v/EPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
October 1984
Air
VOC
From Petroleum
Refinery
Wastewater
Systems-
Draft
EIS
Information for
Proposed Standards
Preliminary Draft
-------
NOTICE
This document has not been formally released by EPA and should not now be construed to represent
agency policy. It is being circulated for comment on its technical accuracy and policy implications.
VOC Emissions from Petroleum Refinery
Wastewater Systems— Background
Information for Proposed Standards
Emission Standards and Engineering Division
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
July 1984
-------
TABLE OF CONTENTS
Chapter/Section Page
1. SUMMARY 1-1
1.1 Regulatory Alternatives 1-1
1.2 Environmental Impact 1-2
1.3 Economic Impact 1-3
2. INTRODUCTION 2-1
2.1 Background and Authority for Standards 2-1
2.2 Selection of Categories of Stationary Sources 2-5
2.3 Procedure for Development of Standards of Performance 2-6
2.4 Consideration of Costs 2-9
2.5 Consideration of Environmental Impacts 2-10
2.6 Impact on Existing Sources 2-11
2.7 Revision of Standards of Performance 2-11
3. DESCRIPTION OF PETROLEUM REFINERY WASTEWATER SYSTEMS AND VOC
EMISSIONS 3-1
3.1 Introduction and General Information 3-1
3.1.1 Petroleum Refining Industry 3-1
3.1.2 Overview of Petroleum Refinery Wastewater Systems. 3-2
3.1.2.1 Sources of Refinery Wastewater 3-8
3.1.2.2 Future Trends in Refinery Wastewater
Generation 3-17
3.2 Petroleum Refinery Wastewater Processes and VOC
Emission 3-19
3.2.1 Process Drains 3-19
3.2.1.1 Description of Process Drain System 3-19
3.2.1.2 Process Drain Types 3-22
3.2.1.3 Junction Box Types 3-24
-------
Chapter/Section Page
3.2.1.4 Factors Affecting Emissions from Process
Drains and Junction Boxes 3-28
3.2.1.5 VOC Emissions from Process Drains 3-30
3.2.1.6 VOC Emissions from Junction Boxes 3-31
3.2.2 Oil-Water Separators 3-32
3.2.2.1 Types of Oil-Water Seperators 3-32
3.2.2.2 Major Factors Affecting VOC Emissions 3-34
3.2.2.3 VOC Emissions from Oil-Water Separator 3-42
3.2.3 Air Flotation Systems 3-47
3.2.3.1 Description of Air Flotation Systems 3-47
3.2.3.2 Factors Affecting Emissions 3-54
3.2.3.3 VOC Emissions from Air Flotation Systems 3-59
3.2.4 Miscellaneous Wastewater Treatment Processes 3-60
3.2.4.1 Intermediate Treatment Processes 3-62
3.2.4.2 Secondary Treatment Processes 3-63
3.2.4.3 Additional Treatment Processes 3-65
3.2.4.4 VOC Emissions from Miscellaneous Wastewater
Treatment Processes 3-66
3.3 Growth of Source Category 3-67
3.3.1 Process Drains and Junction Boxes 3-67
3.3.2 Oi 1 -Water Separators 3-68
3.3.3 Air Flotation 3-70
3.4 Baseline Emissions 3-70
3.4.1 Process Drains and Junction Boxes 3-70
3.4.2 Oil-Water Separators 3-71
3.4.3 Air Flotation Systems 3-77
3.5 References 3"78
n
-------
Chapter/Section Page
4. EMISSION CONTROL TECHNIQUES 4-1
4.1 Methods for Reduction of VOC Emissions 4-2
4.1.1 Process Drains and Junction Boxes 4-2
4.1.1.1 Methods for Controlling VOC Emissions 4-2
4.1.1.2 Effectiveness of VOC Emission Controls 4-4
4.1.2 Oil-Water Separators 4-13
4.1.2.1 Methods for Controlling VOC Emissions 4-14
4.1.2.2 Effectiveness of VOC Emission Controls 4-14
4.1.3 Air Flotation Systems 4-17
4.1.3.1 Methods for Controlling Emissions 4-17
4.1.3.2 Effectiveness of VOC Emission Controls 4-19
4.2 Control of Captured VOC 4-24
4.2.1 Flare Systems 4-25
4.2.1.1 Operating Principles 4-25
4.2.1.2 Factors Affecting Efficiency ...4-28
4.2.1.3 Control Efficiency 4-30
4.2.1.4 Applicability 4-32
4.2.2 Carbon Adsorption 4-32
4.2.2.1 Operating Principles 4-33
4.2.2.2 Factors Affecting Performance and
Applicability 4~34
4.2.2.3 Control Efficiency 4-37
4.2.3 Incineration 4-37
4.2.3.1 Operating Principles 4-37
4.2.3.2 Factors Affecting Performance and
Appl icabil ity 4-39
4.2.3.3 Control Efficiency 4-4Z
-------
Chapter/Section
4.2.4 Catalytic Oxidation 4-42
4.2.4.1 Operating Principles 4-42
4.2.4.2 Factors Affecting Performance and
Applicability 4-44
4.2.4.3 Control Efficiency 4-44
4.2.5 Condensation 4-45
4.2.5.1 Factors Affecting Performance and
Appl i cabi 1 i ty 4-47
4.2.5.2 Control Efficiency 4-49
4.2.6 Industrial Boilers and Process Heaters 4-49
4.2.6.1 Operating Principles 4-49
4.2.6.2 Factors Affecting Performance and
Appl i cabi 1 i ty 4-51
4.2.6.3 Control Efficiency *-w
4.3 References 4"54
5. MODIFICATION AND RECONSTRUCTION 5'1
5.1 General Discussion of Modification and Reconstruction
Provisions 5~1
5.1.1 Modification 5~1
5.1.2 Reconstruction 5~2
5 2 Applicability of Modification and Reconstruction
Provisions to VOC Emissions from Petroleum Refinery
Wastewater Systems 5~J
5.2.1 Modification 5~3
5.2.2 Reconstruction 5"4
6. MODEL UNITS AND REGULATORY ALTERNATIVES 6-1
6.1 Model Units 6"1
6.1.1 Process Drains and Junction Boxes 6-1
6.1.2 Oil-Water Separators 6'2
IV
-------
Chapter/Section Page
6.1.3 Air Flotation Systems 6-5
6.2 Regulatory Alternatives 6-7
6.3 References 6-9
7. ENVIRONMENTAL IMPACTS 7-1
7.1 Introduction 7-1
7.2 Air Pollution Impacts 7-1
7.2.1 Estimated Emissions and Percent Emission
Reduction for Model Units 7-1
7.2.2 Projected VOC Emissions for Petroleum Refinery
Wastewater System Source Category 7-3
7.2.3 Secondary Air Pollution Impacts 7-7
7.2.4 Summary of Air Pollution Impacts 7-9
7.3 Water Pollution Impacts 7-8
7.4 Solid Waste Impacts 7-8
7.5 Energy Impacts and Water Usage 7-10
7.6 Other Environmental Concerns 7-10
7.7 References 7-12
8. COSTS 8-1
8.1 Cost Analysis of Regulatory Alternatives 8-1
8.1.1 Process Drains and Junction Boxes 8-1
8.1.1.1 Regulatory Alternative II - Water Sealed
Drains and Junction Boxes 8-4
8.1.1.2 Regulatory Alternative III - Closed Drain
System 8-7
-------
Chapter/Section Page
8.1.2 Oil-Water Separators 8-11
8.1.2.1 Regulatory Alternative II - Covered
Separators 8-12
8.1.2.2 Regulatory Alternative III - Covered
Separators with Vapor Control Systems 8-12
8.1.3 Air Flotation Systems 8-15
8.1.4 Incremental Cost Effectiveness 8-18
8.2 Other Cost Considerations 8-18
8.3 References 8-23
9. ECONOMIC IMPACTS 9-1
9.1 Industry Characterization 9-1
9.1.1 General Profile 9-1
9.1.1.1 Refinery Capacity 9-1
9.1.1.2 Refinery Production 9-3
9.1.1.3 Refinery Ownership, Vertical Integration
and Diversification 9-3
9.1.1.4 Refinery Employment and Wages 9-7
9.1.2 Refining Processes 9-9
9.1.2.1 Crude Distillation 9-9
9.1.2.2 Thermal Operations 9-9
9.1.2.3 Catlytic Cracking 9-12
9.1.2.4 Reforming 9-12
9.1.2.5 Insomerization .9-12
9.1.2.6 Alkylation 9-12
9.1.2.7 Hydrotreating 9-12
9.1.2.8 Lubes 9-12
9.1.2.9 Hydrogen Manufacture 9-13
9.1.2.10 Solvent Extraction 9-13
9.1.2.11 Asphalt 9-13
9.1.3 Market Factors 9-13
9.1.3.1 Demand Determinants 9-13
9.1.3.2 Supply Determinants 9-16
9.1.3.3 Prices 9-19
VI
-------
Chapter/Section Pa9e
9.1.3.4 Imports 9-19
9,1.3.5 Exports 9-22
9.1.4 Financial Profile 9-22
9.2 Economic Impact Analysis 9-26
9.2.1 Introduction and Summary 9-26
9.2.2 Method 9-26
9.2.3 Analysis 9-29
9.2.4 Concl usi ons 9-34
9.3 Socioeconomic and Inflationary Impacts 9-38
9.3.1 Executive Order 12291 9-38
9.3.1.1 Fifth-year Annualized Costs 9-38
9.3.1.2 Inflationary Impacts -9-42
9.3.1.3 Employment Impacts 9'4^
9.3.2 Small Business Impacts - Regulatory Flexibility Act..9-43
9.4 References 9"44
vn
-------
APPENDICES Page
A. Evolution of the Background Information Document A-l
B. Index to Environmental Considerations B-l
C. Emission Source Test Data C-l
C.I Emission Measurements C-l
C.I.I Chevron, U.S.A., Inc. Refinery - El Segundo,
California C-l
C.I.2 Golden West Refinery - Sante Fe Springs,
California C-3
C.I.3 Phillips Petroleum Company, Sweeny, Texas C-41
C.2 VOC Screening of Process Drains C-41
C.3 References C-56
D. Emission Measurement and Continuous Monitoring D-l
D.I Introduction D-l
D.2 Emission Measurement Experience ; D-l
D.2.1 Air Flotation and Equalization Basin Test D-2
D.2.1.1 Vent Gas Flow Rate ,.. D-2
D.2.1.2 Total Organic Concentration Measurement D-4
D.2.1.3 Gaseous Organics Speciation D-5
D.2.1.4 Wastewater Sampling and Analysis D-6
D.2.1.5 Process Drain Screening Surveys D-8
D.3 Performance Test Methods D-9
D.3.1 VOC Concentration Measurement D-9
D.3.2 Gas Flow Measurement D-ll
D.3.3 Mass Flow D-12
D.3.4 Emission Reduction Efficiency Determination D-12
D.3.5 Performance Test Time and Costs D-12
vm
-------
Appendices Page
D.4 Monitoring Systems and Devices D-13
D.4.1 Monitoring of Vapor Processing Devices D-14
D.4.2 Monitoring of Combustion Devices D-16
D .4.2.1 Inci nerators D-16
D.4.2.2 Boilers or Process Heaters D-17
D.4.2.3 Flares D-18
D.5 References D-19
IX
-------
LIST OF TABLES
Table Page
1-1 Assessment of Environmental, Energy and Economic Impacts
for Each Regulatory Alternative Considered for Petroleum
Refinery Wastewater Systems 1-4
3-1 Classification of Refinery Wastewater Treatment Processes 3-6
3-2 Wastewater Sources and Generation Rates 3-10
3-3 Qualitative Evaluation of Wastewater Characterization by
Fundamental Refinery Processes 3-14
3-4 Factors for Calculating Emission Losses Using the Litchfield
Method 3-44
3-5 Data Used to Calculate Emission Factor 3-46
3-6 Typical DAF Design Characteristics 3-57
3-7 Summary of Results of EPA Tests on Air Flotation Systems 3-61
3-8 Projected Increase in Refinery Wastewater from 1985 to 1989 3-69
3-9 Existing State Regulations Applicable to Oil-Water Separators
in Petroleum Refineries 3-72
3-10 Summary of Baseline Control for Oil-Water Separators 3-75
3-11 Estimate of Crude Throughput at Refineries Having Varying
Emi ssion Control s - .3-76
4-1 Summary of Screening Values for Individual Drains 4-7
4-2 Summary of Emission Rates and Emission Reduction for Drains
With a Leak Rate > 100 PPM 4-8
4-3 Assumptions for Estimating Benzene Emissions from Example
Drain 4-9
4-4 Benzene Emissions from Each Drain Configuration 4-12
4-5 Physical Constants and Condensation Properties of Some
Organic Sol vents 4-46
-------
Table
6-1 Process Drains Model Unit Parameters 6-3
6-2 Oil-Water Separators Model Unit Parameters 6-4
6-3 Air Flotation Model Unit Parameters 6-6
6-4 Regulatory Alternatives 6-8
7-1 Estimated Emissions and Emission Reductions for Each
Model Unit and Regulatory Alternative 7-2
7-2 Projected VOC Emissions from New and Modified/
Reconstructed Process Drain Systems for Regulatory
Alternatives in Period from 1985 - 1989 7-4
7-3 Projected VOC Emissions from New and Modified/
Reconstructed Oil-Water Separators for Regulatory
Alternatives in Period from 1985 - 1989 7-5
7-4 Projected VOC Emissions from New and Modified/
Reconstructed Air Flotation Systems for Regulatory
Alternatives in Period from 1985 - 1989 7-6
7-5 Summary of Annual Emissions and Emission Reduction by
1989 for Source Category (New and Modified/Reconstructed
Units) 7'9
7-6 Energy Requirements and Water Demand - Regulatory
Alternative III for Process Drains and Junction Boxes,
Oil-Water Separators, and Regulatory Alternative II
for Air Flotation Systems 7-11
8-1 Components and Factors of Total Capital Investment 8-2
8-2 Components, Factors and Rate of Total Annual Cost 8-3
8-3 Cost Breakdown of Major Equipment for VOC Control on
Process Drain and Junction Box System 8-5
8-4 Annualized Cost and Cost Effectiveness of Regulatory
Alternatives for New Process Drain and Junction Box
System 8"6
8-5 Annualized Cost and Cost Effectiveness of Regulatory
Alternatives for Retrofitting a Process Drain and
Junction Box Emission Reduction System 8-8
XI
-------
Table
8-6 Basis for Buried Tank Subsystem Cost Estimate for
Regulatory Alternative III 8-9
8-7 Annual Utility Costs for Regulatory Alternatives 8-10
8-8 Annualized Cost and Cost Effectiveness of Regulatory
Alternatives for a Retrofit Control System or an
Oi 1 -Water Separator 8-13
8-9 Annualized Cost and Cost Effectiveness of Regulatory
Alternatives for New Oil-Water Separators 8-14
8-10 Cost Breakdown of Major Equipment for VOC Control for
Oil-Water Separators and Air Flotation Systems 8-16
8-11 Operating Parameters and Costs of Carbon Adsorber 8-17
8-12 Annualized Cost and Cost Effectiveness of Regulatory
Alternatives for DAF Systems 8-19
8-13 Annualized Cost and Cost Effectiveness of Regulatory
Alternatives for IAF Systems 8-20
8-14 Incremental Cost Effectiveness of Regulatory Alternatives 8-21
8-15 Statutes That May Be Applicable to the Petroleum 8-22
Refining Industry 8-21
9-1 Total and Average Crude Distillation Capacity by Year -
United States Refineries, 1973 - 1983 9-2
9-2 Percent Volume Yields of Petroleum Products by Year -
United States Refineries, 1972 - 1981 9-4
9-3 Production of Petroleum Products by Year - United States
Refineries, 1972 - 1981 9~5
9-4 Number and Capacity of Refineries Owned and Operated
by Major Companies - United States Refineries, 1983 9-6
9-5 Employment in Petroleum and Natural Gas Extraction and
Petroleum Refining by Year - United States, 1972 - 1981 9-8
9-6 Average Hourly Earnings of Selected Industries by Year -
United States, 1972 - 1981 9-10
xn
-------
Table
9-7 Estimated Gasoline Pool Composition by Refinery Stream -
United States Refineries, 1981 9-11
9-8 Refined Product Demand Projections for U.S. Refineries
Under Three World Oil Price Scenarios 1983 - 1986 - 1989 9-14
9-9 Price Elasticity Estimates for Major Refinery Products
by Demand Sector - United States, 1990 9-17
9-10 Crude Oil Production and Consumption by Year - United
States, 1970 - 1982 9-18
9-11 Average Wholesale Prices: Gasoline, Distillate Fuel Oil
and Residual Fuel Oil by Year - United States, 1968 - 1982 9-20
9-12 Imports of Selected Petroleum Products by Year - United
States, 1968 - 1982 9-21
9-13 Exports of Selected Petroleum Products by Year - United
States, 1969 - 1981 9-23
9-14 Profit Margins for Major Corporations with Petroleum
Refinery Capacity, 1977 - 1981 (Percent) 9-24
9-15 Return on Investment of Major Corporations with Petroleum
Refinery Capacity 1977 - 1981 9-25
9-16 Total Annualized Control Costs for a New Refinery,
Regulatory Alternative II 9-30
9-17 Total Annualized Control Costs for a New Refinery,
Regulatory Alternative III 9-31
9-18 DOE Projected Prices and Domestic Refinery Demand Under
Three World Oil Price Scenarios, 1989 9-33
9-19 Price and Total Demand Under Regulatory Alternatives
II and III 9'35
9-20 Changes in 1989 Price and Demand Compared with 1983
Basel i ne Level s 9-36
9-21 Summary of Fifth Year Annualized Cost by Model Unit and
Regulatory Al ternative 9-39
9-22 Range of Fifth-Year Annualized Cost of Affected Facilities 9-41
xm
-------
Table Page
C-l Summary of Daily Emission Rate Averages: Continuous
Monitoring Results, Chevron Refinery, El Segundo,
California C-5
C-2 Gas Chromatography Results from DAF System, Chevron
Refinery, El Segundo, California C-6
C-3 Gas Chromatography Results from Equalization Basin, Chevron
Refinery, El Segundo, California C-9
C-4 Gas Chromatography and Emission Rates from IAF System,
Chevron Refinery, El Segundo, California C-12
C-5 Liquid Samples Taken on 8/3/83 - Chevron Refinery, El
Segundo, California C-13
C-6 Liquid Samples Taken on 8/4/84 - Chevron Refinery, El
Segundo, Cal i forni a C-15
C-7 Liquid Samples Taken on 8/5/83 - Chevron Refinery, El
Segundo, Cal i f orni a C-16
C-8 Liquid Samples Taken on 8/8/83 - Chevron Refinery, El
Segundo, California C-17
C-9 Liquid Samples Taken on 8/9/83 - Chevron Refinery, El
Segundo, California C-20
C-10 Liquid Samples Taken on 8/10/83 - Chevron Refinery, El
Segundo, Cal if ornia C-21
C-ll Liquid Samples Taken on 8/11/83 - Chevron Refinery, El
Segundo, Cal if ornia C-22
C-12 Liquid Samples Taken on 8/12/83 - Chevron Refinery, El
Segundo, California C-26
C-13 Daily Emission Rate Averages at IAF Outlet - Golden West
Refinery, Santa Fe Springs, California C-28
C-14 Gas Chromatography Results from IAF System - Golden West
Refinery, Santa Fe Springs, California C-29
C-15 Liquid Samples Taken on 8/16/83 - Golden West Refinery,
Santa Fe Springs, California C-31
xiv
-------
Table Page
C-16 Liquid Samples Taken on 8/17/83 - Golden West Refinery,
Santa Fe Springs, California .................................... C-33
C-17 Liquid Samples Taken on 8/18/83 - Golden West Refinery,
Santa Fe Springs, California .................................... C-38
C-18 Liquid Samples Taken on 8/18/83 - Golden West Refinery,
Santa Fe Springs, California .................................... C-40
C-19 Daily Emission Rate Averages at IAF Outlets - Phillips
Petrol eum , Sweeny , Texas ........................................ C-44
C-20 Gas Chroma tog raphy Results from IAF #1 (South IAF) -
Phillips Petroleum, Sweeny, Texas ............................... C-45
C-21 Gas Chromatography Results from IAF #2 (North IAF) -
Phillips Petroleum, Sweeny, Texas ............................... C-47
C-22 Liquid Samples Taken on 9/20/83 - Phillips Petroleum,
Sweeny , Texas [[[ c~^8
C-23 Liquid Samples Taken on 9/21/83 - Phillips Petroleum,
Sweeny , Texas
C-24 Liquid Samples Taken on 9/22/83 - Phillips Petroleum,
Sweeny, Texas .......................... • ........................ c~50
C-25 Liquid Samples Taken on 9/23/83 - Phillips Petroleum,
Sweeny , Texas [[[ ^-51
C-26 Summary of Emission Rates and Emission Reduction for
Drains with a Leak Rate > 100 PPM ............................... C-53
C-27 Summary of Process Drain Screening - Golden West
Refinery, Santa Fe Springs, California .......................... C-54
C-28 Summary of Process Drains Screening - Total
-------
LIST OF FIGURES
Figure
3-1 Geographical Distribution of Petroleum Refineries
in the United States as of January 1, 1983 3-3
3-2 Block Diagram of a Petroleum Refinery Oily Waste-
water System • 3-5
3-3 Example of a Segregated Wastewater Collection and
Treatment System 3-7
3-4 Atmospheric Distillation System 3-16
3-5 Two Stage Steam Actuated Vacuum Jet System 3-18
3-6 General Refinery Drain System 3-21
3-7 Types of Individual Refinery Drains for Oily Waste-
water 3-23
3-8 Closed Drain and Collection System 3-25
3-9 Refinery Drain System Junction Boxes 3-26
3-10 Gas Trap Manhole 3-27
3-11 Oil/Water Separator 3-33
3-12 Corrugated Plate Separator 3-35
3-13 Effect of Ambient Air Temperature on Evaporation 3-38
3-14 Effects of 10% Point on Evaporation 3-39
3-15 Effect of Influent Temperature on Evaporation 3-40
3-16 Relationship Between Vapor Pressure, Wind Speed and
Loss Rate 3"41
3-17 Interaction of Gas Bubbles with Suspended Solid or
Liquid Phases 3-48
3-18 Dissolved Air Flotation System 3-49
3-19 Mechanism of an Impeller Type IAF 3-52
xvi
-------
Figure Pa9e
3-20 Mechanism of a Nozzle Type IAF 3-53
4-1 Floating Cover on an API Separator 4-15
4-2 Polyurethane Foam Seal on a Floating Cover 4-16
4-3 Example of DAF Emission Control System 4-20
4-4 Examples of DAF and IAF Control Systems 4-21
4-5 Steam-assisted Elevated Flare System 4-26
4-6 Schematic of Non-Regenerative Carbon Adsorption System
for VOC Control 4-35
4-7 Schematic of Incineration System for VOC Control 4-38
4-8 Typical Effect of Combustion Zone Temperature on
Hydrocarbon and Carbon Monoxide Destruction Efficiency 4-40
4-9 Schematic of Catalytic Oxidation System for VOC Control 4-43
4-10 Condensation System 4"48
C-l Dissolved Air Flotation System with Sample Location C-4
C-2 Equalization Basin with Sample Location c~8
C-3 Induced Air Flotation System at Chevron - El Segundo,
California c"11
C-4 Wastewater Treatment Facilities at Santa Fe Springs,
Cal ifornia c'27
C-5 Schematic Representation of the IAF Process with Sample
Points and Induced Air System: Phillips Petroleum, Sweeny,
Texas C~4Z
C-6 IAF - Outlet Sample Locations Fabricated: Phillips
Petrol eum - Sweeny, Texas c"43
xvi i
-------
1. SUMMARY
Standards of performance for new stationary sources are established
under Section 111 of the Clean Air Act (42 U.S.C. 7411), as amended in 1977.
Section 111 directs the Administrator to establish standards of performance
for any category of new stationary source of air pollution which "causes or
contributes significantly to air pollution which may reasonably be
anticipated to endanger public health or welfare."
1.1 REGULATORY ALTERNATIVES
The analysis of environmental, economic, and energy impacts were based
on consideration of three regulatory alternatives for each emission source.
The regulatory alternatives are given below:
Process Drain Systems:
Regulatory Alternative I: No additional control.
Regulatory Alternative II: Require water seals on process drains and
junction boxes.
Regulatory Alternative III: Require completely closed drain systems
with vapors vented to a control device.
Oil-Water Separators:
Regulatory Alternative I: No additional control.
Regulatory Alternative II: Require gasketed and sealed fixed or
floating roofs.
Regulatory Alternative III: Require gasketed and sealed fixed roof with
vapors vented to a control device.
Air Flotation Systems:
Regulatory Alternative I. No additional control.
Regulatory Alternative II. Require gasketed and sealed fixed roofs and
access doors.
Regulatory Alternative III. Require gasketed and sealed fixed roofs and
access doors with vapors vented to a
control device.
1-1
-------
Regulatory Alternative I requires no action. Under this alternative,
emissions would be controlled to levels established by existing State
regulations. Of the sources included in this NSPS, only oil-water
separators are regulated by existing regulations.
Requiring water seals on process drains and junction boxes will result
in emission reductions of 50 percent or more when compared to Regulatory
Alternative I. A fixed or floating roof on an oil-water separator will
result in emission reduction of 85 percent. A fixed roof on a dissolved air
flotation system will result in emission reductions of 77 percent.
Gasketing and sealing an induced air flotation system will result in a
23 percent reduction. Again, these emission reductions are those achieved
in comparison to Regulatory Alternative I.
The more stringent requirements of Regulatory Alternative III result in
a 98 percent reduction in emissions from process drain systems. A fixed
roof on an oil-water separator or dissolved air flotation system with
captured VOC vented to a control device will result in emission reductions
of 94 to 97 percent, depending on the efficiency of the control device.
Gasketing and sealing an IAF system and venting the captured VOC to a
control device will result in emission reductions of 70 to 85 percent, again
depending on the efficiency of the control device. All emission reductions
are those achieved in comparison to Regulatory Alternative I.
1.2 ENVIRONMENTAL IMPACT
Implementation of either Alternative II or Alternative III for all
three emission sources will result in a beneficial impact on air quality.
Implementation of Alternative II will reduce VOC emissions by approximately
1630 Mg/yr in 1989. This represents a 50 percent reduction below Regulatory
Alternative I. Implementation of Alternative III will reduce VOC emissions
by approximately 3055 Mg/yr in 1989. This represents a 95% percent
reduction below Alternative I. It should be noted that the regulatory
alternatives can be independently applied to each of the three emission
sources. Therefore, depending upon the specific regulatory alternative
1-2
-------
picked for each source, the actual emission reduction achieved by the NSPS
can range from 1630 Mg/yr to 3055 Mg/yr. These reductions in VOC emissions
can be accomplished without causing any adverse environmental impacts.
No water pollution impact will result from implementation of any of the
regulatory alternatives. Small quantities of water will be required if
regenerative carbon adsorbers are used to control VOC vented from oil-water
separators and air flotation systems. However, the quantity of water needed
will be insignificant.
Solid waste will be generated by carbon adsorption systems if they are
used for VOC control. Again, the amount of solid waste generated will be
minimal. Energy impacts will result only by implementing Regulatory
Alternative III. These impacts are also expected to be minimal.
Table 1-1 summarizes the environmental and energy impacts of the
regulatory alternatives. A more detailed analysis of these impacts is
presented in Chapter 7.
1.3 ECONOMIC IMPACT
The preliminary economic analysis indicates that the fifth-year
annualized costs of the most stringent regulatory alternatives for each
emission source are $6.3 million dollars. This is well below the $100
million level that Executive Order 12291 identifies as the threshold for
major regulatory actions. Additionally, the price increase and output
reduction due to the most costly alternatives are 0.1 percent and
0.03 percent, respectively.
1-3
-------
TABLE 1-1. ASSESSMENT OF ENVIRONMENTAL, ENERGY, AND ECONOMIC IMPACTS FOR EACH REGULATORY ALTERNATIVE
CONSIDERED FOR PETROLEUM REFINERY WASTEWATER SYSTEMS
Administrative
alternative
Regulatory Alternative I
Regulatory Alternative II
Regulatory Alternative III
Air
impact
0
+2
+3
Water
impact
0
0
0
Solid
waste
impact
0
0
0
Energy
impact
0
0
0
Economic
impact
0
0
-1
KEY: + Beneficial impact
- Adverse impact
0 No impact
1 Negligible impact
2 Small impact
3 Moderate impact
4 Large impact
5 Very large impact
-------
2. INTRODUCTION
2.1 BACKGROUND AND AUTHORITY FOR STANDARDS
Before standards of performance are proposed as a Federal regulation,
air pollution control methods available to the affected industry and the
associated costs of installing and maintaining the control equipment are
examined in detail. Various levels of control based on different
technologies and degrees of efficiency are examined. Each potential level
of control is studied by EPA as a prospective basis for a standard. The
alternatives are investigated in terms of their impacts on the economics and
well-being of the industry, the impacts on the national economy, and the
impacts on the environment. This document summarizes the information
obtained through these studies so that interested persons will be able to
see the information considered by EPA in the development of the proposed
standard.
Standards of performance for new stationary sources are established
under Section 111 of the Clean Air Act (42 U.S.C. 7411) as amended, herein-
after referred to as the Act. Section 111 directs the Administrator to
establish standards of performance for any category of new stationary source
of air pollution which "... causes, or contributes significantly to air
pollution which may reasonably be anticipated to endanger public health or
welfare."
The Act requires that standards of performance for stationary sources
reflect "... the degree of emission reduction achievable which (taking
into consideration the cost of achieving such emission reduction, and any
nonair quality health and environmental impact and energy requirements) the
Administrator determines has been adequately demonstrated for that category
of sources." The standards apply only to stationary sources, the construc-
tion or modification of which commences after regulations are proposed by
publication in the Federal Register.
2-1
-------
The 1977 amendments to the Act altered or added numerous provisions
that apply to the process of establishing standards of performance.
1. EPA is required to list the categories of major stationary sources
that have not already been listed and regulated under standards of
performance. Regulations must be promulgated for these new categories on
the following schedule:
a. 25 percent of the listed categories by August 7, 1980.
b. 75 percent of the listed categories by August 7, 1981.
c. 100 percent of the listed categories by August 7, 1982.
A governor of a State may apply to the Administrator to add a category not
on the list or may apply to the Administrator to have a standard of
performance revised.
2. EPA is required to review the standards of performance every four
years and, if appropriate, revise them.
3. EPA is authorized to promulgate a standard based on design,
equipment, work practice, or operational procedures when a standard based on
emission levels is not feasible.
4. The term "standards of performance" is redefined, and a new term
"technological system of continuous emission reduction" is defined. The new
definitions clarify that the control system must be continuous and may
include a low- or non-polluting process or operation.
5. The time between the proposal and promulgation of a standard under
Section 111 of the Act may be extended to six months.
Standards of performance, by themselves, do not guarantee protection of
health or welfare because they are not designed to achieve any specific air
quality levels. Rather, they are designed to reflect the degree of emission
limitation achievable through application of the best adequately demon-
strated technological system of continuous emission reduction, taking into
consideration the cost of achieving such emission reduction, any non-air-
quality health and environmental impacts, and energy requirements.
Congress had several reasons for including these requirements. First,
standards with a degree of uniformity are needed to avoid situations where
some States may attract industries by relaxing standards relative to other
2-2
-------
States. Second, stringent standards enhance the potential for long-term
growth. Third, stringent standards may help achieve long-term cost savings
by avoiding the need for more expensive retrofitting when pollution ceilings
may be reduced in the future. Fourth, certain types of standards for
coal burning sources can adversely affect the coal market by driving up the
price of low-sulfur coal or effectively excluding certain coals from the
reserve base because their untreated pollution potentials are high.
Congress does not intend that new source performance standards contribute to
these problems. Fifth, the standard-setting process should create incen-
tives for improved technology.
Promulgation of standards of performance does not prevent State or
local agencies from adopting more stringent emission limitations for the
same sources. States are free under Section 116 of the Act to establish
even more stringent emission limits than those established under Section 111
or those necessary to attain or maintain the National Ambient Air Quality
Standards (NAAQS) under Section 110. Thus, new sources may in some cases be
subject to limitations more stringent than standards of performance under
Section 111, and prospective owners and operators of new sources should be
aware of this possibility in planning for such facilities.
A similar situation may arise when a major emitting facility is to be
constructed in a geographic area that falls under the prevention of signifi-
cant deterioration of air quality provisions of Part C of the Act. These
provisions require, among other things, that major emitting facilities to be
constructed in such areas are to be subject to best available control
technology. The term Best Available Control Technology (BACT), as defined
in the Act, means
... an emission limitation based on the maximum degree of
reduction of each pollutant subject to regulation under this Act
emitted from, or which results from, any major emitting facility,
which the permitting authority, on a case-by-case basis, taking
into account energy, environmental, and economic impacts and
other costs, determines is achievable for such facility through
application of production processes and available methods,
systems, and techniques, including fuel cleaning or treatment or
innovative fuel combustion techniques for control of each such
2-3
-------
pollutant. In no event shall application of "best available
control technology" result in emissions of any pollutants which
will exceed the emissions allowed by any applicable standard
established pursuant to Section 111 or 112 of this Act.
(Section 169(3))
Although standards of performance are normally structured in terms of
numerical emission limits where feasible, alternative approaches are
sometimes necessary. In some cases physical measurement of emissions from a
new source may be impractical or exorbitantly expensive. Section lll(h)
provides that the Administrator may promulgate a design or equipment
standard in those cases where it is not feasible to prescribe or enforce a
standard of performance. For example, emissions of hydrocarbons from
storage vessels for petroleum liquids are greatest during tank filling. The
nature of the emissions, high concentrations for short periods during
filling and low concentrations for longer periods during storage, and the
configuration of storage tanks make direct emission measurement impractical.
Therefore, a more practical approach to standards of performance for storage
vessels has been equipment specification.
In addition, Section lll(i) authorizes the Administrator to grant
waivers of compliance to permit a source to use innovative continuous
emission control technology. In order to grant the waiver, the
Administrator must find: (1) a substantial likelihood that the technology
will produce greater emission reductions than the standards require or an
equivalent reduction at lower economic, energy, or environmental cost;
(2) the proposed system has not been adequately demonstrated; (3) the
technology will not cause or contribute to an unreasonable risk to the
public health, welfare, or safety; (4) the governor of the State where the
source is located consents; and (5) the waiver will not prevent the
attainment or maintenance of any ambient standard. A waiver may have
conditions attached to assure the source will not prevent attainment of any
NAAQS. Any such condition will have the force of a performance standard.
Finally, waivers have definite end dates and may be terminated earlier if
the conditions are not met or if the system fails to perform as expected.
In such a case, the source may be given up to 3 years to meet the standards
with a mandatory progress schedule.
2-4
-------
2.2 SELECTION OF CATEGORIES OF STATIONARY SOURCES
Section 111 of the Act directs the Adminstrator to list categories of
stationary sources. The Administrator "... shall include a category of
sources in such list if in his judgment it causes, or contributes signifi-
cantly to, air pollution which may reasonably be anticipated to endanger
public health or welfare." Proposal and promulgation of standards of
performance are to follow.
Since passage of the Clean Air Amendments of 1970, considerable
attention has been given to the development of a system for assigning
priorities to various source categories. The approach specifies areas of
interest by considering the broad strategy of the Agency for implementing
the Clean Air Act. Often, these "areas" are actually pollutants emitted by
stationary sources. Source categories that emit these pollutants are
evaluated and ranked by a process involving such factors as (1) the level of
emission control (if any) already required by State regulations,
(2) estimated levels of control that might be required from standards of
performance for the source category, (3) projections of growth and replace-
ment of existing facilities for the source category, and (4) the estimated
incremental amount of air pollution that could be prevented in a preselected
future year by standards of performance for the source category. Sources
for which new source performance standards were promulgated or under
development during 1977, or earlier, were selected on these criteria.
The Act amendments of August 1977 establish specific criteria to be
used in determining priorities for all major source categories not yet
listed by EPA. These are (1) the quantity of air pollutant emissions that
each such category will emit, or will be designed to emit; (2) the extent to
which each such pollutant may reasonably be anticipated to endanger public
health or welfare; and (3) the mobility and competitive nature of each such
category of sources and the consequent need for nationally applicable new
source standards of performance.
The Administrator is to promulgate standards for these categories
according to the schedule referred to earlier.
2-5
-------
In some cases it may not be feasible immediately to develop a standard
for a source category with a high priority. This might happen when a
program of research is needed to develop control techniques or because
techniques for sampling and measuring emissions may require refinement. In
the developing of standards, differences in the time required to complete
the necessary investigation for different source categories must also be
considered. For example, substantially more time may be necessary if
numerous pollutants must be investigated from a single source category.
Further, even late in the development process the schedule for completion of
a standard may change. For example, inability to obtain emission data from
well-controlled sources in time to pursue the development process in a
systematic fashion may force a change in scheduling. Nevertheless, priority
ranking is, and will continue to be, used to establish the order in which
projects are initiated and resources assigned.
After the source category has been chosen, the types of facilities
within the source category to which the standard will apply must be
determined. A source category may have several facilities that cause air
pollution, and emissions from some of these facilities may vary from
insignificant to very expensive to control. Economic studies of the source
category and of applicable control technology may show that air pollution
control is better served by applying standards to the more severe pollution
sources. For this reason, and because there is no adequately demonstrated
system for controlling emissions from certain facilities, standards often do
not apply to all facilities at a source. For the same reasons, the standards
may not apply to all air pollutants emitted. Thus, although a source
category may be selected to be covered by a standard of performance, not all
pollutants or facilities within that source category may be covered by the
standards.
2.3 PROCEDURE FOR DEVELOPMENT OF STANDARDS OF PERFORMANCE
Standards of performance must (1) realistically reflect best demon-
strated control practice; (2) adequately consider the cost, the non-air-
quality health and environmental impacts, and the energy requirements of
such control; (3) be applicable to existing sources that are modified or
2-6
-------
reconstructed as well as new installations; and (4) meet these conditions
for all variations of operating conditions being considered anywhere in the
country.
The objective of a program for developing standards is to identify the
best technological system of continuous emission reduction that has been
adequately demonstrated. The standard-setting process involves three
principal phases of activity: (1) information gathering, (2) analysis of
the information, and (3) development of the standard of performance.
During the information-gathering phase, industries are queried through
a telephone survey, letters of inquiry, and plant visits by EPA representa-
tives. Information is also gathered from many other sources, and a
literature search is conducted. From the knowledge acquired about the
industry, EPA selects certain plants at which emission tests are conducted
to provide reliable data that characterize the pollutant emissions from
well-controlled existing facilities.
In the second phase of a project, the information about the industry
and the pollutants emitted is used in analytical studies. Hypothetical
"model plants" are defined to provide a common basis for analysis. The
model plant definitions, national pollutant emission data, and existing
State regulations governing emissions from the source category are then used
in establishing "regulatory alternatives." (For the refinery wastewater
standard, there are a few deviations from this model plant and regulatory
analysis approach, as described in Chapters 6 through 8.) These regulatory
alternatives are essentially different levels of emission control.
EPA conducts studies to determine the impact of each regulatory alter-
native on the economics of the industry and on the national economy, on the
environment, and on energy consumption. From several possibly applicable
alternatives, EPA selects the single most plausible regulatory alternative
as the basis for a standard of performance for the source category under
study.
In the third phase of a project, the selected regulatory alternative is
translated into a standard of performance, which, in turn, is written in the
form of a Federal regulation. The Federal regulation, when applied to newly
2-7
-------
constructed plants, will limit emissions to the levels indicated in the
selected regulatory alternative.
As early as is practical in each standard-setting project, EPA repre-
sentatives discuss the possibilities of a standard and the form it might
take with members of the National Air Pollution Control Techniques Advisory
Committee. Industry representatives and other interested parties also
participate in these meetings.
The information acquired in the project is summarized in the Background
Information Document (BID). The BID, the standard, and a preamble
explaining the standard are widely circulated to the industry being
considered for control, environmental groups, other government agencies, and
offices within EPA. Through this extensive review process, the points of
view of expert reviewers are taken into consideration as changes are made to
the documentation.
A "proposal package" is assembled and sent through the offices of EPA
Assistant Administrators for concurrence before the proposed standard is
officially endorsed by the EPA Administrator. After being approved by the
EPA Administrator, the preamble and the proposed regulation are published in
the Federal Register.
As a part of the Federal Register announcement of the proposed
regulation, the public is invited to participate in the standard-setting
process. EPA invites written comments on the proposal and also holds a
public hearing to discuss the proposed standard with interested parties. All
public comments are summarized and incorporated into a second volume of the
BID. All information reviewed and generated in studies in support of the
standard of performance is available to the public in a "docket" on file in
Washington, D.C.
Comments from the public are evaluated, and the standard of performance
may be altered in response to the comments.
The significant comments and EPA's position on the issues raised are
included in the "preamble" of a promulgation package, which also contains
the draft of the final regulation. The regulation is then subjected to
another round of review and refinement until it is approved by the EPA
2-8
-------
Administrator. After the Administrator signs the regulation, it is
published as a "final rule" in the Federal Register.
2.4 CONSIDERATION OF COSTS
Section 317 of the Act requires an economic impact assessment with
respect to any standard of performance established under Section 111 of the
Act. The assessment is required to contain an analysis of: (1) the costs
of compliance with the regulation, including the extent to which the cost of
compliance varies depending on the effective date of the regulation and the
development of less expensive or more efficient methods of compliance;
(2) the potential inflationary or recessionary effects of the regulation;
(3) the effects the regulation might have on small business with respect to
competition; (4) the effects of the regulation on consumer costs; and
(5) the effects of the regulation on energy use. Section 317 also requires
that the economic impact assessment be as extensive as practicable.
The economic impact of a proposed standard upon an industry is usually
addressed both in absolute terms and in terms of the control costs that
would be incurred as a result of compliance with typical, existing State
control regulations. An incremental approach is necessary because both new
and existing plants would be required to comply with State regulations in
the absence of a Federal standard of performance. This approach requires a
detailed analysis of the economic impact from the cost differential that
would exist between a proposed standard of performance and the typical State
standard.
Air pollutant emissions may cause water pollution problems, and
captured potential air pollutants may pose a solid waste disposal problem.
The total environmental impact of an emission source must, therefore, be
analyzed and the costs determined whenever possible.
A thorough study of the profitability and price-setting mechanisms of
the industry is essential to the analysis so that an accurate estimate of
potential adverse economic impacts can be made for proposed standards. It
is also essential to know the capital requirements for pollution control
systems already placed on plants so that the additional capital requirements
2-9
-------
necessitated by these Federal standards can be placed in proper perspective.
Finally, it is necessary to assess the availability of capital to provide
the additional control equipment needed to meet the standards of
performance.
2.5 CONSIDERATION OF ENVIRONMENTAL IMPACTS
Section 102(2)(C) of the National Environmental Policy Act (NEPA) of
1969 requires Federal agencies to prepare detailed environmental impact
statements on proposals for legislation and other major Federal actions
significantly affecting the quality of the human environment. The objective
of NEPA is to build into the decisionmaking process of Federal agencies a
careful consideration of all environmental aspects of proposed actions.
In a number of legal challenges to standards of performance for various
industries, the United States Court of Appeals for the District of Columbia
Circuit has held that environmental impact statements need not be prepared
by the Agency for proposed actions under Section 111 of the Clean Air Act.
Essentially, the Court of Appeals has determined that the best system of
emission reduction requires the Administrator to take into account counter-
productive environmental effects of a proposed standard, as well as economic
costs to the industry. On this basis, therefore, the Court established a
narrow exemption from NEPA for EPA determination under Section 111.
In addition to these judicial determinations, the Energy Supply and
Environmental Coordination Act (ESECA) of 1974 (PL-93-319) specifically
exempted proposed actions under the Clean Air Act from NEPA requirements.
According to Section 7(c)(l), "No action taken under the Clean Air Act shall
be deemed a major Federal action significantly affecting the quality of the
human environment within the meaning of the National Environmental Policy
Act of 1969." (15 U.S.C. 793(c)(l)).
Nevertheless, the Agency has concluded that the preparation of environ-
mental impact statements could have beneficial effects on certain regulatory
actions. Consequently, although not legally required to do so by
Section 102(2)(C) of NEPA, EPA has adopted a policy requiring that environ-
mental impact statements be prepared for various regulatory actions,
2-10
-------
including standards of performance developed under Section 111 of the Act.
This voluntary preparation of environmental impact statements, however, in
no way legally subjects the Agency to NEPA requirements.
To implement this policy, a separate section in this document is
devoted solely to an analysis of the potential environmental impacts
associated with the proposed standards. Both adverse and beneficial impacts
in such areas as air and water pollution, increased solid waste disposal,
and increased energy consumption are discussed.
2.6 IMPACT ON EXISTING SOURCES
Section 111 of the Act defines a new source as ". . . any stationary
source, the construction or modification of which is commenced ..." after
the proposed standards are published. An existing source is redefined as a
new source if "modified" or "reconstructed" as defined in amendments to the
general provisions of Subpart A of 40 CFR Part 60, which were promulgated in
the Federal Register on December 16, 1975 (40 FR 58416).
Promulgation of a standard of performance requires States to establish
standards of performance for existing sources in the same industry under
Section lll(d) of the Act if the standard for new sources limits emissions
of a designated pollutant (i.e., a pollutant for which air quality criteria
have not been issued under Section 108 or which has not been listed as a
hazardous pollutant under Section 112). If a State does not act, EPA must
establish such standards. General provisions outlining procedures for
control of existing sources under Section lll(d) were promulgated on
November 17, 1975, as Subpart B of 40 CFR Part 60 (40 FR 53340).
2.7 REVISION OF STANDARDS OF PERFORMANCE
Congress was aware that the level of air pollution control achievable
by any industry may improve with technological advances. Accordingly,
Section 111 of the Act provides that the Administrator ". . . shall, at
least every four years, review and, if appropriate, revise . . ." the
standards. Revisions are made to assure that the standards continue to
reflect the best systems that become available in the future. Such
revisions will not be retroactive, but will apply to stationary sources
constructed or modified after the proposal of the revised standards.
2-11
-------
3. DESCRIPTION OF PETROLEUM REFINERY WASTEWATER
SYSTEMS AND VOC EMISSIONS
This chapter presents a description of petroleum refinery wastewater
systems. Section 3.1 provides general information about the petroleum
refining industry and also presents an overview of petroleum refinery
wastewater systems. Section 3.2 describes the processes used in the waste-
water system and emissions from these processes. Section 3.3 presents
growth estimates for the source category while Section 3.4 presents baseline
emissions from petroleum refinery wastewater treatment systems.
3.1 INTRODUCTION AND GENERAL INFORMATION
Wastewater is generated by many of the refining processes used by the
petroleum refining industry. This wastewater is collected by a plant wide
sewer system, which carries the flow to a treatment system. An introduction
to petroleum refining processes and the related wastewater collection and
treatment systems is presented in the following sections. Section 3.1.1
presents a general discussion of the petroleum refining industry, while
Section 3.1.2 covers sources of wastewater from petroleum refining.
3.1.1 Petroleum Refining Industry
The petroleum refining industry is defined by Standard Industrial
Classification (SIC) Code 2911 of the U.S. Department of Commerce. SIC
Code 2911 includes facilities primarily engaged in producing hydrocarbon
materials through the distillation of crude petroleum and its fractionation
products. As of January 1, 1984, there were 220 operating refineries in the
United States. They are distributed among 34 states with 44 percent of the
refineries located in Texas, California, and Louisiana. This represents 18,
17, and 9 percent of the total number of refineries, respectively, in these
three states. Approximately 28 percent of the total crude refining capacity
3-1
-------
is located in Texas. California contains 15 percent of the total crude
capacity while Louisiana holds 14 percent. The geographic distribution of
U.S. refineries is shown in Figure 3-1.
The refining industry in the United States has experienced a reversal
in growth trends as a result of the reduction in consumption of petroleum
products that has occurred since 1978. U.S. crude oil runs peaked at
14.7 million barrels per day in that year. Crude oil runs have decreased
each year since then reaching 12.5 million barrels per day for 1981 and
11.5 million barrels per day in early 1982. Since January 1, 1981, more
than 75 refineries have discontinued operations. It is expected that
refinery activity will recover somewhat and projections for 1985 and 1990
estimate crude oil runs of 14.4 million barrels per day and 13.4 million
2
barrels per day, respectively.
Based on the above forecasts, very few, if any, new refining facilities
will be built at undeveloped sites over the next 10 years. However, it will
be necessary for refineries to modernize and expand downstream processes at
existing refinery sites to allow increasingly heavier and higher sulfur
crude oils to be processed. This will allow for the production of lighter
and higher quality products that will be demanded by the marketplace. In
1980, approximately 15 percent of the crude processed in the United States
was heavy, with a sulfur content over 1 percent. This quantity will have to
increase as 85 percent of foreign crude reserves and 58 percent of U.S.
4
crude reserves have a high sulfur content.
3.1.2 Overview of Petroleum Refinery Wastewater Systems
Most petroleum refineries use some type of wastewater collection and
treatment system as part of their operations. These systems are designed to
collect wastewater generated during the refining process as well as storm
water run-off from the facility grounds. Wastewater is treated by various
means to remove contaminants such as hydrocarbons and phenols. The specific
design of such a system will depend on the quantity of wastewater generated,
the contaminant concentration, and the necessary level of treatment.
Generally a wastewater collection and treatment system will consist of the
5
following:
3-2
-------
, Alaska - 4
Hawaii - 2
Figure 3-1. Geographical Distribution of Petroleum Refineries in the United States
as of January 1, 1984.
-------
o A drainage and collection system;
o Gravity oil-water separators;
o Air flotation systems for further oil removal from the
separator effluent, if necessary; and
o Secondary treatment, if needed, following oil removal.
Figure 3-2 illustrates the components of an example petroleum refinery
wastewater system. As shown, wastewater is collected by individual drains
located throughout each process unit area. The drains feed into a series of
lateral sewers which converge into junction boxes. Wastewater from the
junction boxes is led to the oil-water separators by gravity flow or
pumping. These separators can either be small units which handle the flow
from one process unit or a group of process units, or they can be large
separators which handle the wastewater from the whole refinery. Air flota-
tion may also be used after the oil-water separators if secondary oil
removal is necessary. Following oil removal, secondary and tertiary treat-
ment processes can be used to further improve wastewater quality before
discharge. Refineries which dispose of wastewater by direct discharge into
surface waters must meet effluent guidelines established under the authority
of the Clean Water Act (40 CFR 419). Refineries which direct their
wastewater to a Publicly Owned Treatment Works (POTW) must meet pretreatment
standards which have also been established under the authority of the Clean
Water Act.6 Refineries may also dispose of some or all of their wastewater
78
in disposal wells, surface ponds located on site, or through contractors. '
Others not discharge any wastewater. Table 3-1 lists the various
processes which can be used by a refinery and the objectives of each
treatment stage.
A facility's wastewater system can consist of separate collection and
treatment systems each designed to handle wastewater streams containing
similar levels of contamination. ' A simplified flow diagram of a
segregated system handling four basic types of wastewater is shown in
Figure 3-3. The non-oily sewer system collects wastewater that does not
contain significant quantities of oil. This water can be directed through
3-4
-------
I
en
Process
lint f
Process
Unit
oW
Rranrh
Drain
r
i
i
L .
Branch
Drain
[] Possible Location of
1
Slop Oil
Tank
/
-,
J
Junction
Box or Collection Tank
% Possible Location of Unit
Oil Water Separator
Process Drains
,, Water Layer
Trunk
Drain ^ Oil -Hater ' -•
Separator
Solids To
Disposal
i
•
Oil -Water Separation
iqualization
Tank
Float
Treatment
Or Disposal
i
Air Flotation
Sollc
Dispc
-
— — *»•»_
/ Lagoon /
(Or (
-Equalization ) — *• Biological —*
\ Tank J Treatment
I /
1 ^V /
Is To
>sal
Air Flotation
(optional }
^ — -^
i
Miscellaneous Treatment Process
(optional )
Figure 3-2. Block Diagram of a Petroleum Refinery Wastewater System.
-------
TABLE 3-1. CLASSIFICATION OF REFINERY WASTEWATER
TREATMENT PROCESSES
Treatment
Objectives
Example Processes
Primary Treatment
Intermediate Treatment
Secondary Treatment
Tertiary Treatment
Free Oil and Suspended
Solids Removal
Emulsified Oil, Free
Oil, Suspended Solids,
and Colloidal
Solids Removal
Dissolved Organics
Removal, Reduction
in BOD and COD
Final Polishing
API Separators
Parallel Plate Separators
CPI Separators
Dissolved Air Flotation
Induced Air Flotation
Coagulation-Flotation
Coagulation-Preci pi tati on
Filtration
Activated Sludge
Trickling Filters
Aerated Lagoons
Oxidation Ponds
Rotating Biological Contacto
Carbon Adsorption
Filtration
3-6
-------
Non-Oily Water
- cooling tower blowdown (€5 and lighter)
- oil-free storm water ( from non-tank and non-process area)
- once through cooling-water (C5 and lighter)
- steam turbine condenser water
- boiler blowdown
- water treatment plant filter backwash
- roof drainage
Clean Water
Sewer
Emergency
Oil-Water
Separator
Oily Cooling Water (Light Contamination)
- cooling tower blowdown (C6 and heavier)
- once through cooling water (Cq and heavier)
- oily storm water from tank and process area
Oily Coollna ^
Water Sewer
API Separator
Air Flotation
00
Process Water (Oily-Water)
- desalter water
- tank drawoffs
- steam stripper bottoms (sour water strippers)
- cooling water from pumps and compressor jackets, glands and pedestals
- barometric condenser water
- contact process water and condenced stripping steam from fractlona-
tlon columns
Oily Water Sewer
API Separator
Air Flotation
Sanitary Waste
Sanitary Sewer
Secondary
Treatment
Figure 3-3. Example of a Segregated Wastewater Collection and Treatment System12'13
-------
oil-water separators which can remove oil from leaks or spills. The oily
cooling water sewer handles wastewater which has been lightly contaminated
with hydrocarbons from leaks in the heat exchanger equipment and from
stormwater runoff. This water can also be treated by oil separation before
it undergoes secondary treatment or is discharged. Process water
originates from a variety of processes which use water or steam, and may
contain oil, emulsified oil and various chemicals. This wastewater is
usually treated by oil separation and may require further secondary
12
treatment. Sanitary wastewater from lavatories and locker rooms must be
treated by an inplant sewage treatment facility or it can be discharged to a
local POTW.12
3.1.2.1 Sources of Refinery Wastewater. A petroleum refinery is a
complex operation consisting of a number of interdependent processes. Over
150 separate processes were identified in a 1977 EPA survey of the petroleum
15
refining industry. Each refining process consists of a series of unit
operations which cause chemical and physical changes in the feedstock or
products. Each unit operation may have different water usages associated
with it. The wastewater is generated by a variety of sources including
cooling water, condensed stripping steam, tank draw offs, and contact
process water.
The total wastewater flow generated by a refinery varies from one
refinery to another. Some of the factors which influence the quality of
wastewater produced are:
o the process configuration of the refinery;
o age of refinery and degree of good "housekeeping1 practiced
within the refinery;
o the degree of air-cooling and of wastewater reuse to minimize
the overall water demand of the refinery;
o type of cooling water system;
o whether or not the refinery handles tanker ballast water; and
o annual rainfall at the refinery.
3-8
-------
Some of the major sources of wastewater within a refinery are shown in
Table 3-2. This table provides a brief description of the specific
wastewater sources from each of these processes, the U.S. production
capacity for the process, and the estimated wastewater generation rates. As
can be seen from this table, the wastewater may not be directly discharged
to the sewer system. It may first undergo some type of treatment, such as
steam stripping for the removal of sulfides, mercaptans and phenolics.
Additionally, the discharge of cooling water blowdown from the cooling water
system can be considered an indirect discharge to the sewer system. There
are also general sources of wastewater not specific to any one process which
are not listed in the table. These sources include pump and compressor
cooling water, pump and compressor seal water, stormwater runoff, equipment
washing, steam traps, and leaks or spills.
Based on the information presented in Table 3-2, the processes which
generate the largest volume of wastewater are catalytic cracking, vacuum
distillation, crude desalting and crude/product storage. Additionally, the
wastewater streams from these processes contain high concentrations of oil,
emulsified oil and COD as shown in Table 3-3. Thus, these streams may be
the major sources of VOC compounds in the wastewater.
The specific source of wastewater within each process, as shown in
Table 3-2, will vary depending on the process design and operating
characteristics. A general evaluation can be made of some of the major
sources of wastewater, as follows:
Crude Oil and Product Storage. During storage, a water layer accumulates
below the oil and is drained off at intervals. The water layer is likely
saturated with VOC which is often carried along as a water emulsion when the
water layer is drawn off to the sewer.
Water associated with crude may come from the production unit or from
the ballast water used by tankers and product vessels. Tankers used to ship
crude and products generally use water as ballast. The crude is loaded on
top of the ballast water, most of which is displaced during loading.
However, large quantities of water may remain as emulsion. This emulsion
3-9
-------
Table 3-2.
17 18 19 20
Wastewater Sources and Generation Rates. ' ' '
Process
Crude Separation
Crude Storage
Desalting
Atmospheric
Distillation
Process
Description
Store crude oil 1n tanks
Removal of salt, water
and water soluble
compounds from crude
Separates light hydro-
carbons from crude In a
distillation column under
atmospheric pressure
U.S.
