Preliminary Data Summary for the
Petroleum Refining Category
United States Environmental Protection Agency
Office of Water
Engineering and Analysis Division
401 M Street, S.W.
Washington, D.C. 20460
EPA821-R-96-015
April 1996
-------
Table of Contents
List of Figures iv
List of Tables v
Acknowledgments vii
1. Introduction 1
1.1 Background 1
1.2 Status of Categorical Regulations 1
1.3 Software Disk Available 4
2. Description of the Industry 5
2.1 Production Operations 5
2.1.1 Crude Oil and Product Storage 5
2.1.2 Crude Distillation 5
2.1.3 Cracking 6
2.1.4 Hydrocarbon Rebuilding 6
2.1.5 Hydrocarbon Rearrangements 6
2.1.6 Hydrotreating 6
2.1.7 Solvent Refining 7
2.1.8 Asphalt Production 7
2.1.9 Lubricating Oil Manufacture 7
2.1.10 Production of Petrochemicals 8
2.2 Industry Trends 8
3. Summary of Information Sources Used in This Study 13
3.1 Oil And Gas Journal Survey 13
3.2 EPA Office of Air and Radiation Questionnaire 13
3.3 Plant Visits 13
3.4 Permit Compliance System Data 13
3.5 Los Angeles County Sanitation Districts 14
3.6 Other Sewerage Authorities 14
3.7 Province of Ontario, Canada Petroleum Study 14
3.8 Other Data Sources 14
4. Treatment Technologies Used in The Industry 15
4.1 In-Plant Controls 16
4.1.1 Steam Stripping 16
4.1.2 Neutralization of Spent Acids and Caustics 18
4.1.3 Source Control 18
4.1.4 Wastewater S egregati on 18
4.1.5 Boiler Condensate Recovery 19
4.1.6 Treated Effluent Reuse 19
-------
4.1.7 Other General Measures 19
4.1.8 Cooling Water Systems 19
4.1.9 Once-Through Cooling Water Systems 20
4.1.10 Cooling Tower Systems 21
4.2 End-of-Pipe Treatment Technologies 21
4.2.1 Preliminary Treatment 21
4.2.2 Biological Treatment 22
4.2.3 Effluent Polishing 23
4.2.3 Activated Carbon Treatment 24
4.2.5 Technologies Used at EPA/OAR Survey Refineries 24
4.2.6 Performance of End-of-Pipe Systems 26
4.2.7 Storm Water Management 29
5. Water Use 30
6. Pretreatment Standards Review And Catalytic Reformer Issues 39
6.1 Indirect Discharging Refineries 39
6.1.1 Treatment Technologies 39
6.2 Dioxins in Catalytic Reformer Wastewaters 45
6.2.1 Waste Characterization 47
6.2.2 Available Treatment Technologies 54
7. Evaluation of Pollutant Discharges And Environmental Issues 58
7.1 Identification And Quantification of Pollutant Releases 58
7.1.1 Permit Compliance System 63
7.1.2 Estimation of Annual Pollutant Loads from PCS 64
7.1.3 Analysis of Average Measured Pollutant Concentrations from PCS 78
7.1.4 Toxic Release Inventory 83
7.1.5 Reported Annual Pollutant Loads from TRI 83
7.2 Fate and Toxicity Evaluation of Released Pollutants 88
7.2.1 Compilation of Physical-Chemical and Toxicity Data and Information to
Evaluate Indirect Discharges 88
7.2.2 Categorization of Pollutants 90
7.2.3 Toxic Weighting Factor Analysis 97
7.2.4 Whole Effluent Toxicity Testing 100
7.3 Documented Impacts 102
8. Economic Profile of the Petroleum Refining Industry 104
8.1 U.S. Petroleum Refinery Geographic Distribution and Trends 104
8.1.1 Number and Distribution of Refineries 104
8.1.2 Trends in the Number of Refineries 107
8.2 Economic Profile 109
8.2.1 The FRS Companies 109
11
-------
8.2.2 Refined Product Margins 112
8.2.3 Refined Products Imports 112
8.3 Impacts of Environmental Regulations 113
8.3.1 Air Pollution Abatement Expenditures 116
8.3.2 Water Pollution Abatement Expenditures 117
Bibliography 120
Index 123
in
-------
List of Figures
2.1 Number of Refineries per State 10
2.2 Number of Refineries Since 1982 12
5.1 Water Use Trends (Graph) 31
5.2 Comparison of Flow Predicted by BPT and BAT Models (Graph) 33
7.1 Geographic Distribution of 1991 TRI Chemical Releases and Transfers (Map) 85
7.2 1991 TRI Priority Pollutant Releases to Surface Water and POTWs (Graphs) 87
7.3 Ecological Impact Potential (Scatter Plot) 95
7.4 Human Health Impact Potential (Scatter Plot) 97
8.1 Refinery Pollution Abatement Operating Expenditures (Graph) 115
8.2 Refinery Pollution Abatement Capital Expenditures (Graph) 116
IV
-------
List of Tables
2.1 Capacity and Number of United States Refineries in 1991 vs. 1976 8
4.1 Demonstrated Wastewater Technologies for In-Plant Treatment of Refinery Process Streams
17
4.2 Advantages and Disadvantages of Different Types of Cooling Systems 20
4.3 Summary of Current Wastewater Treatment Technologies for 27 Refineries Surveyed 25
4.4 Summary of Effluent Data: Six Site Visits 27
4.5 Summary of Refinery Effluent Data: Canadian Study and PCS Data 28
4.6 Refinery Storm Water Management Practices 29
5.1 Predicted and Actual Wastewater Flows for 27 Refineries 32
5.2 Selected Sources of Wastewaters 36
5.3 Water Use: Six Site Visits 38
6.1 Summary Comparison of Locations and Capacities for Indirect Dischargers Between 1976
and 1991 40
6.2 Summary of Discharge Data for Major Refineries Discharging to Los Angeles County
Sanitation Districts (LACSD) 42
6.3 Summary of Discharge Data for Major Refineries Discharging to Local POTWs 43
6.4 Data Comparison: Indirect Discharging Refineries 45
6.5 Shell Canada Products Limited, Sarnia Refinery Range of Dioxins/Furans in Internal Shell
Wastewaters 48
6.6 Esso Petroleum Canada, Sarnia Refinery Powerformer Regeneration Study 48
6.7 Catalytic Reforming Data: Six Site Visits 50
6.8 Summary of CDD/CDF Data for Chevron Richmond Refinery 51
6.9 Summary of CDD/CDF Data for Tosco Martinez Refinery 52
6.10 Summary of CDD/CDF Data for Unocal Rodeo Refinery 53
6.11 Suncor-Sarnia Catalytic Reformer Wastewater Treatment 56
6.12 Shell Canada-Sarnia, Catalytic Reformer Wastewater Treatment Spent Caustic 57
7.1 Refinery Wastewater Constituents 58
7.2 1992 Annual Loading Data from PCS 65
7.3 Crude Production for 138 Direct Discharging Petroleum Refineries 70
7.4 Comparison of California and Non-California PCS Loads after Production Weighting
(Low-End Estimate) 75
7.5 Comparison of California and Non-California PCS Loads after Production Weighting
(High-End Estimate) 77
7.6 Comparison of California and Non-California PCS Concentration Data (Low End Estimate)
80
7.7 Comparison of California and Non-California PCS Concentration Data (High-End
Estimate) 81
7.8 Direct Discharge Pollutant Concentration Levels Reported in 1982 Effluent Guidelines
Development Document (Current/BPT) 82
7.9 POTW Information for Selected Indirect Discharges 89
-------
7.10 Number of Pollutants by Categorization Group 92
7.11 Number of Pollutants with Health Effect Designations 92
7.12 Petroleum Refining Annual Loads from PCS and TRI 100
8.1 Number and Capacity of Operable Petroleum Refineries 105
8.2 Number and Capacity of Refineries in California, Louisiana, and Texas 106
8.3 States with No Refining Capacity 107
8.4 Number of Operable Refineries 108
8.5 Income from Refining and Marketing Operations 110
8.6 Composite Refiner Margin Ill
8.7 Refined Product Import Volumes 112
8.8 National Petroleum Council Estimates of Incremental Cost to Meet the New CAAA
Requirements 116
8.9 National Petroleum Council Estimate of Emission Control Investment (by Emission Type)
117
8.10 National Petroleum Council Estimates of Incremental Cost to Meet the New Requirements
of Clean Water Act Reauthorization 118
8.11 National Petroleum Council Estimate for Water Pollution Control Investments 119
8.12 National Petroleum Council Projection for Water Pollution Control Investments by Time
Period 119
VI
-------
Acknowledgments
EPA prepared this report with contract support from Science Applications International
Corporation (SAIC) under the direction of Barry S. Langer, Project Manager with assistance by
K.C. Mahesh, Project Engineer.
Annette Huber and Ed Gardetto of EPA's Exposure Assessment Branch, Standards and
Applied Science Division, provided substantial input by analyzing Permit Compliance System
(PCS) data, estimating loadings, and writing Section 7 of the report on environmental issues.
Joe Ford and Bill Anderson of EPA's Economic and Statistical Analysis Branch,
Engineering and Analysis Division (EAD) wrote Section 8 (Economic Information). Marvin
Rubin, Chief, Energy Branch, EAD provided substantial comments on the report. Jim Durham
of EPA's Office of Air Quality, Planning and Standards (OAQPS) helped provide updated data
on wastewater flow by adding questions to the OAR Clean Air Act questionnaire. Ed McHam
and EPA Region 6 staff provided comments pertaining to the permitting program. Many thanks
to Dottie Grosse, Carol Swann and Eric Strassler of EAD for editing and word processing to
prepare the document for distribution.
EPA appreciates the efforts of the American Petroleum Institute (API) and Alison
Kerester for their assistance. Dave Pierce of the Chevron Corporation and the API Refinery
Effluent Task Force helped to set up a number of site visits as requested by EPA.
Paul Martyn and Brent Perry of the Los Angeles County Sanitation Districts (LACSD)
provided much of the data from refineries that discharge to POTWs. API, National Petroleum
Refiners Association, Citizens for a Better Environment, LACSD and staff from EPA's Region
VI and VUI offices provided comments on the draft report.
We also thank the staffs of refineries that granted EPA tours of their facilities and
provided requested data: Chevron refineries located in Richmond, CA and Philadelphia, PA;
Unocal, Rodeo, CA; Shell, Martinez, CA.; Tosco, Martinez, CA.; and Phillips, Borger, TX.
Ron Kirby
EPA Task Manager
vn
-------
1. Introduction
1.1 Background
The purpose of this study is to provide information for determining whether the current
effluent limitations guidelines and standards for the petroleum refining industry contained within
Title 40 of the U.S. Code of Federal Regulations at Part 419 (cited as 40 CFR 419), should be
revised or updated. This study was conducted to meet EPA's obligations under Section 304(m)
of the Clean Water Act (CWA), in accordance with a consent decree in Natural Resources
Defense Council et al v. Reilly (D.D.C. 89-2980, January 31, 1992).
This report is a compilation of data collected during 1992 and 1993, and includes
comparisons with data collected in the late 1970's which formed the basis of the existing
limitations. The industry has changed significantly since the 1970's and this report summarizes
and evaluates these changes.
1.2 Status of Categorical Regulations
EPA's effluent limitations guidelines and standards program was initiated as one of the
major provisions of the 1972 Federal Water Pollution Control Act Amendments (Clean Water
Act Sections 301, 304, 306, 307 and 501). Under these provisions, EPA is required to establish
Best Practicable Control Technology Currently Available (BPT), Best Available Technology
Economically Available (BAT), New Source Performance Standards (NSPS) and Pretreatment
Standards for Existing Sources and New Sources (PSES and PSNS respectively) regulations for
major industrial categories.
In 1974, EPA promulgated BPT and BAT effluent limitations guidelines as well as NSPS
and PSNS for the petroleum refining industry. (US EPA, 1974a). These regulations were
based on the information presented in the 1974 Development Document for the Petroleum
Refining Category (US EPA, 1974b). Data included in this report were gathered from a number
of EPA and American Petroleum Institute (API) sources to identify facilities and unit processes
employed in this industry, to characterize their wastewater discharges, and to review the
performance of wastewater treatment systems within the U.S. petroleum refining industry.
When the 1974 regulations were developed, EPA found that the industry could be divided
into five discrete subcategories:
Topping Refineries (Subcategory A)
Cracking Refineries (Subcategory B)
Petrochemical Refineries (Subcategory C)
Lube Refineries (Subcategory D)
Integrated Refineries (Subcategory E).
-------
These subcategories are defined as follows:
Sub category
Topping
Cracking
Petrochemical
Lube
Integrated
Basic Refinery Operations Included
Topping and catalytic reforming whether or not the facility includes any
other process in addition to topping and catalytic reforming. This
subcategory does not include facilities which include thermal processes
(coking, visbreaking, etc.) or catalytic cracking.
Topping and cracking, whether or not the facility includes any processes in
addition to topping and cracking, unless specified in one of the
subcategories listed below.
Topping, cracking and petrochemical operations, whether or not the facility
includes any process in addition to topping, cracking and petrochemical
operations1, except lube oil manufacturing operations.
Topping, cracking and lube oil manufacturing processes, whether or not the
facility includes any process in addition to topping, cracking and lube oil
manufacturing processes, except petrochemical operations1.
Topping, cracking, lube oil manufacturing processes, and petrochemical
operations, whether or not the facility includes any processes in addition to
topping, cracking, lube oil manufacturing processes and petrochemical
operations1.
From the data, a size and complexity factor were determined, and for each individual refinery
these factors are calculated to account for additional variations within each subcategory.
The BPT limitations determined in 1974 are based on both in-plant and end-of-pipe
technology. BPT in-plant technology consist of control practices widely used within the
petroleum refining industry, and includes the following:
Installation of sour water strippers to reduce the sulfide and ammonia concentrations
entering the treatment plant.
Elimination of once-through barometric condenser water by using surface condensers
or recycle systems with oily water cooling towers.
1 The term "petrochemical operations" means the production of second generation
petrochemicals (i.e., alcohols, ketones, cumene, styrene, etc.) or first generation petrochemicals
and isomerization products (i.e., BTX, olefms, cyclohexane, etc.) when 15 percent or more of
refinery production is as first generation petrochemicals and isomerization products.
-------
Segregation of sewers, so that unpolluted storm run-off and once through cooling
waters are not treated normally with the process and other polluted waters.
Elimination of polluted once-through cooling water, by monitoring and repair of
surface condensers or by use of wet and dry recycle systems.
BPT end-of-pipe treatment technologies consist of equalization and storm water
diversion; initial oil and solids removal (API separators or baffle plate separators); carbonaceous
(biochemical and chemical oxygen demand, i.e., BOD and COD) waste removal using biological
treatment (activated sludge, aerated lagoons, oxidation ponds, trickling filter, or combinations of
these); and effluent polishing (polishing ponds or sand, dual-media, or multi-media filters)
following biological treatment (US EPA, 1974b).
The BPT and BAT limitations, as well as NSPS were established in 1974. The BPT and
BAT limitations, as well as NSPS, were challenged in the U.S. Court of Appeals for the Tenth
Circuit. On August 11, 1976, the Court upheld the BPT limitations and NSPS, but remanded
the BAT limitations, including limitations issued to control storm water discharges from
refineries, to EPA for further consideration.
In 1977, EPA began restudying the BAT and storm water regulations. To update the
information needed to establish BAT effluent limitations guidelines for the petroleum refining
category, questionnaires were sent to all refineries in the United States and its territorial
possessions. The information obtained described petroleum refining industry wastewater
treatment practices for the year 1976 (US EPA, 1982a).
Information received as a result of this questionnaire was combined with existing
information from the 1974 rulemaking in order to develop an industry profile. This profile
included number of plants, their size, geographic location, manufacturing processes, and
wastewater generation, treatment, and discharge methods. Information on number, size, and
geographic location of refineries was later updated with 1980 data from the U.S. Department of
Energy (DOE).
In 1982, EPA determined that BAT for the petroleum refining industry was equivalent to
the 1976 (existing) level of control (US EPA, 1982b). However, as a result of litigation, BAT
limitations were revised in 1985 to reflect additional flow reduction basis and lower attainable
concentrations for phenol and chromium.
For BAT limitations covering phenol and chromium, the revised regulation is based on
the following flow model:
FLOW = 0.0021C + 0.0127A + 0.0236K + 0.0549L + 0.0212R
Where:
-------
FLOW = Net process wastewater in million gallons/day
C = Sum of crude process rates in 1000 bbl/day
A = Sum of asphalt process rates in 1000 bbl/day
K = Sum of cracking and coking process rates in 1000 bbl/day
L = Sum of lube process rates in 1000 bbl/day
R = Sum of reforming and alkylation process rates in 1000 bbl/day
PSES final regulations were promulgated on March 23, 1977 (US EPA, 1977), codifying
the interim final rule published along with BPT in 1974. These regulations established a daily
maximum limitations for oil and grease and ammonia of 100 mg/L each. There are no current
pretreatment standards for toxic pollutants.
1.3 Software Disk Available
The calculations for determining permit limitations are simplified somewhat in that Mr.
Ed McHam of EPA's Region 6 has developed a software program (spreadsheet) to complete the
required calculation. The final spreadsheet is available in Lotus 1-2-3ฎ format, with text in
WordPerfectฎ. The program determines categorical limits and water quality limits after input of
process through-put information. EPA Engineering and Analysis Division will provide a disk
upon request.
-------
2. Description of the Industry
2.1 Production Operations
Generally, a simple petroleum refinery includes catalytic reforming and treating processes
in addition to crude oil distillation. A more complex refinery also includes catalytic cracking,
polymerization, alkylation and asphalt oxidation as well as other selected unit processes. A very
complex refinery may include high vacuum fractionation, solvent extraction, de-asphalting,
de-waxing and treating processes, in addition to those found in simple and complex refineries.
Although many minor products can be produced from crude oil by simple physical
separation processes, such as fractional distillation, the proportions of each product may not
match the desired values, or the quality may not be adequate for the use intended. Therefore,
many sophisticated chemical process operations also take place in a petroleum refinery, in order
to produce the distribution, quality and quantity of products desired.
The following paragraphs describe the basic processes that are used in petroleum
refineries.
2.1.1 Crude Oil and Product Storage
Petroleum refineries require storage facilities for both crude oil and individual final
products. The amount of storage required is quite variable, depending on the type and reliability
of crude supply and on the location and nature of markets. The crude oil storage area of a
refinery serves to provide a working supply, equalize process flow and separate water and
suspended solids from the crude oil.
During storage, water and suspended solids in crude oil and, in lesser quantities, in
products tend to settle out to form a water layer at the tank bottom. This is typically in the form
of a sludge which, in the case of crude oil, usually contains foul sulphur compounds and high
dissolved solids concentrations.
2.1.2 Crude Distillation
Distillation is the basic refining process for the separation of crude petroleum into
intermediate fractions of specified boiling point ranges. This separation or fractionation takes
advantage of the differing boiling points and vapor pressures of the various components in the
crude oil mixture.
In addition to the atmospheric distillation process it is normally necessary to subject the
residual or bottoms from atmospheric distillation to a second and/or third stage distillation,
conducted under vacuum.
-------
The steam applied at the various stages to the process is in direct contact with
hydrocarbons. It is eventually carried over with various fractions and is separated out by gravity
when the fraction is condensed. These steam condensates are invariably foul, and constitute a
foul or sour condensate waste stream, containing sulphides, ammonia, chlorides, mercaptans and
phenols.
2.1.3 Cracking
In this process, heavy oil fractions are converted into lower molecular weight fractions
including domestic heating oils, high octane gasoline stocks and furnace oils. The cracking
process increases the yield of gasoline taken from the crude oil and improves its quality. By
using cracking, refiners can double their gasoline output per unit volume of crude oil charged to
their distillation towers or stills.
The cracking is usually the largest single source of sour and phenolic wastewater in a
large refinery. The wastewater is derived from the steam condensate from the overhead
accumulator and condensate from steam stripping of side streams. The major pollutants are oil,
sulphides, phenols, ammonia and traces of cyanides.
2.1.4 Hydrocarbon Rebuilding
Higher octane products for use in gasoline are manufactured by alkylation. In this
process, small hydrocarbon molecules are combined into large molecules: the reverse of
cracking. The resulting products are valuable components of high quality motor fuel and
aviation gasolines.
This operation produces sour water, high in sulphides, mercaptans, ammonia, suspended
solids and oils.
2.1.5 Hydrocarbon Rearrangements
Isomerization and reforming are two process techniques for obtaining higher octane
gasoline blending stock. Isomerization, a molecular rearrangement process rather than
decomposition process, generates no major pollutant discharge. Catalytic reforming produces
aromatics from naptha in the presence of a catalyst by molecular rearrangement.
Dehydrogenation is the primary reaction.
2.1.6 Hydrotreating
Hydrotreating processes are used to purify and pretreat various feedstocks by reacting
with hydrogen. Product contaminants, including sulphur and nitrogen compounds, odor, color
and gum-forming materials, are removed in this process.
-------
Many different hydrotreating processes are used, depending on the feedstock and intended
use of the product. Common applications include:
Pretreatment of reformer feedstock
Naphtha de-sulphurization
Lube oil polishing
Pretreatment of cat-cracking feedstock
Heavy gas oil and residual desulphurization
Naphtha hydrogenation.
The strength and quantity of wastewaters generated by hydrotreating is largely dependent
upon the specific process and feedstock used. Wastewaters are derived as condensates from
fractionating the product hydrocarbons and are mainly contaminated by ammonia and sulphur
compounds.
2.1.7 Solvent Refining
Various chemicals and solvents are used to improve the quality of a particular feedstock
component. The compounds removed or isolated by this process may be highly objectionable in
the specific product being prepared, but may be desirable in making other products or may be
converted into desirable materials. The major pollutants from solvent refining are the solvents
themselves, many of which can produce a high BOD. Under ideal conditions the solvents are
continually recirculated, but in practice some solvent is always lost, usually through leaks at
pump seals and flanges. Oil and solvent are major wastewater constituents.
2.1.8 Asphalt Production
The reduced crude fraction or residual taken from the bottom of the vacuum still may be
blended into heavy fuel oil or may be made into asphalt by oxidation in an asphalt still.
Wastewater is derived from steam added to the reactor for stripping volatiles, as well as a
small quantity of water produced for oxidation reactions with the asphalt. The water separated
out is very oily, high in BOD and usually sour as a result of the normally high sulphur content of
the residual.
2.1.9 Lubricating Oil Manufacture
Lubricating oils require closely controlled properties and are generally only manufactured
from special high grade feedstocks. However, even with high grade feedstocks, lube oils must
be treated to remove asphalt, wax and hydrocarbons whose viscosity is temperature sensitive
(generally aromatic compounds).
-------
This operation produces acidic rinse waters and acid sludges for disposal, which are high
in dissolved and suspended solids, sulphates and sulphonates and which form stable oil
emulsions.
2.1.10 Production of Petrochemicals
These operations are extremely varied, and include production of a wide range of
products such as alcohols, ketones, cumene, styrene, benzene, toluene, xylene, olefms,
cyclohexane, etc. Many petrochemicals are manufactured directly, while others are derivatives
from intermediate products. Wastewaters from these processes are quite variable and dependent
upon the specific operations employed.
2.2 Industry Trends
EPA identified 256 refineries in 1976. Total average production that year was
16,357,000 barrels per day. During the subsequent 14 years, 64 refineries closed, or
approximately 25 percent of the facilities. However, the production capacity only dropped
1,000,000 barrels per day (bbl/day), or approximately six percent. This is because most of the
facilities that closed were small inefficient refineries. Their capacity was replaced by increasing
production at the larger existing refineries. Table 2.1 presents a summary of the number of
refineries, and their associated production rates by state. This table indicates that the number of
refineries decreased in 26 states; there was no changes in refinery count in nine states; and the
number of refineries actually increased in five states (Alaska, Arizona, Nevada, New Jersey, and
Tennessee). Figure 2.1 graphically presents the number of refineries by state.
Table 2.1 Capacity and Number of United States Refineries in 1991 vs. 1976
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Delaware
Florida
Georgia
1976
Facilities
4
o
6
i
4
33
o
6
i
i
2
1976 Total
Capacity
Crude (b/cd)
54,250
73,000
5,000
61,000
2,269,600
65,000
150,000
4,000
17,000
1991
Facilities
4
5
2
3
30
3
1
0
2
1991 Total
Capacity
Crude
(b/cd)
154,250
224,500
14,210
60,470
2,150,555
91,200
140,000
0
35,500
Difference
in No. of
Facilities
0
2
1
-1
-3
0
0
-1
0
Difference in
Total
Capacity
100,000
151,500
9,210
-530
-119,045
26,200
-10,000
-4,000
18,500
-------
Hawaii
Illinois
Indiana
Kansas
Kentucky
Louisiana
Maryland
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New
Hampshire
New Jersey
New Mexico
New York
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Tennessee
Texas
Utah
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Totals
Source: Thrash
2
12
6
10
4
21
2
6
3
5
1
6
1
0
1
4
7
2
3
7
12
1
9
0
46
7
1
8
o
6
i
13
256
i, 1991
100,300
1,272,000
605,820
410,011
171,200
2,108,173
30,500
155,920
223,900
346,200
107,000
116,500
5,380
0
15,000
671,000
92,620
107,000
60,006
602,000
560,400
14,000
800,200
0
4,231,135
158,500
55,000
379,950
22,700
42,000
194,002
16,357,267
2
7
4
8
2
19
0
4
2
5
0
4
0
1
0
6
4
1
1
4
7
1
7
1
31
6
1
7
2
1
4
192
143,050
948,500
428,900
353,225
218,900
2,299,241
0
125,200
285,600
358,600
0
138,900
0
4,500
0
499,250
76,650
39,900
58,000
454,150
409,500
0
741,300
75,000
3,882,200
154,500
53,000
523,225
29,680
32,000
122,900
15,326,556
0
-5
-2
-2
-2
-2
-2
-2
-1
0
-1
-2
-1
1
-1
2
-3
-1
-2
-3
-5
0
-2
1
-15
-1
0
-1
-1
0
-9
-64
42,750
-323,500
-176,920
-56,786
47,700
191,068
-30,500
-30,720
61,700
12,400
-107,000
22,400
-5,380
4,500
-15,000
-171,750
-15,970
-67,100
-2,006
-147,850
-150,970
-14,000
-58,900
75,000
-348,935
-4,000
-2,000
143,275
6,980
-10,000
-71,102
-1,030,711
-------
Figure 2.1. Number of Refineries per State
In 1976, there were 44 indirect dischargers. As of 1990, 22 remained in operation.
