfur,
low-ash bus tended to meet at least the recon,nended control levels for
moderate control in the oil cleaning category and quite often met even
the intermediate or stringent control levels. The high—sulfur, high-ash
oils, however, often failed to meet even moderate control levels, which
suggests the need for oil cleaning. The use of HDS as an SO 9 control
technology on industrial boilers results in the expenditure bf 2-4% of
the energy generated by the boilers. When the energy consumption of the
hydrogen plant is factored into a total desulfurization energy consumed,
the percentage of energy used increases subst ntially to 4.5 to 10.8%.
The cost impact of providing low sulfur distillate oil for firing
small coninercial boilers is minimal, amounting to just a 6.7% premium
for 0.3% S and 7.7% premium for 0.1% S oil.
The cost impact of using residual fuel oil is much more dramatic,
ranging from a premium of 6.7 to 18.6% when using oil desulfurized to a
leve’ of 1.6% S up to a premium of 39 to 43.1% when using oil desulfurized
to a level of 0.1% S.
The cost of HDS escalates quite rapidly with the degree of desulfuri-
zation in a given oil going from $0.91/B for a level of 1.6° ’ sulfur to a
cost of $5.28/B for desulfurizing to a level of 0.1% sulfur.
iv
-------
EXECUTIVE SUMMARY
1.1 Introduction
1.2 Hydrotreating Processes for
Producing Cleaned Liquid Fuels
1.3 Gasification Processes for
Producing Clean Gaseous Fuels
1.4 Selection of the Best System
1.5 Regulatory Options
1.6 Economic Impact of
Best Emission Control System
1.7 Energy Impact of Best System
1.8 Environmental Impact
2 OIL CLEANING AND CLEAN OIL EMISSION CONTROL
TECHNIQUES FOR OIL-FIRED INDUSTRIAL BOILERS
2.1 Principles of Control
2.2 Hydrotreating Processes for
Producing Cleaned Liquid Fuels
2.3 Performance of Cleaned Liquid
Fuels in Industrial Boilers.
2.4 Gasification Processes for
Producing Clean Gaseous Fuels.
2.5 Performance of Product Fuel
Gases on Industrial Boilers.
3 CANDIDATES FOR BEST SYSTEMS OF
EMISSION REDUCTION FOR CLEAN OIL TECHNOLOGY
3.0 Introduction . . . . 87
3.1 Selection Criteria . 87
3.2 Best Systems . . . . 88
3.3 Regulatory Options . 93
3.4 Suninary 105
ECONOMIC IMPACT OF BEST EMISSION CONTROL SYSTEM . 109
4.0 Introduction 109
4.1 Sun iiary 110
4.2 Process Description 115
4.3 Cost Basis 124
4.4 Sensitivity Analysis 141
4.5 Appendix - Sample Calculations 159
(*) References are tabulated at the end of each section.
CONTENTS *
FOREWORD
PREFACE
ABSTRACT
Section
Page
11
111
iv
1
1
4
9
12
18
20
25
27
36
36
41
64
66
81
87
4
V
-------
CONTENTS (cont.)
Section Page
ENERGY IMPACT OF BEST
EMISSION CONTROL SYSTEM FOR CLEAN OIL
5.1 Introduction
5.2 Energy Impact of Controls for Oil-Fired Boilers.
5.3 Sumary
ENVIRONMENTAL IMPACT OF DESULFURIZATION
TECHNIQUES FOR THE PRODUCTION OF LOW-SULFUR FUEL OIL.
6.1 Introduction
6.2 Air Pollution
6.3 Water Pollution
6.4 Solid Waste
6.5 Other Environmental Impacts
EMISSION SOURCE TEST DATA
7. 1 Introduction
7.2 Test Results
7.3 Test Methods
7.4 Sinnary
Glossary of Terms
Sulfur Recovery Systems for Off-Gas Treatment.
5
6
7
Appendix A
Appendix B
• 165
• 165
• 167
• 185
191
• 191
• 195
• 204
• 211
• 219
222
222
• 222
229
• 230
• 233
• 236
vi
-------
FIGURES
Figure Page
1-1 Basic HDS Process (Gulf HDS - Type II) 6
2-1 Simplified Crude Oil Refinery Flowsheet for Fuel
Oil Production 40
2-2 Basic HDS Process (Gulf HDS - Type II) 44
2-3 Improved HDS Process (Gulf HDS - Type III) 45
2-4 State of the Art HDS Process (Gulf HDS - Type IV) 46
2-5 Chemical Hydrogen Consumption in Desulfurization of Residua 49
2-6 Hydrogen Consumption for Atmospheric and Vacuum Residua . 50
2-7 Effect of Metals on Catalyst Consumption 52
2-8 Use of Guard Reactor in HDS Processes 54
2-9 Catalyst Bed Temperature vs. Time 55
2-10 Integration of Direct HDS Into a Refinery 57
2-11 Utilization of Indirect HDS in a Refinery 58
2-12 Effect of Fuel Sulfur Content on SO 2 Emissions 67
2-13 Effect of Fuel Nitrogen Content on Total Nitrogen Oxides
Emissions 68
2-14 Effect of Fuel Carbon Residue Content on Total Particulate
Emissions 69
2-15 Typical Partial Oxidation Process Flowsheet (Shell
Gasification Process) 71
2-16 Shell Pelletizing System for Soot Recovery and Recycle 72
2-17 Shell Closed Carbon Recovery System 73
2-18 Chemically Active Fluidized Bed Process Schematic Flow
Diagram 78
2-19 CAFB Gasifier/Regenerator 79
3-1 Effect of Fuel Sulfur Content on SO 2 Emissions 94
3-2 Effect of Fuel Nitrogen Content on Total Nitrogen Oxides. . . . 95
3-3 Effect of Fuel Carbon Residue Content on Total Particulate
Emissions 96
vii
-------
FIGURES (continued)
Figure Page
4—1 Cost of Desulfurized Residual Fuel Oil 112
4—2 Cost of Residual Fuel Oil 113
4—3 Typical HDS Unit 119
4—4 Typical Hydrogen Plant 120
4—5 Typical Claus Sulfur Plant 121
4—6 SCOT Off-Gas Treating Process 123
4—7 HDS Plant Cost 128
4-8 Hydrogen Plant Cost 131
4—9 Claus Sulfur Plant Cost 132
4-10 Hydrogen Consumption in Desulfurization of l6°API Residua. . . 137
4—11 Cost Distribution for Desulfurization of Three Residua to
Moderate Sulfur Levels 154
4-12 Cost Distribution for Desulfurization of Three Residua to
Stringent Sulfur Levels. 155
5-1 Energy Consumption vs. Level—of-Control for Distillate
FuelOil 174
5-2 Fossil Fuel Energy Consumption of High/Medium/Low Sulfur
Residual Fuel Oils vs. Level—of-Control 175
5-3 Electrical Energy Consumption of High/Medium/Low Sulfur
Residual Fuel Oils vs. Level-of-Control 176
5-4 HP Steam Consunq,tion of High/Medium/Low Sulfur Residual
Fuel Oils vs. Level-of-Control 177
5—5 LP Steam Consumption of High/Medium/Low Sulfur Residual
Fuel Oils vs. Level-of-Control 178
B-l Typical Packaged Claus Plant 237
B-2 Emission Control Systems for Refinery Claus Plants . . . 240
B-3 Flow Diagram for the Beavon Sulfur Removal Process . . . 241
B-4 Flow Diagram for the Cleanair Claus Tail-Gas Treatment
Process . 243
V ii 1
-------
FIGURES (continued)
Figure
B—5 Flow Diagram for IFP-2 Claus Tail-Gas Clean-up Process . . . 245
B-6 Flow Diagram for the Shell Claus Off-Gas Treating Process. . 246
B-7 Sulfinol Process Flow Diagram 248
B—8 Flow Diagram for the Sulfreen Process 250
B—9 Flow Diagram for the Wellman-LordSO 2 Recovery Process . . . 251
B-lU Allied Chemical SO 2 Reduction Process 253
B-il FWEC RESOX System for Sulfur Recovery 255
ix
-------
TABLES
Table Page
1-1 Estimated Uncontrolled Emissions From the Industrial!
Commercial Boiler Population 2
1-2 Typical Properties of LSFO Product From Gulf HDS
Processes 7
1—3 Typical Product Gas Composition From Gasification
Processes 11
1—4 Rating Matrix for Selection of Best System of Emission
Reduction for Cl ean Oil Technology 14
1—5 Suggested Pollutant Content of Cleaned Fuel Oil to Meet
Recommended Levels 19
1—6 Summary of Costs of Hydrodesulfurization of Residual
Fuel Oil 21
1—7 Cost Distribution for Three Residua 23
1—8 Cost Impact of Low-Sulfur Fuel Oil Firing in Boilers 24
1—9 Cost Effectiveness 26
1—10 Energy Consiinption vs. Control Level in HDS Processes 28
1—11 Environmental Impact of a Fuel Oil Refinery Producing
3.0% S Residual and 0.5% S Distillate Oils 29
1—12 Estimated Air Emissions Summary 30
1—13 Refinery Wastewater Effluent Quality 32
2—1 Estimated Uncontrolled Emissions From the Industrial/
Commercial Boiler Population 37
2-2 Chemistry of Hydrodesulfurization Reactions 42
2-3 Typical Properties of Low-Sulfur Fuel Oils From Gulf
HDS Processes 48
2-4 Survey of U. S. Refinery Desulfurization Capacity -
January 1978 60
2-5 Commercial HDS Results 61
2—6 Contaminant Removal in Hydrotreating Processes 65
2-7 Typical Product Gas Composition From Gasification Processes. 75
x
-------
TABLES (continued)
Table Page
3-1 Rating Matrix for Selection of Best System of Emission
Reduction for Clean Oil Technology . . 89
3-2 Suggested Pollutant Content of Cleaned Fuel Oil to Meet
Recommended Control Levels 97
3-3 Commercial Hydrodesulfurization Technology 98
3-4 Effect of Crude Switches on 1 -IDS Unit Capability 104
3-5 Characteristics of Typical Hydrodesulfurization Processes
by Level of Reduction in Sulfur Content 106
4-1 Sunuiiary Costs of F -Iydrodesulfurization of Residual Fuel
Oil 110
4-2 Cost Distribution for Three Residua 114
4-3 Cost Impact of Low-Sulfur Fuel Oil Firing in Boilers . . 116
4-4 Cost Effectiveness 117
4-5 HDS Plant Details for Hydrodesulfurization of Residual
FuelOils 125
4-6 HDS Plant Investment 126
4—7 Economic Indicators 127
4-8 Distillate Desulfurization Details 130
4-9 HDS Plant Investment 133
4-10 Annual Unit Costs for Operation and Maintenance 136
4-11 Labor and Supervisory Costs for Hydrodesulfurization of
Residual Oil 138
4-12 Hydrogen Production Costs 140
4-13 Total Cost of Desulfurization of Ceuta Residual Fuel Oil . . . 142
4-14 Total Cost of Desulfurization of E. Venezuelan Residual
Fuel Oil 143
4-15 Total Cost of Desulfurization of Kuwait Residual Fuel Oil. 144
4-16 Total Cost of Desulfurization of Khafji Residual Fuel Oil. 145
4-17 Total Cost of Desulfurization of Cold Lake Residual Fuel
Oil 146
xi
-------
TABLES (continued)
Table Page
4-18 Total Cost of Desulfurization of Distillate Fuel Oil . 147
4—19 Utility Consumption and Cost for Desulfurization of
Residual Oil 149
4—20 Hydrogen Consiaiiption and Cost for Desulfurization
of Residual Oil 151
4—21 Catalyst Consumption and Cost for Desulfurization of
Residual Oil 152
4—22 Sulfur Removal for Various Control Levels 157
5—1 Energy Consumption for SO 2 Control Techniques for Oil-
Fired Boilers 168
5-2 Energy Consun tion for Sulfur Level-of—Controls in
Residual Oil 186
6-1 Environmental Impact of a Fuel Oil Refinery Producing
3.0% S Residual and 0.5% S Distillate Oils 194
6-2 Baseline Fuel Oil Refinery Atmospheric Emissions . 197
6—3 Estimated Air Emissions, 3.0% S Fuel Oil 200
6-4 Estimated Air Emissions, 0.8% S Fuel Oil 201
6-5 Estimated Air Emissions, 0.1% S Fue’ Oil 202
6—6 Estimated Air Emissions - Residual Oil Combustion. 203
6-7 Refinery Wastewater Effluent Quality . 205
6—8 Refinery Water Management Plan . . . . 207
6—9 Sludge Incineration Emissions 213
7-1 Emission Source Test Data - S0, . . . . 223
7—2 Emission Source Test Data - N0 225
7—3 Emission Source Test Data - Solid Particulate. . . 227
x l i
-------
SECTION 1
EXECUTIVE SUMMARY
1.1 INTRODUCTION
Purpose of Report
The purpose of this report is to assess the available oil cleaning tech-
nology for control of emissions from oil-fired industrial boilers. Uncontrolled
industrial boilers using oil fuels emit significant amounts of particulates,
SON, and NO to the atmosphere. A recent study by PEDCo provided the data
shown in Table 1-1. PEDCo estimated the consumption of residual and distillate
fuels in the industrial section at 19,881 x 1O 3 m3/yr (125,067 x bbl/yr)
and 7281 x lO m3/yr (45,799 x lO bbl/yr), respectively, in 1975. In 1977,
residual oil fuel supplied 27.3% of the Btu capacity of industrial/commercial
boilers with distillate fuels providing 9.7%. Furthermore, PEDCo projects that
oil-fired industrial/comercial boiler capacity will increase by approximately
2 3 times 1977 values by the year 2000. Obviously, without controls, emissions
would increase roughly in proportion to fuel use.
Pollutant Formation Mechanisms
Sulfur oxide emissions are directly related to the sulfur content of the
fuel. The combustion of cleaner, low-sulfur oil will result in lower SO, emis-
sions to the atmosphere. Nitrogen oxides are formed from both oil-bound nitro-
gen and nitrogen in the combustion air. Decreased nitrogen in fuels will reduce
N0 emissions, but it will not affect thermal fixation of atmospheric nitrogen.
Staged combustion, low excess oxygen, and flue gas recirculation may be used for
controlling NOx emissions originating from thermal fixation of atmospheric ni-
trogen during the combustion of clean oils. 2 Particulate emissions are de-
pendent on the fuel characteristics, such as carbon residue and ash, which are
reduced in oil treatment. Therefore, particulate loadings in general should
be lower when cleaner fuels are being burned.
—1—
-------
TABLE 1—1. ESTIMATED UNCONTROLLED EMISSIONS FROM THE
INDUSTRIAL/COMMERCIAL BOILER POPULATION(l)
Estimated Emissions (1975) Mg/yr (tons/yr)
Boiler Type Particulate SO
Water Tube
Residual Oil Fired 59,900 (66,000) 794,500 (875,800) 198,600 (218,900)
Distillate Oil Fired 2,300 (2,500) 38,700 (42,700) 25,200 (27,800)
Fire Tube
Residual Oil Fired 22,200 (24,500) 294,700 (324,900) 73,700 (81,200)
Distillate Oil Fired 3,200 (3,500) 53,700 (59,200) 34,900 (38,500)
Cast Iron
Residual Oil Fired 14,500 (16,000) 192,200 (211,900) 48,000 (53,000)
Distillate Oil Fired 1,900 (2,100) 32,500 (35,800) 21,100 (23,300 )
TOTALS - Residual Oil Fired 96,600 (106,500) 1,281,400 (1,412,600) 320,300 (353,100)
Distillate Oil Fired 7,400 (8,100) 124,900 (137,700) 81,200 (89,600)
-2-
-------
Identification of Prime Control Mechanisms
There are two general approaches to reducing the emissions of SO> , NO
and particulates from the burning of oil as a fuel in industrial boilers.
These are:
1) by cleaning the flue gas output from the boiler; and
2) by reducing the input of these impurities to the boiler
by precleaning the fuel
The first approach, called Flue Gas Cleaning, is being addressed by other contractors.
This report discusses the oil cleaning techniques applicable to fuel oil frac-
tions which are used as industrial boiler fuels.*
There are various methods of cleaning fuel oils which are used by refin-
eries. The techniques fall into two general categories:
1) processes which produce a liquid fuel with reduced
contaminant content; and
2) processes which produce a gaseous fuel with reduced
contaminant content
The processes which produce cleaned liquid fuels are called hydrotreating.
They are chemical processes involving contact of the oil with catalyst and hy-
drogen. These processes convert much of the chemically-bonded sulfur and ni-
trogen to gaseous hydrogen sulfide and ammonia, respectively, thereby removing
them from the oil stream. In addition, the metals content of the oil is re-
duced, as is the carbon residue portion of the oil.
A second prime technique included in this report is the group of processes
which is designed to convert heavy, high—sulfur residual oils to clean, low-
sulfur gas. These processes include Partial Oxidation (POX) and the Chemically
Active Fluid Bed Process (CAFB). The processes which produce liquid fuels are
discussed under 1.2, and those which produce gaseous fuels are discussed
under 1.3.
* A glossary of terms used in petroleum technology will be found
in Appendix A.
—3—
-------
1.2 HYDROTREATING PROCESSES FOR PRODUCING CLEANED LIQUID FUELS
In the typical hydrotreating process, atmospheric resid is filtered to re-
move rust, coke, and other suspended material. It is then mixed with hydrogen, heated to
6500 to 850°F, and passed over one or more catalytic reaction beds.* Numerous
chemical reactions occur which lead to removal of most of the sulfur as H 2 S.
Since sulfur is the major impurity in petroleum, the general technique is fre-
quently called Hydrodesulfurization (HDS) in the industry.** However, the hy-
drogen reacts with other species besides sulfur compounds. For example, nitro-
gen compounds break down to liberate ammonia from the oil. This is called
denitrogenation or denitrification. The nickel and vanadium in the oil, which
are bound as organo-metal compounds, are liberated by reaction with the hydrogen.
This is called demetallization. Most of the liberated metals deposit (as the
sulfide) on the catalyst surface or in its pores and slowly deactivate the cata-
lyst. Other reactions which take place break up large complex molecules such
as asphaltenes and lead to a reduction in carbon residue of the product oil.
By utilizing catalysts, the reactions with hydrogen can be restricted
largely to the types above which take place under moderate reaction conditions.
Without the catalysts, higher reaction temperatures or pressures would be re-
quired; and, this would lead to greatly-increased hydrogen consumption due to
hydrogenation of aromatic ring systems which are abundant in high—boiling
petroleum fractions.
* The most widely-used catalysts are composites comprised of
cobalt oxide, molybdenum oxide, and alumina, where alumina
is the support carrying the other agents as promoters.
However, other catalyst materials are in use or under de-
velopment. Much of the catalyst technology is prqprietary,
but the patent literature is extensive. Ranney I 3) has com-
piled information from over 200 patents during the period
1970-1975.
A discussion of collection and processing techniques for
the evolved H 2 S will be found in Appendix B.
-4-
-------
Many companies are engaged in developing and using catalytic hydrotreat-
ing (or hydrodesulfurization) processes. All processes are similar
in basic concept and vary only in details such as catalysts, process conditions,
and complexity. A recent paper by Gulf Oil Company investigators traces the
development of one basic process. Figure 1-1 illustrates the simplest com-
mercial version currently marketed by Gulf, and it is known as Gulf II HDS.
Its basic elements are a feed filter, heater, single-stage catalytic reactor,
a gas/liquid separator, a fractionating column, and a gas treatment section.
This simple process system is capable of producing fuel oil of approximately
1% sulfur from a feedstock such as atmospheric resid containing 2-4% sulfur.
To produce a lower sulfur content product, additional catalytic reaction stages
must be added. The most advanced process Gulf has developed is known as
Gulf IV. It uses three catalytic reactors and can produce fuel oils of approx-
imately 0.1% sulfur.
Table 1-2 illustrates the typical product properties obtained when the
three versions of the Gulf process are applied to Kuwait atmospheric resid
containing 3.8% suiqur. It can be seen that the number of catalyst stages
strongly affects both physical and chemical properties of the product oil. In
addition to sulfur removal, other changes are noted: reduction in pour point
and viscosity; and reduction of chemical impurities such as nitrogen, metals,
salts, and ash.
There is a price to pay for such beneficial changes, however. The amount
of hydrogen consumed increases with the degree of desulfurization. So does
catalyst cost. A further cost is a slight reduction in heating value per gal-
lon. Even though heating value/pound increases with the degree of desulfuri-
zation, the density of the product decreases; thus, slightly greater volumes
of cleaned fuel oil must be burned to produce the same amount of heat provided
by an untreated residual oil. The change in heating value is of the order of
-5-
-------
Figure 1-1
BASIC HDS PROCESS (GULF HDS - TYPE
Recycle Gas
Treatment
Fractionating Column
I Off Gas
Acid Gas (H 2 S)
Light H.C.
Frayer, J.A., et al., “Gulf’s HDS Processes for High Metal Stocks,”
Institute, Tokyo, Japan, 8 May 1975.
Hydrogen
Reduced
Recycled Hydrogen Gas
Filter
Heater
Compressor
Catalyst
Vessel
Gas
j Unreacted Hydrogen plus H 2 S
—-—---—- -a
Used ,,ith permission taken from:
presented at the Japan Petroleum
-------
TABLE 1-2. TYPICAL PROPERTIES OF LSFO PRODUCT FROM GULF
HDS PROCESSES
Gulf Gulf Gulf
Untreated II III iv
Product Yield: Vol. % -- 89.4 97.5 97.1
Product Properties:
Cut Point: °F 650 650 375 375
Gravity: °API 16.6 20.0 23.4 24.1
Sulfur: Wt. % 3.8 1.0 0.3 0.1
Carbon Residue: Wt. % 9.0 5.31 3.33 2.75
Nitrogen: Wt. % 0.22 0.13 0.13 0.09
Nickel: PPM 15.0 4.6 1.3 0.4
Vanadium: PPM 45.0 8.2 2.2 1.0
Viscosity: SUV (210°F) 250 107.3 52 45
Ash: Wt. % 0.02 0.004 0.003 0.003
Salt: PPM* 44.9 0 0 0
Heat of Combustion: Btu/LB ** 19,110 19,250 19,375
Hydrogen: Wt. % ** 12.1 12.5 12.7
Carbon: Wt. % ** 86.7 87.1 87.1
Pour Point: °F ** +60 +35 0
Hydrogen Consumption: SCF/BBL 497 663 812
* Salt refers to all water-soluble cations, determined as
halide and reported as NaCl before desalting.
** Not stated.
—7—
-------
1—3% on atmospheric resids and about 5% on vacuum resids. When the degree of
cleanup is considered, this energy impact seems insignificant.
The composition of the feedstock to a hydrotreater strongly influences
the amount of hydrogen and catalyst consumption in the process. Neison(6) has
correlated hydrogen consumption with sulfur reduction for a variety of resid
feeds. For example, to obtain 90% reduction in sulfur for an 18°API feedstock,
about 650 scf of hydrogen are consumed per barrel of oil processed whereas a
4°API feed would require 1200 scf/barrel. In another of Nelson’s correlations,
he plotted hydrogen consumption data for typical low metals atmospheric resid
(16°API) and vacuum resid (6°API) as a function of sulfur desired in the fuel
oil product. Thus, in producing 0.3-1.0% sulfur fuel oils from atmospheric
resid, 345-775 scf of hydrogen per barrel are required, while vacuum resid re-
quires 870 to 1150 scf to accomplish the same task. The problem caused by
metals is deposition onto the catalyst surface or in the pores. This leads to
deactivation of the catalyst, which is only overcome by an increase in bed
temperature and/or hydrogen recirculation rate in order to maintain acceptable
processing rates. Any increase in required severity of process conditions
leads to more hydrocracking with a subsequent increase in hydrogen consumption.
A further complication from the metals content of the feed is a shortening
of catalyst life. Even though some deactivation can be tolerated, the resul-
tant increase in hydrogen uptake means catalyst must be changed out sooner.
The effect of metals was shown in another Nelson corre1ation.’ 7 As an example,
for 90% sulfur removal from a 25 ppm metals content feedstock, about 27 barrels
of oil can be processed per pound of catalyst; to achieve the same sulfur per-
formance with a 100 ppm metals content feedstock, only 4.5 barrels can be pro-
cessed per pound of catalyst; a 300 ppm feedstock requires almost one pound of
catalyst per barrel. Clearly, high metal feedstocks are a costly problem to
-8-
-------
the refiner. Most refiners are using a separate stage of lower cost catalyst
material prior to the special hydrodesulfurization catalysts. These separate
stages may be packed with a material such as alumina or clay, which collect
the metals and “guards” the subsequent high activity catalyst. For this rea-
son, some refiners call this stage a “guard reactor” or “guard vessel.” Exxon
uses such a system in their RESlDfining processJ 8
From the above, it is obvious that catalyst life is quite variable. In
a fixed-bed system, catalyst changes are usually once or twice per year, de-
pending on product requirement and feedstock composition. For example, with
the Gulf IV system, producing 0.1% LSFO from 3.8% S Kuwait Resid (97% HDS), a
catalyst life of six months is the design base. Similarly, using Arabian Heavy
with RESlDfining at 75% HOS, a catalyst life of 400-500 days can be achieved.
Most refiners use fixed catalyst beds, which require a process shutdown
for catalyst replacement. However, there are some advocates of expanded or
ebullated beds with intermittent catalyst feed and bleed systems to maintain
continuous operation at relatively constant conditions. The H-Oil process de-
veloped by Hydrocarbon Research, Inc. and utilized by Cities Service is an ex-
ample of such an operating system. The expanded bed operates as a back mix
reactor with the entire bed at a constant temperature. Such a system is less
prone to plugging but has a slightly greater hydrogen consumption than a fixed
bed system, according to Nelson. 6 It is most suitable for high metals feed-
stocks where catalyst life is shortJ 7
1.3 GASIFICATION PROCESSES FOR PRODUCING CLEAN GASEOUS FUELS
Process Description
Gasification processes can convert high-sulfur feedstocks into a fuel gas
by controlled partial oxidation (POX) with air or oxygen. No catalyst is re-
quired; thus, a wide range of fuels can be converted to clean gaseous fuels.
The desalted feedstock is partially oxidized at high temperature to form a
-9—
-------
gaseous mixture of carbon monoxide (CO) and hydrogen (H 2 ) with a small amount
of methane (CH 4 ). Carbon soot produced as a result of incomplete combustion
is recycled back to the process. Either oxygen or air may be used for the
partial oxidation, depending on the desired heating value of the product gas.
When air is used, the nitrogen remains in the product gas. When oxygen is
used, the peak temperatures are usually controlled by introducing a diluent
such as steam or carbon dioxide.
In a typical application of this process, a heavy residuum and the oxidant
(air or oxygen) are preheated with steam and fed to the reactor. The hot re-
actor effluent [ 2400°F (1316°C)] gas containing some ash from the feed and
soot (1-3% by wt. of fuel) is passed to a specially-designed waste heat boiler
which produces high pressure steam. The crude gas exiting the waste heat sys-
tem [ about 325°F (163°C)] is then passed into a carbon removal system, which
consists of two units. The bulk of the carbon is removed by a special carbon-
water contactor of proprietary design; the remaining carbon is removed by a
cooler/scrubber water wash. The product gas contains less than 5 ppm carbon
and virtually no ash or other particulates. It is directly usable for gas-
turbine fuel after sulfur removal (see Appendix B for a discussion of processes
for removal of H 2 S from hydrocarbon gases).
The desulfurized product gas has a heating value of approximately 300
Btu/Scf when oxygen is used for the oxidation. By comparison, the product from
air oxidation has a heating value of approximately 120 Btu/Scf because of the
nitrogen remaining in the gas. Typical product-gas compositions for oxygen
and air gasification are shown in Table 1-3)
A recent development partially funded by the EPA has resulted in the
design, construction, and testing of the Chemically Active Fluid Bed (CAFB)
process demonstration unit in San Benito, Texas.(10) The CAFB process is de-
signed to produce a clean, low-sulfur gas by partially oxidizing a heavy,
-10-
-------
TABLE 1-3. TYPICAL PRODUCT GAS COMPOSITION FROM GASIFICATION PROCESSES 9
% vol., dry basis
02 Air
Oxidation Oxidation
Hydrogen 48.0 12.0
Carbon Monoxide 51.0 21.0
Methane 0.6 0.6
Nitrogen 0.2 66.0
Argon 0.2 0.4
Sulfur* 5 ppm 5 ppm
Total 100.0 100.0
* after desulfurizing using Shell Sulfinol or ADIP
process
Used with permission taken from: Kuhre, C. J. and J. A. Sykes, Jr.,
Energy Technology Handbook , D. M. Considine, Ed., McGraw-Hill Book Company,
1977, pages 2-173 and 2—174.
—11—
-------
high-sulfur feed in a limestone fluid bed. Hydrogen sulfide and some organic
sulfur are absorbed by the lime. The remaining hot, low-sulfur fuel gas pro-
duced is ready for combustion. The CAFB reactor contains two sections; one
for gasification of the feed, and one for regenerating the sulfur-containing
limestone. During regeneration, air reacts with the spent stone, freeing the
sulfur as sulfur dioxide. The sulfur dioxide is removed from the regeneration
gas and may be recovered in a variety of processes. The limestone is then re-
cycled to the gasification section until it loses its efficiency as an absor-
bent. This unit is currently scheduled for operation in the suniiier of 1979.
Commercial performance and reliability of the CAFB process are yet to be
determined. Until this reliability is demonstrated, the CAFB concept is not
likely to make a significant impact on the boiler-fuel picture. However, the
simplicity and wide fuel flexibility of the CAFB as a front-end add-on make
it a potentially very attractive new technology which could rapidly become im-
portant, especially for large new chemical complexes with several industrial
boilers.
1.4 SELECTION OF THE BEST SYSTEM
The factors considered in the selection of the best system of emission
reduction from those systems discussed above were:
1. Performance
2. Applicability
3. Status of Development
4. Cost Considerations
5. Energy Considerations
6. Environmental Considerations
In the selection process, an effort was made to rate each of the control
systems against each of the selection criteria listed above. For each criter-
ion, the best system was designated 1, the next best 2, etc. The lowest
—12-
-------
overall score for all criteria was adjudged to be the best system. A summary
of the rating evaluation is given in Table 1-4.
From these rating results, it is concluded that Hydrotreating (HDS) of-
fers the best system of emission reduction for clean oil technology. A dis-
cussion of the ratings for the different selection criteria is given below:
a) Performance — All three systems will yield a fuel that is
environmentally acceptable for burning in a boiler. The
overriding consideration in selecting HDS as the best
system is its negligible or minor impact on boiler per-
formance. Cleaned liquid fuels are directly applicable
to existing industrial boilers with little negative im-
pact on boiler performance. The modest reduction (1-3%)
of heat content per gallon of cleaned oil will require
additional fuel consumption to achieve rated boiler out-
put. However, burning cleaner fuels will lower the
severity of operation and maintenance on the boiler.
b) Applicability - The HOS system is a clear selection for
applicability, since the clean oil produced from HDS can
be directly utilized in existing boilers with little
impact on the boiler physical facilities. The gasifica-
tion processes require the addition of equipment with
attendant cost and space impacts. The retrofitting of
either gasification process to an existing industrial
boiler could, in some cases, be extremely difficult, if
not impossible.
—1 3—
-------
TABLE 1-4. RATING MATRIX FOR SELECTION OF BEST SYSTEM OF
EMISSION REDUCTION FOR CLEAN OIL TECHNOLOGY
Control S stem 1 2 3 4 5 6 Total
UDS 1 1 1 1 1 1 6
pox 2 2 2 3 2 3 14
CAFB 3 3 3 2 3 2 16
NOTE: Selection Criteria
1. Performance
2. Applicability
3. Status of Development
4. Cost Considerations
5. Energy Considerations
6. Environmental Considerations
-1 4-.
-------
c) Status of Development - Hydrotreating processing has been
in commercial existence for more than 20 years, and over
20 hydrotreating processes are actively in use. 0 The
current United States refinery desulfurization capacity
is more than 1.8 million barrels per day from 86 plants.
However, only 19 plants have direct resid or heavy gas
oil hydrodesulfurization facilities which provide a total
desulfurization capacity of approximately 0.6 million
barrels per day.l 2)
Hydrotreating is an extremely versatile process which is
used to desulfurize, denitrogenate, and demetallize fuel
oils prior to contustion. It can be adapted to a wide
variety of feedstocks ranging from low-sulfur crude oils
to high-sulfur residual oils. Hydrotreating used in
conjunction with blending can produce fuel oils with
almost any characteristics desired. Because of its ver-
satility and widespread use, hydrotreating has been se-
lected as the best candidate under the status of develop-
ment criterion.
Partial oxidation is a commercially-proven process with
more than 200 installations worldwide. 3 It, too, is
a versatile process that has been used with feedstocks
ranging from natural gas through naphtha, residual oil,
and even coal. Its primary use has been to produce syn-
thesis gas for the manufacture of methanol or ammonia;
however, there is no technical reason that the synthesis
gas cannot be used as a boiler or turbine fuel. The pro-
cess can be designed to use either air or oxygen as the
-1 5—
-------
oxidizing medium, and the product gas will have a heating
value of 120 Btu/SCF or 300 Btu/SCF dependent upon whether
air or oxygen is used.
The Chemically Active Fluid Bed (CAFB) process is an
attractive new technology which could become a significant
factor over the next few years in both the utility and in-
dustrial areas. It, too, is a versatile process which can
be used with a wide selection of feedstocks. The coniner-
cial performance and reliability of the CAFB process are
yet to be determined. Until this reliability is demonstrat-
ed, the CAFB concept is not likely to make a significant
impact on the boiler—fuel picture. A CAFB demonstration
unit has been constructed and is currently undergoing test-
ing at the La Palma station of Central Power and Light
Company at San Benito, Texasil4)
d) Cost Considerations — The cost of upgrading liquid fuels
in a few large refinery complexes is less demanding than
individual emission control techniques at each industrial
boiler. In the first place is the matter of mere ntmibers,
wherein we are comparing the cost of desulfurization fa-
cilities at less than 100 refineries against individual
emission control systems at literally thousands of indus-
trial boiler installations. In addition, the economics of
scale would greatly favor the installation of large central
hydrotreating units at refineries rather than smaller in-
stallations at individual boilers. For example, a 150 x
106 Btu/hr oil-fired boiler requires approximately 25
barrels/hr of oil or 600 barrels per day. A typical
-16-
-------
50,000 BPD HDS unit could supply nearly 85 such boilers.
Preliminary cost figures show that typical hydrotreating
facilities can be installed for investments ranging from
$500 per barrel per stream day capacity for moderate le-
vels of emission control up to $1,600 per barrel per
stream day capacity for the most stringent levels of con-
troi.(15) The costs of producing such cleaned fuels range
from $0.91 to as much as $5.84 per barrel. The cost of
partial oxidation units ranges from $3,000 to $6,000 per
barrel per stream day with operating costs ranging from
$1.50 to $7.00 per barrel. 6) Cost information on CAFB
units is very sketchy, but preliminary figures indicate
an investment cost of $3,500-$4,500 per barrel per stream
day and an operating cost of about $4.00 per barrel)16)
e) Energy Considerations - The selection of hydrotreating
under this criterion closely parallels the reasoning used
for the cost considerations in that the selection is largely
determined by numbers and sizes of units. The energy im-
pacts are reflected by the operating cost figures given in
the previous paragraph, which indicate that hydrotreating
has advantages over both partial oxidation and chemically
active fluid bed processing.
f) Environmental Considerations — There is little difference
between the three control systems from an environmental
viewpoint. From a practical viewpoint, it is more advan-
tageous to burn cleaned liquid fuels in industrial boilers
rather than t* handle the environmental problems of untreated
—17—
-------
fuel oil at each industrial site. The control of poten-
tially-hazardous pollutants can be more effectively manag-
ed at the refinery, and the emissions from combustion at
industrial boilers can then be more effectively controlled
by monitoring the fuel quality.
1 .5 REGULATORY OPTIONS
Three regulatory options, which represent moderate, intermediate, and
stringent levels of control for SO 2 , NOR. and particulate emissions, have been
selected. The selected emission levels were derived from actual emission data
from industrial boilers.
Table 1-5 gives the maximum sulfur, nitrogen, and carbon residue content
of cleaned fuel oil to meet the recommended control levels. The suggested
regulatory options are based upon commercially-available systems for the pro-
duction of low-sulfur fuels from high-sulfur feedstocks. The selected levels•
of control are based upon the use of residual fuel oil and represent the degree
of desulfurization that can be attained using typical refinery processes and
technology.
For the moderate level of control, a suggested fuel content of 0.8% sul-
fur, 0.3% nitrogen, and 12% carbon residue represents a residual fuel oil that
is readily achievable from a niinber of refinery practices. For the stringent
level of control, a suggested fuel content of 0.1% sulfur, 0.2% nitrogen, and
3% carbon residue represents the highest technically achievable residual fuel
oil that can be attained with current technology. No Ii. S. refinery is cur-
rently producing such material, since there is no demand at present. For the
intermediate level of control, the suggested fuel content of 0.3% sulfur,
0.2% nitrogen, and 6% carbon residue represents a technology which can be met
by a number of available refinery, processes, although only about 120,000 BPD
capacity is currently installed in the U. S. An additional 75,000 BPD is
Coming on stream in 198O)
-18-
-------
TABLE 1-5. SUGGESTED POLLUTANT CONTENT OF CLEANED FUEL OIL
TO MEET RECOMMENDED CONTROL LEVELS
Control Level
Stringent Intermediate Moderate
Emission Max. Fuel Emission Max. Fuel Emission Max. Fuel
Pollutant #1106 Btu Content #1106 Btu Content #/106 Btu Content
SO 2 0.1 0.1% S 0.3 0.3% S 0.8 0.8% S
0.2 0.15% N 0.22 0.2% N 0.3 0.3% N
Particulates 0.05 3.0% C.R.* 0.1 6.0% C.R. 0.25 12% C.R.
*NOTE: % C.R. = weight percent carbon residue in fuel oil
-19-
-------
1.6 ECONOMIC iMPACT OF BEST EMISSION CONTROL SYSTEM
As in Section 1.4, we selected hydrodesulfurization (HDS) as the
best system of emission reduction for clean oil technology and recomended
guideline control levels of regulatory options to best achieve moderate (O.8%S),
intermediate (O..3%S), and stringent (O.l%S) levels of control.
In this section, we determine the cost of hydrodesulfurization to produce
cleaned fuel oils to meet the required control limits and assess the economic
impact of burning desulfurized oils in industrial boilers. In our cost analy-
ses, only direct desulfurization of residual fuel oil is considered. Indirect
desulfurization, or the procedure of desulfurizing a light distillate and back
blending with residua to produce the required product level, is not capable of
achieving the intermediate and stringent levels of control and therefore is riot
considered in this study.
The cost of hydrodesulfurization of residual fuel oil is a function not
only of the sulfur content but also of the crude source from whence the residual
was derived and the metal content of the residual. Since there are literally
hundreds of different crude oils and, consequently, a like number of residua,
it is virtually impossible to select a typical residual oil that would be repre-
sentative of all these crudes. Accordingly, we have selected a group of five
residual oils which cover a range of sulfur and metal values and which will
acconinodate virtually all the known crudes within the limits covered by these
five residuals. The five residua considered in this section can be classified
as follows:
-20-
-------
The cost of hydrodesulfurization is also highly dependent upon the degree
of desulfurization. In order to cover as wide a range as possible, the hydro-
desulfurization costs were calculated for the three recorm ended levels of con-
trol, as well as the State Implementation Plan (S.I.P.) level of 1.6% sulfur
currently being used in most of the United States.
A sumary of the hydrodesulfurization costs for the five residual fuel
oils and four levels of sulfur content is given in Table 1-6.
TABLE 1-6. SUMMARY COSTS OF HYDRODESULFURIZATION OF RESIDUAL FUEL OIL
Resi dua
1. Ceuta
2. E. Venezuelan
3. Kuwait
4. Khafji
5. Cold Lake
Classi fication
Low Sulfur, high metals
Low sulfur, high metals
Medium sulfur, low metals
High sulfur, moderate metals
High sulfur, high metals
Residual
Fuel Oil
%
Type Sulfur
Percent_Sulfur_in_Treated_Oil
ppm 1.6 0.8 0.3
(Ni + V) $/bbl $/bbl $/bbl
0.1
$/bbl
Ceuta 2.12
292 0.91 2.28 3.91
5.28
E. Venezuelan 2.38
274 1.17 2.45 3.93
5.71
Kuwait 3.80
60 1.80 2.49 3.14
3.51
Khafji 4.36
118 2.20 2.85 3.60
4.11
Cold Lake 4.55
236 2.52 3.42 4.53
5.84
As evidenced from the
foregoing table, the cost of HDS ranges
from a low
of $0.91 per barrel for the
hydrodesulfurization of a low-sulfur,
high-metals
residua to a high of $5.84
per barrel for the hydrodesulfurization
of a high-
sulfur, high-metals residua.
—21—
-------
It is also evident that the cost of HDS escalates quite rapidly with the
degree of desulfurization,gojng from $0.91/bbl for the desulfurization of Ceuta
residual to a level of 1.6% sulfur to a cost of $5.28/bbl for desulfurizing to
a level of 0.1% sulfur. This represents a cost of $14.46/bbl for the 1.5% S
Ceuta oil, or 39% over the cost of untreated oil.
The foregoing table also indicates that the cost of desulfurization to the
S.1.P. (1.6% S) and moderate (0.8% S) levels is primarily a function of the
sulfur level of the untreated oil; whereas, desulfurizing to the intermediate
(0.3% 5) and stringent (0.1% S) levels clearly reflects the influence of metals
content on desulfurization cost. It further shows that, regardless of the type
of residual feed, the cost of desulfurizing to very low levels such as 0.1% S
is substantial, ranging from $3.51 to $5.84 per barrel or 26 to 43% more than
the cost of untreated oil.
Table 1-7 gives a cost breakdown into the principal cost elements. This
table vividly illustrates the effect of hydrogen and catalyst costs on the over-
all cost of desulfurization which ranges from 33 to 61% of the total cost.
Table 1-8 gives the cost impact of low sulfur fuel oil firing in industrial
boilers. Data are presented for small (4.4MW, 15,000 MBtu/hr) cou,nercial-type
boilers and for large (44MW, 150,000 MBtu/hr) industrial-type boilers. From
Table 1—8, it is evident that the cost impact of providing low-sulfur distil-
late oil for firing small coninercial boilers is minimal, amounting to just a
6.7% premium for 0.3% S and 7.7% premium for 0.1% S oil. This small effect is
primarily brought about as a result of the small amount of desulfurization re-
quired to desulfurize regular No. 2 distillate oil, which usually contains
0.5% (or less) sulfur, to these lower sulfur levels.
The cost impact of using residual fuel oil is much more dramatic, ranging
from a premium of 6.7 to 18.6% when using oil desulfurized to a level of
1.6% S up to a premium of 39 to 43.1% when using oil desulfurized to a level of
0.1% S.
—22—
-------
TABLE 1-7.
COST DISTRIBUTION FOR THREE RESIDUA ( /bbl)
Percent Sulfur in Treated Oil
Residual Oil 1.6 0.8 0.3 0.1
Ceuta
Labor 6.9 6.9 6.9 6.9
Utilities 24.9 49.7 62.0 65.7
Investment, Maint.,
& Waste Disposal 31.5 70.1 105.8 129.6
Hydrogen 21.0 67.0 102.0 117.0
Catalyst 9.0 38.0 121.0 216.0
Total 93.3 231.7 397.7 535.2
Sulfur Credit 2.0 5.0 6.9 7.6
Net Cost 91.3 226.7 390.8 527.6
Kuwait
Labor 6.9 6.9 6.9 6.9
Utilities 47.8 59.4 73.4 76.9
Invest., Maint.,
& Waste Disposal 64.4 91.1 119.6 141.8
Hydrogen 58.0 88.0 107.0 115.0
Catalyst 11.0 15.0 20.0 24.0
Total 188.1 260.4 326.9 364.6
Sulfur Credit 8.3 11.3 13.2 13.9
Net Cost 179.8 249.1 313.7 350.7
Cold Lake
Labor 6.9 6.9 6.9 6.9
Utilities 54.2 64.8 80.4 82.9
Invest., Maint.,
& Waste Disposal 97.1 116.0 139.5 156.9
Hydrogen 70.0 96.0 113.0 120.0
Catalyst 35.0 72.0 129.0 234.0
Total 263.2 355.7 468.8 600.7
Sulfur Credit 11.1 14.1 16.0 16.7
Net Cost 252.1 341.6 452.8 584.0
—23—
-------
TABLE 1-8.
COST IMPACT OF LOW SULFUR FUEL OIL FIRING IN BOILERS
Cost Impact
System
Standard Boilers
Heat Input
MW jMBTU/HR)
44 (150,000)
Crude
Source
Annual Costs -
% Over
$/MBTU/ Iincon-
$LkJIS HR tro lle4
% Over
s.I.P.
Control led
Type
& Level
Type of Control
Watertube LSFO
Control
Efficiency
(%S)
4.4 (15,000)
Low
Ceuta
S.I.P.