Process
Capacity
Waste Mater Sources MMB/SO
Residual water in crude >6.9
Water washing >6.9
Condensed stripping steam >6.9
from overhead accumulator
Waste Water Generation Factors (Gal/bbl)
Direct Indirect
to Via
Sewer Cooling-Tower
2.0
0.002
0.3
Direct Via Direct Via
Sour Water Chemical
Treatment Treatment Total
2.0
2.1 — 2.1
0.04 — 0.3
I
»-*
o
Gas Processing
Separates gases, such as
LPG; fuel gas; isobutane;
butylene and light
naphtha, from the light
ends of the atmospheric
distillation unit
Caustic and water wash
N/A 0.08 0.07
3.2 3.3
Vacuum
Distillation
Hydrogen
Production
Separates heavy gas oil
from the bottoms of the
atmospheric distillation
unit, under a vacuum
Produces hydrogen from
either light hydocarbons
Jet ejectors,
barometric condensers
Partial oxidation:
water quench/wash
6.9
1900.0
(MMcfd)
0.8 1.3
65.0 46.0
(MMcfd) (MMcfd)
5.2 -- 7.3
111.0
(MMcfd)
(steam-hydrocarbon
process) or heavy oils
(partial oxidation
process). Used for hydro
treating processes
Steam-hydrocarbon:
caustic and water wash
Light Hydrocarbon
g
Naphtha flydro-
desulfurization
Processing
Removes sulfur and nitro-
gen from naphtha stream
from atmospheric distil-
lation through catalytic
treatment with hydrogen
Condensed stripping
steam from overhead
accumulator
6.6a
0.06
0.4
1.4
1.9
-------
Table 3-2. (Continued)
Process
Catalytic
Reforming
Process
Description Waste Water Sources
Converts low octane Condensed stripping steam
naphthas into high octane from overhead accumulator
gasoline blending compounds
by contacting feedstock
with hydrogen over a
catalyst
U.S.
Process
Capacity
MMB/SD
3.9
Waste Water Generation Factors (Gal/bbl)
Direct
to
Sewer
0.22
Indirect
Via
Cooling-Tower
1.0
Direct Via
Sour Water
Treatment
0.004
Direct Via
Chemical
Treatment Total
1.2
Isomerization
Converts n-butane,
n-pentane and n-hexane
into their respective
isoparafflns
Caustic washer
N/A 0.24
1.0
1.2
Alkylation
Catalytlcally combines
an olefin with an
isoparaffin to form high
octane gasoline blending
compounds
Overhead accumulator on
fractionation tower,
caustic washer (sulfuric
acid alkylation process)
0.92 0.41
5.7
0.40 6.5
Middle and Heavy
Distillate
Processing
Chemical Sweeting
Chemically removes
mercaptans, hydrogen
sulfide and sulfur
Water washers, caustic N/A
washer, spent caustic
N/A
N/A
N/A
N/A N/A
Hydrodesulfuri-
zation
Removes sulfur, nitrogen
and metallic compounds
through catalytic
treatment with hydrogen
Overhead accumulator on
fractionator (steam
strippers), sour water
stripper bottoms
1.9 0.088
0.12
0.95
0.58
5.2
3.4
0.2
(kerosene)
4.1
(light
gas/oil )
Catalytic Cracker
Converts heavy petroleum
fractions to lighter
products using a high-
temperature catalytic
process
Overhead accumulators
and steam strippers on
the fractionator, catalyst
regeneration
6.0
1.1
3.0
5.4
9.5
-------
Table 3-2. (continued)
Process
Hyd roc rack ing
Process
Description
Converts heavy petroleum
Waste Water Sources
High and low pressure
U.S.
Process
Capacity
MMB/SO
0.94
Waste Water Generation Factors (Gal/bbl)
Direct
to
Sewer
0.64
Indirect
Via
Cooling-Tower
0.81
Direct Via
Sour Water
Treatment
3.0
Direct Via
Chemical
Treatment
Total
4.5
fractions to lighter
products using a cata-
lytic cracking 1n the
presence of hydrogen
separators, accumulator
on fractlonator
Lube 011 Processing
solvent refining
Removal of aroma tics,
unsaturates, naphthenes
and asphalts from lubrl-
catlng-oll base stocks
using solvents such as
furfural or phenol
Bottom from fractlonatlon O.Z3
towers, contact process (est)
water
11.0
1.6
13.0
Dewaxlng
ro
Removal of wax from
lubrlcatlng-oll base
stocks using solvents,
such as HER or propane,
under reduced temperature
conditions.
Compressor cooling
0.23(est) 5.8
6.7
12.5
Lubricating-oil
finishing
(hydrotreating)
Removes sulfur, nitrogen
and metallic compounds
through catalytic treat-
ment with hydrogen
Overhead accumulator 0.23 N/A N/A
on fractionator
N/A
N/A
Residual Hydro-
Carbon Processing
Visbreaking
Reducing the viscosity of
residual feed materials
through mild thermal
cracking
Accumulator on the
fractionator
N/A
•N/A
N/A
N/A
N/A N/A
Coking
Converts crude oil residue
and tar pitch products
Into gas, oil, and
petroleum coke by a
thermal cracking process
Contact process water and N/A
steam overhead accumulators (56 T/D)
31
2.6
0.70
6.4
-------
Table 3-2. (Continued)
Process
Deasphalting
Process
Description Waste Water Sources
Removes asphaltlc Steam jet ejectors,
materials from heavy condensers
oil and residual
fractions using solvent
extraction
U.S.
Process
Capacity
MMB/SD
N/A
Waste Water Generation Factors (Gal/bbl)
Direct
to
Sewer
N/A
Indirect
Via
Cooling-Tower
N/A
Direct Via
Sour Water
Treatment
N/A
Direct Via
Chemical
Treatment
N/A
Total
N/A
alncludes: Pretreating catalytic reformer feeds; naphtha desulfurizing; naphtha, olefin or aromatlcs saturation; straight run distillate;
other distillate; lube-oil polishing.
CO
Notes:
N/A: Not Available
MMB/SD: Million Barrels per Stream Day
-------
OJ
I
Table 3-3. Qualitative Evaluation of Wastewater Characteristics by
Fundamental Refinery Processes (21)
Fundamental Processes
Crude Oil and Product Storage
Crude 011 Desalting
Crude Oil Distillation
Thermal Cracking
Catalytic Cracking
Hyd roc rack ing
Reforming
Polymerization
Alkylation
Isomerization
Solvent Refining
Dewaxing
Hyd rot rea ting
Drying and Sweetening
BOD
1
2
1
1
2
--
0
1
1
—
--
3
1
3
COD
3
2
1
1
2
—
0
1
1
--
1
3
1
1
Phenol
--
1
2
1
3
--
1
0
0
--
1
1
—
2
Sulfide
--
3
3
1
3
2
1
1
2
--
0
0
2
0
Oil
3
1
2
1
1
--
1
1
1
--
--
1
--
0
Emulsified
Oil
2
3
3
--
1
--
0
0
0
--
1
0
0
1
Ph
0
1
1
2
3
—
0
1
2
—
1
--
2
2
Temp.
0
3
2
2
2
2
1
1
1
--
0
--
--
0
Ammonia
0
2
3
2
3
--
1
1
1
--
.-
—
0
1
Chlorides
-
3
1
2
1
—
0
1
2
--
—
--
0
0
Acidity
0
0
0
0
0
--
0
1
2
--
0
--
0
1
Alkalinity
-
1
1
2
3
_.
0
0
0
--
1
--
i
1
Susp.
Solids
2
3
1
1
1
__
0
1
2
—
--
—
0
2
3 - Major Contribution
2 - Moderate Contribution
1 - Minor Contribution
0 - Insignificant Contribution
No data
-------
often does not break and the water cannot be removed by the tanker crew. A
significant quantity often remains and is pumped along with the crude to the
22
refinery.
Crude Desalting. Desalters are a major source of oil and oil-water emulsion
p "3
loss to the refinery sewer system. An oil-water emulsion is purposely
formed in the desalter to allow salt removal. Most emulsions are likely to
pass through oil-water separators and are, therefore, potential sources of
VOC emulsions throughout the refinery wastewater system.
When the emulsion is not completely resolved into two components, an
interface of emulsion forms and builds up to the point where it is period-
ically discharged to the oily sewer system through the water outlet. Such
an emulsion interface is usually stabilized with solids from the repro-
cessing of slop oil and the use of stripped foul water. Additionally,
wastewater containing various removed impurities is discharged from the
desalter to the wastewater system. Some of these desalting processes
require holding the crude at high temperatures. The temperature of the
desalting wastewater often exceeds 95°C.22 Such high temperatures may cause
VOC to volatilize from the wastewater system.
Overhead Accumulator Fractionation Column. Overhead vapors from
fractionation columns are condensed and collected in an accumulator, as
shown in Figure 3-4. The water originates from condensed stripping steam
and residual water in the feed. The water is separated from the product in
the accumulator and discharged to the wastewater treatment system. Since
this water has been in direct contact with the product it can contain
soluble hydrocarbons.25 This type of wastewater source can be found in many
processes which use distillation for product separation. These processes
include atmospheric distillation, catalytic reforming, hydrodesulfurization,
and cracking operations.
3-15
-------
CRUOC CHARGE
GAS TO LPC
RECOVERY
SALT MATER
LSR GASOLINE
TO TREATING
NAPHTHA
GAS OIL
TOPPED CRUDE TO
VACUUM TOWER
Figure 3-4. Atmospheric distillation system.
3-16
-------
Steam Jet Ejectors/Condensers. A steam jet ejector is a device which uses
one fluid to pump another. It is usually used as a vacuum pump for
distillation columns. In this device, high velocity steam is discharged
across a suction chamber that is connected to the equipment being
pc
evacuated. Figure 3-5 shows an example of a steam jet ejector.
oc
After the ejector, a condenser can be used to condense the vapors.
This can either be a direct contact (barometric) or surface type (shell and
tube) condenser. Of the two types, barometric condensers generate the
largest quantity of wastewater, as the vapors from the column are condensed
by direct contact with a water spray. Since the water directly contacts the
26
vapors, it can contain soluble and emulsified oil.
Cooling Tower Slowdown. A portion of the water used for non-contact cooling
water must be regularly discharged in order to control the build up of
dissolved solids in the system. This water may contain VOC from leaks in
14
the heat exchanger equipment.
3.1.2.2 Future Trends in Refinery Wastewater Generation. The future
trends in petroleum refinery wastewater production depend on many variables.
These variables include future environmental regulations, new refinery
technology, new refinery feedstocks, and water reuse and conservation
practices. Environmental regulations relating to both water and air
pollution control will affect wastewater generation. More stringent water
regulations may result in further water conservation practices or addition
of wastewater treatment facilities. Regulations controlling air pollutants
from refinery boilers and process heaters may require flue gas scrubbers
which would result in additional wastewater generation.
New refinery technology is constantly being developed. Although it is
difficult to predict technology development, it can be predicted with some
certainty that refineries will become increasingly complex. Increased
complexity in a refinery has been shown to result in increased wastewater
generation. This has been demonstrated in one study which compared
wastewater production of a topping and integrated refinery.
3-17
-------
urn
svcTira
T
NONCOHOEiSMlCS TO
FWC INCIIEMTOR
IITER AND CONDENSIBLES
Figure 3-5. Two stage steam actuated vacuum jet system.
28
3-18
-------
As mentioned in Section 3.1.1., future crude supplies will be higher in
sulfur content. Processing higher sulfur crude oils will require more
hydrogen synthesis units. Hydrogen synthesis units require large amounts of
steam which will lead to increases in wastewater production. Some of the
increases in wastewater production will be offset by the trend towards water
conservation. Water conservation in a refinery will include practices such
as:
o replacement of once through cooling water systems with circulatory
systems using evaporative cooling towers;
o raising the level of concentration cycles within existing
circulatory cooling water systems by reducing the amount of
blowdown;
o more usage of air-cooling rather than water-cooling, and
o more intensive efforts to reduce water-cooling and steam heating
needs by using more process heat recovery.
3.2 PETROLEUM REFINERY WASTEWATER PROCESSES AND VOC EMISSIONS
As discussed in Section 3.1.2, a basic petroleum refinery wastewater
treatment system consists of a drain system connected to a series of
treatment steps. This section will discuss each of the major components in
this system. The sources and factors affecting emissions, and emission
estimates from major sources will be presented. The components examined
include process drain systems, oil-water separators, air flotation systems
and miscellaneous treatment processes.
3.2.1 Process Drain Systems
Although the number of process drains may vary widely among refineries
and individual process units, the general layouts of process drain systems
are similar. The process drain system, the types of process drains, and the
emissions from process drains and junction boxes are described below.
3.2.1.1 Description of Process Drain System. In petroleum refineries,
oily water from various sources enters the oily water collection system
3-19
-------
through numerous, generally small, individual process drains. Many of these
drains are open to the atmosphere. The numbers of these drains in
refineries have been estimated to be more than 1000 in some medium-sized
29 30
refineries and in excess of 3000 for some large refineries. '
The general principles of refinery drain systems are well
defined.5' ' Details of the individual drain systems do vary, however,
depending on the needs of a specific facility and on the design choices made
by individual refiners. Variations can include pipe size, type of traps,
processes handled, and type of junction boxes.
A generalized refinery drain system is conceptually illustrated in
Figure 3-6. Liquid is collected in individual small drains distributed
throughout each process unit. Some drains may be dedicated to a single
piece of equipment (e.g., a single pump), while others might serve several
sources. In some cases, these drains may be completely closed instead of
open to atmosphere. The individual drains are connected directly to lateral
sewer lines. There may be several lateral lines in a process unit. The
lateral sewers from the process drains flow into junction boxes, which
provide effective vapor seals. The vapor seals prevent hydrocarbons from
backing up into other lateral lines and confine any fire or explosion to a
small area.
The wastewater leaves the junction boxes through branch lines. Branch
lines from refinery units and processing areas generally flow through a
gas-trap manhole before entering the trunk line system. The gas-trap
manhole is often located at the boundaries of the process unit and prevents
vapor from the trunk system from backing up into the sewer lines. Manholes
also serve to isolate the individual branch lines. Because the function and
structure of junction boxes and gas-trap manholes are similar, both will be
referred to collectively as junction boxes in this document.
The trunk sewer system carries wastewater from the branch sewers to the
wastewater treatment system. The number and configuration of lateral,
branch, and trunk lines vary considerably among refineries.
Current design practice normally provides for segregated wastewater
sewers. Storm drainage systems are separated from oily water drains and
3-20
-------
REFINERY PROCESS UNIT
I
i
Lateral
Sewers
\ I
1
I
x 1
Drain
7 I
i
K
r
Risers
K
w
i
r~
i
L -
1
j
|
L-
1
I
A
Junction
Boxes
1
J
r
1
i
i
j
CO
REFINERY PROCESS UNIT
Trunk Sower
I
I Branch
I
I
Junction Box
Sewer
I
r-
I
_ _J
Junction Box
Branch
Sewer
To Waste Water
Treatment
Figure 3-6. General Refinery Drain System.
-------
sewers. Clean process water and condensate may also be drained into the
storm drains. In some cases, additional wastewater streams, such as foul
33
water, may have separate drain and sewer systems . Separate systems, such
as storm drains, may also be configured with lateral, branch, and trunk
sewers. Storm water runoff is generally collected by open troughs or sumps
covered with iron or steel grating and located below grade.
In general, the refinery sewer system is designed for gravity flow of
the liquid. Pumping of wastewater is minimized because of the tendency to
form oil-water emulsions. In cases where pumping cannot be avoided, special
pumps are used to reduce the formation of emulsions.
3.2.1.2 Process Drain Types. Several types of individual drains are
used in petroleum refineries. These types of drains are shown in
Figure 3-7. A configuration common in older refineries is shown in
Configuration A. A straight section of pipe, usually four to six inches in
diameter, extends vertically to a height of 4-6 inches above grade. The
pipe is connected directly to a lateral sewer line with the pipe directed
either straight down or at an offset. There is no liquid seal to prevent
vapors from rising from the lateral line, which is normally connected to
several other drains. Drain lines/piping from the various sources within
the process unit generally terminate just within, at, or slightly above the
mouth of the process drain. There is often more than one drain line
directed to a single drain opening.
Another drain type used in refineries is shown in Configuration B in
Figure 3-7. The straight section of the drain inlet is connected below
grade to a "P"-bend which provides a liquid seal in the individual drain.
Vapors from the downstream drainage system are prevented from escaping by
the liquid seal.
An external liquid seal arrangement is shown in Configuration C. A cap
covers the drain opening, and the bottom edge of the cap extends below the
level of the drain entrance. Liquid from the various drain pipes falls into
the drain area outside of the cap and then flows under the edge of the cap
and into the drain line. Thus, the liquid seal prevents emissions of those
3-22
-------
v///
DRAIN
PIPE
DRAIN
RISER
(ALTERNATE OFFSET
CONFIGURATION)
OPEN, UNSEALED
CONFIGURATION A
/////////// X / / / / /
DRAIN
PIP6
DRAIN
' RISER
^^ .
P-LEG SEAL
CONFIGURATION B
V / / / / /
GRAIN
PIPE
SEAL
M
//////
0
SEAL POT
CONFIGURATION C
DRAIN
PIPE
DRAIN
RISER
'//////
CLOSED DRAIN
CONFIGURATION 0
Figure 3-7 Types of Individual Refinery Drains for Oily Wastewater
3-23
-------
vapors which may be present in the downstream drainage system. A "P"-seal
is not needed in this configuration. The drain cap can be easily removed to
clean the drain entrance and drain line, if necessary.
A completely closed drain system was observed in one refinery process
34
unit. This type of drain is illustrated in Configuration D of Figure 3-7.
The drain riser extends about 12-18" above grade. The top of this riser is
completely sealed with a flange. Drain pipes are welded directly to the
riser at points between grade and the flange seal. In some cases, an
"extra" drain nozzle is also welded to the riser. This line is normally
closed with a valve, but provides access to the closed drain system for
intermittent and infrequent needs such as pump drainage. Hoses or flexible
lines can be connected to the riser valve from the liquid source.
All the drains in this system are connected through lateral and branch
drain lines to an underground collection tank. To avoid the danger of
explosion, the entire system is purged with some type of gas which does not
contain oxygen (such as refinery fuel gas or nitrogen). The underground
tank is vented to the flare system. This closed drain system prevents any
VOC emissions to the atmosphere. The complete system is shown schematically
in Figure 3-8.
3.2.1.3 Junction Box Types. Lateral and branch sewers generally flow
through trapped junction boxes before entering the trunk (and/or branch)
sewers. The purpose of the junction boxes is to permit ready access to the
sewer lines to facilitate cleaning and inspection, as well as to isolate the
branch or lateral sewers from one another. This isolation prevents the
travel of hydrocarbon vapors from one line to another and thus reduces the
area in which a fire or explosion could occur. A typical vented junction
box is shown in Figure 3-9. The junction boxes are normally vented to
prevent siphoning and vapor locks. A junction box equipped with a vent
seal pot is shown in Figure 3-9. A small amount of water flows continually
down the vent pipe and into the seal pot, assuring a continuous seal. A
third type of junction box is shown in Figure 3-10. This type of junction
box is often referred to as a gas trap manhole.
3-24
-------
PROCESS UNIT
BOUNDARY
LATERAL
DRAIN
BRANCH
SEWER
r
VAPOR TO FUEL GAS
FLARE SYSTEM PURGE
INDIVIDUAL
DRAIN
OILY WASTE PUMPED
TO INTERMEDIATE
STORAGE TANKS OR
OIL WATER SEPARATOR
UNDERGROUND
COLLECTION TANK
SUMP
PUMPS
Figure 3-8. Closed Drain and Collection System.
3-25
-------
SEAL
WATER
-VENT
GAS TIGHT
COVER
GRADE
•CONCRETE
WATER-
(a) TYPICAL JUNCTION BOX
VENT
SEAL
POT
(fa) JUNCTION BOX WITH WATER-SEAL POT
Figure 3-9 Refinery Drain System Junction Boxes
3-26
-------
Vent
Gas Tight Lids
Vent
/ *
Figure 3-10. Gas Trap Manhole.
32
3-27
-------
23
Most vents on junction boxes are at least 4 inches in diameter.
Smaller vents can develop problems such as freezing during low temperatures
or clogging from gradual deposition of scale and sediment. The vent usually
drains to the junction box and is free of excessive bends and other
obstructions which might cause blockages.
3.2.1.4 Factors Affecting Emissions From Process Drains and Junction
Boxes. VOC are known to be emitted from refinery process drains. The
factors influencing emissions are the composition of wastewater entering the
drain system, drain design characteristics, and climatic factors.
Specifically, these factors include:
o Rate of molecular diffision of compounds through air and water;
o Rate of convection;
o Solubility and vapor pressure of the compounds found in the
wastewater stream;
o Frequency and composition of wastewater discharge through the drain;
o Wastewater temperature;
o Ambient temperature;
o Wind speed;
o Length of drain or vent pipe;
o Length of water seal; and
o Concentration of compounds in the sewer vapor space and in the waste
water
No predictive theoretical or even semi-theoretical models for process
drain emissions have been published. However, some factors affecting
emissions can be evaluated by theoretical means. These factors include
diffusion and convection.
The rate at which molecular diffusion can transport volatile compounds
through air can be calculated by using the following formula:
AD pm 1-Y
NA= -^ In
A BT !'Y1
3-28
-------
Where:
N. = Flux (mole/sec)
2
A = Exposed surface area (cm )
o
pm = Molar Density (mole/cm )
By = Diffusion path length (cm)
Y.J = Initial concentration (atm)
Y = Final concentration (atm)
2
Dy = Diffusion coefficient (cm /sec)
The density and diffusion coefficient are both controlled by the
temperature of the vapor in the drain pipe. Thus, the factors which control
molecular diffusion through air are temperature, drain design, solution
density, and the concentration gradient. Since the coefficient is inversely
proportional to the diffusion path length, the greater the drain length, the
lower the flux rate. Another controlling factor is the media through which
the compound is diffusing. For example, the diffusion coefficient for
2
benzene through air is 0.085 cm /sec while the diffusion coefficient for
-5 2
benzene through water is 1.02 x 10 cm /sec.
The rate of molecular diffusion is very small and can be overshadowed
by the effects of convection. This effect was demonstrated by one study
which showed that the rate of diffusion of hexane through different size
38
openings was 1.0 to 31.7 times the calculated diffusion rate. This study
was based on the results of laboratory evaluations of the emission rates
from different size and shaped fittings placed into covered drums containing
hexane. These fittings ranged from circular open pipes to complex shaped
steel support structures. The rate was found to depend on the design of the
opening. A small covered opening had less convective flux than a complex
shaped large opening.
Another factor which may influence the convective flux is wind
speed.39 One study showed that the mass transfer coefficient for a spilled
0 78
compound is proportional to u , where u is equal to the wind speed.
Convective flux can therefore increase the total flux through an
3-29
-------
uncontrolled drain pipe. For a water sealed drain (with no VOC
contamination in the water), the molecular diffusion through the water layer
will control the mass flux and convection cannot increase this rate. Thus,
water seals can reduce VOC emissions by eliminating the effects of
convection.
The rate at which compounds can transfer across the wastewater/air
interface and the resulting equilibrium concentration will also control the
emission rate. The faster the mass transfer rate, the greater the potential
for high vapor concentrations. The state of the compounds (i.e., whether
the compound is dissolved in the wastewater or in a separate phase) will
also affect this rate. The effects of film transport can be assumed to be
negligible. To estimate the maximum potential vapor concentration, Henry's
law can then be used to estimate vapor concentrations over solutions while
the vapor pressure can be used to estimate the vapor concentration over an
immiscible phase.
The final controlling factor is the rate and composition of the
wastewater stream entering a water sealed drain. If the wastewater stream
is highly contaminated, the water seal may become saturated with the
compounds in the stream. Additionally, if the compounds are immiscible with
water, they may float on top of the water seal. In either of these cases,
the effectiveness of the water seal will be negated, and the drain will act
as if no seal were present until the VOC are weathered off or drain is
flushed with fresh water. Fresh water flowing into such a drain can flush
out any residual compounds, restoring the effectiveness of the water seal.
3.2.1.5 VOC Emissions From Process Drains. A study sponsored by the
EPA is the only study in which the emission rate from drains has been
oc
measured. A 1958 study of refinery emissions in Los Angeles County
provided an overall emission rate estimate for the combined process drain
40
and wastewater treatment system. However, this estimate was based
primarily on qualitative observations. Little, if any, quantitative
emission data were obtained. Additionally, the VOC emissions from drains
alone cannot be estimated from this information.
3-30
-------
The EPA-sponsored study of atmospheric VOC emissions in petroleum
refineries was published in 1980.36 The results of this study were used to
develop emission factors for fugitive sources, including drains, in
petroleum refineries. These factors have since been included in EPA's
41
AP-42. The emission factor for refinery drains is 0.032 (0.010, 0.091)
kg/hr-drain. The numbers in parentheses are the lower and upper limits of
the 95% confidence interval about the average value of 0.032 kg/hr-drain.
The VOC emission measurements were made on a total of 49 process
drains. The ratio of trapped (liquid-sealed) to untrapped drains in the
sampled population was not determined. These drains were sampled in 13
different refineries, and the sampled population was intended to be
reasonably representative of refinery practices in the 1976-1979 time
period. It seems probable that the majority of the drains were unsealed,
since it was not common practice to install individually sealed drains.
This is borne out in responses to inquiries of refineries by the California
Air Resources Board in 1978. The responses indicated that the majority of
the refinery drains were not equipped with liquid seals. It is assumed in
this document that the emission factor represents emissions from untrapped
drains.
3.2.1.6 VOC Emissions from Junction Boxes. There are no studies of
VOC emissions from junction boxes. For the purposes of this document, it is
assumed that all junction boxes are sealed and vented to atmosphere. Since
the diameters of the vent lines are in the same size range as those of
drains, the mechanism for VOC emissions was assumed to be the same as that
for open, untrapped drains. Under these conditions, the emission rate from
junction box vents was estimated to be the same as the emission rate from
open drains. Thus, the junction box vent emission factor is estimated to be
0.032 kg/hr-junction box.
3-31
-------
3.2.2 Oil-Water Separators
Oil-water separators are commonly used by most refineries as the
primary method of separating and removing oil from oily process water.
Since these separators remove much of the VOC with the skimmed oil, the
units following this process will have lower VOC emissions.42
Oil-water separators are the first step in the treatment of refinery
wastewaters. Most refinery layouts provide sufficient difference in
elevation between the oil-water separator and the various areas being
drained to cause the oily process waters to flow by gravity to and through
the oil-water separator. Some refineries have installed small oil-water
separators close to the source of the oily-water. This minimizes the
formation of emulsions which cannot be removed by a separator and provides
overall improvements in efficiency of VOC recovery.10'43 The operation of
oil-water separators and the emissions from this system will be discussed in
more detail in the following sections of this chapter.
3.2.2.1 Types of Oil-Water Separators. All oil-water separators rely
on the different densities of oil, water, and solids for successful
operation. Within the separator, the wastewater stream is led to a
quiescent zone where the various phases separate. Oils and solids with
specific gravities less than that of water float to the top of the aqueous
phase, while heavy sludges and solids sink to the bottom of the vessel. As
mentioned earlier, oil-water separators will not break emulsions nor will
they separate substances in solution.
The most commonly used type of oil-water separator is the American
Petroleum Institute (API) type separator. A typical API separator is shown
in Figure 3-11. In API separators, the influent wastewater passes through
trash bars and a skimmer (the forebay) before entering the quiescent zone of
the separator (main bay). In this quiescent zone, the wastewater velocity
is kept very low to prevent any turbulent mixing. Here, free oil droplets
rise to the surface where they coalesce. The resulting oil layer is then
skimmed from the water surface at the downstream end of the tank.
3-32
-------
V
OJ
Covir lotcbay
lld«lrtd
O«»|B«»«J for (mining 'ubbcr
b»lloon ilopp«r«. lo» divtrllni Howl ^nd*
fof cUinlng MpMMM Mellon. Slulc* |*IM
or B*I* v»U»« iMy b« liHl«N»d If dulrcd.
CMnuntrf oil pump
Sludge pump
h
ft
E
\
i
•«-
1
1
\
Flight Krapcr chain iprockil
OU *Mrwnm
Oil rtUntlon titl|. Ollfuilon d.ylc. (veilic.l t.tlll.)
rim
Wood fllghl Kf*p«r
«
/
OH iklrmntr
\
Flow
1
— ' ' 'S
-------
Several types of skimmers are currently used including rotary drums,
slotted pipes, and floating oil skimmers. These can be used in both the
main bay and forebay. In the main bay, slowly moving paddles or a water
spray can be used to direct the oil layer to the end of the tank where it
can be skimmed. API separators have been, for many years, constructed with
:re
49
48
reinforced concrete. However, at least one supplier offers fiberglass
packaged units.
Other separator designs have been developed that enhance the coalescing
of oil droplets and therefore improve the oil removal efficiency of the
unit. Collectively, these separators can be referred to as enhanced oil-
water separators. The most commonly used enhanced oil-water separator is
the corrugated plate interceptor (CPI).
A corrugated plate interceptor, shown in Figure 3-12, consists of a
number of parallel corrugated plates mounted from 2 to 4 cm apart at a 45°
to 50° angle to the horizontal. Between 12 and 48 plates are typically
used. Wastewater flows downward between the plates, with the lighter oil
droplets floating upward into the tops of the corrugation, where they
coalesce. The oil droplets move up the plates to form a floating layer that
49
is skimmed from the surface of the treatment tank. These systems do not
use moving paddles to collect the oil on the surface nor are sludge rakes
used.
By using these plates the effective coalescing surface area in a CPI is
increased. Thus, for the same wastewater treatment capacity a CPI will have
a smaller surface area than a corresponding API separator. This smaller
surface area enables the systems to be supplied as prefabricated units,
usually including a cover. Manufacturers offer prefabricated systems which
can handle flow rates from 2 gpm to 2,000 gpm.
3.2.2.2 Major Factors Affecting VOC Emissions Volatilization of
organic compounds from the oily surface of an oil-water separator is a
complex mass transfer phenomenon. The force behind the volatilization
process is the drive to reach equilibrium between the oil layer and the
atmosphere. This driving force can be considered to be the difference in
3-34
-------
00
00
tn
•m^w—. A<U lnl«l wtlr
\
-..^ Pteta twtmWy comlnlng ol
24 or 48 corrugti*d.
Cltan-wiur -
ouitel chtnntl
CoocflU
Figure 3-12. Corrugated Plate Separator
50
-------
partial pressure of a compound between the two phases. The rate at which
volatilization will occur per unit surface area can be assumed to be
proportional to the difference between the vapor pressure of a compound in
the liquid phase.and its partial pressure in the gas phase.
Four studies have examined the physical and chemical factors which
52
control this transfer process. One study, conducted by Litchfield , used
a small hot water bath to simulate the operating conditions of a API
separator. Tests were conducted by placing weighed pans of actual API
separator influent oil in the hot water bath. After 24 hours the pans were
rp
reweighed and the losses calculated. The results of this study related
the percent volume loss of oil in a separator to the ambient temperature,
influent wastewater temperature, and the 10 percent true boiling point of
the influent oil. The 10 percent point is an indication of the oil's vapor
pressure. The lower the 10 percent true boiling point, the higher the vapor
pressure.
52
The relationship developed by Litchfield is as follows:
V = -6.6339 + 0.0319 X -0.0286 Y + 0.2145 Z
where:
V = Percent volume loss after 24 hours
X = Ambient temperature (°F)
Y = 10% point (°F)
Z = Influent temperature (°F)
This equation predicts losses within 2.58 percent with a confidence
limit of 95 percent. These three independent variables accounted for
82 percent of the total losses. The factors not taken into account during
this study include the thickness of the oil layer, the average wind
velocity, and the surface area of the separator, all of which can affect the
emission rate.
The results of the study showed that ambient temperature had the least
effect on the percent volume of oil lost. For each 10°F increase in ambient
temperature, a 0.3 percent increase in losses was experienced, shown in
3-36
-------
Figure 3-13. As shown in Figure 3-14, a 20°F decrease in the 10 percent
point of the influent oil will increase losses by 0.6 percent. Influent
temperature had the greatest effect on the loss rate amounting to a 2.2
percent increase in losses for every 10°F increase in temperature, as shown
in Figure 3-15.
53
The second study, by Jones and Viles , concluded that the variables
controlling air emissions from API separators were the vapor pressure of the
influent oil and the wind speed over the basin. Figure 3-16 shows the
results of this study. As can be seen, an increase in either the wind
velocity or the vapor pressure will increase the emission rate.
Several other factors can also affect the VOC emission rate including
surface area of separator, time of exposure (frequency of oil skimming) and
oil layer thickness. These factors are interrelated, as the size of the
separator and frequency of oil removal will control the oil layer thickness.
This oil layer may suppress VOC emissions because the volatilization of VOC
from the oil layer will change its composition as more volatile compounds
are lost.55 If no fresh oil is mixed with the surface oil layer and the
rate at which VOC can diffuse into this layer is small, the emission rate
could decrease with time. The weathered oil layer could then act as a
blanket and suppress vapor emissions.
Two theoretical models for predicting VOC emissions from separators
were developed by the Shell Oil Company. The first model predicts the mass
transfer of VOC from an open flat oil surface into a well developed wind
profile. The air is assumed to flow over flat terrain before encountering
an oil surface that is level with the terrain. Mass transfer is assumed to
be gas phase controlling. The mass transfer coefficient is calculated based
on an eddy diffusion model that includes a logarithmic distribution of wind
speed with height.
The second model developed by Shell is based on the Sherwood-Pigford
correlation and the Colburn j factor. This correlation is based on a
boundary layer solution of momentum transfer for flow over flat plates. The
Sherwood-Pigford correlation is used to caluclate the average mass transfer
coefficient which is then used to estimte the average mass flux of VOC.
3-37
-------
-
i
ex
I
-
-------
c
a.
a
-
400
380
360
340
320
300
280
260
240
220
200
I
10 11 12 13 14 15 16 17 18 19 20
Vol % Loss
Note: Ambient Temperature of 40°F and influent
temperature of 140°F
CO
Figure 3-14 • Effect of 10% point on evaporation.
3-39
-------
180
170
160
^ 150
o
I 140
i-
c 130
OJ
3
= 120
3 110
2
g- 100
« 90
80
/
/
j/
/
>
/
^
/
/
/
/
/
0 2 46 8 10 12 14 16 18 20
Vol % Loss
Note: 10% point of 300°F and ambient air
temoerature of 40°F
Figure 3-15. Effect of influent temperature on evaporation.
52
3-40
-------
O.iS
0.16
0.14
0.12
^ 0.10
CM
0.08
JQ
2
0.06
0.04
0.02
Vapor presure (psia)
Figure 3-16.
Relationship between vapor pressure, wind speed, and
loss rate. (53)
3-41
-------
A method for applying the second model to predicting emissions from
site specific separators was also developed by Shell. This method is based
on measuring the evaporation rate of a specific liquid hydrocarbon from open
pans placed in the oil-water separator. The measured volatilization rate is
then adjusted by a series of correction factors to estimate the
volatilization rate of the separator oil. Correction factors were developed
for the boiling point of the test liquid, temperature of the liquid surface,
wind speed, height of the measurement of wind speed, and length of the
liquid surface.
3.2.2,3 VOC Emissions From Oil-Water Separators. The earliest
detailed study of VOC emissions from oil-water separators was performed in
1958 in Los Angeles County. This study estimated the emissions from
sumps, drains and API separators to range from 30 kg/1000 m of crude to
600 kg/1000 m of crude with an average refinery emission rate of 2700
CO
kg/day. Based on this average rate and a reported wastewater flow of 31.9
million gallon per day, the emission factor was 85 kg/MM gallons of
58
wastewater flow. The emission factor listed in AP-42 is based on the 600
3 41
kg/1000 m of crude value reported by the Los Angeles County study.
There have been many changes since 1958 in the quantity and quality of
wastewater generated in refineries and the associated emissions. In
addition to decreasing wastewater flow, industry has reduced the amount of
59
oil lost to the wastewater streams. These two trends would indicate that
the emission factors determined in 1958 are higher than today's or at least
that the lower end of the range is more representative of today's
operations.
Due to the large surface area size of oil-water separators and the
physical/chemical characteristics of oil, it is difficult to make direct
measurements of VOC emissions. Recent estimations of VOC emissions have
been based on the study done by Litchfield. A discussion of these emission
estimates follows:
3-42
-------
American Petroleum Institute ; The API estimated an annual
emission rate for an API separator based on the factors
shown in Table 3-4. The results, based on Litchfield's
study, showed an estimated 12 percent volume loss. This
results in an emission factor of 570 kg/MM gallons of
wastewater using an influent oil concentration of 1500 mg/L.
State of California ; The State of California, in
1979, estimated the annual emission rates for the API
separators located in their state. The bases for these
calculations are shown in Table 3-4. California estimated
that about half the separators at refineries in the state
were completely covered. From these, VOC emissions were
thought to be minimal. Most of the oil- water separator
systems at the remainder of the refineries were partially
covered. Often a covered primary separator was followed by
an uncovered seperator. For the oil-water separator systems
that were partially covered, 950 cubic meters (6000 bbls)
per day of oil entered oil water separators in the state.
The State assumed that 80 percent of the 950 cubic meters
per day of oil was recovered in the covered part of
separators. That is, 760 cubic meters per day of oil were
recovered and 190 cubic meters per day entered the uncovered
part of the separator. Litchfield's method was used to
estimate a volume loss rate of 10 percent which equals an
emission factor of 526 kg per MM gallons of wastewater for
the uncovered portion of the separators. The inlet VOC
concentration was assumed to be 2000 mg/L.
3-43
-------
to
I
TABLE 3-4. FACTORS FOR CALCULATING EMISSION LOSSES USING THE
I TTPUPTFI n MFTUnn°U»01
LITCHFIELD METHOD
Study
API
California
Ambient
Temperature
50
65
Influent
Temperature
120
110
10%
Point
300
300
Influent
Cone.
(mg/L)
1,500
2,000
Flow
(gpm)
5,000
17,500a
Refinery
Caoacity
(ms/day)
16,000
192,000b
Emission
Rate .,
(kg/ 1000 irT
of cude)
256
68
Volume
Percent
(*)
12
10
Loss
Flow of wastewater in all of California to uncovered separators.
Total State refining capacity.
-------
The emission factors developed by API and the State of California using
the Litchfield study cannot be used to calculate the current emissions from
API separators for several reasons. Both of these studies use higher
influent oil concentrations than recent industrial contacts and a review of
current data have indicated. As refineries are trying to reduce both the
quantities of wastewater generated and the amount of oil contamination, a
value of 1000 mg/L (0.1%) is a more accurate current estimate. The high
emission factor calculated by the API study was based on wastewater genera-
tion rates which have been significantly reduced since that study was
fi?
conducted. On the other hand, the California study assumed that the first
basin of the API separator was covered and estimated the emission factor
only for the second basin.
The models developed by Shell are more complex than the method
developed by Litchfield. However, these models are more applicable to site
specific applications. Additionally, neither model has been adequately
field tested. Therefore, because the Litchfield method is based on measured
test data, this method is judged to be the best available method for
estimating VOC emissions from oil-water separators.
The Litchfield equation can be used to estimate the percent volume loss
from an API separator under a set of conditions more representative of
present day refineries. The influent temperature was selected based on
actual values found at several refineries. These temperatures ranged from
90°F to 150°F. An average temperature of 120°F was selected based on this
range.63 The 10 percent point of the influent oil was assumed to be 300°F.
This is the value used in the Litchfield study which has been verified by
recent information.64 The ambient temperature is assumed to be 65°F. Based
on the variables listed in Table 3-5, a percent volume loss rate of
12.6 percent was calculated. Assuming an influent VOC concentration of
1000 mg/L (0.1%), an emission factor of 420 kg/MM gallons of wastewater was
calculated.
A recent study by the State of California estimated a wastewater to
crude throughput ratio of O.5.59 Using this estimate, the VOC emission
factor of 420 kg/MM gallon of wastewater is equivalent to 56 kg/1000 m
crude.
3-45
-------
TABLE 3-5. DATA USED TO CALCULATE EMISSION FACTOR
Ambient Temperature: 65°F
Influent Temperature: 120°F
10% True Boiling Point: 300°F
Influent Oil Concentration: 1000 mg/L
Specific Gravity: 0.85
3-46
-------
3.2.3 Air Flotation Systems
Air flotation is commonly used in refinery wastewater treatment systems
to remove free oil, colloidal solids, emulsified oil and suspended solids.
Air flotation usually follows the oil-water separator and precedes
biological treatment. The air flotation process, types of air flotation
systems, and emissions from air flotation systems are described below.
3.2.3.1 Description of Air Flotation Systems. In air flotation
systems, bubbles are formed by introducing gas or air directly into the
wastewater by mechanical means. These bubbles become attached or entrained
with free and emulsified oil, suspended solids, and colloidal solids,
causing the combined density of these substances to be less than the density
of the
liquid phase. The bubbles, therefore, create a buoyancy which allows these
substances to rise to the surface of the flotation chamber where they are
removed. The basic mechanisms by which air or gas bubbles intereact with
suspended substances are shown in Figure 3-17. '
Two types of air flotation systems are used in petroleum refinery
wastewater treatment. These are the dissolved air flotation system (DAF)
and the induced air flotation system (IAF). Both systems rely on basic
flotation principles for removing free and emulsified oil, colloidal and
suspended solids. However, the two systems have a number of mechanical and
structural differences. Each system will be described separately followed
by a general comparison of the two.
Dissolved Air Flotation. In a DAF system, wastewater is saturated with air
or gas under pressure and passed into a flotation chamber at atmospheric
pressure. The reduction in pressure results in the formation of small
bubbles which interact with colloidal and suspended solids and free and
emulsified oil, and carry these to the surface of the flotation chamber.
Here, the floated material is removed by mechanical flight scrapers. A
DAF system is shown in Figure 3-18.
3-47
-------
Precipuaooa of the
(is oo the solid or
SoMputickw
otlgtotmie
Collision of hsiaggu
bubble and suspended
Gms-bubWe
nuclei
formation
Contact angle
Gasbwbbtehu
(rown as pressure
is reieued
o
Contact
Rising air bubbk
A) Adhesion of a bubble to a solid or liquid surface
Floe structure
o
Rising gas bubbles
B) Trapping of gas bubble in a floe structure
Suspended soiids
Gas-bubble
NCW
fonnauoo
Gas bubbles are
trapped withia tbe floe
or in surface irrejuianoes
Gas bubble
Rising gas babbie
C) Incorporation of gas bubbles into floe structure
Figure 3-17. Interaction of gas bubbles with suspended solids or
liquid phases. (65)
3-48
-------
Motor & gear
00
i
(•
Skimmings hopper
Rotating skimmer blade
CD
Compressed
air
Recycle
pump
Aeration
tank
Aerated recycle water
I A( Pressure Releasing Valve
Oily-water Influent
Figure 3-18. Dissolved Air Flotation System (DAF).
-------
The DAF can be divided into a number of sub-processes: 1) pretreatment
of the waste stream, 2) solution of the gas, 3) dissolution of the gas,
4) mixing of the gas bubbles and wastestream; 5) flotation of the colloidal
and suspended solids and free and emulsified oils, and 6) removal and
disposal of the floated material. The overall design of the system varies
from site to site and depends on the needs of the refinery. Pretreatment of
the waste stream can consist of pH adjustment and/or the addition of
chemical coagulants followed by flocculation. The coagulation/flocculation
process assists flotation by breaking the colloidal suspensions and oily
emulsions in the wastewater and by forming a floe which can easily interact
with bubbles in the flotation chamber. Commonly used coagulants include
lime, ferric chloride, alum, and various-cationic polyelectrolytes. '
Air is most commonly used as the flotation gas in a DAF system.
However, nitrogen and natural gas have also been used in refinery applica-
tions.70'71 The choice of the gas is dependent on cost, availability, and
safety considerations. Nitrogen and fuel gas can reduce the likelihood of
an explosion in the flotation system.
Three principal modes are used for pressurizing and mixing gas with the
wastewater stream. In full stream pressurization, the entire influent is
pressurized, aerated, and then released to the flotation tank. In split
stream pressurization, a portion of the influent is pressurized, aerated,
and then mixed with the remainder of the influent after reduction in
pressure. And finally, recycle pressurization involves recycling a portion
of the effluent which is then pressurized and mixed with the influent after
reduction of the pressure.
DAF flotation tanks can be rectangular or circular. Retention times
and quantity of recycle water are variable. Skimming mechanisms also vary
from system to system.
Induced Air Flotation. Induced air flotation has been used extensively in
the mining industry for ore beneficiation. Only recently has the IAF been
introduced as a treatment process for refinery wastewater. In induced air
flotation, bubbles can be produced by the following techniques:
3-50
-------
(1) mechanical shear or propellers; (2) diffusion of gas through a porous
72
medium, or (3) mixing of a gas and liquid stream. The bubbles formed
interact with suspended solids and oils and carry these substances to the
surface of the IAF where they are removed by a surface skimmer. Two types
of IAF systems are commonly used for treating refinery wastewater. These
are the impeller type, which use mechanical shear, and nozzle type systems,
which mix gas and a liquid stream.
The impeller IAF is the older of the two systems. It consists of a
rotating impeller suspended between a cylindrical stand-pipe and draft tube.
Rotation of the impeller generates a liquid vortex flow pattern with a gas
liquid interface. The interface extends from the midpoint of the inner wall
of the standpipe through the interior of the impeller section down to the
upper portion of the tube axis. The gas cavity formed within the vortex
will be at sub-atmospheric pressure. As a result, gas from the vapor space
of the flotation cell is induced through gas inlet ducts into the interior
of the rotor. Impeller rotation causes liquid to circulate upward from the
bottom of the cell. The liquid and gas phases are mixed by the impeller and
gas bubbles are formed. Further gas liquid mixing occurs when the waste-
water passes through a disperser which surrounds the impeller. After
escaping the mixing region, gas bubbles enter a quiescent region of the
cell. Here, the gas bubbles attach to suspended materials and rise to the
73
surface of the cell where they are removed. The mechanisms of an impeller
IAF are shown in Figure 3-19.
The nozzle IAF is mechanically simpler than the impeller type. In the
nozzle IAF, treated effluent is recycled to the flotation cells. Air or gas
is drawn into the liquid by means of the venturi effect and bubbles are
formed through agitation of the liquid-gas mixture. The gas bubbles formed
in the nozzle type are distributed throughout the flotation cell as opposed
to the concentration of bubbles in the upper portion of the impeller type.
A nozzle type IAF is shown in Figure 3-20.
Both the nozzle and impeller IAF systems are multi-staged units usually
consisting of four flotation cells in series. Contaminant removal
efficiency increases as wastewater moves from cell to cell. Chemical
conditioning can also increase the efficiency of both IAF systems.
3-51
-------
Air Induction
4
Control Valve
Two Phase
Mixing
\
S-
tf-
J \
\
J
Float
Figure 3-19. Mechanism of an Impeller Type IAF.
:
3-52
-------
Gas Drawn Down
Standpipe
Delivery Tube From
Recirculation Pump
Float
Skimmer
Figure 3-20. Mechanism of a Nozzle Type IAF.
3-53
-------
Comparison of DAF and IAF Systems. The DAF and IAF systems have been shown
to be equally effective in removing oil and suspended solids from refinery
wastewater when operated properly. For both systems, the factors affecting
flotation performance include influent characteristics, hydraulic loading,
chemical conditioning, and the operation of the skimmer. Additionally, DAF
performance can be influenced by the recycle rate and gas pressure while the
performance of an impeller IAF is influenced by impeller speed and impeller
submergence level. A DAF is characterized by relatively quiescent flotation,
high retention times, and usage of small quantities of (dissolved) gas. An
IAF is a more turbulent system, has lower retention times, and uses large
quantities of recirculated (ambient) gas. Both systems can be improved by
chemical conditioning. A DAF, because of the quiescent flotation, may be
more suitable for use with a wide range of chemical coagulants. An impeller
IAF has a tendency to inhibit floe formation because of the sheering action
of the impellers. However, the nozzle type IAF does not subject the floe
formed to high sheering and is therefore better suited for chemical
68 73
conditioning. *
3.2.3.2 Factors Affecting Emissions. The factors affecting VOC
emissions from air flotation systems are much the same as those affecting
emissions from API separators. Five factors which are the same include:
o quantity of VOC in wastewater entering the air flotation system;
o exposed surface area of the system;
o temperature of the wastewater;
o ambient temperature; and
o wind flow across the surface of the flotation chamber.
The above factors were discussed in detail in Section 3.2.2.2. The
quantity of VOC in wastewater entering the air flotation system is dependent
on the processes preceding air flotation. Most of the light end VOC would
be expected to be removed from the wastewater in preceding processes. An
increase in the concentration of volatile compounds in the influent oil,
52 53
however, will increase the emission rate. '
3-54
-------
Factors affecting emissions which are unique to air flotation include:
o use of air or gas used for flotation; and
o physical design characteristics of the flotation system.
Use of Gas for Flotation
A factor which is unique to air flotation systems is the introduction
of a gas into the wastewater. This gas could act to strip out volatile
hydrocarbons. The factors which control the stripping rate include the
surface area available for transfer (interfacial area), air flow rate,
temperature, and residence time of stripping. This relationship can be
73
expressed as follows:
ct - S - (Co-sr(k)(A)(t)/(v)
where: C. = Final concentration (mg/L)
C = Initial concentration (mg/L)
o
S = Concentration of unstrippable compounds (mg/L)
A = Area available for transfer
V = Volume of liquid (L)
T = Residence time (min)
K = Constant
This equation assumes that the volatilization rate will follow first
order kinetics. Although first order kinetics may not be applicable to all
the compounds in the wastewater stream, it has been shown to be true for
some compounds and waste streams from petroleum refining and petrochemical
manufacturing.75'76 This equation can be simplified by assuming that the
compounds in the wastewater are completely soluble and that an overall
mass-transfer coefficient, K, can be used in place of the term (k)(A)/(V).
This coefficient is a function of many factors including air flow rate,
water temperature, and tank configuration.
3-55
-------
The relative amount of emissions due to air stripping and evaporation
was estimated by examining the properties of an example VOC, benzene.
Theoretical calculations were performed to estimate the emissions of benzene
due to air stripping as well as evaporation from a DAF system. The
operational and design characteristics of the DAF system were assumed to be
the same as an actual refinery DAF system tested by the EPA.77 The
characteristics are given in Table 3-6.
The emissions due to air stripping can be estimated by using the above
equation. The overall mass transfer coefficient was not readily available
in the literature. Experimental studies of another compound, acetone,
indicate a value of 0.006/hr for K at the low air flow used in DAF systems.
Based on this value, the mass-transfer coefficient for benzene can be
7Q
related to that for acetone by the following equation:
KB = (NpR)B2/3 (N$c)B-2/3
KA (NpR)A2/3 (Nsc)A-2/3
where:
KB = mass-transfer coefficient for benzene
K^ = mass-transfer coefficient for acetone
f N }
v PR'B = Prandtl number for benzene = 4.37
'NPR'A = Prandtl number for acetone = 22.3
/u \
v SC'B = Schmidt number for benzene = 0.299
(N }
v SC'A = Schmidt number for acetone = 0.32
Based on this equation, the mass-transfer coefficient for benzene is
0.0096/hr. Using this coefficient and the DAF parameters shown in Table
3-6, the benzene losses due to air stripping are estimated to be 0.3 kg/MM
gallons of wastewater.
3-56
-------
TABLE 3-6. TYPICAL DAF DESIGN CHARACTERISTICS
78,81
Volume of DAF System:
Influent Flow:
Recycle
Air Temperature
Wind Speed
Diameter of DAF
Area
Residence Time:
Initial Concentration:
Concentration of Unstrippable Compounds:
Air Flow Rate:
174,000 gallons
1,800 gallons/minute
520 gallons/minute
70°F
16,000 meters/hr
15.8 meters
2
197 meters
1.25 hr
700 mg/L
0 mg/L
1.5 cfm
3-57
-------
The emissions due to evaporation of benzene from the DAF system can be
estimated by using relationships developed for calculating emissions from
oil spills. One method based on mass transfer theory and laboratory
80
experiments closely agrees with field data. This equation, based on first
order kinetics, is as follows:
c . c . -(kg)(A)(P)(t)/(nt)
0
where:
C = Mass of compound remaining (mg)
C = Initial mass of compound (mg)
k = Mass transfer coefficient (/atm hr)
A = Surface area (m2)
P = Vapor pressure of compound (atm)
t = Time (hr)
n. = Total number of moles of liquid in float
and:
k = 0.0292
g
where:
y = Wind speed (m/hr)
d = Tank diameter (m)
S = Gas-phase Schmidt number =1.76
R = Gas constant = 8.206 x 10" atm m3/(mole K)
T = Temperature (K)
Based on these equations and the input variables given in Table 3-6,
the emission rate of benzene due to evaporation is estimated to be 2.6 Kg/MM
gallons. This shows that emissions due to air stripping are small (less
than 10% of total emissions) compared to the losses due to evaporation. It
3-58
-------
should be noted that the total benzene emissions of 2.9 kg/MM gallons
estimated by the theoretical calculations compares with measured emissions
of 3.1 kg/MM gallons during EPA tests. The details of these tests are
presented in Appendix C.
Design Characteristics
The physical design characteristics of air flotation systems are also
important factors influencing emissions. The flotation chamber in a DAF is
usually open to the atmosphere where ambient conditions such as wind speed
can increase volatility of the VOC. Therefore, the flotation chamber will
be the major emission point for a DAF.
IAF systems, on the other hand, are usually supplied with a cover.
This consists of a roof and two access doors on each of the four flotation
chambers. These doors can be gasketed and sealed to reduce emissions.
Further, lAF's are usually equipped with a pressure/vacuum relief valve so
that the system can be operated gas tight. One study showed that the access
doors and pressure/vacuum relief valves are the major point of emissions
82
from IAF systems.
The action of the skimmer mechanisms in both DAF and IAF systems can
also affect emissions. If a skimmer is not in operation, a film of oil will
form on the surface of the flotation tank and inhibit the release of VOC.
Constant skimming of the oil allows for greater mass transfer of VOC to the
atmosphere. The effect of skimmer operation on VOC emissions was observed
during emissions testing of a DAF.
3.2.3.3 VOC Emissions From Air Flotation Systems. Emissions from air
flotation systems were estimated from the results of EPA tests on five air
flotation systems. These tests were performed on one DAF and four IAF
systems. The details of the tests are included in Appendix C of this
document.
Three of the IAF systems and the DAF system treated oily process
wastewater while one IAF system treated only non-oily wastewater. The
influent wastewater characteristics of the DAF and three lAF's treating oily
process wastewater were similar. As expected, the influent wastewater
3-59
-------
characteristics of the IAF treating non-oily wastewater differed greatly
from the other four systems. Therefore, only emissions results from the
tests of the four systems treating oily wastewater were used to estimate an
emission factor.
The results of the four tests used to estimate the emission factor are
given in Table 3-7. It should be noted that air purging was used to test
all four systems. Therefore, the emission results represent the emission
potential of the systems rather than the actual emissions resulting from a
system operating under normal conditions. The discussion and calculations
given in the preceding sections have shown that air stripping is not a major
cause of VOC emissions from a DAF system. Since evaporation losses are the
major cause of VOC emissions, the emission potential of IAF and DAF systems
would be equal if both are considered to have flotation chambers open to the
atmosphere. The air purging of the systems during the tests created
conditions similar to those that would exist if both types of systems were
open to the atmosphere.
As shown in Table 3-7, the VOC emissions measured at these systems
varied over a wide range. This variation could be due to design and
operational differences between the systems, differences in the concentra-
tion of hydrocarbons in the wastewater, or differences in the purge rate
used during the tests. Therefore, to account for these variations and due
to the fact that the emission tests represent emission potential, an average
uncontrolled emission factor was calculated. This uncontrolled emission
factor for air flotation systems is 15.2 kg/MM gallons of wastewater.
However, as discussed previously, an IAF does not normally operate in a
completely uncontrolled state because a cover is usually provided. The
emission factor for an IAF under normal operating conditions is estimated to
be 3.0 kg/MM gallons of wastewater. The derivation of this emission factor
is presented in Section 4.1.3.2.
3.2.4 Miscellaneous Wastewater Treatment Processes
Following oil-water separation and air flotation, wastewater streams
can be further treated by a number of processes as shown in Table 3-1 and
3-60
-------
TABLE 3-7. SUMMARY OF RESULTS OF EPA TESTS ON
AIR FLOTATION SYSTEMS78'83'84
Refi nery
Chevron
Golden West
Phillips
Phillips
Air Flotation
Type
DAF
IAF
IAF
IAF
Emission
Factor (kg VOC/MM
gal Wastewater)
30.0
21.2
5.0
4.5
T572
3-61
-------
Figure 3-2. The majority of the oil and VOC in the wastewater is removed in
primary and intermediate treatment. Hence, the potential for VOC emissions
from the treatment processes which follow is greatly reduced. There may be
situations, however, where a processs such as equalization precedes air
flotation. In these situations, the emission potential may be higher. A
brief description of the miscellaneous treatment processes is given below.
3.2.4.1 Intermediate Treatment Processes. The intermediate treatment
processes discussed in this section include coagulation-precipitation,
filtration, and equalization. Air flotation, which represents about 75
percent of the intermediate treatment processes, has been discussed in
detail in Section 3.2.3. Coagulation-precipitation and filtration remove
emulsified oil and suspended solids which have not been removed in the
primary treatment processes. Equalization is used to balance the quantity
and quality of the wastewater before entering downstream treatment.