The data presented below indicate that the petroleum refining industry has been going
through a consolidation, which has resulted in a large decrease in the number of refineries in the
United States, but only a slight (six percent) decrease in production. Figure 2.2 graphically
shows the number of refineries by capacity from 1982 through 1993. These data confirm that
there has been a dramatic reduction in small refineries, and an increase in refineries with
capacities over 100,000 bbl/day. It is expected that this trend will continue, with refineries
continuing to close, but expansions occurring at others, keeping the total refinery capacity in line
with demand for refinery products.
A factor affecting this industry is the addition and change of production operations.
Although atmospheric and vacuum distillation, catalytic cracking, and their associated treating
and reforming operations will remain the primary refinery operations, new production operations
continue to be added. These include new innovative coking and desulphurization processes.
10
-------
Many of these process changes occur as a result of the new gasoline reformulation rules
designed to reduce the amount of volatile components in gasoline. These rules are causing
refineries to make process modifications to their catalytic cracker units, as well as installing
additional hydrotreaters and unit processes to manufacture additives. These improvements and
changes may greatly effect the amount and quality of wastewaters generated by refineries.
Other process changes are being made to comply with 1990 Clean Air Act Amendments
requirements, such as the lead phaseout rules, National Emissions Standards for Hazardous Air
Pollutants (NESHAP) requirements covering benzene and hazardous organics and low sulfur
diesel standards are in place or in the works.
11
-------
90
80
70
60
50
40
30
20
Figure 2.2 Number of Refineries Since 1982
(Classified by Capacity)
I
3?
u.
o
s
z
1
\ ปปซ .
\ %
....
1
I
J_
J_
I
I
1S8? 1991
ViAH
Morซ thwi 100,000 bOH. 30,O01 to HMMNMI Mb. 1O.OO1 to 3O.OOO bWs. 1O.OOObfeH.wlws
Source: Energy Information Administration
12
-------
3. Summary of Information Sources Used in This Study
The preponderance of information collected for use in this study was obtained from a
number of readily available data sources. A description of these sources is contained below.
3.1 Oil And Gas Journal Survey
The Oil and Gas Journal publishes a list of all active U. S. and foreign refineries. For
this study, the December 1991 report was used to provide an estimate of the number of refineries
in use in the United States, their location and production capacity (Thrash, 1991). These data
were used to show general industry trends since 1976.
3.2 EPA Office of Air and Radiation Questionnaire
EPA's Office of Air Quality Planning and Standards, in the Office of Air and Radiation
(OAR), surveyed nine companies to obtain information on hazardous air pollutants (HAP) and
volatile organic compounds (VOC) emissions from refineries. A total of 27 refineries were
covered by the responses from these nine companies. For the purpose of this study, an
additional data table was added to the survey form to request information on refinery production,
process throughput, wastewater discharge rates, wastewater treatment systems and wastewater
flow diagrams. The data from these 27 refineries (approximately 15 percent of the industry)
have been used as a sample to represent the industry as a whole.
3.3 Plant Visits
EPA visited six refineries as part of this study. Four facilities in California were visited,
which represent exemplary refineries in terms of water use and existing wastewater treatment
technologies. One refinery in Texas and one in Pennsylvania were also visited to represent
refineries that use greater amounts of process water or that did not have stringent water quality
standards, but meet the existing effluent limitation guidelines. The data from these six sites
have been used to obtain detailed data on site-specific water use practices and treatment system
performance.
3.4 Permit Compliance System Data
EPA maintains a large computerized data base called the Permit Compliance System
(PCS). This data base contains an inventory of National Pollutant Discharge Elimination
System (NPDES) permittees, and discharge monitoring report (DMR) data supplied by industry
permittees as part of their self monitoring program. (Generally DMR data are available on PCS
only for certain facilitiesmajor facilities as identified by their permit authorities). This data
base contains DMR data on 137 refineries, and the data were used to estimate the levels of
pollutants in refinery effluents for the reporting year of 1992.
13
-------
3.5 Los Angeles County Sanitation Districts
Information from the Los Angeles County Sanitation Districts (LACSD) was provided for
14 refineries discharging into their municipal sewer system. These data were used to represent
the status of refinery effluents to a sewer system with an exemplary pretreatment program.
3.6 Other Sewerage Authorities
Three additional sewerage authorities were contacted to obtain data on discharges from
other indirect discharging refineries. These data were used to represent the pollutant levels of
refinery discharges to smaller systems with less comprehensive pretreatment programs.
3.7 Province of Ontario, Canada Petroleum Study
Ontario's Ministry of Environment established the Municipal-Industrial Strategy for
Abatement (MIS A) Program in 1986 with an ultimate goal of achieving "virtual elimination" of
the discharge of persistent toxic pollutants. As part of this program, the Ministry is setting
sector (categorical) specific best available technology limitations. The petroleum refinery sector
(industry) was identified as part of this program.
The data collected as part of this study (data on seven refineries) were used in this effort.
The seven existing Ontario refineries collected extensive effluent quality data for a one year
period on a full range of toxic pollutants. In addition, the refineries collected data on the
presence of dioxins in the wastewaters from the regeneration of catalysts from their catalytic
reformers. These data have been used to supplement the data collected on U.S. refineries.
3.8 Other Data Sources
Data from published literature, industry studies, previous effluent limitations guidelines
studies and other EPA studies have been collected and also used in this study.
14
-------
4. Treatment Technologies Used in The Industry
Historically (prior to the 1960's), process operations used large quantities of water, and
often simple oil separation constituted end-of-pipe treatment. In fact, API separators were
originally installed to economically recover oil rather than treat wastewater discharges. When
EPA studied this industry in the early 1970's, secondary biological treatment was becoming
common, and certain in-plant controls were becoming industry standards. These included sour
water stripping and the replacement of barometric condensers with surface condensers.
As a result of the early 1970's studies, EPA promulgated the BPT and NSPS regulations
in 1974 that were production-based mass limitations based upon the following technologies:
In-Plant Controls
Installation of sour water strippers to reduce the sulfide and ammonia concentrations
entering the treatment plant.
Elimination of once through barometric condenser water by using surface condensers
or recycle systems with oily water cooling tower.
Segregation of sewers, so that unpolluted storm run-off and once through cooling
waters are not treated normally with the process and other polluted waters.
Elimination of polluted once through cooling water, by monitoring and repair of
surface condensers or by use of wet and dry recycle systems.
End-of-Pipe Treatment
Equalization
Additional oil separation using dissolved air flotation (DAF)
Biological treatment
Polishing (polishing ponds, sand filtration).
EPA's 1982 BAT rulemaking confirmed the use of the above technologies as the
framework for setting effluent limitations guidelines for priority pollutants. However, further
water reduction had been experienced between 1972 and 1976, the years data had been collected
by EPA. The new flow data was used to develop a revised (BAT) flow model, which formed
the basis for more stringent chromium and phenolic production-based mass limitations.
The following subsections describe the technologies used by the petroleum refinery
industry, the performance expected by them, and industry trends since the 1982 EPA rulemaking.
15
-------
4.1 In-Plant Controls
In-plant technologies for refinery wastewater include steam stripping, neutralization, and
source control. Table 4.1 summarizes the in-plant treatment technologies currently in use for
treating wastewaters generated from specific refinery operations. A description of the
technologies is contained below.
4.1.1 Steam Stripping
Sour waters generally result from water brought into direct contact with a hydrocarbon
stream. This occurs when steam is used as a stripping or mixing medium or when water is used
as a washing medium. Sour waters contain sulfides, ammonia, phenols and other organic
chemical constituents of the crude oil.
The most common in-plant treatment for sour waters is steam stripping (i.e., sour water
stripping). Sour water stripping is a gas-liquid separation process that uses steam or flue gas to
extract the gases (sulfides and ammonia) from the wastewater. The stripper itself is a
distillation-type column containing either trays or packing material. Columns range from simple
one-pass systems to sophisticated reflux columns with reboilers.
In removing sulfides and ammonia, the efficiency of sour water treatment processes is
greatly influenced by pH. In general, sour water strippers remove between 85 and 99 percent of
the sulfides present. However, when the pH is lowered by means of acid treatment, stripping
efficiency is increased. On the other hand, when caustic is utilized to maintain high pH, up to
95 percent ammonia removal can be achieved. By considering pH in the stripping process, one
can either adjust the pH to optimize removal of one or another of sulfides or ammonia, or use a
two stage sour water stripping process to obtain maximum removal of both pollutants.
Steam stripping can also be used to remove volatile organic compounds from selected
refinery wastewater streams that have high concentrations of these pollutants. Stripping of the
organic constituents of the wastewater stream occurs because the organic volatiles tend to
vaporize into the steam until its concentration in the vapor and liquid phases (within the stripper)
are in equilibrium. The height of the column and the amount of packing material and/or the
number of metal trays along with steam pressure in the column generally determine the amounts
of volatiles that can be removed and the effluent pollutant concentration levels that can be
attained by the stripper.
16
-------
Table 4.1 Demonstrated Wastewater Technologies for In-Plant Treatment of Refinery
Process Streams
Refinery Operation
Crude Desalting
Atmospheric and Vacuum Distillation
Thermal Cracking
Catalytic Cracking
Hydrocracking
Polymerization
Alkylation
Isomerization
Catalytic Reforming
Solvent Refining
Hydrotreating
Grease Manufacturing
Drying and Sweetening
Lube Oil Finishing
Blending and Packaging
Equipment Cleaning;
Spills; Miscellaneous
Utilities
Technologies
Stripped sour water as makeup to the desalter
Sour water stripping. Sour water can be recycled
through crude desalters prior to processing in sour
water strippers.
Sour water stripping.
Sour water stripping.
Sour water stripping.
Neutralization
Acid recovery; neutralization.
Generally not pre-treated.
Granular activated carbon for removal of
CDDs/CDFs.*
Generally not pre-treated.
Generally not pre-treated.
Generally not pre-treated.
Neutralization with acid or FCCU regenerator flue
gas.
Generally not pre-treated.
Generally not pre-treated.
Segregation; slop oil tank.
Equalization of ion exchange regeneration wastes;
others generally not pre-treated.
* Presently, there is very limited capacity in the U.S. to regenerate carbon loaded
with CDDs/CDFs.
17
-------
4.1.2 Neutralization of Spent Acids and Caustics
Spent caustic solutions are generated by various finishing wet treatment processes aimed
at neutralizing and extracting acidic materials occurring naturally in crude, acidic products from
various chemical treatment steps, and acidic materials produced in cracking processes. Spent
caustics generally contain sulfides, mercaptans, sulfates, sulphonates, phenols and naphthionic
acids. The phenol concentrations, in particular, may be high enough to warrant processing of
spent caustic for the recovery of phenols.
Spent acid is reclaimed on site or returned to the vendor for reclamation, if the bottoms
are then sent to crude desalting, the high phenol content may be recovered within the process by
extraction.
4.1.3 Source Control
Source control measures to minimize wastewater generation and contamination can
significantly reduce the volume of effluent and the amount of pollutants discharged from
refineries. Such measures include water use reduction, and wastewater reuse and recycle.
Along with several general measures to reduce water use, major wastewater discharge
reduction techniques address segregation, boiler condensate recovery, and treated effluent reuse.
A report entitled "Water Reuse Studies" (API, 1977) discusses the practicality (and costs) of
specific wastewater reduction techniques. Another study, "Wastewater Reuse and Recycle in
Petroleum Refineries" (Langer, 1983) also presents information on this topic. The Langer study
investigated 15 U.S. refineries: three refineries were considered to be exemplary and the
remaining 12 were candidates for further effluent discharge volume reduction programs. The
report identifies specific wastewater reduction techniques with their anticipated effectiveness and
associated costs. A summary of the findings from these studies is presented below.
4.1.4 Wastewater Segregation
Segregation of refinery wastewaters is important to allow for reuse of wastewaters with
little or no treatment. Additionally, segregation of severely contaminated streams provides the
opportunity for pretreatment, thus reducing the effects of dilution and contamination of the
overall combined process wastewater stream. The API study recommended dividing streams
into three groups, by level of contamination. The first stream is high quality and is suitable for
reuse with only minimal treatment, if any. The second stream has low total dissolved solids
(TDS) and requires some treatment prior to reuse, and the final stream, which has high TDS, is
not suitable for reuse and requires complete treatment before discharge.
18
-------
4.1.5 Boiler Condensate Recovery
Boiler steam condensate recovery and reuse can also significantly reduce the amount of
boiler circuit wastewater requiring discharge. (Note: Many refineries are already recovering and
reusing as much condensate as economically feasible.) Specific measures include the following:
Increase condensate recovery by the installation or expansion of piping systems to
collect steam lost by overheating, tracing, tank heating, traps, utility and leaks.
Reduce vent losses by the elimination of vents at process units, turbines and steam
traps.
4.1.6 Treated Effluent Reuse
For high quality wastestreams, the wastewater may be suitable for direct reuse in
cooling systems or for steam generation without treatment. Other examples of reuse (with little
or no treatment) include using sour water as make-up for desalters and acid gases for the
neutralization of spent caustic solutions.
In the Langer report, several uses of treated effluent were identified. These included
wastewater reuse for: exchanger and barometric condenser cooling, dampening of coke fines for
dust control, firewater, service water and wash water, pump gland cooling, and other machine
cooling processes. Once-through cooling water can be reused as make-up for desalters, cooling
towers, or as process water, but may be unsuitable without prior treatment.
4.1.7 Other General Measures
Other general measures for the reduction of wastewater generation include:
Conversion of barometric condensers to surface condensers.
Improved management of firewater and wash water systems including the elimination
of losses from overflowing sumps, freely running hoses, temporary exchange coolers,
and underground leakage.
4.1.8 Cooling Water Systems
Historically, the primary factor considered in selecting cooling water systems was the
availability of water and its associated cost. However, impacts of thermal discharges, water
conservation, and compliance with discharge limitations have also become relevant
considerations. The advantages and disadvantages of different cooling systems are presented in
Table 4.2.
19
-------
Table 4.2 Advantages and Disadvantages of Different Types of Cooling Systems
Factor
Noise
Cold Weather Plumes
Water Contamination
Potential from Leaks
Soil Contamination
Potential from Leaks
Air Contamination
Potential
Cooling Efficiency
Energy Required
Maintenance and Labor
Potential for Impact for
Thermal Shock to Fish
Potential for Impact of
Water Treatment
Chemicals
Costs
Once-Through
Cooling Water
None
none
high
low
none
high
moderate
low
moderate to high
moderate
low
Cooling Tower
low
potential problem
moderate
low
moderate
moderate to high
high
high
none
moderate
high
Air Cooling
high
none
none
low
moderate
low to moderate
high
low
none
none
high
Although cooling water can be completely eliminated by converting to 100 percent air
cooling systems, this may be impractical or uneconomical for some refineries due to space
availability and the orientation of the heat exchanger systems. There are also real process
limitations in that many processes designed for cooling water systems (more efficient cooling
systems) can not be retrofitted for air cooling systems.
4.1.9 Once-Through Cooling Water Systems
In once-through cooling water (OTCW) systems, due to the use of chlorine and chlorine
derivatives as additives, dechlorination using sulfide or sulfite compounds may be necessary to
remove residual chlorine. In recent years, as concerns about the environmental impacts of
20
-------
chlorine and chlorination by-products have increased, the use of bromine an bromine compounds
for OTCW treatment has also received consideration.
4.1.10 Cooling Tower Systems
There are several methods to minimize cooling tower blowdown streams. Cooling tower
blowdown can contribute up to one third of total refinery wastewaters. Although cooling tower
systems vary from plant to plant, the following general recommendations were made in the study
to reduce cooling tower blowdown:
Recycle cooling water from pumps, compressors, and sample boxes that use
blowdown.
Replace existing oil-leaking pump gland packing with mechanical seals to permit
collection and recycle of blowdown to cooling tower.
Reduce use of pump gland cooling water where presently overused or eliminate
service completely.
Upgrade maintenance of existing systems to reduce leakage and sump overflow.
Refineries can pretreat raw water to improve the initial quality of the influent which in
turn will significantly increase the number of reuse cycles in cooling towers and reduce
blowdown amounts. The toxicity of the blowdown can also be reduced. Water treatment
chemicals containing zinc and chromate compounds used in the recirculating waters can be
replaced with less toxic organic compounds. In the last ten years chromates have been
virtually eliminated by substitution for less toxic chemicals. Zinc levels in the effluent discharge
average 0.15 mg/1 as shown in Table 4.5.
4.2 End-of-Pipe Treatment Technologies
All wastewater treatment that immediately follows the oil/water gravity separators (API
or oily water separator) is considered end-of-pipe treatment. (The API separator is recognized as
part of the refinery process equipment for the economic recovery of oil and, as such, is not
considered a treatment unit.) Conventional end-of-pipe treatment technologies are addressed in
this section and are classified as preliminary, biological, and effluent polishing.
4.2.1 Preliminary Treatment
Preliminary treatment commonly consists of equalization, followed by chemical treatment
and supplemental oil removal. Filtration may also be included as part of the preliminary
treatment system to limit the loading of soils to downstream units.
Equalization is one of the first, and one of the most important steps in the treatment of
wastewater. Fluctuations in contaminant concentrations are leveled and the flow and pH of the
waste stream are adjusted to provide the optimum conditions for further treatment. Unusually
21
-------
high flows or high contaminant concentrations, which cannot be handled by equalization, may be
diverted to auxiliary holding facilities and slowly re-introduced in the treatment system when
conditions warrant.
Supplemental oil removal is often accomplished by using parallel plate separators or
chemically assisted dissolved air flotation (DAF) units, whereby emulsified oil in the waste
stream is dispersed and removed. A parallel plate separator is a device which is very similar to
the API separator. It was developed to improve oil and solids removal by mounting parallel
plates at an angle along the length of the separator. By vastly increasing surface area, this device
permits more efficient collection of oils and solids.
Dissolved air flotation consists of saturating a portion of the wastewater feed, or a portion
of the feed or recycled effluent from the flotation unit, with air. The wastewater or effluent
recycle is held at elevated pressure, typically for one to five minutes, in a retention tank and then
released at atmospheric pressure to the flotation chamber. The sudden reduction in pressure
results in the release of microscopic air bubbles which attach themselves to oil and suspended
particles in the wastewater in the flotation chamber. This results in agglomerates which rise to
the surface to form a froth layer.
Chemical flocculation agents, such as salts of iron and aluminum, with or without organic
polyelectrolytes, are often helpful in improving the effectiveness of the air flotation process and
in obtaining a high degree of clarification. Induced air flotation (IAF) is similar to DAF systems
but IAF adds air to a flotation tank by using impellers rather than by adding dissolved air to a
recirculation tank.
Chemical precipitation can be used to remove metals from selected refinery wastewater
streams, such as cooling tower blowdown. Most metals are relatively insoluble as hydroxides,
sulfides, or carbonates, an can be precipitated in one of these forms. The sludge formed is then
separated from solution by physical means such as clarification or filtration. Hydroxide
precipitation is the conventional method of removing metals from wastewater. Most commonly,
caustic soda (NaOH) or lime (Ca(OH)2) is added to the wastewater to adjust the pH to the point
where metal hydroxides exhibit minimum solubilities and are thus precipitated. Sulfide
precipitation has also been demonstrated to be an alternative to hydroxide precipitation for
removing metals from certain wastewaters. Sulfide, in the form of hydrogen sulfide, sodium
sulfide, or ferrous sulfide, is added to the wastewater to precipitate metal ions as insoluble metal
sulfides.
4.2.2 Biological Treatment
Biological treatment is the basic process for treating oxygen-demanding compounds,
usually measured as biochemical oxygen demand (BOD), chemical oxygen demand (COD) and
total organic carbon (TOC). There are a number of variations of which the most common are
described below.
22
-------
Oxidation Ponds. The oxidation pond is practical where land is plentiful and
relatively inexpensive. An oxidation pond has a large surface area and a shallow
depth, usually not exceeding two meters. These ponds have long detention periods
of 11 to 110 days. This process is not reliable in very cold climates.
Aerated Lagoons. The aerated lagoon is a smaller, deeper oxidation pond equipped
with mechanical aerators or diffused air units. The addition of oxygen enables the
aerated lagoon to have a higher concentration of microbes than the oxidation pond.
Where effluent standards are stringent, final clarification is necessary. However,
since the effectiveness of conventional clarification on such effluent is often poor,
filtration may be necessary to comply with limitations. However, refiners have often
addressed this problem by adding polishing ponds after the lagoon.
Trickling Filters. A trickling filter is an aerobic biological process. It differs from
other processes in that the biomass is attached to the bed medium, which may be rock,
slag or plastic. When the biomass reaches a certain thickness, part of it sloughs off.
When the filter is used as the major treatment process, a clarifier is used to remove
the sloughed biomass.
Rotating Biological Contactors (RBCs). RBCs are analogous to trickling filters, in
that they are fixed-film reactors. Bacterial slime is grown on plastic discs rotating
through the wastewater. Approximately half of the circular disc is out of the water at
any one time, being aerated, and half is under water supporting biological growth.
Activated Sludge. Activated sludge is an aerobic biological treatment process in
which newly grown and recycled microbial biomass are suspended uniformly
throughout a holding tank to which raw wastewaters are added. Oxygen is
introduced by mechanical aerators, diffused air systems or a combination of the two.
The organic materials in the waste are removed from the aqueous phase by the
microbial biomass and stabilized by biochemical synthesis and oxidation reactions.
The basic activated sludge process consists of an aeration tank followed by a
clarification step.
4.2.3 Effluent Polishing
The function of effluent polishing is to remove residual suspended solids (biological floe)
which may be carried over from the clarification step. The biological floe will add BOD and
certain toxic organic compounds (which are adsorbed onto or absorbed into the floe) to the final
effluent, and must be removed.
Most end-of-pipe treatment systems at petroleum refineries include effluent polishing in
the form of polishing filters, polishing ponds, or both. Effluent polishing filters are often
23
-------
single-media (sand), however, dual-media (sand, anthracite) and multi-media filters are also
used. Both gravity and pressure filtration systems are utilized in refinery applications.
Polishing ponds can be equipped with baffles and oil skimmers on overflows to remove
traces of free oil which may have evaded upstream treatment systems. In the event that the final
effluent does not meet discharge limitations or standards, some refinery treatment systems allow
the transfer of effluent from the polishing ponds back to preliminary treatment.
4.2.3 Activated Carbon Treatment
There are two forms of activated carbon treatment, Granular Activated Carbon (GAC)
and Powdered Activated Carbon (PAC). Each is discussed below.
Granular Activated Carbon. Adsorption on granular activated carbon (GAC) is
currently being used for effluent polishing following biological treatment at some
refineries to remove trace level toxic organic pollutants, and at least three U.S.
refineries to meet bioassay permit requirements based on toxicity for trout or fat-head
minnows.
The adsorption process typically requires preliminary filtration or clarification to
remove suspended solids. Next, the wastewaters are placed in contact with carbon so
adsorption can take place. Normally, two or more beds are used so that adsorption
can continue while a depleted bet is reactivated. Reactivation is accomplished by
heating the carbon to 870ฐ to 980ฐC (1600ฐ to 1800ฐF) to volatilize and/or oxidize the
adsorbed contaminants.
Powdered Activated Carbon. This technology consists of the addition of powdered
activated carbon (PAC) to biological treatment systems. The adsorbent quality of the
carbon aids in the removal of soluble organic materials in the biological treatment
unit. This treatment technique also enhances color removal, clarification and system
stability. BOD and COD removal may be enhanced but, it is not certain, depending
on the treatment system. This treatment technology is currently being used at one
U.S. refinery at least.
4.2.5 Technologies Used at EPA/OAR Survey Refineries
A summary of the treatment technologies that are in place at the 27 refineries covered by
the OAR survey plants is presented in Table 4.3. Of the 27 refineries, 20 are direct dischargers
and 7 are indirect dischargers. All of the 20 direct discharging refineries have some form of
biological treatment. Three have sand filtration and one facility has an in-plant activated carbon
system in addition to biological treatment.
24
-------
Table 4.3 Summary of Current Wastewater Treatment Technologies for 27
Refineries Surveyed
In-Plant Controls
Primary Treatment
Secondary
Treatment
Treatment Type
Oil-Water Separator
Stripper
Oxidizer
Activated Carbon
API Separator
Air Flotation
Coagulation
Chemical Precipitation
Dissolved Air Flotation
Equalization
Flocculation
Grit Chamber
Gas Flotation
Induced Air Flotation
Settling & Skimming
Activated Sludge Unit
Bio Treatment Ponds
PAC Bio-Treatment
RBC's
Secondary Clarifier
Lagoons
Filtration (Media & Sand)
Aeration & Other
Biological Treatment
Direct Discharge
Refineries (total 20)
15
16
2
1
9
5
1
1
10
16
1
0
0
4
0
11
6
1
1
12
3
3
5
Indirect Discharge
Refineries (total 7)
4
5
0
1
5
1
0
0
1
4
1
1
1
2
1
0
2
0
1
0
0
1
0
Source : EPA Office of Air and Radiation Survey ( 1 992)
25
-------
4.2.6 Performance of End-of-Pipe Systems
There are virtually no available data from this industry on the performance of individual
treatment units within a treatment system. Therefore, performance must be assessed using
effluent data only. However, since most direct discharging plants use the basic treatment train
of preliminary treatment (oil removal), biological treatment and effluent polishing (filtration or
ponds), effective comparisons of performance can be made using effluent data.
Table 4.4 presents a summary of the effluent data collected from the six refineries visited
as part of this study, and compares the pollutants covered by BPT with the concentrations used as
a basis to develop BPT limitations in 1974. Table 4.5 summarizes effluent concentration data
for a number of pollutants obtained from the following three data sources:
Average concentration data (over a one-year period) collected during Canada's
"Seven Refineries Study" conducted in 1989,
Long-term average data collected from seven U.S. refineries during the Canadian
study,
A summary of PCS data from 138 direct discharging refineries for 1992.
EPA's PCS system was accessed for priority pollutant data only. The data in this table
indicate higher levels of priority pollutants in the PCS data base for chromium (in 1991 only),
benzene, toluene, copper and nickel than from the other data sources.
26
-------
Table 4.4 Summary of Effluent Data: Six Site Visits, 1992
Pollutant Average
Values (in mg/1)
TSS
COD
Oil & Grease
NH3 (as N)
Sulfide
Phenols (4AAP)
Chrome, Total
Lead
Zinc
Benzene
Toluene
Naphthalene
Copper
Nickel
Selenium
4 California
Refineries
8.75
2.7
1.43
<0.05
<0.02
<0.02
0.012
0.04
ND
ND
ND
0.01
0.033
0.06
Pennsylvania
Refinery
11
51
2.7
0.94
0.14
0.005
0.02
0.002
0.147
0.011
0.006
Texas
12
59.5
4.2
1.42
0.018
0.012
0.015
<0.001**
.025**
<005**
<005**
.013**
.039**
.008**
BPT/BAT Equiv.*
Concentrations
10
5
0.1
0.1
0.25
Notes:
* These are concentrations used as a basis to develop the BPT production-based mass
limitations using the BPT flow model.