1.6
4.03
1.19
6.72
N/A
Sulfur
Moderate
0.8
10.03
2.94
16.82
9.47
Resid
Intermediate
0.3
17.23
5.05
28.86
20.75
( 3% S)
Stringent
0.1
23.27
6.82
39.04
30.22
E. Venezuelan
S.I.P.
Moderate
Intermediate
Stringent
1.6
0.8
0.3
0.1
5.15
10.78
17.30
25.15
1.51
3.16
5.07
7.37
8.63
18.08
29.00
42.14
N/A
8.70
18.75
30.85
Medium
Kuwait
S..I.P.
1.6
7.92
2.32
13.28
N/A
Sulfur
Resid
(3.8% S)
Moderate
Intermediate
Stringent
0.8
0.3
0.1
10.95
13.82
15.46
3.21
4.05
4.53
18.38
23.17
25.90
4.50
8.73
11.14
High
Sulfur
Khafji
S.I.P.
Moderate
1.6
0.8
9.69
12.56
2.84
3.68
16.24
21.03
N/A
4.12
Resid
Intermediate
0.3
15.87
4.65
26.57
8.89
(> 4%S)
Stringent
0.1
18.12
5.31
30.33
12.12
Cold Lake
S.I.P.
Moderate
Intermediate
Stringent
1.6
0.8
0.3
0.1
11.09
15.08
19.96
25.73
3.25
4.42
5.85
7.54
18.60
25.24
33.43
43.10
N/A
5.60
12.50
20.66
Firetube
LSFO
Distillate
Fuel
Oil
N/A
S.I.P.
Moderate
Intermediate
Stringent
0.3
0.1
N/A
N/A
5.46
6.21
N/A
N/A
1.60
1.82
N/A
N/A
6.73
7.67
N/A
14/A
N/A
0.88
-------
Table 1-9 shows the cost effectiveness of fuel oil desulfurization for
the five residua considered, as well as the distillate fuel oil. Generally,
these data indicate that the cost effectiveness improves as the sulfur content
of the residuum feed rises, provided that the metal content does not increase
as well.
A comparison of the Kuwait and Khafji data shows the effect of similarity
between sulfur levels combined with relatively similar metal levels. The Cold
Lake data vividly show the strong effect of high-metal levels.
The data of Table 1-9 also indicate that, for a given feedstock, fuel oil
desulfurization tends to be less cost effective as the degree of desulfurization
increases. This effect ranges from 17% to 65%, depending on the specific resi-
duum; but, the trend is quite general.
1.7 ENERGY IMPACT OF BEST SYSTEM
Based upon the rating factors developed in Section 1.4, hydrodesulfurization
was selected as the best system for emission reduction for clean oil technology.
The HDS process cannot be considered on a stand-alone basis for the energy
impact assessment, since auxiliary processes are required to dispose of process
by-products. These auxiliary processes include a hydrogen sulfide absorption
unit (circulating amine type), sulfur recovery with tail-gas scrubbing (Claus
type with reduction system and tail-gas reheat), and sour water stripper
(steam stripping), and a hydrogen plant.
The production of low-sulfur distillate and residual fuel oils by pre-
combustion treatment methods such as hydrodesulfurization has the advantage of
scale from the energy impact point of view. The HDS unit is centralized, with
the attendant benefits of a large-scale operation. However, HDS systems re-
quire significant energy to operate. The process itself consumes about 2-4%
of the energy content of the oil produced. When the energy consumption of the
hydrogen plant is factored in, the percentage of energy used increases
-25-
-------
TABLE 1-9. COST EFFECTIVENESS
Crude
Source ___________ $1 lb $/KG
1.6 0.52 1.14
0.8 0.51 1.13
0.3 0.64 1.40
0.1 0.78 1.71
1.6 0.45 0.99
0.8 0.46 1.01
0.3 0.56 1.23
0.1 0.74 1.63
1.6 0.24 0.53
0.8 0.25 0.55
0.3 0.27 0.59
0.1 0.28 0.62
1.6 0.24 0.53
0.8 0.24 0.53
0.3 0.26 0.57
0.1 0.29 0.64
1.6 0.25 0.55
0.8 0.27 0.59
0.3 0.32 0.70
0.1 0.39 ( .86
0.2 0.20 0.44
0.1 0.21 0.46
Sulfur
In Fuel Oil
Cost/Unit Removal
Ceuta
E. Venezuelan
Kuwait
Khafj i
Co’d Lake
Distillate
-26-
-------
substantially. The data below indicate the total desulfurization energy ex-
penditure as a percentage of total boiler energy generated for the various sul-
fur control levels:
Energy Consumed As
Sulfur in a Percentage of
Fuel Oil Energy Generated
1.6% S 4.5%
0.8% S 5.6%
0.3% S 8.6%
0.1% S 10.8%
Table 1-10 shows energy consumption for the HDS and hydrogen plant utilities
as a function of desulfurization levels. It is apparent that, to achieve the
lower sulfur level fuels (0.3% S or 0.1% S) for industrial boiler combustion with—
out controls, substantial energy inputs are needed as the degree of desulfuriza-
tion increases.
1.8 ENVIRONMENTAL IMPACT
A basic fuel oil refinery without an HOS system produces a variety of waste
products which include air and water pollutants, as well as solid waste products.
As a baseline, the data shownin Table 1-li indicate the amounts of such pollu-
(18)
tants produced.
As a consequence of adding an HDS unit, there are only minor increases in
the quantities of air and water pollutants produced. The major waste product
comes from spent catalyst. With regard to the air emissions, the important
factor to be considered is the net impact, i.e., reduction of emissions in in-
dustrial boilers versus increased refinery emissions.
First to be considered are the emissions to the air. Table 1-12 shows
emissions of SO> , NO, and particulate for uncontrolled (3.0% S), 0.8% S, and
0.1% S for the refinery processes and for industrial boiler combustion. Over
the range from 3.0% to 0.1%, the refinery emissions increase slightly for
-27-
-------
TABLE 1-10. ENERGY CONSUMPTION FOR SULFUR LEVEL-OF-CONTROLS IN RESIDUAL OIL
Sulfur in Fuel Oil
Crude Class Utility 1.6% 0.8% 0.3% 0.1%
Low Sulfur Powe 2.9 6.4 8.2 8.7
3% S Steam 12.8 28.3 44.4 48.6
Fuel 107.7 262.9 360.0 398.7
Cooling Water .31 .63 1.1 1.2
Med. Sulfur Power 5.8 7.7 10.0 10.5
3—4% S Stear 31.0 43.8 63.8 71.2
Fuel 227.5 314.8 364.2 393.1
Cooling Water .65 .94 1.2 1.2
High Sulfur Power 7.0 8.9 10.7 11.0
4% S Steam 37.4 49.2 79.6 84.4
Fuel 257.4 334.0 382.2 403.1
Cooling Water .75 .99 1.15 1.23
NOTE: Utility Units
Power in K 4/8b1
Steam in P .tu/Bbl
Fuel In MBtu/Bbl
Cooling Water in MSal/Bbl
-28-
-------
TABLE 1-11. ENVIRONMENTAL IMPACT OF A FUEL OIL REFINERY
PRODUCING 3.0% S RESIDUAL AND 0.5% S DISTILLATE OILS 18 )
Air Emissions (1b./ 1 BBL Crude Processed)
Particulates 63.2
S0 160.0
NOx 118.3
CO 12.0
Hydrocarbons 740
Water (lb./?i BBL Crude Processed)
Suspended Solids 2.5
Dissolved Solids 92.6
0rg nic Material 0.5
Solid Wastes (lb./M BBL Crude Processed)
Catalysts 20
Other* 60
Source : “Environmental Problem Definition for Petroleum
Refineries, Synthetic Natural Gas Plants, and
Liquefied Natural Gas Plants,” Radian Corp.,
November 1975, EPA-600/2-75-068.
* Includes: (1) entrained solids in the crude; (2) corro-
sion products; (3) silt from drainage and
influent water; (4) maintenance and cleaning
solids; and (5) waste water treatment facil-
ities.
-29-
-------
TABLE 1-12. ESTIMATED AIR EMISSIONS SIJ* ARY
so, NO P rticu1ate
( lb/b 6 Btu Fuel Oil) ( lb/b 6 Btu Fuel Oil) ( lb/b 0 Btu Fuel Oil
Emission
Sources 3.O%S. O.8% O.l%S 3O%S O.8%S O.l%S 3.O%S O.8%S O.T%
Refinery
Processes .023 .030 .05 .019 .019 .019 .010 .010 .011
Industrial
Boiler
Combustion 3.170 0.85 0.11 0.40 0.30 0.1.1 0.22 0.12 0.0
Total 3.193 0.88 0.25 0.419 0.32 0.22 0.23 0.13 0.0
-30-S
-------
SO and remain constant for NO and particulate. In marked contrast, all
three pollutants drop sharply as a result of using cleaner fuels at the in-
dustrial boiler. In the sulfur case, while refinery emissions increased
from .02 #,io6 Btu to 0.05 #,io6 Btu, the industrial boiler emission drops
from 3.2 #1106 Btu to 0.11 #iiO 6 Btu. The other two pollutants drop signifi-
cantly, but not as dramatically.
The water pollution picture is not nearly as significant. Table 1-13
shows the baseline wastewater effluents for a 200,000 Bbl/day refinery. Sul-
fur as H 2 S is about 0.1 ppm, but 3 x 106 gallons are discharged daily, i.e.,
about 2.5 # H 2 S/day. Adding a 50,000 Bbl/day HDS unit would not increase the
water emissions significantly, since so little water actually stays in contact
with the sulfur compounds. Estimates from data in Section 6 indicate that less
than a 1% increase in waterborne H 2 S effluent would result from residual oil
desulfurization to the stringent level. This seems anomalous, but nearly all
the sulfur compounds are converted to elemental sulfur with highly efficient
cleanup processes.
The solid waste situation is quite different. In the past, if HDS cata-
lyst activity could no longer be restored by carbon burnoff, i.e., metal poison
accumulation was the limiting factor, the catalyst was discarded in a suitable
landfill. With the soaring cost of catalyst (due to large recent price in-
creases for cobalt in particular), metal recovery is beginning to alter the
disposal picture. In order to estimate worst case, i.e., all HDS catalyst dis-
posed of as hazardous waste, the following conservative assumptions were made:
1) U. S. HDS capacity of 600,000 Bbl/day for residual and vacuum
gas oil HDS
2) Four types of residual oil used (worse than
being used)
-31—
-------
TABLE 1-13. REFINERY WASTEWATER EFFLUENT QUALITY FOR
3 x lO 1./D AY (200,000 BBL/DAY CRUDE FEED) (18)
Concentration
BOD l5ppm
COD 8Oppm
Amonia 2 ppm
Hydrogen Sulfide 0.1 ppm
Total Phosphorus 2 ppm
Phenols 0.1 ppm
Oil and Grease 2 ppm
Suspended Solids 10 ppm
Dissolved Solids 310 ppm
Source : “Environmental Problem Definition for Petroleian Refineries,
Synthetic Natural Gas Plants, and Liquified Natural Gas
Plants, TM Radian Corporation, November 1975, EPA-60012-75—068.
—32-
-------
3) catalyst consumption rates for desulfurization to 0.8,
0.3, and 0.1% S
The data obtained from these calculations follows:
Spent Catalyst Generation for Sulfur
Control Levels in loris/yr
Residual Oil 0.8% S LSFO 0.3% S LSFO 0.1% S LSFO
1. Low Sulfur,
High Metals 20,000 64,000 115,000
2. High Sulfur,
High Metals 38,000 69,000 125,000
3. Low Sulfur,
Low Metals 8,000 11,000 13,000
4. High Sulfur,
Low Metals 17,000 29,000 38,000
As an indication that the worst case scenario is very conservative, the
actual U. S. HOS catalyst market has been estimated at 10,000-12,000 tons/yr. 09
This value is probably quite a bit more accurate than our calculated range,
primarily because of our using only residual oil in the calculations when, in
fact, much of the U. S. capacity is from vacuum gas oil which has a much lower
metal content than resids. Thus, catalyst life is likely to be several times
longer. Also, catalyst regeneration efficiencies of 90-95% can be achieved by
carbon burnoff, since metal accumulation is very slow with gas oils.
-33-
-------
REFERENCES
1. Devitt, T., et al., PEDCo. Environmental, Inc., “The Population and Charac-
teristics of Industrial/Coninercial Boilers,” May 1979, pp. 30, 43-44, 72-75.
2. Bartok, W., A. R. Crawford, and A. Skopp, “Control of NOx Emissions From
Stationary Sources,” Chemical Engineering Progress , Volume 67, No. 2,
February 1971, pp. 64—72.
3. Ranney, Maurice William, “Desulfurization of Petroleum,” Noyes Data Corpor-
ation, Park Ridge, New Jersey, 1975, pp. 3, 31.
4. Frayer, J. A., A. A. Montagna, and S. J. Yanik, “Gulf’s HDS Processes for
High Metal Stocks,” paper presented at the Japan Petroleum Institute, Tokyo,
Japan, 8 May 1975, Figures 1, 2 and 3.
5. Tyndall, M. F., et al., “Environmental Assessment for Residual Oil Utiliza-
tion,” Catalytic, Inc., EPA—600/7—78—175.
6. Nelson, W. L., “Data Correlation Shows the Amount of Hydrogen Used in Desul-
furizing Residua,” Oil and Gas Journal , 28 February 1977, pp. 126-128.
7. Nelson, W. L., “Catalyst Consumption Required in Desulfurizing Residua,”
Oil and Gas Journal , 15 November 1976, pp. 72-74.
8. Edelman, A. M., et al., “A Flexible Approach to Fuel Oil Desulfurization,”
presented at Japan Petroleum Institute, Tokyo, Japan, 8 May 1975, p. 5 and
Figures 5, 6, 9, 10, and 11.
9. Kuhre, C. J., and J. A. Sykes, Jr., Energy Technology Handbook , D. M.
Considine, Ed., McGraw-Hill Book Company, 1977, pp. 2-172.
10. Turner, p P., S. L. Rakes, and T. W. Petrie, “Advanced Oil Processing Util-
ization Environmental Engineering - EPA Program Status Report,” EPA-600/
7-78-077, May 1978, p. 43.
11. Jimeson, R., and W. Richardson, “Census of Oil Desulfurization to Achieve
Environmental Goals,” AIChE Symposium Series, No. 148, Volume 71, pp. 199-215.
12. Cantrell, Ailleen, “Annual Refining Survey,” Oil and Gas Journal , 20 March
1978, pp. 108 and 113.
13. Streizoff, Samuel, “Partial Oxidation for Syngas and Fuel,” Hydrocarbon
Processing , December 1974, p. 74.
14. Turner, P. p., S. L. Rakes, and T. W. Petrie, “Advanced Oil Processing
Utilization Environmental Engineering - EPA Program Status Report,”
EPA-600/7-78-077, ‘lay 1978, p. 43.
15. “Refining Processes Handbook,” Hydrocarbon Processing , September 1978,
pp. 99—224.
-34-
-------
REFERENCES (continued)
16. Cost studies being performed by Catalytic, Inc., under EPA Contract No.
68-02—2155 (unpublished).
17. “Refining Processes Handbook,” Hydrocarbon Processinq , September 1978.
18. “Environmental Problem Definition for Petroleum Refineries, Synthetic
Natural Gas Plants and Liquified Natural Gas Plants,” Radian Corporation,
November 1975, EPA-600/2-75-068.
19. “Catalysts,” Chemical Week , 28 March 1979, pp. 51-53.
—35-
-------
SECTION 2
OIL CLEANING AND CLEAN OIL EMISSION CONTROL TECHNIQUES
FOR OIL-FIRED INDUSTRIAL BOILERS
2.1 PRINCIPLES OF CONTROL
Sources of Emissions
Uncontrolled industrial boilers using oil fuels emit significant amounts
of particulates, SO, , and NO to the atmosphere. A recent study by PEDCo
provided the data shown in Table 2-1. PEDCo estimated the consumption of resi-
dual and distillate fuels in the industrial section at 19,881 x 1O 3 m3/yr
(125,067 x bbl/yr) and 7281 x 1O 3 m3/yr (45,799 x iD bbl/yr), respectively,
in 1975.(2) In 1977, residual oil fuel supplied ?7•3% of the Btu capacity of
industrial/coninercial boilers with distillate fuels providing 9.7%. Further-
more, PEOCo projects oil-fired industrial/comercial boiler capacity to increase
by approximately 2.3 times 1977 values by the year 20OO) Obviously, without
controls, emissions would increase roughly in proportion to fuel use.
Pollutant Formation Mechanisms
Sulfur oxide emissions are directly related to the sulfur content of the
fuel. The combustion of cleaner, low-sulfur oil will result in lower SO, emis-
sions to the atmosphere. Nitrogen oxides are formed from both oil-bound nitro-
gen and nitrogen in the combustion air. Decreased nitrogen in fuels will reduce
N0 emissions, but it will not affect thermal fixation of atmospheric nitrogen.
Staged con ustion, low excess oxygen, and flue gas recirculation may be used for
controlling NO emissions originating from thermal fixation of atmospheric ni-
trogen during the combustion of clean oi1s. 4 Particulate emissions are depen-
dent on the fuel characteristics, such as carbon residue and ash, which are
reduced in oil treatment. Therefore, particulate loadings in general should
be lower when cleaner fuels are being burned.
-36-
-------
TABLE 2-1. ESTIMATED UNCONTROLLED EMISSIONS FROM THE
INDUSTRIAL/COMMERCIAL BOILER POPULATION (1)
Estimated Emission J1975) Mg/yr (tons/yr)
Boiler Type Particulate SOx NOx
Water Tube
Residual Oil Fired 59,900 (66,000) 794,500 (875,800) 198,600 (218,900)
Distillate Oil Fired 2,300 (2,500) 38,700 (42,700) 25,200 (27,800)
Fire Tube
Residual Oil Fired 22,200 (24,500) 294,700 (324,900) 73,700 (8L200)
Distillate Oil Fired 3,200 (3,500) 53,700 (59,200) 34,900 (38,500)
Cast Iron
Residual Oil Fired 14,500 (16,000) 192,200 (211,900) 48,000 (53,000)
Distillate Oil Fired 1,900 (2,100) 32,500 (35,800) 21,100 (23,300 )
TOTALS Residual Oil Fired 96,600 (106,500) 1,281,400 (1,412,600) 320,300 (353,100)
Distillate Oil Fired 7,400 (8,100) 124,900 (137,700) 81,200 (89,600)
-37-
-------
Relative Emission Levels
As previously indicated, the SON, l O and particulate emission levels from
oil-fired boilers are highly dependent on the quality of fuel being burned. For
example. S0 emissions may range from 3.5 lb. S0 per million Btu for a high-
sulfur residual oil to 0.1 lb. S0, per million Btu for a clean, desulfurized
fuel oil. Similarly, N0 emission levels may range from 0.6 lb. NO, per million
Btu to 0.3 lb. N0 per million Btu for a denitrogenized fuel oil. Total parti-
culate loadings may range from 0.1 lb. particulates per million Btu to 0.03 for
a clean fuel.
Identification of Prime Control Mechanisms
There are two general approaches to reducing the emissions of SOx N0
and particulates from the burning of oil as a fuel in industrial boilers. These
are:
1) by cleaning the flue gas output from the boiler; and
2) by reducing the input of these impurities to the boiler
by precleaning the fuel.
The first approach, called Flue Gas Cleaning, is being addressed by other
This report discusses the oil cleaning techniques applicable to fuel oil frac-
tions which are used as industrial boiler fuels.
There are various methods of cleaning fuel oils which are used by refiner-
ies. The techniques fall into two general categories:
I) processes which process a liquid fuel with reduced contami-
nant content; and
2) processes which produce a gaseous fuel with reduced contami-
nant content.
! brief background on oil refining may be of benefit to readers who are
not fami1iar with the petroleum field. A glossary of terms used in petroleum
technology will be found in Appendix A.
-38-
-------
Petroleum (or crude oil) is a very complex mixture of chemical compounds
which are comprised mainly of carbon, hydrogen, sulfur, nitrogen, oxygen, and
metals. Crude oils vary widely in composition and boiling range depending on
their origin. They also frequently contain inorganic salts from brine found
with the crude or from seawater used as ballast in tankers for shipment.
A very simple flow sheet of a petroleum refinery is given in Figure 2-1.
It can be seen that the first step in processing is desalting, wherein the crude
oil is washed with water to remove inorganic salts which would otherwise corrode
or deposit in later processing equipment such as process heaters or catalyst
beds. Following desalting, the crude is heated and passed into a distillation
column operated at atmospheric pressure, wherein it vaporizes and is split into
a number of fractions with broad boiling ranges. This process is called top-
ping. Fractions are generally grouped as follows:
Crude Oil Fraction Boiling Range 0 F *
Gases below 80
Light Naphtha 80-220
Heavy Naphtha 180- 520
Light Gas Oil 420-650
Atmospheric Bottoms 650+
If the refinery has. a vacuum distillation unit, as well as an atmospheric
unit, the bottoms product (variously called topped crude, reduced crude, atmos-
pheric resid, or atmospheric tower bottoms) may be fractionated further to give
a heavy gas oil (sometimes called vacuum gas oil) with a boiling range from ap-
proximately 700—1100°F and a heavy residue called Vacuum Tower Bottoms (VTB) or
Vacuum Resid (VR) with a boiling range from l050 0 F+. Depending on the refinery’s
design purposes, the various fractions may be sold without further treatment,
blended for fuel oil use, or refined or upgraded into higher-priced products
such as gasoline, lubricants, and cleaner fuel oils.
* Ranges are approximate and vary with crude processed
and individual refinery systems. (5)
Used with permission taken from “Refining
-39- Processes Handbook” September 1978, p. 100.
-------
Figure 2-1
SIMPLIFIED CRUDE OIL REFINERY FLOWSHEET FOR FUEL OIL PRODUCTION
Atmospheric
Distillation
Tower
Gases Optional
I—————— 1
Water Lt. Naphtha I
r— Heavy Gas Oil —.Ø To Hydrotreater
Heavy Naphtha
I Vacuum
Distillation
Tower
I To Coker
or
(Vacuum Resid) i Hydrotreater
L_
No. 6 Fuel Oil (High Sulfur)
or
Feed to Hydrotreater to
Produce Low Sulfur Fuel Oil
Lt. Gas Oil
I
-------
The processes which produce cleaned liquid fuels are called hydrotreating.
They are chemical processes involving contact of the oil with catalyst and
hydrogen. These processes convert much of the chemically-bonded sulfur and
nitrogen to gaseous hydrogen sulfide and aninonia respectively, thereby removing
them from the oil stream. In addition, the metals content o-f the oil is reduc-
ed, as is the carbon residue portion of the oil.
A second prime technique included in this report is the group of processes
which is designed to convert heavy, high-sulfur residual oils to clean, low-
sulfur gas. These processes include Partial Oxidation (POX) and the Chemically
Active Fluid Bed Process (CAFB). The processes which produce liquid fuels are
discussed under 2.2, and those which produce gaseous fuels are discussed
under 2.4.
2.2 HYDROTREATING PROCESSES FOR PRODUCING CLEANED LIQUID FUELS
In the typical hydrotreating process, atmospheric resid is filtered to
remove rust, coke, and other suspended material. It is then mixed with hydrogen, heated
tO 6500 to 650°F, and passed over one or more catalytic reaction beds.* Numerous
chemical reactions occur which lead to removal of most of the sulfur as H 2 S.
Table 2-2 illustrates some of the types of compounds and reactions involved. 7
Since sulfur is the major impurity -in petroleum, the general technique is fre-
quently called Hydrodesulfurization (HDS) in the industry.** However, the hy-
drogen reacts with other species besides sulfur compounds. For example, nitro-
gen compounds break down to liberate aniiionia from the oil. This is called
* The most widely-used catalysts are composites comprised
of cobalt oxide, molybdenum oxide, and alumina, where
alumina is the support carrying the other agents as pro-
moters. However, other catalyst materials are in use or
under development. Much of the catalyst technology is
propriçtary, but the patent literature is extensive.
Ranneyl .6) has compiled information from over 200 patents
during the period 1970-1975.
** A discussion of collection and processing techniques for
the evolved H 2 S will be found in Appendix B.
-41-
-------
Table 2-2
CHEMISTRY OF HYDRODESULFURIZATION REACTIONS IN
PETROLEUM CRUDE OIL (7)
Name Structure Typicai reaction
Th ots (mer aptans) R—SH R—SH + H 2 RH + H 2 S
Disulfides R—S—S—R’ R—S--S—R’ + 3H 2 0 RH + R’H + 2H 2 S
Sulfides R—S--R’ R—S--R’ + 2H 2 0 RH + RH + H 2 S
Thiophefles (3+ 4H 2 On-C 4 H 10 + H 2 S
P + 3142—4 cH 3 CH 2 1J + H 2 S
Dibenzothiophenes EX TJ1TJ + 2H 2 + + H 2 S
Used with permission Taken from: Hastings, K. H. and R. P. Van Drisen,
“Hydrodesulfurization of Petroletiii Crude Oil Fractions and Petroleuii Products,”
Energy Technology Handbook , D. M. Considine, Ed., McGraw-Hill Book Co., 1977,
p. 2—254.
-42-
-------
denitrogenation or denitrification. The nickel and vanadium in the oil, which
are bound as organo-metal compounds, are liberated by reaction with the hydrogen.
This is called demetallization. Most of the liberated metals deposit (as the
sulfide) on the catalyst surface or in its pores and slowly deactivate the cata-
lyst. Other reactions which take place break up large complex molecules such
as asphaltenes and lead to a reduction in carbon residue of the product oil.
By utilizing catalysts, the reactions with hydrogen can be restricted large-
ly to the types above which take place under moderate reaction conditions. With-
out the catalysts, higher reaction temperatures or pressures would be required,
and this would lead to greatly-increased hydrogen consumption due to hydrogena-
tion of aromatic ring systems which are abundant in high boiling petroleum
fractions.
many companies are engaged in developing and using catalytic hydro-
treating (or hydrodesulfurization) processes. All are similar in
basic concept and vary only in details such as catalysts, process conditions,
and complexity. A recent paper by Gulf Oil Company investigators traces the
development of one basic processJ 8 Figure 2-2 illustrates the simplest com-
mercial version currently marketed by Gulf, and it is known as Gulf II. Its
basic elements are a feed filter, heater, single-stage catalytic reactor, a gas!
liquid separator, a fractionating column, and a gas treatment section. This
simple process system is capable of producing fuel oil of approximately 1% sul-
fur from a feedstock such as atmospheric resid containing 2-4% sulfur. To pro-
duce a lower sulfur content product, additional catalytic reaction stages must
be added. Figure 2-3 shows the Gulf III system with two catalytic reaction
stages which can produce fuel of approximately 0.3% sulfur content from the
same feedstocks as above. The most advanced process Gulf has developed is
known as Gulf IV (Figure 2-4). It uses three catalytic reactors and can pro-
duce fuel oils of approximately 0.1% sulfur.
-43-
-------
I Fractionating Column
Acid Gas (H 2 S)
- Light H.C.
Used with permission Taken from: Frayer, J. A., et ah, “Gulf’s HUS Processes for High Metal Stocks,”
presented at the Japan Petroleti Institute, Tokyo, Japan, May 8, 1975.
Figure 22
BASIC HDS PROCESS (GULF lIDS - TYPE II) (8)
Hydrogen
Reduced
Recycled Hydrogen Gas
F liter
Heater
Compressor
Catalyst
Vessel
Gas
4 -----
Unreacted Hydrogen plus H 2 S
-1
Recycle Gas
Treatment
-------
Figure 2-3
IMPROVED HDS PROCESS (GULF HDS -TYPE IIl)(8)
Used with permission Taken from:
presented at the Japan Petroletsn
Fractionating Column
_________ Acid Gas
‘ (H S)
Light H.C.
Frayer, J. A., et al., Gulf’s HDS Processes for High Metal Stocks,
Institute, Tokyo, Japan, May 8, 1975.
Hydrogen
Reduced
Heater
Catalyst
Vessel
Compressor
Catalyst
Vessel
Off Gas
Gas! Liquid Separator Gas! Liquid Separator
I -—--— — -
Fuel
Recycle Gas
Treatment
-------
Figure 2-4
STATE OF THE ART HDS PROCESS (GULF HDS-TYPE lV) 8
— a a — — a — _ a — a — a — n a — — a — — — —
I
drogen I
Hy
Compressor
I
I
I
________________ I
Fractionating Column
Catalyst Off Gas
Vessels
I
I
I
I
I
I
I
Filter Heater
$
Gas/Liquid Separators
I. — — — — as — — — — a —
I I
Recycle Gas Ø Acid Gas
— — — 4 (H” S)
Treatment
Light H.C.
Used with permission Taken from: Frayer, J. A.,, et al., “Gulf’s HDS Processes for High Metal Stocks,”
presented at the Japan Petroleum Institute, Tokyo, Japan, May 8, 1975.
Low Sulfur
-------
Table 2—3 illustrates the typical product properties obtained when the
three versions of the Gulf process are applied to Kuwait atmospheric resid con-
taining 3.8% sulfur. It can be seen that the number of catalyst stages
strongly affects both physical and chemical properties of the product oil. In
addition to sulfur removal, other changes are noted: reduction in pour point
and viscosity; and reduction of chemical impurities such as nitrogen, metals,
salts, and ash.
There is a price to pay for such beneficial changes, however. The amount
of hydrogen consumed increases with the degree of desulfurization. So does
catalyst cost. A further cost is a slight reduction in heating value per gal-
lon. Even though heating value/pound increases with the degree of desulfuri-
zation, the density of the product decreases; thus, slightly greater volumes of
cleaned fuel oil must be burned to produce the same amount of heat provided by
an untreated residual oil. The change in heating value is of the order of 1-3%
on atmospheric resids and about 5% on vacuum resids. When the degree of clean-
up is considered, this energy impact seems insignificant.
The composition of the feedstock to a hydrotreater strongly influences the
amount of hydrogen and catalyst consumption in the process. Nelson 0 has cor-
related hydrogen consumption with sulfur reduction for a variety of resid feeds.
Figure 2-5 illustrates his results on feedstocks varying from 4-l8°API gravity.
It can be seen that to obtain 90% reduction in sulfur for an l8°API feedstock,
about 650 scf of hydrogen are consumed per barrel of oil processed; whereas, a
4°API feed would require 1200 scf/barrel. Another of Nelson’s correlations is
shown in Figure 2-6, wherein he has plotted hydrogen consumption data for typi-
cal low metals atmospheric resid (16°API) and vacuum resid (6°API) as a function
of sulfur desired in the fuel oil product. Thus, in producing 0.3-1.0% sulfur
fuel oils from atmospheric resid, 345-775 scf of hydrogen per barrel are required,
while vacuum resid requires 870 to 1150 scf to accomplish the same task.
-47-
-------
TABLE 2-3. TYPICAL PROPERTIES OF LSFO PRODUCT FROM GULF HDS PROCESSES 9
Gulf Gulf Gulf
Untreated II III IV
Product Yield: Vol. % -- 89.4 97.5 97.1
Product Properties:
Cut Point: 0 F 650 650 375 375
Gravity: °API 16.6 20.0 23.4 24.1
Sulfur: Wt. % 3.8 1.0 0.3 0.1
Carbon Residue: Wt. % 9.0 5.31 3.33 2.75
Nitrogen: Wt. % 022 0.13 0.13 0.09
Nickel: PPM 15.0 4.6 1.3 0.4
Vanadit n: PPM 45.0 8.2 2.2 1.0
Viscosity: SUV (210°F) 250 107.3 52 45
Ash: Wt. % 0.02 0.004 0.003 0.003
Salt: PPM* 44.9 0 0 0
Heat of Conthustion: Btu/LB 19,110 19,250 19,375
Hydrogen: Wt. % 12.1 12.5 12.7
Carbon: Wt. % 86.7 87.1 87.1
Pour Point: 0 F ** +60 +35 0
Hydrogen Consunption: SCF/BBL 497 663 812
* Salt refers to all water-soluble cations, determined as halide and
reported as MaCi before desalting.
Not stated.
-48-
-------
1-igure
CHEMICAL HYDROGEN CONSUMPTION IN DESULFURIZATION OF RESIDUA( 10 )
1,000
900
800
U-
C-)
U)
700
I-
a-
U)
z
3 600
z
w
0
0
500
400
300
200
-
•__
0 AP I of fee
v
<77
.
,,./“
V
/
• 77
/
10
.1277_//
14
“//__
16 / 18/”
30 40 50 60 70 80 90
Note: SULFUR REDUCTION %
1. Reduce by 9% for fixed-bed processes.
2. Apply correction for high-metals feeds.
-49-
Used with p nh1issiQn.of Petroleum Publishing Company Taken from Oil and Gas Journal
magazine reterencea issue
100
-------
Figure 2-6
HYDROGEN CONSUMPTION FOR ATMOSPHERIC AND VACUUM RESIDUA
0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
SULFUR IN FUEL OIL, %
1. Reduce by 9% for fixed-bed processes.
2. Apply correction for high-metals feeds. 50
Used with permission of Petroleuii Publishing
Company Taken from Oil and Gas Journal
magazine referenced issue
1,200
1,100
-J
C-)
C,,
z
w
w
0
>-
I
-J
0
w
I
0
1,000
900
800
700
600
500
400
300
200
Note:
-------
Both figures are based on normal metals content of less than about 200 ppm.
Nelson suggests the following corrections to Figure 2—6 for higher metals con-
tent feedstocks:
V + 1’1i, Corrections,
ppm %
0-100 -2
200 +1
300 2.5
400 4
500 6.5
600 9
700 12
800 16
900 21
1000 28
1100 38
1200 50
The problem caused by metals is deposition onto the catalyst surface or in the
pores. This leads to deactivation of the catalyst, which is only overcome by
a temperature or pressure increase to maintain acceptable processing rates.
The increase in required severity of process conditions leads to more hydro-
cracking with a subsequent increase in hydrogen consumption.
A further complication from the metals content of the feed is a shortening
of catalyst life. Even though some deactivation can be tolerated, the resultant
increase in hydrogen uptake means catalyst must be changed out sooner. The
effect of metals can be seen in another Nelson corre1ation shown in
Figure 2—7. It will be observed that, for 90% sulfur removal from a 25 ppm
metals content feedstock, about 27 barrels of oil can be processed per pound
-51-
-------
Figure 2-7
EFFECT OF METALS ON CATALYST CONSUMPTION
PERCENTAGE OF SULFUR REMOVED
100
Used with permission of Petro1e Publishing Company Taken from Oil and Gas Journal
magazine referenced issue
I-
U)
>-
ii
-52-
-------
of catalyst; to achieve the same sulfur performance with a 100 ppm metals con-
tent feedstock, only 4.5 barrels can be processed per pound of catalyst; a 300
ppm feedstock requires almost 1 pound of catalyst per barrel. Clearly, high
metal feedstocks are a costly problem to the refiner. Most refiners are using
a separate stage of lower cost catalyst material prior to the special hydro-
desulfurization catalysts. These separate stages may be packed with a material
such as alumina or clay, which collects the metals and “guards” the subsequent
high activity catalyst. For this reason, some refiners call this stage a
“guard reactor” or “guard vessel.” Such a process scheme used by Exxon (called
RESlDfining) is shown in Figure 2_8. 2)
From the above, it is obvious that catalyst life is quite variable. In a
fixed-bed system, catalyst changes are usually once or twice per year, depend-
ing on product requirement and feedstock composition. For example, with the
Gulf IV system, producing 0.1% LSFO from 3.8% S Kuwait Resid (97% lIDS), a cata-
lyst life of six months is the design base. Similarly, using Arabian Heavy
with RESlDfining at 75% HDS, a catalyst life of 400-500 days can be achieved.
The catalyst bed temperature is not a constant. As can be seen in Figure 2-9,
note that a bed temperature rise of 80°F is reached after 400 days operation
on resid and about l20 0 F shortly thereafter. This is usually the normal
limit for catalyst life, and changeout is required to eliminate excessive
hydrogen consumption.(12)
Most refiners use fixed catalyst beds, which require a process shutdown
for catalyst replacement. However, there are some advocates 0 f expanded or
ebullated beds with intermittent catalyst feed and bleed systems to maintain
continuous operation at relatively constant conditions. The H-Oil process
developed by Hydrocarbon Research, Inc. and utilized by Cities Service is an
example of such an operating system. The expanded bed operates as a back mix
reactor with the entire bed at a constant temperature. Such a system is less
—53—
-------
-fl-a rii
- .-1 0-Lfl
(D (D
v
V) rP
m •
N) -
•
S
u’r’ (D
•
c_.
• -C
o
4 O=
• 0
-
-
• r1
• •1 D i
o
Di - -m
VD
- — fD 0
•
v
C-) 1. -
o
-5
-40
a,
0
r’i 0
)( c.
x -‘1
—
C).
0 0
_5 -4.
. ,x-’
D i
f -n
- .
C
N
• Di
U, .- .
.0
D i .
0.
Figure 2-8
USE OF GUARD REACTOR (RESIDFINING PROCESS - SIMPLIFIED FLOW DIAGRAM) 12
Naphtha
Specialty
Fuel Oil
Desulfurized
Product
H 2
H 2 S
Fuel Gas
(Sulfur Free)
Reactors
Vessel
Product
Stripper
------- An error occurred while trying to OCR this image.
-------
prone to plugging but has a slightly greater hydrogen consumption than a fixed
bed system, according to Nelson)1 It is most suitable for high metals feed-
stocks where catalyst life is short.. (I )
Hydrotreating is applied in various ways in refineries. 2 The direct
approach is not necessarily the lowest cost method, nor is it best suited to the
fuel oil market mix of a particular refiner. Refiners use both direct lIDS pro-
cesses, as well as indirect HDS processes. Figure 2-10 illustrates how a direct
HDS process is used in a refinery. Note that a charge of 150,000 BBL/day of
heavy Iranian crude produces 71,000 BBL/day of 2.5% atmospheric bottoms. This
is fed to a RESIDfiner (Exxon) to produce 70,000 BBL/day of 0.3% sulfur fuel
oil and 264 tons of sulfur. 12
Figure 2-11 shows how a less costly scheme can produce a pool of fuel oils
by desulfurization of lighter products and back biendingJ 12 This is called
indirect HDS. In the Exxon example shown, a 250,000 BBL/day charge of Arabian
Light crude produced light gas oil, vacuum gas oil, and vacuum resid. Most of
the light gas oil is treated by a hydrofiner to remove the sulfur. The vacuum
gas oil is 95% desulfurized in a GO-finer. The vacuum resid stream is split;
most is b1ended with the desulfurized light gas oil and vacuum gas oil to pro-
duce 0.3% S or 1.0% S fuel oils, while some is blended with light gas oil to
produce 3.3% S bunker fuel oil. In this scheme, the refiner avoids the costly
direct desulfurization of vacuum resid and yet satisfies three fuel oil market
segments. He has taken advantage of the relative ease of desulfurization of
light gas oil and vacuum or heavy gas oil which are low in metals content. The
indirect scheme is limited in that it cannot produce as much very low sulfur
oil as direct resid lIDS can.
Development Status
Hydrotreating processing of heavy oils has been in coimnercial use for
about 15 years! Most of the installed capacity is in Japan because of the
extensive use of fuel oil for electric generation in that country and the very
-56-
-------
Figure 2-10
INTEGRATION OF DIRECT HDS INTO A REFINERY (12)
Typical Application of Residfining for Deep Desulfurization of Fuel Oil
Distillation
Heavy
Iranian
Crude
150 MB/D p
1.55% S
\._ ./ 650°F
T 71 MB/D
2.5% S
Gas and
Naphtha
Residfiner /
Light Ends
Processing
70 MB/D
0.3% S
Gas
Used with permission taken from:
Edelman A. M et al “A Flexible ADprQach to Fuel Oil Desulfurization
present d at 3 pan eDoleum institute, JoKyo, Japan, 3 May 19/b, p. b, nd
Figures 2, 5, 6, 9, 10, and 11. Copyright by Exxon Corp.
Motor
Gasoline
Jet Fuel
• Diesel Oil
Heating Oil
Low Sulfur
• Fuel Oil
Sulfur
264 T/D
-p Gas
01
-1
‘1
1 —
-------
Figure 2—11
UTILIZATION OF INDIRECT HDS IN A REFINERY (12)
Typical Application of Gofining for Fuel Oil Desulfurization
___________________________________________________ Naphtha & Light
I ___________
__________ 116 MB/D
32 MB/D _______________
Arabian Light Crude Hydrofiner
250 MB7 Atmospheric
Pipestill Very Low Sulfur
—0 FuelOii
22 MB/D
—
0.3% S
7 +
95%HDS .4
2.4% S
97 MB/D
2.9% S
_ fj 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1*EJ11111111 l! Low Sulfur
Fuel Oil
_____ Vacuum
____ 97 MB/D
Pipestill
1.0% S
— 1100k
31 MB/D
5 MB/D 4.4% S 21 MB/D
10 MB/D Bunkers
0 15 MB/D
3.3% S
Used with permission taken from:
Edelman, A. M., et al., “A Flexible Approach to Fuel 011 Desulfurization,”
2resented at Japan PetroleLn InstItute 1 Tokyo, Japan 8 May 1975, p. 5, and
-------
tight emission standards adopted by the Japanese government. Most of the pro-
cesses were developed by U. S. oil companies, however, and built under license
in Japan.
In recent years, desulfurization capacity has been installed in U. S. re-
fineries to meet demand created by tightening air quality standards. As shown
in Table 2-4, the current total U. S. refinery desulfurization capacity is over
1.8 million barrels per day from 86 plants ) rbwever, only 19 plants have di-
rect catalytic resid or heavy gas oil hydrodesulfurization capability; and,
they produce less than 0.6 million barrels per day. The other plants use phy-
sical or thermal cracking processes. Products are only slightly desulfurized
when liquids are produced by visbreaking of gas oils. In coking processes,
little or no fuel oil fraction is produced; i.e., gases, naphtha, and coke are
generally the desired products. Gas oils produced are frequently recycled to
the coker or are used as feed to catalytic cracking units for gasoline produc-
tion, although they may be sold as low sulfur fuel oii. 0 2)
The extent of interest and process capability among refiners can be best
appreciated by study of Table 2-5, which compiles commercial hydrodesulfuriza-
tion results obtained by eleven companiesJ 424) The table shows the company
which is the licensor of the process and its tradename, along with the results
obtained with a variety of feedstocks from various crudes. Most of the data
are on atmospheric resids, but there are also data on vacuum resids, vacuum gas
oils, crudes, tar sands, and bitumen.
Applicability
Desulfurized oils are readily applicable to all boiler types and sizes.
There seems to be no limitation on the use of refinery techniques as an emission
control mechanism for S0 , N0 , and particulates on boilers which are currently
burning fuel oils.
-59-
-------
TABLE 2-4. SURVEY OF U. S. REFINERY DESULFURIZING CAPACITY - JANUARY 1978 (13)
No. Plants Capacity
Resid Desulfurizing 5 113,600 b/sd
Heavy Gas—Oil Desulfurizing 14 448,100 b/sd
Residual Visbreaking -0- -0-
Residual Upgrading 1 20,000 b/sd
Visbreaking 13 180,550 b/sd
fluid Coking 6 134,100 b/sd
Delayed Coking 47 936,000 b/sd
Totals 86 1,832,350 b/sd
Used with permission of Petroletin Publishing Company Taken from Oil and Gas Journal
magazine referenced issue
-60-
-------
TABLE 2—5. CONMERCLAL HDS RESULTS (14—24)
Licensor Process Name Feedstock Product Ref. No .
V + Ni Sulfur Sulfur Desulf. H2 Cons.
Source Type ppm/wt. Wt. % Wt. % ______ SCF/BBL
Chevron RDS Hydrotreating Arab Light AR 38 32 0.46 86 560 14
Arab Heavy AR 115 3.9 0.55 85 750 15
VRDS Hydrotreating Arab Light VR 38 4.1 1.0 76 780 14
Arab Heavy VR 115 5.1 1.0 80 960 15
C—E Lummus, LC—Fining Kuwait AR 64 4.0 1.4 65 520 14
Cities Service Kuwait AR 64 4.0 0.4 90 690 14
Gach Saran VR 413 3.5 0.9 75 1400 14
Arab Heavy VR 195 4.9 1.2 75 1410 14
Exxon Go-Fining Arab Heavy Crude 117 3.0 0.1 96 410 14
Athabasca Tar Sands NR 4.0 0.1 97 975 14
Kuwait AR 66 3.0 0.3 90 280 15
Arab Light AR 40 2.3 0.23 90 220 15
Khafji AR 105 2.9 0.3 90 300 15
Gach Saran AR 220 1.9 0.19 90 220 15
Exxon RESIDfining Gach Saran AR 220 2.5 0.3 88 625 14
Arab Heavy AR 120 4.2 0.3 93 915 14
Kuwait AR 55 3.8 0.5 86 640 15
Arab Light AR 40 3.0 0.5 83 530 15
Khafji AR 105 4.0 0.5 88 700 15
Gach Saran AR 210 2.4 0.5 79 480 15
Iran Heavy AR 210 2.5 0.3 88 625 16
Gulf Resid HDS Kuwait AR 60 3.8 1.0 74 490 14
(Gulf [ I) Kuwait VR 162 5.7 0.9 84 1250 14
S. Louisiana AR 5 0.46 0.05 87 17
W. Texas AR 21 2.2 0.33 85 17
Kuwait AR 66 3.8 1.0 74 515 18
Khafji AR 118 4.4 1.0 77 685 18
Iranian AR 183 2.5 0.65 74 440 18
Kuwait VR 199 5.7 0.9 84 1250 18
Agha Jan VR 274 3.8 0.75 80 1125 18
Arab Light VR 101 4.1 0.75 82 1000 18
Kuwait AR 60 3.8 1.0 74 630 15
Mid East A AR 40 3.0 0.6 80 490 15
Mid East B AR 114 4.2 0.8 76 655 15
Mid East C AR 110 5.8 1.5 74 820 15
U. S. Domestic AR 120 1.8 0.5 72 295 15
-------
TABLE 2—5. COMMERCIAL liDS RBSULTS (cont’d.)