Coagulation-Precipitation. Coagulation-precipitation begins with the
addition of chemical coagulants to the wastewater. Chemicals used for
coagulation include lime, ferric chloride, alum, and various cationic
polymers. The wastewater and coagulant are then rapidly mixed in a tank
which is followed by slow agitation of the mixture in a flocculation
chamber. The coagulant breaks the oily emulsion by reducing charge
repulsion between particles and allowing the particles to combine and form a
floe structure. The floe particles are then allowed to settle or float by
QC
gravity in a precipitation or sedimentation tank.
Filtration. Filtration can be used as both an intermediate treatment
process and as a polishing step. Several types of filtration devices have
been developed for removing free and emulsified oil from refining waste-
waters. These filters range from units using a simple sand medium to those
og
containing media which exhibit specific affinities for oil.
The filtering medium is usually contained within a basin or tank and is
supported by an underdrain system. The underdrain system allows the
3-62
-------
filtered water to be drawn off while retaining the filter medium in place.
The filter must be frequently backwashed to prevent a buildup of solids in
the medium which would reduce the filtration rate. The spent backwash water
87
contains the suspended solids removed from the water and must be treated.
Equalization. Flow equalization is used to balance the quantity and
quality of wastewater before further treatment. Equalization has been found
to greatly improve treatment results. Biological processes as well as
physical-chemical systems operate more efficiently if the composition and
flow of the wastewater feed is relatively constant. Periodic and unpre-
dictable large discharges can occur in any refinery. Equalization basins
act to minimize the effects of these increased loadings on downstream
treatment processes.
The size of an equalization system is dependent on the storage capacity
required. Tanks or basins may be used. Equalization basins can consume
large land areas. They are often aerated to maintain aerobic conditions in
the wastewater and to alleviate odor problems.
3.2.4.2 Secondary Treatment Processes. The secondary treatment
processes which will be discussed include activated sludge, trickling
filters, aerated lagoons, oxidation ponds, and rotating biological
contactors. Secondary treatment processes are used to remove dissolved
organics through oxidative decomposition by microorganisms. The processes
used in each refinery are determined by the flow and contaminant
88
characteristics of the wastewater to be treated.
Activated Sludge. Activated sludge is a continuous flow, biological
treatment process which uses microorganisms to remove organic materials by
biochemical synthesis and oxidative reaction. The microorganisms convert
the organics to carbon dioxide, water, and new cell material. The process
is carried out in a reaction tank where the wastewater is mixed with the
microorganisms in the presence of oxygen. Oxygen is supplied to the tank
either by mechanical aerators or a diffused air system. A clarification
3-63
-------
tank follows the reaction tank to allow for liquid-solids separation. A
portion of the microorganisms settled out in the clarifiers is recycled to
the reaction tank while the excess is sent to sludge handling
facilities.88*89
Trickling Filters. Trickling filters can be used as complete secondary
treatment processes or as pretreatment devices to reduce the organic load on
subsequent activated sludge units. A trickling filter consists of a large,
open topped vessel containing a packed medium that provides a growth site
for microorganisms. Wastewater is usually applied to the medium by a rotary
distributor and the treated wastewater is collected in an underdrain system.
Soluble organics are consumed by the microorganisms and converted to carbon
90
dioxide, water, and new protoplasm.
Aerated Lagoons. Aerated lagoons are medium depth basins (about 10
feet) designed for the biological treatment of wastewater on a continuous
basis. Oxygen is supplied to the lagoon by mechanical devices such as
surface aerators and submerged turbine aerators. Microorganisms convert
dissolved or suspended organics in the wastewater to stable organics, carbon
dioxide, and water. Aerated lagoons are often used as a polishing step
following removal of organics.
Oxidation Ponds. The depth of an oxidation pond is normally limited to
three to four feet to assure an adequate supply of oxygen so that aerobic
conditions are maintained without mechanical mixing. Aeration is achieved
by oxygen transfer at the surface and by the photosynthetic action of algae
present in the pond. Microorganisms then cause aerobic degradation of
91
organic contaminants in the wastewater.
Oxidation ponds have been used in the past as the only treatment of
refinery waste and also as a polishing step for the effluent from physical-
chemical or other biological waste treatment processes. Multicellular ponds
are used in some instances, especially if the oxidation pond is used as a
92
basic treatment unit rather than polishing unit.
3-64
-------
Rotating Biological Contactor. A rotating biological contactor (RBC)
is a mechanical process that brings wastewater, air, and microorganisms
together for biological oxidation. This process consists of a series of
closely spaced discs (10-12 feet in diameter) which are mounted on a
horizontal shaft and rotated with about one-third of the surface immersed in
the wastewater. The discs are typically constructed of light-weight
plastic. When the process is placed in operation, the microbes in the
wastewater begin to adhere to the rotating surfaces and grow there until the
entire surface area is covered with a 1/16-1/8 inch layer of biological
growth. As the discs rotate, they carry a film of wastewater into the air
where it trickles down the surface of the discs, absorbing oxygen. Upon
completion of a rotation, the aerated and partially treated wastewater is
mixed with the balance of the wastewater. This adds to the dissolved oxygen
content and reduces the concentration of organic matter in the tank. BOD
removal and oxidation of ammonia nitrogen is inversely proportional to the
90
hydraulic loading on the disc units.
3.2.4.3 Additional Treatment Processes. Following secondary
treatment, a number of processes are used to remove dissolved organics and
suspended solids that remain in the wastewater. These processes include
clarification, polishing ponds, and carbon adsorption. Filtration, which
has been described under intermediate treatment, may also be used in this
stage of treatment.
Clarification is used to remove suspended solids by gravity separation
and always follows biological treatment systems. Clarification tanks can be
circular or rectangular in shape and have a depth of up to 15 feet. The
settled solids are transported along the bottom of the tank by a scraper
mechanism. When following an activated sludge system, clarification helps
to produce a concentrated return sludge flow which helps to sustain
biological treatment.93 Polishing ponds also remove suspended solids by
gravity separation. The depth of a polishing pond is usually 3 to 5 feet.
Carbon adsorption can be used to remove non-biodegradable and toxic
organics which may be present in the wastewater after biological treatment.
3-65
-------
Activated carbon systems have functioned both as polishing units following a
biological system and as the major treatment process in a physical/chemical
treatment system. However, the use of activated carbon adsorption processes
94 95
has not been widespread for refinery wastewater treatment. '
3.2.4.4 VOC Emissions for Miscellaneous Wastewater Treatment
Processes. The majority of the oil in a refinery wastewater is removed by
the oil-water separator. The effluent leaving the oil-water separator
usually contains oil and grease concentrations less than 200 mg/1.
Concentrations may be higher or lower at some plants depending on the design
of the system and the retention time of the wastewater in the oil water
separator. In general, separators can remove 50 to 99 percent of the
90
separable oil in a refinery wastewater.
Because the concentrations of oil and other pollutants are highest when
entering the separator, the greatest potential for VOC emissions from
treatment processes would be from that source. Air flotation systems often
follow oil-water separators. Due to their location in the treatment scheme
air flotation is the next largest potential source of VOC emissions. As
wastewater continues to move through the treatment scheme, additional
quantities of pollutants are removed and the quality of the wastewater
improves. Secondary treatment processes also remove organic material by
biological means which further reduces the potential for air emissions.
A limited amount of emissions data are available for the treatment
processes discussed in this section. One study estimated VOC emissions from
an activated sludge system while a second study described a theoretical
method for estimating emissions from oxidation ponds.
In estimating VOC emissions from an activated sludge system, the air
stripping rate for organics in a typical refinery wastewater was calculated.
The wastewater flowing to the activated sludge system was assumed to have a
chemical oxygen demand (COD) of 600 mg/1. Using these two parameters, mass
VOC emissions were calculated for a 90,000 barrel per day refinery. The
calculated emission factor was 0.006 pounds per barrel of crude throughput
(17 kg per thousand cubic meters).76 This emission factor is based on
3-66
-------
wastewater flow of 50 gallons per barrel of crude. Using the estimated
wastewater flow to crude ratio of 0.5, the emission factor would 0.0025
pounds per barrel of crude. Due to the aeration mechanism and retention
time common in activated sludge systems, this factor can be assumed to
represent the maximum emissions which would result from all of the treatment
processes following oil removal. Very little, if any, VOC would remain in
the wastewater following activated sludge treatment.
One study indicated that VOC emissions from oxidation ponds can be
estimated by determining the surface area of the pond, the concentration of
the various organic compounds in the wastewater, the molecular weight of the
compounds, and by calculating the overall mass transfer coefficient of each
compound. Actual examples of emissions from oxidation ponds used to treat
refinery wastewaters were not given.
3.3 GROWTH OF SOURCE CATEGORY
This section present growth estimates for each emission source in the
source category. Section 3.3.1 will discuss growth estimates for process
drains and junction boxes. Section 3.3.2 and 3.3.3 will discuss growth
estimates for oil-water separators and air flotation systems, respectively.
3.3.1 Process Drains and Junction Boxes
Estimates of new process drains and junction boxes can be made by
evaluating projected refinery construction. Available sources indicate that
approximately 102 new process units will be built in the five year period
from 1985 to 1989.98'99'100 These new process units will include
approximately 4,900 new drains and 1,000 new junction boxes. In addition to
new units, it is also expected that a number of process units will be
expanded and/or modified.98 Approximately 180 process units will be
expanded and/or modified by 1989. It is estimated that 10 percent of the
drain systems of these process units will be affected by the
modification/reconstruction provisions of the NSPS. Therefore,
approximately 5,800 drains and 1,200 junction boxes will be affected by the
NSPS in the five year period from 1985 to 1989.
3-67
-------
3.3.2 Oil-Water Separators
An estimate of new oil-water separators to be built from 1985 to 1989
can be made by evaluating new construction and expansion of existing
refineries. New process units and expansion of existing process units will
result in additional wastewater generation. Using 1983 construction
projections, it is estimated that approximately 136,000 barrels per day
(5.7 MMgpd) of wastewater will be produced by new process units and
101 99
expansion of existing process units. ' Table 3-8 lists these expected
increases for some of the major refinery process units. These units will
account for approximately 124,000 barrels per day of new wastewater. It is
estimated that additional new process units and auxiliary refinery
operations will produce an additional 10 percent increase in wastewater.
Therefore, the total estimated annual increase in wastewater production is
136,000 barrels per day. It is assumed, based on projected construction
rates, that similar wastewater production increases can be expected each
year from 1985 to 1989.
Closer analysis of construction projections shows that a large portion
of the new process units will not significantly increase wastewater
generation at a specific refinery. Unused capacity of existing separators
should handle any small increases in wastewater. However, there are a
number of major construction projects planned which may warrant additional
oil-water separators. These projects include greenfield refineries and
expansion of existing refineries to handle heavy, sour crudes. Large
separators may be needed to treat wastewater produced by these projects.
Further, some refineries use unit oil-water separators to recover oil at the
source of generation. Addition of new process units will therefore call for
the addition of some smaller separators.
Based on projected refinery construction and subsequent wastewater
increases, it is estimated that 30 new oil-water separators can be expected
over the five-year period from 1985-1989. The majority of these separators
are expected to be small in size because most of the constructions projects
are minor. A few large separators will be required by major projects.
Additionally, it was assumed that another 10 percent (3 oil-water separators)
may become modified affected facilities during this time period.
3-68
-------
TABLE 3-8. PROJECTED ANNUAL INCREASE IN REFINERY WASTEWATER FROM 1985 TO 1989
Process
Increased Increased
Capacity From Capacity From
New Units (Mbbl/d) Expansion (Mbbl/d)
Wastewater Increase
Generation In Wastewater
Factor (gal/bbl) (thousand gal/day)
Hydrotreating
Hvdrorefininq
146
136
4.0 584
Light Ends
Cat Reform/PIatformer
Vacuum Distillation
Hydrogen (MM cfd)
Lube Oil
Alkylation
Cat Polymerization
Thermal Cracking/Coking
Hydrocracking
Crude Distillation
FCC
75
243.7
7.7
11.0
61.2
13.0
80.0
101.0
23.7
142.0
95.0
20.1
101.7
99.8
83
19.5
2.5
1.2
7.3
111.1 (MM cfd)
12.1
6.5
6.4
4.4
3.4
9.5
118
1,037
37.6
180.7
1,042
496
554
1,144
5,194 M gal/day
(124,000 bbl/day)
-------
3.3.3. Air Flotation
Although addition of a new oil-water separator may not necessarily
warrant a new air flotation system, increases in wastewater generation may
result in some refineries adding air flotation. Further, air flotation
alone may be added in an effort to upgrade existing wastewater treatment
facilities. Estimates of new air flotation systems can be derived using the
growth estimates for oil-water separators. Available information indicates
that approximately 75 percent of the operating refineries use air flotation
in their wastewater treatment systems.
Assuming that the number of new air flotation systems will be about
75 percent of the new oil-water separators, it is estimated that 25 new air
flotation systems will be built over the five-year period from 1985-1989.
Modified air flotation systems are assumed to equal approximately 10 percent
of the new air flotation systems (i.e. 3 air flotation systems).
3.4 BASELINE EMISSIONS
The baseline emission level is the level of control that is achieved by
industry in the absence of NSPS. Baseline reflects the emission controls
currently required by state regulations. Section 3.4.1 will discuss
baseline control for process drains and junction boxes. Sections 3.4.2 and
3.4.3 will discuss baseline control for oil-water separators and air
flotation systems, respectively.
3.4.1 Process Drains and Junction Boxes
There are presently no specific state regulations controlling VOC
emissions from process drains and junction boxes. A few refineries do exist
that apply various levels of control to process drains for emission offset
purposes. These control measures include water sealed or capped drains.
However, due to absence of state regulations, new drain systems may or may
not use any control measures. Therefore, baseline control for process
drains and junction boxes is assumed to be no control.
Current nationwide VOC emissions from process drains can be estimated
by applying the emission factor given in Section 3.2.1.5 to an estimate of
3-70
-------
the national drain population. The nationwide drain population can be
qc
estimated by extrapolating data from the EPA study and the California
study.30 The uncontrolled emission rate of VOC from an estimated 145,940
drains is 40.6 gigagrams per year (Gg/yr), with an approximate 95 percent
confidence interval range of 6.6 to 174.2 Gg/yr. This estimate does not
include the uncertainty in the estimate of total drain population.
Current nationwide VOC emissions from junction boxes can be estimated
by applying the emission factor given in Section 3.2.1.6 to the nationwide
junction box population. Based on information collected in the California
study30, it is estimated that one junction box is needed for every six
drains. Therefore, the number of junction boxes nationwide is one sixth the
number of drains, or approximately 24,300. The estimated VOC emission rate
from junction boxes is therefore 6.8 Gg/yr.
Based on the emission factors presented in Sections 3.2.1.5 and 3.2.1.6
and the growth projections presented in Section 3.3.1, the baseline
emissions from process drains and junction boxes in the 120 new, modified,
and reconstructed process units will be 1920 Mg per year in 1989.
3.4.2 Oil-Water Separators
Nearly all states where petroleum refineries are presently located have
some regulations controlling VOC emissions from oil-water separators. These
regulations vary considerably due to provisions for various exemptions in
many states. Table 3-9 provides an overview of existing state regulations
applicable to oil-water separators. As shown in the table, some states have
designated minimum separators capacity, emission level, or vapor pressure as
criteria for coverage by regulations.
As a result of state regulations, separators can generally be divided
into three classes. State regulations may require separators to be fully
covered, partially covered, or they may not be regulated. In order to
determine the proportion of each type of separator, state agencies in major
oil refining states were contacted. In addition, information on individual
refineries in a number of states was compliled. Table 3-10 summarizes the
information obtained.
3-71
-------
Table 3-9. Existing State Regulations Applicable To Oil-Water Separators
In Petroleum Refineries
Alabana
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
X
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
ATTAINMENT HO NO COVER COVER MINIMUM SIZE
AREA SOURCES REGULATION SEPARATORS FOREBAYS OTHER CUTOFF
X sources with potential
to emit < 100 TPY
X
X sources with potential
to emit < 100 TPY
X
X X
X
X
X emits < 10 Ib/day
X emits < 15 Ib/day
and < 3 Ib/hr
X sources with potential
to emit < 100 TPY
X
X
X
X
X
X sources with potential
to emit < 100 TPY
X recovers <_ 200 gal /day
f
X sources with potential
to emit < 100 TPY
X
X
X receive > 200 gal/day
VOC
X
X
X
X
X
X
X source with potential
to emit < 100 TPY
X
X
XX >. 200 gal /day
recovered
X
X
X > 200 gal /day
recovered
NOTES
a
b
c
d
e
g
h
i
3-72
-------
Table 3-9. Continued
ATTAINMENT NO
AREA SOURCES
Oklahoma
Oregon
Pennsylvania
Rhode Island X
South Carolina X
South Dakota X
Tennessee
Texas
Utah
Vermont X
Virginia
Washington
West Virginia
Wisconsin
Wyoming X
District of Columbia X
TOTALS 10 10
NO COVER COVER MINIMUM SIZE
REGULATION SEPARATORS FOREBAYS OTHER CUTOFF NOTES
X
X
X receive > 200 gal /day
VOC
a
X X
X receive > 200 gal/day e,k
VOC
X 1
X emissions > 7.3 tons/vr, m
40 Ib/day, and 3 Ib/hr
X emissions < 25 TRY
X
X
2 25 4 2
NOTES
a. No 100 TRY sources exist.
b. California's regulations vary by Air Quality Management Districts (AQMD). Bay Area AQMD exempts separators
processing < 200 gal/day organic liquids or organic liquids with Reid vapor pressure <_ 0.5 psi. San Diego County
has no sources. South Coast AQMD exempts units which handle only coal tar products and gravity separators used
exclusively for the production of crude oil if the water fraction entering contains less than 5 ppm hydrogen
sulfide plus organic sulfides and less than 100 ppm ammonia. The Kern County AQMD exempts separators based on the
surface area of the separator, the oil recovery rate, and the estimated fractional volume loss of oil.
c. Colorado regulation No. 7 provides for VOC emission control for oil separation equipment. Covers listed as an
option for vapor loss control.
d. Must install air pollution control equipment with 85 percent efficiency or more.
e. Exempts separators used exclusively in conjunction with crude oil production.
f. Requires sealed openings, floating roofs with closure seals, vapor disposal systems, or other approved equipment.
In actual applications only the forebay on PRWS OWS is required to be covered although regulation states all
components unless exempted.
g. This reflects the Kansas City area; there are no refineries in the St. Louis area.
h. No regulations have been established because emissions from refinery sources are considered insignificant.
1. New York City Metropolitan Area and upstate New York.
j. Nashville/Davidson county has no sources.
k. In nonattainment areas, VOC must have a true vapor pressure of > 0.5 psia; in certain other counties VOC must have
a true vapor pressure of > 1.5 psia.
1. All VOC contaminated wastewater must be directed to the separator.
m. Vapor control system must be at least 95 percent efficient.
3-73
-------
The information given in Table 3-10 was used to estimate the level of
control required for new separators. The percentage of covered, partially
covered, and uncovered separators in each state was applied to the crude
throughput in that state. For example, if it is known that 100 percent of
the separators in a state are required to be covered, 100 percent of the
crude throughput is assumed to be processed at refineries with covered
separators. Crude throughputs were calculated using 1983 refining capacity
figures and assuming 60 percent capacity utilization (1982 estimate ).
Applying the percentages to crude throughput in each state provided an
estimate of nationwide crude processed at refineries with the different
levels of control. These estimates are shown in Table 3-11.
According to Table 3-11, the nationwide crude throughput in 1983 was
1540 thousand cubic meters of crude per calendar day (10 m /cd). Of this,
1348 x 10 m /cd, or approximately 85 percent, was processed at refineries
which are located in states requiring separators to be covered. Further,
42 x 10 m /cd, or approximately 5 percent was processed at refineries
required to have partially covered separators. And the remaining 10 percent
was processed at refineries in states with no regulations. Assuming that
new refinery construction will be proportional to the current breakdown of
refining capacity by state, it is estimated that 85 percent of the new oil-
water separators will be required to be covered, 5 percent will be required
to be partially covered, and 10 percent will not be covered at all.
Current nationwide VOC emissions from oil-water separators can be
estimated by applying the emission factor given in Section 3.2.2.4 to the
estimates of crude throughput given in Table 3-11. Consideration must be
given to the emission reduction achieved by the various methods of control.
Control efficiencies of the various control techniques are discussed in
Chapter 4. Using this information, current nationwide VOC emissions can be
estimated. Current nationwide VOC emissions from oil-water separators are
estimated to be 7.5 gigagrams per year (Gg/year).
Baseline emissions from the 33 new and modified oil-water separators
are estimated to be 1211 Mg per year in 1989. This estimate is based on the
emission factor presented in Section 3.2.2.4 and on the assumption that
3-74
-------
TABLE 3-10. SUMMARY OF BASELINE CONTROL FOR OIL-WATER SEPARATORS
OJ
I
U1
State
California
- Bay Area
- Kern County
- South Coast
Delaware
Illinois
Indiana
Louisiana
New Jersey
Ohio
Oklahoma
Pennsylvania
Texas
Other States
% Separators % Separators
Fully Covered Partially Covered
100
40
90
100
50
90
80 20
100
100
100
85
100
85
% Separators
Uncovered Comments
60 Only large refineries covered by
regulation
10 Some small refineries may be exempt
50 Some separators exempted by
regulation
10 Smaller refineries may be exempt
Covering forebay only can meet
regulations under exemption
provisions
15
15 85 refineries in these states, 33%
of which are located in attainment
areas
References: 102,103,104,105,106,107,108,109,110,111
-------
TABLE 3-11. ESTIMATE OF CRUDE THROUGHPUT AT REFINERIES HAVING VARYING EMISSION CONTROLS
CO
I
Total
Crude ^Capacity
State (loV/cd)
California
- Bay Area
- Kern County
- South Coast
Delaware
Illinois
Indiana
Louisiana
New Jersey
Ohio
Oklahoma
Pennsylvania
Texas
Other States
397
128
52
217
22
159
74
349
80
82
75
114
721
495
2,568
, Crude Throughput
Crude Throughput At Refineries With
At Refineries With Partially
Covered Separators Covered Separators
_ _
772
133
1174
132
48
42
167 42
482
492
452
58
4332
2385
1,348 42
Crude Throughput
At Refineries
With Uncovered
Separators
-
-
19
13
-
48
2
-
-
_
-
10
_
60
152
Capacity utilization of 60% used to estimate crude throughput (Reference 112)
"State regulations require all separators to be covered.
Only three large refineries covered by regulation requiring covers.
crude throughput.
i
Assumes 90% of crude throughput designated to covered separators.
to be exempt.
3Assumes 85% of crude throughput designated to covered separators.
This accounts for 40% of
Some small refineries assumed
-------
85 percent of the separators will be located in states requiring covered
separators, 5 percent in states requiring partially covered separators, and
10 percent in states with no regulations.
3.4.3 Air Flotation Systems
There are currently no state regulations that apply directly to
controlling VOC emissions from air flotation systems. However, some states
may apply regulations applying to oil recovery facilities to air flotation.
Further, new source reviews of refinery sites may call for control of
emissions from air flotation. California is one state where new source
reviews have been applied to these systems. Two refineries have been
located that control emissions from air flotation for odor control purposes.
Both of these refineries are located in California. '
Control of emissions from air flotation would be on a site specific
basis. Because of this, it is difficult to determine how may, if any, new
air flotation systems would be controlled. Therefore, baseline control for
air flotation systems is assumed to be no control.
Current nationwide VOC emissions from air flotation systems can be
estimated by using the emission factor given in Section 3.2.3.3. It is
assumed that 75 percent of the refineries in the U.S. use air flotation.
Using this information, current baseline VOC emissions are estimated to be
0.64 Gg/year.
Baseline emissions from new and modified air flotation systems are
estimated to be 84 Mg per year in 1989. This estimate is based on the
emission factors presented in Section 3.2.3.3 and the assumption that
50 percent of the new air flotation systems will be DAF systems and
50 percent will be IAF systems. Current information indicates that
approximately 30 percent of existing air flotation system are IAF systems.
However, the number of IAF systems is expected to increase since this
technology is a relatively new application for petroleum refinery wastewater
systems. There is no distinct preference for either type of system and
therefore, new air flotation systems can be expected to be equally
distributed between the two types of systems.
3-77
-------
3.5 REFERENCES
1. Annual Refinery Survey. Oil and Gas Journal. 81J12):128-153.
March 21, 1983.
2. U.S. Environmental Protection Agency. Development Document for
Effluent Limitations Guidelines and Standards for the Petroleum
Refining Point Source Category. Washington, D. C. Publication
No. EPA 440/1-82/014. October 1982. p. 22-23.
3. Changes Ahead for Tomorrow's Refinery to Include 'Uniform Look'
Worldwide. Hydrocarbon Processing. 60(6):13. June 1980.
4. A Heavy, Sour Taste for Crude-Oil Refiners. Chemical Engineering.
86(10):96-100. May 19, 1980.
5. American Petroleum Institute. Manual on Disposal of Refinery Waste -
Volume on Liquid Wastes. Washington, D.C. 1969. p. 3-3.
6. U.S. Environmental Protection Agency. Code of Federal Regulations.
Title 40, Chapter 419, Washington, D.C. Office of the Federal
Register. October 18, 1982.
7. Trip Report. Laube, A.H. and G. DeWolf, Radian Corporation, to
R. J. McDonald, EPA-.CPB. July 1983. Report of March 14, 1983 visit to
Tosco Corporation in Bakersfield, California.
8. Trip Report. McDonald, R. and J. Durham, EPA:CPB, to file. June 1982.
Report of June 8, 1982 visit to Shell Oil Company in Norco, Louisiana.
9. Ref. 2, 184-187.
10. Trip Report. McDonald, R. and J. Durham, EPA:CPB, to file. June 1982.
Report of June 9, 1982 visit to Exxon Company's refinery in
Baton Rouge, Louisiana.
11. Trip Report. Laube, A.H., Radian Corporation, to R.J. McDonald,
EPA:CPB. April 25, 1983. Report of March 18, 1983 visit to Texaco in
Wilmington, California.
12. Ref. 5, p. 3-5.
13. Jones, H.R. Pollution Control in the Petroleum Industry. Pollution
Technology Review No. 4. Park Ridge, New Jersey, Noyes Data
Corporation. 1973. p. 207.
14. Ref. 5, p. 3-4.
15. Ref. 2, p. 49.
3-78
-------
16. Whetherold, R. G., (Radian Corporation). Assessment of Atmospheric
Emissions from Petroleum Refining. Volume 5: Appendix F Technical
Report. Prepared for U.S. Environmental Protection Agency.
Washington, D.C. Publication No. EPA 600/l-80-075e. April 1980.
p. 389.
17. Ref. 13, p. 315.
18. Finelt, S., J.R. Crump. Predict Wastewater Generation. Hydrocarbon
Processing. ^§: (8)159-166, August 1977.
19. Dickerman, J.C., T.D. Raye, J.D. Colley, and R.H. Parsons. (Radian
Corporation) Industrial Process Profiles for Environmental Use:
Chapter 3. Petroleum Refinery Industry. Prepared for U.S. Environ-
mental Protection Agency. Washington, D.C. Publication No. EPA
600/2-77-023C. January 1977. pp. 16-79.
20. Cantrell, A. Annual Refining Survey. Oil and Gas Journal. _18(12):
128-130. March 21, 1983.
21. Ref. 2, p. 55.
22. Ref. 2, p. 25.
23. Willenbrink, R. Wastewater Reuse and In-Plant Treatment. AICHE
Symposium Series-Water. 1973. p. 672.
24. Ref. 16, p. 127.
25. Ref. 19. p. 22.
26 Perry, J.H. Chemical Engineers' Handbook, Fifth ed. New York,
McGraw-Hill. 1973. p. 6-30.
27. Manning, F.S. and E.H. Snider. Environmental Assessment Data Base for
Petroleum Refining Wastewaters and Residuals. U.S. Environmental
Protection Agency. Ada, Oklahoma. Publication No. EPA 600/2-83-010.
February 1983. p. 65-67.
28 Los Angeles County Air Pollution Control District. Air Pollution
Engineering Manual. Second Edition. Prepared for the U.S. Environ-
mental Protection Agency. Research Triangle Park, N.C. Publication
No. AP-40. May 1973. p. 698.
29 Dames and Moore. Economic Impact of Implementing Volatile Organic
Compound Group II Regulations in Ohio. Prepared for U. S Environ-
mental Protection Agency, Region V. Chicago, Illinois. December 1981,
3-79
-------
30. Memo from Mitsch, B.F., Radian Corporation, to file. June 15, 1984.
Response to California Air Resources Board Survey of Refining Industry.
31. Beychock, M.R. Aqueous Wastes from Petroleum and Petrochemical Plants.
New York, John Wiley and Sons 1967.
32. Brown, J.D., and G.T. Shannon. Design Guide to Refinery Sewers.
Hydrocarbon Processing and Petroleum Refiner. 42(5):141-144.
May 1963.
33. Wigren, A.A. and F.L. Burton. Refinery Wastewater Control. Journal of
Water Pollution Control Federation. _44(1): 117-128. January 1972.
34. Trip Report. A.H. Laube and R.G. Wetherold, Radian Corporation, to
R. J. McDonald EPA:CPB July 19, 1983. Report of March 25, 1983 visit
to Sun Oil Refinery in Toledo, Ohio.
35. Powell, D., P. Peterson, K. Luedtke, and L. Levanas. (Pacific
Environmental Services) Development of Petroleum Refinery Plot Plans.
Prepared for U. S. Environmental Protection Agency. Research Triangle
Park, N.C., Publication No. EPA-450/3-78-025. June, 1978.
36. Wetherold, R. G. and D. D. Rosebrook (Radian Corporation). Assessment
of Atmospheric Emissions from Petroleum Refining. Volume 1: Technical
Report. Prepared for U.S. Environmental Protection Agency, Washington,
D.C. Publication No. EPA 600/l-80-075a. April 1980.
37. McCabe, W.C. and J.C. Smith. Unit Operations at Chemical Engineering.
McGraw-Hill Book Company. New York. 1976.
38. Laverman, R.J., T.J. Haynie, and J.F. Newbury. Testing Program to
Measure Hydrocarbon Emissions from a Controlled Internal Floating Roof
Tank. Prepared for American Petroleum Institute. Chicago Bridge and
Iron Company. Chicago, Illinois. March 1982.
39. Drivas, P.J. Calculation of Evaporative Emissions from Multicomponent
Liquid Spills. Environmental Science and Technology. 16_(10):726-728.
October 1982.
40. Air Pollution Control District/County of Los Angeles. Emissions to the
Atmosphere from Petroleum Refineries in Los Angeles County. Report
No. 8. Los Angeles, California. 1958.
41. U.S. Environmental Protection Agency. Compilation of Air Pollutant
Emission Factors. Third ed. Research Triangle Park, N.C. EPA AP-42,
August 1977. p. 9.1-10. (Supplement 11 Update, October 1980)
42. Ref. 24. p. 394.
3-80
-------
43. Letter from Kronenberger, L., Exxon Company, U.S.A., to Goodwin, D. R.,
EPA-.ESED. February 2, 1977. p. 14. Response to Questionnaire.
44. Ref. 13, p. 175.
45. Ref. 5, p. 6-5.
46. Ref. 5, p. 5-3.
47. Ref. 45, p. 6-3, 6-7
48. Ref. 45, p. 6-13.
49. Ref. 13, p. 175.
50. Ford, D.L. and R.L. Elton. Removal of Oil and Grease from Industrial
Wastewater. Chemical Engineering/Deskbook Issue. October 17, 1977.
p. 52.
51. MacKay, D. Solubility, Partition Coefficients, Volatility, and
Evaporation Rates. In: The Handbook of Environmental Chemistry,
Volume 2, Hutzinger, 0. (ed.) Springer-Verlag, 1980. p. 37.
52. Litchfield, O.K. Controlling Odors and Vapors from API Separators.
Oil and Gas Journal. 69(44):60-62. November 1, 1971.
53. Ref. 28. p. 675.
54. American Petroleum Institute. Hydrocarbon Emissions from Refineries.
API Publication No. 928. Washington, D.C. July 1973. p. 35.
55. Ref. 51, p. 43.
56. Letter and attachment from Caughman, W.L., Jr., Shell Oil Company, to
Durham, J., EPA. May 30, 1984. Norco refinery wastewater system.
57. Air Pollution Control District/Los Angeles. Emissions to the
Atmosphere from Petroleum Refineries in Los Angeles County. Final
Report No. 9. Los Angeles, California. 1958. p. 52.
58. Radian Corporation. Control Technique for Volatile Organic Emissions
from Stationary Sources. Prepared for U.S. Environmental Protection
Agency. Research Triangle Park, N.C. Publication No. EPA
450/1-78-022. May 1978. p. 141.
59 Vincent, R. Control of Organic Gas Emissions from Refinery Oil-Water
Separators. California Air Resources Board. Sacramento, California
April 1979. p. 4.
3-31
-------
60. Ref. 54, p. 35-37.
61. Ref. 59, p. 6-8.
62. Ref. 2, p. 76.
63. Memo from Mitsch, B. and Hunt, G., Radian Corporation, to file. June
19, 1984. Influent Temperature to Oil-Water Separators.
64. Letter from Litchfield, D. K., Amoco Oil Company, to Hunt, G. E.,
Radian Corporation. May 8, 1984.
65. Nemerow, N.L. Industrial Water Pollution Origins, Characteristics and
Treatment. Reading, Massachusetts, Addison-Wesley 1978. p. 122.
66. Ref. 50, p. 52-53.
67. Ref. 50, p. 53.
68. Burkhardt, C.W. Control Pollution by Air Flotation. Hydrocarbon
Processing. 72;(5)59-61. May 1983.
69. Luthy, R.G., R.E. Selleck, and T.R. Galloway. Removal of Emulsified
Oil with Organic Coagulants and Dissolved Air Flotation. Journal of
the Water Pollution Control Federation. 50:331-346. February 1978.
70. Telecon. Laube, A.H., Radian Corporation, with Carleton, R. E., IVEC
Refinery. December 3, 1982. Wastewater treatment system at IVEC
Bakersfield.
71. Trip Report. Laube, A.H., Radian Corporation, to EPA:CPB.
May 17, 1983. Report of March 17, 1983 Visit to Mobil Oil in
Torrance, California.
72. Churchill, R.J. and K.J. Tacchi. A Critical Analysis of Flotation
Performance. AICHE Symposium Series. 178 (74):290-299. 1977.
73. United States Filter Fluid Systems Corporation. Flotation General
Catalog. Whittier, CA.
74. Steiner, J.L., G.F. Bennett, E.F. Mohler, and L.T. Clere. Air
Flotation Treatment of Refinery Wastewater. Chemical Engineering
Practice. 74_( 12): 39-45. December 1978.
75. Engelbrecht, R.S., A.F. Gaudy, and J.M. Cederstrand. Diffused Air
Stripping of Volatile Waste Components of Petrochemical Wastes.
Journal of the Water Pollution Control Federation. 13:(2)128-135.
February 1961.
3-82
-------
76. Richardson, C.P., S.O. Ledbetter. Hydrocarbon Emissions from Refinery
Wastewater Aeration. Industrial Waste. 24.(4):26-28.
July/August 1978.
77. Stackhouse, C. and M. Hartman. Emission Test Report Petroleum Refinery
Wastewater Treatment System Chevron U.S.A., Incorporated (El Segundo,
California). TRW Environmental Operations. Research Triangle Park,
North Carolina. EMB Report No. 83WWS2. March 1984.
78. Stackhouse, C. and M. Hartman. Emission Test Report Petroleum Refinery
Wastewater Treatment System, Chevron U.S.A., Incorporated (El Segundo,
California). TRW Environmental Operations. Research Triangle Park,
North Carolina. EMB Report No. 83WWS2. March 1984.
79. Sherwood, T.K., and R. Pigford. Absorption and extraction. New York,
McGraw-Hill. 1952 p. 58-63.
80. Drivas, P.J. Calculation of Evaporative Emissions from Multicomponent
Liquid Spills in 3rd Joint Conference on Applications of Air Pollutant
Meteorology, American Meteorological Society and Air Pollution Control
Association, San Antonio, Texas, January 1982.
81. Adams, C.E., and W.W. Eckenfelder (eds.) (Associated Water and Air
Resources Engineers, Inc.) Process Design Techniques for Industrial
Waste Treatment. Nashville, TN, Enviro Press. 1974.
82. Letter and attachment from Stein, D.A., Envirosphere Company, to
Mitsch, B., Radian Corporation. July 18, 1983. NSPS for Refinery
Wastewater Systems.
83. Stackhouse, C. and M. Hartman. Emission Test Report Petroleum Refinery
Wastewater Treatment System, Golden West Refining Company (Santa Fe
Springs, California). TRW Environmental Operations. Research Triangle
Park, North Carolina. EMB Report No. 83WWS4. March 1984.
84. Stackhouse, C. and M. Hartman. Emission Test Report Petroleum Refinery
Wastewater Treatment System, Golden West Refining Company (Sweeny,
Texas). TRW Environmental Operations. Research Triangle Park, North
Carolina. EMB Report No. 83WWS3. March 1984.
85. U.S. Environmental Protection Agency. Treatability Mannual.
Volume III: Techniques for Control/Removal of Pollutants.
Washington, D.C. Publication No. EPA 600/8-80-042C. July 1980.
p. III.4.3-1.
86. Ref. 24, p. 388.
87. Ref. 81, p. 389.
3-83
-------
88. Ref. 24, p. 390.
89. Ref. 81, p. III. 5.1-1.
90. Ref. 13, p. 193.
91. Ref. 24, p. 392.
92. Ref. 13, p. 202.
93. Ref. 81, p. III. 4.2-4.
94. Ref.81, p. III. 5.3-3.
95. Ref. 2, p. 158.
96. Ref. 81, p. 4.1-1 - 4.1-33.
97. Shen, T.T. Estimation of Organic Compound Emissions from Waste
Lagoons. Journal of the Air Pollution Control Association.
32:(1)79-82. January 1982.
98. HPI Construction Boxscore. Hydrocarbon Processing. October 1983.
99. Cantrell, Aileen. Worldwide Construction Oil and Gas Journal 81(17).
April 25, 1983. ~
100. U.S. Environmental Protection Agency. VOC Fugitive Emissions in
Petroleum Refinery Industry. Background for Proposed Standards.
Research Triangle Park, N.C. Publication No. EPA 450/3-81-015a.
November 1982.
101. HPI Construction Boxscore. Hydrocarbon Processing. June 1983.
102. Telecon. Laube, A.H., Radian Corporation with Nan Kileen, Louisiana
Air Quality Division. August 4, 1983. Baseline information -
Louisiana air quality regulations.
103. Telecon. Mitsch, B.F., Radian Corporation, with Dr. John Reed, State
of Illinois. September 6, 1983. Baseline emissions.
104. Telecon. Laube, A.H., Radian Corporation, with Ken Kearney, State of
Indiana. August 31, 1983. Baseline - Indiana regulations.
105. Telecon. Mitsch, B.F., Radian Corporation, with Dick Rule,
Pennsylvania Bureau of Air Quality Control. September 6, 1983.
Pennsylvania regulations.
3-84
-------
106. Telecon. Laube, A.H., Radian Corporation, with Larry Wonders,
N. W. Pennsylvania Bureau of Air Control. August 31, 1983. Baseline
information.
107. Telecon. Laube, A.H., Radian Corporation, with John Swanson, Bay Area
Air Quality Management District. August 16, 1983. Baseline
information - Bay Area regulations.
108. Telecon. Laube, A.H., Radian Corporation, with John Powell, South
Coast Air Quality Management District. August 2, 1983. Baseline -
South Coast Air Quality Management District regulations.
109. Telecon. Mitsch, B.F., Radian Corporation, with Tom Paxson, Kern
County Air Pollution Control District. September 7, 1983. Baseline
emissions.
110. Memo from Machin, J.L., Radian Corporation, to S.A. Shareef, Radian
Corporation. August 25, 1983. Report of Meeting with Texas Control
Board.
111. Environmental Reporter. State Air Laws. Volumes 1-3. Washington,
D.C., Bureau of National Affairs, Inc. 1983.
112. U.S. Environmental Protection Agency. Code of Federal Regulations.
Title 40, Chapter 419, Washington, D.C. Office of the Federal
Register. October 18, 1982.
3-85
-------
4. EMISSION CONTROL TECHNIQUES
Petroleum refinery wastewater systems contain several sources of
volatile organic compound (VOC) emissions. These emissions result from the
evaporation of VOC from oily wastewater at points, or sources, where the
wastewater is exposed to the atmosphere. Three sources of emissions are
process drain systems, oil-water separators, and air flotation systems.
These sources and their uncontrolled emissions have been described in
Chapter 3.
There are only a limited number of methods available to reduce VOC
emission from refinery wastewater systems. These methods depend upon one or
more of the following basic principles:
o reduction of VOC entering the wastewater system;
o reducing the surface area of wastewater exposed to the
atmosphere; and
o enclosing the system to isolate it from the atmosphere.
The reduction of VOC entering the wastewater system is very desirable
from both an economic and environmental standpoint. Many, if not most,
refineries are actively pursuing this approach, and have found it to be cost
effective. The reduction can be achieved by reducing either the total
quantity of oily water sent to the wastewater system or by reducing the
quantity of VOC in the oily water. One plant reported reductions of 50-55
2
percent in the quantity of fresh water used for cooling towers and boilers.
Another refinery reported a reduction of 90 percent in the volume of
wastewater.
It must be recognized, however, that there is diversity among
refineries in terms of the design and arrangement of their wastewater
systems, as well as the volume and composition of wastewaters. Thus, it is
difficult to quantitatively define either the general effectiveness of such
programs in reducing VOC entering the wastewater system or the resultant
reduction in VOC emissions.
4-1
-------
Other methods are available for reducing VOC emissions by reducing the
surface area of wastewater exposed to the atmosphere and/or enclosing all or
part of the emission sources. In a few cases, the effectiveness of some of
these methods has been measured or estimated. These methods are discussed
in detail in Section 4.1.
There are a number of technologies that are available to either
destroy, collect or recover and/or process VOC from gaseous streams which
have been captured by a control system. Typical VOC control devices which
may be applicable include:
o flares;
o carbon adsorption;
o incineration;
o condensation;
o industrial boilers and heaters, and
o catalytic oxidation.
These control technologies are reviewed and discussed in Section 4.2.
4.1 METHODS FOR REDUCTION OF VOC EMISSIONS
Methods which can be used to reduce and/or capture VOC emissions from
sources in the wastewater system are described in the following sections.
4.1.1. Process Drains and Junction Boxes
Process drains and junction boxes, as described in Section 3.2.1, make
up the wastewater collection system within a refinery. The VOC emissions
result from vaporization from the open surfaces of drains and vents on the
junction boxes. The technologies for reducing these emissions are discussed
below.
4.1.1.1 Methods for Controlling VOC Emissions. The alternatives for
reducing emissions from oily water process drains and junction boxes involve
some type of closure or seal. A common method involves the use of a P-leg
4-2
-------
in the drain line with a water seal. A less common, but more effective
method, is a completely closed drain system. Junction box emissions can be
reduced with a water-filled seal pot.
As described in Section 3.2.1, many refinery drains are connected
directly to lateral sewer lines, which in turn are generally connected to
several other drains. There is no seal or other means for preventing VOC
vapors present in the sewer line from escaping to the atmosphere through the
open drains. A water seal in the drain can result in a reduction in the
emissions from open drains.
A P-leg water seal was discussed in Section 3.2.1.2. Such a seal could
prevent a substantial portion of the VOC in the drain system from entering
the atmosphere. It is possible that some emissions will occur from the
surface of the liquid seal in the leg of the trap which is open to the
atmosphere. Emissions will be less than those from an open drain unless the
drain is allowed to dry out and the water seal is lost.
The vent lines from sealed junction boxes may be equipped with
water-filled seal pots, as discussed and illustrated in Section 3.2.1.3. As
long as the seal pot is filled with liquid, it will provide an effective
barrier for emissions. The only means whereby VOC emissions can occur are
by diffusion through the water seal, a breach of the water seal, or from
leakage around the cover of the junction box. A small, continuous flow of
water can be directed into the seal pot to keep it filled to the desired
level. Leaks around the cover can be eliminated or minimized by proper
seals or caulking. Pressure/vacuum valves could also be used to prevent
emissions from junction box vents. However, use of this control technique
has not been found in an operating refinery.
There are several factors which affect the performance of water-sealed
drains and junction boxes in reducing VOC emissions. Some of these factors
are the drainage rate, composition of the liquid entering the drain,
temperature of the liquid entering the drain, the diameter of the drain, and
4-3
-------
ambient atmospheric conditions. The most important factor in the
performance of the junction box seal pot is the pressure within the junction
box. If a significant pressure buildup occurs, the water seal will be
breached and VOC will be emitted from the vent.
As discussed previously in Section 3.2.1, a completely closed drain
system was observed in a BTX unit at one refinery.4 This system prevents
exposure of any oily wastewater to the atmosphere within the process unit.
Thus, VOC emissions to the air are completely eliminated within the process
unit. This is assuming that the system does not leak.
In this type of control system the mouth of the vertical drain riser is
closed with a flange. Equipment drain lines are piped into the flange or
directly into the perimeter of the drain risers depending on the number of
connecting lines required per drain. The waste liquid flows into the drains
which are connected to lateral sewer lines. Drainage flows through the
underground lateral drains to a buried collection tank. The collected
liquid is pumped to an oil-water separator. A fuel gas purge removes VOC to
a control device. The entire system is purged by the fuel gas and is
maintained at a very slight positive pressure (^ 0.5 - 1.0" FLO).
Since the system is completely closed, there are very few factors which
would seriously affect its performance with the exception of equipment
failures and equipment leaks. Parameters such as wastewater flow rates,
wastewater composition, and system temperature may affect the amount of
material being directed to the control device, but emissions within the unit
will be unaffected.
4.1.1.2 Effectiveness of VOC Emission Controls. The effectiveness of
water seal drain in reducing VOC emissions has been evaluated using two
methods. First, process drains at three petroleum refineries were screened
for VOC concentration with a portable hydrocarbon analyzer. And second, a
theoretical analysis of the effectiveness of water seals was conducted.
These two methods are discussed below.
A portable organic vapor analyzer (OVA) was used to screen drains at
three refineries. The drains at one refinery were uncontrolled. The
4-4
-------
drains at the second refinery were equipped with water seals. And the
drains at the third refinery were equipped with seal pots having caps which
could be manually removed. The drains having seal pots were screened with
the cap in place and after the cap had been removed. Removing the cap broke
the water seal on the drain and left the drain in an uncontrolled state.
The results of the screening study were analyzed using two approaches.
In the first approach, all screening values from uncontrolled drains were
averaged and compared with the average of all screening values from
o
controlled drains. A total of 200 screening values for controlled drains
were included in the analysis and 169 screening values for uncontrolled
drains. The averaged screening values were converted to leak rates using
the correlation established in an EPA study of atmosphere emissions from
o
petroleum refineries. This correlation is as follows:
Log1Q (Non Methane Leak Rate, ppmv) = -4.9 + 1.10 Log,Q (Max. Screening
Value)
The leak rate for controlled drains was 0.00353 Ibs/hr. The leak rate for
uncontrolled drains was 0.00592 Ibs/hr. Based on the leak rates derived
from averaging screened values, the emission reduction achieved by water
seals is approximately 40 percent.
The second approach used to evaluate the screening results was to
evaluate the drains at the refinery having capped drains both before and
after the cap was removed. Seventy-six drains were evaluated using this
method. The number of drains evaluated is smaller than the total number of
drains screening because some drains were already uncapped, the caps could
not be removed, or the data taken were for various reasons unusable
(e.g. cap was not sealed, cap could not be put in place, or another VOC
source was near drain). If multiple readings were taken on one drain, the
last reading was used in the analysis if it was the lowest of a
4-5
-------
consistently declining set of readings. If multiple readings varied
substantially for the same drain, an average value was used. The results of
this approach are shown in Table 4-1. The results indicate an emission
reduction of approximately 50 percent.
A further analysis grouped drains into two categories to see if the
uncontrolled leak rate had any effect on the emission reduction that could
be achieved. Those with uncontrolled screening values less than 100 ppm
were placed in one group while those with values greater than 100 ppm were
placed in a second group. Of the 76 uncontrolled drains that were screened,
18 had values greater than 100 ppm. The screening value, estimated leak
rate, and the emission reduction factor for each of these drains is shown in
Table 4-2.
As shown in the table, the average emission reduction was approximately
50 percent. In most cases, the percentage reduction for individual drains
was greater than 50 percent. One drain had a negative percentage reduction.
If this value is removed, the emission reduction would be 74 percent.
Based on the analyses of drains screening data, emission reductions of
40 percent to 50 percent are achievable by water seal drains. Values for a
specific drain can vary from 0 percent to 99 percent.
A theoretical analysis of the effectiveness of water seal drains was
also conducted. As discussed in Chapter 3, emissions from drains are
primarily influenced by the forces of convection and diffusion. Three types
of drains were evaluated using benzene as an example compound: an
uncontrolled drain, a P-trap water sealed drain with no contaminated water
and a P-trap water sealed drain saturated with benzene from a contaminated
stream.
The benzene emissions due to molecular diffusion through the water seal
were estimated based on the equation presented in Section 3.2.1.3. The
assumptions used to estimate emissions are presented in Table 4-3. The
emissions due to convection were estimated based on a study which showed
that the total emissions due to convection and molecular diffusion were 1.0
to 31.7 (average of 25) times molecular diffusion.14 This value was then
adjusted to account for windspeed by by making three assumptions. First, it
4-6
-------
TABLE 4-1. SUMMARY OF SCREENING VALUES FOR INDIVIDUAL DRAINS
Leak Rate
# of Drains Screened Type of Drain (Ibs/hr)
76 Controlled 0.10184
76 Uncontrolled 0.20484
4-7
-------
TABLE 4-2. SUMMARY OF EMISSION RATES AND EMISSION REDUCTION FOR DRAINS WITH A LEAK RATE >100 PPM
00
Drain
Unit No.
27.1 6
7
17
26.2 3
27.2 1
2
3
11
12
25 11
19
23
69
83
84
85
86
94
Screening Values
Cap On Cap Off*
12
10
10
4
40
2,000
7
50
40
10
8
120
20
12
7
70
70
1,000
8
1,000
100
120
100
110
1,750
300
300
400
178
300
400
120
150
200
100
300
1,500
150
Estimated
Emission Rate, LB/HR
Cap On Cap Off*
0.00019
0.00016
0.00016
0.00005
0.00073
0.05384
0.00011
0.00083
0.00016
0.00012
0.00244
0.00034
0.00019
0.00011
0.00135
0.00135
0.02512
0.00012
0.08737
0.02512
0.00200
0.00244
0.00200
0.00222
0.04649
0.00668
0.00792
0.00376
0.00668
0.00917
0.00244
0.00312
0.00428
0.00200
0.00668
0.03924
0.00312
0.17536
Est. Emission
Reduction
LB/HR %
0.02493
0.00184
0.00228
0.00195
0.00149
-0.00735
0.00657
0.00709
0.00360
0.00656
0.00673
0.00210
0.00293
0.00417
0.00065
0.00533
0.01412
0.00300
0.08800
99.2
91.8
93.4
97.5
66.9
-15.8
98.4
89.5
97.3
98.2
73.4
86.1
93.8
97.4
32.5
79.8
36.0
96.2
5Od
-------
TABLE 4-3. ASSUMPTIONS FOR ESTIMATING BENZENE EMISSIONS FROM EXAMPLE DRAINS
Uncontrolled Drain
Benzene concentration in vapor phase = 0.125 atm
Wastewater temperature =150 F
Ambient temperature = 70 F
Drain diameter = 4 in
Length of drain = 4.25 ft
Average temperature in drain = 110 F 2
Diffusion coefficient in air = 0.097 cm /sec
Total mass transport 150 times molecular diffusion
Benzene concentration at top of drain = 0 mg/L
Wind speed = 10 ft/sec
P-Trap Water Sealed Drain with Clean Wastewater
Length of water seal = 1.6 ft
Temperature of water seal = 68 F
Drain diameter = 4 in
Length of drain above water seal = 2.25 ft 5 2 o
Diffusion coefficient in water = 1.02 x 10 J cm /sec at 68 F
Henry's Law applies 3 2
Henry's Law coefficient = 5.49 x 10"° atm/m mole
Concentration at bottom of water seal in equilibrium with vapor phase
Concentration of benzene at top of water seal = 0 moles/L
No convection (i.e., diffusion through water seal controls mass
transfer)
P-Trap Water Sealed Drain with Contaminated Wastewater
Water seal saturated with benzene
Temperature of water seal = 68 F
Length of drain above water seal = 2.25 ft
Diameter of drain = 4 in
Benzene concentration at top of drain = 0 mg/L
Solubility of benzene in water = 1780 mg/1
Total mass transport 150 times molecular diffusion
Continuous wastewater flow into drain
Wind speed = 10 ft/sec 2
Diffusion coefficient of benzene in air = 0.085 cm /sec
References: 10,11,12,13,14,15
4-9
-------
was assumed that the mass transfer coefficient for benzene is proportional
to y * . where y is the windspeed. Second, it was assumed that the
windspeed at which the convection data was collected was not greater than
one ft/second. And finally, windspeed used for the example calculations was
10 ft/second. Based on the above, the mass flux of benzene was calculated
to be 150 times molecular diffusion.
The benzene emissions due to diffusion through the water seal were
calculated based on the following equation:12
NA - DV A CAV (xA1 - xA2)
\"B1 "DO/
o Pi - a2—
DT T » IV /V \
T In (Xno/Xpi)
Where:
Dy = Diffusion coefficient
BT = Length of water seal
CAV = Average benzene concentration
XAJ = Initial mole fraction of benzene
XA2 = Final mole fraction of benzene
Xgj = Initial mole fraction of water
XR2 = Final mole fraction of water
A = Cross sectional area of drain
4-10
-------
Based on the above discussion along with the assumptions presented in
Table 4-3, the benzene emissions from each drain configuration were
calculated. The results are presented in Table 4-4.
As shown in the table, the clean water seal is estimated to reduce
emissions by about 99.9 percent over the uncontrolled drain. This reduction
is due to the elimination of the effects of convection. The water seal also
acts as a barrier to molecular diffusion, greatly slowing down the movement
of benzene through the drain.
The estimate of emissions from a water seal saturated with benzene show
how the seal could lose its effectiveness. The emissions from a water seal
contaminated with benzene was calculated to be 555 gm/day. This is over 1.7
c
times the rate of an uncontrolled drain and over 2 x 10 times the emission
rate from an uncontaminated water seal. The increase in emissions over an
uncontaminated water seal is due to the fact that benzene does not have to
diffuse through a water seal. The length of the diffusion path is greatly
reduced and the convection effects are not eliminated.
In an actual refinery sewer system, there will be both contaminated and
uncontaminated water seals. The larger percentage will be uncontaminated
water seals as shown by the drain screening data. Of the 76 drains with
caps properly placed, only three had a screening value of 100 ppm or greater
in the controlled states (caps on). The low screening values of the other
73 drains indicate very little or no contamination. Additionally, the vapor
space in the sewer pipe may not be saturated with hydrocarbon as assumed in
the example calculations. Only 19 drains at the refinery having capped
drains were found to have a screening value of 100 ppm or greater with the
cap off, and only six drains had values between 50 and 100 ppm.
Using both the screening analysis and theoretical analysis as bases, it
is estimated that water seal drains reduce VOC emissions by 50 percent. The
screening study indicates emission reductions of 40 to 50 percent are
achievable. The theoretical analysis indicates that emission reduction may
be much greater, particularly with a well maintained water seal. Water
seals can be maintained by periodic inspection of the drains to ensure the
seal is in place.
4-11
-------
TABLE 4-4. BENZENE EMISSIONS FROM EACH DRAIN CONFIGURATION
CONFIGURATION EMISSIONS DUE TO EMISSIONS DUE TO TOTAL
MOLECULAR DIFFUSION CONVECTION EMISSIONS
(gm/day) (gm/day) (gm/day)
Uncontrolled Drain 2.1 312 315
Uncontaminated 0.0026 0.0026
Controlled Drain
Contaminated 3.7 551 555
Controlled Drain
4-12
-------
A completely closed drain system can capture virtually 100% of the VOC
emissions. The overall reduction in VOC emissions will depend on the
efficiency of the control device. For example, a smokeless flare can
achieve about a 98 percent destruction efficiency.
4.1.2 Oil-Water Separators
Oil-water separators, as described in Section 3.2.2, rely on gravity
separation to remove the oil fraction of the wastewater stream. The VOC
emissions occur as a result of vaporization from the open surfaces of
uncontrolled separators. The technologies for reducing these emissions are
described below.
4.1.2.1 Methods for Controlling VOC Emissions. The most effective
method for controlling VOC emissions from oil-water separators is to use
q
either floating or fixed covers. This will reduce VOC emissions by:
o Reducing the oil surface exposed to the atmosphere,
o Reducing the effects of wind velocity,
o Insulating the oil layer from solar radiation.
A fixed cover can be installed on most separators without interfering
with the oil-skimming system. The cover may be constructed of various
materials including truncated case aluminum segments, steel plates, or
10 IQ on 21
concrete. The roof can be mounted on the sides of the separator
18 22
or supported by horizontal steel beams set into the sides of the tank. '
The covers usually have gas tight access doors which are used for inspection
Pi pp
and maintenance. ' The space between the cover and the edge of the
18 22
separator can be sealed using a urethane or neoprene gasket. '
The vapor space present under fixed covers may constitute an explosion
or fire hazard. In order to eliminate this problem the vapor space can be
19
blanketed with either plant gas or an inert gas, such as nitrogen.