** Data from permit renewal application (March 1993)
NDNon Detectable
-- No Data
27
-------
Table 4.5 Summary of Refinery Effluent Data: Canadian Study and PCS Data
Pollutant (mg/1)
TSS
COD
Oil & Grease
NH3 (as N)
Sulfide
phenols (4AAP)
Chrome, Total
Lead
Zinc
Benzene
Toluene
Naphthalene
Copper
Nickel
Arsenic
Cyanide
Selenium
Canadian Study1
Average of 7
Ontario
Refineries
22
49.2
2.17
1.7
0.08
0.0110
0.0068
0.0041
0.29
0.0008
0.0007
0.0011
0.0048
0.0034
0.009
0.007
0.005
Average of 7
U.S. Refineries
31
85
4.08
5.21
<0.03
0.047
0.028
0.09
< 0.005
< 0.004 (max)
< 0.003 (max)
< 0.012
<0.08
138 U.S. Refineries
PCS Data (1992)2
Low
22.3
93.1
3.34
4.83
0.044
0.038
00.011
0.004
0.15
0.0008
<0.02
0.0106
0.0159
0.202
0.041
0.145
High
22.6
93.1
3.42
4.83
0.052
0.040
0.013
0.006
0.15
0.001
0.003
0.0112
0.0166
0.202
0.041
0.145
Notes
1. Source: Best Available Treatment Technology for Ontario Petroleum Refining Sector, August 1991.
2. Source: Appendix 1 and 2. Low non-detects equal zero, High non-detects equal one-half detection.
- No Data
28
-------
4.2.7 Storm Water Management
Storm water management at petroleum refineries can have a significant bearing on the mass
discharge of conventional and toxic pollutants to receiving waters. In addition to increasing
wastewater volumes, stormwater also often contributes high levels of total suspended solids
(TSS).
Under ideal circumstances, all stormwater should be segregated into the categories identified
in Table 4.6 and treated or discharged as indicated.
Storm water segregation can easily be incorporated into grass roots refineries, however,
segregation at existing refineries can be difficult. Segregation measures may include sloped or
curbed process unit pads, individual or discrete drain and piping systems, and holding ponds for
testing and controlled releases of wastewater to treatment systems or direct discharge points.
Table 4.6 Refinery Storm Water Management Practices
Refinery Area
Immediate process areas
Developed areas of refinery, but outside
immediate process areas
Undeveloped areas
Storm Water Management
Collection and co-treatment with refinery
process wastewaters.
Segregation, collection and diversion to
storm water holding pond equipped with
oil baffles and skimmers. Controlled
discharge after examination and testing.
Segregation and direct discharge.
29
-------
5. Water Use
Historically, U.S. petroleum refineries have been large water users. Water is used for
contact and non-contact cooling, steam production, and in process operations such as desalting.
Up until the early 1970's, barometric condensers were commonly used, which generated large
quantities of contaminated wastewater. However, during the 1970's, barometric condensers
were mostly replaced by surface condensers which has eliminated this water source.
Since 1972 (the year of the original EPA study of this industry), the petroleum refining
industry has been steadily reducing the amount of water that it uses, and consequently discharges.
Figure 5.1 (from EPA's 1982 Development Document for this industry) presents water use in
this industry from 1972 projected to 1984 as a percent of the water use in 1972. This graph
indicates a steady reduction of water use, such that only 45 percent of the water used in 1972 was
projected to be used in 1984. The BAT regulations promulgated in 1982 did not require any
further flow reductions, however, as a result of litigation, the 1986 amendment to BAT and
NSPS incorporated additional flow reduction as part of the basis for limitations for phenol and
total chromium.
Since the early 1980's, it is believed that refineries have continued to undertake flow
reductions. Data collected as part of this study show water use at many refineries well below 50
percent of the flows predicted by the BPT and BAT flow models. Some refineries are as low as
15 percent of their water use rates predicted by the BPT flow model. Table 5.1 presents a water
use comparison for the 27 refineries surveyed in the EPA/OAR refineries survey between
reported water use and that predicted by the BPT and BAT flow models. The water use rates
shown in Table 5.1 for the 27 refineries average 62 percent of their predicted BPT flows, and 66
percent of their BAT predicted flow rates.
Further review of the data presented in Table 5.2 indicates that the unit process water use
basis for the BPT and BAT flow models may not represent actual refinery practices. As can be
seen in Figure 5.2, the flows predicted by the BPT and BAT flow models for a given refinery can
vary by a factor of over 2 to 1. Although the model accounts for some of this difference, many
refineries have reduced water consumption by using techniques not directly related to specific
refining processes. Techniques such as water reuse, condensate recovery, elimination of leaks,
etc., are not necessarily process unit-specific.
30
-------
Figure 5.1 Water Use Trends
1
E
o
n
8
rom I9T7 Survey and'BIT 1972
Calculated Flow
1PT Questionnaire
Specsa! Sample
Adjusted for Industry Expansion
(gal.&bl)
Median of Belter TTiaa
Model
1972
74
84
31
-------
Table 5.1 Predicted and Actual Wastewater Flows for 27 Refineries
in Ascending Order of Actual/BPT Ratio
No.
8
27
4
21
22
9
19
7
20
24
5
10
14
11
26
12
25
17
23
15
1
16
3
6
2
18
13
Refiner
y
50202
50801
50102
50602
50603
50301
50503
50201
50601
50703
50103
50302
50402
50303
50705
50304
50704
50501
50701
50403
50001
50404
50101
50104
50002
50502
50401
Size
(bbl/day
111,765
64,000
45,400
127,600
100,000
255,000
209,966
46,467
74,200
161,500
316,600
50,000
175,877
70,000
120,300
68,381
151,359
22,319
217,200
132,187
73,100
44,000
57,000
187,033
50,000
105,000
45,856
BPT
Subcategor
D
B
A
B
B
B
E
D
B
B
B
B
B
B
B
D
C
D
D
D
B
B
B
D
B
B
B
Totals
BPT
Flow
(MGD)
4.40
2.20
1.45
3.80
2.70
9.00
7.00
1.50
1.43
6.20
14.40
1.14
5.45
2.80
6.20
1.80
4.60
0.63
12.67
4.67
2.07
1.14
1.30
8.44
2.50
2.84
0.96
113.29
BAT
Flow
(MGD)
3.90
2.60
0.49
2.70
3.20
9.10
6.20
1.70
1.26
4.24
12.57
1.24
5.05
2.70
4.30
1.10
4.80
0.92
9.41
3.13
1.88
0.95
1.43
6.00
4.30
4.40
1.19
100.76
Actual
Flow
(MGD)
0.80
0.40
0.40
1.06
0.78
2.70
2.20
0.51
0.55
2.60
6.60
0.60
2.90
1.62
3.73
1.10
2.83
0.40
8.10
3.30
1.50
1.00
1.20
7.90
3.30
4.21
1.50
63.79
$
*
*
*
*
*
*
*
Avg:
Ratio
Actual/
BPT
0.18
0.18
0.28
0.28
0.29
0.30
0.31
0.34
0.38
0.42
0.46
0.53
0.53
0.58
0.60
0.61
0.62
0.63
0.64
0.71
0.72
0.88
0.92
0.94
1.32
1.48
1.56
0.62
Ratio
Actual/
BAT
0.21
0.15
0.82
0.39
0.24
0.30
0.35
0.30
0.44
0.61
0.53
0.48
0.57
0.60
0.87
1.00
0.59
0.43
0.86
1.05
0.80
1.05
0.84
1.32
0.77
0.96
1.26
0.66
* Discharge to POTW
Actual flow = Facilities total washwater flow - (stormwater and once through cooling water)
32
-------
Figure 5.2. Comparison of Flow Predicted by BPT and BAT Models
1.6-
1,4-
1.2-
ec
ง
0.8
0.4-
I
ActuQl/BAT
1 2 3 4 5 i ? 8 S 1011 12 13141516171819
Wait Number
33
-------
The EPA/OAR survey obtained water use information from the 27 refineries surveyed. Each
facility supplied a water balance diagram from which specific water flows were obtained. Table
5.2 summarizes these data. Flow information for the wastewater sources are summarized below:
Water Source Percent of Total Discharge
Sour water stripper 19.6
Ballast water 4.2
Cooling tower blowdown 18.4
Pump compressors 1.94
Boiler blowdown 6.9
Water treatment 3.1
Desalter 20
Land farm 0.02
Cat. Reformer Scrubber 2.2
Tank draw down 2.4
Total 7876
The remaining 21 percent cannot be accounted for. This is because most refineries do not
have flow monitoring stations at their in-process discharge points, and therefore cannot complete
a detailed water balance.
Data on water use was also collected from the six refineries visited as part of this study.
Water use, as compared to their BPT and BAT flow model rates, is shown in Table 5.3. The
four California refineries average 0.46 of their BPT flows, and 0.67 of their BAT model flows.
The Texas and Pennsylvania refineries average 1.17 of their BPT model flows and 1.08 of their
BAT model flows. Refineries located where there are water shortages and/or stringent local
water quality standards have made great strides in reducing water usage.
Environmental concerns have driven refineries to produce additional, significant wastewater
as a result of compliance efforts.
Resource Conservation and Recovery Act (RCRA) or Clean Air Act regulations have
resulted in refineries closing certain wastewater ponds, which has reduced evaporation of
wastewater in the system.
34
-------
The need to more frequently and more rigorously test tank and pipeline integrity produces
large quantities of hydrotest water, up to several million gallons at a time.
Some refineries have discovered groundwater contamination (due to past failures in tank
or pipeline integrity). To remediate this contamination may generate up to several
million gallons of groundwater per day, which may need to be treated and discharged
through the NPDES outfall.
Additional U.S. Coast Guard requirements for accepting ballast water from vessels.
35
-------
Table 5.2 Selected Sources of Wastewaters
Facil.
No.
50001
50002
50101
50102
50103
50104
50201
50202
50301
50302
50303
50304
50401
50402
50403
50404
50501
BPT
Sub.
B
B
B
A
B
D
D
D
B
D
B
D
B
B
D
B
B
DryWt
Process
Flow*
0.44
3.3
1.4
0.40
6.6
6.57
0.5
0.5
2.7
0.6
0.33
1.1
0.7
2.9
3.3
1.0
0.40
Wastewater Sources in gallons per day
Stripper
270,720
N/A
N/A
None
1,022,400
475,200
104,000
720
N/A
1,440
142,000
None
172,800
230,400
201,600
1,350,000
None
Ballast
30,420
N/A
23,040
N/A
20,000
N/A
N/A
N/A
N/A
0
N/A
N/A
N/A
330
N/A
N/A
N/A
Cooling
Tower
Blowdn
139,680
N/A
191,520
28,800
706,000
N/A
53,280
193,000
500,000
260,000
N/A
N/A
64,800
900,000
606,240
100,000
115,200
Pumps
Com-
press.
N/A
N/A
27,340
N/A
None
N/A
N/A
N/A
N/A
N/A
N/A
N/A
8,640
N/A
4,320
N/A
N/A
Boiler
B.D.
30,240
N/A
158,400
21,600
None
N/A
40,000
25,200
497,000
86,400
N/A
N/A
7,200
None
20,160
10,000
43,300
Storm
6,067
N/A
604,800
1,380,000
476,640
17,280
214,000
374,000
102,000
N/A
N/A
20,160
170,000
233,300
75,000
390,000
Water
Treatmt
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Desalter
86,000
N/A
130,000
144,000
N/A
N/A
54,900
254,900
625,000
95,000
104,000
144,000
93,600
N/A
N/A
110,000
56,160
Land
Farm
N/A
N/A
0
N/A
N/A
N/A
90
N/A
None
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Cat
Reform
Scrub
23,040
N/A
14,247
N/A
28,800
N/A
9,700
21,600
150,000
N/A
N/A
535
N/A
128,000
1,440
N/A
18,720
Tank
Draw
Down
3,600
N/A
1,370
28,800
864,00
140,00
4,526
5,600
144,00
1,440
613
432
1,440
4,320
N/A
100
2,850
36
-------
Table 5.2 Selected Sources of Wastewaters
Facil.
No.
50502
50503
50601
50602
50603
50701
50702
50703
50704
50705
50801
BPT
Sub.
C
E
B
B
B
D
DryWt
Process
Flow*
5.2
2.2
0.55
0.73
1.64
8.10
Wastewater Sources in gallons per day
Stripper
705,600
None
259,200
259,200
504,000
1,670,000
Ballast
N/A
N/A
72,000
72,000
72,000
89,300
Cooling
Tower
Blowdn
2,400,00
400
144,000
144,000
28,800
1,340,00
Pumps
Com-
press.
N/A
N/A
N/A
N/A
N/A
N/A
Boiler
B.D.
400,000
1,444
21,600
43,200
43,200
735,900
Storm
N/A
720,000
43,200
43,200
72,000
2,082,200
Water
Treatmt
N/A
N/A
36,000
7,200
28,800
N/A
Desalter
280,000
443,000
221,760
161,280
316,810
N/A
Land
Farm
N/A
N/A
N/A
N/A
N/A
N/A
Cat
Reform
Scrub
5,000
3,000
7,200
1,440
7,200
500
Tank
Draw
Down
100,00
288,00
7,200
7,200
Did not answer the questionnaire
D
C
B
B
2.6
2.83
3.73
0.5
1,186,300
221,800
907,200
37,500
N/A
N/A
N/A
1,440
684,000
223,200
82,100
None
N/A
N/A
165,16
None
360,000
278,000
59,040
50,000
43,200
1,641,600
N/A
N/A
N/A
N/A
N/A
N/A
470,000
N/A
527,000
150,000
N/A
N/A
N/A
N/A
55,000
266,300
90,410
7,200
38,500
1,440
116,50
1,472
* million gallons per day
37
-------
Table 5.3 Water Use: Six Site Visits
Refinery
Chevron, Richmond, CA
Shell, Martinez, CA
Unocal, Rodeo, CA
Tosco, Martinez, CA
Phillips, Borger, TX
Chevron, Philadelphia, PA
Flow Ratios
Actual/BPT
0.285
0.51
0.72
0.31
1.48
0.86
Actual/BAT
0.50
0.74
0.80
0.63
0.86
1.31
38
-------
6. Pretreatment Standards Review And Catalytic Reformer Issues
6.1 Indirect Discharging Refineries
EPA's 1976 survey of this industry identified 44 indirect discharging refineries. Table 6.1
lists these facilities, along with their location and refining capacities. This list was compared to
the Oil and Gas Journal 1991 list of operating refineries. Only 22 indirect discharging facilities
are now believed to be in operation, and their 1991 refinery capacities are also shown in Table
6.1. As can be seen, a greater proportion of smaller refineries have closed since 1976.
Although the number of indirect dischargers have been reduced by one half, total capacity of
indirect discharge refineries has only dropped by 25 percent.
6.1.1 Treatment Technologies
The current pretreatment standards (PSES and PSNS) are based on the use of oil/water
gravity separators and in-plant sour water stripping for ammonia. However, indirect discharging
refineries use a variety of technologies. Table 4.3 presents a summary of the technologies in
place at the seven indirect refineries included in the EPA/OAR survey. These seven facilities
use a range of technologies including enhanced oil removal (four facilities), biological treatment
(three facilities), sand filtration and activated carbon (one facility).
Effluent Characteristics
Two data collection efforts were undertaken as part of this study to obtain effluent quality
data from indirect discharging refineries. The first source of data was the Los Angeles County
Sanitation Districts (LACSD) which have 14 indirect discharging refineries. LACSD has
developed a comprehensive pretreatment program, in which the refineries have had to install
enhanced oil removal systems such as dissolved air flotation. Table 6.2 presents a detailed
summary of the data collected from these 14 facilities.
Data was also obtained from three other indirect discharging refineries and is summarized in
Table 6.3. The facilitiesLa Gloria Oil and Gas Company, Tyler, Texas; Derby Refinery,
Wichita, Kansas; and Clark Oil, Blue Island, Illinois-were selected to represent refinery
discharges to smaller sewer systems that do not have as comprehensive a pretreatment program
as LACSD.
39
-------
Table 6.1 Summary Comparison of Locations and Capacities for Indirect
Dischargers Between 1976 and 1991
Name
Flint Chemical Co.
Mid-America Refining Co.,
Inc.
Lunday Thagard Oil Co.
Eddy Refining Co.
Chevron U.S.A., Inc.
Northland Oil & Refining Co.
CRA, Inc.
Lakeside Refining Co.
Crystal Refining Co.
Edgington Oil Co., Inc.
Sigmor Refinery Co.
Western Refining Co.
MacMillan Ring-Free Oil Co.
Beacon Oil Co.
Saber Refining Co.
Chevron U.S.A. Inc.
U.S.A. Petrochem Corp.
Golden Eagle Refining Co.,
Inc.
Amoco Oil Co.
Ashland Petroleum Co.
Fletcher Oil & Refining Co.
Winston Refining Co.
1976 1991
Capacity Capacity
(1000 (1000
Location bbl/day) bbl/day)
San Antonio, TX
Chanute, KS
South Gate, CA
Houston, TX
Richmond Beach,
WA
West Dickinson, ND
Scottsbluff, NE
Kalamazoo, MI
Carson City, MI
Long Beach, CA
Three Rivers, TX
Woods Cross, UT
Long Beach, CA
Hanford, CA
Corpus Christi, TX
Portland, OR
Ventura, CA
Carson, CA
Baltimore, MD
Findlay, OH
Carson, CA
Fort Worth, TX
1.0
3.0
3.2
3.250
5.0
5.25
5.38
5.92
6.0
10.0
10.0
10.0
12.2
12.4
13.0
14.0
15.2
16.0
17.0
20.0
20.0
20.0
1.9
-
7.0
-
-
-
-
5.6
4.0
41.6
53.0
-
-
-
-
16.0
-
-
-
-
29.657
-
40
-------
Table 6.1 Summary Comparison of Locations and Capacities for Indirect
Dischargers Between 1976 and 1991
Name
Continental Oil Co.
Ashland Petroleum Co.
Husky Oil Co. of Delaware
LaGloria Oil & Gas Co.
Derby Refining Co.
Pride Refining, Inc.
Amoco Oil Co.
Delta Refining Co.
Mobil Oil Corp.
Powerine Oil Co.
Rock Island Refining Corp.
Quintana-Howell Joint Venture
Douglas Oil Co.
Gulf Oil Co., U.S.A.
Ashland Petroleum Co.
Marathon Oil Co.
Clark Oil and Refining Corp.
Texaco Inc.
Shell Oil Co.
Crown Central Petroleum
Corp.
Union Oil Co. of California
Mobil Oil Corp.
Totals
Location
Wrenshall, MN
Louisville, KY
North Salt Lake, UT
Tyler, TX
Wichita, KS
Abiline, TX
Salt Lake City, UT
Memphis, TN
Buffalo, NY
Santa Fe Springs,
CA
Indianapolis, IN
Corpus Christi, TX
Paramount, CA
Santa Fe Springs,
CA
Tonawanda, NY
Detroit, MI
Blue Island, IL
Wilmington, CA
Carson, CA
Pasadena, TX
Wilmington, CA
Torrance, CA
1976
Capacity
(1000
bbl/day)
24.0
25.0
25.0
29.3
29.9
36.5
40.4
43.9
44.0
44.1
44.5
46.0
48.0
53.8
63.0
66.0
70.0
80.0
93.0
100.0
111.0
131.1
1,476.3
1991
Capacity
(1000
bbl/day)
-
-
-
49.5
29.925
45.5
40.0
-
-
46.5
50.0
-
42.7
44.0
-
70.0
66.5
95.0
133.3
-
108.0
123.0
1,102.7
41
-------
Table 6.2 Summary of Discharge Data for Major Refineries
Discharging to Los Angeles County Sanitation Districts (LACSD)
January 1,1990 through February 9,1993
Analyte
DH
Suspended Solids (mg/1)
Ammonia Nitrogen (mg/1)
Total Cyanide (mg/1)
Soluble Sulfide (mg/1)
Thiosulfate (mg/1)
Sulfate (mg/1)
Sulfite (mg/1)
Mercaptans (mg/1)
Phenols (mg/1)
Total COD (mg/1)
Oil & Grease (mg/1)
Non-polar Oil & Grease
Benzene (ug/1)
Toluene (ug/1)
Ethyl Benzene (ug/1)
o-xylene (ug/D
p-xylene (ug/D
m+p xylene (ug/D
Chloroform (ug/D
Naphthalene (ug/D
2,4 Dimethyl phenol (ug/D
Chrysene (ug/D
Fluorene (ug/D
Phenanthrene (ug/D
2-Chlorophenol (ug/D
2,4,6-Trichlorophenol (ug/D
Acenaphthene (ug/D
Pyrene (ug/D
Total Chromium (mg/)
Lead (mg/1)
Zinc (mg/1)
Maxim
10.08
222.37
99.97
0.74
1.99
61.19
984.19
2.20
2.25
66.06
1613.05
272.20
53.75
5523.89
7400.56
572.64
1357.78
1366.25
2385.88
800.00
143.60
359.50
10.50
6.50
15.00
22.00
54.00
2.00
1.00
0.16
0.48
2.00
Minim
6.50
11.42
4.09
0.05
0.05
0.80
888.64
0.50
0.10
2.73
159.05
4.12
27.40
87.71
97.30
42.73
47.94
674.37
79.69
800.00
143.60
359.50
10.50
6.50
15.00
22.00
54.00
2.00
1.00
0.03
0.11
0.19
Average
7.89
62.45
29.73
0.12
0.15
9.59
935.57
0.62
0.65
22.77
632.96
51.14
40.59
1419.90
1783.86
167.82
340.39
1066.53
630.60
800.00
143.60
359.50
10.50
6.50
15.00
22.00
54.00
2.00
1.00
0.04
0.36
0.42
Note: In calculating averages, one-half of the detection limit was used when
"Less than" values were reported.
42
-------
Table 6.3 Summary of Discharge Data for Major Refineries
Discharging to Local POTWs
Analyte
Suspended Solids
(mg/1)
Maximum
Minimum
Mean
Ammonia
Nitrogen (mg/1)
Maximum
Minimum
Mean
Total Cyanide
(mg/1)
Maximum
Minimum
Mean
Phenols (mg/1)
Maximum
Minimum
Mean
Total COD (mg/1)
Maximum
Minimum
Mean
Oil & Grease
(mg/1)
Maximum
Minimum
Mean
Total Chromium
(mg/1)
Maximum
Minimum
Mean
LaGloria Oil
and Gas Co.
720.00
40.00
126.61
38.00
0.60
9.72
3.50
<0.02
0.45
2597.00
82.00
328.31
47.00
1.05
8.07
Derby
Refining
248.00
1.00
25.68
32.6
10.6
21.35
0.19
0.01
0.04
190.00
11.70
62.05
1196.22
244.00
557.46
326.00
7.80
78.60
1.36
0.36
0.86
Clark
Oil
85.30
0.90
18.38
0.12
0.09
0.10
95.00
2.00
36.16
2.18
1.01
1.62
Average
484.00
20.50
76.14
51.96
4.03
16.48
0.15
0.05
0.07
96.75
<5.86
31.25
1896.61
163.00
442.88
156.00
3.61
40.94
1.77
0.68
1.24
43
-------
Table 6.3 Summary of Discharge Data for Major Refineries
Discharging to Local POTWs
Analyte
Lead (mg/1)
Maximum
Minimum
Mean
Zinc (mg/1)
Maximum
Minimum
Mean
Sulfide
Maximum
Minimum
Mean
BOD
Maximum
Minimum
Mean
Total VOC's
Maximum
Minimum
Mean
Total Dissolved
Solids
Maximum
Minimum
Mean
pH
LaGloria Oil
and Gas Co.
<0.10
<0.10
<0.10
12.00
<0.50
1.43
246.00
21.50
99.08
2160.00
130.00
1121.61
9.60
5.90
7.54
Derby
Refining
0.10
0.06
0.08
13.10
0.10
3.35
8.72
1.46
4.98
9.20
8.20
8.90
Clark
Oil
0.99
0.13
0.58
9.80
8.00
8.77
Average
<0.10
<0.08
<0.09
0.99
0.13
0.58
12.55
<0.30
2.39
343.00
16.25
165.08
8.72
1.46
4.98
2160.00
130.00
1121.61
9.53
7.36
8.40
A summary is contained in Table 6.4 comparing effluent values for seven selected pollutants
for which the data is shown in Tables 6.2 and 6.3. This comparison of the data indicates that the
LACSD refineries are doing slightly better in the removal of oil (oil and grease) and total
suspended solids.
44
-------
Table 6.4 Data Comparison: Indirect
Discharging Refineries
Pollutant
Oil & Grease
Suspended
Solids
Benzene
Toluene
Phenols
Lead
Zinc
Average Data (mg/1)
LACSD
37.5
62.45
0.83
1.00
21.64
0.36
0.35
Other
Refineries
40.94
76.14
31.25
0.09
0.58
6.2 Dioxins in Catalytic Reformer Wastewaters
Polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (CDDs and CDFs,
respectively) are closely related families of highly toxic and persistent organic chemicals which
are formed as unwanted by-products in some commercially significant chemical reactions, during
high temperature decomposition and combustion of certain chlorinated organic chemicals, and
through other reactions involving chlorine and organic materials. CDDs and CDFs constitute a
family of over 200 related chemical compounds with varying chemical, physical, and
toxicological properties. The congener that appears to be the most toxic and has generally raised
the greatest health concerns is 2,3,7,8-tetrachlorodibenzo-/>-dioxin, abbreviated as 2378-TCDD.
Unfortunately, CDDs and CDFs are among the most persistent as well as the most toxic
pollutants. Certain congeners, including 2378-TCDD, are highly bioaccumulative and
lipophilic. The U.S. Centers for Disease Control has estimated the half life of 2378-TCDD in
the environment to be about 12 years.
45
-------
In 1988, CDDs and CDFs were found in internal waste streams at refineries in Canada.
Further studies at refineries in the United States and Canada located the source of the CDDs and
CDFs to be the regeneration of catalyst for the catalytic reforming operations. In particular the
source was identified as the caustic and rinse wastewaters from certain types of regeneration
processes.
One such study was conducted by the EPA Engineering and Analysis Division. The
objective of this study was to verify the analytical method for measuring CDD's and CDF's in
refinery wastewater matrices, and to screen and characterize the wastewaters from the catalytic
reforming catalyst regeneration processes for formulation of CDD' s and CDF's. This report is
included as Appendix G to this Preliminary Data Summary.
Catalytic reformers can be categorized by the type of catalyst regeneration system employed.