Licensor Process Name Feedatock Product Ref. No .
V + Ni Sulfur Sulfur Desuif. j 2 Cons.
Source Ty ppm/vt. Wt. Z Wt. % ______ SCF/BBL
Gulf Resid RDS Kuwait AR 66 3.8 0.3 92 770 18
(Gulf III) Kuwait AR 66 3.8 0.5 87 660 1.8
Khafji AR 118 4.4 0.5 89 950 18
Kuwait AR 60 3.8 0.3 92 640 14
Arab Light AR 40 3.0 0.3 90 535 14
Iran Light AR 110 2.4 0.3 88 480 14
Kuwait AR 69 3.8 0.3 92 17
Kuwait AR 60 3.8 0.5 87 815 15
Mid East A AR 40 3.0 0.3 90 720 15
Mid East D AR 114 4.2 0.5 88 825 15
Mid East E AR 116 4.6 0.7 85 965 15
Kuwait AR 60 3.8 0.1 97 700 18
Resid liDS Kuwait AR 60 3.8 0.1 97 720 14
(Gulf IV) Kuwait AR 66 3.8 0.1 97 960 18
Kuwait AR 60 3.8 0.1 97 960 19
Ceuta Crude 392 2.1 0.3 86 19
Hydrocarbon 11—Oil Athabasca Bitumen NR 4.9 1.0 80 1400 14
Researcb Cjtje Gach Saran VR 413 3.5 1.0 71 1220 20
Service Kuwait AR 75 3.8 1.0 74 480 20
Kuwait AR 60 3.9 0.2 95 580 15
Kuwait VGO Nil 3.0 0.2 93 450 15
Institute Fuel Hydro— Kuwait AR 63 4.1 0.45 89 760 15
Francais du desulfuri— Kuwait VGO Nil 2.5 0.25 90 232 15
Petrole zatton
Iippon Oil Residua Arab Light AR 37 2.9 0.30 90 592 21
Shell Residual Oil Kuwait AR 66 4.2 0.53 88 683 22
Hydrodesul— Quatar AR 35 2.8 0.23 93 536 22
furization Kuwait AR 66 4.2 0.58 86 747 22
Kuwait AR 66 4.2 0.62 85 753 22
Iran Heavy AR 179 2.7 0.37 86 626 22
Iran Heavy AR 179 2.7 0.38 86 707 22
-------
TABLE 2—5. COMMERCIAL HDS RESULTS (cont’d.)
Licensor Process Name Feedstock Product Ref. No .
V + Ni Sulfur Sulfur Desuif. H2 Cons.
Source ppm/wt. Wt. % Wt. % ______ SCF/BBL
Shell Residual Oil Kuwait VR 130 5.2 1.14 78 845 22
Hydrodesul— Kuwait AR 66 4.2 0.57 86 747 22
furization Arab Heavy AR 115 4.3 0.80 81 749 22
Iran Heavy AR 179 2.7 0.34 87 760 22
Qatar AR 18 2.6 0.37 86 523 22
Oman AR 44 1.8 0.28 84 389 22
Iran Heavy AR 163 2.6 0.34 87 596 22
Mid East AR 50 4.2 0.5 88 725 14
Union Unicracking/HDS Kuwait AR 46 3.8 0.3 92 730 14
Gach Saran AR 220 2.4 0.3 88 570 23
N. Slope AR 44 1.6 0.3 82 340 23
Arab Light AR 40 3.0 0.29 90 530 24
UOP RCD/Unlbon Kuwait AR 60 3.52 0.28 92 706 14
AR = Atmospheric Resid
VR = Vacuum Resid
VGO = Vacuum Gas Oil
-------
Factors Affecting Performance
Boiler performance is not strongly affected by using cleaner oils. As was
discussed earlier, the reduction in heating value per gallon of 1-3% for atmos-
pheric resid and perhaps 5% for vacuum resid is a mild penalty for the substan-
tial improvement in emissions (see Section 2.3 below). This effect would be
further offset by the greater ease of handling (lower pour point and viscosity)
and by lessening of corrosion and deposit formation due to chemical composition
changes in the oil as a result of hydrotreating.
2.3 PERFORMANCE OF CLEANED LIQUID FUELS IN INDUSTRIAL BOILERS
Emission Reductions
Sulfur oxides emissions are directly dependent on the sulfur content of
the fuel. Hydrotreating techniques may be used to reduce sulfur content to the
level required by emission standards. The NO emissions are generated from two
sources: the nitrogen content of the fuel, and the nitrogen present from com-
bustion air. The nitrogen content of the fuel can be reduced by hydrotreating.
Thermal NO emissions are a function of boiler design and operation. Because
of the smaller combustion chamber volume and reduced operating flexibility, the
NOx emissions from industrial boilers will be greater than those of utility
boilers.(25X26) Similarly, due to the less efficient atomization and co itustion
in industrial boilers, particulate emissions are more directly related to the
carbon residue of the fuel than to the ash content. Hydrotreating processes
reduce carbon residue, thus reducing the particulate emission. Table 2-6 shows
the effective sulfur, nitrogen, and carbon residue removal through several hy-
drotreating processes.(27(3 )(29 3 ) In general, the sulfur removal levels
achieved have averaged approximately 80% by effectively reducing the sulfur con-
tent of fuel oils having 3% to 5% sulfur to levels of 0.5% to 1%. More strin-
gent hydrotreating may even reduce the sulfur content of oils down to 0.1%
sulfur. 07 Approximately 40-50% nitrogen removal efficiencies are achieved
by the best hydrotreating processes. This will result in a slight reduction of
-64-
-------
TABLE 2-6. CONTAt4INAIiT REMOVAL IN IIYOROTREATING PROCESSES
Fee s _ tpcj ___ - - — _ 4 j t % Sulfur Weig ht % Nitrog en - Weight % Carbon Residue
Gravity, Feed- Feed- Feed-
Process Name OAPI - stock Product Removal stock Product Removal stock Product Removal
HDS-Gu1f 27 Kuwait 16.6 3.8 1.0 /4 0.21 0.18 14 8.3 4.9 41
0.5 87 0.14 33 4.0 52
0.3 92 0.13 38 3.3 60
0.1 97 0.11 48 2.2 73
(28)
RCD Unibon-
U0P Kuwait 16.4 3.52 0.24 92 0.20 0.12 40 9.45 3.84 59
(29)
Residfining- Gach Saran 2.5 0.3 88
ii Exxon Arab Heavy 4.19 0.3 93
An co(29) West Ix. Sour 3.85 1.0 74
West Tx. Sour 3.85 0.3 92
Amoco 30 Khafji 14.7 4.3 0.90 79 0.27 0.21 22
Gach Saran 17.0 2.4 0.89 63 0.46 0.41 11
-------
NO from industrial boilers burning the cleaned fuel. Greater nitrogen removal
efficiencies will not significantly reduce the emissions from industrial
boilers due to the large contribution of thermal fixation.
Hydrotreating of crudes reduces the ash content of residual oils by 80%
(see Table 2-3), thus lowering particulate in the flue gas from combustion of
these oils. All of these removal efficiencies are highly dependent on the type
of feedstock, and the oil cleaning techniques should be evaluated on a case-by-
case basis to determine the effective emission reduction.
Actual emission data from industrial boilers are shown in Figures 2-12,
2-13, and 2-14 for sulfur oxides, NOR. and particulates, respectively. Figure
2-12 shows the direct relationship of sulfur content in the fuel with sulfur
oxides emissions. Figure 2-13 shows the effect of reduced NO emissions by
using a continuing lower nitrogen content. Figure 2-14 shows how particulate
loadings can be reduced by burning fuels having a lower carbon residue content.
These plots suninarize the available emissions data on industrial boilers.(24)(2631)
Impact on the Boiler
There are or appear to be no significant negative impacts on the boiler operation.
Burning lighter fuels will lower the severity of operation on the boiler. The
modest effects (1-5%) on heat content per gallon will require additional fuel
constanption to achieve rated boiler output.
2.4 GASIFICATION PROCESSES FOR PRODUCI1 G CLEAN GASEOUS FUELS
Process Description
Gasification processes can convert high-sulfur feedstocks into a fuel gas
by controlled partial oxidation (POX) with air or oxygen. No catalyst is re-
quired; thus, a wide range of fuels can be converted to clean gaseous fuels.
The desalted feedstock is partially oxidized at high temperature to form a gas-
eous mixture of carbon monoxide (CO) and hydrogen (H 2 ) with a small amount of
methane (CH 4 ). Carbon soot produced as a result of incomplete combustion is
recycled back to the process. Either oxygen or air may be used for the partial
-66-
-------
FIGURE 2-12
EFFECT OF FUEL SULFUR CONTENT ON SO 2 EMISSIONS
3.2
2.8 -
0
2.4 -
2.0
0
1.6. 0
00
El
1.2
0 0
0
0.8
Legend: ØNo. 6 Fuel Oil
/ No. 2 Fuel Oil
EJN0. 5 Fuel Oil
0.4 Source:
A EPA-650/2-74-078-a, Oct. 1974
EPA•600/2-76.Q86a April 1976
EPA-600/7-78-099a, June 1978
0 — I I I I I I I I
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8
WEIGHT % SULFUR IN FUEL
-------
FIGURE 2-13
EFFECT OF FUEL NITROGEN CONTENT ON TOTAL NITROGEN OXIDES
0.8
0
0.7
0.6- 0
0.5- o
000
0 0
0.3- 8
0.2 - 0 0 Fuel Type: ONo. 6 Fuel Oil
EJNo. 5 Fuel Oil
A No. 2 Fuel Oil
Source:
0.1 EPA-650/2-74-078-a, Oct. 1974
EPA-600/2-76 086a, April 1976
EPA-600/7-78-099a, June 1978
0 .— I I I I I I I
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3
PER CENT NITROGEN IN THE FUEL
-------
FIGURE 2-14
EFFECT OF FUEL CARBON RESIDUE CONTENT ON TOTAL PARTICULATE EMISSION
0.32
0.28 -
0.24
I—
0.20.
C r )
—I Legend: ®No. 6 Fuel Oil
(I )
0.16 . ONo. 5 Fuel Oil
I — i No. 2 Fuel Oil
-J
Source:
0
EPA-650f2-74-073.a Oct. 1974
0.12 EPA-60012-76-086a April 1976
EPA 6OO/7-78-O99a June 1978
0 0
I-
o 0.08 0
0
0
0
0.04 . 0
9
0
0.00 I I I I I I I I I
0 1 2 3 4 5 6 7 8 9 10 11 12
WEIGHT % CARBON RESIDUE
-------
oxidation, depending on the desired heating value of the product gas. When air
is used, the nitrogen remains in the product gas. When oxygen is used, the peak
temperatures are usually controlled by introducing a diluent such as steam or
carbon dioxide.
Figure 2-15 shows a flow diagram of a typical partial oxidation process. 32)
In this process, a heavy residuum and the oxidant (air or oxygen) are preheated
with steam and fed to the reactor. The hot reactor effluent [ 2400°F (1316°C))
gas containing some ash from the feed and soot (1-3% by wt. of fuel) is passed
to a specially-designed waste heat boiler which produces high pressure steam.
The crude gas exiting the waste heat system [ about 325°F (163°F)] is then passed
into a carbon removal system, which consists of two units. The bulk of the car-
bon is removed by a special carbon-water contactor of proprietary design; the
remaining carbon is removed by a cooler/scrubber water wash. The product gas
contains less than 5 ppm carbon and virtually no ash or other particulates. It
is directly usable for gas-turbine fuel after sulfur removal (see Appendix B
for a discussion of processes for removal of H 2 S from hydrocarbon gases).
The carbon produced by gasification is collected as a slurry, dewatered, and
recycled back to the process. In many systems in use, a pelletizing-homogenizing
system is used to collect and recycle soot. This is shown in Figure
If the oil feed is not too heavy, the carbon particles are preferentially coated
with oil to form pellets which then can be separated from the water phase. The
pellets are then ground up again after mixing with more oil feed and recycled
to the gasifier. If very heavy feedstocks (such as pitch from propane deasphal-
ting) are used, then the Shell Closed Carbon Recovery Syster (Figure 2-17) is
needed. Naphtha is used to collect the carbon particles, which are then trans-
ferred to the oil feedstock in a stripping
The desulfurized product gas has a heating value of approximately 300
Btu/Scf when oxygen is used for the oxidation. By comparison, the product from
-70-
-------
Figure 2-15
c oiIer Feedwater
TYPICAL PARTIAL OXIDATION PROCESS FLOW DIAGRAM (SHELL GASI FICATION PROCESS)(32)
Carbon Slurry
Waste-Water
Stripper
Incinerator
L.P. Steam
Waste- H eat
Boiler
Economizer
Waste-Water
Used with permission Taken from: Kuhre, C. J. and J. A. Sykes, Jr. Energy Technology Handbook D. M. Considine, Ed.,
McGraw-Hill Book Company, 1977 pages 2—173 & 2-174 - Copyright by Shell Development Company
Reactor
-------
Soot/Water
Slurry -
Fuel Oil
Pelletizer Waste Water Homogenizer Slurry Pump
Used with permission of Petroleum Publishing Company - Taken from Oil and Gas Journal magazine referenced issue
Figure 2-16
SHELL PELLETIZING SYSTEM FOR SOOT RECOVERY AND RECYCLE 3
Soot/Fuel
Vent Fuel Oil Slurry
Sieve
Screw
Conveyor
Recycle
to Soot
Scrubber
-------
Figure 2—17
SHELL CLOSED CARBON RECOVERY SYSTEM (33)
Cooling Water
Makeup Naphtha
Water
+ Clean
Water
Fuel and
Carbon
Carbon
Carbon
Water
Slurry
Fuel Oil
Steam
*
Recycle
Naphtha
Used with permission of Petroleum Publishing Company - Taken from Oil and Gas Journal magazine referenced issue.
-------
air oxidation has a heating value of approximately 120 Btu/Scf because of the
nitrogen remaining in the gas. Typical product-gas compositions for oxygen and
air gasification are shown in Table 2-7i 32
The process chemistry of the partial oxidation process is complicated,
even though the process system is simple. The overall reaction is approximated
by the following equation:
CnHm + (- —) 02 nCO + (— -) H 2
Following the initial heat-up phase where some cracking of the hydrocarbon
takes place, part of the hydrocarbons react with oxygen during ignition in a
highly exothermic reaction:
CnHm + (n + - -) 02 nCO 2 + (f) H 2 0
Practically all the available oxygen is consumed in this reaction whose
equilibrium is far to the right. The remaining unoxidized hydrocarbons react
with steam and the combustion products from reaction (2) via endothermic
reactions:
CnHm + nCO 2 2nCO +
CnHm + nCO 2 nCO + (- -- + n) H 2
The burner-reactor system is designed for good mixing to prevent excessive
local temperatures and to bring the complex of reactions to a thermal equilibrium
at 2350 to 2550°F (1288-1399°C) in a very short residence time. A soaking
phase takes place in the rest of the reactor where the gas is at a high tempera-
ture. The final gas composition is determined by secondary reactions of methane,
carbon, and the water—gas shift equilibrium.
-74-
-------
TABLE 2-7. TYPICAL PRODUCT GAS COMPOSITION FROM GASIFICATION PROCESSES (32)
% vol., dry basis
02 Air
oxidation oxidation
Hydrogen 48.0 12.0
Carbon Monoxide 51.0 21.0
Methane 0.6 0.6
Nitrogen 0.2 66.0
Argon 0.2 0.4
Sulfur* ppm 5 ppm
Total 100.0 100.0
* after desulfurizing using Shell Sulfinol or ADIP process
Used with permission — Taken from: Kuhre, C. J. and J. A. Sykes, Jr.
Energy Technology Handbook 1). M. Considine, Ed., McGraw-Hill Book
Company, 1977 pages 2-173 & 2-174
-75-
-------
These are slow reactions, and methane contents are higher than equilibrium
calculations predict.
During the soaking phase, a portion of the carbon (soot) disappears, accord-
ing to the reactions:
C+C0 2 2C0
C+H 2 0 CO+H
Some carbon is always present in the product gas from the reactor (approxi-
mately I to 3% of the oil feed), but generally soot is recycled to extinction.
The composition of the fuel gas is determined by the water-gas shift
equilibrium:
rn+un — rn
. l ‘2 112
which becomes fixed as the gas leaves the reactor at 2200 to 2400°F (1204 to
1316°C) and then is rapidly quenched in the waste heat boiler to about 325°F
(163°C).
Development Status
Over 200 con,nercial partial oxidation reactors are in successful operation
worldwide. 4 Two similar commercial processes are primarily in use: the
Shell Gasification process and the Texaco Synthesis Gas Generation Process.
Shell has supplied slightly more than half of the installed units. The corner-
cial application of these gasification processes for producing clean gaseous
fuels was introduced during the early 1950’s. The total capacity of the in-
stalled partial oxidation units is unknown because of the wide ranges of product
gases, ranging from aninonia to methanol to synthesis gas, and because of the
variety in feedstocks currently being used.
A recent development partially funded by the EPA has resulted in the de-
sign, construction, and testing of the Chemically Active Fluid Bed (CAFB) pro-
cess demonstration unit in San Benito, Texas. The CAFB process is designed
—76-
-------
to produce a clean, low-sulfur gas by partially oxidizing a heavy, high-sulfur
feed in a limestone fluid bed. Hydrogen sulfide and some organic sulfur are
absorbed by the lime. The remaining hot, low—sulfur fuel gas produced is ready
for combustion. The CAFB reactor contains two sections; one for gasification
of the feed, and one for regenerating the sulfur-containing limestone. During
regeneration, air reacts with the spent stone, freeing the sulfur as sulfur
dioxide. The sulfur dioxide is removed from the regeneration gas and may be
recovered in a variety of processes. The limestone is then recycled to the
gasification section until it loses its efficiency as an absorbent. This unit
is currently scheduled for operation in the summer of 1979. A schematic flow-
sheet of the CAFB demonstration unit being constructed in San Benito, Texas,
is shown in Figure 2-l8, and a cutaway drawing of the CAFB gasification!
regenerator is shown in Figure 2_19. (36) The San Benito unit is designed to
use either coal (lignite) or residual oil as a feedstock.
Applicability
The gasification processes produce low- to medium-Btu gas products from
liquid fuels. The low-Btu gas can be utilized in most, if not all, boilers
designed for oil, coal, or gas. Modifications to the fuel burners are required
for proper fuel/air control. Boiler development is progressing to the end that
a wide variety of fuels may be utilized. However, derating of an existing boil-
er could result if the flue gas volume is increased in an existing boiler, the
amount depending on the boiler design, e.g., with or without superheat. (36)
It is probable that the smaller-size industrial boiler users would not
find the CAFB useful, since additional capital expenditures and land space are
required. There are no technical constraints to prohibit the use of the CAFB
for any size industrial boilers, but an economic study needs to be made to
determine the minimum size of boiler which would make the use of the CAFB pro-
cess an attractive alternative to other control measures.
-77-
-------
Figure 2-18. CAFB CHEMICALLY-ACTIVE FLUIDIZED BED PROCESS SCHEMATIC FLOW DIAGRAM 9
Flue Gas
Boiler
Limestone Fuel
Limestone Preparation __________
Gasifier Regenerator
& __
Storage Air I -
Air
Spent
Stone ______ ______
Coal
Resox Unit
Steam - l
Ash
S — --____________________________ ___________________
£
Sulfur
Condenser
Sulfur
-------
Figure 2-19
CAFB GASIF,ER/REGENERATOR(36)
Limestone
Injector
Refractory
Air Nozzle
(Typ.)
Air to
Regenerator
Air & Flue
Gas to
Gasifier
Air & Flue Gas
Nozzles (Typ.)
Floor
Used with permission taken from: McMillan, R. E. and F. D. Zoldak, “A Discussion
Fluid Bed Process (CAFB),” Oklahoma State University Frontiers of Power Technology
Oklahoma, 26 & 27 October 1977, p. 11.
of the Chemically Active
Conference, Stillwater,
Product Gas
to Burner
l’roduct Gas
to Burner
Regenerator
Off Gas
to Resox€
Gasifier
Fluidi ;ed
Bed Level
Injector
Drain
Oil Injection Line
Oil Combustion Pit
Transfer Slot
-------
As with the CAFB, smaller-size industrial boiler users may not be attract-
ed to a POX installation due to the additional capital expenditure and land space
required, plus increased system compiexityi 7 The large volumes of compressed
air or oxygen required may be better met with a combined cycle/combustion tur-
bine installation. The application of gasification processes to the production
of clean, low—sulfur fuel gas for combustion seems to be initially restricted
to boilers or gas turbines in power plants rather than industrial boilers. The
economics of gasification processes in terms of retrofit, operation, mainten-
ance, and costs added to the fuel product will probably limit the use of this
technique as a control mechanism to utilities or large industrial complexes.
Coninercial performance and reliability of the CAFB process are yet to be
determined. Until this reliability is demonstrated, the CAFB concept is not
likely to make a significant impact on the boiler-fuel picture. However, given
the simplicity of the CAFB as a front-end add-on, it is an attractive new tech-
nology which could rapidly become important, especially for large new chemical
complexes with several industrial boilers. Its wide fuel flexibility also could
propel it into coninercial use in the next few years.
Factors Affecting Performance
There are numerous design parameters which affect the performance of gasi-
fication processes. The following is a description of some of the important
design factors to consider in producing a clean, low-sulfur fuel gas.
The combustion must be carefully controlled by limiting the oxygen due to
the fact that excess combustion will increase the carbon dioxide produced, and
insufficient combustion will produce excess carbon soot in tie gas. The design
and installation of the reactor refractory lining and the design of the burners
are critical for proper operation of the gasification process.
Much of the efficiency of the gasification process depends upon recovering
the maximum heat possible. In addition to the design of the waste heat boiler
-80-
-------
or exchanger for high—temperature recovery, special design must be used to allow
the carbon by—product to pass through the system without fouling. This fouling
would reduce the amount of steam produced, thus reducing the efficiency of the
process. One prototype unit has been in operation for nearly 20 years without
requiring gas-side cieaning.(32)
Particulate and ash carryover into the product gas are controlled by the
slurry separator and water scrubber. Ash buildup in the reactor normally re-
quires only an annual cleanout. SulfUr content of the product fuel gas is
dependent on the feedstock sulfur content and the H 2 S removal efficiency of the
absorbent in the removal processes employed. An absorbent should be chosen
that would also remove as much of the other traces of sulfur compounds, such as
carbonyl sulfide and carbon disulfide, as possible. All of the sulfur compounds
will normally form sulfur dioxide when the product gas is burned. Generally,
greater than 99% removal of the sulfur compounds is readily achievable with
the existing sulfur removal processes.(38)
In the case of the CAFB, the sulfur content of the fuel gas is affected by
the calcium-to-sulfur ratio in the fuel bed. With the integrated gasifier-
regenerator, this ratio can be maintained within narrow operating limits. It
is, therefore, thought that the SO 2 concentration in the CAFB product gas could
be controlled within narrow limits by controlling the firing rate and the
calcium—to-sulfur ratio in the fuel bed.
2.5 PERFORMANCE OF PRODUCT FUEL GASES ON INDUSTRIAL BOILERS
Emission Reductions
Sulfur oxide emissions from the combustion of fuel gas from gasification
processes are directly dependent on the sulfur content of the product gas.
Gas treatment processes can effectively remove up to 99.9% of the sulfur in the
feedstock to give a typical product gas having a sulfur concentration of about
five parts per million (see Appendix B for a discussion of gas treatment
-81 -
-------
processes). The sulfur content of the fuel gas is based on the efficiency of
the H 2 S removal unit in a typical partial oxidation process.
The N0 emissions are dependent on boiler design and firing characteris-
tics for low-Btu fuel gases produced using air as an oxidant. The nitrogen con-
tent of a typical product gas from gasification processes using air oxidation
approaches 70% by volume on a dry basis. However, 0.2% nitrogen is typical for
gasification processes using oxygen (see Table 2-7). When oxygen is used for
oxidation, the product gas is more costly per Btu; but, the potential for N0
emissions may be greatly reduced due to the low fuel nitrogen content. In
theory, the total NO emissions could be minimized by eliminating the nitrogen
from the partial oxidation process and subsequent combustion. This route is
not considered to be economically attractive at the present time.
Particulate emissions from the cont ustion of fuel gas are directly related
to the carbon soot present in the product gas. The carbon soot may be removed
from the product gas quite effectively by using water scrubber techniques. The
particulate emissions from the combustion of fuel gases from heavy feedstocks
are estimated to be below 0.1 pound per million Btu’s. NO emissions are esti-
mated to be below 0.2 pound per million Btu s using air oxidation, and sulfur
oxide concentrations in the flue gas may be less than five parts per million.(36)
Nitrogen oxide emissions in gasification processes cannot be predicted with
any degree of certainty at this time, except to say that N0 emissionS should be
lower than with conventional systems due to the lower temperatures involved in
the combustion. The estimated emission reductions presented here are based on
the operation of conventional gasification processes producing a wide variety
of product gases.
In act on the Boiler
It should be feasible to burn cleaned, low-sulfur fuel gas in existing gas-
fired boilers. However, because of the low heating value of this gas, some
-82-
-------
derating of the boiler may result, as discussed under Applicability.
Converting an existing oil-fired industrial boiler to a low-Btu fuel gas may
require a major redesign of the system. In a new installation, the boiler de-
sign would consider the heating value of the fuel gas.
For small industrial boilers, it does not appear economically feasible to
utilize fuel gas from partial oxidation processes. The cost of processing
equipment and related controls for the integration of low-sulfur fuel gas in an
industrial boiler could not be justified.
-83-
-------
REFERENCES
1. Devitt, 1., et al., PEDCo. Environmental, Inc., “The Population and Charac-
teristics of Industrial/Coninercial Boilers,” May 1979, Pp. 72-75.
2. Devitt, T., et al., PEDCo. Environmental, Inc., “The Population and Charac-
teristics of Industrial/ConinerCial Boilers,” May 1979, p. 30.
3. Devitt, T., et al., PEOC0. Environmental, Inc., “The Population and Charac-
teristics of Industrial/Coimnercial Boilers,” May 1979, pp. 43-44.
4. Bartok, W., A. R. Crawford, and A. Skopp, “Control of NOx Emissions From
Stationary Sources,” Chemical Engineering Progress , Volume 67, No. 2,
February 1971, Pp. 64-72.
5. “RefIning Process Handbook,” Hydrocarbon Processing , September 1978, p. 100.
6. Ranney, Maurice William, “Desulfurization of Petroleum,” Noyes Data Corpor-
ation, Park Ridge, New Jersey, 1975, pp. 3, 31.
7. Hastings, K. H., and R. P. Van Driesen, “HydrodesulfurizatiOn of Petroleum
Crude Oil Fractions and Petroleum Products,” Enerqy Technology Handbook ,
D. M. Considine, Ed., McGraw-Hill Book Company, 1977, pp. 3-254.
8. Frayer, J. A., et al., “Gulf’s HDS Processes for High Metals Stocks,”
presented at the Japan Petroleum Institute, Tokyo, Japan, 8 May 1975,
Figures 1, 2, and 3.
9. Tyndall, M. F., et al., “Envirormiental Assessment for Residual Oil Utiliza-
tion,” Catalytic, Inc., EPA-600F7-78- 175.
10. Nelson, W. 1., “Data Correlation Shows the Amount of Hydrogen Used in Desul-
furizing Residua,” Oil and Gas Journal , 28 February 1977, pp. 126—128.
11. Nelson, W. L., “Catalyst Consumption Required in Desulfurizing Residua,”
Oil and Gas Journal , 15 November 1976, pp.. 72-74.
12. Edelman, A. M., et al., “A Flexible Approach to Fuel Oil DesulfurizatiOfl,”
presented at Japan Petroleum Institute, Tokyo, Japan, 8 May 1975, p. 5, and
Figures 2, 5, 6, 9, 10, and 11.
13. Cantrell, Ailleen, “Annual Refining Survey,” Oil and Gas Journal , 20 March
1978, pp. 108 and 113.
14. “Refining Processes Handbook,” Hydrocarbon Processing , September 1978,
pp. 99—224.
15. Aalund, Leo, “Hydrodesulfurizatiofl Technology Takes on the Sulfur Challenge,”
Oil and Gas Journal , 11 September 1972.
16. “Technology Improves in Processing Sour Residua,” Oil and Gas Journal ,
19 August 1974.
17. Yanik, S. J., et al., “Gulf HDS Process Paves Way for Residual Oil Upgrading,”
Oil and Gas Journal , 16 May 1977.
-84-
-------
REFERENCES (continued)
18. Paraskos, J. A., et al., “Ecologically-Acceptable Fuels From the Gulf HDS
Process,” 67th Annual Meeting, AIChE, Washington, D. C., 1-5 December 1974.
19. Paraskos, J. A., et al., “Here’s How Residual Oils Are Desulfurized,” Oil
and Gas Journal , 26 May 1975, pp. 90-93. —
20. “Refining Processes Handbook,” Hydrocarbon Processing , September 1974,
pp. 103—214.
21. Kubo, Junichi, “Japanese Residua HDS Process Looks Good,” Oil and Gas JQurnal ,
11 November 1975, pp. 105-108.
22. Von Ginneken, A. J. J., “Shell Process Desulfurizes Resid,” Oil and Gas
Journal , 28 April 1975, pp. 59-63.
23. “HDS Ready With Alaskan North Slope Crude Oil,” Oil and Gas Journal ,
7 February 1977.
24. Young, B. J., “Resid Desulfurizer a Year Later,” Hydrocarbon Processing ,
September 1977, pp. 103-108.
25. Cato, G. A., L. J. Muzio, and D. E. Shore, “Field Testing: Applications of
Combustion Modifications to Control Pollutant Emissions,” KVB Engineering,
Inc., EPA-600/2-76-086(a), April 1976.
26. Cato, G. A., et al., “Field Testing: Applications of Combustion Modifica-
tions to Control Pollutant Emissions,” EPA-65O/2-74-O78(a), October 1974.
27. Ondish, G. F., “The Gulf HDS Process,” Gulf Research and Development Co.,
16 August 1974, p. 20.
28. Yamamato, M. 0., “Deep Desulfurization of Atmospheric Residue by the RCD
Unibon Process,” API Refining Meeting, Chicago, Illinois, 12 May 1977,
pp. 3 and 12.
29. “Residue Desulfurization,” Hydrocarbon Processing , September 1978, p. 130.
30. Oxenreiter, M. F., et al., “Desulfurization of Khafji and Gach Saran Resids,’
American Oil Co., 30 November 1972, pp. 2 and Tables I, II, and III.
31. Carter, W. A., H. J. Buenirig, and S. C. Hunter, “Emission Reduction on Two
Industrial Boilers With Major Combustion Modifications,” KVB Engineering,
Inc., EPA-600/7-78-O99(a), June 1978.
32. Kuhre, C. J. and J. A. Sykes, Jr., Energy Technology Handbook , D. M. Considine,
Ed., McGraw—Hill Book Company, 1977, pages 2-173 & 2—174.
33. Kuhre, C. J., and C. L. Reed, “Shell Non-catalytic Gasification Process
Leads to SNG,” Oil and Gas Journal , 12 January 1976, pp. flO—118.
34. Strelzoff, Samuel, “Partial Oxidation for Syngas and Fuel,” Hydrocarbon
Processing , December 1974, p. 74.
-85-
-------
REFERENCES (conti nued)
35. Turner, P. P., S. L. Rakes, and T. W. Petrie, “Advanced Oil Processing Util-
ization Environmental Engineering — EPA Program Status Report,” EPA-600/
7-78-077, May 1978, p. 43.
36. McMillan, R. E., and F. 0. Zoldak, “A Discussion of the Chemically Active
Fluid Bed Process (CAFB),” Oklahoma State University Frontiers of Power
Technology Conference, Stiliwater, Oklahoma, 26 and 27 October 1977.
37. Werner, Arthur 5., et al., “Preliminary Environmental Assessment of the
GAFB,” GCA Corporation, EPA-600/7-76-017, October 1976.
38. Laengrich, Arthur R., “Tail-Gas Cleanup Addition May Solve Sulfur Plant
Compliance Problem,” Oil and Gas Journal , 27 March 1978, p. 159.
-86-
-------
SECTION 3
CANDIDATES FOR BEST SYSTEMS OF EMISSION REDUCTION
FOR CLEAN OIL TECHNOLOGY
3.0 INTRODUCTION
In Section II, we examined three methods of emission control which are
applicable to oil-fired industrial boilers. The three techniques considered
were:
1. Hydrotreating (HDS) in which fuel oil is treated with
hydrogen to produce a clean oil suitable for combustion;
2. Partial Oxidation (POX) in which fuel oil is partially
oxidized with oxygen or air to produce a gas which is
then scrubbed to remove sulfur, thus making it suitable
for combustion; and
3. Chemically Active Fluid Bed (CAFB) in which fuel oil is
partially oxidized in a fluid bed of limestone to pro-
duce a clean gas suitable for combustion.
In this section, we select the best system of emission reduction and
recommend guideline control levels of regulatory options to best achieve
moderate, stringent, and intermediate levels of control.
3.1 SELECTION CRITERIA
The factors considered in the selection of the best system of emission
reduction from those systems discussed in Section II were:
1. Performance
2. Applicability
3. Status of Development
4. Cost Considerations
5. Energy Considerations
6. Environmental Considerations
-87-
-------
In the selection process, an effort was made to rate each of the control
systems against each of the selection criteria listed above. For each cri-
terion, the best system was designated 1, the next best 2, etc. The lowest
overall score for all criteria was adjudged to be the best system. A summary
of the rating evaluation is given in Table 3-1.
3.2 BEST SYSTEMS
From the rating matrix developed in Section 3.1, it is concluded that
Hydrotreating (HDS) offers the best system of emission reduction for clean oil
technology. A discussion of the ratings for the different selection criteria
is given below:
a) Performance - All three systems will yield a fuel that is
environmentally acceptable for burning in a boiler. The
overriding consideration in selecting HOS as the best
system is its negligible or minor impact on boiler per-
V formance. Cleaned liquid fuels are directly applicable
to existing industrial boilers with little negative im-
pact on boiler performance. The modest reduction (1-3%)
of heat content per gallon of cleaned oil will require
additional fuel consumption to achieve rated boiler out-
put. However, burning cleaner fuels will lower the
severity of operation and maintenance on the boiler.
b) Applicability - The HDS system is a clear selection for
applicability, since the clean oil produced from HDS can
be directly utilized in existing boilers with little im-
pact on the boiler physical facilities. The gasification
-88-
-------
TABLE 3-1. RATING MATRIX FOR SELECTION OF BEST SYSTEM OF
EMISSION REDUCTION FOR CLEAN OIL TECHNOLOGY
Sel ecti on Criteria
Control System 3 4 5 6 Total
HDS 1 1 1 1 1 1 6
POX 2 2 2 3 2 3 14
CAFB 3 3 3 2 3 2 16
NOTE: Selection Criteria
1. Performance
2. Applicability
3. Status of Development
4. Cost Considerations
5. Energy Considerations
6. Environmental Considerations
-89-
-------
processes require the addition of equipment with attendant
cost and space impacts. The retrofitting of either gasi-
fication process to an existing industrial boiler could,
in some cases, be extremely difficult, if not impossible.
c) Status of Development — Hydrotreating processing has been
in commercial existence for more than 20 years,- and over
20 hydrotreating processes are actively in use The
current United States refinery desulfurization capacity is
more than 1.8 million barrels per day from 86 plants.
However, only 19 plants have direct resid or heavy gas oil
hydrodesulfurization facilities which provide a total de-
sulfurization capacity of approximately 0.6 million barrels
per dayc2)
Hydrotreating is an extremely versatile process which is
used to desulfurize, denitrogenate, and demetallize fuel
oils prior to combustion. It can be adapted to a wide
variety of feedstocks ranging from low-sulfur crude oils
to high-sulfur residual oils. Hydrotreating used in
conjunction with blending can produce fuel oils with al-
most any characteristics desired. Because of its versa-
tility and widespread use, hydrotreating has been select-
ed-as the best candidate under the status of development
criterion.
Partial oxidation is a commercially-proven process with
more than 200 installations worldwide It, too, is a
versatile process that has been used with feedstocks
ranging from natural gas through naphtha, residual oil,
and even coal. Its primary use has been to produce
-90-
-------
synthesis gas for the manufacture of methanol or ammonia;
however, there is no technical reason that the synthesis
gas cannot be used as a boiler or turbine fuel. The pro-
cess can be designed to use either air or oxygen as the
oxidizing medium, and the product gas will have a heating
value of 120 Btu/SCF or 300 Btu/SCF dependent upon whe-
ther air or oxygen is used.
The Chemically Active Fluid Bed (CAFB) process is an
attractive new technology which could become a significant
factor over the next few years in both the utility and
industrial areas. It, too, is a versatile process which
can be used with a wide selection of feedstocks. The com-
mercial performance and reliability of the Cl LFB process
are yet to be determined. Until this reliability is dem-
onstrated, the CAFB concept is not likely to make a sig-
nificant impact on the boiler-fuel picture. A CAFB demon-
stration unit has been constructed and is currently under-
going testing at the La Palnia station of Central Power and
Light Company at San Benito, iexasc
d) Cost Considerations - The cost of upgrading liquid fuels
in a few large refinery complexes is less demanding than
individual emission control techniques at each industrial
boiler. In the first place is the matter of mere numbers,
wherein we are comparing the cost of desulfurization
facilities at less than 100 refineries against individual
emission control systems at literally thousands of indus-
trial boiler installations. In addition, the economics of
scale would greatly favor the installation of large central
-91-
-------
hydrotreating units at refineries rather than smaller in-
stallations at individual boilers. For example, a
150 x io6 Btu/hr.oil-fired boiler requires approximately
25 barrels/hr. of oil or 600 barrels per day. A typical
50,000 BPD HDS unit could supply nearly 85 such boilers.
Preliminary cost figures show that typical hydrotreating
facilities can be installed for investments ranging from
$500 per barrel per stream day capacity for moderate
levels of emission control up to $1,600 per barrel per
stream day capacity for the most stringent levels of con-
trol ’ The costs of producing such cleaned fuels range
from $0.91 to as much as $5.84 per barrel. The cost of
partial oxidation units ranges from $3,000 to $6,000 per
barrel per stream day with operating costs ranging from
$1.50 to $7.00 per barrel 6) Cost information on CAFB
units is very sketchy, but preliminary figures indicate
an investment cost of $3,500—$4,500 per barrel per stream
day and an operating cost of about $4.30 per barrel.
e) Energy Considerations - The selection of hydrotreating
under this criterion closely parallels the reasoning used
for the cost considerations in that the selection is
largely determined by numbers and sizes of units. The
energy impacts are reflected by the operating cost fi-
gures given in the previous paragraph, which indicate
that hydrotreating has advantages over both partial oxi—
dation and chemically active fluid bed processing.
f) Environmental Considerations — There is little difference
between the three control systems from an environmental
viewpoint. From a practical viewpoint, it is more
—92—
-------
advantageous to burn cleaned liquid fuels in industrial
boilers rather than handle the environmental problems of
untreated fuel oil at each industrial site. The control
of potentially—hazardous pollutants can be more effective-
ly managed at the refinery, and the emissions from com-
bustion at industrial boilers can then be more effectively
controlled by monitoring the fuel quality.
3.3 REGULATORY OPTIONS
Three regulatory options, which represent moderate, intermediate, and
stringent levels of control for SO 2 , NOR, and particulate emissions, have
been selected. The selected emission levels were derived from actual emis-
sion data from industrial boilers. Figure 3-1 shows the relationship of sul-
fur content of the fuel with sulfur oxides emissions; Figure 3-2 shows a
similar relationship for NO emissions versus nitrogen content of the fuel;
Figure 3-3 shows the effect of carbon residue content of the fuel on the par-
ticulate emissions.
Table 3-2 gives the maximum sulfur, nitrogen, and carbon residue content
of cleaned fuel oil to meet the recommended control levels. The suggested
regulatory options are based upon commercially-available systems for the pro-
duction of low-sulfur fuels from high-sulfur feedstocks. The selected levels
of control are based upon the use of residual fuel oil and represent the degree
of desulfurization that can be attained using typical refinery processes and
technology. Table 3—3 gives the commercially-available hydrodesulfurization
technology and illustrates the degree of desulfurization that can be attained
with the various processes, feedstocks, and treating conditions.
For the moderate level of control, a suggested fuel content of 0.8% sul-
fur, 0.3% nitrogen, and 12% carbon residue represents a residual fuel oil that
is readily achievable from a number of refinery practices. For the stringent
-93-
-------
F$GURE 3-1
EFFECT OF FUEL SULFUR CONTENT ON SO 2 EMISSIONS
3.2
2.8.
2.4 -
2.0. 0
0
0
0
1.6.
0
1.2
0
0.8
Legend: ØNo. 6 Fuel Oil
£No. 2 Fuel Oil
EINo. 5 Fuel Oil
0.4 Source:
A EPA.650/2-74-078-a, Oct. 1974
EPA-600/2-76-086a. April 1976
EPA-600/7 -78-099a. June 1978
o . I I I I I
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8
WEIGHT % SULFUR IN FUEL
-------
FIGURE 3-2
EFFECT OF FUEL NITROGEN CONTENT ON TOTAL NITROGEN OXIDES
0.8 -
0.7 -
0.6- El
U,
i 0.5
01
I o
0
L J
0.4 o ®®
0.3-
0.2 El ( ) Fuel Type: (DNo. 6 Fuel Oil
ElNo. 5 Fuel Oil
i No. 2 Fuel Oil
Source:
0.1 EPA-650/2-74-078-a, Oct. 1974
EPA-600/2-76-086a. April 1976
EPA-600/7-78-099a, June 1978
0 I I I I I I I I I —1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3
PER CENT NITROGEN IN THE FUEL
-------
FIGURE 3-3
EFFECT OF FUEL CARBON RESIDUE CONTENT ON TOTAL PARTICULATE EMISSION
0.32
0.28
0.24
I -.
0.20
U,
‘° — Legend: ONo. 6 Fuel Oil
0.16 9No. 5 Fuel Oil
U)
No. 2 Fuel Oil
D Source:
U
EPA .650/2-74-078 .a, Oct. 1974
0.12 EPA 600/2-76-086a, April 1976
0 El EPA-600/7-78-099a, June 1978
a.
-J 0 0
0.08. El
I. - 0
0
0
0.04 0
0
0
0.00
0 1 2 4 5 6 9 10 11 12
WEIGHT % CARBON RESIDUE
-------
TABLE 3-2. SUGGESTED POLLUTANT CONTENT OF CLEANED FUEL OIL
TO MEET RECOMMENDED CONTROL LEVELS
Pollutant
so 2
N0
Parti cul ates
Stringent
Emission Max. Fuel
#1106 Btu Content
0.1 0.1% S
0.2 0.15% N
0.05 3.0% C.R.*
Control Level
Intermediate
Emission Max. Fuel
#1106 Btu Content
0.3 0.3% S
0.22 0.2% N
0.1 6.0% C.R.
Moderate
Emission Max. Fuel
#1106 Btu Content
0.8 0.8% S
0.3 0.3% N
0.25 12% C.R.
* Note : % C.R. = weight percent carbon residue in fuel oil
—97—
-------
TABLE 3-3. COMMERCIAL HYDRODESULFURIZATION TECHNOLOGY*
Final H2 Feedstock
Sulfur Initial V + NI Consumption
Level (Wt. %) S (Wt. %) Desuif. ppm (Wt j _ ( SCF/BBL) Source Type Process
01 3.9 97 NR 975 Athabasca Tar Sands GO-Fining
3.8 97 60 960 Kuwait AR Gulf IV
3.8 97 60 700 Kuwait AIB Gulf III
2.9 96 117 410 Arab Heavy Crude GO-Fining
0.46 87 5 NR S. Louisiana AR Gulf II
0.2 3.9 95 60 580 KuwaIt AR H-Oil
3.0 93 0 450 KuwaIt VGO H-Oil
2.8 93 35 NR Qatar AR Shell
2.5 90 0 232 Kuwait VGO IFP
2.3 90 40 220 Arab Light AR GO-Fining
1.9 90 ?20 220 Gach Saran AR GO-Fining
0 3 4.2 93 120 915 Arab Heavy ATB RESIDfining
3.8 92 66 770 Kuwait AR Gulf III
3.8 92 46 730 Kuwait AR tinicracking HDS
3.0 90 40 720 Mid East “A” AR Gulf III
3.5 92 60 706 Kuwait AR RCD Unibon
2.1 86 392 NR Ceuta Crude Gulf IV
3.8 92 60 640 Kuwait RIB Gulf III
2.5 88 220 625 Gach Saran ATB RES lOfining
2.5 88 210 625 Iran Heavy ATB RES lDfining
2.9 90 37 592 Arab Light AR Nippon
2.4 88 220 570 Gach Saran AR Union
3.0 90 40 535 Arab Light AR Gulf III
3.0 90 40 530 Arab Light AR Union
3.8 88 110 480 Iran Light AR Gulf IV
1.6 82 44 340 N. Slope AR Union
2.9 90 105 300 Khafji AR GO-Fining
3.0 90 66 280 Kuwait AR GO-Fining
2.6 87 123 NR Iran Heavy AR Shell
2.2 85 21 NR W. Texas AR Gulf II
1.8 84 30 NR Oman AR Shell
* See key at end of table.