Additionally, the vapor space can be purged with air, steam, inert gas or
product gas, and the vapors sent to a recovery or destruction device. Such
a system can greatly reduce VOC emissions. The technologies used to control
VOC gases are discussed in Section 4.2.
4-13
-------
In contrast to fixed roofs, which are always above the oil layer,
floating roofs actually float on the oil surface. This eliminates most of
the vapor space above the liquid, thus greatly reducing the potential for
volatilization from the oil layer. To prevent the roof from interfering
with the operation of the flight scraper, the water level can be raised in
the separator so that the top of the oil surface is above the flight scraper
blades.18 An example of a floating cover on an API separator is shown in
Figure 4-1.
The cover can be constructed of plastic or glass foam blocks, aluminum
pontoons or fiberalass.18'24'25 Gas tight doors can be installed in the
18
roof for inspection and maintenance. To prevent VOC from leaking around
the edges of the cover, seals are used between the cover and the walls of
the separator. These seals are usually resilient foam wrapped with a coated
fabric. The seal is placed in direct contact with the edge of the cover and
the separator wall. One manufacturer of floating roof covers uses a
polyurethane foam wrapped with a nylon-polyurethane fabric. This seal is
shown in Figure 4-2.
There are several factors which can affect the overall performance of
the two types of covers in reducing VOC emissions. The most obvious is the
degree of maintenance. The seals must be kept in good condition to minimize
leakage around joints and seals. With the exception of leakage, the control
effectiveness of closed systems which are vented to recovery or destruction
devices is relatively insensitive to variations in system parameters. The
efficiency of those covered units which are vented to the atmosphere depends
on system variables such as VOC content of the incoming water, the
temperature of the liquid phase, the ambient temperature, amount of solar
insulation, extent of surface area, and thickness of the oil layer. All of
these factors were discussed in detail in Section 3.2.2.2.
4.1.2.2 Effectiveness of VOC Emission Controls Very little data are
available regarding the reduction of VOC emissions which can be achieved by
installing a cover over an oil-water separator. The only published study,
done by Litchfield, found that by using 2 inch thick Foamglas slabs as a
4-14
-------
ADJACENT f LOATWG COVE"
DUOI
oevics
N, 0
SKIMMNO
UAMOJ
l*-f A W«'X'€SAMCf WIO
^A »^ I.SPECTON MAMHOLE
rt
^ 0
ADJACENT cLOAr,i<; COVE"
1
!
I
Figure 4-1. Floating Cover on an API Separator.
23
4-15
-------
Floating Roof
Polyurethane Foam
Wall of Separator Basin
Floating Roof
Mylon-Polyurethane Wrap
Polyurethane Foam
Figure 4-2. Polyurethane Foam Seal on a Floating Cover.
26
-------
23
floating cover, the evaporation losses could be reduced by 85 percent.
Other sources report varying levels of emission reduction but give no
supporting documentation. The American Petroleum Institute stated that a
floating or fixed cover would reduce emissions by 90 percent to
27
98 percent. In AP-42, an emission reduction value of 96 percent was
28
reported. Further, in a recent study the State of California estimated
29
that a 90 percent reduction in emission could be achieved by using covers.
The reduction in VOC emissions which can be obtained using a cover was
assumed to be 85 percent. This factor is based on the only documented
23
study, done by Litchfield. It is assumed that a fixed roof and a floating
roof provide equivalent control efficiency.
The addition of a fixed roof vapor collection system, and direction of
the collected vapor to a control device, will result in a greater overall
21
control of captured VOC emissions. Due to some possible leakage, the
capture efficiency of the roof in this type of control system would be
approximately 99 percent. The actual efficiency of the system will depend
on the efficiency of the control device. For example, the efficiency of a
flare is estimated to be 98 percent. Therefore, the overall efficiency of a
fixed roof with vapors vented to a flare would be 97 percent (0.99 x 0.98 =
97%). The efficiencies of various control devices are discussed in
Section 4.2.
4.1.3 Air Flotation Systems
Air flotation systems are used to remove free and emulsified oil,
suspended solids, and colloidal solids from refinery wastewater. Their
operation has been described in Chapter 3.2.3. VOC emissions occur as a
result of volatilization from the exposed surface of the air flotation
system. The methods for controlling these emissions are described below.
4.1.3.1 Methods for Controlling Emissions Methods for controlling VOC
emissions from air flotation systems differ depending upon the type of air
flotation system. Induced air flotation systems (IAF) usually are equipped
with a cover while dissolved air flotation systems (DAF) are open to the
4-17
-------
atmosphere. Gas or air used for flotation in an IAF is usually recirculated
in the vapor space while the gas or air used for flotation in a DAF is
introduced into the system from an outside source.
Control of VOC emissions from an IAF can be accomplished by operating
the IAF under gas tight conditions. IAF systems usually are equipped with a
cover on top and eight access doors on the sides. The access doors can be
gasketed and tightly sealed during operation of the system. A slight
negative pressure is created in the vapor space of the IAF due to the action
of the impellers or recycled wastewater. The impellers or recycled
wastewater create a vortex which draws gas or air into the wastewater. The
only emissions resulting from a gas tight IAF would be from breathing
losses. The breathing losses would result in VOC being emitted through an
atompheric vent or pressure/vacuum valve located on the roof of the cover.
The pressure/vacuum valve is needed to safely operate the system.
VOC emissions from DAF systems can be controlled by placing a fixed
cover on the flotation chamber. Because of the slight positive pressure
created by the flotation gas or air, the cover must be provided with an
atmosphere vent or vent equipped with a pressure/vacuum valve. Only fixed
covers can be used for DAF systems due to the design of the systems.
Floating covers would interfere with the skimming devices and inhibit the
18
formation of floating oil and suspended solids. Fixed covers would be of
the same type and design as covers discussed for oil-water separators. At
least two refineries presently use fixed covers with atmospheric vents on
DAF systems.30'31
A more stringent level of control for both IAF and DAF systems would be
to completely seal the flotation chamber with a fixed cover and vent the
captured VOC to a control device Incinerators, flares, process heaters, or
carbon absorbers are some of the devices used to control the collected
vapor. VOC emissions captured by a fixed cover are diverted to the control
device using air, inert gas (such as nitrogen), or plant gas to purge the
vapor space.18-20-32'33'6
Four refineries have been identified as using emission control systems
with captured VOC vented to a control device. In one refinery, the two DAF
4-18
-------
systems used in the wastewater treatment system are covered and the vapors
are collected. The collected vapors are directed to an incinerator.
Nitrogen is used as the OAF flotation gas and fuel gas from the plant fuel
gas system is used as the source of fuel for the incinerator. The control
system shown in Figure 4-3 was installed by the refinery to control odors
arising from the wastewater system.
A second refinery uses a segregated wastewater system. The bulk of the
oily wastewater is treated by two DAF's operating in parallel to treat the
effluent from the one oil-water separator. The flotation chambers are
covered, and the vapors are collected and directed to an activated carbon
bed. An IAF unit is also used to treat effluent from a second oil-water
separator. The IAF is also covered, and its vapors are collected and
directed to two 55-gallon drums filled with activated carbon. The system
•JO
was installed to eliminate odor problems, and is shown in Figure 4-4.
The third refinery uses fuel gas in the DAF systems. The flotation
chambers are covered and the vapors are recycled to the refinery fuel gas
33
system. Another refinery uses purge air to direct emissions from the IAF
unit to a process heater.
4.1.3.2 Effectiveness of VOC Emission Controls The effectiveness of
emission control techniques differs between the IAF and DAF systems. An IAF
is usually provided with a cover and some emission reduction results due to
this cover. Operating the IAF with the access doors in a closed state
achieves additional reduction in emissions. The DAF system usually is not
equipped with a cover and is therefore in a totally uncontrolled state.
Emission reduction achieved by covering a DAF will be less than that
for a gas-tight IAF or a covered oil-water separator. This is due to the
slight positive displacement of gas caused by the flotation mechanism.
Theoretical analyses presented in Section 3.2.3.2 examined the effects of
evaporation and air stripping on emissions from a DAF. Example design
specifications for the DAF were chosen and input parameters based on the
test results were used in calculating emissions. These input parameters
included the influent oil concentration and influent benzene
4-19
-------
iva
o
effluent
rover
cover
O
pump
nitrogen gas supply
| pimp
collected vapors from
oil-water separator
I
*»-
re
It
OAF
IZ
cycle
nk '
.
— ~
\
\
i»-
re
ti
UAF
fl
cycle
nk \
,
Mastcwater from
otl-water separator
stack
_L
Incinerator
t
fuel gas
from plant
supply
Dlower
Figure 4-3. Example of DAF Emission Control System.
-------
i
ro
Induced air
Uastewater from
oil-Mater separator
,.
or
Y Cover
IAF
I
I
-_ Cfflunnt
*
I
'I
1
Y
1
1
-«.__r__^
t
i
Plant air
Uastewater from
oil-water separator
Plant air ._
Both DAF's tightly covered
Effluent
Activated
carbon
bed
55 gallon
drums containing
activated carbon
II lower
Tigure 4-4. Examples of DAF and IAF Control Systems.
-------
conconcentration. Appropriate calculations were then used to estimate
benzene losses due to evaporation and air stripping. The analyses indicate
that the major cause of emissions is evaporative losses. Evaporative losses
have been estimated to account for 90 percent of the total losses. It is
assumed that covering a DAF will reduce the evaporative losses by 85 percent,
as determined by Litchfield. The air stripping losses would continue to be
emitted through the atmospheric vent. Therefore, the overall emission
or og
reduction achieved by a fixed roof will be (0.9)(0.85) = 77 percent. '
An estimate of the emission reduction achieved by a completely gasketed
and sealed IAF can be made using test data, a laboratory study, and
engineering judgment. Consideration must first be given to the emission
reduction achieved by an IAF operating under "normal" conditions. A typical
IAF is expected to be operated with the doors closed but not gasketed and
sealed. The emission reduction achieved by a gasketed and sealed IAF can be
estimated by calculating an emission factor for an IAF operating under four
conditions: completely uncovered; covered with the doors open; covered with
the doors closed but not gasketed; and covered with the doors gasketed and
sealed.
As mentioned in Section 3.2.3.3, the emission potential of an uncovered
IAF is approximately 15.2 kg/MM gallons of wastewater. This emission factor
is based on test data. An emission factor for a covered IAF with all the
access doors open can be estimated using engineering judgment. In
Section 4.1.2.2, it is estimated that a tightly sealed cover on an oil-water
separator will reduce emissions by 85 percent. This estimate is based on
the Litchfield Study. It is assumed that a cover on an IAF would reduce the
emissions from the top of the IAF by 85 percent. An IAF system with all the
access doors open would have 50 percent of the surface area exposed. This
estimate is based on design specifications of an IAF provided by a vendor.
Therefore, 50 percent of the emissions from the IAF (through access doors)
are completely uncontrolled while the other 50 percent (through the top) are
controlled by 85 percent. Thus the emissions from an IAF operating under
this condition are 15.2-(15.2) (0.5) (0.85) = 8.7 Kg/MM gal.
4-22
-------
The emission reduction achieved by an IAF with the doors closed can be
estimated using data from a study conducted by the Chicago Bridge and Iron
Company.36 In this study, emissions were measured from drums filled with
hexane. Different levels of control were placed on the drums. One level of
control included a cover having 1/8" gaps between the tank wall and the
cover. The second level of control included a cover with an 8 inch diameter
opening. Extropolation of the emission results from this experiment can be
used to estimate emissions from an IAF with the doors closed (but not
gasketed) and an IAF with the doors open.
In the CBI study, the 8-inch opening in the drum represents 12.6% of
the total surface area of the cover. As discussed above, if all the access
doors in an IAF are open, 50% of the surface area of the IAF is exposed.
Assuming a proportional relationship between exposed surface area and
emissions, the emissions from the drum (with 50 percent of the surface area
exposed) can be estimated as follows:
12.6 % = 50 %
0.02 Ib/hr X
X = 0.079 Ibs/hr
The emission rate from the drum having a cover with a 1/8" gap between
the cover and drum walls was measured to be 0.02 Ibs/hr. This represents a
75 percent reduction over the drum with 50 percent of the surface area
exposed. Extrapolating these data to an IAF systen, it can be estimated
that a 75 percent reduction will occur if the doors are closed (over the
case where the doors are left open). This results in an emission factor of
15 - (15) (0.5)(0.25) - (15) (0.5) (0.75) = 3.0 Kg/MM gallon for an IAF with
the doors closed but not gasketed and sealed.
As mentioned above the emission reduction achieved by an oil-water
separator equipped with a tightly sealed cover is 85 percent. Therefore, it
is assumed that the emission reduction for a tightly sealed IAF would also
be 85 percent. An 85 percent emission reduction over the uncontrolled state
4-23
-------
would result in an emission factor of 2.3 kg/MM gallon for the IAF.
Therefore, the emission reduction achieved by gasketing and sealing an IAF
is 3.0 - 2.3 = 0.7 kg/MM gallons, a 23 percent reduction from the typical
operating condition.
The emission reduction achieved by tightly covering a DAF or IAF and
venting the captured emissions to a control device will be dependent on the
efficiency of the control device. Venting the emissions to a control device
will require some type of purging system. As discussed in Section 3.2.3.3,
the emission potential of the DAF and IAF is equal when both systems are
purged. However, the percentage emission reduction achieved by the vent
system will be less for the IAF because some control is achieved by the
cover normally found on the system. For example, tightly covering a DAF and
venting the emissions to a flare will reduce emissions by approximately
97 percent. This assumes a 99 percent capture efficiency for the roof and a
98 percent destruction efficiency for the flare. The destruction efficiency
of a flare has been established by a number of studies which are discussed
in the following section. Tightly covering an IAF and venting the emissions
to a flare will reduce emissions by 85 percent. Although the amount of VOC
captured and destroyed is equivalent to that for the DAF, the percentage
reduction from the uncontrolled state is less since some control is achieved
by the cover normally found on the "uncontrolled" IAF.
4.2 CONTROL OF CAPTURED VOC
There are several methods that may be used to control VOC emissions,
either by recovery of VOC from gas streams or by destruction of the VOC by
means of combustion. These methods include the following:
o flare systems;
o carbon adsorption;
o incineration;
o condensation;
o industrial boilers and process heaters; and
o catalytic oxidation.
4-24
-------
Some of these control methods, such as flare systems, incineration,
carbon adsorption, and process heaters have been applied to the VOC
emissions from refinery wastewater sources. Others have the potential for
application to these sources. All of the above listed control methods are
described in the section which follow. In addition, factors which affect
their performance are discussed and control efficiencies are defined.
4.2.1 Flare Systems
Flares are a method of controlling VOC emissions by thermal
destruction. This is a proven technology that is used for controlling a
wide range of gaseous emissions. A brief description of the technology,
factors affecting its performance, and the potential as a VOC control method
for refinery wastewater sources are discussed in this section.
4.2.1.1 Operating Principles. For safety and environmental reasons,
refinery discharges of flammable and/or toxic vapors (and liquids) must be
either recovered or removed to an appropriate location and destroyed. The
vapors are collected and transported through a header or blowdown system.
The most widely accepted method of disposing of these vapors is to burn a
flare.
Flaring is an open combustion process in which the oxygen required for
combustion is provided by the air around the flame. Good combustion in a
flare is governed by flame temperature, residence time of components in the
combustion zone, turbulent mixing of the components to complete the
oxidation reaction, and the amount of oxygen available for free radical
formation.
There are two types of flares: ground level flares and elevated
flares. Kalcevic presents a detailed discussion of different types of
flares, flare design and operating considerations, and a method for
OQ
estimating capital and operating costs for flares. The basic elements of
an elevated flare system are shown in Figure 4-5. Process off-gases are
sent to the flare through the collection header (1). The off-gases entering
the header can vary widely in volumetric flow rate, moisture content, VOC
4-25
-------
Flare T1p(8)
Steam Nozzles(9)
Gas Barrier(6)
Gas Collection Header
and Transfer Line (1)
Knock-out
Drum(2)
Pilot Burners(7)
Steam Line
Ignition Device
Air Line
Gas Line
Drain
Figure 4-5. Steam-Assisted Elevated Flare System.
4-26
-------
concentration, and heat value. The knock-out drum (2) removes water or
hydrocarbon droplets that could extinguish the flame or cause irregular
combustion. Off-gases are usually passed through a water seal (3) before
going to the flare. This prevents possible flame flashbacks, caused when
the off-gas flow to the flare is too low and the flame front pulls down into
the stack.
Purge gas (N2, CO^, or natural gas) (4) also helps to prevent flashback
in the flare stack (5) caused by low off-gas flow. The total volumetric
flow to the flame must be carefully controlled to prevent low flow flashback
problems and to avoid a detached flame (a space between the stack and flame
with incomplete combustion) caused by an excessively high flow rate. A gas
barrier (6) or a stack seal is sometimes used just below the flare head to
impede the flow of air into the flare gas network.
The VOC stream enters at the base of the flame where it is heated by
already burning fuel and pilot burners (7) at the flare tip (8). If the gas
has sufficient oxygen and residence time in the flame zone it can be
completely burned. A diffusion flame receives its combustion oxygen by
diffusion of air into the flame from the surrounding atmosphere. The high
volume of fuel flow in a flare requires more combustion air at a faster rate
than simple gas diffusion can supply so flare designers add steam injection
nozzles (9) to increase gas turbulence in the flame boundary zones, drawing
in more combustion air and improving combustion efficiency. This steam
injection promotes smokeless flare operation by minimizing the cracking
reactions that form carbon. Significant disadvantages of steam usage are
the increased noise and cost. The steam requirement depends on the
composition of the gas flared, the steam velocity from the injection nozzle,
and the tip diameter. Although some gases can be flared smokelessly without
any steam, typically 0.15 to 0.5 kg of steam per kg of flare gas is
required. Gases with heating values of below about 18 MJ/scm (500 Btu/scf)
may be flared smokelessly with steam or air assist.
Steam injection is usually controlled manually with the operator
observing the flare (either directly or on a television monitor) and adding
steam as required to maintain smokeless operation. Several flare
4-27
-------
manufacturers offer devices such as infrared sensors which sense flare flame
characteristics and adjust the steam flow rate automatically to maintain
smokeless operation.
Some elevated flares use forced air instead of steam to provide the
combustion air and the mixing required for smokeless operation. These
flares consist of two coaxial flow channels. The combustible gases flow in
the center channel and the combustion air (provided by a fan in the bottom
of the flare stack) flows in the annul us. The principal advantage of air
assisted flares is that expensive steam is not required. Air assist is
rarely used on large flares because air flow is difficult to control when
the gas flow is intermittent. About 0.8 hp of blower capacity is required
for each 100 Ib/hr of gas flared.39
Ground flares are usually enclosed and have multiple burner heads that
are staged to operate based on the quantity of gas released to the flare.
The energy of the flared gas itself (because of the high nozzle pressure
drop) is usually adequate to provide the mixing necessary for smokeless
operation and air or steam assist is not required, A fence or other
enclosure reduces noise and light from the flare and provides some wind
protection. Ground flares are less numerous and have less capacity than
elevated flares. Typically they are used to burn gas "continuously" while
steam-assisted elevated flares are used to dispose of large amounts of gas
40
released in emergencies.
4.2.1.2 Factors affecting efficiency. The flammability limits of the
gases flared influence ignition stability and flame extinction. (Gases must
be within their flammability limits to burn.) When flammability limits are
narrow, the interior of the flame may have insufficient air for the mixture
to burn. Fuels with wide limits of flammability (for instance, H,, and
acetylene) are therefore usually easier to burn. However, in spite of wide
flammability limits, CO is difficult to burn because it has a low heating
value and slow combustion kinetics.
The auto-ignition temperature of a fuel affects combustion because gas
mixtures must be at high enough temperature to burn. A gas with low
4-28
-------
auto-ignition temperature will ignite and burn more easily than a gas with a
high auto-ignition temperature. Hydrogen and acetylene have low
auto-ignition temperatures while CO has a high one.
The heating value of the fuel also affects the flame stability,
emissions, and flame structure. A lower heating value fuel produces a
cooler flame which does not favor combustion kinetics and also is more
easily extinguished. The lower flame temperature will also reduce buoyant
forces, which reduces mixing (especially for large flares on the verge of
smoking). For these reasons, VOC emissions from flares burning gases with
low Btu content may be higher than those from flares which burn high Btu
gases.
The density of the gas flared also affects the structure and stability
of the flame through the effect on buoyancy and mixing. The velocity in
many flares is very low, therefore, most of the flame structure is developed
through buoyant forces as a result of the burning gas. Lighter gases
therefore tend to burn better. The density of the fuel also affects the
minimum purge gas required to prevent flashback and the design of the burner
tip.
Poor mixing at the flare tip or poor flare maintenance can cause
smoking (particulate). Fuels with high carbon to hydrogen ratios (greater
than 0.35) have a greater tendency to smoke and require better mixing if
they are to be burned smokelessly.
Many flare systems are currently operated in conjunction with baseload
gas recovery systems. Such systems are used to recover hydrocarbons from
the flare header system for reuse. Recovered hydrocarbons may be used as a
feedstock in other processes or as a fuel in process heaters, boilers or
other combustion devices. When baseload gas recovery systems are applied,
the flare is generally used to combust process upset and emergency gas
releases which the baseload system is not designed to recover and
unrecoverable hydrocarbons. In some cases, the operation of a baseload gas
recovery system may offer an economic advantage over operation of a flare
alone since sufficient quantity of useable hydrocarbons can be recovered.
4-29
-------
4.2.1.3 Control Efficiency. This section presents a review of the
flares and operating conditions used in five studies of flare combustion
efficiency. Each study summarized in Table 4-1.
Palmer experimented with a 1.3 cm (1/2-inch) ID flare head, the tip of
which was located 1.2 m (4 feet) from the ground. Ethylene was flared at
15 to 76 m/s (50 to 250 ft/sec) at the exit, 0.1 to 0.6 MW (0.4 x 106 to
2.1 x 10 Btu/hr). Helium was added to the ethylene as a tracer at 1 to
3 volume percent and the effect of steam injection was investigated in some
experiments. Destruction efficiency (the percent ethylene converted to some
other compound) was 97.8 percent.
Siegel made the first comprehensive study of a commercial flare system.
He studied burning of refinery gas on a commercial flare head manufactured
by Flaregas Company. The flare gases used consisted primarily of hydrogen
(45.4 to 69.3 percent by volume) and light paraffins (methane to butane).
Traces of H«S were also present in some runs. The flare was operated from
30 to 2900 kilograms of fuel/hr (287 to 6,393 Ib/hr), and the maximum heat
release rate was approximately 68.96 MW (235 x 10 Btu/hr). Combustion
efficiencies (the percent VOC converted to CO^) averaged over 99 percent.
Lee and Whipple studied a bench-scale propane flare. The flare head
was 5.1 cm (2 inches) in diameter with one 13/16-inch center hole surrounded
by two rings of 16 1/8-inch holes, and two rings of 16 3/16-inch holes.
This configuration had an open area of 57.1 percent. The velocity through
the head was approximately 0.9 m/s (3 ft/sec) and the heating rate was
0.1 MW (0.3 x 10 Btu/hr). The effects of steam and crosswind were not
investigated in this study. Destruction efficiencies were 99.9 percent or
greater.
Howes, et al. studied two commercial flare heads at John Zink's flare
test facility. The primary purpose of this test (which was sponsored by the
EPA) was to develop a flare testing procedure. The commercial flare heads
were an LH air assisted head and an LRGO (Linear Relief Gas Oxidizer) head
manufactured by John Zink Company. The LH flare burned 1,043 kg/hr
(2,300 Ib/hr) of commercial propane. The exit gas velocity based on the
pipe diameter was 8.2 m/s (27 ft/sec) and the firing rate was 13 MW
4-30
-------
(44 x 106 Btu/hr). The LRGO flare consisted of 3 burner heads located 0.9 m
(3 feet) apart. The 3 burners combined fired 1,905 kg/hr (4,200 Ibs/hr) of
natural gas. This corresponds to a firing rate of 24.5 MW (83.7 x 10 Btu/hr),
Steam was not used for either flare, but the LH flare head was in some
trials assisted by a forced draft fan. Combustion efficiencies for both
44
flares during normal operation were greater than 99 percent.
A detailed review of all four studies was done by Joseph, et al. in
January 1982.40 A fifth study45 determined the influence on flare
performance of mixing, Btu content, and gas flow velocity. A steam-assisted
flare was tested at the John Zink facility using the procedures developed by
Howes. The test was sponsored by the Chemical Manufacturers Associated
(CMA) with the cooperation and support of the EPA. All of the tests were
with an 80 percent propylene, 20 percent propane mixture diluted as required
with nitrogen to give different heat content values. This was the first
work which determined flare efficiencies at a variety of "nonideal" condi-
tions where lower efficiencies had been predicted. All previous tests were
of flares which burned gases which were very easily combustible and did not
tend to soot (i.e., they tended to burn smokelessly). This was also the
first test which used the sampling and chemical analysis methods developed
for the EPA by Howes. The steam-assisted flare was tested with exit flow
velocities ranging up to about 19 m/s (63 ft/sec), with heat contents from
11 to 84 MJ/scm (300 to 2,200 Btu/scf) and with steam to gas (weight) ratios
varying from 0 (no steam) to 6.86. Air-assisted flares were tested with
fuel gas heat contents as low as 3 MJ/scm (83 Btu/scf). Flares without
assist were tested down to 8 MJ/scm (200 Btu/scf). All of these tests,
except for those with very high steam to gas ratios, showed combustion
efficiencies of over 98 percent. Flares with high steam to gas ratios
(about 10 times more steam than that required for smokeless operation) had
lower efficiencies (69 to 82 percent) when combusting 84 MJ/scm
(2,200 Btu/scf) gas.
After considering the results of these five studies, the EPA has
concluded that 98 percent combustion efficiency can be achieved by steam-
4-31
-------
assisted flares with exit flow velocities less than 19 m/s (63 ft/sec) and
combustion gases with heat contents over 11 MJ/scm (300 Btu/scf) and by
flares operated without assist with exit flow velocities less than 18 m/s
(60 ft/sec) and burning gases with heat contents over 8 MJ/scm
(200 Btu/scf). Flares are not normally operated at the very high steam to
gas ratios that resulted in low efficiency in some tests because steam is
expensive and operators make every effort to keep steam consumption low.
Flares with high steam rates are also noisy and may be a neighborhood
nuisance.
The EPA has a program under way to determine more exactly the
efficiencies of flares used in the petroleum refining industry/SOCMI and a
flare test facility has been constructed. The combustion efficiency of four
flares (1 1/2 inches to 12 inches ID) will be determined and the effect on
efficiency of flare operating parameters, weather factors, and fuel
composition will be established. The efficiency of larger flares will be
estimated by scaling.
4.2.1.4 Applicability . Flares are commonly used at refineries as
emission control devices. They can be used for almost any VOC stream and
can handle fluctuations in VOC concentration, flow rate, and inerts content.
Flares should be applicable to the control of VOC emissions from oil-water
separators, air flotation systems, and closed drains systems. Flares would
be particularly attractive for these processes if existing flares are
accessible at a given refinery. Small ground flares dedicated to the
wastewater treating units might be considered as an alternative to directing
the captured VOC emissions into the refinery flare system.
4.2.2 Carbon Adsorption
Carbon adsorption is a method of controlling VOC emissions by fixation
of the organic compounds to the surface of activated carbon. When the
capacity of the carbon to adsorb VOC is exhausted, the spent carbon is
replaced or regenerated. Carbon adsorption is a proven technology for the
control of numerous organic compounds from a wide variety of industrial
4-32
-------
sources, including refinery wastewater sources. The theory and operating
principles of carbon adsorption have been extensively reviewed in the
literature. A brief description of the technology, factors affecting its
performance, and its potential as a VOC control method for refinery
wastewater sources are discussed in this section.
4.2.2.1 Operating Principles. Two basic configurations of carbon
adsorption systems are typically used for VOC control—regenerative and
non-regenerative systems.
In regenerative systems, multiple and separate carbon beds are
typically used to remove and concentrate organic compounds from a gas
stream. The beds alternate adsorption/regeneration duty in a cyclical
manner. Regeneration of spent carbon is normally accomplished by in situ
thermal desorption of the organics, usually by stripping with low pressure
steam. The desorbed organics and steam are condensed and separated. The
water phase is reused, further processed, or discarded without further
treatment. The recovered organic phase is typically reused. In a refinery
application, the recovered organics would be reprocessed or used as fuel.
In non-regenerative systems, the basic absorption mechanism is
identical. However, when activated carbon in a non-regenerative system
becomes spent, it is simply replaced with a fresh charge. The spent carbon
is discarded or reactivated off-site for eventual reuse. Equipment
requirements are much less complex, but periodic carbon replacement is
necessary.
The feasibility of using regenerative or non-regenerative carbon
adsorption for a particular VOC control application is determined primarily
by operating economics, with the cost difference largely dependent on the
required frequency of regeneration or carbon replacement. VOC sources
within refinery wastewater systems are expected to emit varying
concentrations and types of organics, but at relatively low total mass
rates. Therefore, the activated carbon charge in a VOC control system would
probably become spent only at infrequent intervals. For this reason and
other described in the following discussion, the less complex
4-33
-------
non-regenerative configuration appears to be more applicable to the control
of VOC emissions from refinery wastewater sources.
A typical non-regenerative system is shown in Figure 4-6. The effluent
gas streams are ducted to one or multiple parallel vessels containing
activated carbon particles held in fixed beds. The VOC are adsorbed onto
the surface of the carbon, and the treated gas exits at a very low VOC
concentration. As the capacity of the carbon bed to adsorb additional VOC
is exceeded, the outlet VOC concentration begins to increase. This increase
in concentration is referred to as VOC breakthrough and signals the need for
carbon replacement.
4.2.2.2 Factors Affecting Performance and Applicability. Factors that
affect the adsorption capacity of activated carbon in non-regenerative
systems include:
o VOC type and inlet mass loading;
o moisture content of the inlet gas;
o temperature of the inlet gas; and
o carbon type, amount, and condition.
Similarly, these factors determine the performance and applicability of
carbon adsorption as a VOC control method for refinery wastewater sources.
The types of VOC vented to a carbon adsorption system from wastewater
sources are variable. The majority of the compounds are low boiling
compounds since wastewater system normally operate at temperatures below
140°F. Typical compounds emitted during emissions testing of air flotation
systems included paraffins and aromatics such as benzene, toluene, and
xylene. The nature of the organics emitted would not result in any
significant carbon fouling problems. However, if severe carbon fouling did
occur, off-site carbon reactivation (non-regenerative systems) would be the
most practical choice, since high boiling compounds are difficult to remove
by steam stripping. Furthermore, if the carbon would need regeneration/
replacement only infrequently, the organics on the carbon may become even
more irreversibly adsorbed due to chemical or polymerization reactions that
may occur because of the long residence time on the carbon. While the light
4-34
-------
TREATED
GAS
\ \ \ \ \ \ \ \ \\
ACTIVATED CARBON
\\\ \\x\\\\
Figure 4-6. Schematic of Non-Regenerative Carbon Adsorption
for VOC Control.
4-35
-------
molecular weight of the emitted organics may preclude severe carbon fouling,
the full potential adsorption capacity of the carbon might not be realized.
Activated carbon has a greater affinity for larger nonpolar molecules; very
46
light organics can pass through carbon virtually uncontrolled.
The VOC mass rate is determined by the inlet gas flow rate and the VOC
concentration. The VOC mass rate is of significance in determining the
service life of the carbon. The inlet gas flow rate affects the gas-phase
residence time in the bed and therefore the VOC control efficiency. If VOCs
are conveyed in an oxygen-containing gas stream, the inlet VOC concentration
is of significance for safety reasons—the concentration should be outside
of the explosive range of the mixture. In a refinery wastewater control
application, the source(s) might be purged with nitrogen or refinery fuel
gas to reduce the possibility of oxygen contamination. Nitrogen may be the
preferred purge gas; fuel gas would not only increase the chance of an
explosive situation but would also represent an additional VOC loading for
the carbon adsorption control system.
Moisture content of the inlet gas stream affects the adsorption capacity
of the carbon for VOCs. Water vapor competes with organic compounds for
adsorption sites, particularly at moisture levels corresponding to relative
humidities greater than 50 percent. Therefore, saturation or near-saturation
levels of moisture in VOC-laden gas streams from wastewater sources may
significantly inhibit the ability of carbon adsorption systems to control
VOCs. Demister pads are used by one refinery to remove excess moisture from
47
the VOC gas stream.
VOC adsorption capacity is inversely related to inlet gas temperature.
Most carbon adsorption systems are designed to treat gas streams having
temperature lower than 120°F. The temperatures of VOC-laden gas streams
from refinery wastewater sources should be within the acceptable range.
Finally, the properties of the carbon within the beds significantly
affect the VOC control efficiency. Many types and grades of carbon are
available. Selection of the appropriate carbon types and amount will
determine its adsorption capability and service life. The ease of
4-36
-------
replacement is important to the overall design, whether the carbon is
removed from containment vessels (e.g., by vacuum) or whether the
48
containment vessels themselves are removable (e.g., 55-gallon drums).
4.2.2.3 Control Efficiency. Well-designed and operated state-of-the-
art carbon adsorption systems can reliably remove 95 percent of many types
49
of VOCs from contaminated gas streams. Some systems are capable of
achieving VOC control efficiencies exceeding 99 percent. A non-regenera-
tive system tested at one refinery was operating at 90 percent efficiency.
This system was controlling VOC emissions from an equalization tank of the
47
wastewater treatment system.
A non-regenerative carbon adsorption system must be designed and
operated conservatively and/or be monitored continuously to ensure that it
is controlling VOC emissions efficiently. Frequent replacement of carbon
and continuous monitoring of the treated exhaust gas for VOC content are two
methods whereby maximum VOC control efficiency can be maintained.
4.2.3 Incineration
Incineration, or thermal oxidation, is a method for controlling VOC
emissions by high-temperature oxidation of the organic compounds to carbon
dioxide and water. Incineration is recognized as the most universally
applicable of available VOC control methods because it can be used to
destroy essentially all types of organic compounds from a variety of
*il *5? 5"?
sources, including refinery wastewater sources. ' ' The technology is
described briefly in this section, with emphasis placed upon its potential
as a VOC control device for wastewater sources.
4.2.3.1 Operating Principles Design specifications for incinerators
used for VOC control devices may vary considerably, but the basic design and
operating principles are represented by the schematic system shown in
Figure 4-7. In this system, the VOC-laden gas stream is ducted from the
emission sources to a burner zone. A flame is established in the burner
zone by combustion of auxiliary fuel (e.g., refinery fuel gas) and air. The
4-37
-------
voc-
LAOEN
GAS
EXHAUST TO
ATMOSPHERE
AUXILIARY.
FUEL
1
1
BURNER
A
COMBUSTION
ZONE
OPTIONAL
HEAT
RECOVERY
AIR
Figure 4-7. Schematic of Incineration System for VOC Control.
4-38
-------
high-temperature gases are expanded into a combustion chamber maintained at
a constant temperature, typically in the range of 1000°F to 1600°F. The
gases remain in the combustion zone for a residence time sufficient to
oxidize the VOC, typically 1 second or less. The combustion products are
then exhausted to the atmosphere. Heat recovery (e.g., inlet air preheat)
can be employed to minimize fuel consumption.
4.2.3.2 Factors Affecting Performance and Applicability. A number of
factors determine the effectiveness of incineration as a VOC control method.
These include:
o inlet waste stream characteristics;
o temperature;
o residence time;
o auxiliary fuel/air requirements; and
o other design parameters.
The effect of these factors on incineration systems is discussed below.
Incineration represents a flexible control method in terms of inlet VOC
type and concentration. Factors relevant to induction of the inlet waste
stream from refinery wastewater to an incinerator are similar to those
described for carbon adsorption in Section 4.2.2. In summary, oxygen-free
purge gases would be preferred. One possible handicap inherent with an
incineration system might be the necessity of a relatively constant inlet
flow rate. VOC-laden gases can be allowed to "breathe" through a carbon
adsorber, but an incinerator may require a steadier inlet flow rate of waste
gases from wastewater sources in order to sustain stable flame conditions.
An incinerator can handle minor flow fluctuations, but more severe flow
54
fluctuations might require the use of a flare for VOC control.
Combustion zone temperature can have a pronounced effect on VOC
destruction efficiency and auxiliary fuel consumption. The required
temperature, which is controlled by the auxiliary fuel flow rate, would be
determined by the VOC type and the required level of control. Figure 4-8
represents an example case showing the effect of combustion zone temperature
on VOC destruction efficiency.
4-39
-------
100
I
-e»
O
HYDROCARBONS
ONLY
HYDROCARBON AND CARBON
MONOXIDE (PER LOS ANGELES
AIR POLLUTION CONTROL
DISTRICT RULE 66)
UJ
60
1150 1200 1250 1300 1350 1400
TEMPERATURE. °F
1450
1500
1550
Figure 4-8. Typical Effect of Combustion Zone Temperature on
Hydrocarbon and Carbon Monoxide Destruction Efficiency.
55
-------
In addition to combustion zone temperature, gas-phase residence time in
the combustion zone also contributes to the degree of completion of the
oxidation reaction. Residence times on the order of 0.3 seconds to
1.5 seconds are typical for VOC control applications. ' ' ' ' '
Auxiliary fuel and air requirements also affect the operation of an
incinerator. Fuel type affects the design of an incinerator and fuel rate
determines its operating costs. Some excess air is required for proper
fuel/air mixing and completion of the combustion reaction. However, too
much excess air can have a negative impact on auxiliary fuel requirements
(heat losses) and design size.
Other factors affect the performance and applicability of incineration
as a VOC control method for refinery wastewater sources. A major
consideration is heat recovery. Primary or secondary heat recovery is often
utilized to minimized operating costs. Primary heat recovery refers to heat
exchange between the hot combustion gases and the cool inlet VOC-laden gas
or auxiliary air stream. Secondary heat recovery refers to heat transfer
between an incinerator gas stream and an adjacent, yet separate, process
stream. Use of secondary heat would be limited to those situations in which
such a process stream was adjacent and available to serve as a heat sink.
Incineration represents a simple and reliable method of VOC control,
but several problems can limit its performance. Fouling can occur,
particularly on heat exchange surfaces, although the probability of
significant fouling may be low for a refinery wastewater control
application. Incinerator internals may be subject to corrosion in the
presence of sulfur- or halogen-containing compounds. The existence of the
former would be expected in refinery wastewater effluent gases, but its
potential for causing corrosion problems in an incinerator is unknown.
Also, operation of an incinerator can be expected to result in secondary
emissions of oxides of nitrogen, carbon monoxide, and possibly
combustion-created organic reaction products. However, proper design and
operation of the incinerator should result in negligible secondary emission
problems.
4-41
-------
4.2.3.3 Control Efficiency Incineration of VOCs from refinery
wastewater would be expected to achieve destruction efficiencies equivalent
to those achieved in other applications (i.e., 90 percent to 99+ percent at
temperatures between 1,000°F and l,600°F).54'56'59'60'61 The performance of
incineration with regard to VOC destruction efficiency would not be expected
to degrade over a period of time, as is typically the case for carbon
adsorption and catalytic oxidation systems.
4.2.4 Catalytic Oxidation
Catalytic oxidation is a method of controlling VOC emissions by
oxidation to carbon dioxide and water in the presence of a catalyst. Many
factors important to the design and operation of a catalytic oxidation VOC
control system parallel those of an incineration system, which were
described above. Therefore, the discussion in this section will be limited
to those aspects of catalytic oxidation that cause it to differ
significantly from incineration with regard to VOC control.
4.2.4.1 Operating Principles. Catalytic oxidation featues the use of
a metal- or metallic-alloy based catalyst to promote higher rates of VOC/
oxygen reactions at lower energy (temperature) levels. Thus, temperature
and auxiliary fuel requirements are lowered. A schematic diagram of a
typical catalytic oxidation system is shown in Figure 4-9. It is generally
similar to the incineration system described previously, except for the
presence of a catalyst chamber downstream of the burner zone.
In operation, the VOC-laden gas is typically heated to 500°F to 900°F
by contact with hot combustion products of an auxiliary fuel/air burner.
The heated gas then enters the catalyst chamber. The catalyst chamber
contains the catalyst material fixed on a substrate structure of large
surface area (e.g., pellets or a honeycomb configuration). The catalyst
consists of platinum-, palladium-, copper-, chromium-, nickel-, cobalt-,
managanese-, or rhodium-based material layered onto the substrate. ' VOC
oxidation occurs in the catalyst bed, with subsequent release of heat and an
increase in temperature. The treated gas, at 700°F to 1200°F, exits the
4-42
-------
VOC-LADEN
GAS
•CATALYST
EXHAUST TO
ATMOSPHERE
AUXILIARY
FUEL
J
BURNER
*
1
OPTIONAL
HEAT
RECOVERY
AIR
Figure 4-9.
Schematic of Catalytic Oxidation System for
VOC Control.
4-43
-------
reaction chamber and is exhausted to the atmosphere. Temperature is
controlled by auxiliary fuel flow rate; the controlling temperature can be
measured at the catalyst inlet or outlet or as the average of the inlet and
outlet.52'53'56
4.2.4.2 Factors Affecting Performance and Applicability. Catalytic
oxidation present potential advantages over incineration, but its use is
limited because of its sensitivity to inlet waste stream characteristics.
If inlet VOCs are relatively heavy in molecular weight, they may
collect or polymerize on the catalyst surface, thus reducing the available
surface area of the catalyst. Also, the presence of sulfur-, halogen-, or
heavy metal-containing compound in the inlet gas can poison the catalyst or
suppress its activity. ' The presence of the former could be expected in
waste gas streams from refinery wastewater. When the catalyst is poisoned
or deactivated, a portion of the inlet VOCs can either pass through the
system uncontrolled or be converted to aldehydes, ketones, or organic
CO
acids. Also, typical catalytic oxdiation systems are unable to handle
excursions of high inlet VOC concentrations. Excessive VOC loading can
increase the heat release in the catalyst bed such that temperatures become
high enough to sinter (deactivate) or volatilize the catalyst.
The gradual loss of catalyst activity due to any of the reasons
described above introduces additional maintenance requirements for catalyst
cleaning and/or replacement.
4.2.4.3 Control Efficiency. Catalytic oxidation systems can achieve
VOC destruction efficiencies approaching 99 percent. ' However, certain
data indicate that, to achieve destruction efficiencies approaching or
exceeding 95 percent, operating temperatures have to increase to levels that
56
threaten to sinter or deactivate the catalyst. Recent test data for
catalytic oxidation systems used in other industrial for VOC control
indicate that half of the tested units achieved greater than 90 percent VOC
destruction.57 The remaining tested units were capable of achieving 80 or
57
90 percent VOC destruction.
4-44
-------
4.2.5 Condensation
In a vapor containing two components, one of which is essentially
non-condensible at system conditions, condensation of the condensible
component occurs when its partial pressure exceeds its vapor pressure. Any
component in a vapor mixture can ultimately be condensed if the temperature
is lowered far enough. The point where condensation first occurs is called
the dew point. As the vapor is cooled below the dew point, condensation
will continue until the partial pressure in the vapor phase is once again
equal to the vapor pressure of the liquid phase at the lower temperature.
In the cases where the hydrocarbon concentration in the gas phase is
high, condensation is relatively easy. When concentrations are low,
condensation at reasonably achieved temperatures can be difficult.
Table 4-5 contains some examples of the temperatures required to-achieve
90-95 percent condensation of some organic solvents. It can be seen that
relatively low temperatures are needed, even for compounds such as xylene,
r o
toluene, benzene and hexane. These compounds are commonly found in
gaseous emissions from wastewater systems.
There are two ways to obtain condensation. First, at a given tempera-
ture, the system pressure may be increased until the partial pressure of the
condensible component exceeds its vapor pressure. Alternately, at a fixed
pressure, the temperature of the gaseous mixture may be reduced until the
partial pressure of the condensible component exceeds its liquid-phase vapor
pressure. In practice, condensation is achieved mainly through removal of
heat from the vapor. Also in practice, some components in multicomponent
condensation may dissolve in the condensate even though their boiling points
are below the exit temperature of the condenser.
Condensers employ several methods for cooling the vapor. In surface
condensers, the coolant does not contact the vapors or condensate; condensa-
tion occurs on a wall separating the coolant and the vapor. In contact
condensers, the coolant, vapors, and condensate are intimately mixed.
Most surface condensers are common shell-and-tube heat exchangers. The
coolant usually flows through the tubes and the vapors condenses on the
4-45
-------
Table 4-5. Physical Constants and Condensation Properties of Some Organic Solvents.
25* of LEL
Concentration
Compound
Dodecane
Plnene
(Terpentine)
0-xylene
Toluene
Me
Benzene
Methanol
C2H60
Normal
Boiling
Point, °F
421
360
260
211
175
147
LEL,
0.6
0.7
1.0
1.4
1.3
6.0
Partial
Pressure,
mm of Hg
1.1
1.3
1.9
2.7
2.5
11.4
Dew
Point,
°F
120
53
26
5
-15
2
90* Condensation
From 25* of LEL
Partial
Pressure,
mm of Hg
0.11
0.13
0.19
0.27
0.25
1.14
Temp,
°F
61
116
-31
-51
-69
-41
95% Condensation
From 25* of LEL
Partial
Pressure,
mm of Hg
0.55
0.065
0.095
0.135
0.125
0.57
Temp,
°F
54.4
-31.4
-36.5
-54.3
-96.4
-08.7
90* Condensation
From 200 ppm
Partial
Pressure,
mm of Hg
0.15
0.015
0.015
0.015
0.015
0.015
Temp,
°F
19
-60
-72
-103
-114
-126
Hexane 155 1.2 2.3 -39 0.23 -93 0.115 -108 0.015 -129
-------
outside tube surface. The condensed vapor forms a film on the cool tube and
drains away to storage or disposal. Air-cooled condensers are usually
constructed with extended surface fins; the vapor condenses inside the
finned tubes.
Contact condensers usually cool the vapor by spraying an ambient
temperature or slightly chilled liquid directly into the gas stream. Contact
condensers also act as scrubbers in removing vapors which normally might not
be condensed. The condensed vapor and water are then usually treated and
discarded as waste. Equipment used for contact condensation includes simple
spray towers, high velocity jets, and barometric condensers.
Contact condensers are, in general, less expensive, more flexible and
more efficient in removing organic vapors than surface condensers. On the
other hand, surface condensers may recover marketable condesate and minimize
waste disposal problems. Often condensate from contact condensers cannot be
reused and may require significant wastewater treatment prior to disposal.
The coolant used in surface condensers depends on the saturation
temperature (dew point) of the VOC. Chilled water can be used to bring
temperatures as low as 7°C, brines down to -34°C, and freonsbelow -34°C.
The major pieces of equipment in a condenser system consist of the
condenser, refrigeration system, storage tanks, and pumps. A typical
arrangement is shown in Figure 4-10.
4.2.5.1 Factors Affecting Performance and Applicability. Condensers
are not well suited to treatment of gas streams containing VOC with low
boiling points or streams containing large quantities of inert and/or
noncondensible gases such as air, nitrogen, or fuel gas (methane).
Condensers used for VOC control must often operate at temperatures
below the freezing point of water. Thus, moist vent streams (such as would
be present in gas streams from wastewater sources) must be dehumidified
before treatment to prevent the formation of ice in the condenser.
Particulate matter should be removed because it may deposit on the tube
surfaces and interfere with gas flows and heat transfer. Gas flow rates in
4-47
-------
VOC-LADEN -
GAS -
"**" DE
COOLA
RETUF
R
UNIT ^1
*\
(>
1
A COOLANT
EFRIGERATION
PLANT
CLEANED
GAS OUT
MAIN ^
CONDENSER J
\ \
1 CONDENSED
voc
\ '
STORAGE
1 ^ TO PROCESS
^^ OR DIPOSAL
Figure 4-10. Condensation System
54
4-48
-------
the range of 100-200 cfm are typical of the capacities of condensers used as
emission control devices.
Vent streams containing less than 0.5 percent VOC are generally not
64
considered for control by condensation.
Oil-water separators and air flotation systems usually operate at
temperatures below 140°F. The vapor streams from these sources will
generally be saturated with water and will probably contain a large number
of compounds with a broad range of boiling points. It is doubtful whether a
condenser system can be effective as a primary VOC control device. There
could conceivably be applications in which the gas stream from the emission
sources is first passed through a condenser to recover some of the "higher
boiling" compounds.
4.2.5.2 Control Efficiency. The VOC removal efficiency of a
condenser is highly dependent upon the type of vapor stream entering the
condenser, and on the condenser operating parameters. Efficiencies of
65
condensers usually vary from 50 to 95 percent.
4.2.6 Industrial Boilers and Process Heaters
Industrial boilers and heaters are widely used for the thermal
destruction of captured VOC emissions. A brief description of the
technology, factors affecting its performance and its potential as a VOC
control method for refinery wastewater sources are discussed below.
4.2.6.1 Operating Principles. Boilers and process heaters are used
extensively in petroleum refineries. They represent a potential emissions
control system for combusting captured VOC emissions from sources in
refinery wastewater systems.
Industrial Boilers. Most refineries use boilers to provide steam
for direct use of various processes (e.g., light end strippers), for heating
and for the production of electrical power (via steam turbines). Boilers in
refineries are fired with the most available (and economical) fuel, such as
4-49
-------
purchased natural gas, refinery fuel gas (mostly methane), residual oil, and
and combinations of these various fuel types. Surveys of industrial boilers
used in the chemical industry have shown that the majority are of watertube
design, and it seems reasonable to assume that similar situation prevails in
54
the petroleum industry.
A watertube boiler is designed such that hot combustion gases are
present outside of heat transfer tubes. Water flows inside the tubes and is
vaporized by the heat that is transmitted through the tube walls. The
tubes are interconnected to stream drums in which the steam and hot water
are collected, separated, and stored. The water tubes are relatively small
in diameter (2.0 inch being a typical diameter) to produce high liquid
velocities, good heat transfer, rapid response to steam demands, and
relatively high thermal efficiency. The thermal efficiency of-the tubes
and drum system can be as high as 85 percent. The efficiency can be
increased by recovering heat from the flue gas by exchange with combustion
air or feedwater.
When firing natural gas, forced or natural draft burners are used to
thoroughly mix the incoming fuel and combustion air. If a waste gas stream,
such as that from an oil-water separator vent, is combusted in a boiler, it
can either be mixed with the incoming fuel or fed directly to the furnace
through a separate burner. A particular burner design commonly known as a
high intensity or vortex burner can be effective for waste gas streams with
low heating values (i.e., streams where a conventional burner may not be
applicable). Effective combustion of streams with low heating values is
accomplished in a high intensity burner by passing the combustion air
through a series of spin vanes to generate a strong vortex.
Furnace residence time and temperature profiles for industrial boilers
vary as a function of the furnace and burner configuration, fuel type, heat
input, and excess air level.67 This model predicts mean furnace residence
times of from 0.25 to 0.83 seconds for natural gas-fired water tube boilers
in the size range from 4.4 to 44 MW (15 to 150 x 106 Btu/hr). Furnace exit
temperatures for this range of boiler sizes are at or above 1475°K (2810°F).
Residence times for oil-fired boilers are similar to those of the natural
54
gas-fired boilers.
4-50
-------
Process Heaters. Process heaters are used in petroleum refineries as
reboilers for distillation columns and to provide heat for reaction (naptha
reforming, thermal cracking, coking) and for preheating feed stocks.
Natural gas, refinery fuel gas, and various grades of fuel oil are all used
to fire process heaters.
There are many variations in the design of process heaters, depending
on the application considered. In general, the radiant section consists of
the burner(s), the firebox, and a row of tubular coils containing the
process fluid to be heated. Most heaters also contain a convective heat
transfer to the process fluid.
Process heater applications in the petroleum refining industry can be
broadly classified with respect to firebox temperature: (1) low firebox
temperature applications such as steam superheaters, and (3) high firebox
temperature applications such as thermal cracking furnaces and catalytic
reformers. Firebox temperatures within the refining industry can be
expected to range from about 750°F for preheaters and reboilers to more than
2000°F for coking process furnaces.
4.2.6.2 Factors Affecting Performance and Applicability. The primary
function of boilers and heaters in refineries is to generate steam and
provide process heat, respectively. Their successful operation is critical
for the successful operation of refinery process units. Thus, it is
extremely important that any injection of waste gases be done in a manner
that precludes any reduction in the efficiency, operability, and/or
reliability of the affected heater or boiler. Variability in the flow rate
or composition of gas streams from wastewater sources could have an effect
on the combustion characteristics and heat output if the stream represents a
significant source of fuel relative to the normal fuel rate.
Waste streams containing relatively high concentration of chlorinated
or sulfur-containing compound could cause corrosion problems in
heater/boilers that are not designed to handle either the compounds or their
combustion products. When such VOC compounds are burned, the flue gas
temperature must be maintained above the acid dew point to prevent acid
4-51
-------
condensation and subsequent corrosion. However, the VOC being emitted from
refinery wastewater sources is expected to contain minimal amounts of
sulfur- or halogen-containing compounds.
If the volume of the waste gas stream is significant when compared to
that of the heater/boiler fuel, its injection could affect the heat transfer
characteristics of the furnace. Heat transfer characteristics are dependent
on the flow rate, heating value, and elemental composition of the waste gas
stream, and the size and type of heat generating unit being used. Often,
there is no significant alteration of the heat transfer, and the organic
content of the water gas stream can, in some cases, lead to some reduction
in the amount of fuel required to achieved the desired heat production.
Wastewater streams are expected to be relatively small compared to the total
amount of fuel provided to most heaters and boilers in refineries.
If the waste stream volume is significant, and the heat content
relatively low, the change in heat transfer characteristics after injecting
the waste stream could have an adverse effect on the heater/boiler
performance. Even equipment damage could result. In addition to these
reliability problems, there are also potential safety problems associated
with ducting wastewater emission vent to a boiler or process heater.
Variation in the flow rate and organic content of the vent stream could
cause extensive damage. Another related problem is flame fluttering which
could result from these variations. Potential flashback is another
possibility that must be considered. Presently, there is only one refinery
known to be venting emissions from an air flotation system to a process
CO
heater. No safety problems have been reported by the refinery.
4.2.6.3 Control Efficiency. Some testing has been performed to
evaluate the performance of boilers and heaters in destroying hydrocarbon
gases injected into the flame zones of the combustion devices. The EPA
sponsored a test to determine the capability of an industrial boiler for
destroying polychlorinated biphenyls (PCB). A relatively small quantity
of PCB is added to the fuel oil which is then burned in the boiler. The
test results indicated that more than 99.9 percent of the PCB was destroyed
in the boiler.
4-52
-------
Other tests conducted by EPA measured the efficiency of five processes
heaters for destroying a mixture of benzene off-gas and natural gas. ' '
The heaters were representative of those with both low- and medium-
temperature fireboxes. In both types of heaters, more than 99 percent of
the total C. to Cg hydrocarbons in the gas injected into the flame zone was
destroyed.
Thus, when boilers or process heaters are available, it appears that
they are acceptable control devices for waste gas streams. In general, they
appear to be at least 98 percent efficient for destroying VOC in the vapor
phase. The collected VOC gas streams from refinery wastewater sources may,
in some cases, be suitable for control with this technology.
4-53
-------
4.3 REFERENCES
1. Vincent, R. Control of Organic Gas Emissions from Refinery Oil-Water
Separators. California Air Resources Board. Sacramento, California.
April 1979, p. 4.
2. Racine, W.J. Plant Designed to Protect the Environment. Hydrocarbon
Processing. ^l.(3):115. March 1972.
3. Trip Report. R.J. McDonald to J. Durham, EPA:CPB. June 10, 1982, p.
2. Report of June 9, 1982 visit to Exxon Company, Baton Rouge
Refinery.
4. Trip Report. Laube, A.H., and R.G. Wetherold to R.J. McDonald,
EPA:CPB, July 19, 1983. Report of March 25, 1983 visit to Sun Oil
Company, Toledo, Ohio Refinery.
5. Memo and Attachment from Mitsch, B.F., Radian Corporation, to file.
November 11, 1983. Screening Data from Process Drains at Total
Petroleum, Alma, Michigan.
6. Memo and Attachment from Mitsch, B.F., Radian Corporation, to file.
November 11, 1983. Screening Data from Process Drains at Golden West
Refinery, Santa Fe Springs, California.
7. Memo and Attachment from Mitsch, B.F., Radian Corporation, to file.
November 11, 1983. Screening Data from Process Drains at Phillips
Refinery, Sweeny, Texas.
8. Memo from Wetherold, B. and Mitsch, B. F., Radian Corporation to file.
January 26, 1984. Analysis of Drain Screening Data from Phillips,
Sweeny, Texas.
9. Wetherold, R. G., L. P. Provost, and C. D. Smith. (Radian
Corporation.) Assessment of Atmospheric Emissions from Petroleum
Refining. Volume 3. Appendix B: Detailed Results. Prepared for U.S.
Environmental Protection Agency. Research Triangle Park, N.C.
Publication No. EPA 600/2-80-075C. April 1980.
10. Thibodeaux, L.J. Chemodynamics. New York, John Wiley and Sons. 1979.
11. Dean, J.A. Lange's Handbook of Chemistry. New York, McGraw-Hill Book
Company. 1979.
12. Treyball, R.E. Mass-Transfer Operations. New York, McGraw-Hill Book
Company. 1980.