The three major types of regeneration are:
1. Semi-Regenerative. Characterized by the shutdown of the entire reforming unit at
specified intervals for in situ regeneration of the catalyst. Regenerations are generally
limited to one or two per year.
2. Cyclic. Characterized by continual regeneration of the catalyst in situ in one of several
reactors that is isolated from the naptha feed during regeneration. The remaining reactors
continue reforming naphtha while regeneration of the catalyst occurs in the isolated reactor.
There may be several regeneration cycles each year since one of the reactors is usually being
operated in a regeneration mode.
3. Continuous. A portion of the catalyst is continually removed from the reformer,
regenerated in a separate reactor, and returned to the reformer.
In all cases, the purpose of catalyst regeneration is to remove accumulated coke from the
catalyst under controlled combustion conditions and to replenish the catalyst with chlorine which
is necessary for catalytic reactions to occur. Chlorine may be added in the form of chlorine gas,
hydrochloric acid, or any of a number of chlorinated compounds including carbon tetrachloride,
trichloromethane, dichloropropane, and dichlorethane. Reactions conditions of temperature,
pressure and presence of free chlorine radicals and CDD and CDF precursors (various
unchlorinated polycyclic compounds) are such that the potential exists for the formation of
CDD's and CDF's during the catalyst regeneration cycle.
The off-gases from the regeneration processes contain combustion products, hydrochloric
acid, and water vapor. Depending upon design considerations (materials of construction, etc.)
off-gases may be scrubbed with a caustic or water solution, or vented directly to the atmosphere.
Caustic scrubbing is more common at semi-regenerative reformers and generally not practiced at
refineries with cyclic reformers.
46
-------
6.2.1 Waste Characterization
As a result of these discoveries, refineries in both Canada and the U.S. have conducted waste
characterization studies of their catalytic reformer regeneration wastes and their refinery final
effluents in order to identify the presence of CDD's and CDF's.
Tables 6.5 and 6.6 present data resulting from sample and analysis from two Canadian
refineries.
47
-------
Table 6.5 Shell Canada Products Limited, Sarnia Refinery
Range of Dioxins/Furans in Internal Shell Wastewaters
all figures in parts per billion
Sample
2,3,7,8,8-4 CDD
4CDD (Total)
5CDD
6CDD
7CDD
8CDD
New
Caustic
ND**
ND
ND
ND
ND
ND
Spent
Caustic
0.0046 - 0.0054
0.350-5.900*
0.400 - 8.200*
0.530-5.300*
0.290-1.500*
0.230-1.300*
Scrubber
Water
ND
0.0052-0.012
0.00076 - 0.002
0.00076-0.0012
0.00091 -0.0014
0.00052-0.00071
Biological
Sludge
ND
9.9-15
9.0 - 17
17.0-23
8.7 - 12
7.0-8.4
Combined
Effluent
ND
ND
ND
ND
ND
ND
Total Dioxins *** 1.810-22.200 0.00815-0.01731 51.6-75.4
4CDF
5CDF
6CDF
7CDF
8CDF
ND
ND
ND
ND
ND
0.380-6.1*
0.680-8.9*
1.200-5.6*
1.400-4.2*
0.760-2.5*
0.0051-0.010
0.0016-0.0038
0.0031 -0.0057
0.0049 - 0.0084
0.00071-0.0008
8.2 - 10.0
12.0 - 16.0
24.0-31.0
28.0-40.0
20.0-28.0
ND
ND
ND
0.00034 - 0.00022
ND
Total Furans *** 4.420-27.3 0.01541-0.0287 92.2-125
* July samples by Shell; others are November Samples taken by Ontario MOE; all November samples were done
in duplicate
** Non-detectable
*** Totals may not add because highest and lowest concentrations for individual types of dioxins and furans did
not occur in same samples.
48
-------
Table 6.6 Esso Petroleum Canada, Sarnia Refinery
Powerformer Regeneration Study
Dioxin and Furan Concentrations, December 16, 1988
(parts per trillion)
PCDF/PCDD
Isomer Group
2378-TCDD
4-TCDD
5-PCDD
6-HCDD
7-HCDF
8-OCDF
Total Dioxins
4-TCDF
5-PCDF
6-HCDF
7-HCDF
8-OCDF
Total Furans
2378-TCDD
Toxic
Equivalent
Scrubber Water
1st 24
hours
< 0.0005
0.11
0.030
0.015
0.042
0.037
0.243
0.32
0.14
0.071
0.026
0.071
0.628
0.243
2nd 24
hours
< 0.012
0.111
0.072
0.027
0.041
0.063
0.308
0.48
0.17
0.11
0.077
0.21
1.038
0.345
BIOX Inlet
1st 24
hours
< 0.002
< 0.005
< 0.004
<0.04
<0.04
0.000
< 0.002
< 0.003
<0.02
<0.01
<0.02
0.000
0.000
2nd 24
hours
< 0.009
< 0.005
< 0.007
<0.02
<0.06
0.000
0.051
0.029
< 0.007
< 0.008
<0.03
0.080
0.040
BIOX Outlet
1st 24
Hours
< 0.003
< 0.008
< 0.007
<0.01
<0.03
0.000
< 0.002
< 0.008
< 0.006
< 0.007
<0.02
0.000
0.000
2nd 24
Hours
< 0.008
< 0.004
< 0.004
<0.01
<0.03
0.000
< 0.002
< 0.004
< 0.002
< 0.004
<0.02
0.000
0.000
Notes
4-TCDD includes 2378-TCDD
Totals and toxic equivalents do not include values below MDL. Assume less than
detectable means zero.
CDD and CDF information was also collected as part of the six refinery visits conducted as
part of this study. Table 6.7 presents a summary of catalytic reformer usage in the six refineries
visited, and the availability of CDD/CDF analytical data. Table 6.8 through Table 6.10 present
CDD/CDF analytical data obtained from three of the refineries from which information was
obtained. In all cases, 2378-TCDD and 2378-TCDF levels in the refinery effluents were
non-detectable.
49
-------
Table 6.7 Catalytic Reforming Data: Six Site Visits
Refinery
Chevron
Richmond, CA
Shell
Martinez, CA
Unocal
Rodeo, CA
Tosco
Martinez, CA
Phillips
Borger, TX
Chevron
Philadelphia, PA
Catalytic
Reforming
yes
yes
yes
yes
yes
yes
Semi-
Regeneration
yes
no
yes
yes
yes
yes
Sample Data
Available
yes
N/A
yes
yes
no
no
50
-------
Table 6.8 Summary of CDD/CDF Data for Chevron
Richmond Refinery
Furans
TCDFs (total)
2,3,7,8-TCDF
PeCDFs (total)
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
HxCDFs (total)
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
HpCDFs (total)
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDF
Dioxins
TCDDs (total)
2,3,7,8-TCDD
PeCDDs (total)
1,2,3,7,8-PeCDD
HxCDDs (total)
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
HpCDDs (total)
1,2,3,4,6,7,8-HpCDD
OCDD
Grab - 1
86,000
5,500
120,000
15,000
7,600
87,000
24,000
9,400
2,000
2,600
61,000
28,000
12,000
17,000
Grab - 1
11,000
280
12,000
1,200
13,000
1,200
1,500
710
7,900
4,300
1,900
Grab - 2
19,000
1,200
23,000
3,300
1,600
20,000
5,400
2,200
640
800
14,000
6,500
3,000
3,500
Grab - 2
2,300
58
2,700
270
2,900
280
340
180
1,800
990
440
Notes
1. Two grab samples taken on July 16, 1991, from one batch of all
wastewaters resulting from a catalytic regeneration. The analysis
was by Method 8290.
2. Units in picograms/liter or parts per quadrillion (pg/1 or ppg)
51
-------
Table 6.9 Summary of CDD/CDF Data for Tosco Martinez Refinery
Furans
TCDFs (total)
2,3,7,8-TCDF
PeCDFs (total)
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
HxCDFs (total)
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
HpCDFs (total)
1,2,3,4,6,7,8-HpCD
1,2,3,4,7,8,9-HpCD
OCDF
#2
Ref
Prim.
Burn
170
270
13.0
43.0
450
30.0
37.0
54.0
21.0
240.0
98.0
32.0
59.0
#2
Ref
Mkup
H20
2.00
0.580
2.40
0.120
0.250
3.40
0.290
0.270
0.130
0.290
1.50
0.750
0.150
0.280
#2
Ref
Mkup
H20
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
#3
Ref
Regen
750
1100
75.0
180
2300
227
225
300
29.0
1800
1100
1100
500
#3
Ref
clean
cond.
0.081
ND
0.100
ND
ND
0.140
ND
ND
ND
ND
0.068
ND
ND
ND
E-00
1
Back-
grnd
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
E-00
1
Post
0.023
0.013
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
E-001
Post
Reana
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Dioxins
TCDDs (total)
2,3,7,8-TCDD
PeCDDs (total)
1,2,3,7,8-PeCDD
HxCDDs (total)
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
HpCDDs (total)
1,2,3,4,6,7,8-HpCD
OCDD
ND
ND
18.0
12.90
0.100
ND
0.350
0.020
0.460
0.021
0.042
0.026
0.330
0.170
0.170
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
3.00
ND
ND
ND
270
7.60
29.0
32.0
230
110
130
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
1.20
All results in picograms/milliliter (parts per trillion)
52
-------
Table 6.10 Summary of CDD/CDF Data for Unocal
Rodeo Refinery
Test Parameters
Refinery # 231
9/16/90
2/23/91
Refinery #
244
6/27/91
Furans
TCDFs (total)
2,3,7,8-TCDF
PeCDFs (total)
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
HxCDFs (total)
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
HpCDFs (total)
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDF
1200
230
1500
170
240
3100
830
430
390
340
2600
2100
1300
5300
2300
61
2800
160
260
2000
190
290
190
330
1600
790
260
440
2900
160
270
36
38
48
11
11
5.4
9.3
58
35
13
17
Dioxins
TCDDs (total)
2,3,7,8-TCDD
PeCDDs (total)
1,2,3,7,8-PeCDD
HxCDDs (total)
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
HpCDDs (total)
1,2,3,4,6,7,8-HpCDD
OCDD
550
47
500
120
1700
150
250
370
2600
1700
3300
600
14
1500
130
1100
120
200
190
1700
950
890
65
5.3
22
3.9
15
ND(1.9)
ND(3.1)
ND (2.4)
14
8.3
21
Note : Results in picograms/liter (ppq) for samples of regeneration
wastewater
53
-------
6.2.2 Available Treatment Technologies
The discovery of the presence of CDD/CDF's in refinery wastewaters only occurred in the
last three to five years. As a result, there have been limited studies of this problem, and only
limited data on available technologies. The following sections present a summary of currently
available control information.
Flow Reduction
There is a very large range in water use found during regeneration of the catalyst from
refinery to refinery. As a result, there may be opportunities to minimize the volumes of scrubber
waters used at certain facilities. Modifications to off-gas cooling an scrubbing systems may be
possible.
As can be seen by the waste characterization data presented earlier, there is a wide range in
concentrations found. This may be caused by various process techniques used at each facility.
Investigation into the causes of these variations in pollutant concentrations would be needed in
order to determine whether there are in-process techniques that can reduce the quantities of
CDD/CDF's generated.
Pretreatment
Since available data from Ontario and U.S. refineries indicate the more toxic CDD's and
CDF's are found only in wastewaters from catalytic reforming (regeneration) operations, the
most effective means to achieve minimum mass discharge of these compounds is to isolate an
treat the low volume catalytic reforming regeneration process wastewaters prior to mixing with
other refinery process or cooling waters, or stormwater. Accordingly, the regeneration process
wastewaters should be collected and isolated in each refinery in appropriately sized equalization
or holding tanks prior to treatment. The principal purposes of the holding or equalization tanks
are to provide for temporary storage of reforming regeneration wastewaters and to provide for
low volume constant feeds, thus allowing for design of downstream treatment systems at low
hydraulic loading rates.
Based upon investigations by Shell Canada Products Limited, catalytic reforming
regeneration wastewaters are characterized by relatively low concentrations of very fine
suspended particulates. CDD's and CDF's are most often associated with paniculate matter in
wastewater matrices. Hence, relatively simple technologies such as conventional gravity settling
or mixed media filtration that are incapable of fine particulate removal would not be effective for
removal of CDD's and CDF's from catalytic reforming wastewaters.
The more advanced adsorption and membrane technologies require fairly clean feed streams
in terms of TSS to prevent fouling and plugging. Although it appears that untreated catalytic
reforming regeneration wastewaters do not contain TSS at levels likely to cause operating
54
-------
problems in downstream units, consideration of pretreatment by filtration is recommended.
Also, depending upon the reforming operation and the type of regeneration system and gas
scrubbing system, the untreated wastewaters may be highly alkaline and unsuitable for direct feed
to downstream treatment units. In these cases, neutralization with acid may be necessary.
Granular Activated Carbon
Adsorption on granular activated carbon offers several advantages over membrane
technologies for removal of CDD's and CDF's from catalytic reforming wastewater. The
technology is suitable for treatment of relatively large volumes of wastewaters contaminated with
adsorbable organic contaminants at low levels. Aside from spent carbon, there are no
by-product sludges or concentrated aqueous streams requiring further processing or treatment for
ultimate disposal. Multiple carbon units can be used in parallel or series to ensure maximum
removal. Finally, since catalytic reformer wastewater streams are relatively low in organic
content, the life of the carbon beds should be relatively long, on the order of a few years as
opposed to weeks or months.
Two Canadian refineries have installed temporary activated carbon treatment facilities and
have applied for Certificates of Approval for permanent wastewater treatment facilities.
Treatability and performance data from these systems are summarized in Tables 6.11 and 6.12.
These data indicate consistently high removal rates for CDD's and CDF's (> 95 percent).
Suncor recently reported consistent removal from current operations to non-detectable levels in
the low parts per quadrillion range (ppq). Shell reported more than 96 percent removal during
recent testing. All Ontario refineries have reported consistently no detection of the more toxic
CDD and CDF congeners in treated refinery process wastewater effluents.
55
-------
Table 6.11 Suncor-Sarnia Catalytic Reformer
Wastewater Treatment
August 3,1990 Samples
Analyte
2378-TCDD
TCDDs
PeCDDs
HxCDDs
HpCDDs
OCDD
2378-TCDF
TCDFs
PeCDFs
HxCDFs
HpCDFs
OCDF
2378-TCDD TEQ*
Carbon Filter
Influent
ND (10)
260
310
700
210
83
300
1,400
2,000
3,900
1,300
530
2,520
Effluent
ND (10)
ND (10)
ND (10)
ND (20)
ND (30)
ND (30)
ND (10)
ND (10)
ND (10)
ND (10)
ND (20)
ND (20)
34.5
Results in parts per trillion (ppt)
* 2378-TCDD TEQ computed assuming CDDs and CDFs
were present at detection levels when not-detected results
were obtained.
Removal Efficiency >98.6%
Source: October 18, 1990, letter from T.A. Brown, Suncor
to L.Van Asseldonk, Ontario MOE-Sarnia.
56
-------
Table 6.12 Shell Canada-Sarnia, Catalytic Reformer
Wastewater Treatment
Spent Caustic, May 9,1991
Analyte
2378-TCDD
TCDDs
12378-PeCDD
PeCDDs
123478-HxCDD
123678-HxCDD
123789-HxCDD
HxCDDs
1234678-HpCDD
HpCDDs
OCDD
2378-TCDF
TCDFs
12378-PeCDF
23478-PeCDF
PeCDFs
123478-HxCDF
123678-HxCDF
234678-HxCDF
123799-HxCDF
HxCDFs
1234678-HpCDF
1234789-HpCDF
HpCDF
OCDF
2378-TCDD TEQ
Carbon Filter
Influent
<5.5
95
15
120
8.5
25
< 5.8
140
86
140
90
54
210
13
27
350
220
120
47
27
580
400
69
590
260
431
Effluent
0.21
5.9
0.3
7.3
0.35
0.8
0.84
8
3.9
7.2
4.7
0.43
8.5
1.6
1.3
16
9.2
3.5
1.8
< 0.057
28
18
3.2
29
15
4.54
Results in parts per trillion (ppt) Removal Efficiency 98.9%
Source: June 26, 1991, letter from D. Atwell, Shell Canada to
A. Peterson, Ontario MOE-Sarnia.
57
-------
7. Evaluation of Pollutant Discharges And Environmental Issues
The purpose of this section is to present a preliminary assessment of the pollutant loadings
and potential water quality impacts of discharges from petroleum refining facilities to surface
waters and publicly-owned treatment works (POTWs). Using readily available data and
information sources on refinery wastewater volume and constituents, annual loadings and
average concentration are estimated. In addition, potential aquatic life and human health
impacts are summarized based on a review of documented environmental impacts and a review
of the physical-chemical properties and toxicity of pollutants associated with wastewater
discharges from the petroleum refining industry. The following sections of this report describe
the methodology and results (including data sources and assumptions/limitations) used in the
identification of documented environmental impacts, the identification and quantification of
pollutant releases, and the evaluation of the fate and toxicity of released pollutants. Additional
details on specific information addressed in this section are presented in the Appendices.
7.1 Identification And Quantification of Pollutant Releases
Petroleum refining wastewater constituents are identified using two EPA data bases: the
Permit Compliance System (PCS) and the Toxic Release Inventory (TRI). The identified
constituents are listed on Table 7.1. Annual loadings are obtained from both PCS and TRI data
for a variety of parameters including conventional, priority, and non-conventional pollutants.
TRI encompasses direct and indirect discharges, whereas PCS covers direct discharges only.
Average pollutant concentrations are also retrieved from PCS for analysis. A brief description
of each data base, the methodology to identify and quantify releases, add the assumptions and
limitations of the analyses are described below.
Table 7.1 Refinery Wastewater Constituents
Constituent
1,1,1 -Trichloroethane
1,1,2-Trichloroethane
1 ,2-Dibromoethane
1 ,2-Dichlorobenzene
1 ,2-Dichloroethane
1,2,4-Trimethylbenzene
1,3 -Butadiene
1 ,4-Dichlorobenzene
CAS Number
71556
79005
106934
95501
107062
95636
106990
106467
Number of
1992 PCS
Parameters
0
0
0
1
0
0
0
1
Number of
1991 TRI
Parameters
1
1
1
0
1
1
1
0
Number of 1992
TRI Parameters
1
1
1
0
1
1
1
0
58
-------
Table 7.1 Refinery Wastewater Constituents
Constituent
2 -Methoxy ethanol
2,3,7,8-Tetrachloro-dibenzo-p-di
oxin
2,4-Dimethylphenol
2,4,6-Trichlorophenol
Acetone
Acetonitrile
CAS Number
109864
1746016
105679
88062
67641
75058
Alkalinity
Alkalinity/Hardness (CaCO3)
Aluminum
Ammonia
Ammonium Sulfate (Solution)
Anthracene
Antimony
Arsenic
Barium
Benzene
7429905
7664417
7783202
120127
7440360
7440382
7440393
71432
Benzene,Toluene,Ethylbenzene,Xylene (BTEX)
Benzo(a)anthracene
Benzo(a)pyrene
56553
50328
Biological Oxygen Demand (BOD)
Biphenyl
Bis (2-ethylhexyl) Phthalate
Bromide
Cadmium
Carbon Tetrachloride
92524
117817
24959679
7440439
56235
Chemical Oxygen Demand (COD)
Number of
1992 PCS
Parameters
0
1
0
1
0
0
1
2
1
4
0
1
0
2
0
1
1
1
1
o
J
0
1
1
1
0
5
Number of
1991 TRI
Parameters
1
0
1
0
1
1
0
0
0
1
1
1
1
1
2
1
0
0
0
0
1
0
0
0
1
0
Number of 1992
TRI Parameters
1
0
1
0
1
1
0
0
0
1
1
1
1
2
2
1
0
0
0
0
1
0
0
0
1
0
59
-------
Table 7.1 Refinery Wastewater Constituents
Constituent
Chloride
Chlorine
Chlorine Dioxide
Chloroform
Chromium
Chromium, Hexavalent
Chromium, Trivalent
Chrysene
Cobalt
Copper
Cresol (Mixed Isomers)
Cumene
Cyanide
Cyclohexane
Diethanolamine
CAS Number
16887006
7782505
10049044
67663
7440473
18540299
16065831
218019
7440484
7440508
1319773
98828
57125
110827
111422
Dissolved Oxygen (DO)
Ethylbenzene
Ethylene
Ethylene Glycol
Fluoride
Formaldehyde
100414
74851
107211
16984488
50000
Glycol Ethers
Hexachlorobenzene
118741
Hydrocarbons
Hydrogen Cyanide
Hydrogen Fluoride
Iron
74908
7664393
7439896
Number of
1992 PCS
Parameters
1
1
0
1
1
1
1
1
1
1
0
0
o
J
0
0
1
1
0
0
1
0
0
1
o
J
0
0
o
J
Number of
1991 TRI
Parameters
0
1
1
0
2
0
0
0
2
2
1
1
0
1
1
0
1
1
1
0
1
1
0
0
1
1
0
Number of 1992
TRI Parameters
0
1
1
1
2
0
0
0
1
2
1
1
1
1
1
0
1
1
1
0
1
1
0
0
1
0
0
60
-------
Table 7.1 Refinery Wastewater Constituents
Constituent
Isophorone
Lead
Manganese
M-Cresol
Mercury
Methanol
Methyl Ethyl Ketone
Methyl Isobutyl Ketone
Methyl Tert-Butyl Ether
Molybdenum Trioxide
M-Xylene
Naphthalene
N-Butyl Alcohol
Nickel
Nitrite Plus Nitrate
Nitrogen
O-Cresol
CAS Number
78591
7439921
7439965
108394
7439976
67561
78933
108101
1634044
1313275
108383
91203
71363
7440020
14797558
17778880
95487
Oil And Grease
O-Xylene
95476
PAH Compounds
P-Cresol
Phenanthrene
Phenol
106445
85018
108952
Phenolic Compounds
Phosphate
Phosphoric Acid
Phosphorus
14265442
7664382
7723140
Number of
1992 PCS
Parameters
1
2
o
J
0
1
0
0
0
1
0
0
1
0
2
1
2
0
4
0
1
0
1
1
o
J
1
0
1
Number of
1991 TRI
Parameters
0
2
2
1
1
1
1
1
1
1
1
1
1
2
0
0
1
0
1
0
1
0
1
0
0
1
0
Number of 1992
TRI Parameters
0
2
2
0
0
1
1
1
1
1
1
1
1
2
0
0
0
0
1
0
0
0
1
0
0
0
0
61
-------
Table 7.1 Refinery Wastewater Constituents
Constituent
Polychlorinated Biphenyls
Polyram
Propylene
P-Xylene
CAS Number
1336363
9006422
115071
106423
Residual Oxidants
Selenium
Silver
Sodium Chloride (Salt)
Styrene
Sulfate
Sulfide
Sulfite
Sulfuric Acid
7782492
7440224
7647145
100425
14808798
18496258
14265433
7664939
Surfactants (MB AS)
Tetrachlorodibenzofuran, 2,3,7,8-
Tetrachloroethylene
Thallium
Toluene
51207319
127184
7440280
108883
Total Dissolved Solids (TDS)
Total Organic Carbon (TOC)
Total Oxygen Demand (TOD)
Total Suspended Solids (TSS)
Total Toxic Organics (TTO)
Trichloroethylene
Trichlorophenol
Vanadium
Xylene
79016
25167822
7440622
1330207
Number of
1992 PCS
Parameters
1
1
0
0
1
2
1
1
0
1
2
1
0
1
1
0
1
2
1
1
1
1
1
1
1
1
Number of
1991 TRI
Parameters
0
0
1
1
0
2
1
0
1
0
0
0
1
0
0
1
1
0
0
0
0
0
0
0
1
1
Number of 1992
TRI Parameters
0
0
1
1
0
1
1
0
1
0
0
0
1
0
0
0
1
0
0
0
0
0
0
0
0
1
62
-------
Table 7.1 Refinery Wastewater Constituents
Constituent
Zinc
CAS Number
7440666
Number of
1992 PCS
Parameters
2
Number of
1991 TRI
Parameters
2
Number of 1992
TRI Parameters
2
7.1.1 Permit Compliance System
EPA's Office of Wastewater Management (OWM) oversees the NPDES program on a
national level. EPA has authorized 39 States and the Virgin Islands to administer the NPDES
program. EPA regional offices administer the program in non-delegated States. More than
65,000 active NPDES permits have been issued to facilities throughout the nation. PCS has
extensive records on approximately 7,000 permits which are classified as "major". Facilities are
classified as "major" based on consideration of many factors, including effluent design flow,
physical and chemical characteristics of the wastestream, and location of discharge. Each permit
record in PCS may contain information that:
Identifies and describes the facility to which the permit has been granted (including a
primary Standard Industrial Classification (SIC) code);
Specifies the pollutant discharge limits for that facility;
Records the actual amounts of pollutants measured in the facility's wastewater
discharges; and
Tracks the facility's history of compliance with construction, pollutant limits, and
reporting requirements.
Major facilities must report compliance with NPDES permit limits, usually on a monthly
basis, via Discharge Monitoring Reports (DMRs). DMRs provide detailed information on
measured concentrations, including those that are in violation of established limits for the permit.
DMR data entered into PCS include the type of violation (if any), concentration and quantity
values, and monitoring period. The PCS data base is revised and updated twice weekly and,
therefore, data retrieved at a specific time are subject to subsequent alteration. In addition,
because of data entry delays, a complete set of data for a particular time period may not exist
until a year or more afterwards.
Among the permits listed in PCS are specific discharge limits or monitoring requirements for
it- Onn in/^nri/l-nal /-ปhomi/-ปo1 o
*' A
over 200 individual chemicals.
63
-------
7.1.2 Estimation of Annual Pollutant Loads from PCS
It is important to recognize that, unlike TRI, PCS is a permit tracking system, rather than a
repository of pollutant release amounts. However, an optional report in PCS called "Effluent
Data Statistics" (EDS) can process PCS data to produce annual loading values. EDS uses the
following hierarchy to derive a loading for each measured parameter: (1) reported loading value
in PCS (i.e., mass-based permit limit); and (2) loading estimate based on discharge flow and
concentration measurement. Depending on the monitoring requirements imposed by the permit,
flows and concentrations may be reported in many different ways. Measurements from PCS are
selected in the following order of preference: (1) average concentration; (2) maximum
concentration; and (3) minimum concentration. Estimated loadings are produced for records
with valid concentrations (as defined by PCS-EDS) and corresponding flow data assuming 30
operating days per month for each facility. Loadings are estimated using the following general
equation:
Load = Flow * Cone * Conversion Factors
Where:
Load = Specific pollutant load from a facility per unit time;
Flow = Facility effluent flow per unit time;
Cone = Concentration of a pollutant; and
Conversion Factors = Appropriate factors to convert reported units to standard units.