-------
TABLE 3-3. COMMERCIAL HYDRODESULFtJRIZATION TECFINOLOGY* (Cont’d.)
Final
Sulfur
Level (Wt. %)
Initial
S (Wt. %)
%
Desuif.
V + Ni
ppm (Wt.)
H 2
Consumption
(SCF/BBL)
Feeds toc k
T12
Process
0.4
0.5
0.6
0.7
0.8
0.9
4 .1
4.0
2.7
3.2
2.6
4.4
4.2
4.2
4.2
3.8
4.0
3.8
3.8
3.0
2.4
1 .8
4.2
4.2
4.2
3.9
3.0
2.5
4.6
3.8
4.1
4.2
4.3
3.5
5.7
5.7
89
90
86
86
85
89
88
88
88
87
88
87
86
83
79
72
85
85
85
85
80
74
85
80
82
76
81
75
84
85
63
64
179
38
18
118
50
114
66
60
105
66
55
40
210
120
66
66
66
115
40
183
116
274
101
114
115
413
163
199
760
690
NR
560
NR
950
864
825
NR
815
700
660
640
530
480
295
NR
NR
N R
750
490
440
965
1125
1000
655
NR
1400
1250
1250
Source
Kuwait
Kuwait
Iran Heavy
Arab Light
Qatar
Khafji
Mid East
Mid East “0”
Kuwait
Kuwait
Khafj I
Kuwait
Kuwait
Arab Light
Gach Saran
U. S. Domestic
Kuwait
Kuwait
Kuwait
Arab Heavy
Mid East “A”
Iranian
Mid East “E”
Agha Jan
Arab Light
Mid East “B”
Arab Heavy
Gach Saran
Kuwait
Kuwait
AR
IFP
AR
LC Fining
AR
Shell
AR
ROS
AR
Shell
AR
Gulf III
AR
Shell
AR
Gulf III
AR
Shell
AR
Gulf III
AR
RESlDfining
AR
Gulf III
AR
RESIDfining
AR
RESlDfining
AR
RESlDfining
AR
Gulf II
AR
Shell
AR
Shell
AR
Shell
AR
RDS
AR
Gulf II
AR
Gulf II
AR
Gulf III
VR
Gulf II
VR
Gulf II
AR
Gulf II
VT
VTB + VGO
VR
LC-Fini ng
Gulf II
Gulf II
* See key at end of table.
-------
TABLE 3-3. COMMERCIAL HYDRODESULFURIZATION TECHNOLOGY* (Cont’ d.)
Final H2 Feedstock
Sulfur Initial V + Ni Consumption
Level (Wt.%) S (Wt. % ) Desuif. pm (Wt.1 ( SCF/BBL) Source Type Process
1.0 4.9 80 NR 1400 Athabasca Bitumen H-Oil
3.5 71 413 1220 Gach Saran VTB H-Oil
5.2 79 130 NR Kuwait VR Shell
5.1 80 115 960 Arab Heavy VR VRDS
4.1 76 38 780 Arab Light VR VRDS
4.4’ 77 118 685 Khafji VR Gulf II
3.8 74 60 630 Kuwait AR Gulf II
3.8 74 66 515 Kuwait AR Gulf II
3.8 74 60 490 Kuwait ATB Gulf II
3.8 74 75 480 KuwaIt AR H Oil
1.2 2.9 75 195 1410 Arab Heavy VTB LC-Fining
1.4 4.0 65 64 520 Kuwait AR LC-F lning
1.5 5.8 74 110 820 MId East “C” AR Gulf II
SOURCES : 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
-------
KEY TO TABLE 3-3
Data
NR = Not Reported % Desuif. = Initial S - Final S 10
Initial S 0
Feeds toc ks
AR = Atmospheric Residuum
ATB = Atmospheric Tower Bottoms
VGO = Vacuum Gas Oil
VR = Vacuum Residuum
VTB = Vacuum Tower Bottoms
Licensor Process Tradename
Exxon GO—Fining
Gulf Gulf II HDS
Gulf Gulf III HDS
Gulf Gulf IV HDS
Hydrocarbon Research H-Oil
Institut Francais du Petrole HDS
Cities Service/C. E—Lurmius LC-Fining
Nippon Oil HDS
UOP RCD/Unibon
Chevron RDS Hydrotreating
Exxon RES lDfining
Shell International Residual Oil Hydrodesulfurization
Union Oil Unicracking/HDS
Chevron VRDS Hydrotreating
-101-
-------
level of control, a suggested fuel content of 0.1% sulfur, 0.2% nitrogen, and
3% carbon residue represents the highest technically achievable residual fuel
oil that can be attained with current technology. It is evident from Figures
3—i, 3-2, and 3-3 that more stringent levels of control can be achieved from
the direct use of distillate fuel oils; however, the supply of distillate oils
is limited. Therefore, it is necessary to establish recommended control lev-
els that can be achieved with hydrotreated or cleaned residual fuel oils.
Nearly all processes for producing low-sulfur fuels oils rely on the
vaporization of distillate, gas oil, or cycle stock materials that can be hy-
drodesulfurized (HDS) to very low sulfur contents. These desulfurized oils
can then be blended with high-sulfur residual oils to produce fuel oils of
moderate sulfur contentc
There are rugnerous process variations for producing moderate sulfur con-
tent fuel oils, the selection of which is dependent upon the quality of the
crude, the amount of upgrading required, and the fuel balance at the refinery.
Nearly 95 percent of the low—sulfur fuel oil produced in the U. S. is from one
of the following methods or combination of methods: crude distillation, in-
direct desulfurization of vacuum gas oil, or desulfurizatiOn of feedstock
from catalytic cracking, delayed, or fluid coking.
The lowest cost production method is the straight distillation of low-
sulfur crudes. Some crudes are low enough in sulfur that simple crude dis-
tillation not only yields distillates and gas oils low in sulfur but even the
700°F+ topped crudes may contain less than 0.4% sulfur. These low-sulfur
products can be blended with higher-sulfur products from other refinery pro-
cesses, thus avoiding any direct hydrodesulfuriZation processing.
Another method, called indirect desulfurization, is used as a moderate
reduction control method. This method involves flashing of topped crude under
vacuum and hydrodesulfuriZation of the vacuum gas oil (VGO). Up to 95 percent
of the sulfur in the vacuum gas oil can be removed by hydrodesulfuriZatiOfl
-102-
-------
yielding a desulfurized oil containing from 0.1 to 0.3 percent sulfur. These
desulfurized oils can be blended with residual oils to give moderate control
level fuels.
Increasingly-stringent restrictions on sulfur emissions have reduced the
available markets for high-sulfur fuel oils and have forced the refinery indus-
try into more and more bottoms processing. One such process is residuum desul-
furization, which is used to produce low-sulfur fuel oils directly from atmos-
pheric distillation bottoms. The level of desulfurizatiori required for inter-
mediate reduction control is removal of about 80% of the sulfur from the atmos-
pheric residua, which may contain 3.5 percent, or more, of sulfur. Higher
levels of desulfuri2ation can be achieved but only at greatly-increased costs.
Additional processing steps are needed for handling atmospheric bottoms rather
than vacuum gas oil. For example, feed filters are required to remove various
contaminants such as dirt and carbon usually found in residual oils. In addi-
tion, guard reactors are required to remove metals such as nickel and vanadium
which would otherwise rapidly deactivate the expensive hydrodesulfurization
catalysts.
The crude type also has a significant effect on the processing cost from
atmospheric residuum. The properties of atmospheric residua derived from typi-
cal crudes may vary widely in sulfur content, metals, and carbon. Two major
processing expenses are the cost of hydrogen, which is a function of the sul-
fur and nitrogen content of the feed, and the cost of catalyst, which is a
function of the metals content of the feed. The hydrogen- and catalyst-
• related costs typically represent 65-75% of the total processing cost of mak-
ing low-sulfur fuel oil from atmospheric bottoms 18)
The crude type also has a significant effect on the desulfurization effi-
ciency of a processing unit. Table 3-4 illustrates how the performance of a
unit designed to reduce Kuwait atmospheric residuum to a 0.3% sulfur product
is rated for three other residsc 19
—103-
-------
TABLE 3-4. EFFECT OF CRUDE SUITCHES ON HDS UNIT CAPABILITY
650°FF Atmospheric Bottoms
Product*
Sulfur, fietals, Gravity, Sulfur
Crude bit. % ppm °API Wt. %
Kuwait (design) 4.11 60 14.8 0.30
Light Arabian 2.99 30 17.7 0.19
Heavy Iranian 2.50 221 15.0 0.29
Heavy Arabian 4.19 120 12.3 0.43
* 11 months - cycle length at constant operating conditions
Source : Edelman, A. M., et al, “A Flexible Approach to Fuel Oil Desulfuri-
zation,” Japanese Petroleum Institute Meeting, 8 May 1975L19)
-104-
-------
For the stringent level of control, very rigorous hydrodesulfuriZation of
atmospheric residuum is necessary. To achieve, this high level of desulfuriza-
tion with high product yields, an average desulfurization efficiency of 97% is
required. To reduce the sulfur content of high-sulfur residuals to 0.1% by
weight requires two or more stages of hydrodesulfurizatiOn or one stage of
hydrodesulfurization coupled with other conversion processes such as fluid
coking or demetallization. These techniques for producing 0.1% sulfur fuel
oil from high sulfur resids have been in commercial operation for approximately
3-5 years.
3.4 SUMMARY
Table 3-5 summarizes the characteristics of typical hydrodesulfurizatiOfl
processes by level of sulfur content reduction. As shown in the table, the
average capital investment, as well as the overall energy requirements, in-
crease with increasing degree of desulfurization.
Hydrotreating processes which produce cleaned liquid fuels are considered
the best system of emission reduction applicable to oil-fired industrial boil-
ers. The processes which clean oil by gasification are either not generally
suited to the small scale of industrial boilers (POX) or are not corni ercia1ly
demonstrated (CAFB).
Guideline control levels or regulatory options to best achieve moderate,
stringent, and intermediate levels of control are selected.
Hydrotreating processes are considered the best emission control tech-
niques for all three levels of control.
-105-
-------
TABLE 3—5. CHARACTERISTICS OF TYPICAL HYDRODESULFURIZATION
PROCESSES BY LEVEL OF REDUCTIOfl IN SULFUR CONTENT*
Control Levels
Factor Moderate Intermediate Stringent
Investment,
$ per bpsd capacity 680-1150 1080-1410 1360-1 620
Utilities, units/bbl. feed
fuel fired, MBTU 82-90 92-95 94-96
p er, kwh 5.7-8.0 7.2-9.7 7.6-9.9
cooling water, gal. 120-140 140-150 150-160
steam, MBTU 28-49 44-80 49-84
Wt. Sulfur in product 0.8 0.3 0.1
* Basis - Range of values for five residual oils encompassing low- to high-
sulfur contents (2.1—4.6%) and low to high metals (60-292ppm). Details
are discussed in Section 4.
-106-
-------
REFERENCES
1. “Census of Oil Desulfurization to Achieve Environmental Goals,” AIChE Sym-
posium Series, No. 148, Volume 71.
2. Cantrell, Ailleen, “Annual Refining Survey,” Oil and Gas Journal , 20 March
1978, p. 108.
3. Streizoff, Samuel, “Partial Oxidation for Syngas and Fuel,” Hydrocarbon
Processing , December 1974, p. 74.
4. Turner, P. P., S. L. Rakes, and T. W. Petrie, “Advanced Oil Processing
Utilization Environmental Engineering - EPA Program Status Report,”
EPA-600/7-78-077, May 1978, p. 43.
5. “Refining Processes Handbook,” Hydrocarbon Processing , September 1978,
pp. 99—224.
6. Cost studies being performed by Catalytic, Inc. under EPA Contract No.
68—02—2155 (unpublished).
7. “Refining Processes Handbook,” ydrocarbon Processing , September 1978,
pp. 99-224.
8. Aalund, L., “Hydrodesulfurization Technology Takes On the Sulfur Challenge,”
Oil and Gas Journal , 11 September 1972, pp. 79-104.
9. Aalund, L., “Technology Improves in Processing Sour Residue,” Oil and Gas
Journal , 19 August1974, pp. 62-64.
10. Yanik, S. J., et al., “Gulf HDS Process Paves Way for Residual Oil Up-
grading,” Oil and Gas Journal , 16 May 1977, pp. 139-145.
11. Paraskos, J. A., et al., “Ecologically-Acceptable Fuels From the Gulf HDS
Process,” 67th Annual Meeting AIChE, Washington, D. C., 1—5 December 1974.
12. Paraskos, J. A., et al., “Here’s How Residual Oils Are Desulfurized,”
Oil and Gas Journal , 26 May 1975.
13. “Refining Processes Handbook,” Hydrocarbon Processing , September 1974,
pp. 103-214.
14. Kubo, Junichi, “Japanese Residua HDS Process Looks Good,” Oil and Gas
Journal , 11 November 1975, pp. 105-108.
15. von Ginneken, A. J. J., “Shell Process Desulfurizes Resid,” Oil and Gas
Journal , 28 April 1975, pp. 59-63.
16. von Ginneken, A. J. J., “HDS Ready With North Slope Resid,” Oil and Gas
Journal , 7 February 1977, pp. 67—70.
17. Young, B. J., “Resid Desulfurizer a Year Later,” Hydrocarbon Processing ,
September 1977, pp. 103-108.
18. Nelson, W. L., “What Processes Make Desulfurized Fuel Oils,” Oil and Gas
Journal , 15 August 1977, pp. 32-33.
— 107-
-------
REFERENCES (continued)
19. Edelman, A. M., et al., “A Flexible Approach to Fuel Oil Desulfurization,”
Japanese Petroletin Institute meeting, 8 May 1975, Figures 10 and 11.
-108—
-------
SECTION 4
ECONOMIC IMPACT OF BEST EMISSION CONTROL SYSTEM
4.0 INTRODUCTION
In Section 3, we selected hydrodesulfurization (HDS) as the best system
of emission reduction for clean oil technology and recommended guideline con-
trol levels of regulatory options to best achieve moderate (O.8%S) , interme-
diate (O.3%S), and stringent (O.l%S) levels of control.
In this section, we determine the cost of hydrodesulfurization to produce
cleaned fuel oils to meet the required control limits and assess the economic
impact of burning desulfurized oils in industrial boilers.
In our cost analyses, only direct desulfurization of residual fuel oil is
considered. Indirect desulfurization, or the procedure of desulfurizing a
light distillate and back blending with residua to produce the required product
level, is not capable of achieving the intermediate and stringent levels of
control and therefore is not considered in this study.
The cost of hydrodesulfurization of residual fuel oil is a function not
only of the sulfur content but also of the crude source from whence the resi-
dual was derived and of the metal content of the residual. Since there are liter-
ally hundreds of different crude oils and, consequently, a like number of resi-
dua, it is virtually impossible to select a typical residual oil that would be
representative of all these crudes. Accordingly, we have selected a group of
five residual oils which cover a range of sulfur and metal values and which
will acconinodate virtually all the known crudes within the limits covered by
these five residuals. The five residua considered in this section can be
classified as follows:
-109-
-------
Residua Classification
1. Ceuta Low sulfur, high metals
2. E. Venezuelan Low sulfur, high metals
3. Kuwait Medium sulfur, low metals
4. Khafji High sulfur, moderate metals
5. Cold Lake High sulfur, high metals
The cost of hydrodesulfurization is also highly dependent upon the degree
of desulfurization. In order to cover as wide a range as possible, the hydro-
desulfurization costs were calculated for the three reconinended levels of con-
trol, as well as the State Implementation Plan (S.I.P) level of 1.6% sulfur
currently being used in most of the United States.
4.1 SUII4ARY
A sumary of the hydrodesulfurization costs for the five residual fuel
oils and four levels of sulfur content is given in Table 4-1.
TABLE 4-1. SLJ* ARY COSTS OF HYDRODESULFURIZATION OF RESIDUAL FUEL OIL
Residual
Fuel Oil Percent Sulfur in Treated Oil
ppm 1.6 0.8 0.3 0.1
Type Sulfur ( Ni + V )
Ceuta 2.12 292 0.91 2.28 3.91 5.28
E. Venezuelan 2.38 274 1.17 2.45 3.93 5.71
Kuwait 3.80 60 1.80 2.49 3.14 3.51
Khafji 4.36 118 2.20 2.85 3.60 4.11
Cold Lake 4.55 236 2.52 3.42 4.53 5.84
As evidenced from the foregoing table, the cost of HDS ranges from a low
of $0.91 per barrel for the hydrodesulfurization of a low-sulfur, high metals
residua to a high of $5.84 per barrel for the hydrodesulfurization of a high-
sulfur, high metals residua.
-110—
-------
It is also evident that the cost of HDS escalates quite rapidly with the
degree of desulfurization, going from $0.91/B for the desulfurization of Ceuta
residual to a level of 1.6% sulfur to a cost of $5.28/B for desulfurizing to
a level of 0.1% sulfur. This represents a cost of $14.46/bbl for the 1.5% S
Ceuta oil, or 39% over the cost of untreated oil.
The foregoing table also indicates that the cost of desulfurization to
the S.I.P. (l.6%S) and moderate (0.8%S) levels is primarily a function of the
sulfur level of the untreated oil; whereas, desulfurizing to the intermediate
(0.3%S) and stringent (0.1%) levels clearly reflects the influence of metals
content on desulfurization cost. It further shows that, regardless of the
type of residual feed, the cost of desulfurizing to very low levels such as
0.1%S is substantial, ranging from $3.51 to $5.84 per barrel or 26 to 43
percent more than the cost of untreated oil. This effect is more clearly shown
in Figure 4-1, which shows the cost of high, low, and medium sulfur content
residuals versus the sulfur level of the desulfurized oil. By way of compari-
son, Figure 4-2 gives the prices of residual fuel oil as quoted in the Oil and
Gas Journal , and it can be seen that these agree reasonably well with the cost
curves derived from this study.
Table 4-2 is a further sunnary of the data shown in Figure 4-1 and gives
a cost breakdown into the principal cost elements. This table vividly illus-
trates the effect of hydrogen and catalyst costs on the overall cost of desul-
furization which range from 33 to 61% of the total cost. The effect of these
and the other cost elements is further discussed in later sections of this
report.
Table 4-3 gives the cost impact of low sulfur fuel oil firing in indus-
trial boilers. Data are presented for small (4.4MW, 15,000 MBtu/Hr.) conner-
cial-type boilers and for large (44MW, 150,000 MBtu/Hr.) industrial-type
boilers. These sizes represent typical maximums for these type boilers. It
is assumed that the small boilers are fired with distillate oil (No. 2),
—111—
-------
Figure 41
COST OF DESULFURIZED RESIDUAL FUEL OIL
‘)I
-J
uJ
LU
a.
U)
-J
0
r .’)
£ .
20-
18-
‘
zz —
— -------
—C0
I
—.
______
D LAKE RESI!
—
—Kt
.
JAL
VAITRESIDUA.
14-
12-
-
—.
CE
TA RESIDUAL
in.
—
U
0.4
0.8
1.2 1.6 2.0
PER CENT SULFUR IN TREATED OIL
2.4
2.8
3.2
3.6
-------
Figure 4-2
COST OF RESIDUAL FUEL OIL (NO. 6)
OF VARIOUS SULFUR CONTENT
Source - Oil and Gas Journal May 29, 1979
SULFUR CONTENT - PER CENT BY WEIGHT
-J
LU
LU
U)
4:
- I
-J
0
ci
-J
0
-J
LU
U.
-J
4:
c i
U)
LU
U-
0
I-
U)
0
0
26
24
22
20
18
16
14
12
10
-113-
-------
TABLE 4-2.
COST DISTRIBUTION FOR THREE RESIDUA ( /bbl)
Percent Sulfur in Treated Oil
Residual Oil 1.6 0.8 0.3 0.1
Ceuta
Labor 6.9 6.9 6.9 6.9
Utilities 24.9. 49.7 62.0 65.7
Investment, Maint.,
& Waste Disposal 31.5 70.1 105.8 129.6
Hydrogen 21.0 67.0 102.0 117.0
Catalyst 9.0 38.0 121.0 216.0
Total 93.3 231.7 397.7 535.2
Sulfur Credit 2.0 5.0 6.9 7.6
Net Cost 91.3 226.7 390.8 527.6
Kuwait
Labor 6.9 6.9 6.9 6.9
Utilities 47.8 59.4 73.4 76.9
Invest., Maint.,
& Waste Disposal 64.4 91.1 119.6 141.8
Hydrogen 58.0 88.0 107.0 115.0
Catalyst 11.0 15.0 20.0 24.0
Total 188.1 260.4 326.9 364.6
Sulfur Credit 8.3 11.3 13.2 l3.
Net Cost 179.8 249.1 313.7 350.7
Cold Lake
Labor 6.9 6.9 6.9 6.9
Utilities 54.2 64.8 80.4 82.9
Invest., Maint.,
& Waste Disposal 97.1 116.0 139.5 156.9
Hydrogen 70.0 96.0 113.0 120.0
Catalyst 35.0 72.0 129.0 234.0
Total 263.2 355.7 468.8 600.7
Sulfur Credit 11.1 14.1 16.0 16.7
Net Cost 252.1 341.6 452.8 584.0
-114-
-------
whereas the larger industrial boilers are fired with residual oil (No. 6).
From Table 4-3, it is evident that the cost impact of providing low sulfur
distillate oil for firing small connercial boilers is minimal, amounting to
just a 6.7% premium for 0.3% S and 7.7% premium for 0.1% S oil. This small
effect is primarily brought about as a result of the small amount of desulfur-
ization required to desulfurize regular No. 2 distillate oil, which usually
contains 0.5% (or less) sulfur, to these lower sulfur levels.
The cost impact of using residual fuel oil is much more dramatic, ranging
from a premium of 6.7 to 18.6% when using oil desulfurized to a level of 1.6% S
up to a premium of 39 to 43.1% when using oil desulfurized to a level of
0.1% S.
Table 4-4 shows the cost effectiveness of fuel oil desulfurization for the
five residua considered, as well as for the distillate fuel oil. Generally, these
data indicate that the cost effectiveness improves as the sulfur content of
the residuum feed rises, provided, that the metal content does not increase as
well.
A comparison of the Kuwait and Khafji data shows the effect of similarity
between sulfur levels combined with relatively similar metal levels. The Cold
Lake data vividly show the strong effect of high metal levels.
The data of Table 4-4 also indicate that, for a given feedstock, fuel oil
desulfurization tends to be less cost effective as the degree of desulfurization
increases. This effect ranges from 17% to 65%, depending on the specific resi-
duum; but, the trend is quite general.
4.2 PROCESS DESCRIPTIONS
A modern hydrodesulfurization facility designed to meet EPA Clean Air
Requirements consists of four basic process elements, namely:
1. Hydrodesulfurization (HDS) Unit
2. Hydrogen (H 2 ) Plant
3. Sulfur (5) Plant
4. Sulfur Tail Gas Cleaning (SCOT) Plant
—115—
-------
TABLE 4-3.
COST IMPACT OF LOW SULFUR FUEL OIL FIRING IN BOILERS
Cost Impact
Sy stern
Standard Boilers
Type
Heat Input &
MW (MBTUIHRI yp
Level
of_C ntj’
Crude
Source
Annual Costs
44
(150,000) Watertube LSFO
Low
Sulfur
Re Si d
( 3% S)
Ceuta
E. Venezuelan
Medium
Sulfur
Re si d
(3.8% S)
Kuwait
High
Sulfur
Res I d
(> 4%S)
Khafjl
Cold Lake
4.4
(15,000)
N/A
Control
Efficiency
(%S)
$/kJ]S
$/MBTU/
HR
% Over
Uncon-
trolled
% Over
S.I.P.
Controlled
S.I.P.
Moderate
Intermediate
Stringent
1.6
0.8
0.3
0.1
4.03
10.03
17.23
23.27
1.18
2.94
5.05
6.82
6.72
16.82
28.86
39.04
N/A
9.47
20.75
30.22
S.I.P.
Moderate
IntermedIate
Stringent
1.6
0.8
0.3
0.1
5.15
10.78
17.30
25.15
1.51
3.16
5.07
7.37
8.63
18.08
29.00
42.14
N/A
8.70
18.75
30.85
S.I.P.
MOderate
IntermedIate
Stringent
1.6
0.8
0.3
0.1
7.92
10.95
13.82
15.46
2.32
3.21
4.05
4.53
13.28
18.38
23.17
25.90
N/A
4.50
8.73
11.14
S.I.P.
Moderate
Intermediate
Stringent
1.6
0.8
0.3
0.1
9.69
12.56
15.87
18.12
2.84
3.68
4.65
5.31
16.24
21.03
26.57
30.33
N/A
4.12
8.89
12.12
S.I.P.
Moderate
Intermediate
Stringent
1.6
0.8
0.3
0.1
11.09
15.08
19.96
25.73
3.25
4.42
5.85
7.54
18.60
25.24
33.43
43.10
N/A
5.60
12.50
20.66
S.I.P.
Moderate
Intermediate
Stringent
0.3
0.1
N/A
N/A
5.46
6.21
N/A
N/A
1.60
1.82
N/A
N/A
6.73
7.67
N/A
N/A
N/A
0.88
Firetube LSFO
Distillate
Fuel
Oil
-------
TABLE 4-4. COST EFFECTIVENESS
Crude
Source ___________ $/lb $/KG
1.6 0.52 1.14
0.8 0.51 1.13
0.3 0.64 1.40
0.1 0.78 1.71
1.6 0.45 0.99
0.8 0.46 1.01
0.3 0.56 1.23
0.1 0.74 1.63
1.6 0.24 0.53
0.8 0.25 0.55
0.3 0.27 0.59
0.1 0.28 0.62
1.6 0.24 0.53
0.8 0.24 0.53
0.3 0.26 0.57
0.1 0.29 0.64
1.6 0.25 0.55
0.8 0.27 0.59
0.3 0.32 0.70
0.1 0.39 0.86
0.2 0.20 0.44
0.1 0.21 0.46
Sulfur
In Fuel Oil
0/
/0
Cost/Unit Removal
Ceuta
E. Venezuelan
Kuwait
Khafji
Cold Lake
Distillate
—117—
-------
Figure 4-3 shows a typical HDS unit ‘for the desulfurization of residual
oil. Residual oil feed is heated together with make-up hydrogen and recycle
gas, and the mixture charged to the reactor section. The reactor section con-
sists of one or more reactors in series,with the number being dependent upon
the degree of desulfurization required. When treating feeds containing high
levels of metals (Ni + V), a guard reactor is used to minimize the contamina-
tion of the more valuable catalyst used in the desulfurization reactors. Hy-
drogen-rich gas is flashed from the reactor effluent in a high-pressure separ-
ator and is purified by amine scrubbing prior to recycling to the reactor
section. Liquid from the high pressure separator passes through a low-pressure
separator to remove H 2 S and fuel gas and then goes to the fractionator for
separation of naphtha, middle distillate, and low sulfur fuel oil.
Figure 4-4 shows a typical hydrogen plant for the production of hydrogen
by the steam reforming of natural gas, LPG, or naphtha. Hydrocarbon feed is
first desulfurized to prevent poisoning of the reforming catalyst. The desul-
furized feed is mixed with superheated steam and reformed by passing through
catalyst-filled tubes in the reformer furnace. The reformed gas containing
hydrogen, carbon monoxide, carbon dioxide, and steam is cooled and then passed
through a shift converter where the carbon monoxide is reacted with steam to
produce carbon dioxide and hydrogen. The C0 2 -rich gas is then scrubbed with
amine to remove essentially all the carbon dioxide. The remaining traces of
carbon monoxide and carbon dioxide are removed by passing the gas through the
methanator, wherein the CO 2 and CO are reacted with hydrogen to form methane.
The methanator effluent gas typically contains 95-98% H 2 .
Figure 4-5 shows a typical Claus-type sulfur recovery plant. Hydrogen
sulfide gas is fed to a modified fire-tube boiler where it is partially burned
with air to form sulfur dioxide. The amount of air is controlled to limit the
coii ustion to one-third of the H 2 S fed. Effluent gas from the boiler consist-
ing of 2/3 H 2 S and 1/3 SO 2 is passed through a primary converter wherein H 2 S
-118-
-------
FIGURE 4—3
TYPICAL HDS UNIT
1
FEED
HEATER
> (
RECYCLE
ESSOR
AMINE
SYSTEM
HYDRODESULFURIZATION, RESIDUAL OIL
I
LPG. OFF-GAS
RECYCLE
PUMP
FE ED
FILTER
START
—a
FE ED
PUMP
BOOSTER
PUMP
FEED/E FF.
EXCHANGER
COOLER
-------
Figure 4-4
TYPICAL HYDROGEN PLANT
REFORMER
SHIFT CONVERTER
DESULFURIZER
Start
H
Heat
Recovery
Steam
CO 2 ABSORBER
SOLUTION REGENERATOR
METHANATO R
-------
H 2 S Gas from
Amine Regenerator
and Sour Water
Stripper
Figure 4-5
TYPICAL PACKAGED CLAUS PLANT (2 STAGE)
Secondary
Converter
Steam
Waste Heat
Burner
Air
Tail Gas to
Incinerator
or Tail Gas
Processing
Boiler Feed Water
Steam
Liquid Sulfur Product
E ±E
Sulfur Tank and Sump Pump
-------
is reacted with SO 2 to form sulfur vapor and steam. The converter effluent
passes through a primary condenser wherein the sulfur vapor is condensed out
of the gas stream while steam is generated on the other side of the condenser.
The gas stream then passes through a secondary converter wherein the bulk of
the remaining H 2 S and SO 2 is reacted to form additional sulfur. The converter
effluent passes through a second condenser wherein the sulfur is condensed and
sent to the sulfur storage tank along with the sulfur from the first condenser.
Tail gas from the separator is sent to an incinerator for burning or to a Tail
Gas Unit for further processing.
Figure 4-6 shows a flow diagram for the Shell Claus Off-Gas Treating Pro-
cess (SCOT) which is used to treat the tail gas from the Claus sulfur plant and
to increase the sulfur recovery efficiency from the 95% obtained in the Claus
unit to more than 99.8% overall.
The process essentially consists of a reduction section wherein the SO 2 ,
free sulfur, and other sulfur compounds are reduced to H 2 S by reaction with
hydrogen over a catalyst. The reactor effluent is cooled by indirect exchange
to produce low pressure steam and is further cooled by direct contact with water
in a packed or tray column. Water vapor is condensed from the process gas and
the condensate sent to a sour water stripper. The cooled overhead gas contain-
ing up to 3% U 2 S and 40% CO 2 is sent to an amine absorber wherein essentially
all the H 2 S and little of the CO 2 are absorbed by an ADIP solution which, de-
pending on conditions, uses a secondary or a tertiary amine. The recovered
H 2 S is stripped from the amine solution and recycled to the Claus unit whereas
the treated gas from the absorption column is vented to the air or is burned
in a standard incinerator.
The foregoing process descriptions are meant to be typical rather than
specific and do not imply a preference to any proprietary process. There are
a number of suppliers for each of the foregoing processes; and, while details
of the processes may vary, the overall results for the various processes are
-122—
-------
Figure 4—6
Flow Diagram for the Shell Claus Off-Gas (Scot) Treating Process
Claus plant tail gas
prior to incinerator
Fuel gas
Air
Fat amine to regenerator
Cooling Tower
Packed or Tray
Reactor
Reducing
Gas (H 2 )
Line Heater
Sour gas to Claus unit
Lean amine from regenerator
Tray Tower Absorber
Sour-water
existing sour-
water stripper
-------
the same. For that reason, the processes must be considered equivalent.
4.3 COST BASIS -
Table 4-5 gives HDS plant details for various hydrodesulfurization pro-
cesses as reported in the iiterature.(1)(2) 3) : 4 ) Table 4-6 lists selected
data from Table 4-5 for the five residual fuel oils and four levels of desul-
furization used in this study. The plant investment data shown in Table 4-6
have been updated to a June 1978 base by escalating the costs given in
Table 4-5 by multiplying by the factor derived from the Chemical Engineering
Plant Cost Index given in Table 4-7.
It is seen from Table 4-5 that the units listed range in capacity from
30,000 B/D to 78,000 BID, with the majority having a capacity of 50,000 B/D.
The size of HDS units may range from several thousand barrels per day to one
hundred thousand or more barrels per day; however, the vast majority of new
installations will fall within the 25,000 B/D to 75,000 B/D range. The maxi-
mum capacity that can be built into a single-train unit is approximately
25-30,000 BiD; hence, we see that HOS plant capacities tend to cluster around
25,000 B/D, 50,000 BID, and 75,000 B/D capacities - representing one-, two-,
and three-train plants, respectively. Due to this limit for single-train
capacity, one does not realize much effect on economy of scale in going from
25,000 B/D capacity to 50,000 BID or even 75,000 BID. This fact, coupled with
the fact that most of the data given in Table 4-5 are for 50,000 BID capacity,
has led us to select 50,000 B/D capacity as representative of current refinery
practice.
Table 4-5 further illustrates that single-stage systems are used to re-
duce sulfur levels to about 0.8% (80% removal), two-stage systems to about
0.3% (90% removal), and three-stage or two-stage with guard reactor to about
0.1%. HDS plant cost versus percent sulfur in the treated oil (or percent re-
moval) is assumed to be a continuous function rather than a step function, as
described above, and as illustrated by Figure 4-7, which shows the costs of
-124-
-------
TABLE 4-5
H DS P LANT D ETA ILS
FOR
HYDRODESULFURIZATION OF RESIDUAL FUEL OILS
Reported 1978
Capacity Feed Product Removal Cost Cost
Licensor Type B/CD Residuum Fed %S %S %* MM $ Year MM $
GULF II 50,000 Cold Lake ATB 4.55 1.14 74.9 43.5 1975 52.2
GULF II 50,000 Khafji ATB 4.36 1.09 75.0 30.3 1975 36.4
UOP RCD ISOMAX 40,000 Kuwait ATB 3.8 1.0 73.7 17.1 1972 27.3
GULF II 30,000 Kuwait VTB 5.5 1.0 81.8 34.7 1975 41.6
II 50,000 Kuwait VTB 5.5 1.0 81.8 32.7 1974 43.3
II 50,000 Kuwait ATB 3.8 1.0 73.7 30.6 1975 36.4
ii 50,000 Kuwait ATB 3.8 1.0 73.7 28.8 1974 38.1
50,000 Khafji ATB 4.4 1.0 77.3 34.3 1974 45.4
UOP RCD Isomax 40,000 Tiajuana ATB 2.61 0.9 65.5 20.1 1972 32.1
GULF II 50,000 Iranian ATB 2.47 0.62 74.9 34.0 1975 40.8
GULF II 50,000 E. Venezuela ATB 2.38 0.60 74.8 38.4 1975 46.1
GULF II 50,000 Ceuta ATB 2.12 0.53 75.0 36.0 1975 43.2
VGO/VRDS 78,000 Arab Heavy ATB 4.4 0.5 88.6 34.8 1973 52.8
RDS 78.000 Arab Heavy ATB 4.4 0.5 88.6 38.9 1973 59.1
UOP RCD ISOMAX 40,000 Kuwait ATB 3.8 0.5 86.8 23.0 1972 36.7
UOP RCD ISOMAX 40,000 Kuwait ATB 3.8 0.32 91.6 27.7 1972 44.2
GULF II I 50,000 Kuwait ATB 3.8 0.3 92.1 43.5 1975 52.2
50,000 Iranian ATB 2.47 0.3 87.9 43.5 1975 52.2
III 50,000 Ceuta ATB 2.12 0.3 85.8 44.9 1975 53.9
iii 50,000 E. Venezeula ATB 2.38 0.3 87.4 48.3 1975 58.0
III 50,000 Khafji ATB 4.36 0.3 93.1 51.3 1975 61.6
III 50,000 Kuwait ATB 3.8 0.3 92.1 45.6 1974 60.3
III 50,000 Cold Lake ATB 4.55 0.3 93.4 58.8 1975 70.6
GULF IV 50,000 Kuwait ATB 3.8 0.1 97.4 56.3 1974 74.5
-------
TABLE 4-6. HDS PLANT INVESNENT
Based on HDS of 50,000 BID of Residual Oil Costs Updated to June 1978
Residual Oil
Percent Sulfur in Treated Oil
1.6
0.8
0.3
0.1
Cost
in Millions
of Dollars
Ceuta
2.12%
S
15.0
34.0
54.0
68.0
Venezuelan
2.38%
S
24.0
40.0
58.0
72.0
Kuwait
3.80%
S
28.5
42.5
60.0
74.5
Khafji
4.36%
S
40.0
48.0
62.0
77.0
Cold Lake
4.55%
S
48.5
57.5
70.5
81.0
-126-
-------
TABLE 4-7. ECONOMIC INDICATORS
CE PLANT COST INDEX
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
CE Index
125.5
132.2
137.2
144. 1
165.4
182.4
192.1
204.1
218.8
Multiplier to
Get 1978 Cost
1.743
1 .655
1.595
1 .518
1 .323
1 .200
1.139
1 .072
1 .000
M & S EQUIPMENT COST INDEX
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
M & S Index
303.3
321.3
332.0
344.1
398.4
444.3
472.1
505.4
545.3
Multiplier to
Get 1978 Cost
1.798
1 .697
1 .642
1.585
1.369
1 .228
1.155
1 .079
1.000
Source : Chemical Engineering, 86 p. 7, 7 May 1979
-127—
-------
Figure 4-7 Based on Hydrodesulfurization of
50,000 BiD of Kuwait J esidual
HDS PLANT COST AS FUNCTION OF SULFUR REMOVAL Fuel Oil Containing 3.8% Sulfur
0.5 0.6
PER CENT SULFUR IN TREATED OIL
90
80
70
60
50
40
30
20
10
0
U,
-J
0
0
U..
0
U)
z
0
-J
-J
S
I- .
-j
a-
U,
0
I
-128-
-------
one-, two-, and three—stage HDS units, as represented by the Gulf II, Gulf III,
and Gulf IV processes, respectively.
The selection of these data represents the most widely-reported process
in the literature surveyed. The selection of these data is not intended to
infer a recommendation or approval of the specific process, and it should be
assumed that all similar HDS processes will perform in an equivalent manner.
HDS Unit (Distillate)
Table 4-8 shows similar HDS plant details for the desulfurization of dis-
tillate fuel The Gulf process has again been selected for illustration
because of the completeness of the data provided. The investment cost shown
has been updated to the June 1978 base.
Hydrogen Plant
Investment costs for the production of hydrogen by the steam reforming of
natural gas have been taken from Catalytic’s estimating files and are present-
ed in Figure 4—8. These costs have been updated to a June 1978 base and can
be expressed as a power relationship by the equation:
where
C = installed cost of plant in millions of dollars
x = production capacity of plant in millions of cubic feet
of hydrogen per day
a & b = constants equal to 39.147 and 0.735, respectively
hence Cost ( F l 2 plant) = 39.147 (capacity)° 735
in io 6 dollars in 106 SCF/D
Sulfur Plant
Investment cost of Claus-type sulfur recovery plants is given in Figure 4-9
and is taken fromthe paper “Capital and Operating •Costs for 54 Chemical Pro-
cesses” by K. M. Guthrie of Fluor Corporation as published in the June 15, 1970,
issue of Chemical Engineering magazine.’ 6) These data have been updated to
-129—
-------
TABLE 4-8. DISTILLATE DESULFURIZATION DETAILS
Sulfur Removed
Hydrogen Catalyst Utilities Investment
Feed Product 0.3% 0.1%
Usage Cost Usage Cost Capacity Sulfur Sulfur L Ton/ L Toni
Process Scf/BBL /BBL lb/BBL /BBL JBBL BID MM $ % % Day Day
Gulflning 350 56.0 0.01 1.0 10.3 35,000 18.9 2.2 0.2 93.3 103.1
Basis for Calculation of Sulfur Removed:
35,000 B/D distillate fuel oil of 26°API gravity = 0.898 SPGR
Pounds per day of distillate oil = 0.898 x 8.33 x 42 x 35,000
= 11 x io6 LBS/DAY
Tons per day of sulfur removed = 11 x 106 (%S . - ‘ ( 1
100
= 49.09 (%s - %S
in out)
-------
U,
20
1
.1
0
0
U-
0
U)
z
010
-J
-J
36
I-
Z5
0
Figure 4-8
HYDROGEN PLANT COST
(Updated to June 1978)
10
90
80
70
60
50
40
30
Cost Capacity 0.735
Where
Cost = Millions of Dollars
Capacity = Millions Cubic Ft. Per Day
-
c
-a
—
.,
,
.
/
———----
/
/
/
::::_
4
3
2
.2 .3 .4 .5 .6 .7 .8 .9 1
4
5 6 7 8910
20
30 40 50 60 708090100
HYDROGEN PLANT CAPACITY- 108 SCF/D
-------
Fig. 4.0
INVESTMENT COST OF CLAUS SULFUR RECOVERY PLANTS
Source- Chem. engr. P151, June 1970
Costs Updated to June 1978
C ,,
-J
-j
0
U.
0
U,
z
0
-J
-J
S
z
0-
.1
PLANT CAPACITY - TONS PER DAY
-------
the June 1978 base and can also be represented by a power relationship of the
form:
Cost = a (Capacity)b
where
Cost (io6 dollars) = 0.1877 x 106 (Capacity T/D)° 656
This plant cost correlation differs substantially from the cost data given
for sulfur recovery plants in the EPA report on proposed performance standards
for refinery sulfur recovery piantsJ 7 Data from that report give a value of
0.455 for the exponent “b” for the capacity range from 10 to 100 tons per day.
The cost relationship developed here is more in agreement with the experience
of the senior author, who for years used a power relationship for sulfur plant
costs where a value of 0.65 was used for the exponent. The proposed cost cor-
relation is further substantiated by data from Ne1son 8 which give a value
of 0.544 for the exponent for capacities ranging from 10 to 100 tons per day
and a value of 0.7 for capacities of 100 to 1000 tons per day.
Tail-Gas Treatment Plant
Investment cost of a SCOT tail-gas treatment plant will normally range
from 70% to 100% of the cost of the Claus sulfur plant depending upon whether
it is integrated and constructed with the Claus plant or is added on at some
later dateJ 9 For purposes of this study, we have assumed the Claus unit and
tail-gas unit would be integral parts of the complete HDS installation and
have used a factor of 70% of the sulfur plant cost for the investment cost of
the tail-gas unit.
Table 4-9 gives a sumary of complete HDS investment costs broken down to
show the costs of the HDS unit, hydrogen unit, sulfur unit, and tail-gas
treating unit. Costs are summarized for the five types of residual oil and
four levels of sulfur control. The tabulated costs are order-of-magnitude
costs and have an accuracy of + 25%.
-133-
-------
TABLE 4-9. HDS PLANT INVESTMENT
Residual Oil Percent Sulfur in Treated Oil
1.6 0.8 0.3 0.1
Ceuta
HDS Unit 15.0 34.0 54.0 68.0
Hydrogen Unit 3.9 9.4 12.8 14.1
Sulfur Unit 2.1 3.8 4.6 5.0
SCOT Unit 1.4 2.6 3.2 3.5
Total Investment 22.4 49.8 74.6 90.6
East Venezuelan
HDS Unit 24.0 40.0 58.0 72.0
Hydrogen Unit 5.1 9.9 13.0 14.3
Sulfur Unit 2.7 4.2 5.1 5.4
SCOT Unit 1.9 2.9 3.6 3.8
Total Investment 33.7 57.0 79.7 95.5
Kuwa I t
HDS Unit 28.5 42.5 60.0 74.5
Hydrogen Unit 8.4 11.5 13.2 14.0
Sulfur Unit 5.3 6.4 7.1 7.4
SCOT Unit 3.7 4.5 5.0 5.2
Total Investment 45.9 64.9 85.3 101.1
Khafji
HDS Unit 40.0 48.0 62.0 77.0
Hydrogen Unit 9.3 11.8 13.3 14.1
Sulfur Unit 6.1 7.2 7.8 8.1
SCOT Unit 4.3 5.0 5.5 5.7
Total Investment 59.7 72.0 88.6 104.9
Cold Lake
HDS Unit 48.5 57.5 70.5 81.0
Hydrogen Unit 9.6 12.2 13.7 14.4
Sulfur Unit 6.4 7.4 8.1 8.3
SCOT Unit 4.5 5.2 5.7 5.8
Total Investment 69.0 82.3 98.0 109.5
NOTE : Plant investments in millions of dollars
-134-
-------
Investment Charges
The capital investments presented in Table 4-9 are translated into annual
capital charges following the method outlined in Section 5 of the PEDCo re-
portJ’° Based on a useful life of 15 years (which is typical for chemical
process units) for the HDS, hydrogen, sulfur, and tail—gas units and an annual
interest rate of 10% over the life of the facilities gives a capital recovery
factor of approximately 13%. If we add to this a total of 4% of depreciable
investment to cover general and administrative costs, taxes, and insurance,
we arrive at an annual capital charge of 17% of depreciable investment. This
value has been used as the investment charge in this report.