13. Reid, R.C., J.M. Pransnitz and T.F. Sherwood. The Properties of Gases
and Liquids, New York, McGraw-Hill Book Company. 1977.
4-54
-------
14. McAllister, R.A. (TRW, Incorporated) Internal Floating Roof Technical
Analysis. (Prepared for U.S. Environmental Protection Agency.
Research Triangle Park, North Carolina. January 1983.
15. McCabe, W.C., J.C. Smith. Unit Operations of Chemical Engineering.
New York, McGraw-Hill Book Company. 1976.
Drivas, P.O. Calculation of Evaporative Emissions from Multicomponent
Liquid Spills. Environmental Science and Technology 16(10):726-728.
rt_. _ L 1 f\rt O
16.
October 1982.
17. Los Angeles County Air Pollution Control District. Air Pollution
Engineering Manual. Second Edition. Prepared for the
U. S. Environmental Protection Agency. Research Triangle Park, N.C.
Publication No. AP-40. May 1973. p. 675.
18. American Petroleum Institute. Manual on Disposal of Refinery Wastes;
Volume on Atmospheric Emissions. API Publication 931. Washington D.C.
1976, p. 7-6.
19. Trip Report. Laube, A.H. and G. DeWolf, Radian Corporation,
R.J. McDonald, EPA:CPB. July 12, 1983. Report of March 14, 1983 visit
to Tosco Corporation in Bakersfield, California.
20. Trip Report. Laube, A.H., Radian Corporation, to EPA:CPB.
May 17, 1983. Report of March 17, 1983 visit to Mobil Oil in Torrance,
California.
21 Trip Report. Laube, A.H. and G. DeWolf, Radian Corporation, to
R.J. McDonald, EPA:CPB. June 3, 1983. Report of March 14, 1983 visit
to Champlin Petroleum Company in Wilmington, California.
22. Utah Bureau of Air Quality. Engineering Review Analysis - Summary.
Installation of Covers on Wastewater Separators at Chevron, U.S.A.,
Inc. Salt Lake City, UT. May 1983, p. 1-2.
23. Litchfield, O.K. Controlling Odors and Vapors from API Separators.
Oil and Gas Journal. 69_(44):60-62. November 1, 1971.
24 Trip Report. Wetherold, R.G. and A.H. Laube, Radian Corporation, to
R.J. McDonald, EPA:CPB. July 19, 1983. Report of March 25, 1983 visit
to Sun Oil Company's refinery in Toledo, Ohio.
25 Utah Bureau of Air Quality. Engineering Review Analysis. Summary.
Installation of Covers on Wastewater Separators at Amoco Oil Company.
Salt Lake City, UT. December 1981.
26. Petrex Incorporated. General Plan View of Oil-Water Separator. Woods
Cross, Utah. August 1982.
4-55
-------
27. Ref. 17. p. 7-2.
28. U.S. Environmental Protection Agency. Compilation of Air Pollutant
Emission Factors. Third Edition. Research Triangle Park, N.C.
Publication No. AP-42. August 1977. P. 9.1-19.
29. Ref. 1, p. 10.
30. Telecon, Mitsch, B.F., Radian Corporation, with Bassett, C., Huntway
Refining Company. April 25, 1984. Conversation about DAF system.
31. Telecon, Mitsch, B.F., Radian Corporation, with Crawford, D., Sigmor
Refining. June 29, 1983. Conversation about DAF system.
32. Chevron U.S.A., Inc. (El Segundo, California Refinery). Letter and
Survey Attachment to Terry McGuire, California Air Resources Board.
October 16, 1978.
33. Telecon. Laube, A.H. Radian Corporation with F.E. Carleton, IVEC.
December 3, 1982. Wastewater treatment system.
34. Trip Report. Laube, A.M., Radian Corporation, to McDonald, R.J., EPA.
May 17, 1983. Report of March 17, 1983 visit to Mobil Oil Corporation
Refinery at Torrance, California.
35. Memo from Mitsch, B.F., Radian Corporation, to file. May 16, 1984.
Regulatory Alternative II for Air Flotation Systems.
36. Memo from Hunt, G. and Mitsch, B., Radian Corporation to file. April
16, 1984. Analysis of Emission Potential for Induced and Dissolved Air
Flotation Systems.
37. Laverman, R.J., T.J. Haynie, and J.F. Newbury. Testing Program to
Measure Hydrocarbon Emissions from a Controlled Internal Floating Roof
Tank. Prepared for American Petroleum Institute. Chicago Bridge and
Iron Company. Chicago, Illinois. March 1982.
38. Kalcevic, V. (IT Enviroscience). Control Device Evaluation Flares and
the Use of Emissions as Fuels. In: U.S. Environmental Protection
Agency. Organic Chemical Manufacturing Volume 4: Combustion Control
Devices. Research Triangle Park, N.C. Publication No. EPA
450/3-80-026. December 1980. Report 4.
39. Klett, M.G. and J.B. Galeski. (Lockhead Missiles and Space
Company, Inc.) Flare Systems Study. (Prepared for U. S. Environmental
Protection Agency.) Huntsville, Alabama. Publication No.
EPA-600/2-76-079. March 1976.
4-56
-------
40. Joseph D., et al. Evaluation of the Efficiency of Industrial Flares
Used to Destroy Waste Gases, Phase I Interim Report - Experimental
Design. Prepared for U.S. Environmental Protection Agency. Research
Triangle Park, N.C. EPA Contract No. 68-02-3661. January 1982.
41. Palmer, P.A. A Tracer Technique for Determining Efficiency of an
Elevated Flare. E. I. duPont Nemours and Company, Wilmington, DE.
1972.
42. Siegel, K.D. Degree of Conversion of Flare Gas in Refinery High
Flares. Ph.D. Dissertation, Fridericiana University, Karlsruhe, FRG.
February 1980.
43. Lee, K.C. and Whipple, G.M. Union Carbide Corporation. Waste Gaseous
Hydrocarbon Combustion in a Flare. (Presented at 74th APCA Annual
Meeting, Philadelphia, Pennsylvania. June 21-26, 1981.) Air Pollution
Control Association.
44 Howes, J.E., et al. (Battelle Columbus Laboratories). Development of
Flare Emission Measurement Methodology. Draft Final Report. Prepared
for U.S. Environmental Protection Agency. Research Triangle Park, N.C.
EPA Contract No. 68-02-2682. August 1981.
45. McDaniel, et al. (Engineering-Science.) A Report of a Flare
Efficiency Study Draft. Prepared for U.S. Environmental Protection
Agency. Research Triangle Park, N.C. September 1982.
46 Radian Corporation. Full Scale Carbon Adsorption Applications Study,
Plant 6. Draft Plant Test Report. Prepared for U.S. Environmental
Protection Agency. October 29, 1982.
47 Trip Report. Mitsch, B.F., Radian Corporation. September 30, 1983.
Report on Emissions Test at Chevron, U.S.A., Incorporated in El
Segundo, California.
48 Trip Report. Laube, H.A., Radian Corporation to R.J. McDonald,
EPA:CPB. June 8, 1983. Report of March 16, 1983 visit to Chevron
U.S.A., Incorporated in El Segundo, California.
49. Stern A.C. Air Pollution, Volume IV. Third Edition. New York,
Academic Press. 1977.
50 Radian Corporation. Full Scale Carbon Adsorption Applications Study,
' Plant 2. Draft Plant Test Report. Prepared for U. S. Environmental
Protection Agency. July 30, 1982.
4-57
-------
51. U.S. Environmental Protection Agency. Control of Volatile Organic
Compound Emissions from Air Oxidation Processes in Synthetic Organic
Chemical Manufacturing Industry. Preliminary Draft Report. June 1981.
EPA-450/3-82-001a Air Oxidation Processes in Synthetic Organic
Manufacturing Industry - Background Information for Proposed Standards
October, 1983.
52. Waid, D.E. "Controlling Pollutants Via Thermal Incineration"
Engineering Progress 68_(8):57-58, August 1972.
55.
56.
57.
58.
59.
60.
61.
Chemical
53. Trip Report. Laube, A.H. Radian Corporation, to R.J. McDonald,
EPA:CPB. May 17, 1983. Report of March 17, 1983 visit of Mobil Oil
Corporation, Torrance, California Refinery.
54. U.S. Environmental Protection Agency. Distillation Operations in
Synthetic Organic Chemical Manufacturing Industries. Background
Information for Proposed Standards. Draft. Research Triangle Park,
N.C. October 1982. EPA-450/3-83--005a. December 1983.
U.S. Environmental Protection Agency. Flexible Vinyl Coatings and
Printing Operations. Background Information for Proposed Standards
Draft EIS. January 1983. EPA-450-3-81-016a.
Sittig, M. Incineration of Industrial Hazardous Wastes and Sludges.
Park Ridge, N.J. Noyes Data Corporation, 1979.
Radian Corporation. Characterization of VOC Emissions from Thermal
Incinerators. Test Report, Plant T-2. Prepared for U.S. Environmental
Protection Agency. July 1983.
Radian Corporation. Characterization of VOC Emissions from Thermal
Incinerators, Test Report, Plant T-l. Prepared for U.S. Environmental
Protection Agency. May 1983.
U.S. Environmental Protection Agecny. Background Information Document
for the Pressure Sensitive Tape and Label Surface Coating Industry.
May 1983. EPA-450/2-80-003a. September 1980.
U.S. Environmental Protection Agency. Control of Volatile Organic
Compounds Emissions from Air Oxidation Processes in Synthetic Organic
Chemical Manufacturing Industry. Preliminary Draft Report. June 1981.
Barrett, R.E., and P.R. Sticksel. Preliminary Environmental Assessment
of Afterburner Combustion System. Prepared for the U.S. Environmental
Protection Agency. EPA 600/7-8-153. Research Triangle Park, N.C.
June 1980.
62 Radian Corporation. Performance of Catalytic Incinerators at
Industrial Sites. Final Report. Prepared for U.S. Environmental
Protection Agency. June 15, 1983.
4-58
-------
63. U.S. Environmental Protection Agency. Control of Volatile Organic
Emissions from Existing Stationary Sources - Volume I: Control Methods
for Surface Coating Operations. EPA 450/2-76-028. Research Triangle
Park, N.C. November 1976.
64. Controlling Emissions with Flare Towers. Chemical Week. 132(21):49.
May 25, 1983.
65. Erikson, D.G. (I.T. Enviroscience.) Control Device Evaluation.
Condensation. U.S. Environmental Protection Agency. Organic Chemical
Manufacturing. Volume 5: Adsorption, Condensation, and Absorption
Devices. Research Triangle Park, N.C. Publication No.
EPA-450/3-80-027.
66. U.S. Environmental Protection Agency. Background Information Document
for Industrial Boilers. Research Triangle Park, N.C.Publication No.
450/3-82-006a. March 1982.
67. U.S. Environmental Protection Agency. A Technical Overview of the
Concept of Disposing of Hazardous Wastes in Industrial Boilers. Draft.
Cincinnati, Ohio. EPA Contract No. 68-03-2567. October 1981.
68. Trip Report. Mitsch, B.F., Radian Corporation. September 30, 1983.
Report on Emissions Test at Golden West Refinery, Santa Fe Springs,
California
69. U.S. Environmental Protection Agency. Evaluation of PCB Destruction
Efficiency in an Industrial Boiler. Research Triangle Park, N.C.
EPA Contract No. 600/2—81-055a. April 1981.
70. U.S. Environmental Protection Agency, Emission Test Report on
Ethylbenzene/Styrene. Amoco Chemicals Company (Texas City, Texas).
Reserch Triangle Park, North Carolina. EMB Report No. 79-OCM-13.
August 1979.
71. U.S. Environmental Protection Agency. Emission Test Report. El Paso
Products Company (Odessa, Texas). Research Triangle Park, North
Carolina. EMB Report No. 79-OCM-15. April 1981.
72. U.S. Environmental Protection Agency. Emission Test Report. USS
Chemicals (Houston, Texas). Research Triangle Park, North Carolina.
EMB Report No. 80-OCM-19. August 1980.
4-59
-------
5. MODIFICATION AND RECONSTRUCTION
In accordance with Title 40 of the Code of Federal Regulations (CFR),
Sections 60.14 and 60.15, an existing facility can become an affected
facility and, consequently, subject to applicable standards of performance if
it is modified or reconstructed. An "existing facility," defined in
40 CFR 60.2, is a facility of the type for which a standard of performance is
promulgated and the construction or modification of which was commenced prior
to the proposal date of the applicable standards. The following discussion
examines the modification and reconstruction provisions and their
applicability to petroleum refinery wastewater systems, specifically, to
process drain systems, oil-water separators, and air flotation systems.
5.1 GENERAL DISCUSSION OF MODIFICATION AND RECONSTRUCTION PROVISIONS
5.1.1 Modification
Modification is defined in Section 60.14 as any physical or operational
change to an existing facility which results in an increase in the emission
rate of the pollutant(s) to which the standard applies. Paragraph (e) of
Section 60.14 lists exceptions to this definition which will not be
considered modifications, irrespective of any changes in the emission rate.
These changes include:
1. Routine maintenance, repair, and replacement;
2. An increase in the production rate not requiring a capital
expenditure as defined in Section 60.2;
3. An increase in the hours of operation;
4. Use of an alternative fuel or raw material if, prior to the
standard, the existing facility was designed to accommodate that alternative
fuel or raw material;
5-1
-------
5. The addition or use of any system or device whose primary function
is the reduction of air pollutants, except when an emission control system is
removed or replaced by a system considered to be less environmentally
beneficial.
6. The relocation or change in ownership of an existing facility.
As stated in paragraph (b), emission factors, material balances,
continuous monitoring systems, and manual emission tests are to be used to
determine emission rates expressed as kg/hr of pollutant. Paragraph (c)
affirms that the addition of an affected facility to a stationary source
through any mechanism -- new construction, modification, or reconstruction --
does not make any other facility within the stationary source subject to
standards of performance. Paragraph (f) allows provisions of the applicable
subpart to supersede any conflicting provisions of 40 CFR 60.14. Paragraph
(g) stipulates that compliance be achieved within 180 days of the completion
of any modification.
5.1.2 Reconstruction
Under the provisions of Section 60.15, an existing facility becomes an
affected facility upon reconstruction, irrespective of any change in emission
rate. A source is identified for consideration as a reconstructed source
when: (1) the fixed capital costs of the new components exceed 50 percent of
the fixed capital costs that would be required to construct a comparable
entirely new facility, and (2) it is technologically and economically
feasible to meet the applicable standards set forth in this part. The final
judgment on whether a replacement constitutes reconstruction will be made by
the Administrator of the EPA. As stated in Section 60.15(f), the
Administrator's determination of reconstruction will be based on:
1. The fixed capital cost of the replacement in comparison to the
fixed capital cost of constructing an entirely new facility;
2. The estimated life of the facility after replacements compared to
the life of a comparable entirely new facility;
3. The extent to which the components being replaced cause or
contribute to the emissions from the facility; and
5-2
-------
4. Any economic or technical limitations in compliance with applicable
standards of performance which are inherent in the proposed replacements.
The purpose of the reconstruction provision is to ensure that an owner
or operator does not perpetuate an existing facility by replacing all but
minor components, support structures, frames, housing, etc., rather than
totally replacing it in order to avoid being subject to applicable
performance standards. In accordance with Section 60.5, the EPA will, upon
request, determine if an action taken constitutes construction (including
reconstruction). As with modification, individual standards may include
specific provisions which refine and limit the concept of reconstruction in
40 CFR 60.15.
5.2 APPLICABILITY OF MODIFICATION AND RECONSTRUCTION PROVISIONS TO VOC
EMISSIONS FROM PETROLEUM REFINERY WASTEWATER SYSTEMS
Changes in refinery product demand and in available refinery feedstocks
are expected to result in a number of modernization and alteration projects
at existing refineries over the next several years. Some of these projects
could result in existing process drain systems, oil-water separators, and air
flotation systems becoming subject to regulation under provisions of Sections
60.14 and 60.15. Examples in which this could occur are presented below.
5.2.1 Modification
Refinery modernization and alteration projects will result in new
process units being built and older units being modified. These changes will
allow refineries to process heavier and higher sulfur crude. New and
modified process units could result in increased wastewater production. New
drains will be added along with new or expanded wastewater treatment
facilities.
Modification is defined as any physical or operational change to an
existing facility which results in increased emissions. There are two
general events that would cause an increase in emissions from process drain
systems, oil-water separators, and/or air flotation systems. These events
5-3
-------
are an increase in organic loading of process wastewater and an increase in
the volume of wastewater without necessarily a change in organic loading.
Either or both of these events would be caused by the following:
1. Addition of a process unit to be serviced by the wastewater system.
2. Modification of an existing process unit already serviced by the
wastewater system.
3. Changes in product slates.
4. Changes in the type of crude oil processed.
Increased emissions from affected facilities could result in these
facilities being subject to the NSPS under the modification provisions.
Determination of modification will be made on a case by case basis.
5.2.2 Reconstruction
Expansion of existing process units and renovation of wastewater
treatment facilities could result in affected facilities being subject to the
NSPS under the reconstruction provisions. Reconstruction is determined by
the criteria given in Section 5.1.2. Determination of reconstruction will be
made on a case by case basis.
5-4
-------
6. MODEL UNITS AND REGULATORY ALTERNATIVES
The purpose of this chapter is to define model units and identify
regulatory alternatives. Model units are parametric descriptions of a
representative cross-section of the units that, in the judgment of EPA are
likely to be constructed, modified or reconstructed. The model unit
parameters are used as a basis for estimating the environmental, energy, and
economic impacts associated with the application of the regulatory
alternatives to the model units.
6.1 MODEL UNITS
Petroleum refinery wastewater systems differ considerably from site to
site. Because wastewater characteristics such as flow rate and oil content
may be unique to each refinery, various treatment schemes and techniques may
be employed by each refinery. For this reason, it is difficult to define a
model petroleum refinery wastewater system and more reasonable to define
model units for specific emission sources in petroleum refinery wastewater
systems. Section 6.1.1. discusses model units for process drains and
junction boxes. Sections 6.1.2 and 6.1.3 discuss model units for oil-water
separators and air flotation systems, respectively.
6.1.1 Process Drains and Junction Boxes
An EPA study of emissions in petroleum refineries provided information
on the population of fugitive emission sources. Included in the sources
counted were drains and pumps. Thus, drain populations as well as the
ratios of drains to pumps, were obtained for several refinery process units
of varying complexities. Further, information gathered by California Air
Resources Board has allowed estimates of junction box population, and ratio
of drains to junction boxes to be developed.2 These relationships were used
in developing model units. The number of process drains and junction boxes
6-1
-------
in a process unit was found to be dependent on the complexity of the unit
and independent of unit capacity or size. Therefore, model units are
developed on the basis of drain population.
Model units for process drains and junction boxes are presented in
Table 6-1. Refinery process units have been grouped into three model units
based on the complexity of the process unit. Model Unit A represents
process units of high complexity. It should be noted that within the high
complexity model unit category, process units can be of varying capacity.
Using information acquired in the EPA and California studies, the number of
pumps in these process units is estimated to be ten. Applying a ratio of
2.75 drains per pump, an estimate of 94 drains is derived. Further, using
the ratio of six drains per junction box, it is estimated that sixteen
junction boxes are located in these units.
The number of drains and junction boxes in Model Units B and C are
estimated using the same method. Model Unit B represents process units of
medium complexity while Model Unit C represents units of low complexity.
6.1.2 Oil-Water Separators
Model Units for oil-water separators are presented in Table 6-2. As
discussed in Chapter 3, the major factors affecting emissions are wastewater
flow rate and VOC concentration. The cost of regulatory alternatives
discussed in Section 6.2 depend on the surface area of the oil-water
separator that is open to the atmosphere. Therefore, model units are
characterized according to these three parameters.
In choosing wastewater flow rates for the oil-water separator model
units, consideration was given to crude oil production capacities at
individual refineries, flow rates observed during plant visits, and design
information from vendors. The largest flow rate (1500 gpm) is based on an
actual installation at a large refinery. If a refinery generates a larger
flow rate than 1500 gpm, it is very likely that multiple units will be
installed. The smallest flow rate (50 gpm) is based on information provided
by vendors on the smallest size oil-water separators used in petroleum
refinery applications. A mid-point flow rate (750 gpm) was chosen for the
medium sized model unit.
6-2
-------
TABLE 6-1. PROCESS DRAINS MODEL UNIT PARAMETERS
i
GO
Model
Unit
A
B
C
Representative
Process Unit Types
Crude Distillation
Fluid Catalytic Cracking
Treating Processes
Lube Oil Processing
Alkylation
Catalytic Polymerization
Isomerization
Thermal Cracking/Coking
Solvent Extraction
Hydrocracking
Hydrotreating
Hydrorefining
Light Ends/LPG
Catalytic Reforming
Vacuum Distillation
Hydrogen Manufacture
Model
Range
Small3
Average
Large
Small3
Average
K
Large
Small3
Average
h
Large
Number of sources
Unit Capacities in Model Unit
Capacity Mbpd Pumps Drains Junction . Uncontrolled
Boxes Emissions (Mg/yr)
20
47 34 94 16 30.8
113
3
17 16 44 8 14.6
36
5
28 10 28 5 9.3
67
3Average of smallest 10 percent of representative unit types.
Average of largest 10 percent of representative unit types.
Estimated using factor of 2.75 drains/pump. (Reference 1).
Estimated using factor of 6.0 drains/junction box. (Reference 2).
-------
TABLE 6-2. OIL-WATER SEPARATORS MODEL UNIT PARAMETERS
Model plant
A
B
C
Wastewater
thousand BPD
50
25
2
flow
(gpm)
(1500)
(750)
(50)
Surface3
Arqa
m
107
58
58
Uncontrolled
emissions
kg/hr
37.8
18.9
1.3
.VOC
D
Mg/yr
331.0
165.6
11.0
aRefers to the surface area of the separator that will be open to the
at^sphere Surface areas were calculated using American Petroleum
Institute (API) design specifications (Reference 3).
Calculated using Litchfield Method assuming conditions listed in Table 3-i
6-4
-------
VOC concentration levels were found to range quite widely between
refineries. As discussed in Section 3.2.2.4, a typical level of 1000
mg/liter of VOC at the inlet to the oil-water separator was chosen for
emission calculations.
Surface area is the area of the separator that is open to the
atmosphere. Surface area is dependent on the wastewater flow and was
calculated using API design specifications. However, a broad range of flow
rate conditions can be handled by a given surface area. Model Units B and
C, therefore, have the same surface areas because API design surface area of
58 m includes the 50 to 750 gpm range.
6.1.3. Air Flotation Systems
Model units for air flotation systems are presented in Table 6-3. As
in the case of oil-water separators, air flotation model units were
characterized according to wastewater flow rates and surface areas.
However, instead of calculations based on VOC concentration, uncontrolled
emission estimates are based on actual test data.
The smallest flow rate used in the model units, 50 gpm, approaches the
size of the smallest IAF system available. Conversations with vendors and
industry indicate that DAF systems also approach this size in actual
applications. ' The flow rates of 1500 gpm and 750 gpm shown in Table 6-3
are representative of a large number of actual IAF and DAF systems. Larger
flow rates than 1500 gpm are possible. However, flow rates greater than
1500 gpm would most likely be handled in multiple units to allow for
operating flexibility.
Surface areas for air flotation systems were calculated using an
empirical formula provided by a vendor. The surface areas are only
applicable to DAF systems. Most IAF systems used in refinery applications
come equipped with covers. Surface area represents the area of the DAF
system open to atmosphere. The uncontrolled emission levels for air
flotation systems are based on emissions testing conducted by EPA at three
petroleum refineries.
6-5
-------
TABLE 6-3. AIR FLOTATION MODEL UNIT PARAMETERS
Surface3 Uncontrolled VOC Uncontrolled VOC
Model Wastewater flow Arga emissions - DAF emissions - IAF
Unit thousand BPD (gpm) m
kg/hr
Mg/yr kg/hr
Mg/yr
A
B
C
50
25
2
(1500)
(750)
(50)
70.0
35.0
2.3
1.37
0.63
0.05
12.0
6.0
0.4
0.27
0.14
0.01
2.4
1.2
0.1
aRefers to the surface area of the dissolved air flotation system only.
Surface areas calculated using formula that assumes 1 square foot of
surface area is required for 2 gpm of wastewater flow (Reference 1). The
surface area is given only for a DAF since this area will determine the
cost of control. IAF systems come equipped with covers.
Uncontrolled emissions for a DAF are based on the emission factor
determined by testing. This emission factor is 15.2 kg per MM gallons Of
wastewater flow.
Uncontrolled emissions for an IAF are based on the emission factor
determined by testing. The emission factor has been modified to account
for the cover supplied with the IAF system as explained in Chapter 4,
section 4.1.3.2.
6-6
-------
6.2 REGULATORY ALTERNATIVES
This section presents regulatory alternatives for controlling VOC
emissions from process drains, oil-water separators, and air flotation
systems. These regulatory alternatives are summarized in Table 6-4.
Regulatory Alternative I
Regulatory Alternative I represents no additional control over baseline.
Baseline control is defined as the level of control currently achieved by
industry. This usually reflects the degree of control required by state and
local regulations. Regulatory Alternative I provides the basis for
determining the impacts of other regulatory alternatives.
Regulatory Alternative II
Regulatory Alternative II provides a higher level of control than
required by Regulatory Alternative I. For process drains, this alternative
requires all drains and junction boxes to be water sealed. Oil-water
separators are to be completely covered with either a fixed or floating
roof. Dissolved air flotation systems are also required to be covered with
a fixed roof. For induced air flotation systems, work practices are
required to operate the IAF under gas tight conditions. These control
techniques have been discussed in Chapter 4.
Regulatory Alternative III
Regulatory Alternative III requires the highest level of emission
reduction. For process drains, a completely closed drain system is required
with vapors vented to a control device. Under Regulatory Alternative III,
oil-water separators are also required to be completely covered with a
gasketed and sealed fixed roof with vapors to be vented to a control device.
Air flotation systems, both DAF and IAF, are also required to be completely
covered with a fixed roof with vapors vented to a control device. The
control techniques for Regulatory Alternative III have been discussed in
Chapter 4.
6-7
-------
TABLE 6-4. REGULATORY ALTERNATIVES
Regulatory
Alternative
II.
III.
en
i
oo
Process Drains
No Additional
Control
Oil-Water Separators No Additional
Control
Air Flotation Systems No Additional
Control
Water-sealed process drains
and junction boxes.
Gasketed and sealed fixed or
floating roof.
DAF systems provided with a
gasketed and sealed fixed roof,
vented to atmosphere. IAF
systems maintained gas tight
by gasketing and sealing access
doors.
Completely closed drain system
with vapors led to a control
device.
Gasketed and seal fixed roof
with vapors vented to a
control device.
Gasketed and sealed fixed roof
with vapors vented to a
control device.
-------
6.3 REFERENCES
1. Wetherold, R. G. and D. D. Rosebrook (Radian Corporation). Assessment
of Atmospheric Emissions from Petroleum Refining. Volume 1: Technical
Report. Prepared for U. S. Environmental Protection Agency. EPA
Publication No. 600/2-80-075a. April 1980.
2. Memo from Mitsch, B. F., Radian Corporation, to file. June 15, 1984.
Response to California Air Resources Board Survey of Refining Industry.
3. American Petroleum Institute. Manual on Disposal of Refinery Wastes,
Volume on Liquid Wastes. Chapter 5. Washington, D.C. 1969.
4. U.S. Filter Fluid Systems Corporation. Hydrocell Induced Air Flotation
Separator. Bulletin No. HY-1181-6M.
5. Telecon. Mitsch, B. F., Radian Corporation, with Jim Wahl, AFL
Industries. July 13, 1983. Conversation concerning sizes of DAF
systems.
6. Telecon. Mitsch, B. F., Radian Corporation, with Chuck Bassett,
Huntway Refining Company, Benicia, California. June 29, 1983.
Conversation concerning the wastewater treatment system at Huntway.
7. Komline Sanderson. Dissolved Air Flotation. Bulletin No. KSB
123-8106.
6-9
-------
7. ENVIRONMENTAL IMPACTS
7.1 INTRODUCTION
The purpose of this chapter is to present the environmental impacts of
the regulatory alternatives specified in Chapter 6. The primary emphasis is
on VOC emissions which would result from implementation of each of the
alternatives presented. The impacts of the regulatory alternatives on water
quality, solid waste, and energy are also addressed in this chapter.
7.2 AIR POLLUTION IMPACTS
Implementation of Regulatory Alternatives II and III for each of the
three emission sources will reduce VOC emissions from refinery wastewater
systems. Emission reductions achieved by implementing these alternatives
are estimated for the three emission sources in the source category. These
emission reductions are presented for individual model units on an annual
basis. Additionally, nationwide emission levels resulting from new and
modified/reconstructed process drains and junction boxes, oil-water
separators, and air flotation systems are estimated on a five-year basis.
7.2.1 Estimated Emissions and Percent Emission Reduction for Model Units
Table 7-1 lists the estimated emissions and percent emission reduction
for each model unit and regulatory alternative in the source category.
Regulatory alternatives were described in Chapter 6. Emission factors used
to estimate emissions from each model unit have been given in Chapter 3.
The control efficiencies of the various regulatory alternatives have been
described in Chapter 4. The percent reductions achievable by the regulatory
alternatives for each model unit are given in parenthesis in Table 7-1.
7-1
-------
TABLE 7-1. ESTIMATED EMISSIONS AND EMISSION REDUCTIONS FOR
EACH MODEL UNIT AND REGULATORY ALTERNATIVE
i
ro
Model Units3
Process Drains and Junction Boxes
Oil
Air
Air
A
B
C
-Water Separators
A
B
C
Flotation Systems (DAF)
A
B
C
Flotation Systems (IAF)
A
B
C
Estimated Emissions
Ib
30.8 (0)
14.6 (0)
9.3 (0)
331.0 (0)
165.6 (0)
11.0 (0)
12.0 (0)
6.0 (0)
0.4 (0)
2.36
1.18
0.07
Regulatory Alternatives
» Mg/yr (% Reduction From Reg. Alt. I)
U_
15.4 (50)
7.3 (50)
4.7 (50)
49.7 (85)
24.8 (85)
1.7 (85)
2.8 (77)
1.4 (77)
0.1 (77)
1.81 (23)
0.91 (23)
0.06 (23)
III
0.6 (98)C
0.3 (98)C
0.2 (98)C
9.9 (97)c
5.0 (97)c
0.3 (97)C
0.4 (97)C
0.2 (97)c
0.01 (97)C
0.4 (85)c
0.2 (85)c
0.01 (85)C
Model Units are described in Chapter 6.
Regulatory Alternative I represents no control.
cCaptured VOC emissions vented to an existing flare.
-------
7.2.2 Projected VOC Emissions for Petroleum Refinery Wastewater System
Source Category
Tables 7-2, 7-3, and 7-4 provide estimates of projected VOC emissions
from new and modified/reconstructed model units during the period 1985 to
1989. Table 7-2 lists projections for new and modified/reconstructed
process drain systems. Tables 7-3 and 7-4 list projections for new and
modified/reconstructed oil-water separators and air flotation systems,
respectively.
Growth projections for each emission source were presented in
Chapter 3. Over the next five years, 102 new process units are estimated to
be built with 30 new oil-water separators and 25 new air flotation systems.
Additional estimates of modified/reconstructed models units have been
determined in order to estimate projected VOC emissions from these units.
The number of modified/reconstructed process drain model units was
determined by evaluating the current construction projects at existing
petroleum refineries. It was assumed that the current construction level
would continue over the next five years and that approximately 10 percent of
the drain systems in existing units with ongoing construction projects will
be impacted by the NSPS under the modification/reconstruction provisions.
Estimates of the number of modified/reconstructed oil-water separators
and air flotation systems were determined by assuming that these units will
equal 10 percent of the new units. Therefore, it is estimated that
approximately three oil-water separators and three air flotation systems
will be impacted by the NSPS under the modification/reconstruction
provisions during the five-year period.
In Tables 7-2, 7-3, and 7-4, baseline reflects the level of control
currently required by State regulations. Baseline for the three emission
sources were presented in Section 3.4. Only oil-water separators are
currently controlled by State regulations. As a result of the State
regulations, about 85 percent of the new separators will be covered,
5 percent partially covered, and 10 percent uncovered.
7-3
-------
TABLE 7-2. PROJECTED VOC EMISSIONS FROM NEW AND MODIFIED/RECONSTRUCTED PROCESS DRAIN SYSTEMS FOR
REGULATORY ALTERNATIVES IN PERIOD FROM 1985 - 1989
Year
Number of Affected Model Units
Each Regulatory Alternative (Mg/yr)
A
1985 6
1986 12
1987 18
1988 24
1989 30
B
6
12
18
24
30
C
12
24
36
48
60
Baseline3
384
768
1152
1536
1920
II
192
384
576
768
960
III
8
15
23
31
38
Baseline reflects current level of control required by State regulations. For process drains and junction
V boxes, there is no control required by State regulations.
-------
TABLE 7-3. PROJECTED VOC EMISSIONS FROM NEW AND MODIFIED/RECONSTRUCTED OIL-WATER
SEPARATORS FOR REGULATORY ALTERNATIVES IN PERIOD FROM 1985 - 1989
tn
Year Number
A
1985 1
1986 2
1987 3
1988 4
1989 6
of Affected
B
2
4
6
8
11
Model Units
C
3
6
9
12
16
Total
Each
Baseline
527
828
926
1030
1211
Annual VOC Emissions
Regulatory Alternative
a II
104
208
312
416
597
Projected for
(Mg/yr)
III
21
42
62
83
119
aBaseline reflects the current level of control required by State regulations. The State regulations for
oil-water separators are presented in Section 3.4.
-------
TABLE 7-4. PROJECTED VOC EMISSIONS FROM NEW AND MODIFIED/RECONSTRUCTED AIR FLOTATION
SYSTEMS FOR REGULATORY ALTERNATIVES IN PERIOD FROM 1985 - 1989
Year
1985
1986
1987
1988
1989
Number of Affected Model Units
1
2
3
4
6
B
2
4
6
8
11
C
2
4
6
8
11
lotal Annual VOC Emissions Projected for
Each Regulatory Alternative (Mg/yr)
Baseline3
14.8
29.7
44.5
59.3
85.1
11
4.7
9.5
14.2
18.9
27.1
III
0.7
1.5
2.2
3.0
4.3
aBaseline reflects the current level of control required by State regulations. For air flotation systems,
there is no control required by State regulations.
-------
The projected emissions for process drain systems were estimated using
emission factors determined for drains and junction boxes and the projected
growth estimate discussed above. For oil-water separators, similar
information was used along with information regarding current State
regulations. The projected emissions reflect the current percentage of
separators estimated to be fully covered, partially covered, and uncovered.
Projected emissions from air flotation systems are based on the
emission factors and projected growth estimates. Further, as discussed in
Chapter 3, it is estimated that 50 percent of the new units will be IAF
systems and 50 percent will be DAF systems.
7.2.3 Secondary Air Pollution Impacts
Secondary air pollution impacts are those impacts generated by the
emission control techniques. Control techniques required by Regulatory
Alternative II include water seals for drains and junction boxes, covers for
oil-water separators and DAF systems, and gas-tight operation for IAF
systems. These controls would not create any secondary air pollution
impacts.
Regulatory Alternative III for all three emission sources require VOC
destruction devices. Carbon adsorption systems require steam to be used for
regeneration of the carbon beds. Fuel combustion to produce steam may
result in emissions of some air pollutants. However, the quantity of air
pollutants produced is expected to be minimal. For example, if all new
separators and air flotation systems required a designated carbon adsorber,
the amount of natural gas needed to produce steam to regenerate these units
is estimated to be 1.82 million cubic feet per year. The amount of
secondary pollutants generated by burning this amount of natural gas would
be approximately 1.1 pounds of SOX and 255 pounds of N0x.
Other VOC destruction devices such as flares, boilers, and incinerators
would produce some secondary air pollutants. The quantity of these
pollutants directly attributable to VOC control for refinery wastewater
systems would also be negligible.
7-7
-------
7.2.4 Summary of Air Pollution Impacts
Table 7-5 summarizes the air pollution impacts of the regulatory
alternatives for the source category. Implementation of Regulatory
Alternative II for all emissions sources and Regulatory Alternative III for
process drains and junction boxes and oil-water separators would result in
positive air pollution impacts. The percent reduction from baseline and
incremental emission reduction are also shown in the table.
7.3 WATER POLLUTION IMPACTS
Implementation of any of the regulatory alternatives would not have an
adverse impact on water quality. The control techniques proposed would not
interfere with the basic water treatment functions of oil-water separators
and air flotation systems. Further, as explained below, suppression of VOC
in the wastewater by covering separators and air flotation systems will not
result in a significant increase in organic loading to subsequent treatment
processes.
Data collected in an EPA study2 showed that VOC have a greater affinity
for the oil phase of wastewater than for the water phase. The concentration
of VOC in the oil phase was about one thousand times that in the water
phase. To the extent that control techniques suppress emissions of VOC,
these VOC will mostly be captured in the oil and removed to recovery
processes. Suppression into the oil phase would not be expected to be as
great if the vapor space of a separator or air flotation system is purged
(as required by Regulatory Alternative III for separators and air
flotation). However, when the vapor space is purged, the VOC removed would
be directed to a control device. Again, no adverse impact on water quality
would occur.
7.4 SOLID WASTE IMPACTS
There will not be a significant amount of solid waste produced as a
result of implementing the regulatory alternatives. The only possible
source of solid waste will be from carbon adsorption systems. If activated
carbon is disposed rather than regenerated, small quantities of solid waste
will be produced.
7-8
-------
TABLE 7-5. SUMMARY OF ANNUAL EMISSIONS AND EMISSION REDUCTION BY 1989 FOR SOURCE
CATEGORY (NEW AND MODIFIED/RECONSTRUCTED UNITS)
Emission Source Regulatory Alternative
Process Drains and I
Junction Boxes
II
III
^j
i> Oil -Water Separators I
II
III
Air Flotation Systems I
II
III
Annual
Emissions by 1989
(Mg/yr)
1920
960
38
1211
597
119
84
27
4
% Reduction From
Baseline
-
50
98
-
54
91
-
69
95
-------
7.5 ENERGY IMPACTS AND WATER USAGE
Implementation of Regulatory Alternative II for all three emission
sources would not require high usage of water or energy. Implementation of
Regulatory Alternative III for these sources and Regulatory Alternative II
for air flotation systems would result in consumption of small quantities of
steam, water, electricity and fuel gas. As explained in Chapter 6, these
alternatives require that VOC be captured and vented to a control device.
In some cases, refineries will have existing control devices accessible to
these emission sources. Only blowers would be required to transport the VOC
to the existing control device. Electricity would be required to power the
blowers. If designated control devices are needed, utilities would be
required to operate the control device. In the case of carbon absorbers,
water, steam, and electricity would be needed.
Table 7-6 is a summary of utility requirements which would result from
implementing Regulatory Alternative III for process drain systems, oil-water
separators, and air flotation systems.
7.6 OTHER ENVIRONMENTAL CONCERNS
Implementation of the regulatory alternatives is not expected to result
in a large commitment of energy or other non-renewable resources. As
discussed above, implementation of the regulatory alternatives would not
impact water quality or solid waste generation. However, a delay in the
regulatory action would adversely affect air quality at the rate shown in
Table 7-5.
7-10
-------
TABLE 7-6. ENERGY REQUIREMENTS AND WATER DEMAND - REGULATORY ALTERNATIVE III FOR PROCESS
DRAINS AND JUNCTION BOXES, OIL-WATER SEPARATORS, AND REGULATORY
ALTERNATIVE II FOR AIR FLOTATION SYSTEMS.
Emission Source # Affected Units by
1989
Process Drains
Oil -Water Separators
Oil -Water Separators0
Air Flotation Systems
Air Flotation Systems
120
33
33
28
28
Fuel Gasa Electricity
(MM scf/yr) (kWh/yr)
13 352,350
161,730
330,000
137,230
280,000
Water Steam
(nr/yr) (Mg/yr)
- -
-
12,400 354
-
10,528 300
Fuel gas assumed to be used to purge closed drain system.
Assumes existing control device available. Electricity requirements for blowers to transport VOC to
control device. Cost sharing possible between separators and air flotation systems but has not been
considered in this analysis.
•^
"Electricity, steam, water, needed for blower, carbon adsorption system. Cost sharing possible between
separators and air flotation systems but has not been considered in the analysis.
-------
7.7 REFERENCES
1. U.S. Environmental Protection Agency. Compilation of Air Pollutant
Emission Factors. Third Edition. AP-42, Supplement 13. August 1982.
p. 1.4-1.
2. Wetherold, R. G., L. P. Provost, and C. D. Smith. (Radian
Corporation.) Assessment of Atmospheric Emissions from Petroleum
Refining. Volume 3: Appendix B. Prepared for U.S. Environmental
Protection Agency. Publication No. EPA 600/2-80-075C. April 1980.
7-12
-------
8. COSTS
This chapter presents the methods used to estimate costs for
controlling volatile organic compounds (VOC) from petroleum refinery
wastewater systems. Cost estimates are given for each regulatory
alternative and model unit described in Chapter 6. In Chapter 9, the
results of this cost analysis are used to determine the economic impact of
the regulatory alternatives.
8.1 COST ANALYSIS OF REGULATORY ALTERNATIVES
The costs of major equipment (covers for oil-water separators and air
flotation systems) needed for the regulatory alternatives were acquired from
actual installations in the refining industry. The costs of additional
equipment such as piping, blowers, and vapor control devices were estimated
using engineering references.1'2'3'4'5 Standard costing procedures devised
by Uhl1'2 were then used to estimate capital and annual costs for each model
unit and regulatory alternative. Tables 8-1 and 8-2 present the cost
algorithms used in the analysis. All costs were updated to third quarter
P
1983 dollars using Chemical Engineering Plant Cost Indices.
Section 8.1.1 presents the costs associated with implementing the
regulatory alternatives for process drains and junction boxes. Sections
8.1.2 and 8.1.3 present the costs associated with implementing the
regulatory alternatives for oil-water separators and air flotation systems,
respectively. For all three emission sources, costs for both new and
retrofitted control systems are discussed.
8.1.1 Process Drains and Junction Boxes
Regulatory alternatives for process drains and junction boxes have been
discussed in Section 6.2. Regulatory Alternative I requires no additional
control and, therefore, does not result in any costs. The costs for
implementing Regulatory Alternatives II and III are discussed below.
8-1
-------
TABLE 8-1. COMPONENTS AND FACTORS OF TOTAL CAPITAL INVESTMENT3
Direct Costs
Purchased equipment costs
Installation costs includes:
Piping
Structural Steel
Concrete
Electrical
Instrumentation
Other (paint, insulation, etc.)
Installation labor
Total Direct Capital Cost (TDC)
Indirect Cost
Engineering and supervision
Miscellaneous field expenses
10% of TDC
5% of TDC
Cumulative Subtotal A
Contractors' fees
Contingencies
10% of subtotal A
15% of subtotal A
Cumulative Subtotal B
Interest during construction
Startup
12% of subtotal B
5% of subtotal B
Total Depreciable Investment (TDI)
Subtotal B + interest +
startup
References 1 and 2.
8-2
-------
TABLE 8-2. COMPONENTS, FACTORS, AND RATE OF TOTAL ANNUAL COST3
Basis: 24 hour/day, 365 d/yr.
Direct Annual Operation and Maintenance Expenses (O&M)
Labor - Operating
- Maintenance
- Supervisory
- Other
Materials - Operating
- Maintenance
Fuel gas
Electricity
Other (list as required)
hr/yr x $14.00/hrc
2.5% of TDC
10% of O&M labor
-0-
-0-
2.5% of TDC
annual usage x $3.50/1000 scfc
annual usage x $.05/kWhc
Total Direct O&M (DOM)
Sum of the above
Indirect Annual O&M Expenses
Overhead
General and administration
Insurance and Property Taxes
70% of all labor
2% of TDI
2% of TDI
Total Indirect O&M (IOM)
Sum of the above
Total Annual O&M Expenses (TAOE)
Capital Recovery (CR) (Capital
recovery factor for 10% over
10 years x TDI)
Total Annual Cost
DOM + IOM
0.163 x TDI
TAOE + CR
^References 1 and 2.
Reference 6.
cReference 7.
8-3
-------
8.1.1.1 Regulatory Alternative II - Water Sealed Drains and Junction
Boxes.
New Process Drains and Junction Boxes
A P-trap water sealed drain was used as the basis for estimating the
costs for Regulatory Alternative II. A P-trap drain has been illustrated in
Figure 3-7. The materials needed to construct uncontrolled, P-trap, and
closed drains are given in Table 8-3. The materials needed for these drain
types were derived from actual installations and from engineering judgement.
The cost associated with implementing Regulatory Alternative II is the
additional cost of a P-trap drain over an uncontrolled drain. The difference
in total depreciable investment (TDI) between an uncontrolled drain and a
P-trap is approximately 172 dollars. The difference in cost is due primarily
to additional materials and labor needed for the P-trap. Therefore,
172 dollars represents the cost per drain of implementing Regulatory
Alternative II.
A water seal pot with a water line was used as the VOC reduction
technique for junction boxes. The water seal pot has been illustrated in
Figure 3-9. The materials used to construct a water seal pot and the
associated costs are given in Table 8-3. Using these cost estimates and the
costing algorithms given in Table 8-1, total cost for controlling VOC from
junction boxes was estimated to be $362 dollars per junction box.
The costs for implementing Regulatory Alternative II for new process
drain model units are shown in Table 8-4. These costs were derived by
applying the costs of P-traps drains and controlled junction boxes to the
number of drains and junction boxes in each model unit. Additionally, the
cost effectiveness of controlling VOC emissions from each model unit is
provided in the table. Cost effectiveness estimates for Regulatory
Alternative II are approximately $350 per Mg.
Retrofit Process Drains and Junction Boxes
The cost for retrofitting an existing process unit with P-trap drains
and controlled junction boxes was also estimated. The additional cost
required to retrofit a P-trap drain over installing a new P-trap drain is
8-4
-------
TABLE 8-3. TOTAL DIRECT CAPITAL COST OF MAJOR EQUIPMENT FOR
VOC CONTROL ON PROCESS DRAINS0
Uncontrolled Drain System
1. Straight Pipe (4" diameter, 4.25 ft)
2. Wye (cast iron, no hub)
Water Sealed Drain Systems
P-Trap Drain
1. Straight Pipe (4" diameter, 4.25 ft)
2. Wye (cast iron, no hub)
3. P-trap (4" cast iron, 1/4 bend-3)
4. El bend (4" cast iron)
Water Seal Pot on Junction Box
1. Straight Pipe (4" diameter, 1 ft)
2. 1/4 bend (4" cast iron)
3. Cup (6" welded)
4. Water refill line (20 ft 1/2 steel pipe)
5. Globe value (bronze)
6. 1/4 bends (2) (1/2" steel)
7. Tee (1/2" cast iron)
Total Installed6
Cost ($)
20
58
Total
Total
78
Total
20
58
77
25
180
5
25
65
28
30
20
42
215
Closed Drain System
Closed Drain
1. Straight Pipe (4" diameter, 4.25 ft)
2. Wye (cast iron, no hub)
3. Flange (4" carbon steel #150)
4. Union (3/4" carbon steel)
Underground Tank and Purge Gas System
1. Fabricated tankc
2. Purge Gas System0
Total
20
58
113
13
204
$44,298.00
$ 2,585.00
aCost includes materials and labor, 3rd quarter 1983 dollars.
Reference 3.
Breakdown of materials given in Table 8-6.
8-5
-------
TABLE 8-4. ANNUALIZED COST AND COST EFFECTIVENESS OF REGULATORY ALTERNATIVES
FOR NEW PROCESS DRAIN AND JUNCTION BOX SYSTEM
oo
i
Regulatory
Alternative
I
II
III
Model
Unit
A
B
C
A
B
C
Af
B^
Cc
Drains
94
44
28
94
44
28
94
44
28
Junction Boxes
16
8
5
16
8
5
16
&
5
Total
Depreciable
Investment
($1,000)
NO CONTROL
22.00
10.50
6.60
150.00
90.60
63.40
Annual Cost ($1000)
Direct
Expense
COSTS
0.65
0.31
0.19
11.31
8.93
8.00
Indirect
Expense
1.11
0.53
0.34
11.70
8.60
7.40
Capital
Recovery
3.58
1.71
1.08
24.61
14.77
10.81
Total
Annual
Cost
($1,000)
5.34
2.54
1.61
47.62
32.30
26.16
Emission
Reduction
(Mg/yr)
15.4
7.3
4.6
30.2
14.3
9.1
Cost
Effectiveness
($/Mg)
350
350
350
1580
2260
2880
a.
b.
c.
Regulatory Alternative I - No action
Regulatory Alternative II - Require P-traps on all drains and seal pots on junction boxes.
Regulatory Alternative III - Require a sealed drain system vented to a control device.
Costs are based on the factors and computational algorithms of Table 8.1 and 8.2.
in 3rd quarter 1983 dollars.
All costs are
The capital cost of an underground collection tank was calculated assuming 42 drains. Costs for
other size drain systems were estimated by the following aquation (Reference 9):
Cost = (Cost of tank for a 42 drain system) # of drains
Total depreciable investment for piping equal for all systems.
-------
the cost of materials as well as labor and equipment necessary to remove the
existing drains. Costs were based on a three man crew using a backhoe with
a pneumatic jackhammer to remove concrete around the drain. Using
engineering judgement, it was estimated that each drain would take one-half
hour to excavate. Table 8-5 presents the costs for retrofitting water
sealed drains in each model unit. The cost is $486 per drain. The cost of
retrofitting a junction box with a water seal is considered minimal because
no excavation is necessary.
It is expected that most units which would be affected by the
modification/reconstruction provisions would be down for reasons other than
drain retrofitting. Therefore, no cost due to production losses would
result from implementing the NSPS.
8.1.1.2 Regulatory Alternative III - Closed Drain System.
New Process Drains and Junction Boxes
A completely closed drain system similar to that installed at one
refinery was used as the basis for the cost evaluation. The closed drain
system uses sealed drains and an underground collection tank. The
collection tank is purged with fuel gas to reduce the risk of explosions.
The purge gas is then vented to an existing control device, such as a flare.
The closed drain system has been illustrated in Figure 3-8.
The materials needed to install closed drains are given in Table 8-3.
As with P-trap drains, the difference in cost between installing a closed
drain and an uncontrolled drain is used for all cost calculations. The
difference in cost is approximately $210 per drain.
The materials and methods used to estimate the cost of constructing the
underground collection tank and purge system are shown in Table 8-6. The
tank was sized to handle wastewater from a process unit having 42 drains.
The annual cost for operating the underground tank and purge system includes
the electricity to operate the sump pump and fuel gas for the purge system.
The costs for these utility requirements are shown in Table 8-7. The cost
effectiveness for implementing Regulatory Alternative III for each model
unit is also shown in Table 8-4. The cost effectiveness estimates range
from $1580 per Mg for Model Unit A to $2880 per Mg for Model Unit C.
8-7
-------
TABLE 8-5. ANNUALIZED COST AND COST EFFECTIVENESS OF REGULATORY ALTERNATIVES FOR
RETROFITTING A PROCESS DRAIN AND JUNCTION BOX EMISSION REDUCTION SYSTEM
oo
00
Regulatory
Alternative
I
11
III
Model
Unit
A
B
C
A
B
C
Ac
Br
Cc
Drains
94
44
28
94
44
28
94
44
28
Junction Boxes
16
8
5
16
8
5
16
8
5
Total
Depreciable
Investment
($1,000)
NO CONTROL
51.5
24.3
15.4
182.6
105.4
75.8
i_
Annual Cost ($1000)°
Direct
Expense
COSTS
1.61
0.76
0.48
12.29
9.40
8.29
Indirect
Expense
2.65
1.25
0.79
13.33
9.36
7.83
Capital
Recovery
8.39
3.96
2.51
29.76
17.18
12.35
Total
Annual
Cost
($1,000)
12.65
5.97
3.78
55.38
35.94
28.47
Emission
Reduction
(Mg/yr)
15.4
7.3
4.6
30.2
14.3
9.1
Cost
Effectiveness
($/Mg)
820
820
820
1,830
2,510
3,130
a. Regulatory Alternative I - No action
Regulatory Alternative II - Require P-traps on all drains
Regulatory Alternative III - Require a sealed drain system vented to a control device.
b. Cost assume 1.5 manhour to remove each old drain. Costs are based on the factors and
computational algorithms of Table 8.1 and 8.2. All costs are 1n 3rd quarter 1983 Dollars
c. The capital cost of an underground collection tank was calculated assuming 42 drains.
Costs for other size drain system were estimated by the following equation (Reference 9): '
Cost = (Cost of Tank for a 42 drain system) I of drains '
Total depreciable investment for piping equal for all systems.
-------
TABLE 8-6. BASIS FOR BURIED TANK SUBSYSTEM COST ESTIMATE
FOR REGULATORY ALTERNATIVE III
Direct capital cost based on vessel estimate using methods of
Richardson .
Vessel specifications: 7 feet, i.d., 10.75 feet tangent-to-tangent
length, ellipsoidal head, 5/16 inch thick carbon steel, welds spot
checked. Vessel volume is approximately 400 ft (3000 gal). In
practice an externally coated steel is likely to be used and costs of
such coating are implicitly assumed to be within the overall estimate
contingency allowance.
Vessel buried in excavation 11 feet deep by 14.75 feet long by 11 feet
wide. Vessel rests directly on sand or gravel within excavation, and
backfilled with original overburden.
Vessel contains two manways: 36" diameter and 24" diameter extending
to ground surface. First manway is welded to exterior wall of vessel
to provide access from above ground to piping nozzles attached to
vessel wall. Second manway penetrates wall of vessel to provide access
to vessel interior. Manways are covered with a bolt-on cover.
Two sump pumps each rated at 40 gpm, 25 psig discharge pressure, and
requiring 1 hp motors are used to pump vessel liquid to wastewater
treatment. Motors are located on ground level cover of 36" manway.
Piping and shafts extend through manway, and then through nozzles in
vessel wall.
Piping from plant fuel gas system to tank, installed. Piping between
tank and facility flare system, installed.
4
Installation costs were estimated based on factors in Guthrie for
horizontal process vessels and pumps.
Vessel capacity is directly proportional to the number of drains in the
system. Therefore, the number of drains was used as the sizing factor.
Total Direct Capital Cost of Tank: $44,298, Total Direct Capital Cost
of piping: $2585
8-9
-------
TABLE 8-7. ANNUAL UTILITY COSTS FOR REGULATORY ALTERNATIVES
oo
Process Regulatory
Alternative
Process Drain System - New and III
Retrofit
011 -Water Separator-New and III
Retrofit
CPI System III
DAF System III
IAF System III
Model
Unit
N"
B?
CK
*b
BK
cb
£
BP
CK
AD"
B£
Cb
Ar
B'
ch
A£
BK
cb
Ar
Br
Cb
Ab
BK
cj
Ar
Br
cc
Mater
_
_
-
.
.
0.010
0.010
0.010
_
_
0.010
0.010
0.010
-
_
0.010
0.010
0.010
-
_
—
0.010
0.010
0.010
Utility
Steam
—
-
-
-
.
.
0.574
0.574
0.574
-
_
_
0.574
0.574
0.574
-
_
_
0.574
0.574
0.574
-
_
_
0.574
0.574
0.574
Cost ($1000)
Electricity
0.087
0.136
0.278
0.245
0.245
0.245
0.500
0.500
0.500
0.245
0.245
0.245
0.500
0.500
0.500
0.245
0.245
0.245
0.500
0.500
0.500
0.245
0.245
0.245
0.500
0.500
0.500
Fuel Gas
0.217
0.342
0.696
-
-
-
-
-
-
-
-
_
-
-
.
-
-
_
_
_
-
-
_
—
_
_
-
The electrical requirements are based on a pumping rate of one-half the pumps design capacity for 2,920 hours per year. The fuel
gas usage Is based on a complete turn over of the col leeton tank's vapor space every 24 hours, based on a tank sized for 42
drains. The utility costs were also adjusted for the different tank sizes using the following equation:
( D)
Utility Cost - U42 (47)
Where: U.? * Utility cost for a tank serving 42 drains
D = Number of drains In Model Unit.
Captured VOC emissions vented to an existing control device.
cCaptured VOC emissions vented to a dedicated device (carbon adsorber).
-------
Retrofit Process Drains and Junction Boxes
The cost for retrofitting an existing process unit with a closed drain
system was also estimated. The additional cost of retrofitting a closed
drain system over installing a new drain system is the labor and equipment
needed to excavate the existing uncontrolled drains and weld on the
necessary piping. Additional materials are also needed which add to the
cost of a closed drain system. Costs were based on a three man crew using a
backhoe with a pneumatic jackhammer to remove concrete around the drain.
Field welding was also necessary to attach the piping to the existing drain.
It was estimated that each drain would take one-half hour to excavate and 7
o
manhours to prepare and weld the necessary piping. The cost would be $546
per drain. The cost for installing an underground tank is the same as that
given in Table 8-6. Utility requirements for the purge system are shown in
Table 8-7.
It is expected that most units which would be affected by the
modification/reconstruction provisions would be down for reasons other than
drain retrofitting. Therefore, no costs due to production losses would
result from implementing the NSPS.
Table 8-5 presents the costs of retrofitting closed drain system for
each model unit. Additionally, cost effectiveness estimates for
implementing Regulatory Alternative III for each model unit are given. Cost
effectiveness values range from $1830 per Mg for Model Unit A to $3130 per
Mg for Model Unit C.