The Engineering and Analysis Division identified 138 permitted direct dischargers as
petroleum refineries potentially subject to effluent guidelines regulations. For these facilities,
EDS processed loading data for the calendar year 1992 were retrieved at the discharge pipe level
for each PCS parameter addressed in the permit with sufficient quantity or concentration/
discharge flow information. The loads for each parameter were summed across discharge pipes
to yield the total facility load. Concentration measurements recorded as below a detection limit
were treated in two ways: (1) for a low end estimated loading data set, values below detection
were set equal to zero; and (2) for a high end estimated loading data set, values below detection
were set equal to one-half the recorded detection limit. Parameters loadings based on
concentration measurements always below detection at a given discharge pipe were set equal to
zero for both data sets. The low end and high end data sets are presented in Table 7.2 for all
petroleum refining parameters. Multiple parameters sometimes exist for the same pollutant.
For example, zinc is represented by parameters for "total recoverable zinc" and "total zinc" as
(Zn).
64
-------
Table 7.2 1992 Annual Loading Data from PCS
Sorted by Parameter Name
Parameter
Number
34536
34571
34675
34621
415
410
1105
619
34220
1002
978
34030
30383
34526
34247
39100
310
311
80082
71870
1027
680
81017
940
50060
32106
1032
1034
1033
34320
1037
1042
Parameter Name
1 ,2-dichlorobenzene
1 ,4-dichlorobenzene
2,3,7,8-tetrachloro-dibenzo-ฃ>-dioxin
2,4,6-trichloro-phenol
Alkalinity, Phenol- Phthaline Method
Alkalinity, Total (as CaCO3)
Aluminum, Total (as Al)
Ammonia, unionized
Anthracene
Arsenic, Total (as As)
Arsenic, Total Recoverable
Benzene
Benzene, Ethylbenzenetoluene,
Xylene Combn
Benzo(A)Anthracene
Benzo(A)Pyrene
Bis (2-Ethylhexyl) Phthalate
BOD, 5 -Day (20 Deg. C)
BOD, 5 -Day Dissolved
BOD, Carbonaceous 5 Day, 20 Deg.C
Bromide (as Br)
Cadmium, Total (as Cd)
Carbon, Tot Organic (TOC)
Chemical Oxygen Demand (COD)
Chloride (as Cl)
Chlorine, Total Residual
Chloroform
Chromium, Hexavalent (as Cr)
Chromium, Total (as Cr)
Chromium, Trivalent (as Cr)
Chrysene
Cobalt, Total (as Co)
Copper, Total (as Cu)
Number
of
Facilities
1
1
1
1
2
1
5
1
1
9
1
6
1
1
1
1
100
1
5
1
7
54
8
12
15
3
84
93
o
6
i
5
17
Annual Load
Low-End
(Ibs/yr)
0.00
0.00
0.00
0.00
50,709.60
327,630.01
3,625.64
4.38
0.00
2,512.36
562.30
67.52
0.00
0.00
0.00
14.19
9,552,282.39
47,619.34
81,823.31
4,750.17
33.09
15,728,883.65
5,157,384.29
26,851,306.87
469,150.37
3.76
5,731.06
21,081.95
159.30
0.00
10.25
4,017.84
Annual Load
High-End
(Ibs/yr)
0.00
0.00
0.00
0.00
50,709.60
327,630.01
3,869.49
4.38
0.00
2,524.06
562.30
67.88
0.00
0.00
0.00
14.19
9,579,814.21
47,619.34
81,823.31
4,750.17
48.17
15,811,467.64
5,157,384.29
26,851,306.87
469,165.52
3.76
6,246.07
21,666.80
181.90
0.00
40.35
4,180.82
Selected for
Production-
Weighting
65
-------
Table 7.2 1992 Annual Loading Data from PCS
Sorted by Parameter Name
Parameter
Number
81208
720
722
34371
56
50050
951
900
39700
39942
46116
551
980
1046
1045
34408
1114
1051
1056
1055
11123
71900
22417
34696
1074
1067
630
610
71845
612
Parameter Name
Cyanide, Free (not amenable to
chlorination)
Cyanide, Total (as Cn)
Cyanide, Free (amen, to chlorin.)
Ethylbenzene
Flow Rate
Flow, in conduit or thru treatment
plant
Fluoride, Total (as F)
Hardness, Total (as CaCOs)
Hexachlorobenzene
Hydrocarbons, Aromatic
Hydrocarbons, Total Gas Chromat.
Hydrocarbons, inH2O, IR, CC14 Ext.
Chromat.
Iron, Total Recoverable
Iron, Dissolved (as Fe)
Iron, Total (as Fe)
Isophorone
Lead, Total Recoverable
Lead, Total (as Pb)
Manganese, Dissolved (as Mn)
Manganese, Total (as Mn)
Manganese, Total Recoverable
Mercury, Total (as Hg)
Methyl Tert-Butyl Ether
Naphthalene
Nickel, Total Recoverable
Nickel, Total (as Ni)
Nitrite Plus Nitrate Total
Nitrogen, Ammonia, Total (as N)
Nitrogen, Ammonia, Total (as NH4)
Nitrogen, Ammonia, Tot. Unionized
(asN)
Number
of
Facilities
1
9
4
2
3
108
5
1
1
2
1
o
6
i
i
i
i
i
12
1
1
1
10
2
1
1
7
1
102
1
2
Annual Load
Low-End
(Ibs/yr)
78.13
2,353.27
521.70
224.96
31,776.87
1,583,221.78
208,305.59
1,632,518.41
0.05
10.55
12,293.02
13,509.68
2,377.94
34.78
15.89
0.00
67.80
1,266.47
4,998.97
483.75
5,097.37
827.28
1,334.15
0.00
143.87
2,021.08
12,701.94
3,015,792.35
19,448.10
4,136.28
Annual Load
High-End
(Ibs/yr)
109.16
2,384.00
521.70
1,237.05
31,776.87
1,583,221.78
208,305.59
1,632,518.41
0.05
10.55
12,293.02
13,536.22
2,377.94
49.05
15.89
0.00
140.47
2,688.44
4,998.97
483.75
5,097.37
832.39
1,334.15
0.00
178.51
2,140.30
12,701.94
3,031,910.08
19,448.10
4,136.28
Selected for
Production-
Weighting
66
-------
Table 7.2 1992 Annual Loading Data from PCS
Sorted by Parameter Name
Parameter
Number
600
560
3582
550
556
78141
34044
82210
341
340
335
343
300
34461
34694
78218
32730
46000
650
665
38528
39516
22456
1147
981
1077
32017
70295
70300
530
Parameter Name
Nitrogen, Total (asN)
Oil & Grease (Freon Extr.-IR Meth)
Tot,Rc
Oil And Grease
Oil And Grease (Soxhlet Extr.) Tot.
Oil And Grease Freon Extr-Grav
Meth
Organics, Total Toxic (TTO)
Oxidants, Total Residual
Oxygen Demand First Stage
Oxygen Demand, Chem. (COD),
Dissolved
Oxygen Demand, Chem. (High Level)
(COD)
Oxygen Demand, Chem. (Low Level)
(COD)
Oxygen Demand, Total (TOD)
Oxygen, Dissolved (DO)
Phenanthrene
Phenol, Single Compound
Phenolic Compounds, Unchlorinated
Phenolics, Total Recoverable
Phenols
Phosphate, Total (as PO4)
Phosphorus, Total (as P)
Poly-Nuclear Aromatics (Polyram)
Fob/chlorinated Biphenyls (PCBs)
Polynuc Aromatic HC per Method
610
Selenium, Total (as Se)
Selenium, Total Recoverable
Silver, Total (as Ag)
Sodium Chloride (Salt)
Solids, Total Dissolved
Solids, Total Dissolved- 180 Deg.C
Solids, Total Suspended
Number
of
Facilities
1
1
1
21
89
1
2
2
1
77
7
1
14
1
19
4
87
2
1
5
1
1
1
9
2
9
1
4
3
107
Annual Load
Low-End
(Ibs/yr)
118,681.20
121,980.00
3,001.56
475,731.79
4,805,765.77
1,919.19
9,524.47
601,690.79
133,851.00
48,012,895.91
2,734,385.34
1,206,525.79
9,342,790.99
0.00
1,506.46
226.73
19,157.48
78.57
16,234.99
40,216.87
0.00
1.33
21.59
5,442.65
5.38
43.32
356,129.31
174,189,544.43
14,639,211.69
29,597,956.57
Annual Load
High-End
(Ibs/yr)
118,681.20
121,980.00
3,001.56
478,588.98
4,924,117.65
1,925.73
9,524.47
601,690.79
133,851.00
48,012,952.25
2,734,385.34
1,206,525.79
9,342,790.99
0.00
1,521.75
235.25
19,646.59
78.57
16,234.99
40,216.87
0.00
1.33
86.13
5,483.49
5.38
81.40
356,129.31
174,189,544.43
14,639,211.69
29,613,328.78
Selected for
Production-
Weighting
67
-------
Table 7.2 1992 Annual Loading Data from PCS
Sorted by Parameter Name
Parameter
Number
81395
945
81621
745
741
38260
34475
17
34010
39180
1087
81551
1094
1092
Parameter Name
Storm Water Flow
Sulfate, Total (as SO4)
Sulfide, Total
Sulfide, Total (as S)
Sulfite (as S)
Surfactants (MB AS)
Tetrachloroethylene
Thermal Discharge, million BTUs per
day
Toluene
Trichloroethylene
Vanadium, Total (as V)
Xylene
Zinc, Total Recoverable
Zinc, Total (as Zn)
Number
of
Facilities
3
6
5
93
1
2
1
1
o
3
i
5
1
o
3
19
Annual Load
Low-End
(Ibs/yr)
835.87
1,951,597.35
6,329.89
27,861.12
221.70
828.92
0.00
1,344,075.39
1,093.01
0.00
27,379.00
0.89
1,416.01
13,841.55
Annual Load
High-End
(Ibs/yr)
835.87
1,951,597.35
6,415.89
30,592.20
226.35
828.92
0.00
1,344,075.39
2,022.41
0.00
27,680.95
0.89
1,416.01
14,078.59
Selected for
Production-
Weighting
To support EPA's industry selection process for future effluent guidelines development,
further refinement of the EDS loading data was conducted as follows:
Exclude conventional and classical parameter loads (e.g., TSS, BOD, Oil and Grease,
COD) that represent groups of individual chemicals;
Exclude relatively non-toxic anion and cation parameter loads (e.g., phosphorus,
phosphate, chloride, sulfate, sulfite nitrogen, nitrite, sodium chloride, and sodium);
Exclude nonconventional parameter loads that represent groups of individual chemicals
(e.g., total recoverable phenolics);
Include the parameter with the maximum loading reported if multiple parameters are
reported for the same chemical at the same discharge pipe;
Sum chemical parameter loads across all discharge pipes to calculate a facility pollutant
load; and
Include only reported loadings, representing the high-end estimated data, and not
extrapolated national projections.
A listing of pollutant loadings used in the industry selection is presented in Appendix H.
68
-------
Many of the parameters listed on Table 7.2 are measured at only a portion of the petroleum
refineries. Therefore, the loading values represent a sample of all petroleum refining wastewater
discharges. Based on the assumption that the constituents listed on Table 7.2 are present in the
effluent of all refineries, the loading data for several parameters were extrapolated to a national
level for two sets of facilities: (1) facilities in California, and (2) facilities not in California.
Total loads for California facilities were estimated separately because these facilities employ
significantly different practices with respect to water conservation and treatment systems (i.e.,
activated carbon). To assist in verifying the presence of wastewater constituents in petroleum
refining effluent, EPA undertook a limited review of Form 2C NPDES Applications, which
require a chemical analysis of current or proposed discharges. This review, summarized in
Appendix F, indicates the presence of 37 individual wastewater constituents, 20 of which are
among the constituents selected for production weighting. Five of the chemicals reported as
above detection on at least one Form 2C are represented by less than five facilities in the 1992
PCS loadings data set. The extrapolation procedure is based on the ratio of petroleum
production level at the facilities measuring a given parameter to the overall industry production
level. Production levels (in barrels per day) were obtained from data presented for 138 direct
dischargers (13 in California) in the Oil and Gas Journal (Thrash, 1991). The set of 138
facilities, listed on Table 7.3 with production data, is assumed to represent all direct dischargers.
The total production level for direct dischargers in California is 1,619,950 barrels per day; the
total production level for direct dischargers not in California is 12,376,556 barrels per day. In
general, parameters representing individual chemicals, and parameters representing conventional
pollutants or pollutant groups that have a large sample size, are selected for inclusion in the
production weighted data set. These parameters are identified in Table 7.2. The general
equation for production weighted extrapolation is given as:
f TOTPROD
TOTLOAD = SMPLOAD * SMPPROD
Where: TOTLOAD = Total extrapolated load
SMPLOAD = Sample load based on facilities reporting the parameter in
PCS
TOTPROD = Total production level
SMPPROD = Sample production level
69
-------
Table 7.3 Crude Production for 138 Direct Discharging
Petroleum Refineries
Sorted by State and Company Name
NPDES
Permit No.
AK0000841
AL0000574
AL0000973
AL0055859
AR0000663
AR0000591
AR0000647
CA0000680
CA0005134
CA0000337
CA0005550
CA0005096
CA0057177
CA0005789
CA0003778
CA0055387
CA0004961
CA0005053
CA0000809
CA0000051
CO0001210
CO0001147
CO0000078
DE0000256
GA0001902
HI0000329
IL0001244
IL0004219
IL0004073
IL0002861
IL0000205
IL0001589
IN0000108
IN0002470
IN0001244
Company
Tesoro Alaska Petroleum Co
Gamxx Energy, Inc., Theodore
Hunt Refining Company
LL&E Petroleum Marketing Inc
Berry Petroleum Corp-Stephens
Cross Oil-Smackover
Lion Oil Company
ARCO
Chevron U.S. A. Products Co.
Chevron U.S. A., Inc.
Exxon Co., USA
Pacific Refining Co.
Powerine Oil Co.
Shell Oil Co.
Texaco Refining/Marketing Inc.
Torrance Refinery
Tosco Refining Co.
Union Oil Co. Of Ca.
Unocal
Unocal Corporation
Colorado Refining Company
Conoco, Inc.
Landmark Petroleum, Inc.
Star Enterprises
Young Refining Corp
Chevron U.S. A., Inc.
Clark Oil- Wood River
Indian Refining-Lawrenceville
Marathon Oil-Robinson
Mobil Oil-Joliet Ref
Shell Oil Co.-Wood River
Uno-Ven Company-Lemont
American Oil Company (Amoco)
Countrymark Cooperative, Inc.
Laketon Refining Corporation
City
Kenai
Mobile County
Tuscaloosa
Saraland
Stephens
Smackover
El Dorado
Carson
Richmond
El Segundo
Benicia
Hercules
Santa Fe Springs
Martinez
Wilmington
Torrance
Martinez
Rodeo
Carson
Arroyo Grande
Commerce City
Commerce City
Fruita
Delaware City
Douglasville
Honolulu
Hartford
Lawrenceville
Robinson
Joliet
Roxana
Lemont
Whiting
Mount Vernon
Laketon
State
AK
AL
AL
AL
AR
AR
AR
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
DE
GA
HI
IL
IL
IL
IL
IL
IL
IN
IN
IN
Crude
Production
(bbl/day)
72,000
26,500
33,500
80,000
5,700
6,770
48,000
223,000
205,000
254,000
128,000
52,250
46,500
140,100
95,000
123,000
132,000
56,550
108,000
56,550
28,000
48,000
15,200
140,000
7,500
52,800
57,000
54,000
170,000
180,000
274,000
147,000
350,000
20,600
8,300
70
-------
Table 7.3 Crude Production for 138 Direct Discharging
Petroleum Refineries
Sorted by State and Company Name
NPDES
Permit No.
IN0002364
KS0000205
KS0000248
KS0050997
KS0000761
KS0000434
KY0000388
KY0094579
LA0032417
LA0003115
LA0052370
LA0046612
LA0006963
LA0005941
LA0003026
LA0005584
LA0005312
LA0045683
LA0004260
LA0003646
LA0051942
LA0054216
LA0039390
LA0003522
LA0006041
MI0003778
MI0001066
MN000025
6
MN000041
8
MS0002984
MS0001481
MS0034711
MS0001686
Company
Marathon Petroleum Co., Ird
Coastal Ref. & Marketing
Farmland-Coffeyville Refinery
Farmland-PhiHipsburg Refinery
Texaco Refining & Marketing Inc
Total Petroleum, Inc
Ashland Petroleum Co
Somerset Refinery Inc
Atlas Processing Co-Shreveport
BP Oil Company
Calcasieu Refining Co.
Calumet Refining Co.
Canal Refining-Church Point
Citgo Petroleum Corp.
Conoco Inc -Lake Charles Refine
Exxon Co USA-Baton Rouge
Kerr-Mcgee Corp-Cotton Valley
Marathon Oil Co
Mobil Oil Corp-Chalmette
Murphy Oil USA Inc
Phibro Energy USA, Inc-Krotz
Phibro Energy USA, Inc-St.Rose
Placid Refining Co -Port Allen
Shell Oil Co-Norco
Star Enterprise
Lakeside Refining Co
Total Petroleum Inc
Ashland Oil Inc
Koch Refining Co-Rosemount
Amerada Hess Corp Purvis
Chevron US A
Ergon Refining Incorporated
Southland Oil Company
City
Indianapolis
Butler County
Coffeyville
Phillipsburg
El Dorado
Arkansas City
Ashland
Somerset
Shreveport
Belle Chasse
Lake Charles
Princeton
Church Point
Lake Charles
Westlake
Baton Rouge
Cotton Valley
Garyville
Chalmette
Meraux
Krotz Springs
St Rose
Port Allen
Norco
Convent
Kalamazoo
Alma
Saint Paul Park
Rosemount
Purvis
Pascagoula
Vicksburg
Sandersville
State
IN
KS
KS
KS
KS
KS
KY
KY
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
MI
MI
MN
MN
MS
MS
MS
MS
Crude
Production
(bbl/day)
50,000
30,400
59,600
26,400
80,000
56,000
213,400
5,500
46,200
218,500
13,500
4,376
9,865
320,000
159,500
421,000
7,800
225,000
160,000
97,000
56,700
28,300
48,000
215,000
225,000
5,600
45,600
67,100
218,500
30,000
295,000
16,800
11,000
71
-------
Table 7.3 Crude Production for 138 Direct Discharging
Petroleum Refineries
Sorted by State and Company Name
NPDES
Permit No.
MS0001678
MT0000264
MT0000256
MT0029742
MT0000477
MT0000434
ND0000248
NJ0001511
NJ0000221
NJ0005401
NJ0005029
NJ0028878
NY0028592
OH0005657
OH0002461
OH0002623
OH0002763
OK0000256
OK0000825
OK0001309
OK0000876
OK0001295
PAOO 12637
PA0011533
PA0002551
PAOO 11096
PA0005304
PA0002674
PR0000370
PR0000400
TN0059226
TX0002984
TX0004847
Company
Southland Oil Lumberton
Cenex-Laurel Refinery
Conoco Inc
Conoco, Inc
Exxon Co USA (Billings Refin.)
Montana Refining Co-Blackeagle
Amoco Oil Company
Bayway Refining Company
Chevron US A Inc
Coastal Eagle Point Oil Co
Paulsboro Refinery
Port Reading Refining Fac
Cibro Petroleum Products, Inc
Ashland Oil, Inc.
BP Oil Company
BP Oil Company
Sun Refining & Marketing Co
Conoco Inc.-Ponca City Refiner
Kerr-Mcgee Corp-Garvin
Sinclair Oil Corporation
Sun Refining & Marketing
Compa
Total Petroleum
BP Oil Inc.
Chevron U.S. A. Products, Co.
Pennzoil United Inc Rouseville
Sun Refining & Marketing, Inc
United Refining Co-Warren
Witco Chem Corp
Caribbean Gulf Refining Corp
Puerto Rico Sun Oil Co.
Mapco Petroleumjnc
Amoco Texas Refining Company
Chevron US A Inc
City
Lumberton
Laurel
Billings
Billings
Billings
Black Eagle
Mandan
Linden
Perth Amboy
West Deptford
Paulsboro
Port Reading
Albany
Canton
Toledo
Lima
Toledo
Ponca City
Wynnewood
Tulsa
Tulsa
Ardmore
Marcus Hook
Philadelphia
Rouseville,
Philadelphia
Warren
Bradford
Bayamon
Yabucoa
Shelby County
(Mbo)
Texas City
State
MS
MT
MT
MT
MT
MT
ND
NJ
NJ
NJ
NJ
NJ
NY
OH
OH
OH
OH
OK
OK
OK
OK
OK
PA
PA
PA
PA
PA
PA
PR
PR
TN
TX
TX
Crude
Production
(bbl/day)
5,800
40,400
24,750
24,750
42,000
7,000
58,000
13,000
80,000
109,520
100,000
50,000
39,900
66,000
120,650
142,500
125,000
140,000
43,000
50,000
85,000
68,000
171,000
175,000
15,700
130,000
64,600
10,000
38,000
85,000
75,000
433,000
66,000
72
-------
Table 7.3 Crude Production for 138 Direct Discharging
Petroleum Refineries
Sorted by State and Company Name
NPDES
Permit No.
TX0005991
TX0006211
TX0066591
TX0006904
TX0004626
TX0088331
TX0006271
TX0104515
TX0004201
TX0084778
TX0006289
TX0001449
TX0003247
TX0003697
TX0004227
TX0002976
TX0006009
TX0009148
TX0007536
TX0004871
TX0006599
TX0005835
TX0063355
UT0000175
UT0000507
VA0003018
VI0000019
WA002290
0
WA000298
4
WA000076
1
Company
Chevron USA Inc Port Arthur
Citgo Refining & Chemicals Inc
Coastal Refinig & Marketing, I
Coastal Refining & Marketing,
Crown Central Petr-Houston
Diamond Shamrock Refining &
Ma
Exxon Corp-Houston
Fina Oil & Chem-Big Spr
Fina Oil & Chem-Port A
Howell Hydrocarbons & Chem.
In
Koch Refining Co
La Gloria Oil & Gas Co-Tyler
Lyondell Petrochemical Co.
Marathon Oil Company
Mobil Chem-Beaumont
Phibro Energy USA, Inc -Houston
Phibro Energy USA, Inc-Tx City
Phillips 66 Co-Hutchins
Phillips 66 Co-Sweeny Refinery
Shell Oil Co-Deer Park
Southwestern Refining-Corpus C
Star Enterprise-Port Arthur
Valero Refining Co. -Corpus Chr
Chevron U.S. A., Inc
Phillips 66 Company
American Oil Yorktown
Hess Oil Virgin Islands Corp
ARCO Petroleum Products Co
BP Oil Company
Shell Oil Co
City
Port Arthur
Corpus Christi
Corpus Christi
Corpus Christi
Pasadena
Three Rivers
Baytown
Big Spring
Jefferson County
Houston
Corpus Christi
Tyler
Houston
Texas City
Jefferson County
Houston
Texas City
Borger
Sweeny
Harris County
Corpus Christi
Port Arthur
Corpus Christi
Salt Lake City
West Bountiful
Yorktown
St. Croix
Ferndale
Ferndale
State
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
VA
VI
WA
WA
WA
Crude
Production
(bbl/day)
315,300
132,500
45,125
45,125
100,000
53,000
396,000
55,000
110,000
1,900
125,000
49,500
265,000
70,000
275,000
67,000
123,500
105,000
175,000
215,900
104,000
250,000
27,000
45,000
25,000
53,000
545,000
167,000
90,250
89,300
73
-------
Table 7.3 Crude Production for 138 Direct Discharging
Petroleum Refineries
Sorted by State and Company Name
NPDES
Permit No.
WA000320
4
WA000294
1
WA000178
o
5
WI0003085
WV000462
6
WY000044
2
WY000116
3
Company
Sound Refining Co
Texaco Inc
US Oil & Refining Company
Murphy Oil USA Inc Superior Re
Quaker State Oil Refining Corp
Frontier Oil & Refining Co
Wyoming Refining Co
City
Tacoma
Anacortes
Tacoma
Superior
Newell
Cheyenne
Denver
State
WA
WA
WA
WI
WV
WY
WY
Crude
Production
(bbl/day)
11,900
132,000
32,775
32,000
10,500
35,000
11,900
The low-end (zero for non-detects) production-weighted loadings are given by parameter on
Table 7.4 and for the high-end estimate (half detection limit) is presented on Table 7.5. These
summaries identify pollutants as classical or conventional, organics or metals. The designation
as classical is given to parameters which have been measured historically as part of refinery
discharge permits and are neither an organic nor a metal. Parameters that represent conventional
pollutant parameters such as BOD are listed at the bottom of both Table 7.4 and 7.5. National
totals of metals and organics for 1992 are between 2.1 and 2.6 million Ibs/yr. (Double counting
may occur in the reported national total loadings reported for organics. Hydrocarbons and total
recoverable phenolics are listed with other individual organics and are included in the total).
Over 577,000 pounds of hydrocarbons and approximately 28,000 pounds of total recoverable
phenolics were estimated to be released in 1992. Annual releases of priority pollutants were
projected to be between 1.0 and 1.6 million pounds. In general, the average releases per facility
are greater for non-California facilities. Notable exceptions are nickel, selenium, and phenolics.
Note the significance of values below detection in the estimated ethylbenzene load for
non-California facilities.
Assumptions and Limitations
Several assumptions and limitations of the above described analyses include:
Only refineries considered as "major facilities" that directly discharge to surface waters
and have a NPDES permit are included in PCS. Consequently, PCS may be incomplete
in terms of petroleum refining facilities, pollutants, or wastestreams.
74
-------
Facilities are not required by their NPDES permit to report on all chemicals actually
discharged. A facility is only required to report on a particular chemical if it is specified
in the permit conditions.
Although EDS converts all values to standard units, there are an undetermined number of
unit code/measurement value mismatches in PCS, as well as invalid discharge flow
records and analytical test results, that cannot be readily identified after EDS processing.
National production-weighted load estimates assume that all petroleum refineries in a
particular grouping discharge similar waste streams. The fewer the data points in a
parameter sample set, the greater the uncertainty that the parameter loading rate is
representative of refinery discharge in general.
Table 7.4 Comparison of California and Non-California PCS Loads
after Production Weighting (Low-End Estimate)
Parameter Name
Bromide (as Br)
Chloride (as Cl)
Chlorine, Total Residual
Cyanide, Total (as CN)
Fluoride, Total (as F)
Nitrogen, Ammonia, Total
(asN)
Phosphate, Total (as PO4)
Sulfide, Total (as S)
Poll.