Mai ntenance Charges
Maintenance charges associated with the upkeep of the physical plant
facilities are translated into annual charges by using a factor of 5% of the
capital investment costs given in Table 4-9. This factor was also taken from
the PEDC0 report and covers both maintenance, labor, and materials.
Utility, Labor, and Supervisory Costs
Table 4-10 lists recomended annual unit costs for operating labor and
supervision, and various utilities such as cooling water, electricity, fuel,
etc. This table has also been taken from the PEDCo report. The PEOCo cost
data have been used to provide some degree of consistency so that the cost
results obtained in this study can be compared with similar results given in
the PEDCo report for other technologies.
Labor
Table 4-11 gives the manning requirements and costs for operating labor
and supervision for all the process units.
Hydrogen
Hydrogen consumption for desulfurization of residual fuel oil is calculated
from published data and is shown in Figure 4-10 as a function of percent
-135-
-------
TABLE 4-10. ANNUAL UNIT COSTS FOR OPERATION AND MAINTENANCE
(June 1978 Dollars)
Cost Factors
Direct labor, $/man-hour
Supervision, $/man-hour
Maintenance, labor, $/man-hour
Electricity, mills/kWh
Untreated water, gal
Process water, gal
Cooling water, $/1O 3 gal
Boiler feed water, $/1O 3 gal
No. 2 fuel oil, Btu
No. 6 fuel oil, $1106 Btu
Natural gas, $/i0 6 Btu
Recommended
Value
12 02 a
15 • 63 b
14. 63 a
25.8 c
015 d
o•
1 .oo
2.21 g
195 h
a Engineering News-Record, June 29, 1978, pp. 52-52. Average
for Chicago, Cincinnati, Cleveland, Detroit, and St. Louis.
b Estimated at 30 percent over direct labor rate.
c EEl members publication for June 1978. Average for Boston,
Chicago, Indianapolis, Houston, San Francisco, and Los
Angeles.
d Peters, M. S., and K. 0. Tiimierhaus, Plant Design and Econo-
mics for Chemical Engineers , 2nd Edition, McGraw-Hill Book
Co., New York, 1968, P. 772. Adjusted to 1978 prices using
Nelson Refinery Operating Cost Indexes for Chemicals, July
1978.
e . .
Perry, J. H., et al., Chemical Engineer s Handbook, McGraw-
Hill Book Co., New York, 1963, pp. 26-29.
Nelson, W. L., Guide to Refinery Operating Costs, The Petro-
leum Publishing Company, 1966, p. 27.
g Electrical Week, May issues, 1978. Spot market prices.
h Gas Facts, 1977, American Gas Association. Average U. S.
price.
-136—
-------
Figure 4-10
CHEMICAL HYDROGEN CONSUMPTION
IN DESULFURIZATION OF 16°API RESIDUAL
Source - Oil & Gas Journal P126 Feb. 28, 1977
SULFUR REDUCTION—%
800
700
600
500
400
300
200
100
-J
U-
0
U)
z
0
F-
0
U,
z
0
0
z
w
0
0
>-
I
0
0 20 40 60 80 100
120
—137-
-------
TABLE 4-11. LABOR AND SUPERVISORY COSTS FOR
HYDRODESULFURIZATION OF RESIDUAL OIL
Shift Shift
Process Unit Operators Supervision
HDS Unit 4 1/2
Hydrogen Plant 1 1/4
Sulfur Plant 1 1/4
SCOT Unit md. w/Sulfur Plant
Total Per Shift 6 1
Total 4 Shifts 24 4
Unit Costs Per Manhour $12.02 $15.63
Annual Costs
(Based on 2080 MH/Yr Per Man) $600,038 $130,042
Annualized Costs /BBL 3.6 0.8
-138—
-------
sulfur reduction. Because of its pronounced effect on the overall cost of
hydrodesulfurization, the cost of hydrogen has not been included in the cost
of raw materials or utilities,but has been computed as a separate operating
cost. Table 4-12 gives the raw material and utility requirements for produc-
ing hydrogen and derives a value to be used to determine the cost of hydrogen.
The unit costs used in Table 4-12 are taken from Table 4-10. The effect of
hydrogen cost on overall hydrodesulfurization costs is discussed in further
detail under the sections on cost and sensitivity analysis.
Catalyst
Catalyst consumption for desulfurization of residual fuel oil is calcu-
lated from published data) 2 The cost of catalyst consumed also has a major
impact on the overall cost of hydrodesulfurization and is also computed as a
separate operating cost. The effect of catalyst cost on overall hydrodesulfur-
ization costs is discussed in further detail in later sections of this report.
It should be pointed out that catalyst cost as used herein refers only to the
consumption or usage of catalyst and is thereby considered as an operating
cost. The cost of the initial charge of catalyst is considered a capital cost
and has been included in the HDS unit investment cost.
Waste Treatment
Waste treatment facilities are assumed to represent about 5% of the HDS
unit cost. 3 This is limited to water treatment and does not include provi-
sion for catalyst disposal. It is estimated that a similar relationship would
exist for the other process units and has so been used in this study.
Catalyst Disposal
For purposes of this study, it has been assumed that spent catalyst will
be disposed to landfill with no provision for recovery of any metal values.
Because of the toxic nature of the metals absorbed on the catalyst, it is
assumed that the catalyst would be sent to a hazardous waste landfill. Quota-
tions received from one company engaged in such hazardous waste disposal gave
-139-
-------
TABLE 4-12. HYDROGEN PRODUCTION COSTS
Item
Process Feed (Natural Gas)
Fuel (Natural Gas)
Cooling Water
Power
Boiler Feedwater
TOTAL
Reqmt’s. Per
lO SCF H 2
260 SCF
415 x BTU
1450 Gal.
1.5 KWH
5 Gal.
Unit Cost
$l.95/lO SCF
$l.95/106 BTU
$0.18/l0 Gal.
25.8 Mils/KWH
$l.00/lO Gal.
CQst Per
iO’ SCF H 2
$0.51
0.78
0.26
0.04
0.01
$1.60
NOTE : Based on steam reforming of natural gas with one stage of CO
conversion to produce 95-98% H 2 . CO 2 removed by amine scrubbing.
-140-
-------
a cost range of $30 to $50 per ton of catalyst disposed. This cost includes
the cost of transportation of the catalyst to the landfill and the cost of the
landfill facility. For purposes of this study, a median cost of $40 per ton
was used. It should be pointed out that recent increases in the cost of metals
used in desulfurization have made the recovery of these metals attractive in
some cases, and several companies are now engaged in this new business. If
this trend continues, it is possible that catalyst disposal may change from an
item of cost to an item of return.
Sulfur Recovery
It is assumed that a steady market exists for all sulfur produced and that
the sulfur can be sold for a net price of $25 per long ton.
4.4 SENSITIVITY ANALYSIS
Tables 4-13 through 4-17 give detailed cost breakdowns for the hydrodesul-
furization of five residual fuel oils for four levels of sulfur in the desul—
furized oils. Table 4-18 gives a similar breakdown for the desulfurization of
distillate fuel oil to two levels of sulfur in the treated oil.
A most striking statistic to be gathered from these tables is that of the
ten cost elements listed; five of these have a very minor effect on the overall
cost of desulfurization. These five items - waste treatment, catalyst disposal,
labor, supervision, and overhead - exert a cumulative effect of less than ten
percent (10%) on the overall cost and, in fact, ranges to as low as three per-
cent (3%) in some cases. For this reason, we will concentrate our analysis
and discussion on the other five cost elements; namely, investment charges,
maintenance charges, utilities costs, hydrogen cost, and catalyst costs.
Investment Charges
As might be expected, investment charges have a major impact on the overall
cost of desulfurization ranging from 17% to nearly 29% of the overall costs.
This is easily understood when one realizes that a complete 50,000 BID hydro-
desulfurization unit consists not only of the NDS unit but also of a hydrogen
-141-
-------
TABLE 4-13. TOTAL COST OF HYDRODESULFURIZI\TION OF RESIDUAL FUEL OIL
Based on 50,000 B/SD Residual Oil Feed
Residual Type: Ceuta
Containing 2.12% S and 292 ppm (Ni + V)
Cost Per Barrel of Treated Oil for
Cost Item Following % S in Treated Oil
1.6% 0.8% 0.3% 0.1%
Investment Charge 23.1 t/BBL 51.3 /BBL 75 .9 /BBL 93.3 /BBL
Maintenance Charge 6.8 15.1 22.6 27.5
Utilities Cost 24.9 49.7 62.0 65.7
Hydrogen Cost 21.0 67.0 102.0 117.0
Catalyst Cost 9.0 38.0 121.0 216.0
Waste Treatment 1.5 3.3 5.0 6.0
Catalyst Disposal 0.1 0.4 1.3 2.8
Labor 3.6 3.6 3.6 3.6
Supervision 0.8 0.8 0.8 0.8
Overhead 2.5 2.5 2.5 2.5
TOTAL 93.3 231.7 397.7 535.2
Sulfur Credit 2.0 5.0 6.9 7.6
NET COST 91.3 226.7 390.8 527.6
($0.91) $$2.28) ($3.91) ($5.28)
NOTES : Investment charge at 17% of total plant investment.
Maintenance charge at 5% of total plant investment.
Waste treatment at 1.1% of total plant investment.
Overhead at 56% of labor plus supervision cost.
Sulfur credit at $25 per ton of sulfur recovered.
-142-
-------
TABLE 4-14. TOTAL COST OF 1-IYDRODESULFURIZATION OF RESIDUAL FUEL OIL
Cost Item
Investment Charge
Maintenance Charge
Utilities Cost
Hydrogen Cost
Catalyst Cost
Waste Treatment
Catalyst Disposal
Labor
Supervision
Overhead
TOTAL
Sulfur Credit
NET COST
0.1%
98.4 /BBL
28.9
65.7
118.0
252.0
6.4
2.8
3.6
0.8
2.5
579.1
8.6
570.5
($5.71)
Based on 50,000 B/Si Residual Oil Feed
Residual Type: East Venezuelan
Containing 2.38% S and 274 ppm (Ni + V)
Cost Per Barrel of Treated Oil for
_________ Following_%_S_in_Treated_Oil
1.6% 0.8% 0.3% ____
34.7 /BBL 58.7 /BBL 82.1 /BBL
10.2 17.3 24.2
24.9 49.7 62.0
29.0 73.0 104.0
12.0 41.0 115.0
2.2 3.8 5.3
0.1 0.4 1.3
3.6 3.6 3.6
0.8 0.8 0.8
2.5 2.5 2.5 _____
120.0 250.8 400.8
3.0 6.0 7.8
117.0 244.8. 393.0
($1.17) ($2.45) ($3.93)
NOTES : Investment charge at 17% of total plant investment.
Maintenance charge at 5% of total plant investment.
Waste treatment at 1.1% of total plant investment.
Overhead at 56% of labor plus supervision cost.
Sulfur credit at $25 per ton of sulfur recovered.
-143-
-------
TABLE 4-15. TOTAL COST OF HYDRODESULFURIZATION OF RESIDUAL FUEL OIL
Based on 50,000 B/SD Residual Oil Feed
Residual Type: Kuwait
Containing 3.8% S and 60 ppm (Ni + V)
Cost Per Barrel of Treated Oil for
Cost Item Followinç % S in Treated Oil.
1.6% 0.8% 0.3% 0.1%
Investment Charge 47.34/BBL 66.9 t/BBL 87.9 UBBL 104.2 /BBL
Maintenance Charge 13.9 19.7 25.8 30.6
Utilities Cost 47.8 59.4 73.4 76.9
Hydrogen Cost 580 88.0 107.0 115.0
Catalyst Cost 11.0 15.0 20.0 24.0
Waste Treatment 3.1 4.3 5.7 6.7
Catalyst Disposal 0.1 0.2 0.2 0.3
Labor 3.6 3.6 3.6 3.6
Supervision 0.8 0.8 0.8 0.8
Overhead 2.5 2.5 2.5 2.5
TOTAL 188.1 260.4 326.9 364.6
Sulfur Credit 8.3 11.3 13.2 13.9
NET COST 179.8 249.1 313.7 350.7
($1.80) ($2.49) ($3.14) ($3.51)
NOTES : Investment •;harge at 17% of total plant investment.
Maintenance charge at 5% of total plant investment.
Waste treatment at 1.1% of total plant investment.
Overhead at 56% of labor plus supervision cost.
Sulfur credit at $25 per ton of sulfur recovered.
-144-
-------
TABLE 4—16. TOTAL COST OF HYDRODESULFURIZATION OF RESIDUAL FUEL OIL
Cost Item
Investment Charge
Mai ntenance Charge
Utilities Cost
Hydrogen Cost
Catalyst Cost
Waste Treatment
Catalyst Disposal
Labor
Supervision
Overhead
TOTAL
Sulfur Credit
NET COST
0.1%
108.1 /BBL
31.8
82.9
117.0
72.0
7.0
0.8
3.6
0.8
2.5
426.5
16.0
410.5
($4.11)
Based on 50,000 B/SD Residual Oil Feed
Residual Type: Khafji
Containing 4.36% S and 118 ppm (Ni + V)
Cost Per Barrel of Treated Oil for
_________ Following % S in Treated Oil
1.6% 0.8% 0.3% ____
6l.5 /BBL 74.2i /BBL 91.3 /BBL
13.1 21.8 26.8
54.2 64.8 80.4
66.0 92.0 109.0
19.0 33.0 54.0
4.0 4.8 5.9
0.2 0.4 0.6
3.6 3.6 3.6
0.8 0.8 0.8
2.5 2.5 2.5 _____
229.9 297.9 374.9
10.4 13.4 15.3
219.5 284.5 359.6
($2.20) ($2.85) ($3.60)
NOTES : Investment charge at 17% of total plant investment.
Maintenance charge at 5% of total plant investmcnt.
Waste treatment at 1.1% of total plant investment.
Overhead at 56% of labor plus supervision cost.
Sulfur credit at $25 per ton of sulfur recovered.
-145—
-------
TABLE 4-17. TOTAL COST OF HYDRODESULFURIZATION OF RESIDUAL FUEL OIL
Cost Item
Investment Charge
Maintenance Charge
Utilities Cost
Hydrogen Cost
Catalyst Cost
Waste Treatment
Catalyst Disposal
Labor
Supervision
Overhead
TOTAL
Sulfur Credit
NET COST
1 .6%
71.1 /BBL
21.0
54.2
70.0
35.0
4.6
0.4
0.1%
112.8 /BBL
34.2
82.9
120.0
234.0
7.3
2.6
3.6
0.8
2.5
600.7
16.7
584.0
($5.84)
Based on 50,000 B/SD Residual Oil Feed
Residual Type: Cold Lake
Containing 4.55% S and 236 ppm (Ni + V)
Cost Per Barrel of Treated Oil for
Following % S in Treated Oil
____ 0.8% 0.3% ____
84.8 4/BBL 101.1 /BBL
24.9 30.6
64.8 80.4
96.0 113.0
72.0 l29.0
5.5 6.5
0.8 1.4
3.6 3.6 3.6
0.8 0.8 0.8
2.5 2.5 2.5
263.2 355.7 468.8
11.1 14.1 16.0
252.1 341.6 452.8
($2.52) ($3.42) ($4.53)
NOTES : Investment charge at 17% of total plant investment.
Maintenance charge at 5% of total plant investment.
Waste treatment at 1.1% of total plant investment.
Overhead at 56% of labor plus supervision cost.
Sulfur credit at $25 per ton of sulfur recovered.
-146-
-------
TABLE 4-18. COST OF HYDRODESULFURIZATION OF DISTILLATE FUEL OIL
(Basis: 35,000 B/D)
Fuel Oil Sulfur
Item 0.3% 0.1%
/BBL /BBL
Hydrogen 47.7 58.8
Catalyst 1.0 1.0
Utilities 9.8 10.8
Investment Charge 44.5 48.7
Maintenance Charge 13.1 14.3
Catalyst Disposal 0.1 0.1
W. Water Treatment 2.9 3.1
Labor 5.1 5.1
Supervision 1.2 1.2
Overhead (56%) 3.5 3.5
Total 128.9 146.6
Less Sulfur Credit 6.7 7.4
Net Cost 122.2 139.2
($1.22) ($1.39)
Total Plant Investment
HDS Unit
Hydrogen Plant
Sulfur Plant
SCOT Unit __________
TOTAL
$
18 x io6
$19.8 x io6
6xl0 6
6.5x10 6
3.7 x 106
4.0 io6
2.6 x
$30.3 x io6
2.8 x io6
$33.1 x
-147—
-------
plant with capacities ranging from six million cubic feet per day (6 x 106 SCFD)
to nearly thirty-eight million cubic feet per day (38 x 106 SCFD) and a sulfur
plant and tail gas unit with capacities ranging from nearly forty tons per day
(40 l/D) to more than three hundred thirty tons per day (330 T/D). These are
large and expensive plants.
Another interesting fact to be noted from these tables is that the invest-
ment charges increase quite rapidly with the degree of desulfurization. This
is brought about by the fact that going to higher degrees of desulfurization,
or lower levels of sulfur in the treated oil, requires more and more stages of
desulfurization. For example, one stage of desulfurization is usually suffi-
cient to obtain an oil of 0.8% sulfur, whereas two stages are required to get
0.3% sulfur and three stages are required for 0.1% sulfur. This effect of
staging on cost is perhaps better shown in Table 4-6, which gives the capital
investment of the HDS units in dollars rather than annualized costs per barrel.
This table clearly shows the jumps in investment in going from one to two to
three stages of desulfurization.
Maintenance Charges
Maintenance charges also play a major role in the overall cost of desul-
furization ranging from 5 to 8.5% of the overall cost. As might be expected,
the maintenance charges parallel the investment charges, since they are simply
a fixed percentage of investment.
Utilities Cost
Utility consLiuption and costs are presented in Table 4-19. As presented
here, utility costs are relatively independent of feed sulfur level but are
directly sensitive to the degree of sulfur removal. For this reason, the data
is presented on the basis of low, medium, and high sulfur levels corresponding
to less than 3% S, 3-4% S, and more than 4% S. Thus, the Ceuta and E. Venezuelan
residuals fall under the low sulfur category, the Kuwait residual under the
medium, and the Khafji and Cold Lake residuals are classified as high sulfur
-148-
-------
TABLE 4-19. UTILITY CONSUMPTION AND COST
NOTE : Utility Units
Power in KWFI/Bbl
Steam in MBTU/Bbl
Fuel in MBTU/Bbl
Cooling Water in MGa1/Bbl
1.6
__ Sulfur in Fuel Oil
0.8 0.3
0.1
Utility
Un i t
Costs
Cost
Usage /BBL
Crude Class
Low Sulfur
<3% S
Med. Sulfur
3-4% S
High Sulfur
> 4% S
Cost
Usage /BBL
Cost
Usage /BBL
Cost
Usage /BBL
Power
Steam
Fuel
Cooling
Water
25.8 Mils/KWH
$3.15 x lO
$2.91 x 10-6
$O.18/MGa1
BTU
BTU
2,7
12.8
43.0
0.08
7.0
4.0
12.5
1.4
5.7
28.3
82.0
0.12
Total
Cost
Power
Steam
Fuel
Cooling
Water
25.8 Mils/KWH
$3.15 x 10-6
$2.91 x 10-6
$O.18/MGa1
BTU
BTU
5.3
31.0
76.0
0.12
6.9
43.8
87.0
0.14
Total
Cost
Power
Steam
Fuel
Cooling
Water
25.8 Mils/KWH
$3.15 x i 6
$2.91 x 10-6
$O.18/MGa1
BTU
BTU
6.4
37.4
81.0
0.13
8.0
49.2
90.0
0.14
14.7
8.9
23.9
2.2
49.7
17.8
13.8
25.3
2.5
59.4
20.6
15.5
26.2
2.5
64.8
24.9
13.7
9.8
22.1
2.2
47.8
16.5
11.8
23.6
2.3
54.2
7.2
44.4
92.3
0.14
9.0
63.8
94.0
0.15
9.7
79.6
95.0
0.15
Total Cost
18.6
14.0
26.9
2.5
62.0
23.2
20.1
27.4
2.7
73.4
25.0
25.1
27.6
2.7
80.4
7.6
48.6
96.0
0.16
9.4
71 .2
94.3
0.15
9.9
84.4
96.0
0.16
19.6
15.3
27.9
2.9
65.7
24.3
22.4
27.5
2.7
76.9
25.5
26.6
27.9
2.9
82.9
-------
oils. One important item of Table 4-19 is that it does not include the utili-
ties associated with the production of hydrogen but does include all power,
steam, fuel, and cooling water used in the desulfurization unit, and the sulfur
recovery and tail-gas units. The consumption and cost of hydrogen has such an
appreciable impact on the overall cost of desulfurization that hydrogen cost
is treated as a separate cost element in our cost analysis.
Even without the hydrogen utilities, the cost of utilities is a major
cost element ranging from about 12% to more than 25% of the overall desulfuri-
zation costs.
Hydrogen Cost
Hydrogen usage is primarily a function of the sulfur and asphaltene content
of the residuum fed to the HDS unit. As seen in Tables 4-13 through 4-18 at
S.I.P. (1.6%) and moderate (0.8%) sulfur levels, hydrogen represents the most
important variable cost in the overall cost picture. Table 4-20 gives hydrogen
consumption and cost for desulfurization of the five residual fuel oils to the
four control levels of sulfur.
As seen in Table 4-20, hydrogen usage varies almost directly with the de-
gree of desulfurization. Hence, although the cost of hydrogen is a major cost
element at all levels of desulfurization, the impact on overall cost is somewhat
reduced at higher degrees of sulfur removal or lower sulfur levels. For exam-
ple, the cost of hydrogen to desuifurize a Ceuta residua to the moderate level
of control (0.8% S) is $O.67/bbi or 28.9% of the total cost of desulfurization;
whereas, to desulfurize the same residua to the intermediate level of control
(0.3% S) and stringent level of control (0.1%) results in hydrogen costs of
$i.02/bbl or 25.6% of the total and $l.l7/bbl or 21.9% of the total, respectively.
Catalyst Cost
Catalyst usage is an exponential function of the degree of sulfur removal
and the metal (primarily Ni + V) content of the residuum feed. Table 4-21 gives
-150-
-------
TABLE 4-20. HYDROGEN CONSUMPTION AND COST FOR DESULFURIZATION OF RESIDUAL FUEL OIL
— Residual Oil Percent Sulfur in Residual Oil
Source 1.6 0.8 0.3 0.1
% S H Use % S 112 Use % S H Use % S H Use
Removal SCF/BBL Removal SCF/BBL Removal SCF/BBL Removal SCF/BBL
Ceuta 2.12 24.5 128 62.3 418 85.8 638 95.3 728
292 ppm (V + Ni) ($0.21) ($0.67) ($1.02) ($1.17)
E. Venezuelan 2.38 32.8 184 66.4 454 87.4 6&1 95.8 735
274 ppm (V + Ni) ($0.29) ($0.73) ($1.04) (S1.18)
(u’aait 3.80 57.9 365 78.9 549 92.1 666 97.4 720
‘60 ppm (V + Ni) ($0.58) ($0.88) ($1.07) ($1.15)
Khafji 4.36 63.3 414 81.7 576 93.1 680 97.7 729
118 ppm (V + Ni) ($0.66) ($0.92) ($1.09) ($1.17)
Cold Lake 4.55 64.8 436 82.4 599 93.4 705 97.8 751
236 ppm (V + Ni) ($0.70) (SO.96) ($1.13) ($1.20)
NOTE : Hydrogen costs are shown in parentheses under hydrogen use and are based on
cost of $1.60 per SCF.
-------
TABLE 4-21. CATALYST CONSUMPTION AND COST FOR DESULFURIZATION OF RESIDUAL FUEL OIL
Residual Oil Percent Sulfur in Residual Oil
Source 1.6 0.8 0.3 0.1
Catalyst Catalyst Catalyst Catalyst
% S Use % S Use % S Use % S Use
Removal LBS/BBL Removal LBS/BBL Removal LBS/BBL Removal LBS/BBL
Ceuta 2.12 24.5 0.51 62.3 0.21 85.8 0.67 95.3 1.2
292 ppm (V + Ni) ($0.09) ($0.38) ($1.21) ($2.16)
E. Venezuelan 2.38 32.8 0.069 66.4 0.227 87.4 0.637 95.8 1.4
.!. 274 ppm (V + Ni) ($0.12) ($0.41) ($1.15) ($2.52)
Kuwait 3.80 57.9 0.06 78.9 0.083 92.1 0.11 97.4 0.135
60 ppm (V + Ni) ($0.11) ($0.15) ($0.20) ($0.24)
Khafji 4.36 63.3 0.105 81.7 0.182 93.1 0.30 97.7 0.40
118 ppm (V + Ni) ($0.19) ($0.33) ($0.54) ($0.72)
Cold Lake 4.55 64.8 0.192 82.4 0.40 93.4 0.714 97.8 1.3
236 ppm (V + Ni) ($0.35) ($0.72) ($1.29) ($2.34)
NOTE : Catalyst costs are shown in parentheses under catalyst use and are based on
cost of $1.80 per pound.
-------
the catalyst consumption and cost for desulfurization of the five selected
residual fuel oils and the four control levels of sulfur. It should be pointed
out that this table represents a worse case basis in that the catalyst usage
was calculated on the basis of no guard reactor being used. In the case of
high metal feeds or severe desulfurization to extremely low levels such as
0.1% S, a refinery would probably use a guard reactor for removing metals which
would reduce the desulfurization catalyst usage and cost by perhaps 25-30%.
As seen from Tables 4-13 through 4-18, catalyst costs vary more than any
other cost element, ranging from a low of $0.09/bbl or 9.6% of the total cost
for desulfurization of a low sulfur Ceuta residua to a high of $2.52/bbl or
43.5% of the total cost for desulfurization of a high metal (274 ppm Ni + V)
E. Venezuelan residual. It is also evident from Tables 4-13 through 4-17 that
desulfurization to S.I.P. (1.6% S) or moderate (0.8% S) levels of control re-
suit in modest catalyst costs ranging from $0.11/bbl or 5.8% to $0.15/bbl or
5.8% for desulfurization of a low metals Kuwait residual to these sulfur levels.
These tables dramatically show the effect of metal content as the degree of
desulfurization or level of control approaches the intermediate control level
of 0.3% S. At this level, catalyst usage begins to dominate the cost profile,
except for the low metal feedstocks. At stringent levels of desulfurization
(0.1% S), the metal content becomes of prime importance; and, in the case of
high metal feedstocks such as the Ceuta and E. Venezuelan feeds, can become the
highest single cost element.
Figures 4-11 and 4-12 vividly illustrate the relative importance of the
above cost segments. Figure 4-11 gives the cost distribution for the desulfur-
ization of three residua to S.I.P. (1.6%) and moderate (0.8%) sulfur levels,
whereas Figure 4-12 presents similar information for desulfurization to inter-
mediate (0.3%) and stringent (0.1%) sulfur levels.
—153—
-------
Figure 4-11
COST DISTRIBUTION FOR DESULFURIZATION OF
THREE RESIDUA TO MODERATE SULFUR LEVELS
LABOR HYDROGEN [ I
UTIUTIES 111111111111 CATALYST _______
INVESTMENT SULFUR
MAINTENANCE PLANT
WASTE DISPOSAL
aIfur Plant erations Result in Net Credit
Note: Figures Sh % o( Total Cost for Eath Element
260
240
220
200
180
160
140
120
100
80
60
40
20
0
Ceuta
Kuwait
S.I.P. (1.6% 5)
Cold
Lake
-J
w
4:
U)
I-
z
w
0
I-
Cl)
0
C)
-154-
360-
340 -
320 -
300
280 -
260
240
220
200
180
160
140
120
100
80-
60
40
20
0
5.0).
16.8
30.9
21.9
3.0
35.3
36
231
2.8
I
21.1
?8.1
34.(
19.(
2.0
Ceuta Kuwait
MODERATE (0.8% 5)
6.0
-J
uJ
4:
U)
I-
z
U i
C)
0
C)
N
Cold
Lake
-------
Figure 4-12
COST DISTRIBUTION FOR DESULFURIZATION OF
THREE RESIDUA TO INTERMEDIATE & STRINGENT SULFUR LEVELS
LABOR V /VA HYDROGEN f
UTIlITIES Jfflffl CATALYST 1 1
INVESTMENT SULFUR
MAINTENANC PLANT
WASTE DISPOSAL
Sulfur Plant Operations Result in Net Credit
Note: Figures Show % of Total Cost for Each Element
480-
440
400— (1.8)
360—
31.0 (4.2) _____
6.4 ______
-J _____
LU _____
280- ____ ___
240- _____ _____
z _____ _____
LU 261 ____ ____
0 - ___ ___
200_ ____
160.
120-
27.1
80-
40_ 15.9
1.8
I
34.1
38.1
23.4
2.2
I
(3.5)
28.5
25.0
30.8
17.8
1.5
-J
w
C,)
I—
z
w
0
F:.
C /)
8
600
560-
(1.4)
520_
480-
440- 40.9
400—
360—
320—
280-
240— 22.2
200-
160..
24.6
120-
80
40— 12.5
1.3
(4.0)
6.8
32.8
10.4
21.9
2.0
(2.9)
40.1
20.5 ________
26.9
‘I
14.2
1.2 A
CEUTA KUWAIT COLD LAKE CEUTA KUWAIT COLD LAKE
INTERMEDIATE (0.3% S)
STRINGENT (0.1% S
-155-
-------
The extreme effect of catalyst consumption is clearly shown in these fig-
ures by comparing the low—metals Kuwait figures to the Ceuta and Cold Lake resi-
duals, both of which are hi h-meta1 feedstocks. These charts also clearly dem-
onstrate the lesser importance of catalyst cost at the lower degrees of desul-
furization or moderate to high levels of control.
Sulfur Credits
Table 4—22 gives the sulfur removal (in long tons per day) for the five
residual fuel oils at the four selected control levels. Sulfur removals are
based on a 100% yield and, for cost purposes, is credited at $25 per ton.
Sulfur credits are modest, ranging from as low as 24/bbl for low sulfur
residua and moderate desulfurization to nearly l7it/bbl for the high sulfur resi-
dua and stringent levels of desulfurization. As seen from Figures 4-11 and
4-12, these credits reduce the overall cost of desulfurization from 2.2 to as
much as 4.6%.
-156-
-------
TABLE 4-22. SULFUR REMOVAL FOR VARIOUS CONTROL LEVELS
BASED ON 50,000 B/D RESIDUAL FUEL OIL
Basis for calculation 50,000 B/D residual fuel oil of 15.7°API
gravity (0.961 SP. Gr.)
Pounds/day of fuel
oil = 0.961 x 8.33 x 42 x 50,000
= 16.81 x 106
Tons/day of sulfur removed = 16.81 x 106
( % S 1 - % SOut ) (22 O
100
= 75.04 (% S 1 - % so )
Residual
Oil
Tons/Day Sulfur
Removed
ForVarious %
Source
S
1.6
0.8
0.3
AS
0.1
T/D
AS
T/D
AS
T/D
AS
T/D
Ceuta
2.12
0.52
39.1
1.32
99.2
1.82
136.7
2.02
151.8
E. Venezuela
2.38
0.78
58.6
1.58
118.7
2.08
156.3
2.28
171.3
Kuwait
3.80
2.20
165.3
3.00
225.4
3.50
263.0
3.70
278.0
Khafji
4.36
2.76
207.4
3.56
267.5
4.06
305.0
4.26
320.1
Cold Lake
4.55
2.95
221.6
3.75
281.8
4.25
319.3
4.45
334.3
-157-
-------
REFERENCES
1. Paraskos, J. A., et al., “Ecologically Acceptable Fuels From the Gulf HOS
Process,” 67th Annual Meeting of the American Institute of Chemical Engi-
neers, 1—5 December 1974.
2. Frayer, J. A., et al., “Gulf’s HDS Processes for High Metal Stocks,” present-
ed at the Japan Petroleum Institute, Tokyo, Japan, 8 May 1975.
3. Krueding, A. P., “RCD Isomax: Production Route to Today’s and Tomorrow’s
Low Sulfur Residual Fuels,” AIChE 71st National Meeting, Dallas, Texas,
February 1972, pp. 20-23.
4. Reed, E. M., P. W. Tamm, and R. F. Goldstein, “HDS Goes Deeper Into Barrel
Bottom,” Oil and Gas Journal , 17 July 1972, pp. 103-108.
5. Fowler, C., “How Gulfining Works for Daikyo,” Hydrocarbon Processing ,
September 1973, PP. 131-133.
6. Guthrie, K. M., “Capital and Operating Costs for 54 Chemical Processes,”
Chemical Engineering , 15 June 1970, p. 140.
7. U. S. EPA Report, EPA-450/2-76-016(a), September 1976.
8. Nelson, W. L., “A Look at Sulfur Recovery Costs,” Oil and Gas Journal ,
18 March 1974, p. 120.
9. “Gas Processing Handbook,” Hydrocarbon Processing , April 1975, pp. 107-111.
10. Devitt, 1., P. Spaite, and L. Gibbs, “Background Study in Support of New
Performance Standards for Industrial Boilers,” PEDC0. Environmental, Inc.,
EPA Contract No. 68-02-2603, Task No. 19, March 1979.
11. Nelson, W. L., “Data Correlation Shows the Amount of Hydrogen Used in Desul-
furizing Residua,” Oil and Gas Journal , 28 February 1977, p. 126.
12. Nelson, W. L.., “Catalyst Consumption Required in Desulfurizing Residua,”
Oil and Gas Journal , 15 November 1976, p. 72.
13. Oil and Gas Journal , 7 July 1972, pp. 131—133.
-158-
-------
4.5 APPENDIX - SAMPLE CALCULATIONS
The following calculations are based on the desulfurization of Kuwait
residuum to a 0.1% product sulfur level and are presented here to illustrate
the bases and methods of calculations used in this report.
1. Per Cent Desulfurization
% Desulfurization = % Sj, - % Sp j
x lOO
% Desulfurization = 3.8 — 0.1
3.8 X 100 = 97.4%
2. Hydrogen Usage and Cost
From Ref 1. API gravity = 15.7°
From Ref. 11, Fig. 1
For API gravity of 15.7° and 97.4% sulfur removal the base hydrogen
consumption = 810 SCF/Bbl
Packed bed credit = 9%
Correction for metals content = -2% hence:
Corrected H 2 consumption = 810 x 0.91 x 0.98
= 720 SCF/Bbl
Hydrogen cost @ $1.60/1O 3 SCF = $1.60 x 720
10
= $1.15
3. Catalyst Usage and Cost
From Ref. 12 Fig. 1
At 97.4% sulfur removal and 6oppm metals in the feed
Barrels feed/# catalyst = 7.4
or lbs. catalyst consumed per barrel feed 1 0.135
7.4
Catalyst cost @ $1.80/lb. = $1.80 x 0.135 = $0.24
-159—
-------
4. Investment Charges
From Table 4-9
Total Investment at 0.1 % S = $101.1 x 106
Investment Charge @ 17%/yr.
= $101.1 x 106 x 0.17 x 100
50,000 x 330
= $1.042/Bbl.
5. Maintenance Charges
At 5% of investment
Maintenance ctiarge = $101. I x i0 6 x 0.05 x 100
50,000 x 330
= $0.306
6. Sulfur Plant Costs
From Ref. 1 API gravity 15.7°
Specific gravity = 141.5 0.961
131.5 + 15.7
% sulfur in = 3.8%
% sulfur out = 0.1%
Sulfur removal 3.8 - 0.1 x ( 0 .961)(8.33)(42) 5O000 )
100 2240
= 277.7 tons/day
Frvm Fig. 4-9
Sulfur plant cost = $7.5 x i06
Check cost by use of equation
C = 0.1877 x (capacity) 0 656
C = 0.1877 x (277.7)0.656
C = $7.56 x 106
-160-
-------
7. Sulfur Credit
at $25/ton
Sulfur credit = 277.7 x 25 = $0. 139/Bbl.
50000
8. Waste Disposal Charges
Basis 1.1% of Total Plant Investment
Cost = 0.011 x $101.1 x 106
50000 x 330
= $ 0.067/Bbl.
9. Labor, Supervision and Overhead
a) Labor
Men/shift = 6
No. shifts = 4
Unit cost per manhour = $12.02
Annual Cost = 4 x 6 x 40 x 52 x $12.02
50000 x 330
= $ 0.036/Bbl.
b) Supervision
Basis 1 man/shift, 4 shifts and $15.63 per manhour.
Cost = 1 x 4 x 40 x 52 x 15.63
50000 x 330
= $ 0.008/Bbl.
c) Overhead
Basis - Payroll overhead 30% of (a + b)
Plant overhead 26% of (a + b)
or total of 56% of (a + b)
Cost = 0.56 x ($0.036 + $0.008)
= 0.56 x $0.044
= $O.025
-------
10. Fuel Oil Costs
Basis: Residual Fuel Oil Gravity 22.O0API
Gross Heat of Combustion 19,000 BTU/lb.
Cost = $221,106 BTU (PEDC0)
Distillate Fuel Oil Gravity 26.O0API
Gross Heat of Combustion 19,200 BTU/lb.
Cost = 3.00/106 BTU (PEDC0)
NOTE: Gravities are typical of source data
Gross heats from API data
a) Residual Fuel Oil Specific Gravity
S.G. = 141.5
131.5 + 22.0
= 0.922
b) Residual Fuel Oil Heat Content
Heat Content = 0.922 (8.33) (42) (19,000)
= 6,135,000 BTU1BB1
c) Residual Fuel Oil Cost
Cost = 6,135,000 x 2.21
lob
= $13.55/BBl
d) Distillate Fuel Oil Specific Gravity
S.G. = 141.5
131.5 + 26
= 0.898
e) Distillate Fuel Oil Heat Content
Heat Content = 0.898 (8.33) (42) (19,200)
= 6,042,200 BTU/BBL
Residual Fuel Oil Cost
Cost = 6,042 200 x 3.00
10
= $18. l3/BB1
-162-
-------
11. Annual Costs
a) Oil Cohsumption (Residual)
Basis: 150,000 M BTU/hr.
Unit Cost $2.410/BB1
BB1/hr = 150,000,000/6. 135 x 106
= 24.45 BB1/hr.
b) Oil Consumption (Distillate)
Basis: 15,000 M BTU/hr.
BB1/hr. = 15,000,000/6.0422 x 106
= 2.48 BB1/hr.
c) Incremental Cost
Basis Kuwait Resid, 0.1 Wt. % S
Cost=24.45x24x330x3.51
= 679690/yr.
d) Annual Cost/Il BTU/hr.
Cost = 679690/150 x 10
= $4.53/M BTU/hr.
e) Cost/KJ/Sec.
Conversion M BTU/hr. x 0.293 = KJ/Sec.
Cost = 4.53/0.293
= $15.46/KJ/Sec.
12. Cost Impacts
Basis: Residual Fuel Oil Cost $13. 55/BB1
a) % Increase over Uncontrolled
= 3.51 x 100
13.55
= 25.90%
-163-
-------
b) % Increase over S.I.P. Controlled
Basis: S.I.P. Control Cost $1.80/BB1
= 13.55 + 3.51 - 1 x 100
13.55 + 1.80
= 11.14%
13. Cost Effectiveness
Basis: 277.7 Long Ton/Day Sulfur
3.51/BB1 Oil
a) C.E. = 3.51 (50,000 )
277.7 (2240)
b) C.E. = 0.28 (2.2)
$0.62/KG S Removed
- 64-
-------
SECTION 5
ENERGY IMPACT OF BEST EMISSION CONTROL SYSTEM FOR CLEAN OIL
5.1 INTRODUCTION
Based upon the rating factors developed in Section 3, hydrodesulfurization
was selected as the best system for emission reduction for clean oil technology.
To allow for a complete energy impact assessment, a brief description of how
and where energy is used in the fuel oil hydrodesulfurization process (HDS) over-
all is presented here. The description is applicable to both distillate and
residual fuel oil, since the HDS processes are similar for both oils.
A small energy debit,which must be assessed against the HDS process is the
reduced calorific value of the lighter product resulting from the desulfuriza-
tion process. This results from a hydrocracking reaction which occurs to a
limited extent along with the HDS reaction and which breaks down the heavier
hydrocarbons yielding a lighter product of lower calorific value (on a volume
basis) than the feed. The normal caloric reduction from HDS usually amounts
to about 1% of the nominal 6.15 MMBtu/barrel value. Thus, a greater volume of
desulfurized fuel oil is necessary to produce the required heat release in the
boiler. However, slight boiler atomizing nozzle modifications can easily com-
pensate for this minor caloric reduction so that there is no noticeable boiler
derating as a result of burning the desulfurized fuel oil. No definitive
assessment is made in this section of the energy impact of this debit.
The HDS process has not been considered on a stand-alone basis for the
energy impact assessment, since auxiliary processes are required to dispose of
process by-products. These auxiliary processes include a hydrogen sulfide
absorption unit (circulating amine type), sulfur recovery with tail-gas scrub-
bing (Claus type with reduction system and tail-gas reheat), a sour water
stripper (steam stripping), and a hydrogen plant.
-165-
-------
HDS Process
The process utilizes a catalytic reaction of oil feed and hydrogen at high
temperature and pressure in one or more reactor vessels and hence requires
shaft energy to pump liquid and compress hydrogen to reaction pressure. Fossil
fuel is required in a preheat furnace to heat the reactants to reaction temper-
ature, while shaft power and cooling water are required by air coolers and water
coolers, respectively, to cool the reaction products.
Separation of the reaction products is achieved in both flash drums and
fractionation tower, the latter requiring shaft energy to pump feedstream,
reflux, and products and shaft energy and/or cooling water to condense over-
head products. Low-pressure (LP) steam is used for stripping products in this
fractionation tower. This steam is condensed with overhead tower products,
separated and routed to the sour water stripper.
1125 Absorption Process
A circulating amine process is usually used to remove the H 2 S produced
in the catalytic reaction from both the reaction products and the recycled
112 returning to the reactor(s). Shaft energy is required to pump the amine
between absorption and stripping towers and provide reflux at the latter, while
shaft power and/or cooling water provides condensation of stripping steam and
cooling of the regenerated amine stream. High-pressure (HP) steam is used to
reboil the stripping tower and provide stripping vapors.
Sulfur Recovery and Tail-Gas Scrubbing
Part of the H 2 S recovered from the circulating amine is combusted with
air to fonn SO in a cootined furnace-waste heat boiler requiring shaft energy
to compress air for cothustion and pump water to the boiler and producing steam
for an energy credit. The H 2 S-S0 2 mixture is then reacted over a series of three
converters with fixed catalyst beds; sulfur formed in the reactors is condensed
in water-cooled condensors. The ml xture is then reheated with hot reaction
-------
gases. Tail-gas from the third condensor is combined with combustion products
of fossil fuel and compressed air to provide reheat and reducing gas before
entering a further reactor to reduce the remaining SO 2 back to H 2 S before cool-
ing and absorption in a circulating solution such as used in the Shell Claus
Off-Gas Treating Process (SCOT). This process requires shaft energy for
pumping the circulating solution and LP steam for regenerating this solution.
Sour waste water is released in the cooling process and is sent to the sour
water stripper. Fossil fuel is required to incinerate the absorber off-gas
before release to the atmosphere.
Sour Water Stripping
Sour water containing H 2 S from either the 1 -IDS fractionation tower or SCOT
process is stripped with LP steam in a stripping tower prior to release to the
plant effluent treatment system.
5.2 ENERGY IMPACT OF CONTROLS FOR OIL-FIRED BOILERS
New Facilities
Table 5-1 included in this section provides a detailed listing of the forms
of energy utilized in the HDS and auxiliary processes at the various control
levels specified viz. SIP, Moderate, Intermediate, and Stringent.
An example calculation for medium sulfur residual-stringent level of con-
trol is shown in the Appendix of this section.