8.1.2 Oil-Water Separators
Regulatory Alternatives for oil-water separators have been discussed in
Section 6.2. Regulatory Alternative I requires no additional control and
therefore does not result in any costs. The costs for implementing
Regulatory Alternatives II and III are discussed below.
The costs of covers for separators were provided by industry and
represent retrofit costs. The costs for providing a cover on a newly
installed separator were derived from the retrofit costs. For this reason,
retrofit costs are presented first.
8-11
-------
8.1.2.1 Regulatory Alternative II - Covered Separators. Information
was provided by the refining industry regarding costs of actual installations
of fixed and floating roofs on existing oil-water separators. These costs
ranged from Sll/ft2 to $45/ft2 for fixed roofs and from $46/ft2 to $93/ft
12
for floating roofs. The wide range in costs is due to differences in
material of construction, size of the roof, type of roof, and problems
encountered during installation. To account for all of these factors, an
average cost for installing a fixed or floating roof was developed using all
available information. The average cost for installing a fixed or floating
2
roof on an existing oil-water separator is $56/ft . The total depreciable
investment for Regulatory Alternative II was calculated by applying this
unit cost to the size roof required by each model unit. Annual costs were
then derived using the cost algorithms given in Table 8-2. Table 8-8
presents these costs as well as cost effectiveness estimates for each model
unit.
A roof which is part of a newly installed oil-water separator would be
expected to cost less than a roof retrofitted on an existing separator. A
detailed cost breakdown of a retrofitted roof was provided by one refinery.
It was determined that 33 percent of the costs for retrofitting would not
have been required for a roof on a newly installed separator. This figure
is supported by standard engineering estimations that consider retrofit
construction to be 25 to 40 percent higher than new construction . Applying
this reasoning, it was estimated that the total cost assignable to a roof on
2
a new separator would be $37/ft .
Table 8-9 presents the costs for Regulatory Alternative II for new
oil-water separators. Cost effectiveness estimates for each model unit are
also presented. These estimates range from $40 per Mg for Model Unit A to
$610 per Mg for Model Unit C.
8.1.2.2 Regulatory Alternative III - Covered Separators with Vapor
Control Systems. Two situations have been considered in estimating costs
for Regulatory Alternative III. It is expected that an existing control
device will be accessible to the separator. Therefore, costs have been
8-12
-------
TABLE 8-8. ANNUALIZED COST AND COST EFFECTIVENESS OF REGULATORY ALTERNATIVES FOR
A RETROFIT CONTROL SYSTEM ON AN API OIL-WATER SEPARATOR
00
I
I—>
CO
Regulatory
Alternative
I
II
III
Model
Unit
A
B
C
A
B
C
*r
Br
cc
AH
BH
Cd
Total
Depreciablg
Investment
($1,000)
NO CONTROL
64.50
34.90
34.90
70.50
40.50
40.50
134.70
105.10
105.10
Annual Cost
Direct
Expense
COST
2.01
1.09
1.09
10.87
9.95
9.95
13.94
12.92
12.92
Indirect
Expense
3.31
1.80
1.80
9.52
8.01
8.01
12.81
11.31
11.31
i
($1000)°
Capital
Recovery
10.51
5.70
5.70
11.49
6.78
6.78
21.96
17.15
17.15
Total
Annual
Cost
($1,000)
15.83
8.59
8.59
31.88
24.74
24.74
48.56
41.38
41.38
Emission
Reduction
(Mg/yr)
281.3
140.8
9.3
321.1
160.6
10.7
311.4
155.7
10.3
Cost
Effectiveness
($/Mg)
60
60
920
100
150
2,310
160
270
4020
a. Regulatory Alternative I - No action
Regulatory Alternative II - Require all oil-water separators to be covered with a fixed or
floating roof.
Regulatory Alternative III - As alternative II plus a vapor collection and control system
b. Costs based on the factors and computational algorithms of Table 8-1 and Table 8-2.
All costs are 3rd quarter 1983 dollars.
c. VOC emissions vented to an existing control device.
d. VOC emissions vented to a dedicated control device (carbon adsorber system).
-------
TABLE 8-9. ANNUALIZED COST AND COST EFFECTIVENESS OF REGULATORY
ALTERNATIVES FOR NEW API OIL-WATER SEPARATORS
00
I
Regulatory Model
Alternative Unit
I A
B
C
II A
B
C
in A<;
Bc
Cc
Ad
Bd
Ca
Total
Depreciable
Investment
($1,000)
42.6
23.1
23.1
48.6
29.1
29.1
112.8
93.3
93.3
Annual
Direct
Expense
NO
1.
0.
0.
10.
9.
9.
13.
12.
12.
Cost
Indirect
Expense
CONTROL
32
72
72
19
58
58
16
55
55
2
1
1
8
7
7
11
10
10
COSTS
.20
.19
.19
.41
.40
.40
.71
.70
.70
L
($1000)°
Capital
Recovery
6.94
3.76
3.76
7.92
4.74
4.74
18.39
15.21
15.21
Total
Annual
Cost
($1,000)
10.
5.
5.
26.
21.
21.
43.
38.
38.
47
67
67
52
72
72
26
46
46
Emission
Reduction
(Mg/yr)
281
140
9
321
160
10
311
155
10
.3
.8
.3
.1
.6
.7
.4
.7
.3
Cost
Effectiveness
($/Mg)
40
40
610
80
140
2,030
140
250
3,730
a. Regulatory Alternative I - No action
Regulatory Alternative II - Require all oil-water separators to be covered with a fixed or floating roof
Regulatory Alternative III - As alternative II plus a vapor collection and control system
b. Total Depreciable Investment costs assumed to be 66% of the retrofit total depreciable investment cost.
Costs are based on the factors and computational algoroithms of Table 8-1 and Table 8-2. All costs are 3rd
quarter 1983 dollars.
c. VOC emissions vented to an existing control device.
d. VOC emissions vented to a dedicated control device (carbon adsorber system).
-------
calculated for this situation. However, cases may be found where an accessi-
ble control device is not available. For this reason, costs have also been
calculated to include the cost of a dedicated control device. In the cost
calculation, the dedicated control device is assumed to be a carbon adsorber.
The equipment needed to vent the captured VOC to an existing control
device and the associated costs are given in Table 8-10. The materials and
installation costs associated with a control system using a carbon adsorber
are presented in Table 8-11. These costs are based on the design and
operating parameters also given in the table. Utility requirements for
these systems and associated costs have been shown in Table 8-7.
The costs for implementing Regulatory Alternative III for oil-water
separators are presented in Tables 8-8 and 8-9. Table 8-8 presents the
costs for separators retrofitted with covers. Table 8-9 presents costs for
covers installed on new separators.
8.1.3 Air Flotation Systems
Three regulatory alternatives for air flotation systems have been
discussed in Section 6.2. Regulatory Alternative I requires no additional
control and therefore results in no costs. Regulatory Alternative II for
DAF systems requires the flotation chamber to be covered with a fixed roof.
For IAF systems, this alternative requires the system to be operated
gas-tight. Regulatory Alternative III requires the flotation chamber of
both types of systems to be tightly covered with captured VOC vented to a
control device.
For purposes of the cost analysis, DAF and IAF systems are considered
separately. IAF system are constructed with covers and, therefore, do not
incur the cost for adding a cover. DAF systems have open flotation tanks
and must have a cover installed. For this reason, control costs for DAF
systems are higher than IAF systems.
The major equipment costs for controlling VOC from air flotation
systems are listed in Table 8-10. The cost for a fiberglass roof was^ ^
acquired from information provided by industry and equipment vendors. •
The unit cost for installing a roof on a DAF system is $20/ft . This cost
can be applied to both new and retrofitted units due to the minimal
modifications which would be required for a retrofitted roof.
8-15
-------
TABLE 8-10. COST BREAKDOWN OF MAJOR EQUIPMENT FOR VOC CONTROL FOR
OIL-WATER SEPARATORS AND AIR FLOTATION SYSTEMS
Unit Cost ($/ft*
Oil-Water Separators
1. Cover - New (Fixed or Floating) 37
2. Cover - Retrofit (Fixed or Floating) 56
Dissolved Air Flotation Systems
1. Cover - Fiberglass fixed 20
Induced Air Flotation Systems Unit Cost ($)
1. Pressure/Vacuum Valve 290
2. Latches 100
Fittings for Vapor Collection System .
(Oil-Water Separators and Air Flotation) Total Installed Cost * ($)
1. Carbon Steel pipe (200'x 2" 40 std) 725
2. Tees (4) (2" carbon steel 40 std) 278
3. Flame arrester (2" aluminum) 370
4. Flanges (2" carbon steel) 62
5. Blower and Motor (3/4 Hp) 2130
^Reference 3.
3rd quarter 1983 dollars.
8-16
-------
TABLE 8-11. OPERATING PARAMETERS AND COSTS OF CARBON ADSORBER3
1. Operating Parameters
a) VOC concentration = 1000 ppm
b) Operating capacity = 7 lb/1000 Ib carbon
c) VOC content = 0.25 Ib VOC/1000 scf
d) Carbon requirement = 0.5 Ib carbon/1000 scf
e) Flow rate of gas = 300 scfm
f) Temperature = 100°F
g) Gas velocity = 100 fpm
h) Bed depth = 3 ft.
i) Pressure drop =?6.5 in. H?0/ft. of carbon
j) Bed area = 3 ft
k) Carbon = 270 Ibs
i) Steam = 0.3 Ibs/lb carbon (93% efficiency)
= 23652 Ibs/yr
2. Costs
a) Total Depreciable Investment $70,213.00
b) Annual Cost
- carbon replacement $ 126.36
- steam $ 573.56
- electricity $ 500.15
- cooling water $ 9-90
- labor (0.5 mhr/shift) $ 7,665.00
aReference 5.
8-17
-------
IAF systems can be made gas tight by gasketing the access doors which
serve to cover the system. For Regulatory Alternative II, costs are added
for the pressure/vacuum valve, latches, and gasketing. Additional costs for
the piping and blower are included for Regulatory Alternative III.
Two situations have been considered in estimating costs for Regulatory
Alternative III. As with oil-water separators, it is expected that an
existing control device may be accessible to the air flotation system.
However, some cases may exist where a dedicated control device is needed.
Therefore, costs have been calculated for both situations. Again, the
dedicated control device is assumed to be a carbon adsorber.
Tables of 8-12 and 8-13 present the annual costs and cost effectiveness
estimates for DAF and IAF systems, respectively. Costs for utility
requirements for the control system are shown in Table 8-7.
8.1.4 Incremental Cost Effectiveness
The incremental cost effectiveness between Regulatory Alternative II
and III was calculated for new and retrofit process drain systems, new and
retrofit oil-water separators and both types of air flotation system. The
results of these calculations are presented in Table 8-14.
8.2 OTHER COST CONSIDERATIONS
Environmental, safety, and health statutes that may cause an
expenditure of funds by the petroleum refining industry are listed in
Table 8-15. Specific costs to the industry to comply with the provisions,
requirements, and regulations of the statutes are unavailable. However,
some references are listed which provide cost estimates for complying with
specific regulations.15'16'17
Few refineries are expected to close solely due to the cost of
compliance with the total regulatory burden. The costs incurred by the
petroleum refining industry to comply with all health, safety, and
environmental regulations are not expected to prevent compliance with the
proposed NSPS for refinery wastewater systems.
8-18
-------
TABLE 8-12. ANNUALIZED COST AND COST EFFECTIVENESS OF REGULATORY ALTERNATIVES FOR DAF SYSTEMS
Regulatory Model
Alternative Unit
I A
B
C
II A
B
C
III Ac
Bc
Cc
Ad
Ra
Bd
Cd
Total
Depreciable
Investment
($1,000)
NO CONTROL
15.0
7.5
0.5
21.1
13.5
6.5
85.3
77.8
70.7
Annual Cost
Direct
Expense
COSTS
0.47
0.24
0.02
9.43
9.20
8.98
12.30
12.07
11.85
Indirect
Expense
0.77
0.39
0.03
6.98
6.60
6.24
10.29
9.90
9.54
L
($1000)D
Capital
Recovery
2.44
1.22
0.08
3.45
2.21
1.06
13.89
12.67
11.52
Total
Annual
Cost
($1,000)
3.69
1.85
0.12
19.86
18.01
16.28
36.48
34.64
32.91
hmission
Reduction
(Mg/yr)
9.2
4.6
0.3
11.6
5.8
0.39
11.3
5.6
0.38
Cost
Effectiveness
($/Mg).
400
400
400
1,710
3,110
41,740
3,230
6,190
86,600
C.
d.
Regulatory Alternative I - No action
Regulatory Alternative II - Requires a fixed cover
Regulatory Alternative III - Requires a fixed cover and vapor collection and control system on all
DAF systems
Costs are based on the factors and computational algorithms of Table 8-1 and Table 8-2.
All costs are in 3rd quarter 1983 dollars
VOC emissions vented to an existing control device
VOC emissions vented to a dedicated control devices (carbon adsorber system).
-------
TABLE 8-13. ANNUALIZED COST AND COST EFFECTIVENESS OF REGULATORY ALTERNATIVES FOR IAF SYSTEMS*
IN)
O
Regulatory h Model
Alternative0 Unit
I A
6
C
II A
B
C
III A<|
Bj
Cd
A*;
B!
Ce
Total
Depreciable
Investment
($1,000)
NO CONTROL
0.4
0.4
0.4
6.0
6.0
6.0
70.2
70.2
70.2
Annual Cost ($1000)c
Direct Indirect Capital
Expense Expense Recovery
0.01
0.01
0.01
8.96
8.96
8.96
11.83
11.83
11.83
0.02
0.02
0.02
6.21
6.21
6.21
9.51
9.51
9.51
0.06
0.06
0.06
0.98
0.98
0.98
11.44
11.44
11.44
Total
Annual Emission
Cost Reduction
($1,000) (Mg/yr)
0.10
0.10
0.10
16.15
16.15
16.15
32.78
32.78
32.78
0.55
0.27
0.02
1.96
0.98
0.06
1.66
0.83
0.05
Cost
Effectiveness
($/Mg)
180
370
5560
8,240
16,480
269,170
19,750
39,350
655,600
a. Cost for vapor control device only, system assumed to be covered.
b. Regulatory Alternative I - No action
Regulatory Alternative II - Gas tight system
Regulatory Alternative III - Vapor collection and control system
c. Costs are based on the factors and computational algorithms of Table 8-1 and 8-2.
All costs are 3rd quarter 1983 dollars.
d. VOC emissions vented to an existing control device.
e. VOC emissions vented to a dedicated control device (carbon adsorber system).
-------
TABLE 8-14. INCREMENTAL COST EFFECTIVENESS OF REGULATORY ALTERNATIVES
Co
ro
Model
Process Unit
Drain System - New
Drain System - Retrofit
Oil -Water Separator - New
Oil -Water Separator-Retrofit
Dissolved Air Flotation
Induced A1r Flotation
a. Regulatory Alternative II
b. Regulatory Alternative II
(carbon adsorber).
A
B
C
A
B
C
A;
£
Ab
k
cb
A!
B5
ca
$
cb
Aa
?:
Sb
lb
Aa
?
Ab
Sb
Cb
: Cover;
; Cover;
Regulatory Alternative II Regulatory Alternative III
Annual Cost Emission Reduction Annual Cost Emission Reduction
($1.000) (Mg/yr) ($1,000) (Mg/yr)
5.34
2.54
1.61
12.65
5.97
3.78
10.47
5.67
5.67
10.47
5.67
5.67
15.83
8.59
8.59
15.83
8.59
8.59
3.7
1.8
0.1
3.7
1.8
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Regulatory
Regulatory
15.4
7.3
4.6
15.4
7.3
4.6
281.3
140.8
9.3
281.3
140.8
9.3
281.3
140.8
9.3
281.3
140.8
9.3
9.2
4.6
0.3
9.2
4.6
0.3
0.55
0.27
0.01
0.55
0.27
0.01
Alternative III:
Alternative III:
47.62
32.30
26.21
55.38
35.94
28.47
26.52
21.72
21.72
43.26
38.46
38.46
31.88
24.74
24.74
48.56
41.38
41.38
19.9
18.0
16.3
36.5
34.6
32.9
16.2
16.2
16.2
32.8
32.8
32.8
Captured VOC emissions
Captured VOC emissions
30.2
14.3
9.1
30.2
14.3
9.1
321.1
160.6
10.7
311.4
155.7
10.3
321.1
160.6
10.7
311.4
155.7
10.3
11.6
5.8
0.4
11.3
5.6
0.4
1.96
0.98
0.06
1.66
0.83
0.06
vented to an existing
vented to a dedicated
Incremental
Cost ($/Mg)
2,860
4.250
5.470
2,890
4.280
5.490
400
810
11,460
1,090
2,200
32,790
400
810
11,460
1,090
2,200
32,790
6,750
13,500
162.000
15.620
32.800
328.000
11,420
22,680
. 322.000
29,460
58,390
654.000
control device.
control device
-------
Table 8-15 STATUTES TIIAT KAY BE APPLICABLE TO THE
PETROLEUN REFINING INDUSTRY
StatHta
Aaftlcafcla aravlalaa. rtynUtlM ar
raajalraM*)t af ttawta
Statute
Appllcabla arwlalaii. ragalatlaii ar
raqulrMMftt 0f atatata
CltM Air Act mi Aiin*iinli
oo
f\3
ro
CUM Ntlar Act (Ft4tral
Matar NttntlM Act)
AtM«rca
Act
ta«lc SubtUnca* Canlral
Act
NatlMMt aaltila*
air
0 llala ka»l0MMt0tla« plain
Ihr
fafltlva Mlitlaa*
0 NM aaarca aarfaraaaca tUarfaNt
Air 0aMatla«
Valatlla arfanlc llapM atarata
0 fSt canitractlai aaraltt
0 Haa-attalaMHt camtractlaii panalta
0 tlacltarfa faraltt
• Ittiklltliet ijrtt«a ta tract
Mattai
a Ittab!Ithat
raportlnft
•onllarlnt
racor4taa«lM.
. laaalliM airf
« ivstM far
a PrtMiMfactara Mtlflcatlan
a Itbtllnf, racordkaealaf
a Mporllnj rcqulramntt
• Iwlclty tattliif
nM
laMy A Naaltfc
a Haw aaarca parfanMnca ataa4ar4i
0 CaNtrat af 011 a«4ll> an4 tflacMrfaa Caaatal tana
a rratraatMat raajalrawmta
a ranaltllMf af Industrial arajacta
that taalajfa aa vat I and* ar
public Malar*
a twIroMMNtal l«aact atataNaati
0 Paralti far traatMNt. atarafa, an4
Act
National CwlraaMa)tal Nllcy
Act
Safa aYlnklna Malar Act
Narlaa SaMtitary Act
• MalklHf-mrkliif Mr faca ttaiataNi
• Naam af afrait itaiMfaNl
• 0cc«ratlanal kaaltk ana* an*lraa>-
Matal cantral lUMaNt
0 NaiaNam awtarlal itaMaNt
0 fartanal yratactlva aajHlfnant
•tamlaNa
a fiaaaral anvlranMaUl cantral
0 IMIcal aM flrit aM
0 Hra fratactla* ataa4ar4a
0 Co*pra*aa4 fat a
air aqulaaaat
0 NalilM. krailnf. 0*4 cattily
•taniaNt
0 Stataa awy vato laJaral aanalta
far flaatt U ba «IM la
caaital
a Raa^lra* aw»lranaa»tal
ttataMftta
a Maqalrat aadararavM lajactlaai
caatral paralt*
0 Ocaaa a\Mplnf paralta
a Aacar4kaa|»liif aa4 rapartlnf
-------
8.3 REFERENCES
1. Uhl, V.W. A Standard Procedure for Cost Analysis of Pollution Control
Operations. Volume 1: User Guide. Research Triangle Park, North
Carolina. Publication No. EPA 600/8-79-018a.
2. Uhl, V.W. A Standard Procedure for Cost Analysis of Pollution Control
Operations. Volume II: Appendices. Research Triangle Park, North
Carolina. Publication No. EPA 600/8-79-018b.
3. Richardson Engineering Services, Inc. The Richardson Rapid
Construction Cost Estimating System. 1982-1983 edition. Richardson
Engineering Services, Inc., San Marcos, Ca.
4. Guthrie, K.M. Process Plant Estimating Evaluation and Control.
Craftsman Book Company of America, Solana Beach, Ca., 1974.
5. U.S. Environmental Protection Agency. Organic Chemical Manufacturing
Volume 5: Adsorption, Condensation, and Absorption Devices. Report 1.
U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina. Publication No. EPA 450/3-80-027. December 1980.
6. U.S. Bureau of Labor Statistics. National Employment, Hours and
Earnings, Average Hourly Earnings of Production Workers: Petroleum
Refining. Dialog Data Base File #178. March 1983.
7. Energy Information Administration. Monthly Energy Review. Washington
D.C. DOE/EIA-0035(83/09). September 1983.
8. C.E. Plant Cost Index. Chemical Engineering. 90(20):7.
October 3, 1983.
9. Perry, R. H. and C. H. Chilton. Chemical Engineers' Handbook. Fifth
edition. New York, McGraw-Hill Book Company. 1973. p. 25-18.
10. Trip Report. A. H. Laube and R. G. Wetherold, Radian Corporation, to
R J McDonald EPA: Chemicals and Petroleum Branch, Research Triangle
Park, N.C., July 19, 1983. Report of March 25, 1983 visit to Sun Oil
Refinery in Toledo, Ohio.
11 Memo from G. Hunt, Radian Corporation, to file. January 4, 1984. Cost
of Installing a Roof During Construction of a new Oil-Water Separator.
12 Memo from G. Hunt, Radian Corporation, to file. January 4, 1984. Cost
of Retrofitting a Roof in an Existing Oil-Water Separator.
8-23
-------
13. Telecon. Mitsch, B. F.t Radian Corporation, with Jim Monroe, EIMCO
Envirotech. Salt Lake City, UT. December 8, 1983. Conversation
regarding cost to install DAF system at Chevron, El Segundo,
California.
14. Telecon. Mitsch, B. F., Radian Corporation, with Jim Strong, Heil
Process Equipment. Avon, Ohio. July 13, 1983. Conversation regarding
cost of covers for air flotation system.
15. U.S. Environmental Protection Agency. VOC Fugitive Emissions in
Petroleum Refinery Industry - Background for Proposed Standards.
Research Triangle Park, NC. Publication No. EPA-450/3-81-015a.
November 1982.
16. U.S. Environmental Protection Agency. Development Document for
Effluent Limitations Guidelines and Standards for the Petroleum.
Refining Point Source Category. Washington, D.C. Publication No.
EPA-440/1-82-014. October 1982.
17. U.S. Environmental Protection Agency. Sulfur Oxides Emissions from
Fluid Catalytic Cracking Unit Regenerators - Background Information for
Proposed Standard. Research Triangle Park, NC. April 1982.
8-24
-------
9.0 ECONOMIC IMPACT
9.1 INDUSTRY CHARACTERIZATION
9.1.1 General Profile
9.1.1.1 Refinery Capacity. On January 1, 1984, there were 220 petro-
leum refineries operating in the United States with a total crude capa-
city of 2,653,000m3 per stream day.1 With respect to location, refining
capacity is fairly well-concentrated, with 57 percent of domestic crude
throughput capacity located in three states: Texas (28%), California (15%),
and Louisiana (14%).
Although refining capacity grew steadily through the 1970s, a similar
trend in capacity growth has not continued into the 1980s, as noted by Table
9-1. The decrease in the rate of capacity expansion can be traced to reduced
consumption resulting from rising prices, the slowdown of economic growth,
the availability of substitutes in some applications, and the increasing fuel
efficiency of newer automobiles and industrial facilities. Those additions
to capacity that have been made in the recent past and which will be made in
the future will occur at existing refineries to allow the processing of
lower-quality high-sulfur crudes, and increase the output of unleaded gaso-
line.^
While the number of refineries operating has declined dramatically
in recent years (i.e. 1981 to 1984) the average capacity of existing refin-
eries has increased. These trends indicate that many of the closing refin-
eries are of relatively small capacity. Small refinery closures have been
due largely to the elimination of Federal subsidies, such as the "small
refiner bias" built into the Department of Energy's crude oil entitlements
program. This program, as well as all Federal price controls on domestic
crude oil and refined petroleum products, was eliminated in 1981 through
Executive Order 12287.
9-1
-------
Table 9-1. TOTAL AND AVERAGE CRUDE DISTILLATION CAPACITY BY YEARa
UNITED STATES REFINERIES, 1973-1983
Year
(January 1)
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
Number of
Refineries
247
259
256
266
285
289
297
303
273
225
220
Total Capacity
(m3/sd)b»c
2,365,000
2,459,000
2,494,000
2,689,000
2,801,000
2,870,000
2,975,000
3,080,000
2,957,000
2,704,000
2,653,000
Average Refinery
Capacity
(m3/sd)D
9,600
9,500
9,700
10,100
9,800
9,900
10,000
10,200
10,800
12,000
12,000
aReferences 1 and 3 through 12.
bNote: Capacity in stream days.
cl m3 = 6.29 barrels.
9-2
-------
It should be noted that in the production and capacity tables that
follow, a distinction is often made between stream days (i.e., sd) and
calendar days (i.e., cd). The basic difference between the two terms is that
"stream day" refers to the maximum capacity of a refinery or unit on a given
operating day, while "calendar day" production represents the average daily
production over a one-year period. Since most refineries do not operate 365
days each year, stream day numbers are always slightly larger than those for
calendar days.
9.1.1.2 Refinery Production. In terms of total national output,
the percentage yields of most refined petroleum products have remained
constant over recent years, although several exceptions are noted below.
The percentage yields of refined petroleum products from crude oil for the
years 1974 through 1981 are summarized in Table 9-2, while Table 9-3 lists
the average daily output of the major products.
The diversity of refinery product output varies with refinery capacity.
Large integrated refineries operate a wide variety of processing units,
thus enabling the production of many or all of the products noted in Table
9-2. Other refineries are relatively small, have only a few processing units,
and produce selected products such as distillate oil and asphalt.
9.1.1.3 Refinery Ownership, Vertical Integration and Diversification.
A large portion of domestic refining capacity is owned and operated by
large, vertically integrated oil companies, both domestic and international.
The remainder is controlled by independent refiners such as Ashland, Charter,
Crown Central Petroleum, Holly, Quaker State, Tesoro Petroleum and Tosco.
Table 9-4 lists twenty companies with the greatest capacity to process
crude oil . Based upon the capacities noted, and a total domestic capacity
of 2,704,000 m3 per stream day, the 4- and 8-firm concentration ratios
are 27 and 47 percent, respectively. These ratios indicate a relatively high
degree of ownership concentration of refinery capacity.
Refinery ownership is but one aspect of the vertical integration of
the major oil companies. Such companies are integrated "backward" in that
they own or lease crude oil production facilities, both domestic and inter-
national, as well as the means to transport crude by way of pipeline and
tankers. On the international level, access to Saudi Arabian crude is
maintained through Aramco which is owned by four international companies:
Exxon, Standard Oil of California, Texaco, and Mobil.
9-3
-------
Table 9-2. PERCENT VOLUME YIELDS OF PETROLEUM PRODUCTS BY YEAR*
UNITED STATES REFINERIES, 1974-1981
(Percent)
Product
Motor Gasoline
Jet Fuel
Ethane
Liquefied Gases
Kerosene
Distillate Fuel Oil
Residual Fuel Oil
Petrochem. Feedstocks
Special Naphthas
Lubricants
Wax
Coke
Asphalt
Road Oil
Still Gas
Miscellaneous
Processing Gainb
Total c
1974
45.9
6.8
0.1
2.6
1.3
21.8
8.7
3.0
0.8
1.6
0.2
2.8
3.7
0.2
3.9
0.5
3.9
103.9
1975
46.
7.
0.
2.
1.
21.
9.
2.
0.
1.
0.
2.
3.
0.
3.
0.
3.
103.
5
0
1
4
2
3
9
7
6
2
1
8
2
1
9
7
7
7
1976
45.5
6.8
0.1
2.4
1.1
21.8
10.3
3.3
0.7
1.3
0.1
2.6
2.8
0.0
3.7
1.0
3.5
103.5
1977
43.4
6.6
0.1
2.3
1.2
22.4
12.0
3.6
0.6
1.2
0.1
2.5
2.9
0.1
3.6
1.0
3.6
103.6
1978
44.1
6.6
0.1
2.3
1.2
22.4
12.0
3.6
0.6
1.2
0.1
2.5
2.9
0.1
3.6
1.0
3.6
103.6
1979
43
6
0
2
1
21
11
4
0
1
0
2
3
3
0
3
103
.0
.9
.1
.3
.3
.5
.5
.7
.6
.3
.1
.6
.1
—
.8
.8
.6
.6
1980
44
7
2
1
19
11
5
0
1
0
2
2
0
4
0
4
104
.5
.4
—
.4
.0
.7
.7
.1
.7
.3
.1
.7
.9
.1
.0
.8
.4
.4
1981
44.8
7.6
0.1
2.4
0.9
20.5
10.4
4.7
0.6
1.3
0.1
3.1
2.7
—
4.3
0.7
4.2
104.2
aReference 13. Section VIII, Tables 4-4a.
bProcessing Gain = Product Yield - Process Feed (Input)
cTotals exceed 100 percent because product yields are greater than process
feeds by an amount equal to the processing gain. In the catalytic reforming
process, for example, straight-chain hydrocarbons are converted to branched
configurations with hydrogen as a by-product, resulting in an overall net
increase in volume.
9-4
-------
Table 9-3. PRODUCTION OF PETROLEUM PRODUCTS BY
UNITED STATES REFINERIES, 1972-1981
(1,000 m3/cd)c
Year
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
Motor
Gasoline
1,000
1,039
1,011
1,037
1,088
1,118
1,140
1,132
1,083
1,019
Distillate
Fuel Oil
419
449
424
422
465
521
501
503
440
416
Residual
Fuel Oil
127
154
170
197
219
279
266
270
262
209
Jet Fuel
135
137
133
138
146
155
155
161
159
154
Kerosene
35
35
25
24
24
27
24
29
22
19
NGL and LRQd
57
60
54
49
54
56
N.A.
54
N.A.
N.A.
aReference 13. Section VII. Tables 5, 6, 6a, 7, 7a, 14, 15, 16, 16a,
17, and 17a.
^Total and product output reports may vary slightly by data source.
clm3 = 6.29 barrels.
dNGL = Natural Gas Liquids; LRG = Liquefied Refinery Gases.
9-5
-------
Table 9-4. NUMBER AND CAPACITY OF REFINERIES OWNED AND OPERATED
BY MAJOR COMPANIES3 »b
UNITED STATES REFINERIES, 1983
Company
Chevron
Exxon
Shell
Amoco
Texaco
Gulf
Mobil
ARCO
Marathon
Union Oil
Sohio/BP
Conoco
Ashland
Sun
Cities Service
Phillips
Champl in
Getty
Tosco
Koch
Number of
Refineries
12
5
7
7
9
5
6
5
4
4
3
5
5
3
1
3
3
3
3
2
Crude Capacity
(1,000 m3/cd)
212
191
176
161
149
140
135
113
93
78
72
61
59
57
51
47
46
45
41
38
aReference 14.
bRecent mergers have combined Chevron with Gulf, and
Texaco with Getty.
9-6
-------
With regard to transportation by pipeline, the major oil companies have
been the main source of capital for the construction and operation of these
facilities, due largely to the huge investments required. On the other hand,
tanker ownership is split among the major oil companies and independent oper-
ators who charter tankers to oil companies and traders.15 The presence of
independent tanker operators is a result of the relatively small financial
requirements, compared to pipeline ownership. However, the profitability of
such operations has declined along with the volume of crude refined.
While many of the low-volume refinery products are marketed directly by
the refiners themselves, the sale of gasoline on the retail level is handled
primarily by franchised dealers and independent operators. The major refiners
do, however, have a high degree of control over the distribution of their
products with regard to market area. This is so because the major refiners
select sites for the construction of service stations before the facilities
are leased to independent operators under franchise agreements. The major
refiners do maintain the direct operation of some service stations for
purpose of measuring the strength of the retail market. However, no more
than 5 percent of all facilities in operation are managed in this fashion.16
Many of the firms that operate refineries, notably the larger oil compa-
nies, are diversified as well as vertically integrated. Several refiners are
vertically integrated through the manufacture of petrochemicals and resins.
Among the firms that have interests in these areas are Getty Oil, Occidental
Petroleum, and Phillips Petroleum. Ashland Oil's construction division
operates the nation's largest highway paving company.
Several instances of diversification can be observed. Exxon Enter-
prises develops and manufactures various high-technology products. The
Kerr-McGee Corporation is the largest supplier of commercial grade uranium
for electricity generation and also manufactures agricultural and industrial
chemicals. Mobil Oil Corp. is owned by Mobil Corp. which owns both Montgom-
ery Ward and Co. and The Container Corporation of America. The Charter Co.,
the largest of the independent refiners, is also engaged in broadcasting,
insurance, publishing, and commercial printing.
9.1.1.4 Refinery Employment and Wages. Total employment at domestic
petroleum refineries has grown steadily since the mid-1960's, with minor dis-
ruptions during periods of economic contractions. As Table 9-5 demonstrates,
9-7
-------
Table 9-5. EMPLOYMENT IN PETROLEUM AND NATURAL GAS EXTRACTION
AND PETROLEUM REFINING BY YEAR*
UNITED STATES, 1972-1981
(1,000 Workers)
Year
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
Petroleum and
Natural Gas Extraction
268.2
277.7
304.5
335.7
360.3
404.5
417.1
476.3
547.4
657.2
Petroleum
Refining
152.3
149.9
155.4
154.2
157.1
160.3
163.0
168.5
154.2
169.6
aReference 13. Section V. Table 2,
9-8
-------
there were 170 thousand workers employed at refineries in 1981.17 With 303
refineries operating that year,H average employment at each refinery is
approximately 560 persons.
The average hourly earnings of petroleum refinery workers have consis-
tently exceeded average wage rates for both the mining and manufacturing
industries.18 Petroleum refinery hourly earnings have also exceeded those
for other sectors of the oil industry as noted in Table 9-6.
9.1.2 Refining Processes
Refineries process crude oil through a series of physical and chemical
processes into many individual products. The four major product areas are as
follows:
o Transportation fuels -- motor gasoline, aviation fuel;
o Residential/commercial fuels --middle distillates;
o Industrial/utility fuels -- residual fuel oils; and
o Other products -- liquified gases and chemical process feeds.
As noted in Table 9-2, motor gasoline is by far the largest volume product of
U.S. refineries. Motor gasoline is produced through blending the products of
various refinery units such as those described below. Estimated 1981 gasoline
pool composition is presented in Table 9-7.19
9.1.2.1 Crude Distillation. The initial step in refining crude oil is
to physically separate the oil into distinct components or fractions through
distillation at atmospheric pressure. There are several possible combina-
tions of fractions and quantities available from crude distillation dependent
upon the type of crude being processed and the products desired.20 High
boiling point components are often further separated by vacuum flashing or
vacuum distillation. The crude oil still provides feedstock for downstream
processing and some final products.21
9.1.2.2 Thermal Operations. Thermal cracking operations include regu-
lar coking as well as visbreaking. In each of these operations, heavy oil
fractions are broken down into lighter fractions by the action of heat and
pressure while heavy fuels and coke are produced from the uncracked residue.22
Visbreaking is a mild form of thermal cracking that causes very little reduc-
tion in boiling point but significantly lowers the viscosity of the feed.
The furnace effluent is quenched with light gas oil and flashed in the bottom
of a fractionator while gas, gasoline, and heavier fractions are recycled.
9-9
-------
Table 9-6. AVERAGE HOURLY EARNINGS OF SELECTED INDUSTRIES BY YEAR*
UNITED STATES, 1972-19813
($/Hour)b
Year
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
Petroleum
Refining
5.25
5.54
5.96
6.90
7.75
8.44
9.32
10.08
10.94
12.17
Petroleum and
Natural Gas Extraction
4.00
4.29
4.82
5.34
5.76
6.23
7.01
7.73
8.55
9.49
Total
Manufacturing
3.81
4.08
4.41
4.81
5.19
5.63
6.17
6.69
7.27
7.98
Total
Mining
4.41
4.73
5.21
5.90
6.42
6.88
7.67
8.48
9.18
10.06
^Reference 13. Section V.
bCurrent dollars.
Table 2.
9-10
-------
Table 9-7. ESTIMATED GASOLINE POOL COMPOSITION BY REFINERY STREAM*
UNITED STATES REFINERIES, 1981
Stream
Reformate
FCC Gasoline
Alkylate
Raffinate
Butanes
Coker Gasoline
Natural Gasoline
Light Hydrocrackate
I some rate
Straight Run Naphtha
Total
Amount
(m3/cd)
355,000
408,000
162,000
17,000
75,000
15,000
30,000
22,000
16,000
86,000
1,186,000
% Of
Total
29.9
34.4
13.7
1.4
6.3
1.3
2.5
1.9
1.3
7.3
100.0
Reference 19.
9-11
-------
Coking is a severe form of thermal cracking in which the feed is held
at a high cracking temperature long enough for coke to form and settle out.
The cracked products are separated and drawn off and heavier materials are
recycled to the coking operations.20
9.1.2.3 Catalytic Cracking. Catalytic cracking is used to increase the
yield and quality of gasoline blending stocks and produce furnace oils and
other useful middle distillates.22 By this process the large hydrocarbon
molecules of the heavy distillate feedstocks are selectively fractured into
smaller olefinic molecules. The use of a catalyst permits operations at lower
temperatures and pressures than those required in thermal cracking. In the
fluidized catalytic cracking processes, a finely-powdered catalyst is handled
as a fluid as opposed to the beaded or pelleted catalysts employed in fixed
and moving bed processes.20
9.1.2.4 Reforming. Reforming is a molecular rearrangement process to
convert low-octane feedstocks to high octane gasoline blending stocks or to
produce aromatics for petrochemical uses.20 Hydrogen is a significant
co-product of reforming, and is in turn, the major source of hydrogen for
processes such as hydrotreating and isomerization.
9.1.2.5 Isomerizaton. Isomerization, like reforming, is a molecular
rearrangement process used to obtain higher octane blending stocks. In this
process, light gasoline materials (primarily butane, pentane, and hexane),
are converted to their higher octane isomers.
9.1.2.6 Alky!ation. Alkylation involves the reaction of an isoparaffin
(usually isobutane) and an olefin (propylene or butylenes) in the presence of
a catalyst to produce a high octane alkylate, an important gasoline blending
stock.20*22
9.1.2.7 Hydrotreating. Hydrotreating is used to saturate olefins and
improve hydrocarbon streams by removing unwanted materials such as nitrogen,
sulfur, and metals. The process uses a selected catalyst in a hydrogen
environment.20 Hydrofining and hydrodesulfurization are two subprocesses
used primarily for the removal of sulfur from feedstock and finished pro-
ducts. Sulfur removal is typically referred to as "sweetening".
9.1.2.8 Lubes. In addition to or in place of drying and sweetening of
hydrotreating units, petroleum fractions in the lubricating oil range are
further processed through solvent, acid, or clay treatment in the production
9-12
-------
of motor oils and other lubricants. These subprocesses can be used to finish
waxes and for other functions.20
9.1.2.9 Hydrogen Manufacture. The manufacture of hydrogen has become
increasingly necessary to maintain growing hydrotreating operations. Natural
gas and by-products from reforming and other processes may serve as charge
stocks. The gases are purified of sulfur (a catalyst poison) and processed
to yield moderate to high purity hydrogen. A small amount of hydrocarbon
impurity is usually not detrimental to processes where hydrogen will be
used.20
9.1.2.10 Solvent Extraction. Solvent extraction processes separate
petroleum fractions or remove impurities through the use of differential
solubilities in particular solvents. Desalting is an example whereby water
is used to wash water soluble salts from crude.21 Several complex refining
processes employ solvent extraction during the production of benzene-related
compounds.
9.1.2.11 Asphalt. Asphalt is a residual product of crude distillation.
It is also generated from deasphalting and solvent decarbonizing -- two spe-
cialized steps that increase the quantity of cracking feedstock.21
9.1.3 Market Factors
9.1.3.1 Demand Determinants. Most projections of refined product
demand conclude that in terms of total refinery output, existing capacity is
capable of satisfying demand over the foreseeable future.23*24 However,
expansions and modifications will be undertaken at existing refineries in
order to allow the processing of greater proportions of high-sulfur crudes,
and to permit the production of increasing levels of high-octane unleaded
gasoline. It is also possible that shifts in demand on the regional level
may allow the construction of a few new small refineries, and several of
these projects are currently known to be planned or under construction.
In Table 9-8 DOE estimates of daily demand levels for the four major
refinery products are presented under several assumptions regarding the world
price of oil. Reduced driving and greater vehicle efficiency have combined
to reduce the future demand for motor gasoline. As Table 9-8 indicates, it
is unlikely that gasoline demand will, within the forecast period, reach
those levels observed during 1983. This conclusion holds true for all
assumptions regarding the future of world oil prices with the exception of
the low price scenario for 1985.
9-13
-------
Table 9-8. REFINED PRODUCT DEMAND PROJECTIONS FOR U.S.
REFINERIES UNDER THREE WORLD OIL PRICE SCENARIOS3
1983-1986-1989
World Crude
Oil Price b»c
Year
1983
1986
Low
Mid
High
1989
Low
Mid
High
$/BBL.
30.00
21.00
28.00
38.00
26.00
36.00
45.00
S/m^
188.70
132.09
176.12
239.02
163.54
226.44
283.05
Demand (1,000 m3/cd)
Motor Distillate Residual
Gasoline Oil Oil
988.7
1,015.7
941.7
869.7
883.8
814.3
764.9
425.8
609.3
539.0
482.4
625.5
534.3
485.2
209.1
422.0
388.8
329.2
425.9
361.0
276.1
Jet
Fuel
160.5
184.6
180.0
173.4
196.9
189.5
183.6
Total d
2,320.79
2,880.53
2,657.94
2,419.68
2,796.68
2,514.29
2,287.66
Reference 23, pp. 68, 103, 138.
bReference 23, p. 17.
C1982 dollars.
dTotal includes the four products listed plus all other refined products.
9-14
-------
Reduced total gasoline demand does not, however, imply that existing
gasoline production facilities are currently capable of meeting future
gasoline requirements. In particular, the continued phase-out of leaded
gasoline and demand for higher octane ratings will require some additions
to refinery capacity. Consequently, refiners can be expected to increase
cracking, catalytic reforming, and alkylation capacities in order to main-
tain octane requirements.25
Distillate fuel oils are used in home heating, utility and industrial
boilers, and as diesel fuel. Unlike the other three major petroleum product
categories noted in Table 9-8, demand for distillate fuel oil is projected to
increase under all price scenarios. The expected increase can be traced to
two major factors namely, the growing popularity of diesel-powered automobiles
and light trucks and the phased deregulation of natural gas prices. The
shift from gasoline toward diesel fuel, along with a projected increase in
vehicle miles traveled by heavy diesel-powered trucks, accounts for the
expected increase in distillate fuel demand in the transportation sector. In
the residential sector it is expected that the continued deregulation of
natural gas prices will reduce the price advantage previously held by natural
gas in space heating applications.
Residual fuel oil is used as a bunker fuel in large ships, large utility
and industrial boilers, and in the heating of some buildings. Residual fuel
oil competes with coal for use as a fuel in the applications noted above.
Table 9-8 shows that the most recent recession depressed residual fuel demand
in 1982, and that little growth in demand is expected in the near future.
This lack of growth is attributable to the increasing ability of refiners
to crack residual fuel into more valuable lighter products as well as a
general decline in demand from industrial and utility consumers. Among the
factors that are adversely affecting the demand for residual fuel oil are: a
slowdown in the generation of electricity and conversions to coal and
nuclear energy by major utilities, and increased fuel efficiency and closing
of obsolete plants in the industrial sector.26
Finally, the demand for some products not shown on Table 9-8 remains
promising for the foreseeable future. Such products include solvents, lubes,
and petrochemical feedstocks.27
The elasticity of demand is a measure of the relative change in
quantity demanded of a product, in response to a relative change in its
9-15
-------
price. With regard to the elasticity of demand for various petroleum pro-
ducts, most analysts agree that in the short-term, quantity demanded is not
very sensitive to price changes due to the inability of consumers to easily
shift to other technologies. However, as the focus shifts to the longer
term, the elasticity of demand increases as consumers have increased ability
to shift to other fuels or more fuel-efficient products. DOE estimates of
longer term (i.e. to 1990) demand elasticities are summarized in Table 9-9.
9.1.3.2 Supply Determinants. As noted in the previous section, it
is unlikely that the supply of refined petroleum products will be restricted
for reason of inadequate domestic refining capacity. It is, however, possible
that disruptions in the flow of imported oil could result from international
developments, in particular, political instability in the Middle East.
Attempts to reduce dependence upon imported oil have focused upon
four major areas: reduced consumption through conservation, increased
domestic production through the decontrol of domestic oil prices, domestic
stockpiling of imported oil, and the development of a synthetic fuels indus-
try. While price decontrol and synthetic fuels development may have a
significant impact in terms of import reductions, these measures are essen-
tially mid- to long-term solutions. Conservation, on the other hand, has
offered more immediate results.
The effects of higher prices and recent conservation efforts, in-
cluding decreased gasoline consumption, and conversion of facilities to coal
and natural gas, can be observed in Table 9-10. In particular, imports of
crude oil have declined significantly after reaching a historic high of 384
million m^ in 1977, and the reduction of imports has continued into the
1980's. Domestic consumption has also fallen considerably since the peak
levels observed during 1978. However, it should be noted that some portion
of the decline in both imports and domestic consumption may be attributed to
the recession of 1981-82.
Price controls on domestic crude oil and refined petroleum products
were revoked by Executive Order 12287 (January 28, 1981). This Order essen-
tially rescinded the price and allocation authority granted to the Department
of Energy under the Emergency Petroleum Allocation Act of 1973. The progres-
sive decontrol of domestic crude oil prices has been accompanied by increased
exploration, and is expected to increase stocks of already proven reserves.
9-16
-------
Table 9-9. PRICE ELASTICITY ESTIMATES FOR MAJOR REFINERY PRODUCTS
BY DEMAND SECTOR3
UNITED STATES, 1990
Demand Sector
Residential
Commerical
Industrial
Transportation
Refinery Product
Distillate Oil
Distillate Oil
Distillate Oil
Residual Oil
Gasoline
Distillate Oil
Residual Oil
Jet Fuel
Price Elasticity^
-0.46
-0.45
-0.64
-0.45
-0.45
-0.89
-0.09
-0.52
Reference 28. p. 333.
bPercent change in quantity demanded in response to a one percent
increase in price.
9-17
-------
Table 9-10. CRUDE OIL PRODUCTION AND CONSUMPTION BY YEARa
UNITED STATES, 1970-1982
(1,000,000 m3/year)D
Domestic
Year Productionc
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
559
549
549
534
486
465
452
457
485
474
500
497
503d
Importsc
77
98
129
188
202
238
308
384
369
376
303
240
20 1^
Domestic
Consumption6 Exports6
633
649
680
723
688
703
760
841
854
850
802
753
703
0.8
0.1
0.1
0.1
0.2
0.3
0.5
2.9
9.2
13.6
16.7
13.2
13.7
Year -End
Stocks6
44
41
39
39
42
43
45
55
60
68
27
34
37
Stocks as Percent
of Consumption
6.94
6.36
5.76
5.33
6.13
6.14
5.97
6.57
7.01
8.05
3.37
4.51
5.26
aReference 2. p. 073 (1970-1979 data).
blm3 = 6.29 barrels.
cReference 13. (1980-1981 data).
dReference 29. Table 2.
6Reference 29. Table 22, (1980-1982 data).
9-18
-------
Finally, evidence of changing supply conditions in the industry can
be seen in the fate of the synthetic fuels industry. The development of
such an industry was a priority during the energy short years of the mid to
late 1970's. However, the incentive to develop technologies capable of
converting oil shale, gas, and coal to liquid fuels has been reduced due
largely to abundant oil supplies, reduced Federal funds, and high interest
rates. Consequently, it is not expected that the availablity of synthetic
fuels will affect the oil supply situation in this decade.
9.1.3.3 Prices. Table 9-11 indicates historic wholesale price levels
for gasoline, distillate fuel oil, and residual fuel oil. For each product,
a pattern of stable prices, followed by rapid price increases in 1974 and
1979 through 1981, can be observed. The increases observed during both
periods can be attributed to the pass-through of increases in the price of
crude oil supplied by the OPEC nations.
Future prices of refined products will continue to rise in response to
increases in the price of both imported and domestic crude. The Department
of Energy expects that average worldwide crude oil prices should increase at
an annual rate of about 3.1 percent up to 1989 (see Table 9-19).
9.1.3.4 Imports. Imports of both crude oil and refined products are
expected to continue to decline through the 1980's. In the case of crude
oil, the fall in import levels can be attributed to increases in the price of
OPEC oil, and the increased production of domestic crude prompted by its
price decontrol.
Low sulfur (sweet) crudes are generally more desirable than high sulfur
(sour) crudes because the refining of the latter requires a larger investment
in desulfurization capacity to meet process as well as environmental needs.
While more than half of the current crude imports are sweet, only 15 percent
of OPEC's total oil reserve is sweet crude.30 Consequently, it is most
likely that future imports will contain higher proportions of sour crudes and
thus make sour crude processing a more profitable investment for many refineries.
With regard to refined petroleum products, the importation of most
of these products is expected to decline as it has since the mid-1970's.
Table 9-12 shows that for the major refined products, imports peaked during
1973-1974. In general, imports of refined products have been relatively
small compared with production at domestic refineries. One notable exception
is residual fuel oil. The relatively high ratio of imports to domestic
9-19
-------
Table 9-11. AVERAGE WHOLESALE PRICES: GASOLINE, DISTILLATE FUEL
OIL, AND RESIDUAL FUEL OIL BY YEAR*
UNITED STATES, 1968-1982
(<71iter)
Year
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
Gasolineb»c
4.4
4.5
4.7
4.8
4.7
5.2
8.1
9.5
10.3
11.2
11.8
16.4
24.0
26.9
24.7
Distillate Fuel Oil&,c
2.7
2.7
2.9
3.1
3.1
3.6
5.6
8.2
8.7
9.8
9.9
14.3
21.3
26.0
24.4
Residual Fuel Oilb»c
1.5
1.5
1.9
2.6
3.0
3.4
6.8
6.8
6.6
7.9
7.4
10.2
14.6
18.2
16.7
aCurrent dollars
bReference 12, P. 079 (1968-1979)
CReference 29. Table 42 (1980-1982)
9-20
-------
Table 9-12.
IMPORTS OF SELECTED PETROLEUM PRODUCTS BY YEAR3
UNITED STATES, 1969-1981
(1,000 m3/cd)
Year
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
Motor .
Gasoline
10
11
9
11
21
32
29
21
34
31
29
22
24
Distillate
Fuel Oil
22
24
24
29
62
46
25
23
40
27
31
22
27
Residual
Fuel Oil
201
243
252
277
295
252
194
225
216
214
182
146
127
Jet Fuel
20
23
29
31
34
26
21
12
12
14
14
13
6
Kerosene
0.5
0.6
0.2
0.2
0.3
0.8
0.5
1.4
3.0
1.7
1.4
1.5
1.1
NGL and LRG
6
8
17
28
38
34
29
31
32
22
37
NA
NA
^Reference 13. Section VII, Table 5, 6, 6a, 7, 7a, 14, 15, 16, 16a, 17, 17a
NA - not available.
9-21
-------
production of this product is attributed to the orientation of U.S. refiner-
ies toward the production of higher levels of more valuable lighter products,
such as motor gasoline, through the "cracking" of residual oil. The importa-
tion of greater amounts of residual oil is therefore required to satisfy the
requirements of utilities and large industrial boilers in this country.
9.1.3.5 Exports. Exports of crude oil and refined petroleum products
are a small portion of total U.S. production, and amount to less than eight
percent of the volume imported.31 All exports are controlled by a strict
licensing policy administered by the U.S. Department of Commerce. Recently,
crude oil exports have increased in response to the Canada-United States
Crude Oil Exchange Program. The program is mutually beneficial in that
acquisition costs are minimized through improved efficiency of transporta-
tion.
Table 9-13 summarizes recent trends in major refined product exports.
The decline in exports through the 1970s can be attributed to both increased
domestic demand and the expansion of foreign refining capacity.
9.1.4 Financial Profile
The financial status of the oil industry is generally regarded as
strong, although recent supply/demand imbalances have affected profitability.
Recent below average performance has been attributed to a number of factors
including, reduced demand due to conservation, oversupply due to new dis-
coveries, and major recessions in Western Europe and the United States.32
Profit margins and returns on investment for both major oil companies
and independent refiners are summarized in Tables 9-14 and 9-15. In those
tables, profit margin refers to net (after-tax) income as a percentage of
sales, while return on investment expresses net (after-tax) income as a
percentage of total investment or total assets.
9-22
-------
Table 9-13. EXPORTS OF SELECTED PETROLEUM PRODUCTS BY YEARa
UNITED STATES, 1969-1981
(1,000 m3/cd)
Year
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
Motor
Gasoline
0.3
0.2
0.2
0.2
0.6
0.3
0.3
0.5
0.3
0.2
0.0
0.2
0.3
Distillate
Fuel Oil
0.5
0.5
1.3
0.5
1.4
0.3
0.2
0.2
0.2
0.5
0.5
0.5
0.8
Residual
Fuel Oil
7.3
8.6
5.7
5.2
3.7
2.2
2.4
1.9
1.0
2.1
1.4
5.2
18.8
Jet Fuel
0.8
1.0
0.6
0.3
0.8
0.3
0.3
0.3
0.3
0.2
0.2
0.2
0.3
Kerosene NGL and LRG
0.2 5.6
4.3
0.2 4.1
4.9
4.3
4.0
4.1
4.0
2.9
3.2
NA
NA
NA
^Reference 13. Section VII, Tables 5, 6, 6a, 7, 7a, 14, 15, 16, 16a, 17 and 17a,
NA - not available.
9-23
-------
Table 9-14. PROFIT MARGINS FOR MAJOR CORPORATIONS WITH
PETROLEUM REFINERY CAPACITY, 1977-19813
(Percent)
Company
Integrated-International
British Petroleum
Exxon Corp.
Gulf Oil
Mobil Corp.
Royal Dutch Petroleum
Standard Oil of California
Texaco, Inc.
Integrated -Domestic
Amerada Hess
American Petrofina
Atlantic Richfield
Diamond Shamrock
Getty Oil
Kerr-McGee
Murphy Oil
Occidental Petroleum
Phillips Petroleum
Shell Oil
Standard Oil (Indiana)
Standard Oil (Ohio)
Sun Co.
Union Oil of California
Refiners
Ashland Oil
Charter Co.
Crown Central Petroleum
Holly Corp.
Quaker State
Tesoro Petroleum
Tosco Corp.
1977
3.0
4.5
4.2
3.1
6.0
4.9
3.3
3.9
3.6
6.4
10.6
9.9
5.5
4.2
3.6
8.2
7.3
7.8
5.2
5.6
5.9
3.4
1.3
2.0
3.8
6.0
0.1
1.2
1978
3.1
4.6
4.4
3.2
5.0
4.8
3.0
3.0
2.8
6.5
7.8
9.3
5.7
3.9
0.1
10.2
7.4
7.2
8.7
4.9
6.4
4.7
1.2
2.8
3.5
4.9
2.4
1.6
1979
8.9
5.4
5.5
4.5
11.1
6.0
4.6
7r
.5
5.2
7.2
7.6
12.5
6.0
6.6
5.9
9.4
7.8
8.1
15.0
6.6
6.6
81
.1
8.7
6.8
2.6
4.9
2.5
4.1
1980
6.9
5.5
5.3
4.7
6.3
5.9
4.4
6f\
.9
4.9
7.0
6.4
8.6
5.2
7.7
5.7
8.0
7.8
7.3
16.4
5.6
6.5
2c
.5
1.1
1.5
2.2
3.1
2.9
1.9
1981
4.2
5.2
4.4
3.8
4.7
5.4
4.0
2-5
.0
2.9
6.0
6.8
6.6
5.5
6.7
4.9
5.5
7.9
6.4
14.5
7.2
7.4
In
.U
1.1
0.2
1.4
3.0
2.6
0.7
Reference 14, p. 088.
9-24
-------
Table 9-15. RETURN ON INVESTMENT OF MAJOR CORPORATIONS
WITH PETROLEUM REFINERY CAPACITY 1977-19813
(Percent)
Company
Integrated-International
British Petroleum
Exxon Corp.
Gulf Oil
Mobil Corp.
Royal Dutch Petroleum
Standard Oil of California
Texaco, Inc.
Integrated -Domestic
Amerada Hess
American Petrofina
Atlantic Richfield
Diamond Shamrock
Getty Oil
Kerr-McGee
Murphy Oil
Occidental Petroleum
Phillips Petroleum
Shell Oil
Standard Oil (Indiana)
Standard Oil (Ohio)
Sun Co.
Union Oil of California
Refiners
Ashland Oil
Charter Co.
Crown Central Petroleum
Holly Corp.
Quaker State
Tesoro Petroleum
Tosco Corp.
1977
4.3
6.5
5.4
5.1
8.0
7.1
5.0
6.0
4.6
6.8
9.4
8.0
6.9
3.7
5.0
9.5
8.3
8.4
2.3
6.6
7.0
6^
.7
3.2
5.1
10.6
7.3
0.7
2.8
1978
4.1
6.9
5.4
5.2
6.0
7.0
4.4
4.2
3.5
6.7
6.7
7.4
6.1
3.2
0.2
11.1
8.4
8.0
5.0
6.8
7.3
80
.0
3.4
6.4
9.9
6.3
5.3
4.2
1979
11.8
9.5
8.2
8.0
13.5
10.2
8.1
11.3
8.0
8.9
7.0
11.2
7.3
6.1
10.9
11.5
9.1
9.6
13.4
10.2
8.7
on o
C\J tf.