Type
*
C
C
C
C
C
C
C
C
Subtotal
Aluminum, Total (as Al)
Arsenic, Total (as As)
Cadmium, Total (as Cd)
Chromium, Hexavalent (as
Cr)
Chromium, Total (as Cr)
Cobalt, Total (as Co)
Copper, Total (as Cu)
Iron, Total (as Fe)
Lead, Total (as Pb)
Manganese, Total (as Mn)
Mercury, Total (as Hg)
Nickel, Total (as Ni)
Selenium, Total (as Se)
M
M
M
M
M
M
M
M
M
M
M
M
M
Non-Calif. Loads
Total
(lb/yr)
1,187,691
146,706,014
3,717,957
10,050
6,911,812
3,552,591
1,148,190
35,723
163,270,028
11,368
78,851
780
7,433
24,450
27,787
2,459
13,840
70,437
14,600
11,298
7,758
Average
(Ib/fac/yr)
9,502
1,173,648
29,744
80
55,294
28,421
9,186
286
1,306,160
91
631
6
59
196
222
20
111
563
117
90
62
California Loads
Total
(lb/yr)
107,420,130
39,083
5,488
178,232
1,738
107,644,672
14,736
764
19
89
3,088
33
625
117
9
3,975
11,691
Average
(Ib/fac/yr)
8,263,087
3,006
422
13,710
134
8,280,359
1,134
59
1
7
238
o
3
48
9
1
306
899
Total
Load
(lb/yr)
1,187,691
254,126,144
3,757,040
15,538
6,911,812
3,730,823
1,148,190
37,461
270,914,700
26,104
79,615
799
7,522
27,537
33
28,412
2,459
13,957
70,437
14,609
15,274
19,449
Priority
Pollutants
(lb/yr)
15,538
15,538
79,615
799
7,522
27,537
28,412
13,957
14,609
15,274
19,449
75
-------
Table 7.4 Comparison of California and Non-California PCS Loads
after Production Weighting (Low-End Estimate)
Parameter Name
Silver, Total (as Ag)
Vanadium, Total (as V)
Zinc, Total (as Zn)
Poll.
Type
*
M
M
M
Subtotal
Benzene
Bis (2-ethylhexyl)
Phthalate
Chloroform
Ethylbenzene
Hexachlorobenzene
Hydrocarbons, in H2O, IR,
CC14 Ext. Chromat
Methyl Tert-butyl Ether
Phenolics, Total Recov.
Phenol, Single Compound
Toluene
Xylene
0
O
0
O
O
0
O
0
0
O
O
Subtotal
BOD, 5 -day (20 Deg. C)
Carbon, Total Organic
(TOC)
Oil and Grease Freon
Extr-grav Meth
Oxygen Demand, Chem.
(High Level) (COD)
Solids, Total Suspended
C
C
C
C
C
Non-Calif. Loads
Total
(Ib/yr)
520
76,744
348,326
1,507
1,405
250
214,124
12
577,519
217,266
23,769
10,070
46,957
1,517,949
11,399,374
27,928,894
6,709,057
86,050,803
33,744,172
Average
(Ib/fac/yr)
4
614
2,787
12
11
2
1,713
0
4,620
1,738
190
81
376
12,144
91,195
223,431
53,672
688,406
269,953
California Loads
Total
(Ib/yr)
24
87,154
9,308
131,631
1
3,411
4
12
144,366
970,502
2,591,416
207,957
5,213,221
1,385,076
Average
(Ib/fac/yr)
2
6,704
716
10,125
0.1
262
0.3
1
11,105
74,654
199,340
15,997
401,017
106,544
Total
Load
(Ib/yr)
544
87,154
86,053
479,957
1,507
1,405
250
214,125
12
577,519
217,266
27,180
10,070
46,962
12
1,662,316
12,369,876
30,520,310
6,917,013
91,264,025
35,129,248
Priority
Pollutants
(Ib/yr)
544
86,053
293,771
1,507
1,405
250
214,125
12
10,070
46,962
654,153
* Pollutant Type: (C) Classical/Conventional; (M) Metal; (O) Organic
76
-------
Table 7.5 Comparison of California and Non-California PCS Loads
after Production Weighting (High-End Estimate)
Parameter Name
Bromide (as Br)
Chloride (as Cl)
Chlorine, Total Residual
Cyanide, Total (as CN)
Fluoride, Total (as F)
Nitrogen, Ammonia, Total
(asN)
Phosphate, Total (as PO4)
Sulfide, Total (as S)
Poll.
Type
*
C
C
C
C
C
C
C
C
Subtotal
Aluminum, Total (as Al)
Arsenic, Total (as As)
Cadmium, Total (as Cd)
Chromium, Hexavalent (as
Cr)
Chromium, Total (as Cr)
Cobalt, Total (as Co)
Copper, Total (as Cu)
Iron, Total (as Fe)
Lead, Total (as Pb)
Manganese, Total (as Mn)
Mercury, Total (as Hg)
Nickel, Total (as Ni)
Selenium, Total (as Se)
Silver, Total (as Ag)
Vanadium, Total (as V)
Zinc, Total (as Zn)
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
Subtotal
Benzene
Bis (2-ethylhexyl)
Phthalate
Chloroform
Ethylbenzene
O
O
O
O
Non-Calif.Loads
Total
(Ib/yr)
1,187,691
146,706,014
3,717,957
10,050
6,911,812
3,561,088
1,148,190
38,317
163,281,118
11,368
78,851
788
7,896
25,107
27,910
2,459
28,882
70,437
14,679
11,298
7,758
685
76,755
364,874
1,515
1,405
250
1,177,682
Average
(Ib/fac/yr)
9,502
1,173,648
29,744
80
55,294
28,489
9,186
307
1,306,249
91
631
6
63
201
223
20
231
563
117
90
62
5
614
2,919
12
11
2
9,421
California Loads
Total
(Ib/yr)
107,420,130
39,220
5,586
197,629
3,930
107,666,496
15,773
790
55
556
3,209
128
957
360
11
4,246
11,784
87
88,115
9,810
135,882
1
Average
(Ib/fac/yr)
8,263,087
3,017
430
15,202
302
8,282,038
1,213
61
4
43
247
10
74
28
1
327
906
7
6,778
755
10,452
0
Total
Load
(Ib/yr)
1,187,691
254,126,144
3,757,177
15,636
6,911,812
3,758,717
1,148,190
42,247
270,947,615
27,141
79,642
843
8,452
28,316
128
28,867
2,459
29,243
70,437
14,690
15,544
19,542
772
88,115
86,565
500,755
1,515
1,405
250
1,177,682
Priority
Pollutants
(Ib/yr)
15,636
15,636
79,642
843
8,452
28,316
28,867
29,243
14,690
15,544
19,542
772
86,565
312,475
1,515
1,405
250
1,177,682
77
-------
Table 7.5 Comparison of California and Non-California PCS Loads
after Production Weighting (High-End Estimate)
Hexachlorobenzene
Hydrocarbons, in H2O, IR,
CC14 Ext. Chromat
Methyl Tert-butyl Ether
Phenolics, Total Recov.
Phenol, Single Compound
Toluene
Xylene
0
O
0
O
O
0
0
Subtotal
BOD, 5-day (20 Deg. C)
Carbon, Tot Organic
(TOC)
Oil and Grease Freon
Extr-grav Meth
Oxygen Demand, Chem.
(High Level) (COD)
Solids, Total Suspended
C
C
C
C
C
Non-Calif.Loads
12
578,654
217,266
24,308
10,172
86,898
2,098,161
11,402,040
28,081,112
6,863,165
86,050,909
33,746,916
0
4,629
1,738
194
81
695
16,785
91,216
224,649
54,905
688,407
269,975
California Loads
3,577
4
12
3,593
1,017,120
2,591,416
223,355
5,213,221
1,402,484
275
0
1
276
78,240
199,340
17,181
401,017
107,883
Total
12
578,654
217,266
27,885
10,172
86,902
12
2,101,755
12,419,160
30,672,527
7,086,520
91,264,131
35,149,400
Priority
12
10,172
86,902
1,277,938
* Pollutant Type: (C) Classical/Conventional; (M) Metal; (O) Organic
7.1.3 Analysis of Average Measured Pollutant Concentrations from PCS
In addition to EDS-generated loadings, measured average concentration values from monthly
monitoring data were retrieved separately from PCS for 1992. Theres data set may include data
not participating in the loadings analysis because of lack of corresponding discharge flow data.
Conversely, parameters with large predicted loads may not have corresponding high
concentrations because the load estimate may be a direct measurement (mass based limit) or may
be based on a minimum or maximum concentration. Also, because the concentration data set is
based on single measured values (rather than a formula as with the loadings data set) limited
QA/QC procedures can be employed. For the concentration data, the range of measured values
at the pollutant/discharge pipe level and at the overall pollutant level (i.e., for all discharge pipes)
were examined. If either range exceeded three orders of magnitude, the concentration data at the
discharge pipe level were checked for potential unit code/measurement value mismatches. In
eight cases, obvious unit code errors were corrected; in an additional 13 cases, there were not
enough data points to clearly identify a unit code error, yet the values were so extreme that they
were considered highly questionable and were excluded from subsequent analysis. As with the
loadings analysis, two sets of data were prepared: (1) a low-end estimate based on assigning zero
to measurements below detection, and (2) a high-end estimate based on assuming half the
78
-------
detection limit. Measurements always below detection for a particular parameter at a particular
discharge pipe were set equal to zero for both data sets.
Comparisons of average concentrations at California and non-California facilities at the
low-end and high-end for the 20 parameters with data available for both groups are presented on
Tables 7.6 and 7.7. Most parameters are measured at higher concentrations outside of
California. Some parameters, such as cadmium, mercury, and arsenic are significantly higher on
the average at non-California facilities. A few parameters, including selenium and cyanide, are
significantly higher at California facilities. Additional tables summarizing concentration data by
parameter for California, non-California, and all facilities combined are presented in Appendix
B. These tables include all of the 74 parameters with measured average concentration data
available, and also present the range and standard deviation. For comparison, Table 7.8 presents
the average concentrations for 27 pollutants reported in the petroleum refining development
document (US EPA, 1982a). These data reflect sampling done at current/BPT treatment levels
prior to implementation of the 1982 BAT regulations (US EPA, 1982b). As with the PCS data,
these data are limited by the frequency of detection. In general, the high-end PCS concentration
averages for all facilities are on-line with or slightly lower than the levels reported in the 1982
development document. Exceptions to this are arsenic at both California and non-California
facilities, mercury and cadmium at non-California facilities, and selenium at California facilities.
In the case of arsenic and mercury, the 1982 document reports high variability in the data that
may reflect low confidence. In addition, the PCS parameters are measuring total mercury, total
arsenic, and total cadmium. Much of this may include metal bound in compounds and
complexes that may not be included in the 1982 measures. It is interesting to note that the
high-end "total recoverable selenium" average concentration for non-California facilities (see
Appendix B) matches the 1982 average almost exactly, whereas the high-end "total selenium"
average concentration for California facilities is much higher.
79
-------
Table 7.6 Comparison of California and Non-California PCS Concentration
(Low-End Estimate)
(Parameters Available for Both Groups, all concentrations in ug/L)
Data
Parameter
Arsenic, Total (as
As)
BOD, 5-day (20
Deg. C)
Cadmium, Total (as
Cd)
Carbon, Tot. Organic
(TOC)
Chromium,
Hexavalent (as Cr)
Chromium, Total (as
Cr)
Copper, Total (as
Cu)
Cyanide, Total (as
Cn)
Lead, Total (as Pb)
Mercury, Total (as
Hg)
Nickel, Total (as Ni)
Nitrogen, Ammonia
Total (as N)
Oil and Grease Freon
Extr-grav Meth
Oxygen Demand,
Chem. (High Level)
(COD)
Phenolics, Total
Recoverable
Selenium, Total (as
Se)
Silver, Total (as Ag)
Solids, Total
Suspended
Sulfide, Total (as S)
Zinc, Total (as Zn)
California
Number
of
Facilities
3
6
3
4
5
6
4
3
3
3
4
5
7
5
6
4
3
7
5
5
Number
of
Observ.
25
58
25
65
84
63
36
31
32
25
39
43
76
45
89
40
25
69
45
50
Number of
Non-
Detects
11
20
24
1
72
42
9
12
29
23
9
o
6
24
0
31
1
24
12
44
8
Measured
Average
3.60
6,281.03
0.02
9,177.54
0.41
2.06
9.53
54.77
3.44
0.03
11.38
3,110.70
2,626.84
72,844.44
18.62
155.50
0.02
13,872.46
1.11
165.50
Non-California
Number of
Facilities
2
25
2
15
21
22
4
2
1
2
2
26
31
17
25
1
1
33
20
4
Number
of
Observ.
24
241
8
366
189
192
44
24
8
16
13
234
464
165
231
3
3
457
175
50
Number of
Non-
Detects
0
2
1
17
65
59
6
7
0
0
1
6
43
0
21
1
1
17
23
1
Measured
Average
408.42
14,264.70
34.79
10,219.05
2.92
13.53
11.39
23.42
8.38
8.75
29.69
5,140.18
3,451.35
126,609.21
45.76
3.90
2.10
23,981.16
55.48
136.36
80
-------
Table 7.7 Comparison of California and Non-California PCS Concentration Data
(High-End Estimate)
(Sorted by Parameter Name for Parameters Available for Both Groups, All concentrations in ug/L)
Parameter
Arsenic, Total
(as As)
BOD, 5-day
(20 Deg. C)
Cadmium, Total
(as Cd)
Carbon, Tot.
Organic (TOC)
Chromium,
Hexavalent (as Cr)
Chromium, Total
(as Cr)
Copper, Total
(as Cu)
Cyanide, Total
(as Cn)
Lead, Total (as Pb)
Mercury, Total
(as Hg)
Nickel, Total (as Ni)
Nitrogen, Ammonia
Total (as N)
Oil and Grease
Freon Extr-grav
Meth
Oxygen Demand,
Chem. (High Level)
(COD)
Phenolics, Total
Recoverable
Selenium, Total
(as Se)
Silver, Total (as Ag)
Solids, Total
Suspended
Sulfide, Total (as S)
Zinc, Total (as Zn)
California
Number
of
Facilities
3
6
o
J
4
5
6
4
o
J
3
o
3
4
5
7
5
6
4
3
7
5
5
Number
of
Observ.
25
58
25
65
84
63
36
31
32
25
39
43
76
45
89
40
25
69
45
50
Number
of
Non-detects
11
20
24
1
72
42
9
12
29
23
9
3
24
0
31
1
24
12
44
8
Measured
Average
4.46
7,462.07
0.02
9,189.85
4.24
3.10
10.45
54.77
5.13
0.09
12.22
3,110.70
2,705.79
72,844.44
21.47
155.50
0.02
13,944.93
8.44
170.77
Non-California
Number
of
Facilities
2
25
2
15
21
22
4
2
1
2
2
26
31
17
25
1
1
33
20
4
Number
of
Observ.
24
241
8
366
189
192
44
24
8
16
13
234
464
165
231
o
J
3
457
175
50
Number
of
Non-detects
0
2
1
17
65
59
6
7
0
0
1
6
43
0
21
1
1
17
23
1
Measured
Average
408.42
14,273.00
37.91
10,244.59
3.88
17.19
11.93
25.67
8.38
8.75
29.69
5,149.82
3,533.83
126,609.21
46.79
4.23
5.43
24,261.24
62.65
136.86
81
-------
Table 7.8. Direct Discharge Pollutant Concentration Levels Reported in
1982 Effluent Guidelines Development Document (Current/BPT)
Pollutant
Chloroform
Benzene
Toluene
2,4-Dichlorophenol0
p-chloro-m-cresol0
Dimethyl phthalate0
Diethyl phthalate
Di-n-butyl phthalate
Acanaphthene
Benzo(a)pyreneฐ
Chyrsene
Phenanthrene
Pyrene
Arsenicb
Beryllium
Cadmium0
Chromium (Trivalent)
Chromium (Hexavalent)
Copper
Cyanide
Lead
Mercury
Nickel
Selenium
Silver0
Thallium0
Zinc
Average
Flow-Weighted
Pollutant
Concentration (jig/L)
3.1
2.3
10.1
0.2
0.3
0.1
1.4
0.04
1.1
0.1
0.02
0.2
0.1
0.01
0.04
0.25
107.8
7.7
9.8
45.5
5.2
0.9
3.4
17.2
0.04
3.2
104.6
Maximum Pollutant
Concentration
(Mg/L)
66
11
35
10
10
3
30
10
6
3
1
1
7
31
2
20
1230
110
199
320
113
6
74
32
4
12
620
Frequency
of
Detection
2/17
3/17
1/17
1/17
1/17
1/17
1/17
2/17
1/17
2/17
2/17
1/17
1/17
3/17
2/51
3/93
41/53
8/48
25/50
26/39
10/54
20/45
13/55
17/20
1/47
5/14
43/59
82
-------
Notes
a. All 129 priority pollutants were analyzed during the sampling of the Current/BPT
wastestream. Thirteen organic pollutants and fourteen inorganic pollutants were detected. The
Current/BPT concentrations were calculated by flow-weighting the data available for the
seventeen direct dischargers sampled.
b. Low values were not included, and were assumed to be not quantifiable. High values were
not included because laboratory contamination was suspected; therefore, data were assumed to be
invalid.
c. The Current/BPT pollutant concentration is greater than that in the Pretreated Raw
wastestream because of the variability of the data during sampling.
7.1.4 Toxic Release Inventory
In October 1986, Congress enacted the Emergency Planning and Community Right-to-know
Act (EPCRA), as Title HI of the Superfund Amendments and Reauthorization Act (SARA).
Section 313 of EPCRA requires manufacturing facilities to report their annual use and releases of
more than 300 toxic chemicals to State and local Emergency Planning Commissions, and to the
EPA's Toxic Release Inventory (TRI).
Facilities are required to report releases and offsite transfers of EPCRA Section 313
chemicals to TRI, if they meet all of the following criteria:
They conduct manufacturing operations in SIC codes 20 through 39.
They have ten or more full-time employees or the equivalent.
They manufacture or process the EPCRA Section 313 chemical in an amount greater than
25,000 pounds/year.
They otherwise use the EPCRA Section 313 chemical in an amount greater than 10,000
pounds/year.
EPCRA requires reporting of five types of onsite releases: (1) fugitive air emissions;
(2) stack air emissions; (3) surface water discharges; (4) underground injections; and (5) land
disposal. EPCRA also requires reporting of two types of offsite waste transfers containing listed
chemicals; (1) transfers to POTWs; and (2) transfers to other treatment or disposal facilities. In
addition, EPCRA specifies that EPA must compile these release reports into an annual inventory
of releases and transfers and make this inventory available to the public. EPA stores reported
release data in the Toxics Release Inventory System (TRIS) which is maintained by EPA's Office
of Pollution Prevention and Toxics (OPPT).
7.1.5 Reported Annual Pollutant Loads from TRI
Two hundred five (205) petroleum refining facilities, identified by primary SIC code 2911,
reported data to TRI in 1991; some facilities may not have been required to report because they
did not meet employment or chemical use thresholds during 1991. Releases and transfers from
83
-------
the 205 facilities, located in 37 states, total 738.8 million pounds. The highest releases are from
ten states: California, Indiana, Louisiana, Mississippi, New Jersey, Ohio, Pennsylvania, Texas,
Utah, and Washington (Figure 7.1). The top two reporting states are Indiana and Pennsylvania
with 201.8 and 110.9 million pounds, respectively.
The largest releases and transfers are to offsite locations other than POTWs (e.g. to recycling,
energy recovery or treatment and disposal) (85 percent), air (10 percent) and underground
injections (3 percent). Releases to surface waters, transfers to POTWs, and land disposal each
represent less than 1 percent. The chemical with the highest reported amount of releases or
transfers is sulfuric acid (a byproduct of processing), accounting for 84 percent of the total
chemicals released or transferred. However, because refineries are required to neutralize their
wastes prior to discharge, the sulfate would not be in an acidic form and would be present as a
more benign the salt of inorganic cations in the wastewater. A total of 119 chemicals and
compounds are identified, 65 of which are released to either surface waters or POTWs.
84
-------
Figure 7.1 Geographic Distribution of 1991 TRI Chemical Releases and Transfers
CD 0-0.99 f3 1-99
Ibs/yr Total
100-999 iH 1,000-9,999
> =
85
-------
Forty-four priority pollutants discharged by petroleum refining facilities are reported to TRI
in 1991. Releases and transfers from these facilities total 33.5 million pounds. The largest
releases and transfers are to air (70 percent), offsite locations other than POTWs (15 percent),
POTWs (8 percent), underground injections (5 percent), and land (2 percent). Releases to
surface waters are less than 1 percent. The priority pollutant with the highest reported amount of
releases or transfers is toluene (44 percent). Seventy-seven (77) percent of the releases are from
three priority pollutants (toluene, benzene, and phenol).
Releases and transfers to POTWs and surface waters are identified from 112 direct
dischargers, 32 indirect dischargers, and 15 that discharge to both surface water and POTWs.
National totals for 1991 are 4.33 million Ibs/yr, or 34,100 Ibs/yr/facility to surface waters, and
6.94 million Ibs/yr, or 148,000 Ibs/yr/facility to POTWs. A table presenting TRI surface water
releases and POTW transfers by pollutant is presented in Appendix C. Annual loads of priority
pollutants are 207,000 Ibs/yr, or 1,600 Ibs/yr/facility to surface waters, and 2.52 million Ibs/yr, or
53,200 Ibs/yr/facility to POTWs. Transfers to POTWs (92 percent of all releases) far exceed
releases to surface waters. Figure 7.2 depicts the priority pollutants with the highest reported
amount of surface water releases or POTW transfers. Three priority pollutants (phenol, toluene,
benzene) account for 91 percent of the total releases and transfers. The vast majority of this load
(over 90 percent) would likely biodegrade or volatilize during typical secondary wastewater
treatment.
Assumptions and Limitations
Several assumptions and limitations of this analysis include:
Only facilities reporting releases of Section 313 chemicals, and meeting threshold
requirements, are required to report in TRI. Consequently, TRI may be incomplete in
terms of petroleum refining facilities, pollutants, or wastestreams.
Facilities in TRI with releases under 1,000 pounds for any one chemical may submit a
range of the release/transfer amount for that chemical. For this study, OPPT standards
are followed by using 5 pounds for loads of 0 to 10 pounds; 250 pounds for loads of 10 to
499; and 750 pounds for loads of 500 to 999.
Data reported by industrial facilities are determined by a variety of methods. Therefore,
the accuracy, precision, and comparability of TRI data are unknown.
Data are based only on facilities identified by primary SIC code 2911. Because many
facilities engage in numerous industrial activities, this may exclude some releases
generated from petroleum refining and include some releases not attributable to refining
activities.
86
-------
Pounds
2,000,000
1,500.000 -
1,000,000 --
500,000 -
Figure 12 1991 TRI Priority Pollutant Releases to Surface Water and POTWs
Pounds
r
Phenol Toluene
Surface Water H POTW
ป toMM M reported <> bur mrtri งMlpซt*it
Total Priority Pollutant Surface Water/POTW Releases = 2.73 Million Pounds
87
-------
7.2 Fate and Toxicity Evaluation of Released Pollutants
The environmental fate and toxicity of pollutant releases are evaluated by: (1) compiling
physical-chemical and toxicity data, and POTW inhibition and sludge contamination values for
identified pollutants; (2) categorizing the pollutants based on their potential toxicity and
environmental fate; (3) weighting loads from PCS and TRI according to toxicity and
bioaccumulation potential; and (4) evaluating whole effluent toxicity test data.
7.2.1 Compilation of Physical-Chemical and Toxicity Data and Information to Evaluate
Indirect Discharges
Physical-chemical properties and toxicity data, both measured and estimated, are compiled
for toxic pollutants currently being discharged by petroleum refining facilities according to
available sources. These data are compiled from Standards and Applied Science Division's
(SASD) Toxics Data Base (TDB) which contains aquatic toxicity, human health,
physical-chemical properties, and other information for over 1,600 toxic chemicals. The
chemical specific data needed to conduct the fate and toxicity evaluation for this study include:
Aquatic life criteria or toxic effect data for native aquatic species;
Human health reference doses (RfD);
Human health cancer potency slope factors;
EPA maximum contaminant levels (MCLs) for drinking water protection;
Henry's Law constants, vapor pressure and solubility values;
Soil/sediment adsorption coefficients (Koc);
Octanol-water partition coefficients (Kow);
Bioconcentration factors (BCF) for native aquatic species; and
Aqueous aerobic biodegradation rate constants.
Sources for the TDB include EPA ambient water quality criteria documents and updates,
EPA's Assessment Tools for the Evaluation of Risk (ASTER) and the associated Aquatic
Information Retrieval System (AQUTRE) and Environmental Research Laboratory-Duluth
fathead minnow data base, EPA's Integrated Risk Information System (IRIS), the Registry of
Toxic Effects of Chemical Substances (RTECS) data base, the Superfund Chemical Data Matrix
(SCDM), Syracuse Research Corporation's CHEMFATE and BIODEG data bases, EPA and
other government reports, scientific literature, and other primary and secondary data sources. To
ensure that the examination is as comprehensive as possible, alternative measures are taken to
compile data for chemicals for which physical-chemical property and/or toxicity data do not
exist. Therefore, where necessary, values are estimated using quantitative structure-activity
relationship (QSAR) models, or for some physical-chemical properties, utilizing published linear
regression correlation equations, if available.
Information needed to evaluate adverse effects on POTW operations and sewage sludge
quality includes inhibition values, sludge partitioning factors, and sludge contamination levels.
88
-------
The lower values for POTW removal rate indicate less removal at the POTW, and therefore a
higher portion of the pollutant reaches the receiving water. Inhibition values are the
concentration of influent to the POTW likely to interfere with treatment. Sludge partitioning
factors represent the proportion of a constituent load that will be found in primary or secondary
sewage sludge. EPA recently established sludge criteria for lead, chromium and zinc, which
restrict certain applications of sludge above criteria values. Inhibition values and sludge
partitioning factors are obtained from the Domestic Sewage Study (US EPA, 1986), Guidance
Manual for Preventing Interference atPOTWs (US EPA, 1987), and CERCLA Site Discharges to
POTWs guidance (US EPA, 1990). Data to determine allowable sludge contamination levels are
obtained from the Agency's final rule on "Standards for the Use or Disposal of Sewage Sludge"
(US EPA, 1993). Pollutant limits established for the final use or disposal of sewage sludge via
application to agricultural and non-agricultural land are reported.