Energy vs. Level-of-Control
Low-capacity steam boilers 0 f the fire-tube and Scotch types use distil-
late fuel oil (i.e., #2) which is presently limited by ASTM specification 0
at 0.5% W/W sulfur. This represents an Intermediate level-of-control and is
presently used for about 10% of the #2 oil nationwide requirements. The Strin-
gent level-of-control of 0.10 lbs. SO 2 per MM Btu/hr., which represents approx-
imately 0.10% W/W sulfur fuel oil, necessitates a deeper level of desulfuriza-
tion which requires a lower catalyst space velocity, increased H 2 consumption,
and higher temperature levels in the catalyst bed. The plot of energy
-167-
-------
TABLE 5-1. ENERGY CONSUMPTION FOR SO 2 CONTROL TECHNIQUES FOR OIL-FIRED BOILERS
Energy Consumption
System b
% Change In
Standard Boiler Energy Consumed % Increase Energy
Control by Control Device in Energy Use Over SIP
Heat Input Type & Level Efficiency Energy Use Over Uncon- Controlled
MW (MBtu/hr) Type of Control _________ Types SI English trolled Boiler Boiler
44 (150,000) Firetube Desulfurized
Distillate
Fuel 011
SIP 0 Fossil Fuel
Moderate 0 Fossil Fuel - - -
IntermedIate 40 Fossil Fuel 325 MJ/m 3 49 MBtu/bbl 0.8 0.8
Stringent 80 Fossil Fuel 372 MJ/m3 56 MBtu/bbl 1.0 0.1
SIP 0 Electrical
Moderate 0 Electrical
IntermedIate 40 Electrical 25 MJ/m3 1.1 Kwh/bbl 0.06 0.06
Stringent 80 ElectrIcal 32 MJ/m3 1.4 Kwh/bbl 0.08 0.08
SIP 0 HP Steam
Moderate 0 HP Steam - -
Intermediate 40 HP Steam —21 Kg/rn 3 -7.2 lb/bbl -0.11 -0.11
Stringent 80 HP Steam 6.9 Kg/rn 3 2.4 lb/bbl 0.04 0.04
SIP 0 LP Steam
Moderate 0 LP Steam - -
Intermediate 40 LP Steam 15 Kg/rn 3 5.4 lb/bbl 0.09 0.09
Stringent 80 LP Steam 15 Kb/rn 3 5.4 lb/bbl 0.09 0.09
-------
TABLE 5-1.
ENERGY CONSUMPTION FOR SO 2 CONTROL TECHNIQUES FOR OIL-FIRED BOILERS (cont’d.)
Energy Consumption
Type & Level
Type of Control
Watertube Desul furl zed
Residual
Fuel Oil
Control
Efficiency
Energy
Types
Energy Consumed
by Control Device
Lri9l ish
% Increasea .
in Energy
Use Over Uncon-
troUed Boiler
% Change 1 gb.
Energy
Use Over SIP
Control led
Boiler
—High Sulfur
Residuals
SIP
Moderate
Intermediate
Stringent
-Medium Sulfur
Residuals
SIP
Moderate
Intermedi ate
Stringent
-Low Sulfur
Residuals
SIP
Moderate
Interrnedi ate
Stringent
81 MBtu/bbl
90 MBtu/bbl
95 MBtu/bbl
95 MBtu/bbl
76 MBtu/bbl
87 MBtu/bbl
93 MBtu/bbl
94 MBtu/bbl
43 MBtu/bbl
82 MBtu/bbl
89 MBtu/bbl
94 MBtu/bbl
1.3
1.5
1.6
1.6
1.2
1.4
1.5
1.6
0.7
1.4
1.5
1.6
0.2
0.3
0.3
Standard Boiler
Heat Input
MW (MBtu/hr )
44 (150,000)
64
76
94
98
Fossil
Fossil
Fossil
Fossil
Fuel
Fuel
Fuel
Fuel
537 d MJ/m 3
597 MJ/m 3
630 MJ/m 3
630 d MJ/m 3
59
75
93
98
Fossil
Fossil
Fossil
Fossil
Fuel
Fuel
Fuel
Fuel
504 d R i/n i 3
577 Ri/rn 3
617 MJ/ni
624 MJ/m
37
75
88
96
Fossil
Fossil
Fossil
Fossil
Fuel
Fuel
Fuel
Fuel
285 d MJ/rn 3
544 MJ/m 3
S91dMJ/m 3
623 MJ/m 3
0.1
0.2
0.2
0.6
0.8
0.8
-------
TABLE 5-1.
ENERGY CONSUMPTION FOR SO 2 CONTROL TECHNIQUES FOR OIL-FIRED BOILERS (cont’d.)
Heat Input
MW (MBtu/hr )
44 (150,000)
Type & Level
Type of Control
Watertube Des Ui furized
Residual
Fuel Oil
Energy Consumed
by Control Device
% Change
Energy
Use Over SIP
Control led
Boi1e ’
System
Standard Boiler
Control
Efficiency
Energy
Types
Energy Consumption
% Increase
in Energy
Use Over Uncon-
English trolled BojIei ’
-High Sulfur
Residuals
SIP
Moderate
Intermediate
Stringent
64
76
94
98
Electrical
Electrical
Electrical
Electrical
d 3
145 MJ/
181 MJ/m
213dMJ/m 3
222 MJ/m 3
6.4
8.0
9.4
9.8
KWH/bbl
KWH/bbi
KWH/bbi
KWH/bbl
0.4
0.5
0.5
0.6
-
0.1
0.2
0.2
-Medium Sulfur
‘Residuals
SIP
Moderate
Intermediate
Stringent
59
75
93
96
Electrical
Electrical
Electrical
Electrical
d
120 MJ/m 3
156 MJ/m 3
197 MJ/m 3
208 MJ/m 3
5.3
6.9
8.7
9.2
KWH/bbl
KWH/bbl
KWH/bbl
KWH/bbl
0.3
0.4
0.5
0.5
-
0.1
0.2
0.2
-Low Sulfur
Residuals
SIP
Moderate
Intermediate
Stringent
37
75
88
96
Electrical
Electrical
Electrical
Electrical
61 d MJ/m 3
129 MJ/m 3
154 MJ/m 3
168d MJ/m 3
2.7
5.7
6.8
7.4
KWIi/bbl
KWH/bbl
KWH/bbl
KWH/bbl
0.1
0.3
0.4
0.4
-
0.2
0.2
0.3
-------
TABLE 5-1.
ENERGY CONSUMPTION FOR SO 2 CONTROL TECHNIQUES FOR OIL-FIRED BOILERS (cont’d.)
Energy Consumption
Heat Input
MW (MBtu/nr )
44 (150,000) Watertube
Type & Level
of Contrul
Desul fun zed
Resi dual
Fuel Oil
Control
Efficiency
Energy
Typ
% Increase
in Energy
Use Over Uncon-
trolled Buler
% Change
Energy
Use Over SIP
Controlled
Bc;ler
System
Standard_Boik!ii
Energy Consumed
y Control Device
9 d
17
17
KG/rn 3
KG/rn 3
KG/rn 3
KG/rn 3
-High Sulfur
Residuals
SIP
Moderate
Intermediate
Stringent
-Medium Sulfur
Residuals
SIP
Moderate
Intermedi ate
Stringent
-Low Sulfur
Residuals
SIP
Moderate
Intermediate
Stringent
64
76
94
98
59
75
93
96
37
75
88
96
English
3 LB/BBL
4 LB/BBL
6 LB/BBL
6 LB/BBL
1 LB/BBL
2 LB/BBL
0 LB/BBL
5 LB/BBL
1 LB/BBL
1 LB/BBL
2 LB/BBL
2 LB/BBL
HP Steame
HP Steam
HP Steam
HP Steam
HP Steam
HP Steam
HP Steam
HP Steam
HP Steam
HP Steam
HP Steam
HP Steam
3 d KG/rn 3
6 KG/rn 3
0 KG/rn 3
14 KG/rn 3
3 d KG/rn 3
3 d KG/rn 3
6 d KG/rn 3
6 d KG/rn 3
0.04
0.06
0.09
0 .09
0.01
0.03
0
0.08
0.01
0.01
0.03
0.03
0.01
0.04
0.05
0.01
0
0.06
0
0.01
0.01
-------
TABLE 5-1. ENERGY CONSUMPTION FOR SO 2 CONTROL TECHNIQUES FOR OIL-FIRED BOILERS (cont’d,)
Energy Consumption
System b
% Change in
Standard Boiler Energy Consumed % Increasea. Energy
Control by Control Device in Energy Use Over SIP
Heat Input Type & Level Efficiency Energy Use Over Uncon- Controlled
MW (MRtu/hr) Type of Control _________ Types SI English trolled Boiler Boiler
44 (150,000) Watertube Desulfurized
Residual
Fuel Oil
-High Sulfur
Residuals f d 3
SIP 64 LP Steam 108 KG/rn 38 LB/BBL 0.6 -
Moderate 76 LP Steam 143 KG/ni 3 50 LB/BBL 0.8 0.2
IntermedIate 94 LP Steam 217dKG/rn 3 76 LB/BBL 1.2 0.6
Stringent 98 LP Steam 240 KG/rn 3 84 LB/BBL 1.3 0.7
-Medium Sulfur
Residuals
SIP 59 LP Steam 94 J KG/W ’ 33 LB/BBL 0.5 -
F4oderate 75 LP Steam 131 KG/rn” 46 LB/BBL 0.7 0.2
Intermediate 93 LP Steam 177 KG/rn’ 3 62 LB/BBL 0.9 0.4
Stringent 96 LP Steam 197 KG/rn 3 69 LB/BBL 1.0 0.5
-Low Sulfur
Residuals d 3
SIP 37 LI’ Steam 37 KG/rn 13 LB/BBL 0.2 -
Moderate 75 LI’ Steam 86 KG/rn 3 30 LB/BBL 0.5 0.3
Intermediate 88 LP Steam l2OdKG/m 3 42 LB/BBL 0.6 0.4
Stringent 96 LI’ Steam 140 KG/rn 3 49 LB/BBL 0.7 0.5
a. Energy consumed by device, MW ÷ standard Boiler Heat c. High Sulfur Residuals - 4+% W/W S
Input, MW x 100 (%) Medium Sulfur Residuals - 3.4% W/W S
Low Sulfur Residuals - 0-3% W/W S
b. Moderate, Intermediate, Stringent, energy consumed by
control device, MW - SIP energy consumed, MW ÷ standard d. Denotes estimated figures
Boiler heat input, MW + SIP energy consumed, MW X 100 e. 87 KPa, 400°C (600 psig, 750°F) steam
F. 7 KPa, Satd. (50 psig)
-------
consumption versus level—of—control curve is shown in Figure 5-1.
Those utilities related to the HDS process (i.e., fossil fuel and elec-
trical energy) show a predictable increase in consumption with increased sever-
ity of desulfurization, while the utilities related to the HDS auxiliary pro-
cesses (i.e., high- and low-pressure steam) do not follow a logical consumption
pattern but rather a change in data source and thus do not provide meaningful
information.
With only two data points for each form of energy, it is not possible to
locate an optimum point for level-of-control, since the increased SO 2 removal
takes progressively more energy to achieve.
The plot of energy consumption versus level-of-control curve for three
levels of sulfur content in residual fuel oil is shown in Figures 5-2 through
5-5. Each plot reflects SIP, Moderate, Intermediate, and Stringent level-of-
control points and shows a distinct knee” between Moderate and Intermediate
levels for all sulfur content fuel oils, indicating an increased rate of energy
consumption between these points.
From these plots, it is not possible to locate an optimum point for an
Intermediate level-of-control between Moderate and Stringent levels without
developing further data points, the data for which were not available in the
public literature.
Each of the energy consumption plots reflects a higher consumption for
high-sulfur content residuals than for medium or low and results from the in-
creased amount of sulfur or sulfur by-products produced which requires addi-
tional energy consumption both in the HDS unit and the auxiliary processes
serving the UDS unit.
Energy Use in Fuel Oil Desulfurization
HDS Process-—
Fuel-fired process heaters required to elevate the temperature of the oil
feed and H 2 before reaction over catalyst are the largest users of energy in
-173—
-------
Figure 5 -1
U. V)
. I
. C’ )
0
Energy Consumption vs. Level-of-Control
for Distillate Fuel Oil
0 ——— LPSteam
400 40 10
00 30 0
200 20 —10
100 10 —20
0 0
0
I I
0.2 0.4
Fossil Fuel
0
Electrical Energy
HP Steam
0.6 0.8
Level of Control
lb S0 2 /10 6 Btu
i..0 1.2
1.4 1.6
-------
Figure 5-2
700 -
600 -
50o -
cI,
u-
(.1
LL
400-
300
Fossil Fuel Energy Consumption
of High/Medium/Low Sulfur Residual
Fuel Oils vs. Level-of-Control
0 0.2 0.4
0.6
-I - - - - - - I I
1.0 1.2 1.4
High Sulfur Resid.
0
200
0.8
Level of Control
lb S0 2 /10 6 Btu
1.6
-------
Figure 5-3
300 -
Electrical Energy Consumption
of High/Medium/Low Sulfur
Residual Fuel Oils vs. Level-of-Control
200 -
w 100 -
.
8
High Sulfur Resid.
__ i111:11ii1:1ii:::::::::iii:::iiiiiiiiiiiiiiiiiiiii_iii__iiiiiii_iiiiiiii_ii_iiiiii:i Medium Sulfur Resid.
Low Sulfur Resid.
0
0
0 0.2 0.4
I I I I
0.6
1.0
0.8
Level of Control
lb S0 2 110 6 Btu
1.2
1.4
1.6
-------
Figure 5-4
HP Steam Consumption
of High/Medium/Low Sulfur
Residual Fuel Oils vs. Level-of-Control
15 -
C I
o
U-
C/)
-
E ’
+ - )
(I)
I
5.
° High Sulfur Resid.
AD Medium Sulfur Resid.
Low Sulfur Resid.
U I I
0 0.2 0.4 0.6
0.8 1.0
Level of Control
lb S0 2 /10 6 Btu
20
0
0
A
0
1.2
1.6
-------
FIgure 5-5
LP Steam Consumption of
High/Medium/Low Sulfur
Residual Fuel Oil vs. Level-of-Control
0 0.2 0.4
0.6 0.8 1.0 1.2
Level of Control
lb S0 2 /1 06 Btu
1.4
1.6
High Sulfur Resid.
Medium Sulfur Resid
Low Sulfur Resid.
4.3
—J o CI
I
w
4.3
p.4
250 -
200 -
150-
100-
50-
0
0
I
-------
the HDS process. The fuel used depends on the refinery product balance and
can range from fuel gas to a residual fuel, the latter being a captive use for
less desirable by-products of the refining process. Residual fuel oil is the
most cotmion type of fuel in this duty. Fuel gas is universally used as a
source for pilot gas, the usage being small.
Motor—driven high—pressure compressors for compressing both fresh and
recycle H 2 before mixing with oil feed are large users of power in this pro-
cess. Standby machines are usually provided with steam turbine drive as de-
scribed below.
Air-coolers are often used to condense and cool both reactor products and
fractionation tower overheads, and the motor-driven fans represent a major
user of power.
Motor-driven feed pumps are used to transfer oil feed to the high-reaction
pressure condition necessary for hydrodesulfurization and represent a major use
of power. Other pump duties such as fractionation tower feed and reflux, pro-
duct pumping, are less severe and use relatively less power.
Lighting, instrumentation, and line tracing are other small users of
power.
High-pressure (600 psi) steam is used to provide a steam turbine drive
for standby critical equipment such as the H 2 compressors and oil feed pump
and are operated normally only when required by the refinery steam balance.
Low-pressure (50 psi) steam is used mainly as stripping steam in the pro-
duct fractionation tower which represents a major use of steam.
Steam heating coils and tracing to maintain heavy oils above their pour
point represent a minor use of steam.
Amine Process--
Motor-driven pumps are used to circulate the H 2 S-absorbing solution of
amine between absorption and stripping towers, as well as providing reflux
-179—
-------
for the stripping tower. Anti-foam injection pumps and amine make-up pumps
also draw a small quantity of power.
Air-coolers used to condense stripping tower reflux and cool lean amine
utilize motor-driven fans and are major users of power.
Lighting and instrumentation are minor users of power in this process.
Low-pressure (50 psi) steam is used as a reboiler heating medium on the
stripping tower and represents a major use of steam. Steam heating coils and
tracing used to prevent the amine from freezing are small users of steam.
Sulfur Plant and Tail—Gas Scrubbing--
A fuel-fired line burner is used in the Tail Gas Scrubbing unit to pro-
vide reducing gases to reduce SO 2 back to H 2 S in the Reduction type of Tail-
Gas unit used as a basis for this discussion. A fuel-fired incinerator is
used after the Tail-Gas Scrubbing unit to convert all traces of H 2 S back to
SO 2 and provide a stack exit temperature high enough to provide stack—gas
bouyancy upon release to the atmosphere. The incinerator represents the major
user of fuel which is a refinery fuel gas rather than fuel oil type.
Motor-driven air compressors are necessary to provide combustion air for
the Sulfur Plant Furnace reaction of H 2 S to SO 2 before passing the gases over
catalyst beds and represent the major user of power in this process.
Motor-driven pumps are used to load liquid sulfur into tank cars and
circulate the absorbing solution in the Tail Gas Reduction process.
Air-coolers used in the Tail Gas Reduction process to reduce process gas
temperature prior to absorption in the circulating amine utilize motor-driven
fans. Lighting and instrumentation are minor users of power in this process.
High-pressure (600 psi) steam is produced in the combination furnace-waste
heat boiler in the Sulfur Plant and results from the highly-exothermic reaction
of H 2 S to SO 2 in the furnace. This steam is not used in the process and is thus
utilized elsewhere in the refinery steam balance.
-180-
-------
Low-pressure (50 psi) steam is produced in the sulfur condensers down-
stream of each catalyst bed of the Sulfur Plant and condenses liquid sulfur
from the reaction product gases. Only a small amount of this steam is used
in the Sulfur Plant for heating coils and jacketted piping, the balance being
exported to the refinery steam balance.
Low-pressure (50 psi) steam is also utilized downstream of the Reduction
type of Tail-Gas Scrubbing Unit catalyst bed and as reboiler heating medium in
a circulating amine stripping tower shared with other amine absorbers on the
refinery.
Sour Water Stripper--
Low-pressure (50 psi) steam is used as both a heating and stripping me-
dium in the sour water stripping tower to remove H 2 S and NH 3 from contaminated
waste water before releasing the water to the refinery effluent reating
system.
Hydrogen Plant--
Most refinery hydrogen requirements are produced in catalytic reforming units;
the advent of desulfurization requirements has meant the installation of a
hydrogen facility. Within the context of previous discussion in Section 4, the
energy impact of such a separate hydrogen plant will be examined. A typical
hydrogen plant consists of a sulfur guard unit, reformer furnace, CO shift
reactor, CO 2 removal unit, methanation unit, and make-up gas compressor. The
most common feedstock for hydrogen production is a hydrocarbon source, usually
naphtha, from the refinery complex. The naphtha is vaporized, mixed with steam,
and passed over a nickel catalyst at temperatures of 1200-1800°F.
The energy requirements of the hydrogen plant concentrate on the delivery
of the naphtha and steam input to the reformer,Tø achieve 1200-1800°F in the
reformer and high pressure, about 3000 lb/hr high-pressure steam (500 spig, 750°F)
and 5-7,000 lb/hr medium-pressure i350 psig, saturated) steam have to be supplied
for a 50 MIISLFD H 2 plant. Some medium-pressure steam is produced during
-18]-
-------
methanation, which can be used elsewhere or recycled prior to reforming to
serve as an overall credit. This level of steam production can easily utilize
1400-1500 gal/MCF cooling water and 4-6 gal/MCF boiler feed water.’ 2
Multi-stage compression of hydrogen to be supplied to the HDS unit repre-
sents a substantial energy outlay. Depending upon the pressure requirement of
the HDS, the hydrogen compressor output can range between 150 psig and 330
psig. At a normal compression ratio of 2.0 (outlet/inlet pressure), about
45 bhp is required for rotary or reciprocating compressors. This gives a
range of power required between 0.02-0.08 hp-hr/barrel processed solely for
the hydrogen plant, depending upon the compression level.
Electric power consumption for a hydrogen plant ranges between 1.4
KWH/MSCF H 2 and 1.6 KWH/MSCF H 2 , which means 1.4-1.6 KWH/barrel for the average
HDS hydrogen consumption of 1000 CF/barrel processed. 13
The largest energy impact from hydrogen production comes from the natural
gas fuel and process feed natural gas consumed during steam reforming. Process
feed gas utilizes about 260 SCF natural gas/MSCF H 2 , and the natural gas fuel
requirements are about 415 SCF natural gas/MSCF H 2 (415 MBtu/MSCF H 2 ).
Energy Conservation Options
A significant reduction in fuel use is achieved where control of excess
air used in the heater is practiced by means of stack gas monitoring using
analytical instriinentation either on an intermittent or continuous basis. The
accompanying reduction in combustion air fan power also results from this ex-
cess air control. The provision of combustion air preheating, additional con-
vection zone heat transfer surface, as well as soot blowing facilities, would
also improve furnace efficiency.
A reduction in energy use by improved maintenance practices results from
higher-efficiency operation of furnace burners, furnace fans, air-cooler fans,
gas compressors, liquid pumps by cleaning burner tips, and fluid filters
-182-
-------
throughout the unit. These operations are usually already performed on a
regular basis.
Optional maintenance practices imply a trade-off of energy saved versus
the cost of implementing maintenance procedures and is a difficult topic to
categorize. Generally, regular preventative maintenance programs will perform
the necessary energy-saving activities at a cost, while breakdown maintenance
programs only replace items no longer serviceable and are not the most energy—
efficient method. The choice of a maintenance program is very subjective, as
detailed payout studies are not available, since industry is not uniform in
the choice of which maintenance program is superior.
Stack gas heat recovery is not applicable to the fuel pretreatment pro-
cess discussed here.
Less energy intensive process is not applicable, since an HDS process
is the subject for discussion.
FueT switching Is not applicable, since both distillate and residual fuel
oil have been selected for discussion.
A considerable amount of heat previously released to the atmosphere via
air or water-condensers/coolers is now being recovered in exchange with pro-
cess- or steam—raising duties. In addition,. closer temperature approaches
than previously used are now being designed.
As air-coolers are designed for the hottest part of the year, considerable
savings in power are possible where the fan pitch is changed to reduce air rate
by means of variable-pitch controllers during cooler weather.
High-pressure streams that are depressurized via a hydraulic turbine re-
cover part of the energy necessary to achieve these high pressures and are
usually coupled with a pump and motor—driver to reduce the overall power con-
sumption of the pump.
-183-
-------
Energy Conservation Savings
Specific estimates of energy savings resulting from a conservation program
depend on the standard of operation before the program is instituted which var-
ies between various refineries, but general savings found are available. A re-
cent survey of 12 furnaces revamped for energy conservation shows average in-
crease in efficiency from 76% to 88%(13), while using preheated combustion air
from air-coolers could save 2% of the energy.(14)
Variable—pitch fans or multiple-fan units can lead to use of less than
half the design power consiinption required by most air-cooiers) General
savings for improved maintenance and increased heat exchange are very specific
to the installation, and no general figures are presented here.
Energy Conservation Drawbacks
The overriding drawback to all energy conservation measures is the in-
creased cost associated with additional equipment or hardware necessary and the
payout time to realize a cost saving. The price and availability of fuel is
the economic pointer in the selection of conservation measures, and most refin-
ers will use an expected payout time of 2-5 years or a discounted cash flow of
15% in the choice of possible options.
Other drawbacks encountered include temperature level of waste heat which
cannot be reused directly in the process,but may be utilized where a heat pump
is supplied to boost temperature levels of heat available.
Modified and Reconstructed Facilities
Where provision has not been made at the design stage to upgrade the ex-
tent of sulfur removal in the HDS process, the retrofit of an existing process
is extremely costly and often not considered economic in the refining indus-
try. Where provision for future upgrading is made, large equipment and pipe-
work are usually installed during initial construction, making the retrofit a
relatively simple procedure. The coninents for New Facilities would then apply
in this section.
-184-
-------
5.3 SUMMARY
The production of low-sulfur distillate and residual fuel oils by pre-
combustion treatment methods, such as the hydrodesulfurization (HDS) process
discussed in detail here,will have the advantage of scale from the energy im-
pact point of view, since the HDS process is centralized; and, advantage can
be taken of the large scale of utility consumption/production. The HDS pro-
cess, however, is a high—pressure, high-temperature process and, consequently,
the energy consumption to operate the equipment is extensive. The use of HDS
as an SO2 control technology on industrial boilers results in the expenditure
of 2-4% of the energy generated by the boilers. When the energy consumption
of the hydrogen plant is factored into a total desulfurization energy consumed,
the percentage of energy used increases substantially. The following table
shows the desulfurization energy expenditure as a percentage of total boiler
energy generated for the various sulfur control levels:
Energy Consumed As
Sulfur in A Percentage of
Fuel Oil Energy Generated
1.6% S 4.5%
0.8% S 5.6%
0.3% S 8.6%
0.1% S 10.8%
Table 5—2 shows energy consumption for the HDS and hydrogen plant untilities
as a function of desulfurization levels. It is apparent that, to achieve low
sulfur levels (0.3% S or 0.1% S) necessary for industrial boiler combustion with-
out controls, substantial energy inputs are needed as the degree of desulfuriza-
tion increases.
-185-
-------
TABLE 5-2. ENERGY CONSUMPTION FOR SULFUR LEVEL-OF-CONTROLS IN RESIDUAL OIL
Sulfur in Fuel Oil
Crude Class Utility 1.6% 0.3% 0.3% 0.1%
Low Sulfur Power 2.9 6.4 8,2 8.7
<3% S Steam 12.8 28.3 44.4 48.6
Fuel 107.7 262.9 360.0 398.7
Cooling Water .31 .63 1.1 1.2
Med. Sulfur Power 5.8 7.7 10.0 10.5
3—4% S Steam 31.0 43.8 63.8 71.2
Fuel 227.5 314.8 364.2 393.1
Cooling Water .65 .94 1.2 1.2
High Sulfur Power 7.0 8.9 10.7 11.0
>4% S Steam 37.4 49.2 79.6 84.4
Fuel 257.4 334.0 382.2 403.1
Cooling Water .75 .99 1.15 1.23
NOTE: Utility Units
Power in KWH/Bbl
Steam in 1 tu/Bbl
Fuel in MBtu/Bbl
Cooling Water in MGa1/Bbl
-186-
-------
REFERENCES
1. “Gas Processing Handbook, SCOT Tail-Gas Cleanup,” Hydrocarbon Processing ,
April 1975, P. 109.
2. “Refining Processes Handbook,” Hydrocarbon Processing , September 1978,
pp. 99-224.
3. “Gas Processes Handbook, ADIP Recovery System,” Hydrocarbon Processing ,
April 1975, p. 84.
4. “Proposed Standards of Performance for Petroleum Refinery Sulfur Recovery
Plants,” EPA—450/2-76-016(a).
5. Bridge, A. G., E. M. Reed, P. W. Tanun, and D. R. Cash, “Chevron Hydrotreating
Processes Desulfurize Arabian Heavy Residua, ” AIChE Symposium Series, No.
148, Vohtne 71, pp. 225-233.
6. Frayer, J. A., A. A. Montagna, and S. J. Yanik, “Gulf’s HDS Processes for
High Metal Stocks,” paper presented at the Japan Petroleum Institute,
Tokyo, Japan, 9 May 1975.
7. Paraskos, J. A., A. A. Montagna, and L. W. Brunn, “Ecologically—Acceptable
Fuels From the Gulf HDS Process.,” paper presented at the 67th Annual Meeting
of the American Institute of Chemical Engineers, December 1974.
8. Aalund, L. R., “U. S. Refining Industry Still Tied to Sweet Crude,” Oil and
Gas Journal , 10 October 1977, pp. 39-43.
9. Watkins, R. N.., “Petroleum Refinery Distillation,” Gulf, 1972.
10. Beychok, M. R., “Aqueous Wastes From Petroleum and Petrochemical Plants,”
Wiley, 1967.
11. A.S.T.M. Standards - 1973, D396.
12. Energy Technology Handbook , Douglas M. Considine, editor, McGraw-Hill Book
Company, New York, Mew York, 1977.
13. Uchida, Hiroshi, “HDS Unit Affects H 2 Plant Conditions,” Oil and Gas Journal ,
11 October 1976, pp. 126-137.
14. Duckham, H., and J. Fleming, “Better Plant Design Saves Energy,” Hydrocarbon
Processing , July 1976, pp. 78-84.
15. Fleming, J., H. Duckham, and J. Styslinger, “Recover Energy With Exchangers,”
Hydrocarbon Processing , July 1976, pp. 101-104.
-187-
-------
APPENDIX
1. Control Efficiency
Stringent level of control = 0.10 lbs. SO 2 per IVVI BTU
= 0.10% W/W sulfur
Typical medium sulfur content residual
= 3.8% W/W at 16.6° API (Kuwait)
0+
Typical HDS 375 F Product for Stringent Level
= 97.4% yield with 0.1% W/W sulfur at 25.0 API
Control Efficiency = sulfur removed per barrel x 100
total sulfur per barrel
= ( 0.038 x 0.96 x 42 x 8.33 - 0.001 x 0.90 x 42 x 8.33 x 0.974 )
0.038 x 0.96 x 42 x 8.33
= ( 12.76 - 0.31 ) x 100
12. 76
= 98%
2. Energy Const ned by Control Device
A. HDS Process Energy Consumption per Barrel Feed
Fuel - 84 M BTU
Power - 8.4 KWh
6001 Steam - 29 lbs.
50# Steam - 52 lbs.
B. Amine Process
H 2 S absorbed per barrel lIDS feed = (12.76 - 0.31) x 34
32
= 13.23 lbs.
Energy Constiuption:
Power = 0.025 KWh per lb. H 2 S absorbed
= 0.33 KWh/bbl HDS feed
501 Steam = 2 lb. per lb. H 2 S absorbed
= 26.5 lbs./bbl HDS feed
-------
C. Sulfur Plant and Tail Gas Scrubbing-Utility Consumption
Sulfur plant feed per barrel HDS feed = 13.23 x 32
34x2240
0.006 Long Tons
Energy Consumption:
Fuel = 1.78 MM BTU/Long Tori
= 9.9 M BTU/bbl HDS feed
Power = 88.8 KWh/Long Ton
0.49 KWh/bbl NDS feed
600# Steam 4,320 lbs/Long Ton produced
(24) lbs/bbl HDS feed
50# Steam = 1,800 lbs/Long Ton produced
(10) lbs/bbl HOS feed
D. Sour Water Stripper
Sour water produced from stripping steam used in
produced fractionation tower (steam rate = 10 lbs/bbl
product rate)
Water treated per barrel HDS feed
= 10 = 1.2 gal.
8.33
Energy Consumption:
50# Steam = 0.8 lbs/gal. treated
= 1.0 lbs/bbl HDS feed
Total Energy Consumed per barrel HDS feed
Fuel = 94 M Btu
Power = 9.2 KWh
600# Steam = 5 lbs.
50# Steam = 69 lbs.
-189-
-------
3. Increase in Energy use over Uncontrolled Boiler-Residual Fuel Oil
rate for 150 III Btu/hr. Boiler = 24.97 bbl/hr. of 25.00 API product
% Increases:
Fuel = 94 x 1000 x 24.97 x 100
150,000,000
= 1.6%
Power = 9.2 x 24.97 x 3412.2 x 100
150,000,000
= 0.53%
600# Steam = 5 x 24.97 x 903* x 100
150,000,000
= 0.08%
50# Steam = 69 x 24.97 x 911 x 100
150,000,000
= 1.0%
4. Change in Energy Use over SIP Controlled Boiler Residual Fuel Oil rate
for 150 MM Btu/hr. Boiler = 24.23 bbl/hr at SIP control level
% Increases:
Fuel = ( 94-76) x 1000 x 24.97 x 100
(150,000,000 + 76 x 1000 x 24.23)
= 0.3%
Power = ( 9.2-5.3) x 24.97 x 3412.2 x 100
( 150,000,000 + 5.3 x 24.23 x 3412.2 )
= 0.2%
600# Steam = ( 5-1) x 24.97 x 903 x 100
( 150,000,000 + 1 x 24.23 x 903 )
= 0.06%
50# Steam = ( 69-33) x 24.97 x 911 x 100
( 150,000,000 + 33 x 24.23 x 911 )
= 0.5%
* Enthalpy - 600 psig, 750°F steam = 1,378 BTU/lb.
- 600 psig saturated liquid = 475 BTU/lb.
Enthalpy - 50 psig, saturated steam = 1,179 BTU/lb.
50 psig, saturated liq 1 uid 268 BTU/lb.
-------
SECTION 6
ENVIRONMENTAL IMPACT OF DESULFURIZATION TECHNIQUES
FOR THE PRODUCTION OF LOW-SULFUR FUEL OIL
6.1 INTRODUCTION
This section deals with the multimedia environmental impacts associated
with the production of low-sulfur fuel oil in the petroleum refining industry.
Also included are the environmental impacts associated with the use of low—
sulfur fuel oils at the industrial boiler site. The basic intent of this sec-
tion is to identify the major environmental concerns, both beneficial and ad-
verse, at the boiler and at the refinery for air emissions, water emissions,
and solid waste emissions.
Fuel Oil Refinery
The fuel oil refinery as a basic processing sequence is defined by a yield
of product fuel oil between 40 and 60% of the liquid processed with the bulk of
the remainder of the products being gasoline. To identify the major environ-
mental concerns associated with producing a clean fuel, one must first explore
the basic fuel oil refinery processing scheme in order to understand the base
emissions associated with producing fuel oil. It is not within the scope of
this section to describe or define all the basic refinery processing steps, but
rather to compare the environmental impacts of burning uncleaned fuels with the
combined impacts of oil cleaning at the refinery plus burning cleaned fuels in
an industrial boiler. Once the basic waste stream characterization has been
made and the pollutants have been identified for the basic refinery producing
fuel oil, then the major environmental impacts can be determined for cleaning
the fuel oil from a normal high-sulfur concentration down to a concentration
which would meet stringent, moderate, or intermediate control requirements.
Then, the incremental environmental effects for this clean oil techno’ogy can
be assessed on a multimedia basis.
—191—
-------
Air Emissions
The major air pollution sources for particulates, SO, , carbon monoxide,
and flO within a fuel oil refinery are the process heaters and boilers. The
heaters and boilers used for these units are directly fired by refinery fuel
gas, heavy fuel oil, or coke gas. The major sources of hydrocarbon emissions
are general fugitive emissions throughout the refinery. The exact sources are
difficult to identify and even harder to quafltifyç 2 ) Another major source of
hydrocarbons are the crude and petroleum products storage tanks. Another pri-
mary pollution source is the tail gas from the acid gas treating plant. New
methods of controlling emissions can remove up to 99.8% of the hydrogen sul-
fide originally introduced to the acid gas piantS However, due to the rela-
tively large volume of the acid gas stream, the total emission of SO, to the
atmosphere after treating is still very substantial. Another air emission
source within the fuel oil refinery may be associated with the incineration of
sludge from wastewater treatment processing. If not incinerated, this sludge
would then represent a solid waste disposal problem.
Water Emissions -
Major sources of contaminated water within the fuel oil refinery are sour
water stripper condensate, process water, cooling tower blowdown, 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 the refin-
ery is located near a docking facility. The combined wastewater from these
sources is usually treated in a wastewater treating plant. Sanitary wastewater
is also generated within the refinery. This water is handled separately from
the contaminated water by a segregated wastewater system. The sanitary waste-
water, along with treated process wastewater, is retained in a holding pond
for a certain period of time before discharge. The wastewater from different
sources within the refinery are segregated and treated together in coim on equip-
ment based on the most cost effective treatment scheme.
-192—
-------
Solid Emissions
The major sources of solid wastes within the fuel oil refinery include en-
trained solids in the crude, silt from surface drainage, silt from water sup-
ply, corrosion products from process units and sewer systems, solids from main-
tenance and cleaning operations, sludge from water treatment facilities, and
spent catalysts. With the exception of spent catalysts, the generated solid
wastes are usually considered inert and acceptable for landfill. The solids
collected in the API separator and wastewater treating facilities represent
about 3/4 of the total solid wastes from a fuel oil ref inery 4 The balance is
from spent catalysts from hydrotreating and hydrodesulfurization units and
represents an average of the intermittent catalyst regenerations.
Environmental Effects
There are numerous beneficial environmental effects resulting from the
combustion of clean fuel at the industrial boiler. Simply stated, the combus-
tion of clean fuel oil at the industrial boiler will result in lower sulfur
oxide, NO)( and particulate emissions, as well as potentially—reduced solid
wastes.
Although no two refineries are exactly alike, the environmental impacts
presented in this chapter are based upon the production of low-sulfur fuel oil
in a fuel oil refinery. Table 6-1 sumarizes the multimedia environmental im-
pact at the fuel oil refinery for a base case of producing a residual fuel oil
having a sulfur content of 3.0% and a distillate fuel oil having a sulfur con-
tent of 0.5%. These emissions would then be representative of a fuel oil re-
finery which utilizes a processing sequence established to promote the yield
of product fuel oil. The fuel oil produced is considered to be approximately
50% by volume of the total liquid product with the remainder of the products
being gasoline. The remainder of this chapter will first define the incre-
mental adverse environmental effects of producing varying grades of clean oil
at the refinery. Secondly, these incremental adverse effects at the refinery
-193-
-------
TABLE 6-1. ENVIRONMENTAL IMPACT OF A FUEL OIL REFINER
PRODUCING 3.0% S RESIDUAL AND 0.5% S DISTILLATE OILS 4)
Air Emissions (lb./M BBL Crude Processed)
Particulates 63.2
SO, 160.0
NOx 118.3
CO 12.0
Hydrocarbons 740
Water (lb./M BBL Crude Processed)
Suspended Solids 2.5
Dissolved Solids 92.6
Organic Material 0.5
Solid Wastes (lb./M BBL Crude Processed)
Catalysts 20
Other* 60
Source : “Environmental Problem Definition for Petroleum
Refineries, Synthetic Natural Gas Plants, and
Liquefied Natural Gas Plants,” Radian Corp.,
November 1975, EPA-600/2-75-068.
* Includes: (1) entrained solids in the crude; (2) corro-
sion products; (3) silt from drainage and
influent water; (4) maintenance and cleaning
solids; and (5) waste water treatment facil-
ities.
-1 94—
-------
will be compared with the beneficial environmental effects of burning clean
fuel in an industrial boiler. This impact analysis will then be used to
determine the overall environmental benefit of cleaning fuel oil at the re-
finery followed by combustion of the fuel oil at the boiler.
Present Capability
Current refinery feedstocks permit production of 0.8 wt. % sulfur fuels
by indirect means. These processes involve desulfurizationof distillate
fuels or vacuum gas oil combined with back-blending to produce residual fuel
oils of acceptable sulfur content. Sulfur levels below this or higher sulfur
crudes will require the installation of direct desulfurization facilities to
meet lower sulfur limits. This is valid only for the domestic refiners, how-
ever. Offshore refiners are already faced with direct desulfurization of at-
mospheric residua in order to produce residual fuel oils of adequate quality
for some areas in the United States requiring 0.5 wt. sulfur residual fuel
oil or
6.2 AIR POLLUTION
The refinery airborne emissions of major consequence are generated by
four types of sources: combustion sources, process units, effluent control
systems, and storage facilities. Each of these source categories is discussed
below with emphasis on its contribution to the incremental environmental im-
pact of producing low-sulfur residual fuel oils.
Combustion Sources
Direct-fired process heaters and boilers are the major air pollution
sources within the refinery. The emissions from each of the n ny refinery
processing units which contain direct-fired heaters are dependent on the unit
heating requirement and on the fuel used. Total refinery heating denw*d is
approximately 4.5 x 1O 5 BTIJ/bbl of fuel oil produced. Specific unit heating
demands vary widely, and the specific fuels fired vary from heavy fuel oil to
-195-
-------
refinery fuel gas or mixtures thereof. The atmospheric emissions are highest
from large duty heaters firing heavy fuel oils.
In order to produce low-sulfur fuel oils, it is necessary to add hydrode-
sulfurization processing capacity. The additional processing heating require-
ments include: hydrodesulfurization, amine scrubbing, Claus plant, tail gas
treating, sour water stripping, and hydrogen plant. The total heating demand
for this processing ranges from 5 to 8 x l0 BTU/bbl of fuel oil depending on
the degree of desulfurization.
The environmental impact associated with combustion sources in the refin-
ery s clean oil processing is therefore only 1 to 1.5% greater than the base-
line refinery atmospheric emissions from direct-fired heaters presented in
Table 6-2. The emission rates of the primary pollutants tabulated in Table 6-2
are for a baseline fuel oil refinery producing 3.0% S residual fuel oil. The
sulfur oxide emissions are based on refinery sulfur balances, and the emission
quantities for particulate and N0 were determined based on emission factors.
Process Units
Emissions from the process units would be associated, for example, with
the regeneration of spent catalysts in the reformer isomeriser, hydrotreaters,
or hydrodesulfurization processes (HDS). The catalyst loses its activity due
to the accumulation of carbonaceous deposits and to the deposition of trace
metals. As the unit continues to operate, the pressure drop across the bed
builds up; and, eventually, the process must be shut down. The catalyst may be
regenerated by burning off the carbonaceous material or it may be replaced by
new catalyst. With a mild treatment, regeneration may be required at yearly
intervals; but, where treatment conditions are severe and with the older-type
catalysts, regeneration will be required at more frequent intervals.
Additional HDS capacity for producing low-sulfur fuels may result in some
minimal increase in particulate loading. However, this emission loading is
-196-
-------
TABLE 6—2. BASELINE FUEL OIL REFINERY ATMOSPHERIC
EMISSIONS (LB/M BBL CRUDE PROCESSED)
Crude Distillation
Gas Oil Hydrotreater
Naphtha Hydrotreater
Heavy Naphtha Hydrotreater
Propane Deasphal ting
Deasphalted Oil Hydrotreater
Tail Gas Treating
Light Ends Recovery
C 5 /C 6 Isomerization
Storage
Sludge Incineration
Miscellaneous Emissions
53
11
2
34
7
4
Source : HEnvirOnmental Problem Definition for Petroleum Refineries, Synthetic
Natural Gas Plants, and Liquefied Natural Gas Plants, Radian Corp-
oration, November 1975, EPA-600/2-75—068.
Sources
Parti-
Hydro-
culates
CO carbons
Total
31
3
64
5
5
1
5
1
20
5
——
41
7
——
34
3
1
——
3
1
1
- -
1
4
3
-—
6
--
-—
-—
1
--
——
119
1
609
2
——
5
--
160
12
740
118
64
-197-
-------
intermittent and only a minor contributor.
Effluent Control Systems
Emissions from the effluent control systems would be represented by stack
gas from the tail gas treatingsystem on the Claus plant. Conversion efficien-
cies in the Claus plant from 95% to 98% can be attained but will depend on the
hydrogen sulfide concentrations in the acid gas feed 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 Claus unit is
large and requires further treatment by tail gas treating unit. Some of the
tail gas units available are the Bevon (Union Oil of California), Clean Air
(J. F. Pritchard & Co.), I.F.P. process, Shell Claus off-gas treating - SCOT
(Shell Development Co.), Suifrene (SNPA-Lurgi - the R. M. Parsons Co.), and
W—L SO 2 recovery (Weliman-Power Gas, IflC.)c8) All of these units will increase
the Claus recovery of the equivalent sulfur in the tail gas to greater than
99.5%. The use of different units is determined by the characteristics of the
tail gas, the operating conditions, and the economics of the situation. Be-
cause of the high sulfur recovery efficiency realized in the tail gas treating
process, the sulfur oxides emissions representative of the production of low-
sulfur fuel oils are minimized; and, therefore, the relative adverse environ-
mental impact for sulfur oxides at the refinery is minimal when compared with
the direct emission of all sulfur contained in the fuel oil when combustion
takes place at the industrial boiler site.
Storage Facilities
Hydrocarbon emissions from the storage of heavy and light fuel oils are
negligible when compared with the hydrocarbon emissions from crude and gasoline
storage tanks. However, there may be a minor adverse environmental impact
resulting from the larger fraction of light ends produced in hydrodesulfuriza-
tion processing. These processing changes will have a negligible effect on
total hydrocarbon emissions from the combined refinery storage facilities.
-198-
-------
The storage emissions included in the baseline pollutant profile (Table 6-2)
are based on emission factors; however, general fugitive emissions included
under miscellaneous sources throughout the refinery are the major source of
hydrocarbon emissions. 2 With the increasing value of hydrocarbons, more
effective systems for controlling storage tank losses are becoming economically
feasible.
Overall Environmental Analysis
In Tables 6-3, 6-4, and 6-5, an approximation is given of the relative
adverse environmental effects of burning high sulfur residual oil in an indus-
trial boiler or of producing low-sulfur residual fuel oil and burning in the
same industrial boiler. The primary air pollutants (SOt , NO,( , and particulate)
emitted by the refinery cleaning processes are identified by the processing
sources needed for removing sulfur from the fuel oil. The estimated air emis-
sions at the refinery for these processing requirements are given in units which
may be directly compared with the air emissions expected at the industrial boil-
er firing the same fuel with no environmental controls.
Table 6-3 shows that without hydrodesulfurizatiOn (no controls), the emis-
sion levels for industrial botler combustion will lead to a significant deter-
ioration of air quality. Table 6-4 shows that, for moderate hydrodesulfuriza—
tion producing a 0.8% sulfur fuel oil, the incremental S0 emissions at the
refinery are estimated to be less than 4% of the SO emissions at the indus-
trial boiler firing residual fuel oil containing 0.8% wt. sulfur. Similarly,
Table 6-5 shows the relative air emissions for producing 0.1% sulfur residual
fuel oil and for burning this fuel in an industrial boiler.
In sumary, there is a significant overall improvement in air quality as
a result of burning lower sulfur residual oil fuel. The beneficial effect
of lower sulfur fuel is suninarized in Table 6-6.