29.1
16.8
8.0
7.6
9.9
14.2
1980
8.8
10.7
7.8
9.3
7.7
11.9
9.1
9.6
9.2
10.7
7.5
12.2
7.1
7.2
11.1
11.7
N/A
10.4
17.0
7.8
10.1
6Q
• O
2.8
3.6
8.0
5.3
1.7
6.2
1981
4.4
9.4
6.5
7.2
5.5
10.4
8.7
3.5
6.4
9.1
7.5
9.6
6.8
6.4
9.0
8.3
9.0
8.9
14.0
9.5
11.0
? 4
C. .H
3.4
0.6
3r\
.8
5.4
10.5
2.3
Reference 14, p. 087.
9-25
-------
9.2 ECONOMIC IMPACT ANALYSIS
9.2.1 Introduction and Summary
In the following sections the economic impacts of the regulatory
alternatives noted in Chapter 6 are discussed. Also presented is a summary
of the method used to estimate such impacts. In general, economic impacts
are described in terms of the price increases that may be prompted by the
various regulatory alternatives, and the potential reductions in petroleum
product output that could result as consumers respond to increased prices.
The socioeconomic impacts of the proposed NSPS including inflationary,
employment, balance of trade, and small business impacts are addressed in
Section 9.3. As noted in that section, the fifth-year annualized costs of
the most costly regulatory alternatives are $6.3 million, well below the $100
million level that Executive Order 12291 identifies as the threshold for
major regulatory actions.
With regard to the price increases and industry-wide output reductions
that could result from the costs of this NSPS, all price and output changes
are very small. If Regulatory Alternative II is required for the three
sources described in the previous section, price increases would be less than
$0.03 per m^ ($0.005/Bbl) and industry-wide output reductions would be
about 110 m3 per day (about 710 Bbl/day). These changes represent a
0.01 percent increase in price and a 0.004 percent decrease in quantity
demanded. With the higher costs of Regulatory Alternative III for the three
sources, price increases would be less than $0.34 per m^ ($0.05/Bbl) and
output reductions would be about 1,200 m^ per day (about 7,560 Bbl/day).
In this case the price increase is about 0.13 percent, and the quantity
demanded is reduced 0.05 percent.
9.2.2 Method
As explained in Chapter 3, the petroleum refinery wastewater system
collects wastewater from numerous points throughout the refinery, and
treats it by way of the separation and flotation processes previously
described. Such wastewater is generated through the operation of various
process units and may also be the result of storm water runoff at the refinery
site. For these reasons, the costs of operating a specific wastewater system
cannot be attributed to the production of an individual refined petroleum
9-26
-------
product, or group of products, but should be allocated over all products
produced at the refinery. Likewise, the total annualized costs incurred by
the refinery in the control of VOC emissions from the wastewater system
should also be evaluated from the perspective of total refinery output,
rather than the output of an individual product, or group of products.
The method used to evaluate potential price and output impacts has
three basic parts:
o the estimation of the annualized control cost per unit of output
produced at a new refinery (i.e. required price increase),
o the estimation of the price per unit of refinery output and
total output demanded in 1989, as well as the demand curve for
petroleum products in that year, and
o the estimation of product prices and demand from domestic refineries
in 1989 both with and without the costs related to the wastewater
NSPS.
Each of these tasks is discussed in greater detail below.
For purposes of this analysis it is assumed that the market for
refined petroleum products is basically competitive, and that there is little
competition from imports of refined products. It is also assumed that, as
projected by the U.S. Department of Energy (DOE), continued economic growth
will result in 1989 prices and production levels that are higher than
those currently observed. Under such conditions, 1989 prices and output
will be influenced by changes in the cost structure of the few totally new
refineries expected to be constructed over the next five years. This is true
because these refineries will have higher average total costs relative to
existing refineries, and as such, will determine the point of intersection
between the industry supply and demand curves. Consequently, even though
most new unit constructions and modifications will occur at existing refin-
eries, the major focus of this analysis is upon the extent to which NSPS
costs will increase the total per unit cost of new refineries.
The estimation of the extent to which the cost/price structure of a
new refinery will be affected, entails the approximation of the annual
capacity of a new refinery, the number of process units that will comprise
9-27
-------
such a refinery, and the total annualized costs to the refinery to control
VOC emissions from all process drain systems, the oil-water separator, and
the air flotation system. In this regard it has been assumed that any new
refinery will be relatively small with daily capacity of 4,000 m3 (about
25,000 Bbl), and will require controls on drains at six process units, two
each for Model Units A, B, and C. The refinery is also assumed to have one
oil-water separator and one air flotation system. It should be noted that in
summarizing NSPS control costs for the refinery, three "worst case" assumptions
are made. That is, it is assumed that dedicated control devices are needed
for both the oil-water separator and air flotation systems, and that these
systems are of the API and DAF types respectively. All three assumptions
imply higher NSPS control costs.
Both the average size of the expected new refinery and number of
process units were selected after review of the capacity and complexity of
those new refineries currently under construction, as reported in published
summaries of new refinery construction activities.33 To the extent that a
new refinery may have fewer process units, total costs to the refinery will
be lower. Finally, per unit annualized costs are estimated through the
division of total annualized NSPS control costs for the refinery by its
expected annual volume of output.
The next step in this method entails the estimation of price per unit
and total domestic refinery output for the year 1989. This year is of
concern because it represents the fifth complete year after proposal, and
because the current planning horizon of the industry extends to about that
point, given the time required to plan, design and construct completely new
refineries.
The estimation of 1989 price and output, as well as the demand curve
for refined products in that year, has been made possible through the results
of DOE econometric models. In particular, published results generated by
OOE's Intermediate Future Forecasting System (IFFS) allow the estimation of
equilibrium price and quantity under several assumptions regarding future
world crude oil prices.34
Some results of the IFFS model have been noted in Table 9-8 and are
used in the following section to approximate the demand curve for refined
9-28
-------
products as it might exist in 1989. The equation for the demand curve for
refined petroleum preducts in 1989, has been estimated in this analysis by
observing two points that lie on the curve, and solving for the straight line
that includes those two points. As shown in the following section, the
points selected are quantity demanded at the most likely 1989 price and
quantity demanded if the 1989 price is about 25 percent higher. The straight
line connecting these two points provides an approximation of the 1989 demand
curve because the two points estimate the level of demand expected in that
year if all factors other than price are held constant. In reality the
demand curve is probably not linear, but for the purpose of this analysis
linearity is assumed because the control costs will add very little to 1989
baseline prices. Consequently, the movement up the demand curve that will
result as consumers respond to slightly higher prices will be very small,
thus reducing the significance of the precise shape of the demand curve in
that area.
Finally, estimates of prices and the demand curve for the industry in
1989, together with estimates of the costs per unit attributable to the NSPS,
will allow approximations of the degree to which industry-wide output will
fall short of the output level that would be expected without the NSPS. Such
lower industry-wide output will have implications for the amount of new
capacity required to meet the future demand for refined petroleum products.
Estimates of 1989 demand under the two regulatory alternatives are made by
simply solving the equation for the 1989 demand curve, under the assumption
that 1989 prices will be higher by the amount of the NSPS control costs. A
horizontal supply curve is implicitly assumed by this part of the analysis,
and the extent to which the NSPS costs shift this curve upward is determined
by the annualized control costs. The following section details the quanti-
tative application of the method outlined above.
9.2.3 Analysis
As explained in the previous section the focus of this analysis is
upon the cost structure of a hypothetical new refinery, and in particular the
extent to which the NSPS costs will increase the per unit cost of the refinery,
and ultimately the market clearing price of all refined petroleum products.
Tables 9-16 and 9-17 demonstrate the calculation of annualized cost on a
9-29
-------
Table 9-16. TOTAL ANNUALIZED CONTROL COSTS FOR A
NEW REFINERY, REGULATORY ALTERNATIVE II*
($1,000 1983)
Model
Unit
Annualized
Cost/Unit
Number
of Units
Annualized
Cost/Refinery
Process Drain Systems
A $5.34b
B 2.54°
C 1.61&
Oil-Water Separator 5.67C
Air Flotation System 1.85d
2
2
2
1
1
TOTAL
$10.68
5.08
3.22
5.67
1.85
26.50
Capacity = 4,000 m3.
bTable 8-4.
CTable 8-9.
dTable 8-13.
9-30
-------
Table 9-17. TOTAL ANNUALIZED CONTROL COSTS FOR A
NEW REFINERY, REGULATORY ALTERNATIVE Ilia
($1,000 1983)
Model
Unit
Annuali zed
Cost/Unit
Number
of Units
Annual ized
Cost/Refinery
Process Drain Systems
Oil
Air
A
B
C
-Water Separator
Flotation System
$47.62b
32.30°
26.21b
38.46C
34.64d
2
2
2
1
1
TOTAL
$ 95.24
64.60
52.42
38.46
34.64
285.36
Capacity = 4,000 m3.
bTable 8-4.
cTable 8-9. API separator with emissions vented to a dedicated control device.
dTable 8-13. OAF system with emissions vented to a dedicated control device.
9-31
-------
refinery basis assuming that the new refinery will have daily capacity
of 4,000 m3 (about 25,000 Bbl/day) and will have six process units and
both an oil-water separation and an air flotation system. According to the
data shown in these tables, total annualized control costs for the refinery
are $26.50 thousand and $285.36 thousand for Regulatory Alternatives II and
III respectively.
In order to express these costs on a per unit output basis, the
annualized costs are divided by total annual output. Assuming the refinery
operates 350 days per year and at 60 percent of the designed capacity, annual
output is 840,000 m3 (5,283,600 Bbl). Thus on a per unit basis the
annualized cost are $0.03 and $0.34 per m3 for Regulatory Alternatives II
and III respectively ($0.005 and $0.05/Bbl).
As noted in the previous section, the results of DOE modelling activi-
ties have allowed the estimation of equilibrium prices and quantities in
1989. While DOE has projected United States refinery demand under three
possible world crude oil prices (in 1982 dollars) these prices have been
converted to domestic wholesale prices for refined products to allow the
approximation of the 1989 demand curve.
The relevant price and quantity data are summarized in Table 9-18.
The world crude oil prices are those reported by DOE, and are also noted in
Table 9-8 of Section 9.1. To convert crude prices to wholesale prices for
refined products, the crude prices have been increased by 8.55 percent
according to recently observed price differences between the two products.3->
The 1989 wholesale price estimates (in 1982 dollars) are presented in the
third and fourth columns of Table 9-18. Finally, because the control costs
presented in Chapter 8 are expressed in terms of third quarter 1983 dollars,
the 1989 wholesale prices (in 1982 dollars) are updated according to the 6NP
price deflator.
The equilibrium price and quantity for 1989 are assumed to be those
represented by the mid-level price scenario. Table 9-18 shows this equili-
brium price and quantity level to be $257.16 per m3 ($40.88/Bbl) and
2,514.29 thousand m3 per day (15,814.90 thousand Bbl/day). The slope
of the demand curve in the immediate area of this equilibrium can be approxi-
mated from the data provided by Table 9-18. Because the table summarizes
demand levels expected when all factors other than price are held constant,
9-32
-------
Table 9-18. DOE PROJECTED PRICES AND DOMESTIC REFINERY DEMAND
UNDER THREE WORLD OIL PRICE SCENARIOS, 1989
World Crude
Oil Price, 19893
(1982 $'s)
$/Bbl $/m3
Low 26.00 163.54
Mid 36.00 226.44
High 45.00 283.05
U.S. Wholesale U.S. Wholesale Total U.S. Refinery
Prices 1989b Prices 1989C Demand 1989d
(1982 $'s) (1983 $'s) (1,000 m3/day)
$/Bbl $/m3 $/Bbl $/m3 1,000 Bbl 1,000m3
28.22 177.52 29.52 185.72 17,591.09 2,796.68
39.08 245.80 40.88 257.16 15,814.90 2,514.29
48.85 307.25 51.11 321.45 14,389.40 2,287.66
aTable 9-8.
bCrude prices converted to wholesale prices for refined products, by applying a factor of
^ 1.0855.
co
cPrices converted to 3rd quarter 1983 dollars through GNP Implicit Price Deflator where
1982 = 206.88, and 3rd quarter 1983 = 216.44.
dTable 9-8.
-------
the demand curve in the area immediately above the mid-price equilibrium can
be approximated by solving for the straight line between the price/quantity
points defined by the high and mid-price scenarios. When the two points
($257.16, 2,514.29 thousand m3/day) and ($321.45, 2,287.66 thousand m3/day)
are considered the following equation for the demand curve is obtained;
Quantity (1,000 m3/day) = 3,420.811 - 3.525125 Price,
where price and quantity are the independent and dependent variables
respectively.
The final step in the analysis is to add the NSPS costs per refinery
to the 1989 equilibrium price for refined products, and estimate 1989 demand
levels from the demand equation noted above. With regard to prices, it has
been shown that the 1989 industry baseline price of $257.16 per m3 would
increase to $257.19 and $257.50 per m3 under Regulatory Alternatives II and
III respectively, if all costs are passed through in the form of higher
prices. Solving the demand equation for these prices decreases the estimate
of 1989 quantity demanded from the 1989 baseline of 2,514.29 thousand m3
per day to 2,514.18 thousand m3 per day and 2,513.09 thousand m3 per day
under Regulatory Alternatives II and III respectively. All 1989 prices and
demand levels are summarized in Table 9-19.
9.2.4 Conclusions
The general conclusion to be derived from the preceding analysis is
that the NSPS for refinery wastewater systems will have very little impact
upon either the firms that refine petroleum products or the consuming public.
Table 9-20 summarizes the changes in price and quantity demanded that can be
expected as both the demand for and supply of petroleum products from domestic
refineries grows until the year 1989. As indicated, market forces alone will
increase the price of refined products by about $42.86 per m3 ($6.81/Bbl)
over that period (i.e., from $214.30/m3 in 1983, to $257.16/m3 in 1989 as shown
in Table 9-19). Such forces will determine the market clearing price
and quantity in 1989 and include such factors as: the price of imported and
domestic crude oil and the proportions of each used by domestic refineries;
the prices of alternative sources of energy; the growth of the United States
9-34
-------
CO
cn
Table 9-19. PRICE AND TOTAL DEMAND
UNDER REGULATORY ALTERNATIVES II AND III
(3rd quarter 1983 dollars, 1,000 m3/day, 1,000 Bbl/day)
1983 Baseline3.b 1989 Baselinec Reg. Alt. II Reg. Alt. Ill
Price Demand Price Demand Price DemandPrice Demand
Cubic Meters (m3) $214.30 2,320.79 $257.16 2,514.29 $257.19 2,514.18 $257.50 2,513.09
Barrels (Bbl) $ 34.07 14,597.79 $ 40.88 15,814.90 $ 40.89 15.814.19 $ 40.93 15,807.34
3Table 9-8, prices converted to 3rd quarter 1983 dollars through 6NP Implicit Price Deflator where
1982 = 206.88, and 3rd quarter 1983 = 216.44.
bCrude prices converted to wholesale prices for refined products by factor of 1.0855.
CTable 9-18.
-------
Table 9-20. CHANGES IN 1989 PRICE AND DEMAND
COMPARED WITH 1983 BASELINE LEVELS
(3rd quarter 1983 dollars, 1,000 m3/day, 1,000 Bbl/day)
Changes Under Changes Under Changes Under
Reg. Alt. I« Reg. Alt. II Reg. Alt. Ill
Price Demand Price Demand Price Demand
Cubic Meters (m3) $42.86 193.50 $42.89 193.39 $43.20 192.30
Barrels (Bbl) $ 6.81 1,217.11 $ 6.82 1,216.40 $ 6.86 1,209.55
aNo NSPS control, thus these increases in price and quantity demanded are
due to market forces alone.
9-36
-------
and international economies; and the costs of other inputs into the refinery
industry (e.g. labor-and capital).
If the NSPS costs are also considered in addition to the factors noted
above, the prices of refined products would show very little additional
increases. If the industry incurs the costs related to Regulatory Alternative
II, the price of refined products would increase about $42.89 per m3 ($6.82/Bbl),
or $0.03 per m3 (less than $0.01/Bbl) more than they would without the NSPS.
If the higher costs of Regulatory Alternative III are incurred the increase
would be about $0.34 per m3 ($0.05/Bbl).
Although the increases noted above are very low, and may in fact be
imperceptible to the average consumer, the method used in this analysis
allows some approximation of sales decreases that would occur as consumers
encounter the slightly higher prices. Table 9-20 shows that in 1989, demand
would be 193.50 thousand m3 per day (1,217.11 thousand Bbl/day) higher than
in 1983, if the NSPS is not promulgated. However, with the standard, demand
would be 193.39 thousand m3 per day (1,216.40 thousand Bbl/day) higher
under Regulatory Alternative II, and 192.30 thousand m3 per day (1,209.55
thousand Bbl/day) higher under Regulatory Alternative III. Thus Regulatory
Alternative II would reduce 1989 demand by about 110 m3 per day (about 710
Bbl/day) and Regulatory Alternative III by 1,200 m3 per day (about 7,560
Bbl/day). Under the competitive market and capacity utilization assumptions
made in this analysis, it should be concluded that planned additions to
industry-wide capacity would be reduced by these small amounts if either
Regulatory Alternative II or III is promulgated.
9-37
-------
9.3 SOCIOECONOMIC AND INFLATIONARY IMPACTS
The previous section has described how the petroleum refining segment
of the national economy might be affected by this NSPS. In this section the
scope of the analysis is expanded so that the probability of broader economic
effects might be assessed. Among the issues examined are those related to
inflation, employment, the balance of trade, and the potential for adverse
impacts upon small businesses.
9.3.1 Executive Order 12291
According to the guidelines established by Executive Order 12291
"major rules" are those that are projected to have any of the following
impacts:
o an annual effect on the economy of $100 million or more,
o a major increase in costs or prices for consumers, individual
industries, federal, state, or local government agencies, or
geographic regions, or
o significant adverse effects on competition, employment,
investment, productivity, innovation, or on the ability of the
United States - based enterprises to compete with foreign-based
enterprises in domestic or export markets.
Each of these topics are examined in the following sections.
9.3.1.1 Fifth-Year Annualized Costs. The determination of fifth-year
annualized costs is demonstrated in Tables 9-21 and 9-22. Table 9-21 shows
the expected fifth-year cost for each model unit under each regulatory
alternative. The total costs noted in this table are determined through
consideration of the annualized costs presented in Chapter 8 and the number
of new unit constructions, reconstructions and modifications noted in
Chapter 7. The costs presented in both tables are the highest that should
be incurred under the regulatory alternatives, because it has been assumed
that control devices do not exist at the refineries that will be affected by
the NSPS.
9-38
-------
Table 9-21. SUMMARY OF FIFTH YEAR ANNUALIZED COST
BY MODEL UNIT AND REGULATORY ALTERNATIVE
(1,000 - 3rd quarter 1983 Dollars)
CO
Model
Unit
Process Drain Systems (New)a
Process Drain Systems (Retrofit)b
Oil -Water Separators (New)c
Oil-Water Separators (Retrofit)d
A
B
C
A
B
C
A
B
C
A
B
C
Regul atory
Alternative
II
III
II
III
II
III
II
III
II
III
II
III
II
III
II
III
II
III
II
III
II
III
II
III
Annual i zed Cost
Per Unit
$ 5.34
47.62
2.54
32.30
1.61
26.21
12.65
55.38
5.97
35.94
3.78
28.47
10.47
43.26
5.67
38.46
5.67
38.46
15.83
48.56
8.59
41.38
8.59
41.38
Number of
Units
27
27
27
27
51
51
3
3
3
3
9
9
5
5
10
10
15
15
1
1
1
1
1
1
Total
Annuali zed Cost
$ 114.18
1,285.74
68.58
872.10
82.11
1,336.71
37.95
166.14
17.91
107.82
34.02
256.23
52.35
216.30
56.70
384.60
85.05
576.90
15.83
48.56
8.59
41.38
8.59
41.38
-------
Table 9-21. (Continued)
Air Flotation (New)e
Air Flotation (Retrofit)f
Model
Unit
A
B
C
A
B
C
Regulatory
Alternative
II
III
II
III
II
III
II
III
II
III
II
III
Annual ized Cost
Per Unit
3.69
36.48
1.85
34.64
0.12
32.91
3.69
36.48
1.85
34.64
0.12
32.91
Number of
Units
5
5
10
10
10
10
1
1
1
1
1
1
Total
Annual ized Cost
18.45
182.40
18.50
346.40
1.20
329.10
3.69
36.48
1.85
34.64
0.12
32.91
vo
*Table 8-4.
bTable 8-5.
CTable 8-9.
dTable 8-8.
eTable 8-13.
fTable 8-13.
-------
Table 9-22. RANGE OF FIFTH-YEAR ANNUALIZED
COST OF AFFECTED FACILITIES
(1,000 - 3rd quarter 1983 Dollars)
Regulatory Alternative
ary
II
I II III
Process Drain Systems (New) $0 $294.87 $3,494.55
Process Drain Systems (Retrofit) 0 89.88 530.19
Oil-Water Separators (New) 0 194.10 1,177.80
Oil-Water Separators (Retrofit) 0 33.01 131.32
Air Flotation (New) 0 38.15 857.90
Air Flotation (Retrofit) Q_ 5.66 104.03
TOTAL 0 655.67 6,295.79
9-41
-------
Table 9-22 summarizes the fifth-year costs in terms of extremes.
Because Regulatory Alternative I entails no controls above those already
employed, no incremental fifth-year costs are incurred. If Regulatory
Alternative II is proposed for all model units, the total annualized costs in
the fifth-year after proposal would be about $0.7 million. Finally, under
Regulatory Alternative III, the most stringent and costly alternative,
fifth-year costs are about $6.3 million.
It should be noted that the fifth-year costs under all regulatory
alternatives are well below the $100 million threshold specified in the
Executive Order.
9.3.1.2 Inflationary Impacts. The proposal of this NSPS will have
virtually no effect upon the rate of inflation in the domestic economy. Even
if consumers eventually bear all of the fifth-year costs noted above, price
increases would be imperceptable as the total annual value of the industry's
output exceeds $100 billion.
9.3.1.3 Employment Impacts. The costs related to this NSPS would
have little effect upon the level of employment in the petroleum refining
industry. Table 9-5 shows that about 169,600 persons were employed in the
industry in 1981. Based upon industry capacity of about 3,000,000 m3 per
day during that year, the approximate capacity per worker is 18 m3 per day.
As reported in Section 9.2.4 the regulatory alternatives evaluated would
reduce the need for planned expansions in capacity up to 1989 by 110 and
1,200 m3 per day for Regulatory Alternatives II and III respectively.
Using the 18 m3 to 1 ratio of daily capacity to workers noted above, and
the expected baseline increase in demand of 193.5 thousand m3 per day
(Table 9-20), the growth in refinery employment over the next five years
would be about 10,750 workers without the NSPS. Because the decreases in
demand from the 1989 baseline are 110 and 1,200 m3 per day for Regulatory
Alternatives II and III respectively, these alternatives could reduce the
growth in employment by six and 67 workers.
9-42
-------
9.3.2 Small Business Impacts - Regulatory Flexibility Act
The Regulatory Flexibility Act, which became effective on January 1,
1981, requires the identification of potentially adverse impacts of Federal
regulations upon small entities including small businesses. The Act requires
that a Regulatory Flexibility Analysis (RFA) be completed for all Federal
regulations that could have a significant adverse economic impact on a
substantial number of small entities. The following discussion will show
that this NSPS will not affect a substantial number of small businesses.
For purpose of this discussion a small refinery is defined as one that
has crude oil capacity of less than 3,180 m3 per day (20,000 Bbl per day).
This level is based upon the recent definition of "small refiner" made by EPA
in establishing lead content rules for gasoline refiners. In those rules a
small refinery is defined as one that produces fewer than 1,590 m3 per day
(10,000 Bbl per day) of gasoline. Because on a national level about half of
total refinery throughput is gasoline, the crude oil capacity of the small
refinery is in this analysis, assumed to be twice the gasoline output or
3,180m3 per day (20,000 Bbl per day).
According to the most recent OAQPS/Economic Analysis Branch guidelines,
the NSPS must affect more than 20 percent of all small businesses in the
industry in order to be defined as one that affects a "substantial" number of
small businesses. Currently about one-third of all domestic refineries have
crude oil capacity of less than 3,180 m3 per day (20,000 Bbl per day).
Because there are about 220 petroleum refineries operating (Table 9-1), about
75 are considered to be small refineries. However, the most recent survey
of refinery construction and reconstruction activities shows that of about 75
current refinery construction and reconstruction projects, only five are
being undertaken at small refineries as defined above. Therefore fewer than
seven percent of the small refineries will be affected by the standard, if
the current distribution of construction activity continues. Because there
is no reason to presume that the current distribution of construction activity
among firms of various sizes will change, it is concluded that this standard
will not affect a substantial number of small refineries, and for this reason
a Regulatory Flexibility Analysis is not required.
9-43
-------
9.4 REFERENCES
1. Cantrell, A. Annual Refining Survey. Oil and Gas Journal. 82(13):112-123.
March 26, 1984.
2. Standard and Poor's. Industry Surveys - Oil, August 7, 1980 (Section 2)
p. 074.
3. Cantrell, A. Annual Refining Survey. Oil and Gas Journal. 72(13).
April 1, 1974.
4. Cantrell, A. Annual Refining Survey. Oil and Gas Journal. 73(14):98.
April 7, 1975.
5. Cantrell, A. Annual Refining Survey. Oil and Gas Journal. 74(13):129.
March 29, 1976.
6. Cantrell, A. Annual Refining Survey. Oil and Gas Journal. 75(13):98.
March 28, 1977.
7. Cantrell, A. Annual Refining Survey. Oil and Gas Journal. 76(12):113.
March 20, 1978.
8. Cantrell, A. Annual Refining Survey. Oil and Gas Journal. 77(3):127.
March 26, 1979.
9. Cantrell, A. Annual Refining Survey. Oil and Gas Journal. 78(12):130.
March 24, 1980.
10. Cantrell, A. Annual Refining Survey. Oil and Gas Journal. 79(12):110.
March 30, 1981.
11. Cantrell, A. Annual Refining Survey. Oil and Gas Journal. 80(12):128.
March 22, 1982.
12. Cantrell, A. Annual Refining Survey. Oil and Gas Journal. 81(12):128.
March 21, 1983.
13. American Petroleum Institute. Basic Petroleum Data Book - 1983.
14. Standard and Poor's. Industry Surveys - Oil. November 4, 1982 (Section 2)
p. 075.
15. Reference 2. p. 081.
16. Reference 2. p. 079.
17. Reference 13. Section V. Table 2.
18. Reference 13. Section V. Table 1.
9-44
-------
19. Cost of Benzene Reduction in Gasoline to the Petroleum Refining Industry.
U.S. Environmental Protection Agency. Office of Air Quality Planning and
Standards. EPA-450/2-78-021. April 1978, p. 1-3.
20. Jones, Harold. Pollution Controls in the Petroleum Industry. Noyes Data
Corporation. Park Ridge, NJ. 1973. 332 p.
21. 1978 Refining Process Handbook. Hydrocarbon Processing. 56(g):97-224.
September 1978.
22. Boland, R.F., et al. Screening Study for Miscellaneous Sources of
Hydrocarbon Emissions in Petroleum Refineries. EPA Report No. 450/3-76-041,
23. Energy Information Administration. U.S. Department of Energy. Supplement
to the 1982 Annual Energy Outlook. DOE/EIA-0408(82).
24. GAO Sees U.S. Refining Capacity Adequate for Future. Oil and Gas Journal.
81(7):60. February 14, 1983.
25. Hoffman, H.C. Components for Unleaded Gasoline. Hydrocarbon Processing.
59(2):57.
26. Reference 14. p. .075.
27. Reference 14. p. .075.
28. Energy Information Administration. U.S. Department of Energy. Annual
Report to Congress 1979. Vol.3. DOE/EIA-0173(79)/3. 359 p.
29. Energy Information Administration. U.S. Department of Energy. 1982
Annual Energy Review. April 1983.
30. Johnson, Axel R. Refining for the Next 20 Years. Hydrocarbon Processing.
59(2):57. February 1980.
31. Beck, J.R. Production Flat; Demand, Imports Off. Oil and Gas Journal.
78(4):108. January 28, 1980.
32. Reference 14. p. 057.
33. HPI Construction Boxscore. Hydrocarbon Processing Section 2. October
1983. pp. 3-8.
34. Reference 23. pp. 17, 68, 103, and 138.
35. The Petroleum Situation. The Chase Manhattan Bank, N.A. 7(1) :4.
March 1983.
9-45
-------
APPENDIX A
EVOLUTION OF THE BACKGROUND INFORMATION DOCUMENT
The purpose of this study was to develop background information to
support New Source Performance Standards (NSPS) for petroleum refinery
wastewater systems. Work on this study was performed by Radian Corporation
under contract to the United States Environmental Protection Agency (EPA),
specifically, under the direction of the Office of Air Quality Planning and
Standards (OAQPS), Emission Standards and Engineering Division (ESED).
In October 1982, Radian Corporation was contracted to develop a Source
Category Survey (Phase I). This phase of the study was a screening study of
refinery wastewater systems. From the screening study it was concluded that
NSPS should be developed for this source category. Radian Corporation then
began work on Phase II of this study, development of the Background
Information Document (BID). Phase II entailed a more complete and up to
date literature search and survey of the industry, including plant visits.
The feasibility of conducting emissions testing was determined during the
plant visits. Emissions testing was then conducted at three refineries.
The chronology which follows lists the major events which have occurred
in the development of background information for New Source Performance
Standards for petroleum refinery wastewater systems.
June 8, 1982 Plant Visit to Gulf Oil, Belle Chasse, Louisiana
June 8, 1982 Plant Visit ot Shell Oil, Norco, Louisiana
June 9, 1982 Plant Visit to Exxon, Baton Rouge, Louisiana
October 26-28, 1982 Plant Visit to Phillips Petroleum, Woods Cross, Utah
November 3, 1982 Meeting held between Radian Corporation and the EPA to
discuss Phase I of project
A-l
-------
November 10, 1982
January 25, 1983
February 2, 1983
March 14, 1983
March 15, 1983
March 16, 1983
March 16, 1983
March 17, 1983
March 18, 1983
March 25, 1983
March 30, 1983
April 6, 1983
April 6, 1983
May 3, 1983
May 11, 1983
May 12, 1983
May 13, 1983
June 2, 1983
July 28, 1983
August 1-12, 1983
August 15-19, 1983
August 30, 1983
September 19-23,
1983
October 7-8, 1983
Outline for Source Category Survey Report Submitted to
the EPA
Findings of Source Category Survey Report presented to
the EPA
Workplan for Phase II submitted to the EPA
Plant Visit to Champlin Oil, Wilmington, California
Plant Visit to Tosco, Bakersfield, California
Plant Visit to Chevron U.S.A., El Segundo, California
Plant Visit to Union Oil, Wilmington, California
Plant Visit to Mobil Oil, Torrance, California
Plant Visit to Texaco, Wilmington, California
Plant Visit to Sun Oil, Toledo, Ohio
Meeting with the EPA to discuss Testing Program
Plant Visit to Phillips Petroleum, Sweeny, Texas
Test Request submitted to Emission Measurement Branch of
the EPA
Meeting with the EPA to discuss inclusion of air
flotation systems and process drain systems in NSPS
Test Request sent to Phillips Petroleum, Sweeny, Texas
Test Request sent to-Chevron U.S.A., Inc., El Segundo, California
Test Request sent Mobil Oil, Torrance, California
Meeting held with the EPA to discuss test plans
Test Request sent to Golden West, Santa Fe Springs,
California
Emission Test at Chevron, U.S.A., El Segundo, California
Emission Test at Golden West, Santa Fe Springs, August
California
Concurrence Memorandum submitted to the EPA for Model
Plant Parameters and Regulatory Alternatives
Emission Test at Phillips Petroleum, Sweeny, Texas
Information requests sent to industry concerning fixed
roofs installed on API oil-water separators
A-2
-------
November 23, 1983
March 14, 1984
July 12, 1984
August 29, 1984
BID Chapters 3-6 Sent to Industry
Concerrence Meeting on Regulatory Approach to NSPS
Development
BID, Preamble, and Regulation sent to NAPCTAC Committee
Members
NAPCTAC Meeting
A-3
-------
DRAFT
April 20, 1984
APPENDIX B
INDEX TO ENVIRONMENTAL CONSIDERATIONS
This appendix consists of a reference system which is cross indexed
with the October 21, 1974, Federal Register (39 FR 37419) containing EPA
guidelines for the preparation of Environmental Impact Statements. This
index can be used to identify sections of the document which contain data
and information germane to any portion of the Federal Register guidelines.
B-l
-------
APPENDIX B
CROSS-INDEX TO ENVIRONMENTAL IMPACT CONSIDERATION
Agency Guidelines for Preparing
Regulatory Action Environmental
Impact Statements (39 FR 37419)
Location Within the Background
Information Document (BID)
1.
2.
Background and Summary of
Regulatory Alternatives
Statutory Basis for the
Standard
Industry Affected
Processes Affected
Availability of Control
Technology
Existing Regulations
at State or Local Level
Environmental, Energy, and
Economic Impacts of Regulatory
Alternatives
Health and Welfare Impact
The regulatory alternatives from
which standards will be chosen for
proposal are given in Chapter 6,
Section 6.2.
The statutory basis for proposing
standards is summarized in Chapter
2, Section 2.1.
A description of the industry to
be affected is given in Chapter 3,
Section 3.1.
A description of the process to be
affected is given in Chapter 3,
Section 3.2.
Information on the availability
of control technology is given
in Chapter 4.
A dicussion of existing regulations
for the industry to be affected by
the standards are included in
Chapter 3, Section 3.4.
The impact of emission control
systems on health and welfare
is considered in Chapter 7,
Section 7.2.
Continued
B-2
-------
CROSS-INDEX TO ENVIRONMENTAL IMPACT CONSIDERATIONS (Concluded)
Agency Guidelines for Preparing
Regulatory Action Environmental
Impact Statements (39 FR 37419)
Location Within the Background
Information Document (BID)
Air Pollution
Water Pollution
Solid Waste Disposal
Energy
Costs
Economics
The air pollutant impact of the
regulatory alternatives are
considered in Chapter 7,
Section 7.2.
The impacts of the regulatory
alternatives on water pollution are
considered in Chapter 7,
Section 7.3.
The impact of the regulatory
alternatives on solid waste
disposal are considered in
Chapter 7, Section 7.4.
The impacts of the regulatory
alternatives on energy use are
considered in Chapter 7,
Section 7.5.
The cost impact of the emission
control systems is considered in
Chapter 8.
Economic impacts of the regulatory
alternatives are considered in
Chapter 9.
B-3
-------
APPENDIX C
EMISSION SOURCE TEST DATA
The purpose of this appendix is to present VOC emissions test data used
in the development of this background information document. VOC emissions
test data were obtained from three refineries by the U.S. Environmental
Protection Agency. At one refinery, tests were conducted on a dissolved air
flotation system (DAF), an induced air flotation system (IAF), and an
equalization basin. At a second refinery, tests were conducted on an IAF
system. At a third refinery, tests were conducted on two IAF systems. In
addition to the emission tests, screening of process drains with a portable
VOC analyzer was performed at three refineries. The results of the tests
are described below along with the methodology used to conduct the tests.
C.I EMISSION MEASUREMENTS
C.I.I Chevron, U.S.A., Inc. Refinery - El Segundo, California.
The refinery wastewater system at Chevron is divided into segregated
and unsegregated systems. The segregated system handles the majority of the
oily wastewater while the unsegregated system handles mostly non-oily
wastewater. , Continuous monitoring of VOC emissions from the DAF and
equalization basin in the segregated system was performed. Continuous
monitoring of VOC emissions from the IAF system in the segregated system was
also conducted.
The DAF and equalization basin are located in the Effluent Treating
Plant (ETP) at Chevron. Two DAF systems are included in the effluent
treatment system, but only one was in operation during the test. The DAF
system treats oily wastewater from the API separators after the wastewater
has been held in a storage tank preceding the ETP. Effluent from the DAF
was discharged to the equalization basin before undergoing biological
treatment.
C-l
-------
Figure C-l shows the DAF system tested at Chevron. The DAF is equipped
with a fiberglass cover which consists of a series of ventilation holes
around its side. The cover also has three access doors and a center
ventilation hole. The DAF flotation chamber is connected to a vapor
recovery system. Two blowers rated at 4,000 scfm create a vacuum which
draws VOC and ventilation air from the flotation chamber. The captured VOC
is vented to an activated carbon bed located near the system.
As shown in Figure C-l, continuous monitoring of VOC from the DAF was
conducted at a sample point located between the DAF and the carbon house.
EPA Method 25A was used to measure the VOC level. In addition, gas
chromatography was used to identify the major volatile components of the
vent stream. EPA Method 18 was used for this purpose. A summary of the
results of the continuous monitoring of the DAF are shown in Table C-l. The
total hydrocarbon increments include methane. Gas chromatography results
are shown in Table C-2.
The equalization basin is shown in Figure C-2. As with the DAF, this
basin is completely covered. Ventilation holes are located on one side of
the basin and outlet ports are located on the opposing side. Two blowers
rated at 4,000 scfm create a vacuum which draws VOC and ventilation air from
the basin. The captured VOC is vented to an activated carbon bed similar to
that on the DAF system. Continuous monitoring at VOC level was conducted at
a sample point located between the equalization basin and the carbon house.
The sample point is shown in the figure.
The same analytical methods used on the DAF were used to monitor the
VOC and identify major volatile components being emitted from the
equalization basin. A summary of the results of the continuous monitoring
are shown in Table C-l. Gas chromatography results are shown in Table C-3.
' The IAF at Chevron receives effluent from an API separator which
handles mostly non-oily wastewater. The IAF is designed to be gas-tight and
the gaseous emissions are vented to a 55 gallon drum of activated carbon.
The IAF system is shown in Figure C-3.
The vapor space in the IAF was initially designed to be purged with
plant air. Evaluations of the system by Chevron found that purging was not
necessary to maintain safe operating conditions. Because of this, a steady
C-2
-------
flow of gas from the IAF to the carbon drum was not maintained. A small
flow of gas from the IAF did result from breathing losses in the system.
This flow was recorded with a 4" vane anemometer. The positive gas
displacement was calculated and used as the IAF outlet flow. Outlet VOC
concentration could then be calculated using EPA Method 25A. The emission
rates and gas chromatography results from the IAF are shown in Table C-4.
In addition to the gaseous samples taken at Chevron, liquid samples of
the wastewater going to and from the API separators, DAF, IAF, and
equalization basin were obtained. These samples were analyzed for chemical
oxygen demand (COD), oil and grease, total organic carbon (TOC), and total
chromatographic organics (TCO). The results of the analyses are shown in
Tables C-5 to C-12. These samples were obtained in an attempt to correlate
VOC emissions with conventions at wastewater pollutant parameters.
2
C.I.2 Golden West Refinery - Santa Fe Springs, California
The refinery wastewater system at Golden West consists of two API
separators followed by an IAF system. The IAF system is operated gas-tight
and the vapor space is purged with plant air. The captured and purged VOC
are vented to a fired heater located near the treatment system. A small
blower serves to drive the VOC from the IAF to the fired heater.
Continuous monitoring of VOC from the IAF to the fired heater was
conducted at a sample point located on the outlet duct of the IAF. The IAF
system and sample point are shown in Figure C-4. EPA Method 25A was used in
monitoring the VOC. Gas chromatography was used to identify the major
volatile components of the vent stream. A summary of the results of the
continuous monitoring of the IAF is shown in Table C-13. The total
hydrocarbon measurements include methane. Gas chromatography results are
shown in Table C-14.
In addition to the gaseous samples taken at Golden West, liquid samples
of wastewater going to and from the API separators and IAF system were
obtained. As with the samples acquired at Chevron, these samples were
analyzed for COD, oil and grease, TOC, and TCO. The results of the analyses
are shown in Table C-15 to C-18.
(text continues on page C-41)
C-3
-------
Dissolved Air Flotation T-302
(Not operating during test period)
o
Dissolved A1r Flotation
T-202
Sampling
Location
-X—
Flash Mix Flocculatlon Tank
T-200 T-201
2000 scfm
blower
Activated
carbon beds
2000 scfm
blower
Figure C-l.i Dissolved air flotation system with sample location
-------
TABLE C-l SUMMARY OF DAILY EMISSION RATE AVERAGES: CONTINUOUS MONITORING
TABLE C 1. Jg|™fUCHEYRON REFINERY, EL SEGUNDO, CALIFORNIA
TEST DAY 8/3/83 8/4/83 8/5/83 8/8/83 8/9/83 8/10/83 8/11/83
SAMPLE LOCATION
DAF Outlet
(Ibs/hr Total Hydrocarbon) 7.18 6.37 6.85 6.75 8.11 6.17 9.01
o
dn Equalization Tank
(Ibs/hr Total Hydrocarbon) 4.18 4.65 4.24
-------
TABLE C-2. GAS CHROMATOGRAPHY RESULTS FROM DAF SYSTEM
CHEVRON REFINERY, EL SEGUNDO, CALIFORNIA
DATE
TIME
ANALYTICAL RESULTS
(pp*v as compound)
C-l
C-2
C-3
C-4
C-5
Hexane
Benzene
Heptane
Toluene
m-Xylene
o-Xylene
TOTAL HYDROCARBON
(ppnv as compound)
CONTINUOUS MONITOR
DATA
Hydrocarbon Level
(ppwv as C3H8)
Emission Rate
Hbs/hr Total
i hydrocarbon)
8/3
1135-
1235
46.8
5.7
6.8
3.8
1.9
10.1
11.0
10. 0
39.2
6.8
3.4
145
510
6.69
8/3
1445-
1545
46.5
7.0
8.1
5.0
3.4
16.9
15.1
11.8
45.3
6.1
3.0
168
526
6.88
8/4
930-
1010
53.6
6.4
8.3
4.9
4.9
23.0
19.8
21.3
55.5
15.9
7.9
217
668
8.59
8/4
1430-
1515
45.5
5.3
6.2
4.4
3.8
15.1
13.2
6.6
32.4
7.7
3.0
143
339
4.35
8/5
900-
945
53.8
6.7
7.1
4.2
4.6
10.7
24.4
2.6
46.7
13.6
5.0
179
583
7.82
8/5
1500-
1530
58.3
6.5
8.3
0.6
18.0
35.0
44.4
10.4
3.8
185
482
6.38
(CONTINUED)
C-6
-------
TABLE C-2. GAS CHROMATOGRAPHY RESULTS FROM DAF SYSTEM
CHEVRON REFINERY, EL SEGUNDO, CALIFORNIA (CONTINUED)
DATE
TIME
ANALYTICAL RESULTS
(ppmv as compound)
C-l
C-2
C-3
C-4
C-5
Hexane
Benzene
Heptane
Tol uene
m-Xylene
o-Xylene
TOTAL HYDROCARBON
(ppmv as compound)
CONTINUOUS MONITOR
DATA
Hydrocarbon Level
(ppmv as C3H8)
Emission Rate
8/8
1100-
1300
55.3
4.5
5.6
4.0
3.4
16.1
39.8
46.4
11.3
3.9
190
495
6.72
8/8
1500-
1530
52.9
3.9
5.0
4.8
4.0
26.2
63.6
75.1
20.7
8.2
264
580
7.87
8/9
915-
1040
37.5
2.4
2.2
3.6
4.8
12.8
49.2
28.3
17.1
6.0
22.4
186
709
9.68
8/9
1400-
1455
34.8
1.8
2.6
3.2
4.8
0
8.0
44.4
17.4
7.0
24.2
148
592
8.09
8/10
904-
1004
26.4
2.1
2.0
1.7
0
6.7
23.7
7.0
0
12.7
5.2
87
460
5.28
8/11
1315-
1415
29.2
0
2.1
6.5
9.2
19.1
55.2
0
61.5
10.0
10.2
203
622
8.2;
(Ibs/hr Total
Hydrocarbon)
C-7
-------
Ventilation Holes
2000 scfm
blower
Activated
Carbon Bed
o
CO
2000
scfm Blower
Sample Location
Equalization Basin T-500
n..Hpt Pnrts
r t i i-
Figure C-2. Equalization Basin with Sample Location.
-------
TABLE C-3. GAS CHROMATOGRAPHY RESULTS FROM EQUALIZATION BASIN
CHEVRON REFINERY, EL SEGUNDO, CALIFORNIA
DATE
TIME
LOCATION
RUN NO.
8/3
1600-
1700
8/4
1053-
1235
8/4
1431-
1510
8/5
930-
1000
Ventilation air
TOTAL HYDROCARBON
(ppmv as compound) 72
CONTINUOUS MONITOR
DATA
Hydrocarbon Level
(ppmv as CH) 150
74
47
50
Emission Rate
(Ib/hr)
4.07
182
4.87
167
4.45
155
3.98
8/5
1228-
1252
76
155
3.98
8/5
1400-
1510
Carbon
house
outlet
OUT
ANALYTICAL RESULTS
(ppmv as compound)
C-l
C-2
C-3
C-4
C-5
Hexane
Benzene
Heptane
Toluene
m-Xylene
o-Xylene
27.0
2.0
0
0
0
0
7.7
29.2
4.6
1.7
29.4
1.2
0
0
0
2.3
9.7
25.5
4.0
1.5
24.6
0
0
0
0
2.1
4.9
13.6
1.7
0
17.7
0
0
0
0
1.4
7.8
18.7
3.6
1.1
20.4
1.8
. 0
0
0
2.1
12.5
29.8
7.0
2.4
22.3
1.6
0
0
0
0
20.4
26.8
0
0
72
179
4.65
C-9
-------
TABLE C-3. (Continued)
DATE
TIME
LOCATION
RUN NO.
ANALYTICAL RESULTS
(ppmv as compound)
C-l
C-2
C-3
C-4
C-5
Hexane
Benzene
Heptane
Toluene
m-Xylene
o-Xylene
TOTAL HYDROCARBON
(ppmv as compound)
CONTINUOUS MONITOR
DATA
Hydrocarbon Level
(ppmv as C3Hg)
Emission Rate
(Ib/hr)
8/12/83
Ventilation
air
8/12/83
8/12/83
Carbon house exhaust
15.4
0
0
0
0
5.8
38.6
0
0
14.8
5.6
89
284
24.4
0
0
0
0
0
0
0
0
0
0
24
23.
0
0
0
0
0
0
0
0
0
0
23
29
7.54
0.77
C-10
-------
Wastewater from API
separator
Induced Air Flotation System
ourtn
Jl
r
JSL
Anemometer
Flow Measurement
Adaptation
Activated
Carbon Drum
Gaseous
Emissions
Sampling Location
Mobile Lab
Figure C-3. Induced air flotation system at Chevron - El Segundo, California.
-------
TABLE C-4. GAS CHROMATOGRAPHY AND EMISSION RATES FROM IAF SYSTEM
CHEVRON REFINERY, ELSEGUNDO, CALIFORNIA
DATE
TIME
LOCATION
RUN NO.
ANALYTICAL RESULTS
(ppmv as compound)
C-l
C-2
C-3
C-4
C-5
Hexane
Benzene
Heptane
Toluene
m-Xylene
0-Xylene
8/11
0924-
0942
Ventilation
1
1602
7.6
18.2
42.0
283
1288
835
826
421
252
145
8/11
1213-
1245
air
2
2818
3217
2913
80.5
220
6127
2642
938
0
105
31.7
8/12
1213-
1254
Carbon
1
2156
8.2
21.8
72.1
510
2005
2101
793
0
385
106
8/12
1040-
1120
drum outlet
2
1762
4.5
12.8
36.4
110
2033
1074
449
0
168
67.8
TOTAL HYDROCARBON
(ppmv as compound) 5720 19,092 8158 5717
CONTINUOUS MONITOR
DATA
Hydrocarbon Level
(ppmv as C3Hg) 6950 7300 7222 . 6601
Emission Rate
(lb/hr) 0.20 0.21 0.18 0.16
C-12
-------
TABLE C-5. LIQUID SAMPLES TAKEN ON 8/3/83 -
CHEVRON REFINERY, EL SEGUNDO, CALIFORNIA
Liquid Composite Samples
DAF-in
DAF-out
EQ-out
COD
mg/L
2,969
3,008
1,748
1,911
1,870
Oil /grease
mg/L
491
535
133
144
123
120
TOC TCO
mg/L mg/L
— 71. 56
— 30.90
— 21. 00
Volatile Organic Samples
DAF-in #1 VOA (1650)a — — 611 —
DAF-out #1 VOA (1650) — — 365 —
EQ-out VOA (1650) — _ 661 —
aTime sample taken.
(continued)
C-13
-------
TABLE C-5. LIQUID SAMPLES TAKEN ON 8/3/83 - CHEVRON REFINERY,
EL SEGUNDO, CALIFORNIA (CONTINUED)
Compound
ng/1
Liauid Composite Samples
DAF Influent
DAF Effluent
Equilization Basin Effluent
Toluene
C9
C9
C9
Cio
Cn
C12
C12
C12
C12
C12
C12
Cl2
Cl3
Cl3
Cl4
Cis
Cis
Toluene
C9
C9
Cio
Cio
Cn
Cn
C12
Cl3
Toluene
C9
C9
Cio
Cio
13.302
2.278
1.328
1.040
17.709
2.679
4.207
4.940
5.339
12.214
2.932
1.436
1.930
1.487
10.496
3.128
4.838
3.570
3.066
3.643
2.595
15.412
4.972
5.549
0.828
1.383
2.679
2.232
2.257
3.301
2.460
11. 538
3.927
3.617
1.180
Note: Benzene could not be determined due to a co-eluting peak in the
solvent.
Note: These values were calculated using average response factors of
CT-CH, Cu-C16, and C17 to C2s hydrocarbons. Due to the reduced
response of C17 to C25 hydrocarbons as compared to C7-Cn, high
values of some C17-C2S compounds were found.
C-14
-------
TABLE C-6. LIQUID SAMPLES TAKEN ON 8/4/83 -
CHEVRON REFINERY, EL SEGUNDO, CALIFORNIA
COD Oil /grease TOC
rag/L mg/L mg/L
Liquid Composite Samples
OAF- In
OAF-out
EQ-out
Volatile Organic Samples
DAF-in-VOA pm (1500)
DAF-in-VOA (1000)
DAF-out VOA pm (1500)
OAF-out VOA (1000)
EQ-out VPA (1000)
EQ-out VOA (1500)
4,024 440 —
4,228 441 —
1,545 125 —
1,585 94 —
1,565 126 —
2,033 148 —
2,155 142 —
— — 484
a
— ' — 478
— — 475
— — 550
— — 542
— — 464
— — 455
— — 511
aSample lost; replaced with aliquot from DAF-in liquid composite samples.
Result was 1,096 mg/L.
C-15
-------
TABLE C-7. LIQUID SAMPLES TAKEN ON 8/5/83 -
CHEVRON REFINERY, EL SEGUNDO, CALIFORNIA
COD Oil/grease TOC TCO
mg/L mg/L mg/L mg/L
Liquid Composite Samples
DAF-in
DAF-out
EQ-out
Volatile Organic Samples
DAF-in VOA (0915)
DAF-in VOA (1530)
DAF-out VOA (0915)
DAF-out VOA (1530)
EQ-out VOA (1530)
EQ-out VOA (0915)
8,056 6.14 — —
2,179 2.37 — —
1,240 110 — —
1,301 109 — —
— — a —
— — 722 —
_ _ 578 —
— — 713 —
_ — 600 —
— — b —
aSample lost: replaced with aliquot from DAF-in liquid composite samples.
Results are 849,940,860 mg/L.
bSample lost; aliquot from EQ-out liquid composite samples. Results are
416,398,476 mg/L.
C-16
-------
TABLE C-8. LIQUID SAMPLES TAKEN ON 8/8/83 -
CHEVRON REFINERY, EL SEGUNDO, CALIFORNIA
COD Oil/grease TOC TCO
mg/L mg/L mg/L mg/L
Liquid Composite Samples
DAF-in
OAF- out
API-2 Inlet A (201)
API-2 Inlet B (202)
API-2 Inlet C (203)
API-2 Inlet D (204)
API -4
Volatile Organic Samples
DAF-in VOA (1100)
DAF-in VOA (1500)
DAF-out VOA (1100)
DAF-out VOA (1500)
2,155
2,114
1,470
20.3
2,560
463
480
2,440
—
—
—
—
383
376
0.21
6.4
65.49
20.9
26.97
18.26
—
—
—
—
— 41. 94
— —
— 22. 38
— 1.74
— 84.00
— 9.30
— '8.26
— 45.66
538 —
a —
622 —
b ~~~
aSample lost; replaced with aliquot from DAF-in liquid composite samples.
TOC result is 016 mg/L.
bSample lost; replaced with aliquot from DAF-out liquid composite samples.
TOC result is 774 mg/L.
(Continued)
C-17
-------
TABLE C-8. LIQUID SAMPLES TAKEN ON 8/8/83 - CHEVRON REFINERY,
EL SEGUNDO, CALIFORNIA (CONTINUED)
Compound
mg/1
Liquid Composite Samples
DAF Influent Toluene
Cg
c*
Cio
Cio
Cio
Ci2
Cn
Cia
Cl4
Cis
Cis
Cjg
Toluene
DAF Effluent C9
C?o
Cio
Toluene
API *2 Influent C8
(Site 202) C9
C9
Cio
Cn
Cn
Cn
Cn
C\l
9.920
2.312
13.518
3.935
3.901
1.871
4.727
1.407
0.783
0.801
4.496
2.837
0.838
3.285
3.136
5.085
10.601
3.697
3.284
1.210
2.571
1.005
2.065
23.039
1.858
7.464
12.990
5.835
0.932
0.051
1.153
4.145
14. 226
(CONTINUED)
C-18
-------
TABLE C-8. LIQUID SAMPLES TAKEN ON 8/8/83 - CHEVRON REFINERY
EL SEGUNDO, CALIFORNIA (CONTINUED)
Compound
mg/1
API #2 Influent
(Site 203)
M
Cl4
C15
C16
els
C19
Toluene
C8
13.544
4.316
8.411
2.306
9.465
7.679
59.638
45.744
65.488
2.165
1.034
API #4 Influent
Toluene
Cio
C12
C12
C12
C12
C13
Cl4
Cis
Cis
6.595
1.848
12.555
3.390
3.291
3.341
8.448
2.436
1.395
1.447
7.986
1.654
5.173
1.388
5.558
4.977
46.394
C-19
-------
TABLE C-9. LIQUID SAMPLES TAKEN ON 8/9/83 -
CHEVRON REFINERY, EL SEGUNDO, CALIFORNIA
COD Oil/grease TOC TCO
mg/L mg/L mg/L mg/L
Liquid Composite Samples
DAF-out
API-2 Inlet A (201)
API-2 Inlet B (202)
API-2 Inlet C (203)
API-2 Inlet D (204)
API-4
Volatile Organic Samples
DAF-in VOA (0900)
OAF-in VOA (1342)
DAF-out VOA (0900)
DAF-out VOA (1340)
1,579
693
3,155
5,179
2,230
620
154
61.56
19.50
32.27
18.28
23.90
482
440
341
509
C-20
-------
TABLE C-10. LIQUID SAMPLES TAKEN ON 8/10/83 -
CHEVRON REFINERY, EL SEGUNDO, CALIFORNIA
Liquid Composite Samples
DAF-in
DAF-in
DAF-out
API-2 SP 201
API-2 SP 202
API-2 SP 203
API-2 SP 204
COO
mg/L
2,170
2,121
2,078
594
2,764
950
2,635
Oil /grease TOC
mg/L mg/L
23.80 —
53.98 —
47.75 —
33.80 —
42.48 -^-
70.03 —
32.62 —
TCO
mg/L
—
—
—
—
—
—
—
Volatile Organic Samples
DAF-in VOA (0920) — — 619
DAF-in VOA (1600) — — 471
DAF-out VOA (0920) — — 546
DAF-out VOA (1600) — — 511
C-21
-------
TABLE C-ll. LIQUID SAMPLES TAKEN ON 8/11/83 -
CHEVRON REFINERY, EL SEGUNDO, CALIFORNIA
COD Oil /grease TOC TCO
rog/L mg/L mg/L mg/L
Liquid Composite Samples
OAF- in
OAF-out
lAF-in
lAF-out
API-4
API-2 SP 201
API-2 SP 202
API-2 SP 203
API-2 SP 204
Volatile Organic Samples
DAF-in VOA (0900)
OAF- in VOA (1530)
DAF-out VOA (0900)
DAF-out VOA (1530)
lAF-in VOA (1000)
lAF-in VOA (1600)
lAF-out VOA (1000)
lAF-out VOA (1600)
2,316 43.74 — 95.26
1,410 54.92 — 22.42
811 61.58 — 12.58
201 46.73 — 11.06
1,616 43.59 — 96.20
100 17.97 — 9.20
1,700 37.24 — 30.68
99 24.45 — 8.60
450 33.06 — 51.98
— — 530 —
— — 355 —
— — 454 —
— — 343 —
— — 64.5 —
— — 402 —
134
— — 52.0 —
(continued)
C-22
-------
TABLE C-ll. LIQUID SAMPLES TAKEN ON 8/11/83 - CHEVRON REFINERY
EL SEGUNDO, CALIFORNIA (CONTINUED)
Compound
mg/1
Liquid Composite Sampleg
DAF Influent
Toluene
14.141
Cg
C8
C8
C8
Cg
CQ
c*
Cjo
Cio
CIQ
C12
Cj.3
Cl4
Cis
Cl6
Cl7
Cig
C2o
Toluene
DAF Effluent C8
Co
C9
Cio
c
C12
Cis
Tol uene
IAF Influent C8
Toluene
IAF Effluent C8
1.211
1.471
5.429
1. 901
2.553
6.035
3.027
5.068
7.398
6.526
15.370
14.351
4.388
9.436
.10.194
6.915
58.459
47.247
44.281
28.031
4.430
0.838
0.805
7.528
4.021
3.658
1. 375
0.852
0.920
1.549
0.668
1.334
0.581
(continued)
C-23
-------
TABLE C-ll. LIQUID SAMPLES TAKEN ON 8/11/83 - CHEVRON REFINERY
EL SEGUNDO, CALIFORNIA (CONTINUED)
Compound
anr JL» T ft * Toluene
API #4 Influent £
Cg
C9
C9
C9
C9
Cio
Cio
Cio
Cio
Cio
Cio
Cn
Cn
Cn
Cl2
Cis
Cl6
ADI 12 Influent Toluene
(Site 202) £8
C9
C9
Cio
C10
Cn
els
c"
Cis
ADI #2 Influent Toluene
(Site 203)
mg/1
39.430
28.123
11.348
4.708
2.586
0.954
13.200
3.242
1.512
1.126
4.686
3.127
2.379
1.349
1.502
1.561
1.976
1.679
1.832
2.025
2.221
1.434
1.188
3.697
3.205
3.147
1.684
4.622
1.450
2.900
4.285
3.544
0.902
(CONTINUED)
C-24
-------
TABLE C-11. LIQUID SAMPLES TAKEN ON 8/11/83 - CHEVRON REFINERY
EL SE6UNDO, CALIFORNIA (CONTINUED)
Compound mg/1
API #2 Influent Toluene <0.5
(Site 204) Cn 4.055
Cw 1.755
Cn 1.505
Cn 1.002
Cn 1.395
Cn 2.130
C12 12.261
C12 3.872
C12 4.312
C13 10.914
C14 7.363
C1S 3.839
C16 70.078
C-25
-------
TABLE C-12. LIQUID SAMPLES TAKEN ON 8/12/83 -
CHEVRON REFINERY, EL SEGUNDO, CALIFORNIA
COD Oil /grease
ng/L mg/L
Liquid Composite Samples
lAF-in
lAF-out
API-4
API-2 SP 201
API-2 SP 202
API-2 SP 203
API-2 SP 204
320
302
202
405
1,584
1,000
a
14.14
64.95
26.5
12.0
70.71
36.74
a
TOC TCO
mg/L mg/L
— —
— —
— —
— —
— —
— — •
a a
Volatile Organic Samples
lAF-in VOA (0900)
lAF-in VOA (1250)
lAF-out VOA (0900)
lAF-out VOA (1330)
86.0 —
57.0 —
162 —
46.0 —
aSaaple broken in laboratory.