Information is compiled and summarized from the TDB for pollutants regulated in the
discharge from 14 petroleum refining facilities located in Los Angeles County, California, which
discharge their wastewaters to POTWs (Table 7.9). Removal rates for these 25 pollutants vary
from a low of 0.52 (52 percent) for 2,4,6-trichlorophenol to a high of 0.98 (98 percent) for both
toluene and acenaphthene. The inhibition concentrations for these pollutants are generally high,
with the exception of lead and chromium. The sludge partitioning factors indicate that lead,
chromium, zinc and cyanide will remain in sludge. Sludge with contaminant levels that exceed
EPA criteria for less expensive disposal by land application must be disposed of through higher
cost alternatives, such as incineration.
Table 7.9. POTW Information for Selected Indirect Discharges
(Sorted by CAS Number)
CAS
Number
57125
67663
71432
74931
83329
85018
86737
88062
91203
95476
95578
100414
105679
106423
Pollutant Name
Cyanide
Trichloromethane
Benzene
Methanethiol*
Acenaphthene
Phenanthrene
Fluorene
Trichlorophenol, 2,4,6-
Naphthalene
Xylene, o-
Chlorophenol, 2-
Ethylbenzene
Dimethylphenol, 2,4-
Xylene, p-
POTW
Removal
0.704
0.676
0.941
0.77
0.983
0.949
0.698
0.516
0.947
0.951
0.65
0.938
0.85
Inhibition
Value
(ng/D
5,000
500,000
125,000
500,000
500,000
500,000
50,000
500,000
5,000
5,000
200,000
40,000
5,000
Partition
Factor
1
0.015
0.019
0.1
0.079
0.275
0.149
0.079
0.06
0.079
0.149
Sludge Criteria
Value (mg/kg)
89
-------
Table 7.9. POTW Information for Selected Indirect Discharges
(Sorted by CAS Number)
108383
108883
108952
129000
218019
7439921
7440473
7440666
7664417
14808798
18496258
Xylene, m-
Toluene
Phenol
Pyrene
Chrysene
Lead
Chromium
Zinc
Ammonia
Sulfate
Sulfide
0.654
0.976
0.967
0.95
0.97
0.918
0.754
0.78
0.6319
5,000
200,000
200,000
500,000
500,000
100
1,000
5,000
480,000
25,000
0.149
0.278
0.146
1
1
1
300
1200
2800
Note: * = Representative of Mercaptans
Assumptions and Limitations
Several assumptions and limitations of this compilation include:
Data are used from readily available electronic data bases; other primary and secondary
sources are not searched.
Many of the data are estimated and therefore have a high degree of associated uncertainty.
For some chemicals, neither measured nor estimated data are available for key
categorization parameters. As a result, this study is an incomplete assessment of
potential fate and toxicity of petroleum refining discharge.
7.2.2 Categorization of Pollutants
Human and ecological exposure and risk from toxic chemical releases is primarily a function
of toxic potency, inter-media partitioning, and chemical persistence. These factors are
dependant on chemical-specific properties relating to pharmocokinetic effects on living
organisms, physical state, hydrophobicity/lipophilicity, and reactivity; as well as the mechanism
and media of release and site-specific environmental conditions. The potential fate and toxicity
of pollutants associated with petroleum refining, based on chemical-specific data, are examined
in this portion of the study.
The objective of this generalized evaluation of fate and toxicity potential is to place
chemicals into qualitative groups based on their potential environmental fate and impact.
categorization groups are based on techniques derived for:
Acute aquatic toxicity (highly, moderately, or slightly toxic);
These
90
-------
Volatility from water (highly, moderately, slightly, or non volatile);
Adsorption to soil/sediment (highly, moderately, slightly, or non adsorptive);
Bioaccumulation potential (high, moderate, slight, or no significant potential); and
Biodegradation potential (fast, moderately fast, slow, or resistant).
The primary advantage of the categorization methods is that the results allow the user to
identify the potential impact/threat of a chemical. The methods effectively group chemicals
based on their potential to harm the environment. Using key parameters, these categorization
methods identify the relative aquatic toxicity and bioaccumulation fate for each chemical
constituent (with sufficient data) of petroleum refining discharges. In addition, the potential to
partition to various media (air, sediment/sludge, or water) and persist in the environment is
identified for each organic constituent. The acute aquatic toxicity, volatility from water,
soil/sediment adsorption, and bioconcentration categorization methods have been reviewed by
EPA staff (Offices of Water; Health and Environmental Assessment; and the former Office of
Toxic Substances), as well as by Dr. Warren J. Lyman, principal author of Handbook of
Chemical Property Estimation Methods (Lyman et al, 1982). The biodegradation categorization
method is based on Handbook of Environmental Degradation Rates (Howard et al, 1991).
These methods are used for screening purposes only, and do not take the place of detailed
pollutant assessments that analyze all fate and transport mechanisms.
This evaluation also identifies chemicals which (1) are known, probable, or possible human
carcinogens; (2) are systemic human health toxicants; and (3) have EPA human health drinking
water standards. The results of this analysis can provide a qualitative indication of potential risk
posed by the release of these chemicals. Actual risk depends on the magnitude, frequency, and
duration of pollutant loading; site-specific environmental conditions; proximity and number of
human and ecological receptors; and relevant exposure pathways. The categorization schemes
and ranges of parameter values defining the categories are presented in Appendix D.
The categorization assessment addresses the 96 individual pollutants identified from PCS
data and TRI releases to surface water and POTW transfers. Inorganic constituents include
heavy metals, halogens and other anionic species. The organic constituents encompass a broad
class of aliphatic and aromatic alkanes, alcohols, acids, ketones, and ethers. Also represented
are several polynuclear aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs), and
2378-TCDD.
Aquatic toxicity data are available for most pollutants, with the exception of inorganic acids
and anionic species. Discharges of these chemicals may cause indirect adverse ecological
effects by altering receiving water chemistry. However, these potential effects are not addressed
in this study. Fate and transport data (i.e., volatility from water, adsorption to soil/sediment, and
biodegradation potential) are available for most organic pollutants and some inorganic pollutants
(e.g., Henry's Law constant for mercury in its methylated form), therefore, the fate assessment is
applicable primarily to organic pollutants. Bioconcentration factors are available for over half of
the inorganic pollutants and all but two organic pollutants.
91
-------
A summary of the categorization results is presented on Tables 7.10 and 7.11.
Approximately one quarter of the pollutants exhibit high or moderate acute toxicity to aquatic
life (Table 7.10). Pollutants notable for their aquatic toxicity include CDD/CDF, mercury,
anthracene, cadmium, silver, and hexachlorobenzene (HCB). The most potent carcinogens
include CDD/CDF, hexavalent chromium, arsenic, 1,3-butadiene, and HCB. Antimony,
cadmium, and mercury are highly potent systemic toxicants. In total, 29 of the pollutants are
potential carcinogens, 56 are recognized by EPA as human systemic toxicants, and 45 have
EPA-assigned concentration limits for drinking water protection (Table 7.11).
Table 7.10 Number of Pollutants by Categorization Group
(96 pollutants evaluated)
Environmental Effects
and
Projected Fate
Acute Aquatic Toxicity
Volatility from Water
Adsorption to Solids
Bioaccumulation
Potential
Biodegradation
Potential
High
18
25
11
10
12
Moderate
7
17
2
18
21
Slight or
Slow
57
12
43
26
12
Not
Significant
4
6
22
15
No Data
14
38
34
20
36
Table 7.11 Number of Pollutants with Health Effect Designations
(96 pollutants evaluated)
Health Effect Designation
Carcinogenic Effects3
Human Systemic Effects'3
Drinking Water Values0
Number of Pollutants
29
56
45
a. Chemicals with EPA classification as a human carcinogen (A), probable human
carcinogen (B 1, B2), or possible human carcinogen (C). Dioxins/furans with a TEF
are also considered to be carcinogens.
b. Chemicals for which EPA has established a verified or provisional chronic
reference dose (RfD).
c. Chemicals for which EPA has established a concentration limit or goal under the
Safe Drinking Water Act (SDWA).
About three-quarters of the organic pollutants have a high to moderate potential to volatilize.
Most of the likely volatile chemicals are only slightly toxic to aquatic life. Notable exceptions
are HCB and mercury. Only 14 percent of the pollutants with data are highly or moderately
adsorptive to soil/sediment. However, most of the highly adsorptive chemicals are also highly
or moderately toxic to aquatic life. These include PAH compounds, phthalates, TCDD/TCDF,
92
-------
HCB, and mercury. One third of the pollutants with data have high to moderate
bioaccumulation potential, which is strongly correlated with soil/sediment adsorption.
TCDD/TCDF, HCB, mercury, and phenanthrene have the greatest potential to bioaccumulate.
Approximately half of the pollutants are expected to biodegrade fast or moderately fast in
oxygenated water. However, several highly to moderately toxic pollutants are resistant to
biodegradation or only slowly biodegrade. These chemicals include HCB, pyrene,
TCDD/TCDF, anthracene, and phenanthrene.
Combining toxicity and fate information can assist in identifying chemicals that have the
greatest potential to cause adverse impacts upon release. In the categorization methods, chronic
aquatic toxicity indicates the potential to reduce the viability of aquatic species populations and
adversely affect ecosystem stability downstream of a discharge. Aqueous aerobic
biodegradation half-life is used as a measure of persistence in the environment. Chemicals that
can cause chronic toxic effects in small amounts and persist for a long period of time are likely to
pose the greatest ecological threat. Figure 7.3 depicts a scatter plot of biodegradation half-life
data versus chronic aquatic toxicity levels. The gridlines show the categorization groups, with
high toxicity associated with low chronic aquatic toxicity levels and high persistence associated
with long half-lives. The names of chemicals falling in the moderate or high toxicity and the
resistant or slow biodegradation ranges (shaded region) are provided.
Based on high-end production-weighted 1992 PCS data, approximately 16 thousand
pounds/year of total cyanide and 12 pounds/year of hexachlorobenzene are released to surface
water. Based on 1991 TRI, 570 pounds of anthracene and hydrogen cyanide are released to
surface water annually.
93
-------
Figure 7.3. Ecological Impact Potential
/
r/1
!S
TJ
v*.
***
, ~
til
\
\
*
180
i
i
3
B
g 28
R
&
c
"8
i
m
Heuthhttfea
ซmn WW ปซ t*ซs|ป)ซn*ซปซซซ
IVl/O f^OH JB.
* * ป *
/ /*
AMtotteiwl
" ; T^
*
c*
i
ป^ซ^K
ซ * *
*
f ป
*
* ป
*
10 100
Shading indicates pollutants in high/moderate toxicity and resistant/slow degradation ranges
94
-------
An additional 13 inorganic pollutants that may or may not have long-term bioavailability,
have high or moderate chronic toxicity. The high-end production-weighted 1992 PCS load of 9
inorganic chemicals (aluminum, chlorine, selenium, lead, copper, mercury, chromium
hexavalent, silver, and cadmium) is 3.9 million pounds/year. According to 1991 TRI data,
45,000 pounds/year of inorganic chemicals (chlorine, antimony compounds, lead and
compounds, copper and compounds, cobalt and compounds, silver compounds, selenium and
compounds, and mercury) are released to surface water. Of particular note is selenium.
Selenium is shown to cause mortality, deformities, lack of embryonic development, and severe
reproductive impacts in a range of species (Harris 1991), including plants, amphibians, fish
(Hermanutz et al., 1992; Marcogliese et al., 1992), and aquatic birds (Hoffman et al., 1988;
Ohlendorf et al., 1986; Ohlendorf, et al., 1987; Ohlendorf et al., 1989; US EPA, 1989).
Cancer slope factor, reference dose, and mammalian LDso are used in the categorization
methods to indicate potential to cause adverse health effects on exposed human populations. A
primary human exposure route for chemicals released to surface water is the consumption of
contaminated fish. In the categorization methods, bioconcentration factor indicates the degree to
which a chemical may accumulate in fish tissue. Chemicals that accumulate in edible fish tissue
and may cause adverse health effects in small amounts are likely to pose the greatest threat to
human health. Figure 7.4 depicts a scatter plot of bioconcentration factors versus "critical
doses" for human health toxicity. Critical human health doses are derived by converting cancer
slope factors and LDso values to an equivalent reference dose unit (mg/kg-day) based on the
categorization methods. The names of chemicals falling into the high or moderate ranges for
both parameters (shaded region) are provided. According to production-weighted 1992 PCS
data, as much as 16,000 pounds/year of these chemicals are released to surface water.
95
-------
Figure 7.4. Human Health Impact Potential
1
^500
a
s
g 50
ง
a
ง 5
3
#^m|
ffl
TCD&
* Wta
*
TW ' HปnปฃMeซ*ป
Mtweary /
. 2,-M
-- - ' - ' ' *
, ^
ป *
*
<
*
ซM.
"~*~~~~~~1
'*A'*4i.*. "- i
M*MMI _
*
*
~-~ ^ ,.ป .^p. ~-
* *
*
_^^-:. ^
?*, -.
* i *
* * '
j
t
i * ~
* *
* *
*
0.05
Shading indicates pollutants in high/moderate toxicity and bioaccumulation potential ranges
96
-------
Assumptions and Limitations
Assumptions and limitations of this analysis include:
Placement into groups is based on arbitrary order of magnitude delineations for several
categorization schemes. Combined with data uncertainty, this may lead to an
overstatement or understatement of the characteristics of a chemical.
Receiving waterbody characteristics, pollutant loading amounts, exposed populations, and
potential exposure routes explicitly are not considered.
Bioavailability of inorganic pollutants is not assessed. Ionic specification,
dissolved-solid phase equilibrium, and attachment to clay particles or organic matter are
largely functions of waterbody characteristics.
Human health toxicity assessment is based on an ingestion exposure route, and may not
accurately reflect the hazard posed by inhalation or dermal contact.
Biodegradation potential may not be a good indicator of persistence for organic chemicals
that rapidly photooxidize or hydrolyze, since these degradation mechanisms are not
considered.
Available aquatic toxicity and bioconcentration test data may not represent the most
sensitive species.
Data derived from laboratory tests may not accurately reflect conditions in the field.
7.2.3 Toxic Weighting Factor Analysis
EPA's Office of Water uses toxic weighting factors (TWFs) analysis to compare the relative
toxicity of industrial effluent discharges. Toxic weighting factors are derived using the same
methodology employed for other effluent guidelines (US EPA, 1992a), but are based on updated
toxicity information.
Originally, TWFs were used to calculate copper based pound-equivalents, and were derived
from chronic aquatic life criteria (or toxic effect levels) and human health criteria (or toxic effect
levels) established for the consumption offish. For carcinogenic substances, the human health
risk level was set at 10"5, i.e. protective to a level allowing 1 in 100,000 excess cancer cases over
background. These toxicity levels were related to a benchmark value, or toxicity level
associated with a single pollutant. Copper, a toxic metal pollutant commonly detected and
removed from industrial effluent, was selected as the benchmark pollutant (i.e., the basis to
which others are compared). EPA had used copper previously in TWF calculation for the
cost-effectiveness analysis of effluent guidelines. While the water quality criterion for copper
has been revised (to 12.0 |ig/L), the TWF method used the former criterion (5.6 |ig/L) to
facilitate comparisons with cost-effectiveness values calculated for other regulations. The
criterion for copper was reported in the Ambient Water Quality Criteria for Copper (US EPA,
1980).
97
-------
In the original method, a TWF for aquatic life effects and a TWF for human health effects
were added for pollutants of concern. The calculation was performed by dividing aquatic life
and human health criteria (or toxic effect levels) for each pollutant, expressed as a concentration
in micrograms per liter (|ig/L), into the former copper criterion of 5.6 |ig/L:
5.6 5.6
TWF = z- AQ + -^-
Where:
TWF = Original toxic weighting factor
AQ= Chronic aquatic life value (|ig/L)
HHOO= Human health (ingesting organisms only) value (|ig/L)
With the new method, pollutant weighting factors (PWFs) are derived from either chronic
aquatic life criteria (or toxic effect levels), or human health criteria (or toxic effect levels)
established for the consumption of water and fish. For carcinogenic substances, the human
health risk level is 10"6, that is, protective to a level allowing 1 in 1,000,000 excess cancer cases
over background. In contrast to original TWFs, PWFs are not related to a benchmark pollutant.
PWFs are derived by taking the reciprocal of the more stringent (smallest value) of the aquatic
life or human health criterion or toxic effect level, both expressed in concentration units of
micrograms per liter (|ig/L):
1
PWF = -AQ , if AQ < HHWO
1
PWF = -HHWO , if HHWO < AQ
or
Where:
PWF = Pollutant weighting factor
AQ = Aquatic life chronic value (|ig/L)
HHWO = Human health (ingesting water and organisms) value (|ig/L)
Individual TWFs and PWFs for petroleum refinery wastewater constituents are presented in
Appendix E. The differences between original TWFs and new PWFs are summarized below:
98
-------
Feature
Benchmark Value
(numerator)
Carcinogenic Risk
Level
Human Health
Exposure
Aquatic Life Effects
vs. Human Health
Effects
TWF
5.6 (former
freshwater chronic
criterion for copper)
10-5 (1 in 100,000
excess cancer cases)
Fish consumption
only
TWFs are added
PWF
1
10-6 (1 in 1,000,000
excess cancer cases)
Drinking water and fish
consumption
More stringent PWF is
used
Application to PCS and TRI Load Estimates
TWFs are applied to the PCS and TRI load estimates given in Section 7.1. Toxic-weighted
loads provide a measure for comparison between industries. Based on TWFs, approximately 75
percent of the high-end PCS production-weighted load is categorized as being a priority pollutant
and a much smaller percentage is classified as carcinogenic. Based on PWFs, more than 90
percent of the PCS weighted load is from priority pollutants, and carcinogens dominate this
total (Table 7.12). TRI loads to surface water show a lower proportion of priority pollutants,
due to the difference in the set of pollutants reported to TRI and those that are reported to PCS.
Transfers to POTWs have a higher percentage of priority pollutant TWF load (80 percent) than
those to surface water (60 percent). Average TWF loads per facility to surface water and to
POTWs show that indirect wastestreams have hazard potentials that are almost three times
greater than direct wastestreams (1,854 vs. 728 Ibs-eq/yr/fac).
99
-------
Table 7.12 Petroleum Refining Annual Loads from PCS and TRI
Unweighted
PCS Surface Water Releases
(1992)
Extrapolated PCS Surface Water
Releases (1992)
TRI Surface Water Releases
(1991)
TRI POTW Transfers (1991)
All
Pollutants
Total
Ib/yr
30,711,578
272,726,320
4,330,091
6,942,533
All
Pollutants
Average*
Ib/yr/fac
1,976,278
34,095
147,713
Priority
Pollutants
Total
Ib/yr
67,222
1,606,049
206,553
2,522,607
Priority
Pollutants
Average*
Ib/yr/fac
11,638
1,626
53,672
EPA
Classified
Carcinogens
Total
Ib/yr
11,593
121,361
33,362
282,726
EPA
Classified
Carcinogens
Average*
Ib/yr/fac
879
263
6,015
Standard TWFs Ib-eq/yr Ib-eq/yr/fac Ib-eq/yr Ib-eq/yr/fac Ib-eq/yr Ib-eq/yr/fac
Extrapolated PCS Surface Water
Releases (1992)
TRI Surface Water Releases
(1991)
TRI POTW Transfers (1991)
10,157,542
92,454
87,123
73,605
728
1,854
7,889,816
54,564
69,821
57,173
430
1,486
389,684
23,450
10,490
2,824
185
223
Optional TWFs Ib-eq/yr Ib-eq/yr/fac Ib-eq/yr Ib-eq/yr/fac Ib-eq/yr Ib-eq/yr/fac
Extrapolated PCS Surface Water
Releases (1992)
TRI Surface Water Releases
(1991)
TRI POTW Transfers (1991)
6,226,774
497,817
242,157
45,122
3,920
5,152
5,821,318
59,992
233,967
42,183
472
4,978
4,579,868
484,834
228,292
33,187
3,818
4,857
* Derived by dividing individual pollutant loads (Ib/yr or Ib-eq/yr) by:
1. 138 (number of facilities included in the PCS extrapolation), or
2. 127 (number of facilities reporting surface water releases in TRI), or
3. 47 (number of facilities reporting POTW transfers in TRI);
and summing for all pollutants.
7.2.4 Whole Effluent Toxicity Testing
EPA has advocated an integrated approach to water quality-based toxics control via
chemical-specific testing, bioassessment, and whole effluent toxicity testing. These methods,
taken in combination, are expected to provide a comprehensive biological evaluation of a water
body (US EPA, 1992b). Whole effluent toxicity (WET) refers to the evaluation of toxic effects
of an effluent on living organisms, and has primarily been used for the protection of aquatic life.
Whole effluent toxicity is defined as the "aggregate toxic effect of an effluent as measured
directly by a toxicity test."
100
-------
Using the whole effluent approach for the protection of aquatic life involves using acute (usually
96-hours or less in duration with lethality as the typical endpoint) and/or chronic (generally 7-day
with lethality, reproduction, and growth effects as test endpoints) toxicity tests to measure the
aggregate effects of the pollutant discharge. See EPA's Technical Support Document for Water
Quality-Based Toxics Control (US EPA, 1991). The acute toxicity endpoint (ATE) values are
generally reported as LCso values, defined as the concentration at which 50 percent of the test
organisms died. In addition, an exposure duration is often reported along with the lethal
concentration value such as a 96-hour LCso value.
Chronic toxicity test results may be reported in terms of a number of different endpoint
values. These commonly include the No Observed Effect Concentration (NOEC), Lowest
Observed Effect Concentration (LOEC), or the Effect Concentration (EC). These and other
chronic toxicity endpoints (CTEs) are defined by EPA in the Technical Support Document.
WET tests provide an indication of the ecological impacts of pollutants on receiving waters.
Tests conducted by EPA's Complex Effluent Toxicity Testing Program, the University of
Kentucky, the University of North Texas, and North Carolina Division of Environmental
Management showed a strong correlation between actual receiving water impact and the
predicted results from whole effluent toxicity tests. As acknowledged in the EPA Technical
Support Document, the correlation has been strongest when related to maximum impact
responses, or acute WET tests.
EPA's Region 6 Office typically requires that there be no statistically significant lethality in
the 7-day chronic WET test at the critical effluent concentration (low flow). Effluent from 18
out of 47 petroleum refining facilities, Texas, Louisiana and Oklahoma (approximately 40
percent) failed at least one WET test for acute, chronic or sublethal effects. All but two of these
facilities also showed a statistically significant lethality in their test results at least once, and 11
of these facilities (approximately 25%) showed persistent lethality (i.e., the facility also failed a
re-test). California has stringent acute and chronic WET test requirements in place. These
requirements typically stipulate that the median test result not reduce survival below 90 percent
and the percentile test result not reduce survival below 70 percent during a 96 hour test
conducted on fingerling trout, stickleback, and fathead minnows. Shallow water discharges
must use 100 percent effluent, whereas deep water dischargers are typically given a 10:1 dilution.
As a result of these testing requirements, many petroleum refineries now have activated carbon
systems in place to achieve compliance. The State of California is currently compiling a data
base of statewide WET test results, which may be available for data summarizing in the future.
However, as of January 1994, all five refineries in the San Francisco Bay region with chronic
WET test requirements were in compliance.
Assumptions and Limitations
There are several potential limitations to using WET tests as a measure of receiving water
impact:
101
-------
WET test results are assumed to be valid independent of water body type.
Biological, physical and chemical factors at any site will affect the true toxic effect of a
given effluent, giving both false positives and false negatives.
7.3 Documented Impacts
Tables with supporting data for the following documented environmental impacts are
presented in Appendix A. In a review of over 60 literature abstracts (accessed through the
DIALOG data service), EPA found that four laboratory studies reported potential environmental
impacts from petroleum refinery wastewaters. Impacts include aquatic life effects such as spinal
curvature, co-carcinogenic activity, behavioral pattern changes in fish, and mutagenic activity.
Tests using treated refinery effluent report a 48-hour LCso for Daphniapulex at 76 percent
effluent (a 3:1 mixture of effluent to water), and a 14-day LCso at 6.4 percent effluent,
representing a threshold value for mortality. Sublethal effects, such as reproductive failure,
ranged from a 14-day ECso at 3.1 percent effluent to a 14-day ECs at 0.52 percent effluent.
Chemical characterization studies using Daphnia magna static bioassays determine that the most
toxic fraction of petroleum refinery wastewaters are steam volatiles, base-neutrals, and aromatic
organics.
Twenty-three petroleum refining facilities (17 percent of the 138 direct dischargers identified
in the Oil and Gas Journal (Thrash, 1991)) are identified by States as point sources impairing (or
contributing to impairment of) water quality and are included on their CWA Section 304(1) Short
List. Pollutants of concern include 10 metals (antimony, cadmium, chromium, hexavalent
chromium, trivalent chromium, copper, lead, mercury, nickel, selenium, silver, and zinc),
cyanide, phenol and toxicity as reflected by whole effluent-toxicity (WET). Section 304(1)
requires States to identify waterbodies impaired by the presence of toxic substances, identify
point source dischargers of these toxics, and develop Individual Control Strategies (ICSs) which
identified dischargers were required to implement by July 30, 1993. In accordance with the
statutory provisions, states must submit to EPA three lists of water bodies, one of which is
termed the "short list." The "short list" (Section 304(1)(B)) is a list of waters for which a state
does not expect applicable water quality standards (numeric or narrative) to be achieved after
technology-based requirements have been met, due entirely or substantially to point source
discharges of Section 307(a) toxics.
Three cases of sediment contamination are identified with petroleum refineries from An
Overview of Sediment Quality in the United States (EPA, 1987b). The associated contaminants
are cadmium, chromium, copper, cyanide, nickel, lead, zinc, PCBs, PAHs, petroleum
hydrocarbons, and oil and grease. This report presents an overview of sediment quality and
qualitatively describes the nature and extent of contaminated sediments (i.e., bottom deposits in
rivers, lakes, harbors, and oceans) polluted from anthropogenic sources. Information for this
report is from a review of the published literature and inquiries to environmental agencies. The
102
-------
data collection effort is not statistically designed or geographically complete; sites are chosen for
inclusion based on the sources of information available.
Petroleum refining ranks second among 40 industrial categories (based on TRI data) as a
source of sediment contaminants in EPA's 1995 draft report on sediment contamination (US
EPA, 1995)(see Appendix I, "Analysis of TRI Data by Industrial Category"). In addition, the
report ranks petroleum refining 11 out of 43 industrial categories based on 1992 PCS data
(Appendix I, "Analysis of PCS Data by Industrial Category"). The rankings are based on
industry loads weighted by individual chemical toxicity and fate specific to potential sediment
contamination. Unitless relative hazard scores are developed for each chemical by multiplying
chemical loads retrieved from PCS and TRI by a toxicity factor, based on relative potential
toxicity to aquatic life or human health when present in sediment, and a fate factor, based on
relevant fate and transport factors.