-199-
-------
TABLE 6-3. ESTIMATED AIR EMISSIONS, 3.0% S FUEL OIL (NO CONTROL LEVEL)
LB/10 6 BTU FUEL OIL
6 SO ( 6 NOx Particulate
Emission Sources ( ib/lO Btu Fuel Oil) ( lbJlO Btu Fuel Oj]j ( lb/b 6 Btu Fuel Oil)
Refinery Processes
Combustion (heaters
and boilers) 0.020 0.019 0.010
- Gas Fired 0.005 0.001
- Oil Fired 0.014 0.009
lIDS
Tail Gas
Treating 0.003
Subtotal 0.023 0.019 0.010
Industrial Boiler Combustion
Residual Fuel Oil
(Mo Control
level) 3.170 0.40 0.22
Total 3.193 0.419 0.230
-200-
-------
TABLE 6-4. ESTIMATED AIR EMISSIONS, 0.8% S FUEL OIL (MODERATE CONTROL LEVEL)
LB/b 6 BTU FUEL OIL
- SO 6 NOx Particulate
Emission Sources ( lb/lOb Btu Fuel Oil) ( lb/b Btu Fuel OilJ ( lb/10 6 Btu Fuel Oil)
Refinery. Processes
Combustion (heaters
and boilers) .020 .019 .010
- Gas Fired .005 .001
- Oil Fired .014 .009
liDS .005
Tail Gas Treating!
Claus Plant . 005
Subtotal 0.030 0.019 0.010
Lndustrial Boiler Combustion
Residual Fuel Oil
(moderate control
level) 0.85 0.3 0.12
Total 0.88 0.32 0.13
-201-
-------
TABLE 6-5. ESTIMATED AIR EMISSI NS, 0.1% S FUEL OIL (STRINGENT CONTROL)
LB/lO BTU FUEL OIL
SO NOx Particulate
Emission Sources ( lb/b 6 Btu Fuel Oij) ( lbJlO 6 Btu Fuel Oil) ( lb/b 6 Btu Fuel Oil)
Ref iner Processes
Combustion (heaters
and boIlers) 0.020 0.019 0.010
— Gas Fired 0.005 0.001
- Oil Fired 0.014 0.009
HDS 0.015
Tail Gas Treating!
Claus Plant 0.015
Subtotal 0.050 0.019 0.010
Industrial Boiler Combustion
Residual Fuel Oil
(stringent control
level) 0.11 0.11 0.05
Total 0.16 0.13 0.06
-202-
-------
TABLE 6-6. ESTIMATED AIR EMISSIONS - RESIDUAL FUEL OIL COMBUSTION
Fuel SOX NOx Particulate
Description ( LB/b 6 BTU) ( LB/b 6 BTU) ( LB/b 0 BTU )
3.0% Sulfur 3.18 0.40 0.22
1.0% Sulfur 1.06 0.33 0.14
0.8% Sulfur .85 0.30 0.12
0.5% Sulfur .53 0.25 0.09
0.2% Sulfur .21 0.20 0.05
0.1% Sulfur .11 0.18 0.03
-203-
-------
6.3 WATER POLLUTION
The major use of water in petroleum refining is for steam generation and
heat transfer. The volume of water coming in direct contact with process streams
is small when compared with the water for indirect cooling and heat transfer.
Nevertheless, nearly every major refining operation produces a wastewater stream
containing various pollutants.
Major sources of contaminated water within the fuel oil refinery are sour
water stripper condensate, contaminated process water, cooling tower blowdown,
and desalter water. Each of these major sources is discussed below. Other
potential contaminated water sources are oily process area storm water, oily
cleaning water, and oily water from a ship’s ballast, if the refinery is locat-
ed near a docking facility. The contined wastewater from these sources is
usually treated in a wastewater treating plant. 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 cer-
tain period of time before discharge.
It is a typical practice in most refineriEs to collect all contaminated
process wastewater and to conthine it into a single wastewater stream and then to
treat it in a central treatment facility. As a result of this, it is unnecessary
to deal with the volimie and characteristics of each of the component wastewater
streams. Therefore, in assessing the water pollution implications of alterations
to the base case refinery, it is not necessary to examine the individual waste-
water sources. Rather, the volume and characteristics of the total refinery
wastewater effluent will be examined. Table 6-7 gives the refinery effluent
based upon 15 gallons of water required per barrel of crude feed.
Water Management
Water management at the refinery can exercise a number of strategies by
direct inçlementation of several wastewater treatment processes. There are
-204-
-------
TABLE 6-7. REFINERY WASTEWATER EFFLUENT QUALITY FOR
3 x 106 GAL/DAY (200,000 BBL/DAY CRUDE FEED)
Concentration
BOD 15 ppm
COD 80 ppm
Ammonia 2 ppm
Hydrogen Sulfide 0.1 ppm
Total Phosphorus 2 ppm
Phenols 0.1 ppm
Oil and Grease 2 ppm
Suspended Solids 10 ppm
Dissolved Solids 370 ppm
Source : “Environmental Problem Definition for Petroleum Refineries,
Synthetic Natural Gas Plants, and Liquified Natural Gas
Plants,” Radian Corporation, November 1975, EPA-600/2-75-068.
—205—
-------
four types of wastewater treatments applicable at the refinery: in-plant, pri-
mary, 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
desired. Table 6-8 shows a water management plan for a refinery and emphasizes
some of the ways in which the wastewaters from various processes are segregated.
Sour Water Stripper Condensate
Sour or acid waters are produced in a refinery when steam is used as a
stripping medium in the various cracking processes. The hydrogen sulfide,
alTnnonia, and phenols distribute themselves between the water and hydrocarbon
phases in the condensate. The concentrations of these pollutants in the water
vary widely depending on crude sources and processing involved.
The purpose of the treatment of sour water is to remove sulfides (as hydro-
gen sulfide, aninonium su1fide, and polysulfides) before the waste enters the
sewer. The sour water can be treated by: stripping with steam or flue gas;
air oxidation to convert hydrogen sulfide to thiosulfates; or varporization and
incineration. Due to the nature of the hydrodesulfurization process, the vol-
ume of direct contact process water discharged from the fuel oil refinery will
not be significantly increased by producing lower sulfur fuels. The two main
pollutants formed in hydrodesulfurization (hydrogen sulfide and ammonia) are
both soluble in water and could be carried out with any wastewater streams. It
is expected that these would be treated in the normal refinery ammonia and sour
water strippers and that their contribution to the environmental impact can be
estimated based on normal removal efficiencies of sulfide and ammonia using a
well-controlled water management system.
Sour water strippers are designed primarily for the removal of sulfides
and can be expected to achieve 85 to 99% removal. If acid is not required to
enhance sulfide stripping, ammonia will also be stripped with the percentage
varying widely with stripping temperature and pH. If acid is added to the
-206-
-------
TABLE 6-8. REFINERY WATER MANAGEMENT PLAN
OILY WASTES
CONTAM I HATED
WITH H 2 S
SEWAGE
SOUR
CONDENSATE
NON-OILY
WASTES
WASTE WATER
SOURCES
IN- PLANT
TREATMENT
I SOUR WATER L
STRIPPER _ f
PRIMARY
TREATMENT
API SEPARATOR,
DISSOLVED AIR
FLOTATION
J SOLIDS
I DEWATERING
SECO I DARY
TREATMENT
TERTIARY
TREATMENT
Source : Industrial Wastewater Management Handbook, H. S. Azad, IVS Corporation,
i’lcGraw-Hill Book Coripany, 1976, pp. 8-1 to C-74.
OILY WASTES
PRECIPITATION
RUNOFF
CRUDE DESALTER
PENTANE DEASPHALT
MISC. OPERATIONS
K
ATM DISTILLATION
DEASPHALTED OIL
HDS
PARTIAL OXIDATION
HYDROCRACKUIG
FLUID CAT. CRACK-
I HG
HF ALKYLATION
I SAN ITARY !rEI
J BIOLOGICALI
¶OXIDATION j
REF I NERY
EFFLUENT
COOLING
TOWER
R I ()WDOWtI
SULFUR REC
SOUR WATER
STRIPPER
STEAM GENERATION
POWER GENERATION
PRECIPITATION
RUNOFF
LIME SOFTENER
SODIUM Z ULITE
UNIT
IOU EXCHANGE
UNIT
COOLING
TOWER
DESALTED
WATER
(FOR
REUSE)
DESALINATION
PROCESS
F
( NEUTRALIZATION
TREATME 1T
WASTES
-------
wastewater, essentially none of the amonia will be removed. Thus, arnonia re-
movals in sour water strippers vary from 0 to 99%!:9)
Depending upon such conditions as wastewater pH, temperature, and contami-
nant partial pressure, phenols and cyanides can also be stripped with removal as
high as The bottoms from the stripper usually go to the desalter where
most of the phenols are extracted, and the wastewater can be sent to the regular
process water treating plant. Chemical oxygen demand (COD) and biochemical oxy-
gen demand (BOD) are reduced because of the stripping out of phenol and oxidiz-
able sulfur compounds.
Contaminated Process Water
Petroleum refinery wastewaters vary in quantity and quality depending on
the refinery. However, the wastes are readily treatable. The processes used
for treating refinery wastewater are designed to maximize oil recovery and min-
imize the discharge of other pollutants. Wastewater will be generated at multi-
ple sources in the refinery. The primary contaminants present in the refinery’s
wastes include sulfides, aninonia, phenols, oil, dissolved and suspended solids,
BOD and COD.
Tank farms used for the storage of refinery products produce wastewater streams
due to storm water runoff contacting petroleum contaminated exposed areas.
Storm water runoff from process areas is another significant source of waste-
water. Within the confines of the refinery itself, there are numerous process-
ing steps in which steam, condensate, or cooling water come in contact with pe-
troleum or petroleum products. In addition to these major wastewater generation
points, there are countless leaks and spills which eventually drain into the
central refinery sewer system.
Since petroleum and petroleum products are the major source of pollutants
in the refinery wastewater, it is not surprising that chemical constituents
found in the petroleum appear in the wastewater. Raw refinery wastewater con-
tains large quantities of oil. The oil is present both as free oil in a float-
able form and as emulsified oil. In addition, water-soluble organics, such as
-208-
-------
phenolic compounds which are present in the petroleum, will also be present in
the wastewater from caustic scrubbing. Crude petroleum contains a variety of
sulfur compounds which are removed from the finished product in varying degrees
depending on product specifications.
For lower sulfur fuel oils, more of the sulfur compounds must be removed.
Due to the oil/water contacting at various stages in the refinery operation,
a significant quantity of sulfur compounds enters the wastewater stream.C
The most objectionable of these sulfur compounds are sulfides which are typi-
cally present in the wastewater as sulfide ions. Petroleum also contains a
nunter of nitrogen-bearing compounds; and, therefore, refinery wastewater is
typically contaminated with appreciable quantities of anuiionia. Small amounts
of cyanide compounds may also be expected to be present Carbonaceous and
inorganic particulate matter from a variety of sources are also present in
refinery wastewater, thus contributing to the suspended solids level. Typical
sources for the suspended solids emissions are incomplete combustion, soil,
and the like. Most of the organic chemical compounds mentioned are oxidizable;
and, therefore, refinery wastewater will exert a chemical oxygen demand. A
certain fraction of the same compounds are biodegradable; and, therefore, ref in-
ery wastewater will also exert a biochemical oxygen demand. Petroleum also
contains a variety of trace heavy metals; for example, mercury, cadmium, lead,
and so forth, which vary greatly from crude to crude and have not been exten-
sively quantified.
Hydrotreating processes in the refinery do not have any known effect on
the quantity or quality of contaminated process water, except if flexicoking
or delayed coking is used rather than direct hydrodesulfurization of the resi-
dua. The only adverse environmental effect is from the process water contacting
the coke produced in flexicoking or delayed coking. The flexicoker produces a
water purge from the Stretford process which incrementally increases the efflu-
ent wastewater load.
-209—
-------
Cooling Tower Blowdown
All petroleum refineries use large amounts of non-contact cooling water.
Due to the large volumes required, a cooling tower system is generally used
to reuse most of the water. To prevent the buildup of naturally-occurring
salts in the cooling water system, it is necessary to purge a portion of the
total cooling water flow. Since it is a comon practice to use corrosion in-
hibitors in the cooling circuit, cooling tower blowdown will contain such sub-
stances.
Depending on the type and quantity of corrosion inhibitor used in the non-
contact cooling water circuit, the cooling water blowdown can present pollu-
tion problems of varying significance. The most coim on types of corrosion in-
hibitors contain chromate salts; thus, blowdown from cooling towers can often
contain chromium in the form of hexevalent chromium, the more objectionable
fonnP0) In the petroleum refining industry, there has been a continuing trend
toward wastewater volume reduction to meet the zero discharge effluent guide-
lines. Large modern refineries generate far less wastewater than small old
refineries because process improvements have decreased the volume of wastewater
that must be treated. Improved leak and spill management, coupled with other
preventive measures, has also contributed to the overall volume reduction.
There has also been a trend toward tighter, non-contact cooling water circuits,
in addition to more widespread use of air cooling.
The cooling water blowdown stream may be increased from 4 to 7% over that
of the base case refinery due to the additional heat removal requirements asso-
ciated with clean oil technologies such as direct hydrodesulfurizatiOn and
Crude Desalter Water
Comon to all types of desalting are an emulsifier and settling tank.
Salts can be separated from oil by either of two methods. In the first method,
water wash desalting in the presence of chemicals (specific to the type of
-210-
-------
salts present and the nature of the crude oil) is followed by heating and gra-
vity separation.
The continuous wastewater stream from a desalter contains emulsified, and
occasionally free oil, amonia, phenol, sulfides, and suspended solids. These
pollutants produce a relatively high BOD 5 and COD. This wastewater also con-
tains enough chlorides and other dissolved materials to contribute to the dis-
solved solids problem in the areas where the wastewater is discharged to fresh
water bodies. There are also potential thermal pollution problems because the
temperature of the desalting wastewater often exceeds 95°C (200°F).
However, hydrotreating processes make no further adverse environmental
impact to the base refinery case.
6.4 SOLID WASTE
A petroleum refinery generates a wide variety of solid waste streams, many
of which contain materials on the EPA toxic substances list. The nature and
quantity of solid wastes eminating from refineries are highly variable and still
the subject of much investigation. Basically, refinery solid waste streams
fall into two main groups: those that are intermittently generated and those
that are continuously generated.
Intermittent Wastes
Intermittent wastes are generally those that result from cleaning within
the process areas and off-site facilities of the refinery. Typical intermittent
waste streams may be processing vessel sludges, vessel scale, and other deposits
generally removed during plant turnarounds, storage tank sediments, and product
treatment wastes such as spent filter clay and spent catalysts from certain pro-
cessing units such as hydrodesulfurization. The annual volume of refinery inter-
mittant wastes is strongly a function of the individual refinery waste management
and housekeeping practices.
-211-
-------
Continuous Wastes
Continous wastes, those that require disposal at least at two-week inter-
vals, can be further broken down into two groups: process unit wastes and
wastewater treatment wastes. Major process unit wastes include coker wastes,
such as coke fines from fluidized cokers and spilled coke from unloading facil-
ities, spent catalysts and catalyst fines from the fluid catalytic cracking
units, and other spent or spilled wastes from processing plants. Water treat-
ment wastes may include biological sludges from activated sludge units and dis-
solved air flotation “float” (DAF). Wastes from water treatment are generally
dewatered by means of sludge thickeners, coupled with vacuum filters or centri-
fuges. The dewatered sludge can then either be land disposed or incinerated.
Low concentrations of heavy metals are usually present in the sludges, which
could affect the level of control required. There are many different types of
refinery solid wastes, and there are many levels of technology for treatment
and disposal of these petroleum refinery wastes.
The sludges present from washdown or processing operations require com-
pliance with RCRA standards for toxic wastes, since they can contain trace
ameunts of nickel, vanadium, nickel carbonyl (or cobalt carbonyl depending on
catalyst selection). Other compounds have been found in “floats” at levels
which make its disposal complex: phenols, sulfides, phosphorus, amonia, etc.
Although incineration is the most desirable disposal alternative, the presence
of heavy metals requires treatment prior to incineration and may make encap-
sulation/land disposal the viable disposal method. The existence of RCRA
standards, the cost of land, and disposal operations costs place an additional
penalty on the cost of desulfurization. Table 6-9 estimates emissions created
durinq oily sludge incineration.
Wastes From Clean Oil Technology
There are only two individual refinery solid waste streams which have ad-
verse environmental impacts at the refinery due to the production of low-sulfur
-212-
-------
TABLE 6-9. SLUDGE INCINERATION EMISSIONS
Emission
Pollutant ( lb/boo Gal Sludge )
Particulate 23
SO 2 47
CO 4
Hydrocarbons 3
40
Source : “Environmental Problem Definition for Petroleum
Refineries, Synthetic Natural Gas Plants, and
Liquefied Natural Gas Plants,” Radian Corpora-
tion, November 1975, EPA-600/2-75-068.
-213-
-------
fuel oils. The worst of these individual solid waste streams is the increased
amount of spent desulfurization catalysts which is sent to disposal. The second
solid waste stream of concern is the increased waste from hydrogen manufactur-
ing, due to the fact that additional hydrogen is needed to reduce the sulfur
level in the product fuel oil.
A number of refinery processes require the use of a fixed-bed catalyst.
These processes include catalytic reforming, hydrodesulfurization, hydrotreat-
ing, hydrocracking, steam hydrocarbon reforming for hydrogen production, sul-
fur production from H 2 S and/or SO 2 , and others. These catalysts eventually
become inactive and are replaced in the reactors with fresh catalysts during a
unit shutdown. Catalyst life may extend from six months to as long as three
years. Many of these catalysts contain valuable metals which can be recovered
economically. Some of these metals, such as platinum and palladium, represent
the active catalytic component. Others are contaminants in the feed, which are
adsorbed on the catalyst during use. Usually, the more valuable metals are re-
covered by an outside company before the spent catalysts are disposed of as
solid wastes by these companies. In addition to hydrodesulfurization and de-
metallization catalysts, additional sulfur removal from the product fuel oil
will produce an increase in the amount of spent Claus catalyst to be disposed.
The disposal of hydrodesulfurization spent catalysts is clearly the larg-
est solid waste problem associated with hydrotreating; however, an estimate of
finite amounts of contamination from catalyst disposal is not available from
the literature surveyed. This impact study is limited to the point at which
the spent catalyst quantity emanating from the refinery has been identified
as a solid waste. The impact of spent catalyst use is the subject of an exist-
ing EPA contract on Environmental Assessment for Residual Oil Utilization) 2)
-214-
-------
Without going into specific disposal techniques, several general statements
can be made about the HOS catalyst use for the U. S. Based upon 600,000 BPSD
Residual and VGO hydrodesulfurization capacity, an estimation of total nation-
wide spent catalyst can be made. The following table estimates the total
possible spent catalyst generation for the 600,000 BPSD HDS processes:
Spent Catalyst Generation for Sulfur
Control Levels in Tons/Yr
0.8% S LSFO 0.3% S LSFO 0.1% S LSFO
1. Low Sulfur,
High Metals 20,000 64,000 115,000
2. High Sulfur,
High Metals 38,000 69,000 125,000
3. Low Sulfur,
Low Metals 8,000 11,000 13,000
4. High Sulfur,
Low Metals 17,000 29,000 38,000
Since most off-site or in—situ regeneration techniques recover 90—96% of the
catalyst activity, the estimated make-up amounts equal the catalyst requiring
ultimate disposal. Thus, for.a worst-case scenario, 10% catalyst loss per
year would be a 12,500 tons per year disposal amount. This closely corre-
lates with an estimated market of 10,000-12,000 tons/yr for HDS catalysts
made by catalyst manufacturers and regeneration firmsS 13)
The rapid growth of HOS processes has increased the demand for hydrogen
beyond the level of byproduct H 2 available from steam reforming and other pro-
cesses. Refinery off-gases are utilized as a charge stock for H 2 production
by passing the gas over a three-bed desulfurization unit containing a first—
stage zinc oxide, second-stage cobalt/moly catalyst, followed by a third-stage
zinc oxide bed. The spent zinc oxide and cobalt/moly catalysts then bec a
solid waste due to HDS requiring additional hydrogen.
—215—
-------
Catalyst disposal presents much the same problem as the sludge or “float
wastes previously mentioned except that its relative toxicity is far greater,
since much of the material is heavy metals. The effect of the metallic nature
for disposal means that very stringent RCRA standards will be applied to any
disposal operation. The rising costs for nickel, cobalt, and molybdenum make
in-situ or off-site regeneration more cost-effective alternatives. The oper-
ational cost for an RCRA quality disposal site for these contaminated catalysts
could be in the $30-50/ton range.
The most coninon regeneration method is steam-air application at tempera-
tures approaching 900°F in order to burn off the carbon/coke build-up. Gener-
ally, a single burn with a steam-air mixture will reduce the residual carbon
from 12-15% by weight to 1.5-3.0% by weight. If lower residual carbon levels
are required by the HDS operation, at least two or three modest increases in
inlet temperature and oxygen levels should be made to see if any secondary com-
bustion can be achieved. Steam rates should -be maximized for the best possible
flow distribution, and an indicator of steam flow is usually about 1 lb steam!
hour! lb catalyst. Inlet oxygen concentrations should not exceed 2%, since
hitting a pocket of previously-untouched catalyst could trigger runaway burn-
ing. After regeneration, the cooling medium should not be switched to air Un-
til the catalyst is cooled to about 400°FJ 4
Both off—site andin-situ regeneration are receiving rapidly-growing accept-
ance for several reasons:
(1) Quality - The catalyst suffers severe activity loss due to
the action of the steam. In the case of nitrogen-air in-
situ regeneration, unscreened catalyst will have poor flow
distribution; and, the resulting localized high temperatures
destroy catalyst activity.
-216-
-------
(2) Cost - Off-site and in-situ regeneration require minimal
catalyst replacement; thus, the savings due to regenera-
tion from a large refinery complex can be considerable if
the alternative is to throw away the catalyst. In-situ
regeneration causes extensive unit downtime unless multi-
ple parallel units exist, which increases initial capital
outlay. If approximately 15 days are needed for in-situ
regeneration, the lost downtime could easily approach
$1,500,000. Although off-site regeneration minimizes
plant downtime, additional replacement catalyst is requir-
ed. At a rate of $1.80/lb for catalyst and using a 75,000
barrel/day atmospheric residual HDS unit requiring
.4,000,000 lbs catalyst, the inventory cost of catalyst
needed for off-site regeneration is $7,200,000. As pre-
viously stated, multiple units would allow for regenera-
tion while minimizing the need for spare catalyst; but,
capital expenditures for the HDS facility would increase
due to multiple equipment requirements.
Using 25,000 barrels/day/desulfurization train as
a guideline, the three trains required for the 75,000
BPSD refinery would allow for off-site or in-situ regen-
eration to be performed with minimal downtime. By rotat-
ing regeneration sequence, the actual monetary loss from
such downtime is considerably smaller than complete shut-
down or catalyst replacement expenses.
0ff-site regeneration charges of 45-55 /lb are ap-
proximately 25-30% of new catalyst cost. Thus, the sav-
ings for the refiner are in the $l.25-$l.35/lb
-217-
-------
neighborhood, (15) which makes regeneration worth examina-
tion when compared to complete catalyst disposal.
(3) Pollution - Off-site regeneration facilities are equipped
to control air and water pollution problems. Opting for
off-site regeneration removes a pollution problem from
the refiner, since unusable catalyst disposal is no longer
his responsibility. In addition, the construction of ex-
tra pollution abatement facilities in the refinery, where
they will be used only during infrequent catalyst regener-
ation, is not economically sound.
Although this discussion centers on regeneration for removal of carbonf
coke build-up, the deposition of metals in the catalyst pores from HDS process-
ing of a high metals content residual enhances the need for restoring catalyst
activity periodically and for recovery of saleable metals. With the HDS pro-
cessing activity increase in recent years, a substantial market has developed
for metals recovery. Several firms now provide off-site metals reclamation
services. Higher prices for cobalt and nickel-tungsten catalysts provide a
credit for the IIDS processor in that the amount of metal recovered offsets
the cost of replacement catalyst. However, the off-site recovery operation
means that refiners will have to carry extra catalyst inventory, which, in
the case of large HDS units, could amount to substantial capital outlay, as
previously discussed in the case of off-site regeneration. The same off-site
regeneration cost figures apply for metals recovery, but metals-value credits
will help reduce the regeneration cost.
Factors Affecting Solid Waste
There are several important factors affecting solid waste generation at
the refinery. The type of crude stock is one of the most important. The con-
stituents of crude oil can vary widely. The heavy metal content, for example,
-218-
-------
is of major importance in determining the hazardous or potentially-hazardous
metal content in crude oil storage tank bottoms, in spent catalysts, and in the
various wastewater treatment plant sludges. It is therefore reasonable to
expect that solid wastes will contain different concentrations of potentially-
hazardous materials and that such differences may even be reflected in the
solid waste loads of two refineries of equal capacity which produce the same
products but utilize different crude mixes.
A second factor affecting solid waste generation is the variation of pro-
cess types; for example, differences in wastewater and air pollution control
processes will affect the quantity as well as the composition of potentially-
hazardous waste material. There are differences, also, in the degree and type
of wastewater treatment processes employed by refineries. A refinery using an
extended aeration sludge activation system will generate smaller quantities of
biological sludges than will a refinery which utilizes a conventional activated
sludge system. Refineries using only primary wastewater treatment before dis-
charging into a municipal treating system do not generate the biological
sludges which are associated with secondary treatment.
A third factor affecting solid waste generation is the age of the process-
es in the refinery. Process age refers to the general technology used in the
process rather than to the length of time the process has been in service.
This technology includes methods that will increase or decrease the quantity
of solid waste. For example, the use of air instead of water cooling will re-
duce or eliminate a solid waste problem from the cooling tower sludges. 6
6.5 OTHER ENVIRONMENTAL IMPACTS
Most refineries generate fairly high noise levels within the battery limits
because of equipment such as pumps, compressors, steam jets, flare stacks, and
so forth. Any equipment or equipment changes associated with the production
of low-sulfur fuel oil or associated with in-process control systems would not
-219—
-------
significantly add to these noise levels. There are many practical industrial
approaches to controlling refinery noise. However, the cost of noise abate-
ment in the overall refinery pollution abatement program would be a substantial
item. (17)(18)
-220-
-------
REFERENCES
1. Reed, E. M., et al. , “HDS Goes Deeper Into Barrel Bottom,” Oil and Gas Journal ,
17 July 1972, pp. 103—108.
2. “Statistical Analysis of Fugitive Emission Changes Due to Refinery Expansion,”
Radian Corporation, September 1978, EPA-600/2-78-195.
3. “Shell Claus Off—Gas Treating (SCOT),” Gas Processing Handbook, Hydrocarbon
Processing , April 1975, p. 109.
4. “Environmental Problem Definition for Petroleum Refineries, Synthetic Natural
Gas Plants and Liquified Natural Gas Plants,” Radian Corporation, November
1975, EPA—600/2—75-068.
5. Laengrich, Arthur R., “Tail Gas Cleanup Addition May Solve Sulfur-Plant Com-
pliance Problem,” Oil and Gas Journal , 27 March 1978, pp. 159—162.
6. Jimeson, R., et al., AIChE Symposium Series, “Census of Oil DesulfurizatiOn
To Achieve Environmental Goals,” Volume 71, No. 148, pp. 199-215.
7. Aalund, L. R. , “U. S. Refining industry Still Tied to Sweet Crude,” Oil and
Gas Journal , 10 October 1977, pp. 39-43.
8. “Gas Processing Handbook,” Hydrocarbon Processing , April 1975, pp. 107-111.
9. “Development Document for Proposed Effluent Limitations Guidelines and New
Source Performance Standards for the Petroleum Refining Point Source Category,”
ii. S. Environmental Protection Agency, Washington, D. C., EPA-440/1-73-014,
September 1973.
10. Azad, H. S., Industrial Wastewater Management Handbook , NVS Corporation,
McGraw-Hill Book Company, 1976, pp. 8-1 to 8-74.
11. Griffel, J., “Flexicoking Clean Products From Dirty Feeds,” 13 May 1976,
Exxon Research and Engineering Company.
12. Tyndall, M. F., et al., “Environmental Assessment for Residual Oil Utilization,”
Second Annual Report, Catalytic, Inc., EPA-600/7—78—l75, September 1978.
13. Bruch, W. R., “How Refineries Work With HDS Catalysts,” Oil and Gas Journal ,
14 June 1976, pp. 83-86.
14. Burke, Donald R., “Catalysts,” Chemical Week , 28 March 1979, pp. 42-54.
15. “Better Way to Give Catalysts a Facelift,” Chemical Week , 21 February 1979,
pp. 53-55.
16. Fleming, James, “Recover Energy With Exchangers,” Pullman-Kellogg Company,
Hydrocarbon Processing , July 1976, p. 101.
17. Seebold, J. F., “Control Plant Noise This Way,” Standard Oil of California,
Hydrocarbon Processing , August 1975, pp. 80-82.
18. Robinson, G., “Assess and Control HPI Noise,” Gulf Oil Refining, Ltd.,
Hydrocarbon Processing , June 1977, pp. 223-226.
—221-
-------
EMISSION SOURCE TEST DATA
7.1 INTRODUCTION
This section contains a sumary of the available emission source test data
gathered during the development of the proposed industrial boiler standards.
Included are descriptions of the facilities tested, identification of the test
methods, and the data obtained.
The majority of the data available for oil-fired industrial boiler emis-
sions was found in References 1-3, which document a study of pollution control
by conthustion modification, with emphasis on NO control. The test methods
used were selected according to the pollutant being measured. Particulate emis-
sions were measured by EPA Reference Method 5, SO by wet chemical analysis us-
ing the ‘ She1l-Emeryvi11e” method, and NO by a chemiluminescent nitric oxide
analyzer. Opacity was measured in some cases, using Bacharach Smoke Spots.
The test facilities were oil—fired industrial boilers with capacities less
than 250 MBtu/hr.
Some particulate emission data were also available from Reference 5, a
study of particulate emission control systems for oil-fired boilers.
7.2 TEST RESULTS
The available source test data for SO, , NOR , and particulate emissions
fron oil-fired industrial boilers are presented in Tables 7-1, 7-2, and 7-3,
respectively. Most of the control methods indicated refer to contustion modi-
fication for N0 control. Baseline refers to boiler runs at approximately
80% of rated capacity, with no emission control techniques employed. Only one
set of data is available for each set of conditions at each location, so the
single values are listed under the “average” column. Data on the duration of
the test runs were not available. Percent control and design control efficiency
are not meaningful, since the boilers are uncontrolled. A value for percent
control could be calculated, however, based on the emission of all of the fuel
-222-
-------
TABLE 7—1 . EMISSION SOURCE TEST DATA - SO,
Fuel Characteristics No. or Emissions % Control Design
Act al Our— (ib/F iBtu) Based on Control Control
E oiinr Heat atlon Longest ( ng/J ) Average Effi- Levels Reference
Size Test Lnad ontro1 Value of Cont. of ciency Sup- Unit 1.0. Test
i Li L r. lO ’b/ tiethod Ash Test Method Te ts Duration_ow Huh y e All Tests of Device rted* & Location
110 88 Baseline 19,340 .18 .004 Shell Emeryville 1 No Data .195 jo Data No Data S Unit T-8; Loc. 17 7-10 1
80 51 Baseline 18,800 1.49 .019 1 .858 SIP Unit 4; Loc. 20 8—5 1
90 71 Baseline 18,420 1.04 .031 1 .863 SIP Unit 2; Loc. 18 9-1 1
65 54 Baseline & 18,930 -— .043 1 .869 SIP Unit 2; Loc. 16 10-2 1
Hi Load
105 80 Baseline 18,910 1.03 .032 1 .902 SIP Unit 3; Loc. 18 21-6 1
160 130 Baseline 18,910 1.03 .032 1 .969 SIP Unit 4; Loc. 18 22-1 1
500 400 Baseline 18,330 2.43 .12 1 2.68 -- Unit 2; bc. 13 29-5 1
10 7 Baseline & -— .23 —- 1 .159 S Unit 2; Loc. 3 33-3 1
Hi Load
7 6.7 Baseline & 18,280 1.85 .038 1 2.01 -- Unit 1; Loc. 23 34-11 1
s Hi Load
14 Baseline 19,470 .06 .001 1 .111 5 Unit 1; Loc. 19 52—5 1
14 Hi Air 19,470 .06 .001 1 .123 S Unit 1; Loc. 19 53-6 1
20 16 Baseline -— .30 —- 1 .198 I Unit 4; Loc. 4 59-6 1
150 115 Baseline -- -— .005 1 .168 S Unit 3; Loc. 6 65-1 1
33 23 Baseline 19,440 .40 .003 1 1 .198 S Unit 3; Loc. 1 66—1 1
17.5 14.5 Baseline 19,680 .16 00l 1 .180 S Loc. 19 19-5 3
17.5 14.5 Low 02 19,610 .14 .001 1 .238 I Loc. 19 19-74 3
17.5 14.5 FOR; Low 02 19,610 .14 .001 1 .177 S Loc. 19 19-83 3
17.5 14.5 Staged Air 19,610 .14 .001 1 .159 S Loc. 19 19-116 3
17.5 14.3 FGR & SA 19,610 .14 .001 1 .200 S Loc. 19 19-179 3
LowO
17.5 13.8 Baselin 19 000 .54 .019 1 580 M Loc 19 19-97 3
17.5 14.2 Low 02 18,780 .60 .034 1 .744 M Loc. 19 19-132 3
17.5 13.8 stagea Air 18,780 .60 .034 1 .660 M Loc. 19 19-143 3
17.5 14.0 Max. FOR 18,780 .60 .034 1 .627 M Loc. 19 19-159 3
17.6 14.3 FOR &SA 18,780 .60 .034 1 .665 M Loc. 19 19-170 3
45 38.0 Baseline 18,467 1.88 .05 1 1.77 -- Loc. 33 200—24 3
45 38.8 Low 02 18,467 1.88 .05 1 1.75 -- Loc. 38 201-12 3
45 38 Stagea Comb. 18,467 1.88 .05 1 1.86 -- Loc. 38 203-26A 3
Al r
45 38.8 Low 0 18,467 1.88 .05 1 1.89 -- Loc. 38 201-15 3
29 17 Base1 ne 19,440 .40 .003 Absorption! 1 .438 I Loc. 1 1026 2
Titration
29 25 Baseline 19,440 .40 .003 1 .699 N Loc. 1 107—2 2
-------
TABLE 7-1. EMISSION SOURCE TEST DATA - SO (cot,t’d.)
Fuel characteristics No. or Emissions % Control Design
Actual Dur- (lb/MBtu) Based on Control Control
Boiler Heat ation Longest ( n /J ) Average Effi- Levels Referents
Size Te&t Load Control Value of Cont. of clency Sup- Unit 1.0. Test
1O lb ,thr 10 ’ lbJhD Method Btu/lb %1 I.k. od Tests Duration !!ii! . Avera All Tests of Devic p r !d & Locatj j9 j.
150 70 BOOS 18,213 2.14 .03 Absorption/ 1 No Data 2.37 No Data No Data -- Loc. 29 119-6 2
Titration
10 30 Variable —- -- —- “ i 2.42 - Loc. 28 1 30-1 2
Preheat
40 32 Baseline l8 773 1.91 .07 “ 1 “ 1.90 —— Loc. 37 176—2 2
45 36 BaselIne 19,227 .19 .08 ‘ 1 1.72 IS -— Loc. 38 1861 2
17.5 14 Baseline 19,365 .37 .009 1 .340 I Loc. 19 200-3 2
* S Stringent • .2 lb/MBtu
I • Intermediate • .5 lb/MBtu
M • Moderate • .8 lb/MBtu
Thus, S. I, and M all meet SIP levels.
5ource: Personal comunication from Acurex, 8-29—18.
t•4
-------
TABLE 7-2. EMISSION SOURCE TEST DATA - MO
Ref.
Actual
Fuel Characteristics
No. or
Dur—
Emissions
(lb/MBtu)
% Control
Based on
Design
Control
Control
Boiler
Size
l0 lb/hr
Te t Load
iO lb/hr
Control
Method
Heat
Value
Btu/lb
%S
%
Ash
Test Method
ation
of
Tests
Longest
Cont.
Duration
(ng/J)
Average
of
All Tests
Effi-
ciency
of Device
Levels
Sup-
ported*
Reference
Unit I.D.
& Location
Test
No.
Low
Average
110
88
Baseline
19,340
.18
.004
Chemiluminescence
1
No Data
.227
No Data
No Data
SIP
Unit T-8
Loc. 17
7-10
80
51
Baseline
18,560
1.53
.032
Cheniiluminescence
1
No Data
.391
No Data
tb Data
--
Unit 4
Loc. 20
8-5
1
90
71
Baseline
18,420
1.04
.031
Chemiluminescence
1
No Data
.316
No Data
No Data
--
Unit 2
Loc. 18
9-1
1
65
54
Baseline
iii Load
18,930
1.03
.043
Chemlluminescence
1
No Data
.239
No Data
No Data
M
Unit 2
Loc. 16
10—2
1
105
80
Baseline
18,910
1.03
.032
Chemiluminescence
1
No Data
.322
No Data
No Data
--
Unit 3
Loc. 18
21-6
1
105
76
#4 BOOS
18,910
1.03
.032
Chemiluminescence
1
No Data
.285
bo Data
No Data
M
Unit 3
21-20
1
.
Loc. 18
160
130
Baseline
18,910
1.03
.032
Chemiluminescence
1
No Data
.308
No Data
Ho Data
--
Unit 4
Loc. 18
22-1
1
160
cji 500
105
400
#2 BOOS
Baseline
18,910
.18,330
1.03
2.43
.032
.12
Chemiluminescence
Chemiluminescence
1
1
No Data
No Data
.258
.342
Ho Data
No Data
No Data
No Data
M
-—
Unit 4
Loc. 18
Unit 2
22—16
29—5
1
1
‘
Loc. 13
10
7
Baseline
Hi Load
-—
.23
-—
Chemiluminescence
1
No Data
.217
No Data
No Data
SIP
Unit 2
Loc. 3
33-3
1
7
6.7
Baseline
UI Load
18,280
1.85
.038
Chemiluminescence
1
No Data
.383
No Data
No Data
--
Unit 1
Loc. 23
34-11
1
17.5
14
Baseline
19,470
.06
.001
Cheniiluminescence
1
No Data
.084
No Data
No Data
S
Unit 1
Loc. 19
52-5
1
17.5
14
Hi Air
19,470
.06
.001
Chemiluminescence
1
Mo Data
.131
No Data
Ho Data
I
Unit 1
53-6
1
17.5
12
Baseline
.
19,470
.06
.001
Chemiluminescence
1
No Data
.102
No Data
No Data
I
Loc. 19
Unit 1
Loc. 19
54-5
1
20
16
Baseline
-—
.30
——
Chemilurninescence
1
No Data
.248
No Data
No Data
SIP
Unit 4
Loc. 4
59—6
1
158
115
Baseline
-—
.23
.005
Chemiluminescence
1
ito Data
.232
No Data
No Data
SIP
Unit 3
Loc. 6
65-1
1
33
23
Baseline
-—
.22
.001
Chemiluminescence
1
tb Data
.158
No Data
No Data
N
Unit 3
66-1
1
17.5
17.5
14.5
14.5
Baseline
Low 02
19,680
19,610
.16
.14
<.001
.001
Chemiluniinescence
Chemiluminescence
1
1
No Data
No Data
.157
.127
No Data
No Data
No Data
Ho Data
M
I
Loc. 1
Loc. 19
Loc. 19
19-5
19—74
3
3
-------
TABLE 7-2. EMISSION SOURCE TEST DATA - NO (cont ’d.)
Actual
Boiler
Size Teat Load
iO lbJhr lO lb/hr _______ ______
17.5 14.5 FGR
Low 02
17.5 14.5 Staged
Air
17.5 1.4.3 FGR & SA
Low 02
17.5 13.8 BaseiTne
17.5 14.2 Low 0
17.5 13.8 Stage
Air
14.0 Max. FGR
14.3 FGR & SA
17 Baseline
25 Baseline
70 Low Load
70 BOOS
29 Baseline
30 Variable
Coot. Air
Temp.
200 68 BaselIne
80 62 Baseline
40 32 Baseline
40 32 Low Air
45 36 Baseline
45 36 SCA
17.5 14 Baseline
17.5 14.3 SCA
17.5 14.0 Baseline
17.5 14.0 SCA
* S = Stringent = .1 lb/MBtu for F.O. #6 and F.O. #2
Intermediate = .2 lb/MBtu for F.0. #6;
.15 lb/MBtu for F.0. #2
M = Moderate = .3 lb/MBtu for F.0. #6
.2 lb/MBTU for F.O. #2
SIP = 0.3 lb/MBTU
Thus, S, I, and M all meet SIP levels.
Source: Personal comunicatlon from Acurex, 8-29-78.
No. or
Dur-
ation
of
Test Method Tests
Design
Control
Effi-
ciency
of Device
Control
1.evels Reference
Sup- Unit I.D.
ported*& Location
2
2
2
2
2
2
2
2
2
2
Fuel Characteristics
Emissions % Control
(lb/MBtu) Based on
( ng/J ) Average
of
Low Average All Tests
17.5
17.5
29
29
ti,, 150
C 150
70
70
Ref.
Control
Method
Heat
Value
BtuJlb
%S
%
Ash
Longest
Cont.
Duration
Test
19,610
.14
.001
Chemilumlnescence
1
No Data
.041
No
Data
No
Data
S
Loc.
19
19-83
3
19,610
.14
.001
Chemiluminescence
1
No Data
.126
No
Data
No
Data
I
Loc.
19
19-116
3
19,610
.14
.001
Chemlluminescence
1
No Data
.042
No
Data
No
Data
S
Loc.
19
19-179
3
19,000
18,780
18,780
.54
.60
.60
.019
.034
.034
Chemllumjnescence
Chemlluminescence
Chemlluninescence
1
1
1
No Data
No Data
No Data
.280
.196
.198
No
No
No
Data
Data
Data
No
No
No
Data
Data
Data
M
I
I
Loc.
Loc.
Loc.
19
19
19
19-97
19-132
19-143
3
3
3
18,780
18,780
19,440
19,440
.60
.60
.40
.40
.034
.034
.003
.003
Chemilumlnescence
Chemllumjnescence
Chemiluminescence
Chemlluminescence
1
1
1
1
No Data
No Data
No Data
No Data
.196
.211
.118
.111
No
No
No
No
Data
Data
Data
Data
No
No
No
No
Data
Data
Data
Data
I
M
I
I
Loc.
Loc.
bc.
Loc.
19
19
1
1
19-159
19-170
102-6
107-2
3
3
2
2
18,213
2.74
.03
Chemlluminescence
1
No Data
.324
No
Data
No
Data
--
bc.
29
119—1
2
18,213
2.74
.03
Chemlluminescence
1
No Data
.242
No
Data
No
Data
N
Loc.
29
119—6
2
--
—-
—-
Chemlluminescence
1
No Data
.268
No
Data
tlO
Data
M
Loc.
28
126—2
2
--
—-
--
Chemiluminescence
1
No Data
.322
No
Data
No
Data
--
Loc.
28
130—1
2
19,390
.31
.O01
Chemlluminescence
1
No Data
.135
No
Data
No
Data
I
bc.
36
160-1
18,660
1.6
.25
Chemiluminescence
1
No Data
.345
No
Data
No
Data
--
Loc.
20
170—3
18,773
1.91
.07
Chemiluminescence
1
No Data
.254
No
Data
No
Data
M
Loc.
37
176-2
18,773
1.91
.07
Chemilumlnescence
1
No Data
.227
No
Data
No
Data
M
bc.
37
179-4
19,227
.19
.08
Chemiluminescence
1
No Data
.426
No
Data
No
Data
--
Loc.
38
186-1
19,227
.19
.08
Chemiluminescence
1
No Data
.226
No
Data
No
Data
N
Loc.
38
188—1
19,365
.37
.009
Chemilumlnescence
1
No Data
.221
No
Data
No
Data
M
Loc.
19
195-1
19,365
.37
.009
Chemiluminescence
1
No Data
.175
No
Data
No
Data
I
Loc.
19
198—12
19,365
.37
.009
Chemiluminescence
1
No Data
.212
No
Data
No
Data
N
Loc.
19
200—3
19,365
.37
.009
Chemiluininescence
1
No Data
.170
No
Data
No
Data
I
Loc.
19
203-7
p
-------
TABLE 7-3. EMISSION SOURCE TEST DATA - SOLID PARTICULATE
Fuel Characteristics
Flea t
Value
LL
19,340
18,500
18,420
1 8,930
18,910
18,910
1 8,910
18,910
18,330
Test Loa Control
10 ’ lb/h Method
88 Baseline
51 Baseline
71 Baseline
54 Baseline
Ill Load
80 Baseline
76 BOOS
130 Baseline
105 BOOS
400 Baseline
7 Baseline
Hi Load
6.7 Baseline
lii Load
14 Baseline
Emissions
(lb/MBtu)
% Control
Based on
Average
of
Design
Control
Effi-
ciency
Control
Levels
Sup-
Low Hiq % Averag
All Tests
of Device
ported*
.18
1.49
1.04
1.03
1.03
1.03
1.03
2.43
.23
Actual
Boiler
ize
110
80
90
65
105
105
160
160
500
10
7
17.5
b l7.5
l7.5
• 20
158
33
17.5
17.5
17.5
17.5
17.5
17.5
17.5
17.5
17.5
17.5
45
45
45
10 Data
iO Data
18,280 1.35
Opacity
Snoke Reference
Spot Unit I.D.