C-26
-------
Covered and Sealed IAF
Air 9 1"
IAF-INLET,
o
Water
Mater
API-INLET
Covered
API Separator
Covered
API Separator
IAF
Q
Platform
IAF-OUTLET
(GAS SAMPLE)
IAF-OUTLET
(PROCESS SAMPLE)
Open Bays
fired
Heater
Blower
Mater
Discharge
Figure C-4. Wastewater treatment facilities at Santa Fe Springs, California.
-------
TABLE C-13. DAILY EMISSION RATE AVERAGES AT IAF OUTLET -
GOLDEN WEST REFINERY, SANTA FE SPRINGS, CALIFORNIA
Average Emission Rate
Test Day (Ib/hr Total Hydrocarbon as C3Hg)
8/15/83 1.40
8A6/83 1.39
8A7/83 1.14
8/18/83 1.23
8/19/83 1.39
C-28
-------
TABLE C-14.
GAS CHROMATOGRAPHY RESULTS FROM IAF SYSTEM -
GOLDEN WEST REFINERY, SANTA FE SPRINGS, CALIFORNIA
DATE
TIME
ANALYTICAL RESULTS
(ppmv as compound)
C-l
C-2
C-3
C-4
C-5
Hexane
Benzene
Toluene
m-Xylene
o-Xylene
8/16
735-
835
74.0
6.8
14.2
38.6
52.0
115
1357
1346
933
326
8/16
1020-
1120
110
9.4
22.1
269
250
370
2851
2486
1458
467
8/16
1235-
1335
90.8
9.6
14.4
108
130
1068
2424
2321
1578
510
8/17
0745-
0845
138
7.8
19.0
140
183
180
1758
1629
905
305
8/17
1000-
1100
135
20.9
78.5
315
685
577
3638
2376
813
283
8/17
1153-
1253
262
122
365
341
524
3530 "
2476
885
308
TOTAL HYDROCARBON
(ppnv as compound)
CONTINUOUS MONITOR
DATA
Hydrocarbon Level
(ppmv as C3H8)
Emission Rate
(Tbs/hr Total
hydrocarbon)
4262
8292
8253
5265
8921
8813
6772 7104 7087 7008 8675 8811
1.47 . 1.54 1.54 1.15 1.42 1.45
(CONTINUED)
C-29
-------
TABLE C-14. GAS CHROMATOGRAPHY RESULTS FROM IAF SYSTEM -
C-OLDEN WEST REFINERY, SANTA FE SPRINGS, CALIFORNIA (CONTINUED)
DATE
TIME
8/18
1030-
1146
8/18
1310-
1410
8/19
850-
950
8/19
1030-
1130
ANALYTICAL RESULTS
(pp*v as coopound)
C-l
C-2
C-3
C-4
C-5
Hexane
Benzene
Toluene
nrXylene
o-Xylene
TOTAL HYDROCARBON
(ppnv as compound)
CONTINUOUS MONITOR
DATA
Hydrocarbon Level
(pp«v as as C3H8)
Emission Rate
(Ibs/hr Total
Hydrocarbons)
44.5
3.0
4.2
10.5
14.9
49.7
547
889
647
236
94.7
4.1
8.0
96.5
71.0
81.4
1106
1661
1164
407
66.0
5.3
8.1
28.4
90.3
93.5
865
1110
640
228
72.8
6.8
10.9
50.7
78.9
116
1236
1785
890
297
2446
4695
3135
4544
5975
1.08
6725
1.21
5205
1.37
6327
1.43
C-30
-------
TABLE C-15. LIQUID SAMPLES TAKEN ON 8/16/83 - GOLDEN WEST
REFINERY, SANTA FE SPRINGS, CALIFORNIA
COD Oil/grease TOC TCO
mg/L mg/L rog/L mg/L
Liquid Composite Samples
lAF-in
lAF-out
API- in
2,323
909
2,020
11.31
21.89X
23.37
— 104.46
— 40.78
— 25.64
Volatile Organic Samples
lAF-in VOA (0805) — —• 344
lAF-in VOA (1400) — — 411
lAF-out VOA (0805) — — 237
lAF-out VOA (1400) — — 304
(continued)
C-31
-------
TABLE C-15. LIQUID SAMPLES TAKEN ON 8/16/83 - GOLDEN WEST REFINERY,
SANTA FE SPRINGS, CALIFORNIA (CONTINUED)
Compound mg/1
Liauiti Composite Samples
Toluene 7.611
IAF Influent C8 5.581
Cio 28^782
C10 8.904
6.967
11.572
12.999
3.990
6.041
11.920
5.032
C;T 229.816
C18 60.938
C19 65.569
C20 34.653
C21 34.247
24.253
r.r- r^i * Toluene 3.721
IAF Effluent Cg I.Q4I
C9 o!899
C9 21.115
C10 6.998
Cjo 13.501
1.888
Toluene 2.546
API Influent C8 1.632
C9 5.749
CJO 3.522
Cjo 4.173
C12 2.765
CJ3 2.646
C14 1.699
C1S 2.621
1.395
65.244
C-32
-------
TABLE C-16. LIQUID SAMPLES TAKEN ON 8/17/83 - GOLDEN WEST REFINERY,
SANTA FE SPRINGS, CALIFORNIA
Volatile Organic Samples
lAF-in VOA (0740)
lAF-in VOA (1300)
lAF-out VOA (1300)
lAF-out VOA (0740)
COD Oil/grease TOC TCO
mg/L mg/L mg/L rag/L
Liquid Composite Samples
lAF-in
lAF-out
API- in
4,089
2,328
5,628
14.09
4.59
17.62
— 158.5
— 109.32
— 244.30
554
426
323
137
(continued)
C-33
-------
TABLE C-16.
LIQUID SAMPLES TAKEN ON 8/17/83 - GOLDEN WEST REFINERY
SANTAFE SPRINGS, CALIFORNIA (CONTINUED)
Compound
mg/1
Liquid Composite Samples
To! uene
CT
Cg
IAF Influent £«
C8
Cg
Cg
Cg
C9
C9
C9
C9
C9
C9
C9
cj
C10
C10
CIQ
Cn
c
Cn
c"
C12
C12
C12
C12
Cia
C13
Cl4
Q
Cis
C15
Cie
cl?
CIT
76.223
1. 835
3.602
2.422
2.066
5.420
17.959
6.712
3.833
1.632
2.160
2.644
3.057
4.577
2.640
5.201
5.709
3.968
8.078
11. 172
4.848
2.108
3.772
1.906
1.556
2.039
7.783
2.979
2.162
2.496
13.111
14.532
7.058
3.105
4.510
3.376
10.791
4.026
5.481
2.347
91.409
224.621
(CONTINUED)
C-34
-------
TABLE C-16. LIQUID SAMPLES TAKEN ON 8/17/83 - GOLDEN WEST REFINERY,
SANTE FE SPRINGS, CALIFORNIA (CONTINUED)
Compound
Cj.8
c"
C20
C22
C23
£1
Toluene
IAF Effluent C7
c?
c?
Cg
Cg
cj
Cg
Cg
C9
C9
C9
C9
C9
C9
C9
Cio
CIQ
Cio
CIQ
CIQ
Cio
cJi
c"
Cn
Cn
Cn
Cn
mg/1
87.140
84.054
110.444
73.046
90.032
73.718
46.656
55.906
30.594
50.025
0.482
0.516
0.957
0.688
0.563
2.543
10.277
3.919
1.296
0.628
0.618
1.126
1.611
2.743
1.290
30.117
2.226
2.117
0.971
0.588
0.889
9.658
20.001
2.108
0.666
1.663
2.282
0.674
2.144
0.726
0.916
0.681
1.092
:n 2.921
C-35
-------
TABLE C-16. LIQUID SAMPLES TAKEN ON 8/17/83 - GOLDEN WEST REFINERY
SANTA FE SPRINGS, CALIFORNIA (CONTINUED)
Compound mg/1
Cl2
C12
Cl2
Cl3
Cl3
Cl3
Cl4
Cis
Cl6
Cl7
Cl*
C2O
C2!
C23
C24
API Influent Toluene
C7
C7
C8
C8
c«
Co
C9
C9
C9
C9
C9
Cio
Cio
Cio
Cio
Cio
Cio
(CONTINUED)
C-36
1.337
1.231
1.445
7.804
8.226
1.390
1.850
2.598
1.808
5.846
2.174
84.094
105. 381
39.690
50.973
36.077
29.241
20.598
23.798
14. 621
23.873
1/593
2.085
2.157
5.764
24. 131
9.263
2.470
3.303
4.726
6.821
3.696
1.205
4.956
9.215
5.188
2.297
2.867
1.772
8.807
4.265
2.081
-------
TABLE C-16. LIQUID SAMPLES TAKEN ON 8/17/83 - GOLDEN WEST REFINERY
SANTA FE SPRINGS, CALIFORNIA (CONTINUED)
Compound mg/1
Cai 3.670
Cn 1.726
Cn 3.837
Cn 4.H6
Cn 1.931
Cu 1.812
Cn 5.883
Cn 2.842
Cn 6.898
C12 2.667
C12 3.212
C12 3.528
C12 2.250
C12 15.183
C12 15.331
C13 7.276
C14 15.577
C14 7.765
C15 3.512
C16 63.229
C17 180.452
Ci« 86.216
C-37
-------
TABLE C-17. LIQUID SAMPLES TAKEN ON 8/18/83 - GOLDEN WEST REFINERY
SANTA FE SPRINGS* CALIFORNIA
COD Oil/grease TOC TCO
mg/L mg/L mg/L mg/L
Liquid Composite Samples
lAF-in
lAF-out
API-4 (1130)
1,162
1,111
1,364
31.83
16. 71
15.16
— 46.48
— 34. 34
— 36.04
Volatile Organic Samples
UF-in VOA (1050) — — 204 —
lAF-in VOA (1500) — — 283 —
lAF-out VOA (1050) — — — —
lAF-out VOA (1500) _ — 315 —
(continued)
C-38
-------
TABLE C-17.
LIQUID SAMPLES TAKEN ON 8/18/83 - GOLDEN WEST REFINERY
SANTA FE SPRINGS, CALIFORNIA
Compound
mg/1
IAF Influent
IAF Effluent
API Influent
To!uene
Ca
Co
C10
C10
els
C18
To!uene
C8
Cs
C*
C10
Cto
Toluene
C8
9.752
4.435
1.832
1.299
22.145
7.012
14.987
2.081
1.203
29.697
5.949
2.174
1.071
16,975
5.575
10.822
0.853
5.477
2.531
0.971
1.052
17.101
5.889
12.505
1.399
0.976
25.959
C-39
-------
TABLE C-18, LIQUID SAMPLES TAKEN ON 8/19/83 - GOLDEN WEST REFINERY
SANTA.FE SPRINGS, CALIFORNIA
COD Oil/grease TOC TCO
•g/L rog/L ng/L ng/L
Liquid Composite Samples
lAF-in 1,194 348 —
lAF-out 830 332 —
960 20* —
API-in 3,482 1,321 —
Volatile Organic Samples
IAF-1n VOA (0830) — — 289
lAF-in VOA (1400) — — 509
lAF-out VOA (0830) — — 293
lAF-out VOA (140) — —
C-40
-------
o
C.I.3 Phillips Petroleum Company - Sweeny, Texas
The refinery wastewater system at Phillips consists of two separate
oil-wastewater separation facilities. Wastewater generated in the older
sections of the refinery is first treated by dual API separators which are
followed by a dissolved air flotation system. Wastewater generated by the
new process units is treated in three corrugated plate interceptor (CPI)
type separators which are followed by two IAF systems. The VOC emission
tests were conducted on the two IAF systems.
The IAF systems operate in parallel and are identical in size and
structure. Both are designed to be operated gas tight, and each has eight
access doors located on the sides of the units. In order to test VOC
emissions from the two systems, the access doors were tightly secured. A
steady air flow was introduced into the units using a blower. An outlet
location was fabricated so that continuous monitoring of the VOC
concentrations from the IAF could be measured. Figures C-5 and C-6 show the
IAF systems and sample locations.
EPA Method 25A was used to measure VOC concentrations from the IAF
systems. A summary of the results are shown in Table C-19. The total
hydrocarbon measurements include methane. In addition, gas chromatography
(EPA Method 18) was used to identify the major volatile components of the
vent stream. The gas chromatography results are shown in Table C-20 for the
south IAF system and in Table C-21 for the north IAF system.
In addition to the gaseous samples taken at Sweeny, liquid samples of
wastewater going to and from the CPI separators and IAF systems were
obtained. As with the samples acquired at Chevron and Golden West, these
samples were analyzed for COD, oil and grease, TOC, and TCO. The results of
the analyses are shown in Table C-22 to C-25.
C.2 VOC SCREENING OF PROCESS DRAINS
Process drains at three refineries were screened using a portable VOC
analyzer (Century Systems OVA-108). Process drains were screened at
Phillips Petroleum in Sweeny, Texas, Golden West in Santa Fe Springs,
California, and Total Petroleum in Alma, Michigan.
(Text continues on Page C-52)
C-41
-------
TOP VIEW
•IOUKICM. TKATUNT
UTIHAHD MC IMVU POINT
i—ii—ii i en
INTEiMTfO MS SMTU MINT
IU t I - SOUTH
WATIO SAMPLE LINES FOR
CONTINUOUS TNC ANALYZERS
SIDE VIEW
1 n
t m
1
1
**• «»
r~
^
1 1
«_
r — "-
ii
. „ - -
IDf
— i K
rt
i
|
EDO VIEM
K
T X
V*
u \
Figure C-5. Schematic representation of the IAF process with sample points
and induced air system: Phillips Petroleum - Sweeny, Texas.
-------
END VIEW
4- REIOUCT TO NOOtTER.
THEN TO EXHAUST
-£»
CO
FAUICATEO KTM. KOUClMg
IN ma.
IAFUNIT
Figure C-G. lAF-outlet sample locations fabricated:
Phillips Petroleum - Sweeny, Texas.
-------
TABLE C-19. DAILY EMISSION RATE AVERAGES AT IAF OUTLETS -
PHILLIPS PETROLEUM, SWEENY, TEXAS
Test Day Average Emission Rate
(Ib/hr Total Hydrocarbon as C,Hg)
IAF fl IAF #2
8/15/83 0.51 a
8/16/83 0.47 0.34
8/17/83 0.71 0.54
8/18/83 0.93 0.80
8/19/83 0.36 0.42
aIAF #2 not on-line for monitoring on 9/19/83.
C-44
-------
TABLE C-20. GAS CHROMATOGRAPHY RESULTS FROM IAF #1 (SOUTH IAF) -
PHILLIPS PETROLEUM, SWEENY, TEXAS
DATE
TIME
ANALYTICAL RESULTS
(ppmv as compound)
C-l
C-2
C-3
C-4
C-5
Hexane
Benzene
Heptane
Toluene
m-Xylene
o-Xylene
TOTAL HYDROCARBON
(ppmv as compound)
CONTINUOUS MONITOR DATA
Hydrocarbon Level
(ppmv as C3H8)
9/20/83
1500
87.2
4.9
6.7
18.4
20.4
145.3
161.1
25.9
139.4
45.4
20.7
675.4
1834
Emission Rate
(Ib/hr) (Total Hydrocarbon)0.72
9/20/83
1645
57.7
—
4.2
11.7
17.6
85.9
99.0
16.8
95.2
34.2
12.4
434.7
1577
0.62
9/21/83
1100
65.1
4.3
3.9
15.2
20.3
110.0
135.2
37.0
94.1
33.3
10.3
528.7
1625
0.67
9/21/83
1430
57.5
6.0
4.7
1.1
3.9
63.6
95.1
21.1
67.0
21.1
8.5
349.6
1508
0.62
(continued)
C-45
-------
TABLE C-20. GAS CHROMATOGRAPHY RESULTS FROM IAF #1 (SOUTH IAF)
PHILLIPS PETROLEUM, SWEENY, TEXAS (COMTIMUED)
DATE
TIME
ANALYTICAL RESULTS
(pp»v as compound)
C-l
C-2
C-3
C-4
C-5
Hexane
Benzene
Heptane
Toluene
orXylene
o-Xylene
9/22/83
0930
218.2
6.2
5.6
21.2
52.4
352.2
353.4
—
217.4
118.4
43.2
9/22/83
1430
197.5
5.7
6.0
15.5
16.2
213.5
201.1
78.7
140.2
62.4
18.9
9/23/83
0915
115.7
4.0
2.7
4.6
10.5
41.3
60.9
20.2
53.7
26.2
10.0
TOTAL HYDROCARBON
(ppav as compound) 1388.2 955.7 349.8
CONTINUOUS MONITOR DATA
Hydrocarbon Level
(ppav as C3HS) 3358 2087 1199
Emission Rate
(Ib/hr) (Total Hydrocarbon)!. 41 0.87 0.41
C-46
-------
TABLE C-21. GAS CHROMATOGRAPHY RESULTS FROM IAF #2 (NORTH IAF)
PHILLIPS PETROLEUM, SWEENY, TEXAS
DATE
TIME
ANALYTICAL RESULTS
(ppmv as compound)
C-l
C-2
C-3
C-4
C-5
Hexane
Benzene
Heptane
Toluene
w-Xylene
o-Xylene
TOTAL HYDROCARBON
(ppmv as compound)
CONTINUOUS MONITOR
Hydrocarbon Level
(ppmv as C3H8)
Emission Rate
(lb/hr)(Total
Hydrocarbon)
9/21/83
0930
58.7
4.2
4.4
17.5
21.5
128.5
134.3
35.9
84.0
26.1
8.1
523.2
DATA
1739
0.55
9/21/83
1545
78.6
7.5
5.9
22.6
10.5
133.7
171.8
46.6
116.5
43.9
13.6
651.2
2319
0.74
9/22/83
1050
226.2
7.3
5.6
21.5
59.5
292.5
287.0
113.1
178.2
73.9
20.0
1284.8
3428
1.11
9/22/83
1550
167.2
3.8
3.6
8.6
7.7
109.7
122.4
50.2
96.5
46.9
14.5
631.1
2892
0.94
9/23/83
1015
93.0
3.4
2.2
3.5
8.9
33.1
53.4
20.3
52.2
26.1
8.5
251.2
1278
0.52
aIAF #2 not monitored on 9/20/83 during Run No. 1 and Run No. 2.
C-47
-------
TABLE C-22. LIQUID SAMPLES TAKEN ON 9/20/83 -
PHILLIPS PETROLEUM, SWEENY, TEXAS
Liquid Composite and Grab Samples
IAF #2-out-D
IAF fl-out-C
IAF- inlet- A1
CPI-3-in (1700)
CPI-2-out (1700)
CPI-2-out (1700)
CPI-3-in (1700)
Void of Air Samples
COO
mg/1
539.3
628.4
4221.8
2061.4
681.2
2267.1
2810. 7
Oil /grease
mg/1
40.6
150.1
3059.5
1065.1
69.6
121.0
339.9
TOC
mg/1
CPI-2-out (1813)
IAF f2-out-C (1830)
CPI-3-in (1700)
lAF-in-A (1830)
IAF f2-out-C (1030)
CPI-2-in (1700)
lAF-in-A (1030)
CPI-l-in (1700)
CPI-3-out (1700)
IAF f2-out-D (1830)
IAF #l-out-C (1030)
502.5
308.5
205
478.5
107
664.5
358
478.5
204
138
229.5
C-48
-------
TABLE C-23. LIQUID SAMPLES TAKEN ON 9/21/83 -
PHILLIPS PETROLEUM, SWEENY, TEXAS
COO
wg/1
Oil /grease
mg/1
TOC
mg/1
Liquid Composite and Grab Samples
CPI-3-out (0930)
CPI-2-in (0945)
CPI-l-in (0945)
lAF-in-A1
IAF #2-out-D
IAF #l-out-C
CPI-2-inlet (0945)
CPI-1-out (0930)
CPI-3-out (0930)
Void of Air Samples
CPI-l-in (1600)
CPI-3-in (1600)
CPI-2-in (1600)
CPI-2-out (1600)
CPI-3-out (1600)
CPI-1-out (1600)
CPI-2-inlet (0945)
IAF #2-out-D (1445)
IAF fl-out-C (0855)
CPI-1-inlet (0945)
lAF-in-A (0855)
IAF *2-out-D (0855)
CPI-2-outlet (0930)
CPI-3-outlet (0930)
CPI-3-inlet (0945)
IAF fl-out-C (1445)
lAF-in-A1 (1445)
CPI-1-outlet (0930)
1991.0
2149.1
2697.8
1476.6
2300.7
1369.5
1042.7
2114.8
2395.0
269.6
267.4
687.7
126.0
34.2
58.0
40.5
168.3
209.4
310
259
250
157.5
198
549
36
218.5
129.5
155.5
237
226.5
223.5
194.5
451.5
242
278
262.5
C-49
-------
TABLE C-24. LIQUID SAMPLES TAKEN ON 9/22/83
PHILLIPS PETROLEUM, SWEENY, TEXAS
COD
mg/1
Oil/grease
mg/1
TOC
mg/1
Liquid Composite and Grab Samples
CPI #3-outlet (0930) 3000.5 232.5
lAF-in-A1 2941.7 262.8
IAF-*l-out-C 1312.9 152.3
CPI-fl-inlet (0940) 1811.2 32.1
CPI-*l-outlet (0930) 3400.2 705.3
CPI-#3-inlet (0940) 2290.5 31.7
CPI-#2-inlet (0940) 2065.1 34.8
CPl-#2-out1et (0940) 5045.2 4293.6
IAF-#2-out-D 1140.3 74.4
Void of Air Samples
CPI-f3-out1et (0920) 192-5
IAF-#2-out-D (0920) 41°
CPI-f2-outlet (1600) 80
CPI-f2-inlet (1600) 199-5
CPI-*2-inlet (0930) 302-5
lAF-fl-out-C (0920) 366
lAF-fl-out-C (1600) 688-5
lAF-in-A1 (0920) 531-5
CPI-fl-outlet (0920) 146-5
CPI-f2-outlet (0920) 194-5
CPI-fl-inlet (1600) 166
lAF-in-A1 (1600) 274
CPI-f3-outlet (1600) 242'5
IAF-f2-out-D (1600) 335
CPI-fl-outlet (1600) 396
CPI-#3-imet (1600) 210'5
CPI-fl-lnlet (0930) 297
CPI-#3-Inlet (0930) 208
C-50
-------
TABLE C-25.
LIQUID SAMPLES TAKEN ON 9/23/83
PHILLIPS PETROLEUM, SWEENY, TEXAS
Liquid Composite and Grab Samples
CPI-#3-outlet (1000)
lAF-in-A1
CPI-#l-1nlet (0930)
CPI-#2- Inlet (0930)
CPI-#3-outlet (1000)
CPI-#3-1nlet (0930)
CPI-#l-out1et (1000)
CPI-#2-outlet (0930)
IAF-#2-out-D
IAF-#l-out-C
Void of Air Samples
CPI-#3-in (1000)
CPI-#l-outlet (1000)
lAF-in-A1 (0900)
CPI-#2-outlet (1000)
IAF-#2-out-D (0900)
IAF-#l-out-C (0900)
CPI-#3-outlet (1000)
CPI-#l-in (1000)
CPI-#2-in (1000)
COO
mg/1
1503.3
160.9
1604.4
29194
1352.2
1135.2
2230.3
2354.4
1927.6
1910.7
Oil/grease
mg/1
469.4
250.0
107.4
10617
90.0
48.3
405.6
336.2
21.2
26.6
TOC
mg/1
204.5
105
224.5
444.5
248
225.5
251
107
153 5
XW J .
-------
At Phillips Petroleum, the process drains are sealed with steel caps.
The caps have a handle for manual removal and rest on supports over the
drain inlet. The drain inlet consists of a circular sump about 6-8 inches
deep and about 12 inches in diameter. Within the sump is the opening of the
vertical drain pipe which connects below grade to the drain line for the
process unit. A water seal is formed between the inside annulus formed by
the drain pipe and the side of the cap, and the cap side and circular watts
of the sump.
Screening values were taken at each drain while the drain was capped.
These screening values represent emissions from controlled drains. The caps
were then removed and left off for a period of time. The screening values
recorded after the cap had been removed for a period of time represented
emissions from uncontrolled drains. Only drains that were properly sealed
and maintained were included in the analysis.
The screening values of the controlled and uncontrolled drains can be
converted to leak rates (Ibs VOC/hr) using the correlation established in a*
EPA study of atmospheric emissions from petroleum refineries. This
correlation is as follows:
Log1Q (Non Methane Leak) = -4.0 + 1.10 Log1Q (Max. Screening Value)
A summary of the screening values is given in Table C-26.
Process drains were also screened at Golden West (Santa Fe Springs,
California) and Total Petroleum (Alma, Michigan). The process drains at
Golden West are designed with water seals. However, it was difficult to
determine if the water seals were being maintained at the time of the
screening. The process drains at Total Petroleum were not sealed.
Summaries of the screening results from these refineries are given in
Tables C-27 and C-28.
C-52
-------
TABLE C-26. SUMMARY OF EMISSION RATES AND EMISSION REDUCTION FOR DRAINS WITH A LEAK RATE >100 PPM
o
I
en
CO
Drain
Unit No.
27.1 6
7
17
26.2 3
27.2 1
2
3
11
12
25 11
19
23
69
83
84
85
86
94
Screening Values
Cap On Cap Off*
12
10
10
4
40
2,000
7
50
40
10
8
120
20
12
7
70
70
1,000
8
1,000
100
120
100
110
1,750
300
300
400
178
300
400
120
150
200
100
300
1,500
150
Estimated
Emission Rate, LB/HR
Cap On Cap Off*
0.00019
0.00016
0.00016
0.00005
0.00073
0.05384
0.00011
0.00083
0.00016
0.00012
0.00244
0.00034
0.00019
0.00011
0.00135
0.00135
0.02512
0.00012
0.08737
0.02512
0.00200
0.00244
0.00200
0.00222
0.04649
0.00668
0.00792
0.00376
0.00668
0.00917
0.00244
0.00312
0.00428
0.00200
0.00668
0.03924
0.00312
0.17536
Est. Emission
Reduction
LB/HR %
0.02493
0.00184
0.00228
0.00195
0.00149
-0.00735
0.00657
0.00709
0.00360
0.00656
0.00673
0.00210
0.00293
0.00417
0.00065
0.00533
0.01412
0.00300
0.08800
99.2
91.8
93.4
97.5
66.9
-15.8
98.4
89.5
97.3
98.2
73.4
86.1
93.8
97.4
32.5
79.8
36.0
96.2
*00
-------
TABLE C-27. SUMMARY OF PROCESS DRAIN SCREENING - GOLDEN WEST REFINERY,
SANTA FE SPRINGS, CALIFORNIA
Drain
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
Total Drains Screened
Average Screening Value
Avg. Non-Methane Leak Rate
Screening Value (ppmv)
30
30
70
20
15
15
20
10
10
70
-
700
15
30
70
10
20
10
50
.
-
> 10 ,000
>10,000
>10,000
300
200
50
700
500
1,000
30
150
-
>10,000
20
15
20
15
10
80
20
20
40
50
10
10
15
10
10
49
725
= 0.14 Ibs VOC/hr
C-54
-------
TABLE C-28. SUMMARY OF PROCESS DRAINS SCREENING
TOTAL PETROLEUM, ALMA, MICHIGAN
Drain
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
Total Drains Screened
Average Screening Value
Avg. Non-Methane Leak Rate
Screening Value (ppmv)
800
0
0
120
260
0
0
—
0
180
>10,000
>10,000
4,500
1,000
>10,000
>10,000
0
0
640
450
3,500
>10,000
0
3,000
60
1,000
10
10
10
50
3,500
150
10
10
10
10
10
10
10
10
10
10
10
100
600
10
50
200
48
= 1470
= 0.30 Ibs VOC/hr
C-55
-------
C.3 References
1. Stackhouse, C. and M. Hartman. Emission Test Report Petroleum Refinery
Wastewater Treatment System, Chevron U.S.A., Incorporated (El Segundo,
California). TRW Environmental Operations. Research Triangle Park,
North Carolina. EMB Report No. 83WWS2. March 1984.
2. Stackhouse, C. and M. Hartman. Emission Test Report Petroleum Refinery
Wastewater Treatment System, Golden West Refining Company (Santa Fe
Springs, California). TRW Environmental Operations. Research Triangle
Park, North Carolina. EMB Report No. 83WWS4. March 1984.
3. Stackhouse, C. and M. Hartman. Emission Test Report Petroleum Refinery
Wastewater Treatment System, Phillips Petroleum Company (Sweeny,
Texas). TRW Environmental Operations. Research Triangle Park, North
Carolina. EMB Report No. 83WWS3. March 1984.
C-56
-------
PETROLEUM REFINERY WASTEWATER TREATMENT SYSTEMS
APPENDIX D: EMISSION MEASUREMENT AND CONTINUOUS MONITORING
0.1 INTRODUCTION
This appendix describes the measurement method experience that was gained
during the emission testing portion of this study, the potential continuous
monitoring procedures, and the recommended performance test procedures. The
purpose of this appendix is to define the methodologies used to collect the
data to support a new source performance standard, to recommend procedures to
demonstrate compliance with a standard, and to describe alternatives for monitoring
either process parameters or emissions to indicate continued compliance with a
standard.
0.2 EMISSION MEASUREMENT EXPERIENCE
The purpose of the field study in this project was to provide estimates of
the organic compound release rates from several types of devices used in
wastewater treatment plants. There was insufficient information available to
estimate the uncontrolled volatile organic compound emission rate from induced
air flotation devices, dissolved air flotation devices, and equalization basins.
Testing was performed at three refineries that use these devices. However,
the true "uncontrolled" emission rate could not be measured because none of the
devices were open directly to the atmosphere. All of the devices were equipped
with a cover, and four of the six devices tested were equipped with an add-on
emission control system. These devices were selected for testing because the
D-l
-------
organic compounds released from the wastewater in the device were or could be
collected in a duct or'vent and the mass flow rate could be measured. This
approach was used to estimate what the emission rate would have been from an
uncovered device because of the difficulty of measuring a dispersed fugitive
emission. It is necessary to assume that the dominant factors affecting
the organic emission rates from these type devices are wastewater and device-
related, and that meteorological variables such as air temperature and wind speed
are secondary parameters.
Tests were conducted at one dissolved air flotation (OAF) unit, three
induced air flotation (IAF) units, and one equalization basin. These tests
included measurements of the gaseous flow rate and organic content, and
various tests to characterize the wastewater organic content before and after
the treatment units. Screening surveys were conducted on the drain systems in
various process units at three refineries to estimate the occurrence of the
fugitive emissions for various drain designs. Emission rate measurements were
not made for drains, junction boxes, oil/water separators, and uncovered or
open primary or secondary treatment processes.
D.2.1 Air Flotation and Equalization Basin Tests
The procedures used to characterize the emissions prior to control at the
two types of air flotation devices and the covered equalization basin were similar
and are discussed below in terms of the parameters that were measured.
D.2.1.1 Vent Gas Flow Rate
At the dissolved air flotation unit, the equalization basin, and one of
the induced air flotation devices, the covered head spaces were ventilated by
induced draft blowers. At the units with relatively high flow rates, EPA
D-2
-------
Method 2^1) was used to measure the gas velocity. This method is based on
the use of a pi tot tube, to traverse the flow area to calculate an average gas
velocity. The gas density was calculated based on a fixed gas (02, C02, N2, CO)
analysis by gas chromatography with thermal conductivity detection. Using
the duct area, the gas volumetric flow rate was calculated. Since the blowers
operated at constant speed with no changes in the ventilation area, the
measured flows were relatively constant. No problems were experienced using
Method 2 at these sources.
At one IAF that was equipped with an induced draft blower, the flow rate
was expected to be too low to measure with a pi tot tube, so a positive
displacement volumetric flow meter was installed. This procedure is essentially
EPA Method 2A. Due to a small pressure head and large amounts of water condensate,
the flow meter approach did not work. At another IAF where no induced blower
was used, a similar volumetric flow meter (a turbine meter) was installed. Itr
was found that the actual flow was less than the minimum rating of the smallest
meter that was commercially available.
The procedure finally used at these two sites was to construct a flow
meter system using a vane anemometer in a housing of the same diameter. This
system routed all of the vent stream through the anemometer at velocities
sufficient to be detectible by the anemometer, with a negligible meter pressure
differential. This measurement system is described in more detail in Reference 2.
The final type flow measurement was at an induced air flotation unit that
normally did not have an induced or a forced ventilation system. The inspection
doors on the unit cover were temporarily sealed and a portable blower was used
to establish positive ventilation. Flow measurements were made using the
TTTSee Reference 1.
D-3
-------
anemometer system described above. No problems were encountered in the actual
measurement of the flow rate, but it was found that the doors could not be
perfectly sealed and that the flow supply and exhaust rates had to be measured
to account for the leakage at the doors.
In summary, it was found that for systems equipped with large capacity
blowers, EPA Method 2 (pitot tube traverses) can be used successfully to
determine volumetric gas flow. Where there is no forced ventilation or the
ventilation rate is deliberately maintained at low levels, large volumes of
condensate can be present, low pressure heads may not drive a flow meter, and
the flow rate may be below the range of commercially available volumetric flow
meters. These conditions existed at several facilities and commercially available
meters could not be used. A fabricated meter based on an anemometer normally
used for low velocity air flows was used with success at these difficult sources.
D.2.1.2 Total Organic Concentration Measurement
Procedures similiar to EPA Method 25A were used to measure the total
organic or hydrocarbon concentration in the vent stream. A sample was
continuously withdrawn from the vent stream through a heated Teflon® sample
line to a flame ionization analyzer. Propane in nitrogen mixtures were used to
calibrate the analyzers. For aliphatic and aromatic hydrocarbons, such as are
expected at a refinery, the total instrument response is relatively proportional
to carbon content and can be used as a measure of total hydrocarbon concentration.
The result of this measurement is a gaseous hydrocarbon equivalent concentration
as propane. The molar density of propane was used to calculate a mass per unit
volume result.
D-4
-------
The analyzers were zeroed and calibrated with propane standards before,
during, and after testfng each day. For those systems that operated continuously
during a multiple-day test, calibrations were performed at 4- to 8-hour intervals.
The zero and calibration drifts were within the acceptable range in Method 25A.
The only problems encountered with the use of this method was the eventual
condensation of high molecular weight organic aerosols in the instruments which
led to instability, noise, and flameout. When these conditions occurred, the
instruments had to be purged with clean air until the signal stabilized. This
problem was minimized when an instrument equipped with a totally heated enclosure
was used.
0.2.1.3 Gaseous Organics Speciation
Gas chromatographic techniques were used to identify the major volatile
components of the vent streams prior to control. The basic techniques described
by EPA Method 18 were used. An integrated sample was collected into an inert,
flexible plastic bag and these samples were analyzed by two chromatograph systems,
The purpose of these determinations was to identify the major components and to
estimate an average flame ionization response factor to evaluate the carbon
proportionality of the total hydrocarbon analyzer result.
One of the gas chromatograph systems was used to separate methane through
pentane. The calibration mixture for this analyzer consisted of Cj. - C5 species
so that specific identification and quantification was possible. The second
system was used to separate higher boiling point compounds in the range of C6
to C9. Benzene and m-xylene were used as calibration species. Specific identi-
fication and quantification was possible for these two compounds. The other
D-5
-------
compounds were identified by retention time and quantified by using the closer
(benzene or xylene) calibration factor based on the number of carbon atoms in
the molecule.
Mo specific problems were encountered in conducting these tests. The
collection of the samples into bags was straightforward. In some cases,
condensate was observed in the bags, but analysis of this material indicated
negligible organic content. The only uncertainty is whether or not any significant
amounts of compounds with a higher boiling point than Cg were present.
This is unlikely because of the relatively high boiling points of compounds
heavier than Cg, and the relatively low source temperatures.
D.2.1.4 Wastewater Sampling and Analysis
Water samples were collected before and after the wastewater treatment
devices that were tested in order to characterize the wastewater and to determine
if there were any simple tests that could be used as an indicator of expected
hydrocarbon emission rates.
Samples were collected using techniques similar to those used by the
refineries for process operation control. Composites were made from individual
grab samples taken periodically during in the test day. The composite sample
volume was approximately 1 gallon. The samples were stored and shipped on ice
to minimize the loss of volatile components. Additional samples were collected
into void-of-air (VOA) vials where all the head space could be eliminated to
obtain a sample for total carbon analysis.
Ho specific problems were encountered with the collection of samples from
flowing streams in pipes. Where samples had to be collected from a quiescent
pool (e.g., an API separator forebay), there is some uncertainty about the
D-6
-------
representativeness of a dipped grab sample. During sample shipment, several
of the void-of-air (VOA) sample vials were broken because of freezing. Since
no expansion area was left in the bottle, the container broke when the sample
remained in direct contact with ice for extended periods. Also, it is possible
that during a storage period of several weeks, coagulation and settling occurred
so that a homogenous mixture could not be regenerated for analysis. This problem
may not have occurred if the analysis had been performed within 1 day and the
samples could have been stored at nearly ambient conditions.
The water samples were analyzed for total organic carbon, chemical oxygen
demand, oil and grease, total chromatographical organics (organic speciation),
and volatile organics by a purge and trap technique.
Total organic carbon was determined using an automatic analyzer that
measures the carbon dioxide resulting from the photochemical oxidation of
organic carbon after the inorganic carbon has been removed by purging. This
procedure does not measure the volatile compounds that are removed by the purge
stream. Variation can also be caused by nonrepresentative collection of heavy
organics in the aliquot transfer syringe used to inject the sample into the
analyzer.
The chemical oxygen demand method is based on the quantity of oxygen
required to oxidize the organic matter in the sample under controlled conditions.
Organic and oxidizable inorganic carbon is measured. Volatile straight chain
aliphatics are not appreciably oxidized, partly due to their presence as
volatiles in the head space where they do not come into contact with the oxidizing
liquid.
D-7
-------
Oil and grease content was determined by a gravimetric determination of
fluorocarbon-113 extractible compounds. The solvent evaporation step of the
process removes short chain hydrocarbons and simple aromatics due to evaporation.
Total chromatographicable organics was performed by gas chromatography with
flame ionization detection. The sample was prepared by extracting the water
with methylene chloride and injecting the extract to the chromatograph. This
procedure allowed speciation of Cy to C25 compounds. A solvent volume reduction
step in the analysis tends to volatilize short straight chain aliphatics and
simple aromatics with a boiling point less than 100°C.
The purge and trap procedure used was EPA Method 624 (see Reference 5)
with component identification by mass spectrometry.
The results of all the analyses were highly variable from day-to-day. There
did not appear to be any one procedure that yielded consistently reasonable
results. These were also significant variations from the results obtained by
the treatment system operators for those parameters that were measured for
process control. The sample storage time and storing the sample on ice may
have contributed to the inconsistencies. Also, all of the routine procedures
that were performed tend to exclude the more volatile compounds from the result.
Because of these inconsistencies, it is not possible to determine if any of the
test procedures would yield results that would predict hydrocarbon emission
factors.
Further studies would be necessary to determine if the inconsistencies
were caused by field sampling, storage, or analysis techniques.
D.2.1.5 Process Drain Screening Surveys
Portable analyzers were used at three refineries to survey the unit drain
systems. The purpose of these surveys was to determine if there was a significant
D-8
-------
difference in the occurrence of fugitive emissions from drain systems of
different designs. EPA Method 21 techniques were used. The meter reading at
the centroid of the cross-sectional opening to atmosphere was recorded. A
leaking source was tentatively identified when the meter reading at the source
exceeded the ambient meter reading.
There were no problems encountered in conducting the field tests.
However, the identification of the source of some detected emissions was difficult.
In some cases it was found that the source of a detected emission was an open-
ended line that terminated at the drain, rather than from the underground drainage
system. Also, since the source of the detected emission was not necessarily con-
centrated or steady, the variability of a meter reading at a source was more
than was observed at other types of fugitive emission sources.
D.3 PERFORMANCE TEST METHODS
The specific combination of measurements that would be necessary to
demonstrate compliance depends on the format of a standard. The options
include specification of a VOC emission concentration limit, a VOC mass rate
limit, or a minimum VOC removal efficiency requirement. The procedures
recommended for determination of each of these values are described in this
section. The estimated cost of each type of performance test is also presented.
0.3.1 VOC Concentration Measurement
The recommended VOC measurement method is Reference Method 25A or 25B.
Method 25A, "Determination of Total Gaseous Organic Concentration Using a Flame
lonization Analyzer," applies to the measurement of total gaseous organic
concentration of vapors consisting of alkanes and aromatic hydrocarbons. The
D-9
-------
instrument is calibrated in terms of propane or another appropriate organic
compound. A sample is.extracted from the source through a heated sample line
and glass fiber filter and routed to a flame ionization analyzer (FIA). Provisions
are included for eliminating the heated sampling line and glass fiber filter
under some sampling conditions. Results are reported as concentration equivalents
of the calibration gas or organic carbon.
Method 25B, "Determination of Total Gaseous Organic Concentration Using
a Nondispersive Infrared Analyzer," is identical to Method 25A except that a
different instrument is used. Method 25B applies to the measurement of total
gaseous organic concentration of vapor consisting primarily of alkanes. The
sample is extracted as described in Method 25A and is analyzed with a non-
dispersive infrared analyzer (NDIR).
In both the FIA and NDIR analysis approaches, instrument calibrations are
based on a single reference compound. For refinery wastewater systems propane"
is the recommended calibration compound. As a result, the sample concentration
measurements are on the basis of that reference and are not necessarily true
hydrocarbon concentrations. Calculation of emissions on a mass basis will not be
affected because the response of the instruments is proportional to carbon content
for similar compounds, which in this case, are crude petroleum components. Mass
results would be equivalent using either the concentration and molecular weight
based on a reference gas or the true concentration and true average molecular
weight of the hydrocarbons. The advantage of using a single component calibration
is that chromatographic techniques are not required to isolate and quantify the
individual compounds present.
The VOC analysis techniques discussed above measure total hydrocarbons
including methane and ethane. Chromatographic analyses during prior field tests
have indicated that significant quantities of methane and ethane may sometimes be
D-10
-------
present in the vapors emitted. If it is expected that methane or ethane is
present in significant quantities, appropriate samples are required for
chromatographic analysis to adjust the results to a nonmethane-nonethane basis.
"Reference Method 18: Measurement of Gaseous Organic Compounds by Gas Chroma-
tography" would be applicable for this measurement.
0.3.2 Gas Flow Measurement
Reference Methods 2, 2C, 2A, and 20 are recommended as applicable for
measurement of gaseous flow rate. "Method 2: Determination of Stack Gas Velocity
and Volumetric Flow Rate (Type S Pi tot Tube)" applies when the duct or pipe
diameter is larger than 12 inches and the flow is constant and continuous.
"Method 2C: Determination of Stack Gas Velocity and Volumetric Flow Rate from
Small Stacks or Ducts (Standard Pi tot Tube)" applies when the duct diameter is
less than 12 inches and the flow is constant and continuous. "Method 2A: Direct
Measurement of Gas Volume Through Pipes and Small Ducts" applies to the measurement
of volumetric flow where a totalizing gas volume meter is installed in the duct
and a direct reading is obtained. This method can be used in the general
temperature range of 0-50°C, with a flow range dependent on the meter size.
Temperature and pressure measurements are made to correct the volume to standard
conditions. "Method 2D: Measurement of Gas Volume Flow Rates in Small Pipes and
Ducts" applies when Method 2A cannot be used because the vent size is too large
or when pressure drop restrictions prevent reducing the duct size to that of a
volumetric meter. This method incorporates the use of a device to measure gas
flow rate, such as an orifice, a venturi, or a rotameter. The flow rate is
integrated with time to compute an average volume flow. This method must be
applied with caution to intermittant or variable gas flow rates.
D-ll
-------
D.3.3 Mass Flow
The VOC concentration and volume measurements are combined to determine the
mass flow. To determine the total VOC mass during the entire test period, the
VOC mass flow is determined for small incremental periods; each 5-minute interval
and increment thereof when the processor is operating, and each 15-minute
interval and increment thereof during non-operation. These incremental flows are
then summed for the entire test period. Because VOC concentrations and flow rate
may vary significantly within a brief time period, these short incremental
calculation intervals are needed so that short-term variations in flow rates can
be properly weighted in the calculations.
D.3.4 Emission Reduction Efficiency Determination
The recommended procedures for determining the VOC concentration and gas
flow would be performed simultaneously at the control device inlet and outlet.
The measurements would be combined to compute a VOC mass flow before and after
the control device. The mass flows would be used to calculate a VOC removal
efficiency.
0.3.5 Performance Test Time and Costs
The length of a performance test is specified in the applicable regulation
and is selected to be representative for the process being tested. Wastewater
treatment operations are generally steady, although there may be periods where
intermittent high organic content wastes are treated. In general, a performance
test would consist of three to six runs, each lasting about 2 hours.
It is estimated that for most operations, the field testing could be
completed in 2 to 3 days (i.e., two or three 8-hour work shifts) with an extra
day for setup, instrument preparation, and cleanup.
D-12
-------
The cost of the testing varies with the length of the test and the number
of vents to be tested. The cost is estimated at $6,000 - $10,000 for VOC
concentration determination at one vent, and $12,000 - $15,000 for the
determination of VOC removal efficiency.
D.4 MONITORING SYSTEMS AND DEVICES
The purpose of monitoring is to ensure that the emission control system is
being properly operated and maintained after the performance test. One can either
directly monitor the regulated pollutant, or instead, monitor an operational
parameter of the emission control system. The aim is to select a relatively
inexpensive and simple method that will indicate that the facility is in continual
compliance with the standard.
The use of monitoring data is the same regardless of whether the VOC outlet
concentration or an operational parameter is selected to be monitored. The
monitor should be installed and operating properly before the first performance
test. Continual surveillance is achieved by comparing the monitored value of
the concentration or parameter to the value which occurred during the last
successful performance test, or alternatively, to a preselected value which is
indicative of good operation. It is important to note that a high monitoring
value does not positively confirm that the facility is out of compliance; instead,
it indicates that the emission control system is operating in a different manner
than during the last successful performance test.
Two types of emission reduction systems can be used to control vent streams
from covered water treatment devices. These are combustion and vapor processing.
Potential monitoring approaches for these control systems are discussed below.
D-13
-------
D.4.1 Monitoring of Vapor Processing Devices
There are presently no demonstrated continuous monitoring systems commercially
available which monitor vapor processor operation in the units of VOC removal
efficiency. This monitoring would require measuring not only inlet and exhaust
VOC concentrations, but also inlet and exhaust volumetric flow rates. An overall
cost for a complete monitoring system is difficult to estimate due to the number
of component combinations possible. The purchase and installation cost of an
entire monitoring system (including VOC concentration monitors, flow measurement
devices, recording devices, and automatic data reduction) is estimated to be
$25,000. Operating costs are estimated at $25,000 per year. Thus, monitoring in
the units of efficiency is not recommended due to the potentially high cost and
lack of a demonstrated monitoring system.
Monitoring in units of mass of VOC emitted would require measurements only
at the exhaust location, as discussed above. The cost is estimated at $12,000
for purchase and installation plus $12,500 annually for operation, maintenance,
calibration, and reduction.
Monitoring equipment is commercially available, however, to monitor the
operational or process variables associated with vapor control system operation.
The variable which would yield the best indication of system operation is VOC
concentration at the processor outlet. Extremely accurate measurements would not
be required because the purpose of the monitoring is not to determine the exact
outlet emissions but rather to indicate operational and maintenance practices
regarding the vapor processor. Thus, the accuracy of a FIA (Method 25A) type
instrument is not needed, and less accurate, less costly instruments which use
different detection principles are acceptable. Monitors for this type of continuous
VOC measurement, including a continuous recorder, typically cost about $6,000
D-14
-------
to purchase and install, and $6,000 annually to calibrate, operate, maintain, and
reduce the data. To achieve representative VOC concentration measurements at the
processor outlet, the concentration monitoring device should be installed in the
exhaust vent at least two equivalent stack diameters from the exit point, and
protected from any interferences due to wind, weather, or other processes.
The EPA does not currently have any experience with continuous monitoring of
VOC exhaust concentration of vapor processing units at wastewater treatment units
in petroleum refineries. Therefore, performance specifications for the sensing
instruments cannot be recommended at this time. Examples of such specifications
that were developed for sulfur dioxide and nitrogen oxides continuous instrument
systems can be found in Appendix B of 40 CFR 60.
For some vapor processing systems, there may be another process parameter
besides the exhaust VOC concentration which is an accurate indicator of system
operation. However, all acceptable process parameters for all systems cannot be
specified. Substituting the monitoring of vapor processing system process
parameters for the monitoring of exhaust VOC concentration is valid and acceptable
if it can be demonstrated that the value of the process parameter is an indicator
of proper operation of the processing system. Monitoring of any such parameters
would have to be approved by enforcement officials on a case-by-case basis.
Parameter monitoring equipment would typically cost about $3,000 plus $3,000
annually to operate, maintain, periodically calibrate, and reduce the data into
the desired format.
D-15
-------
D.4.2 Monitoring of Combustion Devices
0.4.2.1 Incinerators
Incinerators used to comply with a standard need to be maintained and operated
properly if the standard is to be achieved on a continuous basis. Continuous
inlet and outlet emission monitoring would be the preferred method of monitoring
because it would provide a continuous, direct measurement of actual emissions and
destruction efficiency. However, no continuous monitor measuring total VOC has
been demonstrated for incinerators controlling vent streams. Moreover, such a
monitoring system would be extremely complex and labor-intensive, and it would be
relatively expensive when two monitors are required to ensure that a certain
destruction efficiency is maintained.
The incinerator operating parameters that affect performance are temperature,
type of compound, residence time, inlet concentration, and flow regime. Of these
variables, the last two have the smallest impact on incinerator performance."
Residence time is essentially set after incinerator construction unless the vent
stream flow rate is changed. Moreover, at temperatures above 760°C, compound
type has little effect on combustion efficiency.
Test results and theoretical calculations show that lower temperatures can
cause significant decreases in control device efficiency. Test results also
indicate that temperature increases can also adversely affect control device
efficiency. In terms of cost, temperature monitors are relatively inexpensive,
costing less than $5,000 installed with strip charts, and are easily and cheaply
operated. Given the large effect of temperature on efficiency and the low cost
of temperature monitors, this variable is clearly an effective parameter to monitor.
Where a combustion device is used to incinerate waste VOC streams alone,
flow rate can be an important measure of destruction efficiency since it relates
directly to residence time in the combustion device. Flow rates of fugitive
D-16
-------
emission vent streams are typically small in comparison to other streams that may
be ducted to the same incinerator. As a result, flow rate may not always give a
reliable indication of the vent stream residence time in the incinerator. But an
indication of emission vent stream flow rate to the incinerator gives assurance
that VOC is being routed for proper destruction. Flow rate monitors, at an
estimated installed cost of less than $2,000, are inexpensive and easy to operate.
Therefore, since flow rate monitors give an indication that organics-laden streams
are being routed for destruction and since they are inexpensive, flow rate is
also an effective parameter to monitor for incinerators. Flow rate meters should
be installed, calibrated, maintained, and operated according to the manufacturer's
specifications and should be equipped with a continuous recorder. They should
have an accuracy of 5 percent of the flow rate being measured and should be
installed on combustion device inlets.
0.4.2.2 Boilers or Process Heaters
If an emissions vent stream is introduced into the flame zone of a boiler or
process heater, it is necessary to know that the boiler or heater is operating
and that the waste gas is being introduced into the boiler or heater. Maintenance
of records such as steam production records would indicate periods of operation.
Flow indicators could provide a record of flow of the vent stream to the boiler
or heater. For smaller heat producing units less than 44 MW (150 million Btu/hr
heat input), temperature should also be measured to ensure optimum operation.
Monitoring temperature for boilers or heaters with heat design capacities greater
than 44 MW would not be necessary. These larger units always operate at high
temperatures (>1100°C) and stable flow rates to avoid upsets and to maximize
steam generation rates. Maintenance of records that indicate periods of operation
would be sufficient for these larger boilers or heaters.
D-17
-------
D.4.2.3 Flares
Because flares are not enclosed combustion devices, it is not feasible to
measure combustion parameters. Moreover, temperatures and residence times are
more variable throughout the combustion zone for flares than for enclosed devices
and, therefore, such measurements would not necessarily provide a good indicator
of flare performance even if measurable.
The typical method of monitoring continuous operation of a flare is visual
inspection. However, if a flare is operating smokelessly, it can be difficult to
determine if a flame is present, and it may take several hours to discover. The
presence of a flame can also be determined through the use of a heat sensing
device, such as a thermocouple or ultra-violet (U-V) beam sensor on a flare's
pilot flame. If a flame is absent, the temperature probe can be used to alert
the plant operator. The cost of available thermocouple sensors ranges in price
from $800 to $3,000 per pilot. (The more expensive sensors in this price range
have elaborate automatic relight and alarm systems.) One drawback of thermocouples
is that they burn out if not installed properly. The cost of a U-V sensor is
approximately $2,000. However, the U-V system would not be as accurate as a
thermocouple in indicating the presence of a flame. The U-V beam is influenced
by ambient infrared radiation that could affect the accuracy. Interference
between different U-V beams would make it difficult to monitor flares with multiple
pilots. U-V sensors are designed primarily to monitor flames within enclosre
combustion devices. Therefore, thermocouples are a superior monitoring methodology
for flares. To ensure that a vent stream is being continuously vented to a flare,
a flow indicator can be installed on the vent stream.
D-18
-------
D.5 References
1. U.S. Environmental Protection Agency. Code of Federal Regulations.
Title 40, Part 60, Appendix A: Reference Methods. Washington, D.C.
Office of the Federal Register. July 1, 1983. p. 347-558.
2. Stackhouse, C. and M. Hartman. Emission Test Report Petroleum Refinery
Wastewater Treatment System, Chevron U.S.A., Incorporated (El Sequndo,
California). TRW Environmental Operations. Research Triangle Park,
North Carolina. EMB Report No. 83WWS. March 1984.
3. Stackhouse, C. and M. Hartman. Emission Test Report Petroleum Refinery
Wastewater Treatment System, Phillips Petroleum Company (Sweeny,
Texas). TRW Environmental Operations. Research Triangle Park, North
Carolina. EMB Report No. WWS3. March 1984.
4. Stackhouse, C. and M. Hartman. Emission Test Report Petroleum Refinery
Wastewater Treatment System, Golden West Refining Company (Santa Fe
Springs, California). TRW Environmental Operations. Research Triangle
Park, North Carolina. EMB Report No. WWS4. March 1984.
D-19
------- |