Number & Location
-- Unit 1-8; Loc. 17
-— Unit 4; Loc. 20
- - Unit 2; Loc. 18
-— Unit 2; Loc. 16
S
Fl
I sh
.004
.019
.031
.043
.032
.032
.032
.032
.12
.038
.001
.001
.001
.005
.003
.001
.001
.001
.001
.001
.019
.034
.034
.034
.034
.05
.05
.05
No. or
Dur-
ation Longest
of Cont.
Test Method Tests Duration
EPA 5 1 110 Data
14
12
16
115
23
14.6
14.5
14.5
14.5
14.3
13.8
14.2
13.8
14.0
14.3
38.0
38.8
34
Iii Air
Basel me
Baseline
Basel me
Basel me
Basel inc
Low 02
FGR & Low 02
Staged Air
FGR & SA
Low 02
Baseline
Low 02
Staged Air
Max. FGR
FGR & SA
Basel inc
Low 02
Variable
Preheat
19,470
19,470
19,470
19,440
19,680
19,610
19,610
19,610
19,610
19,000
18,780
18,780
18,780
18,780
18,467
18,467
13,467
.06
.06
.06
30
.40
.16
.14
.14
.14
.14
.54
.60
.60
.60
.60
1.83
1 .88
1.88
.0103
.0704
0912
.1045
.0581
.0901
.0 335
.0485
.35 96
.0206
.207
.0 339
.0163
.0151
.0145
.0037
.0246
.006
.0138
.0045
.0210
.00 77
.064
.060
.064
.068
.021
.035
.088
.037
I -- Unit 3; Loc. 18
I -- Unit 3; Loc. 18
-- Unit 4; Loc. 18
I -- Unit 4; Loc. 18
—- - — Unit 2; Loc. 13
S — - Unit 2; Loc. 3
N 2.3 Unit 1; Loc. 23
I —— Unit 1; Loc. 19
S —- Unit 1; Loc. 19
S -- Unit 1; Loc. 19
S -- Unit 4; Loc. 4
S — - Unit 3; Loc. 6
S - - Unit 3; Loc. 1
S 0% Loc. 19
S 0% Loc. 19
S 0% Loc. 19
S 13% Loc. 19
S 0% Loc. 19
1 0% Loc. 19
I 6% Loc. 19
1 22% Loc. 19
I 0% Loc. 19
S 0% Loc. 19
I -- loc. 38
I -- Loc. 38
I — - Loc. 38
Test No .
7-10
8-5
9—1
10-2
21-6
21-20
22-1
22-16
29-5
33-3
34-11
52—5
53-6
54-5
59-6
65-1
66-1
19-5
19-74
19-83
19-116
19-179
19-97
19-132
19-143
19—159
19-170
200 -24
201 -12
2 02-4
Ref.
3
3
3
3
3
3
3
3
3
3
3
3
3
-------
lADLE 7-3. EMISSION SOURCE TEST ,kTA - SOLID PARTICULATE (contd.)
Fuel Characteristics No. or Emissions S Control Design
Actual Dur- (ib/liBtu) Based on Control C. ntrol Opacity
Roller Heat atlon Longest ( 09 /J ) Average Effi- Le els Smoke Reference
Size T 9 t ba r n rol Value S of Cont. of ciency p- Spot Unit 1.0.
th1i Q. pLh ,lethod utuLlb S Ash Test Method Tests Duration I1’oh Av a e All Tests of Device ported* Number & Location Test No . Ref.
45 38 Staged Comb. 18,467 1.88 .05 EPA 5 1 No Data .09 No Data No Data I -- Loc. 38 203-26A 3
Air
45 39 Staged Comb. 18,467 1.88 .05 1 .092 I -- Loc. 38 2O3-26B 3
Al r
45 38.8 Low O , 18,467 1.88 .05 i .089 I -- Loc. 38 201-15 3
29 17 BaselIne 19,440 .40 .003 i .017 S -- Loc. 1 102-6 2
29 25 Baseline 19,440 .40 .003 1 .011 S -- Loc. 1 107-2 2
150 70 Low Load 18,213 2.74 .03 1 .044 I -- Loc. 29 119-1 2
70 29 Baseline -- —- -- i .296 -- 5.5 Loc. 28 126-2 2
70 30 Variable -- — . -- 1 .108 M 3.0 Loc. 28 130-1 2
Preheat
200 68 Baseline 19,390 .31 .001 1 .016 S 0.0 Loc. 36 160-1 2
80 62 Baseline 18,660 1.6 .25 1 .074 1 3.0 Loc. 20 170-3 2
40 32 Baseline 18,773 1.91 .07 i .118 H 5.0 Loc. 37 176-2 2
40 32 Low Air 18,773 1.91 .07 1 .081 0 7.0 Loc. 37 179-4 2
45 36 BaselIne 19,227 .19 .08 1 .090 1 - - Loc. 38 186-1 2
t 45 36 SCA 19,227 .19 .08 1 .128 M 8.0 Loc. 38 188-1 2
17.5 14 Baseline 19,365 .37 .009 1 .020 S 1.5 LoC. 19 195-1 2
17.5 14.4 FOR 19,365 .37 .009 1 .022 S -- Loc. 19 197-8 2
17.6 14.3 SCA 19,365 .37 .009 1 .023 S 4.0 Loc. 19 198-12 2
17.5 14.0 Baseline 19,365 .37 .009 1 .032 I 1.0 Loc. 19 200-3 2
17.5 14.0 SCA 19,365 .37 .009 1 .042 I 8.0 Loc. 19 203-7 2
169 No Data Electrostatic -- .86 -- 1 .074 I •- Co. A, Plant 1 -- 5
Precipitator
169 —- 1.18 1 .142 H -- Co. A, Plant 1 —- 5
169 — — 1.13 i .150 M —— Co. A, Plant 1 —- 5
188 -- 1.00 1 .097 1 -- Co. A, Plant 1 -- 5
94 ‘ -- .70 ‘ 1 .055 I -- Polaroid,New Bedford -- 5
94 .70 1 .070 0 I -- Polaroid,New Bedford -- 5
* S Stringent .03 lb/MBtu
I Intermediate .10 lb/MBtu
M Moderate .25 lb/MBtu
Thus, S and I meet SIP level.
Source: Personal comunication from Acurex, 8-29-78.
-------
nitrogen or sulfur as NO or SO ; but, the value of this number is question-
able. A blank in the “Control Levels Supported” column indicates that no con-
trol leve1s are supported.
7.3 TEST METHODS
Particulate samples were taken with a Joy Manufacturing Co. Portable Ef-
fluent Sampler, which meets design specifications for EPA Test Method 5 (Fed-
eral Register, Volume 36, No. 27, p. 24888, December 23, 1971). Dry particu-
lates were collected in a heated case containing a cyclone for separation of
particles larger than 5 microns, followed by a 125 nm glass fiber filter for
retention of particles down to 0.3 microns. A train of 4 Greenburg-Smith im-
pingers in a chilled water bath was used to collect condensible particulates.
Since EPA standards are based on solid (dry) particulate, only the dry parti-
culate values are considered in this report.
The measured values of particulate in lb/ft 3 were converted to lb/MBtu,
using the revised method promulgated by EPA (Federal Register, Volume 39,
No. 177, Part II, paragraph 60.46, September 11, 1974), which utilizes a fuel
analysis and the measured excess 02 in the exhaust, to calculate the gas volume
generated in liberating a million Btu’s. This method also includes excess air
dilution.
SO 3 concentrations were measured by wet chemical analysis using the “Shell-
Emeryville” method. The gas sample is drawn through a heated glass probe con-
taining a quartz wool filter to remove particulate matter, into three sintered
glass plate absorbers. The sulfur trioxide is removed by the first two absorb-
ers containing aqueous isopropyl alcohol, and the sulfur dioxide is removed by
the last absorber containing aqueous hydrogen peroxide. Separation of the com-
ponents is completed by a nitrogen purge of the absorbers, to transfer all re-
maining SO 2 to the third absorber. The sulfate from the SO 3 and SO 2 absorbers
is then titrated with standard lead perchlorate solution, using Sulfonazo III
indicator.
-Z29-
-------
Total nitrogen oxides were measured by a Thermo-Electron brand chemilumi-
nescent nitric oxide analyzer. t40 and NO were measured using a “hot line”,
heated to about 120°C, to conduct the gas sample to the analyzer. Also, NO
was measured using an unheated “cold line” to the analyzer. Hot line and cold
line measurements were compared statistically as an indicator of measurement,
and it was found that the hot line and cold line measurements agree very close-
ly (see Reference 1, p. 32, 33).
Opacity was measured using Bacharach Smoke Spots, obtained with a Research
Appliance Company Transmittance Particulate Monitor, modified to measure re-
flectance. The Monitor measured the reflected light from a spot on a paper
tape that was soiled by passing flue gas through it for a fixed time period.
Most of the reported data were taken with a standard hand pump device to pass
the gas through the tape.
The analyzers for particulate, smoke spot, and sulfur oxides measurements
were taken to the sample port. The weighing and titration were done in or near
the EPA Mobile Air Pollution Reduction Laboratory trailer, and the NO analyzer
was also located in the trailer. Further detail on the san ling and analysis
procedures can be found in References 1-3.
7.4 SUMMARY
The available data were not collected for the purpose of developing
alternate control options. Most of the data was developed with the emphasis
on nitrogen oxide emission reduction by contustion modification and tended to
neglect control measures for SO and particulate emissions. The bulk of the
available data is for SON, NOR. and particulate emissions from uncontrolled
industrial boilers. There is a little data on electrostatic precipitators for
control of particulate emissions from oil-fired industrial boilers, but there
are no significant data on the emission reduction potential of POX, UDS, CAFB,
and FGD, as applied to SO , NOx, and particulate emissions from oil-fired
industrial boilers.
-230-
-------
Some useful trends for uncontrolled industrial boiler emissions may be
found from the data, however. Sulfur oxide emissions and particulate emissions
are highly dependent on the sulfur and ash contents of the oil, respectively.
Nitrogen oxide emissions are dependent on fuel nitrogen content, as well as ex-
cess 02 level and boiler size. In general, N0 emissions decrease with de-
creasing excess 02 and with increasing boiler size. The naturally low-sulfur,
low-ash oils tended to meet at least the recomended control levels for moder-
ate control in the oil cleaning category and quite often met even the inter-
mediate or stringent control levels. The high-sulfur, high-ash oils, however,
often failed to meet even moderate control levels, which suggests the need for
oil cleaning or flue gas treatment.
The presently-available data, presented in this report, may thus prove
valuable as a basis for comparison with emissions data from oil treatment tech-
nologies, which are not yet available.
-231-’
-------
RE F EREN CES
1. Cato, G. A., et al, “Field Testing: Application of Combustion Modifica-
tions to Control Pollutant Emissions From Industrial Boilers - Phase I,”
EPA-650/2-74-078a, October 1974.
2. Cato, G. A., et al, “Field Testing: Application of Combustion Modifica-
tions to Control Pollutant Emissions From Industrial Boilers - Phase II,”
EPA-600/2-76-086a, April 1976.
3. Carter, W. A., et al, “Emission Reduction on Two Industrial Boilers With
Major Combustion Modifications,” EPA-600/7-78-099a, June 1978.
4. Hunter, S. C., and H. J. Buening, “Field Testing: Application of Combus-
tion Modifications to Control Pollutant Emissions From Industrial Boil-
ers — Phases I and II (Data Supplement),” EPA-600/2-77-122, June 1977.
5. GCA Corporation, “Particulate Emission Control Systems for Oil-Fired
Boilers,” EPA-450/3-74-063, December 1974.
-232-
-------
APPENDIX A
GLOSSARY OF TERMS
API gravity (°API): An arbitrary scale expressing the gravity or density of
liquid petroleum products where the measuring scale is calculated in
terms of degrees API by the following formula:
API = 141.5 — 131.5
sp. gr. 60°F/60°F
atmospheric residue: The heavy, less volatile liquid produced from distilla-
tion of petroleum at atmospheric pressure (frequently called resid or
topped crude) whose boiling point is 650+°F.
barrel: A common unit of liquid measurement in the petroleum industry; it
equals 42 U. S. standard gallons.
catalyst: A substance which influences the rate of a chemical reaction but is
not one of the original reactants or final products. A catalyst partici-
pates in intermediate chemical reaction steps in such a manner as to
facilitate the over-all course of the reaction.
catalytic cracking: The process of selective decomposition of heavy distillate
oils over a catalyst to produce gasoline, C 3 /C 4 olefins, and isobutane.
catalytic reforming: The process of converting low octane naphthas into high
octane naphthas by catalytically rearranging and dehydrogenating naph-
thenes and paraffins to form aromatic compounds.
coking: Thermal cracking of heavy, low—grade oils into lighter products and
a solid residue of coke.
cracking: the reactions in which a hydrocarbon molecule is broken or fractured
into two or more smaller fragments or the process of converting heavy oils
into petroleum fractions of lower boiling range and corresponding lower
molecular weight by thermally or catalytically fracturing the hydrocarbon
molecule.
delayed coking: A thermal process applied to a residual oil stream which uses
severe conditions (1800-2000°F) to crack the feedstock to a coke gas, dis-
tillates, and coke.
demetallization: Removal of the organo-metal compounds by use of catalysts and
heat in crude before processing so valuable catalysts are not poisoned by
unwanted metals deposition.
denitrogenation: Removal of nitrogen compounds by catalysts to improve the
quality of the petroleum product.
desalting: The removal of either sodium chloride and/or compounds that act
like sodium chloride to prevent clogging, accumulation of undesirable
compounds, decomposition and corrosion of refinery equipment.
-233-
-------
GLOSSARY (continued)
desulfurization: The removal of undesirable sulfur and sulfur compounds from
crude or residual oils so that end-use applications avoid violation of
environmental regulations limiting sulfur levels in oil or emission
standards.
distillation: Vaporization of a liquid and its subsequent condensation in a
different chamber which allows separation of petroleum hydrocarbons by
boiling point ranges.
hydrocracking: A process conbining cracking or pyrolysis with hydrogenation
over a catalyst bed to meet various product demands.
hydrodesulfurization: A catalytic process whereby a hydrocarbon feedstock and
hydrogen are passed through a catalyst bed at elevated temperatures and
pressures so that sulfur in the feedstock reacts with the hydrogen on the
catalyst surface to produce hydrogen sulfide (H 2 S) and a desulfurized
hydrocarbon product.
hydrogenation: The chemical addition of hydrogen to a material. Hydrogen can
be added to: (1) unsaturated compounds; or (2) in a destructive cracking
case where the hydrocarbon chains have been broken.
hydrotreating: The use of hydrogen and a catalyst to purify, cleanse, and
improve the quality of the feedstock with minimal reduction in molecular.
size of feed.
petroleum: A material occurring naturally in the earth, composed mainly of
mixtures of chemical compounds of carbon and hydrogen with lesser amounts
of sulfur, nitrogen, oxygen, and metals.
pour point: The lowest temperature at which a petroleum liquid will pour or
flow when it is chilled under definite conditions.
recycle: That portion of a feedstock which has passed through a refining pro-
cess and is recirculated through the process to achieve complete reaction
or accumulate unwanted byproducts.
residual oil: Thick, heavy, semi-solid stream (produced as bottoms in distilla-
tion) which is high boiling and contains undesirable levels of organo-
metallic and organo-sulfur compounds. Comonly called resid.
residue: Heavy oil or bottoms left in the still after gasoline and other rela-
tively low-boiling hydrocarbons have been distilled off.
thermal cracking: The use of heat to achieve cracking or fracturing of the
hydrocarbon molecule.
vacuum gas oil: The heavy fuel oil produced when atmospheric residue is dis-
ti1led at a pressure of 50 nm Hg. Boiling point range is 343-566°C
(650—1050°F).
vacuum residue: The heavy, thick bottoms (pitch) produced by distillation of
atmospheric residue under 50 nw Hg pressure whose boiling point exceeds
9800 F.
-234-
-------
GLOSSARY (continued)
visbreaking: A thermal cracking process which lowers the viscosity of residual
oil to lessen the amount of blending stock required to upgrade the resi-
dua’ to fuel oil specifications.
-235—
-------
APPENDIX B
SULFUR RECOVERY SYSTEMS FOR OFF-GAS TREATMENT
Hydrogen sulfide gases are released during the regeneration of amine or
other scrubbing solutions which are used to desulfurize refinery process
gases such as those produced by hydrodesulfurization. In addition, some H 2 S
is removed from process water by sour water strippers. Most refineries in-
clude facilities for steam stripping H 2 S from sour water streams as part of
the waste water system. Where sulfur recovery is practiced, the off-gases
from the stripper are normally routed to the sulfur recovery plant.
In the CAFB process, SO 2 is produced in the regeneration cycle; and, dif-
ferent techniques are used to collect it. These are discussed below after the
methods for H 2 S recovery.
The Claus Process
The Claus process has been used almost exclusively in petroleum refiner-
ies to recover sulfur. The basic exothermic reactions for the Claus process
are:
(1) H 2 S+l/20 2 ,, .H 2 0+S
(2) FI 2 S + 3/2 02 • SO 2 + H 2 0
(3) 2H 2 S + 2 • ‘ 3S + 2H 2 0
A typical two—stage Claus plant is shown in Figure B—l. Hydrogen sulfide
gas enters the burner with sufficient air to convert all H 2 S to sulfur. As
much as 50 to 60 percent conversion of the hydrogen sulfide to sulfur takes
place in the initial reaction chamber by Reaction (1). Reaction (2) also takes
place, forming SO 2 . After cooling, condensing, and removing sulfur, the gases
are reheated by mixing with a portion of the gases bypassed around the sulfur
condenser and introduced into the first catalytic reactor, where the Claus re-
action (Reaction (3)) occurs. From the first catalytic reactor, the effluent
-236-
-------
F4 S Gas from
Amine Regenerator
and Sour Water
Stripper
Figure B-i
TYPICAL PACKAGED CLAUS PLANT (2 STAGE)
Tail Gas to
Incinerator or
Tail Gas
Processing
Liquid Sulfur Product
Secondary
Converter
Steam
Waste Heat
Burner
Air
Boiler Feed Water
Steam
—--— ---.—-——--—\
El-
‘ ——---— ———-.‘
)
Sulfur Tank and Sump Pump
-------
gas is cooled, sulfur condensed and removed, and the gases reheated again.
The process is repeated in the second catalytic converter. If needed, addi-
tional catalytic stages may be utilized to remove H 2 S as sulfur.
Some carbonyl sulfide (COS) and carbon disulfide (CS 2 ) are formed in the
reaction furnace in the presence of carbon dioxide and hydrocarbons:
(4) CO 2 + H 2 S 4 H O + COS
(5) COS + H 2 S 1 , ‘ H O + CS 2
(6) CH 4 + 2S 2 CS 2 + 2H S
Depending on the exact nature of the sour-gas feed stream and the operating
conditions in the upstream reaction furnace and catalyst beds, combined COS
and CS 2 levels as high as 5000 ppmv may exist in the tail-gas. Values of
600-1500 ppmv are more comon.
The emissions of H 2 S, SO 2 , and sulfur vapor from Claus plants are directly
dependent on the efficiency of sulfur recovery in the Claus plant. Claus plant
efficiencies are dependent on the following variables:
(1) number of catalytic conversion stages;
(2) inlet feed stream composition;
(3) operating temperatures and catalyst maintenance;
(4) maintaining proper stoichiometric ratio of FI 2 S/S0 2 ; and
(5) operating capacity factor
For Claus plants fed with 90 mole percent H 2 S, the sulfur recovery is
approximately 85% for one catalytic stage and 95% for two or three stages.
The percentage sulfur recovery also increases with increasing concentration of
the acid gas fed to the Claus unit. For plants having two or three catalytic
stages, the sulfur recoveries for various acid gas concentrations are approxi-
mately 90% for a 15 mole percent FI 2 S feed stream, 93% for a 50 mole percent
H 2 S stream, and 95% for 90 mole percent H 2 S concentration. 2
-238-
-------
Contaminants in the feed gas reduce Claus sulfur recovery efficiency.
Hydrocarbons in Claus feedstocks require extra air for contustion. The added
water and inert gas associated with burning hydrocarbons increases the size
of the sulfur plant equipment and lowers sulfur recovery, since the sulfur gas
concentrations are decreased. (3) High molecular weight hydrocarbons in the
feed also reduce Claus efficiencies because of carbon soot deposition on the
catalyst. Since the reactions in a Claus plant are exothennic, sulfur recovery
is enhanced by removing heat; hence, operation at as low a temperature as prac-
tical in the reactors without condensing sulfur on the catalyst is necessary.
Although sulfur recovery efficiencies of 94—96% are possible in standard
multiple—stage Claus units, these are insufficient to meet most air pollution
control regulations. The tail-gas, containing 4—6% of the input sulfur value,
is usually incinerated by thermal or catalytic means. Desulfurization of Claus
tail—gas has received increasing attention for two reasons: (1) the increase
in desulfurization activity has resulted in an increase in sulfur dioxide emis-
sions from the Claus system; and (2) the off-gas is particulate free and high
in hydrogen sulfide concentration (up to 1500 ppm) and is thus easier to treat
than ordinary fine gas. The treatment of Claus off-gas can fol 1 one of many
schemes whose basic treatment methodologies are shown in Figure B-2. With appli-
cation of the Claus off-gas cleanup units, the overall sulfur re val effi-
ciency (Claus included) generally exceeds 99.5% for the treatment systems whose
descriptions follow.
Beavon Process (Figure B-3) 4
The Beavon Process involves hydrogenation of the other sulfurous gases
(carbon disulfide, carbonyl sulfide) in the Claus tail-gas to hydrogen sulfide
over a cobal t-molybdate catalyst at moderate temperature and pressure. This
catalyst effectively promotes the reaction between water vapor and carbon
-239-
-------
Figure B-2
EMISSION CONTROL SYSTEMS FOR REFINERY CLAUS PLANTS
Control System
(A) (B) Absorber
______________________ Weilman-Lord
Absorption Off-Gas A: IFP-2
-$ &
Regeneration
SO 2 _Recycle_(A) _____________
Incineration
(B)
I- I______
(A)
Gas Tw Stage (B) (C) (D) (E) ____ J Tail Gas C:
Solution
_____ ________ _________________________ _____ ______ ________________ IF P-i
Claus
from I Claus _________ ___________________________ _____ ______ _________________
Sulfreen
Sulfur
Refinery
Plant Reactor
___________ (F) _______________
(G) (E)
Sulfur (H)
Sulfur
Conversion ___________________________ Stretford Absorber Off-Gas
___________________________ Sulfur E: Clean Air
(F)
Plant F: Beavon
to H 2 S
______________ Sulfur
H 2 S
Absorption Absorber Off-Gas
& G:Scot
H: Sulfinol
Regeneration
H 2 S Recycle (G) (H)
-------
Figure B-3
Air
Claus Plant
Tail Gas before
Incinerator
FLOW DIAGRAM FOR THE BEAVON SULFUR REMOVAL PROCESS
Fuel Gas
Absorber Off-Gas to Incineration or Stack
Fixed Bed
Reactor
Hydrogenated
Tail Gas
Sulfur
Melter
Stretford
Gas PurIfying
Tower Absorber
Oxidizer
F liter
or
Centrifuge
Sour Water
Sulfur
(To Waste
Purge Stream
-------
monoxide to form hydrogen. It also increases the rate of reaction between
water vapor and carbonyl sulfide and carbon disulfide to yield hydrogen sul-
fide. Excess water is condensed after the reactor to avoid corrosion and
plugging problems in the hydrogen sulfide conversion step by the Stretford
Process. That process can be conducted at atmospheric pressure and produces
a high-purity elemental sulfur with a low concentration of hydrogen sulfide
effluent. Air and fuel gas are reacted in an in-line burner where the combus-
tion products are mixed with the Claus off-gas to produce a reducing environ-
ment in the Beavon reactor where the hydrogenation occurs. The hydrogen sul-
fide product stream is converted to sodium hydrosulfide in the Stretford Pro-
cess. Oxidation to elemental sulfur is accomplished over sodium vanadate in
solution. The product sulfur is recovered by conventional fine—solids concen-
tration steps: washing, filtration or centrifugation, followed by decantation
after melting and coagulation of the sulfur.
Cleanair Process (Figure B—4) 5
The Cleanair Process includes the Stretford Process and two confidential
processes. An optional part of the Cleanair unit includes a modification of
the Claus plant first stage to include a reducing and hydrolysis catalyst.
This causes the conversion of COS and CS 2 to H 2 S, according to the following
reaction:
COS + 1120 ‘ H 2 S + CO 2
CS 2 + 21120 ‘2H 2 S + CO 2
Claus tail-gas, with essentially all the gaseous sulfur as sulfur vapor,
H 2 S, and SO 2 , is quenched to reduce temperature and remove water and entrained
sulfur. The cooled gas is fed into a reactor where H 2 S and SO 2 react, lowering
SO 2 to less than 250 ppmv. Both water and sulfur are removed. Next, the tail-
gas is sent to a Stretford unit, where the remaining H 2 S is removed and oxidiz-
ed to elemental sulfur. Residual SO 2 , although absorbed by the Stretford solu-
tion, decomposes the solution to increase chemical consumption and liquid purge
-242-
-------
Figure B -4
Claus Plant Tail
Gas Before
Incineration
FLOW DIAGRAM FOR THE CLEANAIR CLAUS TAI L-GAS TREATMENT PROCESS
Tank
Stretford Purge
Fixed-Bed
Catalyst Reactor
Absorber Off-Gas
to Incineration or Stack
Stretford Solution
L
Depurator
Liquid Sulfur to Storage
-------
rate. Residual COS and CS 2 will pass through the Stretford unit unaffected.
Purified gas is then sent to an incinerator to oxidize residual sulfur to SO 2
and CO to CO 2 .
IFP-2 (Figure B-5) 6
In the IFP—2 Process, Claus plant tail-gas is incinerated to convert all
sulfur species to S0 2 .* The incinerated gas is cooled and then fed to an am-
monia scrubber, where SO 2 is absorbed and converted to amonium sulfite and
animonium bisulfite by the following reactions:
2NH 4 OH + SO 2 = H 2 0 + (NH 4 ) 2 SO 3
NH 4 OH + SO 2 . NH 4 HSO
Gas leaving the absorber is reheated and vented to the atmosphere at less than
250 ppmv SO 2 concentration. The S0 2 -rich solution is fed to an SO 2 regenerator
where the sulfite and bisulfite are thermally decomposed to SO 2 , NH 3 , and 1120.
A saturated solution containing ammonium sulfate and thiosulfate is drawn from
the bottom of the regenerator and fed to a sulfate reducer where it is thermally
reduced, creating SO 2 , NH , S, 1120, and NH 4 HSO 4 . Gases from the sulfate reducer
and SO 2 regenerator are combined with the H 2 S-rich feed stream from the Claus unit
and fed to a catalytic reactor where they are contacted with polyethylene glycol
solvent. The H 2 S and SO 2 react in the solution to form elemental sulfur, which
is withd awn in the molten state. Gases from the reactor are cooled to condense
water and NH 3 and NH 4 OH. The NH 4 OH solution is returned to the amonia scrubber.
Shell Claus Off-Gas Treating Process (Figure B-6) 7 8
The Shell Claus 0ff-Gas Treating (SCOT) Process consists of a reduction
section and an amine absorption section. In the reduction section, all sulfur
values in the Claus off-gas are hydrogenated to hydrogen sulfide over a cobalt!
molybdenum catalyst (supported on alumina) at 300°C. The reactor effluent is
cooled, and the water is condensed. The effluent, containing about 3% H 2 S and
* IFP-2 has replaced IFP-l, which was less efficient
and is not discussed here.
-244 -
-------
Figure B-5
• FLOW DIAGRAM FOR IFP-2 CLAUS TAIL-GAS CLEAN-UP PROCESS
Absorber Off-Gas
to Stack
NH 3 Make-up
Claus Plant
Tail Gas
After
Incineration
Ammonia
Scrubber
cJ
H 2 S Gas from Catalytic
Claus Plant Feed Reactor
Make-up
Ammoniacal Brine
Evaporator and
Regenerator
Sulfur
SO 2 /NH 3 /H 2 O
ISulfate
Reducer
SO 4 =
-------
Figure B•6
FLOW DIAGRAM FOR THE SHELL CLAUS OFF-GAS TREATNG PROCESS
Cooling Tower
Packed or Tray
— Sour gas to Claus unit
Lean amine from regenerator
Tray Tower Absorber
Fat amine to regenerator
Reactor
Reducing Gas -
Line Heater
Claus plant tail gas
prior to incinerator
Fuel gas
Air
Air or
Water
to
existing sour-
water stripper
-------
20% GO 2 , is scrubbed with an alkanolarnine solution in an absorber. COS and
CS 2 are reduced in reactions identical with those in the Beavon catalytic re-
actor. The absorber off-gas contains about 300 ppmv hydrogen sulfide (H 2 S),
which is incinerated. Selective absorption of H 2 S over carbon dioxide is ob-
tained because of a difference in mass transfer rates. The I-1 2 S absorption is
diffusion-limited in the gas phase and is removed by treatment with
di-isopropanolamine solution. The H 2 5—rich solution is regenerated by strip-
ping H 2 S in a conventional steam stripping column. Regenerator off-gas, mainly
and some GO 2 , is recycled as feed to the first stage of the Claus unit.
The SCOT Process can result in up to 99.8% recovery of all the input sulfur.
Sulfinol Process (Figure
The Sulfinol Process uses conventional solvent absorption and regeneration
to remove carbonyl sulfide, hydrogen sulfide, and carbon disulfide from Claus
tail-gas by countercurrent contact with a lean solvent stream under pressure.
The absorbed impurities are removed from the rich solvent by steam stripping
in a heated regenerator column. The hot, lean solvent is cooled for reuse in
the absorber. Sulfinol solvent consists of a mixture of water, an alkanolamine
(di-isopropanolamine), and sulfolane (tetrahydrothiophene dioxide). The alka-
nolamine absorbs acid gases by chemical combination. The sulfolane allows high
solution loadings and low regeneration heat requirements. The Sulfinol Process
can achieve I( 2 S reductions to less than I ppm and CO 2 reductions to less than
100 ppm. The’ combination of the two solvents gives good absorptive properties
for sulfur gaseous compounds at low- to medium-partial pressures and very high
absorption at high partial pressures. The process is nonfoaming and noncorrosive
to steel, and equipment costs can be minimized. Loss of solvent components is
minimal, since sulfolane does not degrade; and, the alkanolamine regeneration
step minimizes those losses.
—247-
-------
Figure B-i
S&JLFINOL PROCESS (9)
SWEET GAS
REF LU X
ACCUMULATOR
REFLUX
CONDENSER
CONTACTOR
SOUR_GAS
RE BOIL ER
FURNACE
HIGH-PRESSURE
SOLVENT PUMP
SOLVENT COOLER
ACID
GAS
FUEL
GAS
STRIPPER
F LASH
VESSEL
SO LV ENT
BOOSTER
PUMP
Ii
DISPOSAL
-------
Suifreen Process (Figure B-8)
The Suifreen Process reduces the sulfur content in Claus plant tail gas
by further promoting the Claus reaction on a catalytic surface in a gas/solid
batch reactor. Claus tail gas is first scrubbed with liquid to wash out en-
trained sulfur liquid and sulfur vapor. The tail gas is then introduced to a
battery of reactors where the Claus reactions are carried out at lower temper-
atures (260-300°F) than those utilized in the sulfur plant. Lower temperatures
push the Claus reaction toward completion due to favorable equilibrium condi-
tions. The catalyst is usually activated carbon, though alumina can be used.
A regeneration gas, usually nitrogen, periodically desorbs the sulfur-laden
catalyst beds. Nitrogen is treated and cycles through the catalyst bed at
approximately 570 0 F until all water and CO 2 are driven off. For the desorp-
tion of sulfur, the temperature is raised to 750°F, where sulfur vaporizes, is
swept away with the nitrogen, and precipitates in a condenser. The carrier
gas is further scrubbed in a sulfur wash before returning to the regeneration
cycle.
The process reduces entrained sulfur, since the catalyst acts as an ab-
sorbent for liquid sulfur. COS and CS 2 are not affected by the Suifreen Pro-
cess. A Sulfreen unit may consist of as little as three reactors; two in ab-
sorption, and one in desorption service. The gases from the desorption service
are incinerated before discharge to the atmosphere.
Weilman-Lord SO 2 Recovery Process (Figure B-9) 02
The Weliman-Lord Process uses a wet regenerative system to reduce the
stack gas sulfur concentration to less than 250 pprnv. Sulfur constituents in
Claus plant tail-gas are oxidized to SO 2 in an incinerator, then cooled and
quenched to reduce the gas temperature and remove excess water. The S0 2 -rich
gas is then contacted countercurrently with a sodium sulfite (Na 2 SO 3 ) and
sodium bisulfite (NaHSO 3 ) solution which absorbs SO 2 to form additional bisul-
fite. The principal reaction between SO 2 and the absorbent solution is:
-249-
-------
Figure B-B
C’aus Plant Tail
Gas before
Incineration
FLOW DIAGRAM FOR THE SULFREEN PROCESS
Converters
Adsorbing Cooling after
Sulfur Regeneration Regeneration
Absorber Off-Gas to
Incineration
Hot
Generation
Gas
Furnace
-------
Figure B-9
FLOW DIAGRAM FOR THE WELLMAN-LORD SO 2 RECOVERY PROCESS
Quench and Gas
Cooling Section
Evaporator
and Steam Stripping
To Stack
Claus Plant
Tail Gas after
Incineration
Product SO 2
Recycle to
Claus Plant
SO Absorber
Dissolving
Tank
‘4
NaOH
Make-up
H 2 0 Recycle
Acid Water Purge
to Neutralization
-------
SO 2 + Na SO 3 + H 2 0 2NaHSO 3
The absorber off—gas is reheated and vented to the atmosphere at less than
250 ppmv SO 2 and negligible amounts of other sulfur compounds. S0 2 -rich solu-
tion is boiled in an evaporator-crystallizer, where the bisulfite solution de-
composes to SO 2 and water vapor; and, sodium sulfite is precipitated according
to the reaction:
2NaHSO 2 ‘ Na 2 SO + H 2 0 + so 2 l
Sulfite crystals are separated and redissolved for reuse as lean solution to
the scrubber. The wet SO 2 flows to a partial condenser where most of the water
is condensed and reused to dissolve sulfite crystals. The enriched SO 2 stream
is recycled to the Claus plant for conversion to elemental sulfur.
Special Cases - Chemically Active Fluidized Bed (CAFB) Residual Oil Gasi-
fication Systems
In regenerative CAFB processes, sulfided lime is oxidized to lime in the
regenerator prior to its recirculation into the gasifier. SO 2 is given off as
a result of this oxidation, and the regenerator off-gases will contain as much
as 6 to 8 percent SO 2 . Since release of these gases is environmentally unac-
ceptable, the sulfur content must be removed. The following discussion covers
the Allied Chemical (AC) and Foster Wheeler Energy Corporation’s RESOX systems
for conversion of the SO 2 into elemental sulfur.
Allied Chemical Process (Figure
The gases from the regenerator must be cooled to about 300°F, and particu-
late material must be removed. A blower then forces the SO 2 through the reduc-
tion system with natural gas as a reducing agent. Although the S0 2 -natural gas
reduction is exothermic, some oxygen in the form of air must be admitted at
the blower inlet to maintain a thermal balance. This mixture then passes
through the feed gas preheater where its temperature is raised above the dew
-252-
-------
Figure B-1O
ALLIED CHEMICAL SO 2 REDUCTION PROCESS
Reducing Gas
Air
Regenerator
C)1
Main
Primary
Reactor
System
(Catalytic
Reduction)
Sulfur
Condenser
C’aus Converters
Steam
To Gasifier
To Storage
Sulfur Holding
Pit
Steam
-------
point of the sulfur formed in the primary reactor system. The gas then enters
the catalytic system where hydrogen sulfide and sulfur are produced. The H 2 S/
SO 2 ratio in the gas stream leaving the system is essentially that required for
the subsequent Claus reaction. The reactions taking place in the primary reac-
tion system are:
CH 4 + 2S0 2 —* CO 2 + 2H 2 0 + S 2
4CH 4 + 6S0 2 —+ 4C0 2 + 4H 2 0 + S 2 + 4H 2 S
Because of the exothermic reaction, a portion of the available heat is used to
preheat the incoming gases in the feed gas heater. The elemental sulfur formed
is condensed in a horizontal shell-and-tube condenser where over 40% of the
total sulfur is recovered. The process gas next enters a two-stage Claus reac-
tor where the following exothermic reaction occurs:
2H 2 S + SO 2 ——p. 1.5 S 2 + 2H 2 0
After the first stage of Claus conversion, the gas is cooled; and, addi-
tional sulfur is condensed by passage through a vertical condenser. Further
conversion of H 2 S and SO 2 to sulfur takes place in the second stage Claus reac-
tor. This sulfur is condensed in a third condenser. Residual sulfur gases
are returned to the CAFB gasifier where they react and are retained by the ac-
tive lime bed.
Foster-Wheeler Energy Corporation - RESOX (Figure B-ll) 3
Gases from the regenerator are reduced to about 1200°F, the required
reactor inlet temperature, by the injection of steam or water. This steam in-
jection also produces the proper H 2 0/S0 2 ratio. Control air and the process
is fed to the RESOX reactor. The air furnishes the limited amount of oxygen
needed for maintaining and controlling reactor temperature 200-300°F above the
inlet conditions. The overall reaction in the RESOX reactor is:
-254..
-------
FIgure B-il
Regeneration Off Gas
I
I
I
I
I
I
• I
I
I
I
I
I
I
I
FWEC RESOXTM SYSTEM FOR SULFUR RECOVERY
Resox Ash
Sulfur Storage
Return Steam
Water
I ¼. _ — — — — — — — —
r i
c 1
I -
Make-Up
Coal Hopper
Tail Gas
— — — — — —
To Regenerator
Control Air
Ambient Conditions
-------
so 2 + C — ‘ Co 2 + S
Although this reaction is exothermic, it cannot maintain the temperature neces-
sary for the high conversion of sulfur dioxide to sulfur. Carbon, the reducing
agent required, can be furnished by any coal with performance characteristics
compatible with the countercurrent, moving-bed, reactor system.
Reduction of 75 to 90 percent of the sulfur dioxide to elemental sulfur
is made in the reactor, depending on the SO 2 inlet concentration. The elemental
sulfur formed is condensed in a horizontal shell-and-tube condenser. An ID fan
downstream of this condenser moves the process gas through the system and re-
turns off-gas to the gasifier where the residual sulfur gases are absorbed by
the fluidized bed.
-256-
-------
RE FERENCES
1. “Standards Support and Environmental Impact Statement, Volume I: Proposed
Standards of Perfor,iiance for Petroleum Refinery Sulfur Recovery Plants,”
EPA—450/2-76—0l6(a), September 1976.
2. Beers, W. D., “Characterization of Claus Plant Emissions,” EPA Contract
No. 68-02—0242, Task No. 2, April 1973.
3. Grekel, H., et al., “Why Recover Sulfur From H 2 S?” Oil and Gas Journal ,
28 October 1968, p. 95, refer to p. 3.15, reference 12, “Standards Support
and Environmental Impact Statement, Volume I: Proposed Standards of Per-
formance for Petroleum Refinery Sulfur Recovery Plants,” EPA-450/2-76-016(a),
September 1976.
4. Beavon, David D., and Raoul P. Vaell, “The Beavon Sulfur Removal Process
for Purifying Claus Plant Tail Gas,” Proceedings, API Division of Refining,
1972, Volume 52, pp. 267-276, refer to p. 3.15, reference 12, “Standards
Support and Environmental Impact Statement, Volume I: Proposed Standards
of Performance for Petroleum Refinery Sulfur Recovery Plants,” EPA-450/
2-76-016(a), September 1976.
5. Landrum, L. H., L. H. Corn, and W. E. Fernald, “The Cleanair Sulfur Pro-
cess,” presented at the 74th AIChE Meeting, New Orleans, Louisiana,
11-15 March 1973, refer to p. 4.31, reference 12, “Standards Support and
Environmental Impact Statement, Volume I: Proposed Standards of Perform-
ance for Petroleum Refinery Sulfur Recovery Plants,” EPA-450/2-76-0l6(a),
September 1976.
6. Barthel, et a]., “IFP Processes for Recovering H 2 S and S02 From Claus Unit
Tail Gas and for Cleaning SO 2 From Stack Gas,” APCA Paper 73-304, refer to
p. 4.31, reference 14, “Standards Support and Environmental Impact State-
ment, Volume I: Proposed Standards of Performance for Petroleum Refinery
Sulfur Recovery Plants,” EPA—450/2-76-016(a), September 1976.
7. Bethea, Robert M., Air Pollution Control Technology , Van Nostrand Reinhold
Co., New York, 1978.
8. Naber, et al., “The Shell Claus Off-Gas Treating Process,” presented at the
74th AIChE Meeting, New Orleans, Louisiana, 11-15 March 1973.
9. Fisch, E. J., et al., Energy Technology Handbook , Douglas M. Considine,
Editor, McGraw-Hill Book Co., New York, New York, 1977, pp. 2-133 to 2-136.
10. Krili, H., and K. Storp, “H2S Absorbed From Tail Gas,” Chemical Engineering ,
80 (17), 23 July 1973, refer to p. 4.31, reference 4, “Standards Support and
Environmental Impact Statement, Volume I: Proposed Standards of Performance
for Petroleum Refinery Sulfur Recovery Plants,” EPA-450/2-76-0l6(a),
September 1976.
11. Genco, Joseph M., and Samuel S. Tam, “Characterization of Sulfur From Refin-
ery Fuel Gas,” EPA Contract No. 68-02-0611, Task No. 4, 28 June 1974.
-257-
-------
REFERENCES (continued)
12. Potter, Brian H., and Christopher B. Earl, “The Weliman-Lord S02 Recovery
Process,” presented to the 1973 Gas Conditioning Conference, refer to
p. 4.31, reference 6, “Standards Support and Environmental Impact State-
ment, Volume I: Proposed Standards of Performance for Petroleum Refinery
Sulfur Recovery Plants,” EPA-450/2-76-016(a), September 1976.
13. “Chemically Active Fluid Bed (CAFB) Process — Preliminary Process Design
Manual,” Foster Wheeler Energy Corporation, prepared for U. S. EPA Office
of Research and Development, Contract No. 68-02-2106, 1 April 1976.
-258-
-------
TECHNICAL REPORT DATA
(Please read Ths.rructions on the reverse before completing)
1. REPORT NO. 2.
EPA-600/7-79-l78b
3. RECIPIENT S ACCESSION NO.
4. TITLE AND SUBTITLE
Technology Assessment Report for Industrial Boiler
Applications: Oil Cleaning
5. REPORT DATE
November 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
E.A.Comley, R.T.Keen, and M. F. Tyndall
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING OROANIZATION NAME AND ADDRESS
Catalytic, Inc. .
P.O. Box 240232
Charlotte, North Carolina 28224
10. PROGRAM ELEMENT NO.
1NE825
11. CONTRACT/GRANT NO.
68-02-2604, Task 2
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Res earch Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT ND PERIO COVERED
EPA/600/ 13
15. SUPPLEMENTARY NO T ES RL.RTP project offici is Samuel L. Rakes, Mail Drop 61, 919/
541—2825.
16. D I ACT The report gives results of an assessment of the applicability of oil clean-
ing technology to industrial boilers. It gives the status of development and perfor-
mance of alternative oil cleaning techniques and the cost, energy, and environmental
impacts of the most promising processes. Hydrotreating processes (HDS, hydrode-
sulfurization) which produce cleaned liquid fuels are considered the best system of
emission reduction applicable to oil-fired industrial boilers. Processes which clean
oil by gasification are either not generally suited to the small scale of industrial
boilers (POX) or are not commercially demonstrated (CAFB), The average capital
investment, as well as the overall energy requirements, increase with increasing
degree of desulfurization. The cost impact o providing low sulfur distillate oil for
firing small commercial boilers is minimal. The cost impact of using residual fuel
oil is much more dramatic. The cost of HDS escalates quite rapidly with the degree
of desulfurization in a given oil. For the distillate oil, there is a 6.7% premium for
0. 3% S oil, and 7. 7% for 0. i% S. For residual oil the premium ranges from 6.7-.
18. 6% (for oil desulfurjzed to 1. 6% S) to 39-43. 1% (desulfurized to 0. 1% S). The cost
of HDS ranges from 0. 91/B (for 1. 6% S) to ‘5. 28/B (for 0.1% S).
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b.IDENTIFIERS!OPEN ENDED TERMS
C. COSATI Field/Group
Pollution Assessments
Fuel Oil Desulfurization
Cleaning
Gasification
Boilers
Emission
Pollution Control
Stationary Sources
Oil Cleaning
Industrial Boilers
Hydrotreating
l3B
T1H, 2lD
13H
07A
13A
14B
18. DtSTRIBUT!ON STATEMENT
Release to Public
.
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO, OF PAGES
271
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-2 59-
------- |