A review of the EPA/Army Corps of Engineers draft report, Evaluation of Dredged Material
Proposed for Discharge in Waters of the United States - Testing Manual (US EPA, 1994a),
reveal contaminants associated with petroleum refineries include ammonia, lead, selenium,
2,3,7,8-tetrachlorodibenzo:p-dioxin (TCDD) and 2,3,7,8-tetrachlorodibenzofuran (TCDF). This
report provides a matrix of potential correlations between industrial sources and specific
contaminants of sediments based on a compilation of existing information. This matrix is,
however, not all inclusive and makes no accounting for current pollution control practices. See
Appendix I.
103
-------
8. Economic Profile of the Petroleum Refining Industry
8.1 U.S. Petroleum Refinery Geographic Distribution and Trends
8.1.1 Number and Distribution of Refineries
Petroleum refineries in the United States are classified into five geographic groups called
Petroleum Administration for Defense (PAD) Districts. All 50 states and the District of
Columbia are distributed among these five districts (see Table 8.1). However, this does not
imply that there are refineries in all 50 states. These districts were originally created for
economic and geographic reasons as "Petroleum Administration for War (PAW)" districts, which
were first established in 1942. In 1950 the nomenclature was changed to "PAD."
As of January 1, 19932 there were 187 operable refineries in the United States with a total
atmospheric crude oil distillation capacity of 15,120,630 barrels per day (see Table 8.2).
However, only 175 refineries were operating with an atmospheric crude oil distillation capacity
of 14,776,880 barrels per day. The remaining 12 refineries were idle. During 1992, 2 new
refineries were put into operation, 1 was reactivated, 15 were shut down, and 2 refineries were
sold to new operators.
As observed in Table 8.1, PAD District in, the Gulf Coast, is the largest in terms of number
of refineries and also in capacity, with 66 operable refineries with a capacity of 6,764,450 barrels
per day, which amounts to 35 percent of the number of refineries and 45 percent of the entire
capacity in the United States. PAD District II, the Midwest, is the second largest in terms of
capacity, producing 3,398,800 barrels per day (22.5 percent) from 38 refineries (20 percent).
The second highest number of refineries is found in PAD HI, the West Coast, with 45 operable
refineries (24 percent). This district ranks third in production with a capacity of 2,895,800
barrels per day (19 percent). The smallest district is PAD IV, the Mountain States, with 17
operable refineries (9 percent), and a capacity of 519,375 barrels per day (3 percent).
2 While 1993 data are available and exhibit a declining trend, the 1992 data from the draft
report were not updated so as to maintain comparability with the technical data on refineries and
the PCS system from 1992 summarized in the rest of the final report.
104
-------
Table 8.1 Number and Capacity of Operable Petroleum
Refineries by PAD District as of January 1,1993
Number o
PAD District Refin
and State
Total Oper
PAD District I 21
Delaware 1
Georgia 2
New Jersey 6
New York 1
Pennsylvania 8
Virginia 1
West Virginia 2
PAD District II 38
Illinois 7
Indiana 4
Kansas 7
Kentucky 2
Michigan 3
Minnesota 2
North Dakota 1
Ohio 4
Oklahoma 6
Tennessee 1
Wisconsin 1
PAD District III 66
Alabama 3
Arkansas 3
Louisiana 20
Mississippi 6
New Mexico 3
Texas 3 1
PAD District IV 17
Colorado 3
' Operable Atmospheric Crude Oil Distillation
icries Capacity (bbl/calendar day)
ating Idle Total
17 4 1,542,805
1 0 140,000
1 0 33,540
4 2 527,500
0 2 41,850
8 0 731,415
1 0 53,000
2 0 15,500
36 2 3,398,200
7 0 965,600
4 0 474,900
6 1 327,300
2 0 218,900
2 1 118,600
2 0 267,100
1 0 58,000
4 0 462,100
6 0 396,500
1 0 76,000
1 0 33,200
64 2 6,764,450
2 1 119,500
3 0 61,700
20 0 2,358,900
6 0 371,800
3 0 94,600
30 1 3,757,950
16 1 519,375
2 1 95,500
Operating
1,352,955
140,000
5,540
407,500
0
731,415
53,000
15,500
3,364,800
965,600
474,900
296,900
218,900
115,600
267,100
58,000
462,100
396,500
76,000
33,200
6,722,450
104,500
61,700
2,358,900
371,800
94,600
3,730,950
509,375
85,500
Idle
189,850
0
28,000
120,000
41,850
0
0
0
33,400
0
0
30,400
0
3,000
0
0
0
0
0
0
42,000
15,000
0
0
0
0
27,000
10,000
10,000
105
-------
Montana
Utah
Wyoming
PAD District V
Alaska
Arizona
California
Hawaii
Nevada
Oregon
Washington
U.S. Total
4
6
4
45
6
2
26
2
1
1
7
187
4
6
4
42
6
1
24
2
1
1
7
175
0
0
0
3
0
1
2
0
0
0
0
12
139,650
154,500
129,725
2,895,800
256,300
14,000
1,933,900
146,300
7,000
0
538,300
15,120,630
139,650
154,500
129,725
2,827,300
256,300
10,000
1,869,400
146,300
7,000
0
538,300
14,776,880
0
0
0
68,500
0
4,000
64,500
0
0
0
0
343,750
Source: U.S. DOE, Energy Information Administration
Table 8.2 Number and Capacity of Refineries in
California, Louisiana, and Texas as of January 1,1993
California
Louisiana
Texas
Total CA, LA, TX
% of U.S. Total
Source: US DOE/EIA,
Number of Refineries
26
20
31
77
41%
Petroleum Supply Annual
Capacity
(bbl/day)
1,933,900
2,358,900
3,757,950
8,050,750
53%
Fifty-three percent of the nation's refining capacity is concentrated in 3 states: California,
Louisiana, and Texas. These 3 states combined contain 77 refineries with a combined capacity
of 8,050,750 barrels per day (see Table 8.2). The remaining 110 refineries (59 percent) are
distributed among 32 other states with a combined capacity of 7,069,080 barrels per day (47
percent). As of January 1, 1993, 15 states did not have refining capacity (see Table 8.3).
106
-------
Table 8.3 States with No Refining Capacity
as of January 1,1993
Connecticut
Florida
Idaho
Iowa
Maine
Maryland
Massachusetts
Missouri
Nebraska
New Hampshire
North Carolina
Rhode Island
South Carolina
South Dakota
Vermont
8.1.2 Trends in the Number of Refineries
Since 1981 the number of refineries has decreased from 324 to 187. During this period 156
refineries have shut down, with 69 closures (44 percent) in PAD District HI (Gulf Coast), making
this the highest number of closures among all 5 districts.
The number of refineries reached a historical high of 324 in 1981. The growth in the
number of refineries was largely the result of government regulatory policy. The combination of
price controls and non-market allocative mechanisms were introduced in the Emergency
Petroleum Allocation Act (EPAA) of 1973 and continued under the Energy Policy and
Conservation Act (EPCA) of 1975 (Vogely, 1985; Bohi and Russell, 1978). With the adoption
of EPAA, the primary problem was assigning rights to price controlled oil among refiners. An
"entitlements" program was designed to equalize the effective cost of crude oil to all refiners at a
level equal to the national weighted average of controlled and uncontrolled prices. Cash
settlements were made from refiners with lower-than-average crude oil acquisition costs to
refiners with higher-than-average acquisition costs. Under EPCA, refiners with crude runs of
less than 50,000 bbl/day were exempted from any entitlements obligation, regardless of the small
refiner's access to low-cost crude oil (Piccini, 1992). This was the so-called "small refiner
bias." As a result of the program, many small refineries were created particularly in PAD
district III (Gulf Coast), for access to low-cost crude oil, and typically with only crude distillation
capability.
The program ended in 1981, eliminating special treatment for small refineries which eventually
caused many of them to go out of business. Since 1981 the number of small refineries that
produce under 10,000 barrels per day has decreased from 82 to 32 (a 61 percent decrease) (see
107
-------
Table 8.4). However, the number of large refineries with capacities over 100,000 bbl/day has
actually increased from 52 to 54 refineries between 1982 and 1993 (see Table 8.4).
Table 8.4 Number of Operable Refineries
Classified by Operable Atmospheric Crude Oil Distillation Capacity
Date
1/1/82
1/1/83
1/1/84
1/1/85
1/1/86
1/1/87
1/1/88
1/1/89
1/1/90
1/1/91
1/1/92
1/1/93
Refinery Capacity (bbl/day)
100,000
+
52
48
47
47
47
50
54
53
56
55
54
54
30,001 -
100,000
87
84
82
77
74
75
72
71
70
70
68
65
10,001 -
30,000
80
59
55
43
46
42
41
37
41
42
38
33
10,000 or
less
82
67
63
56
49
52
46
38
43
33
36
32
Total per
year
301
258
247
223
216
219
213
204*
205 *
202*
199*
187*
* The sum of the columns do not equal the total as some operable refineries possess
only vacuum distillation capacity with no atmospheric crude oil distillation
capacity.
Source: US DOE/EIA, Petroleum Supply Annual
108
-------
8.2 Economic Profile
8.2.1 The FRS Companies
Financial Reporting System (FRS) companies comprise a number of the largest oil producing
companies in the U.S. who report their annual performance to the Energy Information
Administration (EIA). Thirty-three companies have submitted data to EIA since EIA started to
collect this type of information in 1977.
These companies occupy a major position in the U.S. economy. In 1991 their sales were
equal to 21 percent of the $2.3 trillion in sales by Fortune magazine's list of the 500 largest U.S.
industrial corporations, and their profits and assets were each equal to 27 percent and 18 percent
respectively of those of the "Fortune 500" companies. In 1991, The FRS companies accounted
for 56 percent of total U.S. crude oil and natural gas liquids production, 44 percent of U.S.
natural gas production, and 64 percent of the U.S. refinery capacity.
The Energy Information Administration uses the financial data from the FRS companies as a
surrogate for the refining industry as a whole because of the ready availability, completeness,
reliability, and continuity of the data. During the 1980's, FRS refineries accounted for between
75 and 80 percent of total domestic refining capacity, although this has fallen during the early
1990's to approximately 70 percent. As can be seen in Table 8.5, the net income derived from
refining and marketing operations has experienced wide fluctuations during the period from 1979
to 1992, with the only loss recorded in 1992. Although FRS companies are large integrated
entities, economic factors such as increases in raw material costs have a similar effect (in this
case increased costs) on both FRS and non-FRS companies. However, the financial health of
FRS and non-FRS companies may be significantly different.
109
-------
Table 8.5 Income from Refining
and Marketing Operations (FRS)
(in Millions of dollars)
Year
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
Net Income
2,301
2,518
1,278
1,913
1,636
105
2,281
1,641
1,073
5,443
4,522
2,184
903
(200)
Source: US DOE/EIA, Annual Energy Review
110
-------
8.2.2 Refined Product Margins
Refined product margins are a good indicator of overall refinery financial performance.
Refined product margin is defined as the difference between the composite refiner acquisition
price of crude oil and the price of refined products to resellers (i.e., wholesale prices). The
composite price of refined products includes aviation gasoline, kerosene-type jet fuel, kerosene,
motor gasoline, distillate fuel nos. 1, 2, and 4, and residual fuel. Price controls were in effect in
late 1970's and early 1980's (ending in 1981), thus making interpretation of the margin difficult.
As can be seen in Table 8.6, thereafter, margins have experienced significant volatility, reaching
a peak of 22.1 cents per gallon in 1990 and a trough of 13.7 cents per gallon in 1984. The
trough in 1984 can be largely attributed to weakened prices for motor gasoline. The peak in
1990 (the year of the Persian Gulf crisis) can be attributed to significant increases in prices for
aviation gasoline, motor gasoline, and jet fuel.
Table 8.6 Composite Refiner Margin
Year
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
Margin (Cents per Gal)
19.4
22.4
19.4
19.4
16.0
13.7
17.0
15.8
13.8
18.7
18.8
22.1
20.7
19.8
Source: US DOE/EIA, Annual Energy Review
111
-------
8.2.3 Refined Products Imports
The level of imports of refined products has remained stable throughout the 1980's and early
1990's, as shown in Table 8.7. Imports reached a peak of 2,295,000 barrels per day in 1988 and
reached a trough of 1,625,000 barrels per day in 1982. No discernible long-term trend can be
observed in the period.
Table 8.7 Refined Product Import Volumes
(in thousands of barrels per day)
Year
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
Imports
1,937
1,646
1,599
1,625
1,722
2,011
1,866
2,045
2,004
2,295
2,217
2,123
1,844
1,805
Source: US DOE/EIA, Annual Energy Review
112
-------
8.3 Impacts of Environmental Regulations
As observed in Figures 8.1 and 8.2, capital and operation and maintenance (O & M) costs for
pollution abatement have been increasing since 1976. Solid waste pollution abatement
expenditures have increased only slightly while pollution abatement expenditures for water and
air have shown significant increases, especially in 1990 and 1991.
113
-------
Figure 8.1. Refinery Pollution Abatement Operating Expenditures
Millions of Dollars
3,000
2,500
2,000
1,500
1,000
500
1976 1977 1878 197S 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991
Year
Air Water Solid Waste
Source: Bureau of the Census. 1987 costs are estimated.
114
-------
Figure 8.2. Refinery Pollution Abatement Capital Expenditures
Millions of Dollars
1,600
1.400
1.2OI
1,000
800
600
400
200
1977 1978 1979 1981 1SS2 1S84 1SS? 1990 1991
Air m m Solid
Source: Bureau of the Census. 1987 costs are estimated.
115
-------
8.3.1 Air Pollution Abatement Expenditures
A major driving force in increased air pollutant abatement expenditures are the Clean Air Act
Amendments of 1990 (CAAA). The Act establishes deadlines and procedures for bringing areas
into compliance with National Ambient Air Quality Standards.
Ozone and carbon monoxide (CO) are two air pollutants targeted by the CAAA. EPA has
identified areas of the country that are not in compliance with the ozone and CO standards
("non-attainment areas") and has categorized these areas according to the severity of their
non-attainment. Petroleum refineries in these areas will be required to reduce air emissions
from stationary sources. Refineries will be required to undergo modifications to equipment,
increase inspection and maintenance programs in order to reduce fugitive emissions.
Table 8.8 shows one such estimate of the incremental costs for refineries to meet the new
requirements of the CAAA. This estimate was a result of a 1993 environmental study covering
the years 1991-2010 conducted by the National Petroleum Council (NPC), an advisory
committee to the Secretary of Energy.
Table 8.8 National Petroleum Council Estimates of Incremental Cost to Meet the New
CAAA Requirements
Item
Capital
Investments
One-Time Costs
Total
O & M Expenses
$ Millions
1991-1995
3,537
9
3,546
-
1995
..
_
228
1996-2000
1,874
29
1,903
-
2000
..
_
454
2001-2010
2,090
..
2,090
-
2010
..
_
152
Total
7,501
38
7,539
~
Note: Costs are expressed in mid-1990 Gulf Coast Dollars
Source: National Petroleum Council
According to the NPC, the estimated capital investment of $7,501 million (or $7.5 billion)
will be spread over four types of emissions as identified in Table 8.9.
116
-------
Table 8.9 National Petroleum Council Estimate of
Emission Control Investments (by Emission Type)
Emissions
Volatile Organic Compounds (VOC)
Particulate Matter (PMio)
Sulfur Dioxide (SO2)
Nitrogen Oxides (NOx)
Toxics
$ millions
3,760
1,628
965
921
227
Percent
50.1
21.7
12.9
12.3
3.0
8.3.2 Water Pollution Abatement Expenditures
As observed in Table 8.10, pollution abatement capital expenditures for water have also been
on the increase. The NPC expected that a forthcoming reauthorization of the Clean Water Act
would be a major driving force in future increased water pollution abatement expenditures.
Table 8.10 shows the incremental cost estimates for the U.S. refining industry to meet the
provisions of a revised Clean Water Act according to the NPC environmental study covering
1991-2010.
Table 8.10 National Petroleum Council Estimates of Incremental Cost to Meet the New
Requirements of Clean Water Act Reauthorization
Item
Capital
Investments
One-Time Costs
Total
O&M
Expenses
$ Millions
1991-1995
1,251
1,251
-
1995
~
44
1996-2000
4,478
4,478
-
2000
~
405
2001-2010
6,602
8
6,610
-
2010
~
573
Total
12,331
8
12,339
-
Note: Costs are expressed in mid-1990 Gulf Coast Dollars
Source: National Petroleum Council
117
-------
According to the NPC, these figures have been developed from existing and anticipated
wastewater regulations.
The assumptions used to develop these costs based on Clean Water Act reauthorization work
at the time of the NPC study include the following group of requirements:
Reduction of wastewater toxicity and biomonitoring
Elimination of chromium compounds from cooling towers
Storm water permit requirements to exclude oil (in storm water) from tank draw offs
Storm water requirements to exclude oil from sampling (in storm water)
Storm water requirements to exclude exchanger cleaning wastes (from storm water)
Storm water permit requirements to reduce discharge of suspended solids (in storm water)
Store and treat quantity of contaminated storm water from 10-year storm
Other assumptions anticipated additional regulations applicable to water, wastewater and
groundwater and include:
Anticipated requirements for process wastewater reuse
Mandated application of a revised Best Available Technology (Effluent Limitations
Guidelines)
Anticipated requirements to assess and remediate sediments in outfall areas
Prevent ground water pollution from storage tank areas
Prevent ground water pollution from underground process piping
Prevent groundwater pollution from underground process sewers
The NPC estimate for capital investment over these three areas is shown in Table 8.11. Table
8.12 shows the time frame for the projected investments, estimated to total $12.3 billion.
Table 8.11 National Petroleum Council Estimate for
Water Pollution Control Investments
Item
CWA Reauthorization
Storm Water Quality
Ground Water Issues
$ Million
1,251
1,196
3,549
Percent
10.1
9.7
28.8
Table 8.12 National Petroleum Council Projection
118
-------
for Water Pollution Control Investments
by Time Period
Period
1991-1995
1996-2000
2001-2010
$ Million
1,251
4,478
6,602
Percent
10.1
36.3
33.6
The NPC study predicts that the major area of wastewater investment will be made to reduce and
control the toxicity of refinery wastewater effluent during 1996 through 2010 time frame.
119
-------
Bibliography
American Petroleum Institute. 1977. Water Reuse Studies. API Publication 949, Washington,
D.C., 1977. 132pp.
Bohi, D.R. and M. Russell. 1978. Limiting Oil Imports: An Economic History and Analysis.
Resources for the Future, Washington, D.C.
Harris,!. 1991. Death in the Marsh. Island Press, Washington, D.C., 1991.
Hermanutz, R.O.; Allen, K.N.; Roush, T.H.; Hedtke, S.F. 1992. "Effects of Elevated
Selenium Concentrations on Bluegills (Lepomis macrochirus) in Outdoor Experimental
Streams." Environ.-Toxicol.-Chem. Vol. 11, No. 2, pp. 217-224.
Hoffman, D.J.; Ohlendorf, H.M.; Aldrich, T.W. 1988. "Selenium Teratogenesis in Natural
Populations of Aquatic Birds in Central California." Arch.-Environ.-Contam.-Toxicol, Vol. 17,
No. 4, pp. 519-525.
Howard, P.H.; Boethling, R.S.; Jarvis, W.F.; Meyland, W.M.; Michelenko, E.M. 1991.
Handbook of Environmental Degradation Rates.
Langer, B.S. 1983. "Wastewater Reuse and Recycle in Petroleum Refineries." Chemical
Engineering Progress, May 1983, pp. 67-76.
Lyman, W.J.; Reehl, W.F.; Rosenblatt, D.H. 1982. Handbook of Chemical Property
Estimation Methods: Envionmental Behavior of Organic Compounds. McGraw-Hill, New
York, 1982.
Marcogliese, D.J.; Esch, G.W.; Dimock, R.V., Jr. 1992. "Alterations in Zooplankton
Community Structure after Selenium-Induced Replacement of a Fish Community: A Natural
Whole-Lake Experiment." Hydrobiologia, Vol. 242, No. 1, pp. 19-32.
Ohlendorf, H.M.; Hoffman, D.J.; Saiki, M.K.; Aldrich, T.W. 1986. "Embryonic Mortality and
Abnormalities of Aquatic Birds: Apparent Impacts of Selenium from Irrigation Drainwater."
Sci.-Total-Environ. Vol. 52, No. 1-2, pp. 49-63.
Ohlendorf, H.M.; Hothem, R.L.; Aldrich, T.W.; Krynitsky, A.J. 1987. "Selenium
Contamination of the Grasslands, A Major California Waterfowl Area." Sci.-Total-Environ. Vol.
66, pp. 169-183.
Ohlendorf, H.M.; Hothem, R.L.; Welsh, D. 1989. "Nest Success, Cause-Specific Nest Failure,
and Hatchability of Aquatic Birds at Selenium-Contaminated Kesterson Reservoir and a
Reference Site." Condor. VI. 91, No. 4, pp. 787-796.
120
-------
Piccini, R.A., ed. 1992. Petroleum and Public Policy: The Post-World War II Experience.
American Petroleum Institute, Washington, D.C. API Discussion Paper 071.
Thrash, L. A. 1991. " Survey of Operating Refineries Worldwide." Oil and Gas Journal,
December 23, 1991.
U.S. Department of Commerce, Bureau of the Census. Current Industrial Reports, Pollution
Abatement Costs and Expenditures. Various years.
U.S. Department of Energy. Performance Profile of Major Energy Producers. 1991.
. Energy Information Administration. Annual Energy Review. Various years.
. Energy Information Administration. Petroleum Supply Annual. Various years.
U.S. Environmental Protection Agency. Office of Water. 1974a. "Effluent Guidelines and
Standards for the Petroleum Refining Point Source Category." Final Rule. Code of Federal
Regulations (CFR), Title 40, Part 419; Federal Register, Vol. 39, p. 16560, May 9, 1974.
. Office of Water. 1974b. Development Document for Effluent Limitations Guidelines
and New Source Performance Standards for Petroleum Refining. EP A-440/1 -74-014a., April
1974.
. Office of Water. 1977. "Effluent Guidelines and Standards for the Petroleum
Refining Point Source Category." Final Rule. Code of Federal Regulations (CFR), Title 40,
Part 419; Federal Register, Vol. 42, p. 15684, March 23, 1977.
. Office of Water. 1980. Ambient Water Quality Criteria for Copper.
EPA/440/5-80-036, October 1980.
. Office of Water. 1982a. Development Document for Effluent Limitations
Guidelines and Standards for the Petroleum Refining Point Source Category.
EPA/440/1-82/014, October 1982.
. Office of Water. 1982b. "Effluent Guidelines and Standards for the Petroleum
Refining Point Source Category." Final Rule. Code of Federal Regulations (CFR), Title 40,
Part 419; Federal Register, Vol. 47, p. 46434, October 18, 1982.
. Office of Water. 1986. Report to Congress on the Discharge of Hazardous Waste
to Publicly Owned Treatment Works. ("Domestic Sewage Study") EPA/530-SW-86-004,
February 1986.
121
-------
. Office of Water. 1987a. Guidance Manual for Preventing Interference at POTWs.
EPA/833/B-87-201, September 1987.
. Office of Water. 1987b. An Overview of Sediment Quality in the United States.
EPA/905/9-88/002, June 1987.
. Office of Research and Development. 1989. Effects of selenium on mallard duck
reproduction and immune function. Whiteley, P.L.; Yuill, T.M.; Fairbrother, A. EPA
Environmental Research Laboratory, Corvallis, OR. EPA/600/3-89/078.
. Office of Water and Office of Emergency and Remedial Response. 1990.
CERCLA Site Discharges to POTWs: Guidance Manual. EPA/540/G-90/005, August 1990.
. Office of Water. 1991. Technical Support Document for Water Quality-Based
Toxics Control. EPA/505/2-90-001, March 1991.
. Office of Water. 1992a. Toxic Weighting Factors for Offshore Oil and Gas
Extraction Industry Pollutants. Rulemaking Record for Final Effluent Guidelines for Offshore
Oil and Gas Extraction, October 15, 1992.
. Office of Water. 1992b. Introduction To Water Quality-Based Toxics Control For
The NPDES Program. EPA/83 l/S-92/002.
. Office of Water. 1993. "Standards for the Use or Disposal of Sewage Sludge."
Final Rule. Code of Federal Regulations (CFR), Title 40, Part 503; Federal Register Vol. 58, p.
9387, February 9, 1993.
. Office of Water and U.S. Department of the Army, Corps of Engineers. 1994a.
Evaluation of Dredged Material Proposed for Discharge in Waters of the United States.
Testing Manual, draft ("Inland Testing Manual"). EPA/823-B/94-002, June 1994.
. Office of Water. 1995. National Sediment Contaminant Point Source Inventory:
Analysis of Release Data for 1992. Staff working draft, Standards and Applied Science
Division, May 1995.
Vogely, William, ed. 1985. Economics of the Mineral Industries. 4th Ed. AJJVIE, New York,
1985.
122
-------
Index
activated carbon, 17, 24, 39, 55, 69, 101
activated sludge, 3, 23
API separator, 3, 15,21,22
asphalt, 4, 5, 7
baffle plate separator, 3
best available technology (BAT), 1, 3, 14, 15,
27, 30, 32, 33, 34, 38, 79
biological treatment, 3, 15, 23, 24, 26, 39
boiler condensate, 18
best practicable technology (BPT), 1, 2, 3, 4, 15,
26, 27, 30, 32, 33, 34, 36, 38, 79, 82, 83
Clean Air Act, vii, 11, 35, 116
Clean Water Act, 1, 117, 118
cooling tower, 3, 15, 19, 21, 22, 118
cooling water, 3, 15, 19, 20, 21, 32, 54
cracking, 2, 4, 5, 6, 7, 10, 18
development document, 79
dioxin, 46, 59, 65, 103
dissolved air flotation (DAF), 15, 22, 39
distillation, 5, 6, 10, 16, 104, 107, 108
discharge monitoring report (DMR), 13, 63
ecological impact, 101
employment, 84
end-of-pipe treatment, 3, 15, 21, 23
equalization, 3, 21, 22, 54
exposure, 90, 91, 95, 97, 101
filters, 3, 23
flow reduction, 3,30
Financial Reporting System (FRS) companies,
109
health effects, 95, 97
hydrotreating, 6, 7
induced air flotation (IAF), 22
impacts, 19, 21, 58, 93, 95, 102
imports, 112
indirect discharges, 58
industry trends, 13, 15
in-plant controls, 15
integrated, 100, 109
lagoons, 3
Los Angeles County, 14, 39, 42, 89
lube, 2, 4, 7
National Emissions Standards for Hazardous
Air Pollutants (NESHAP), 11
National Pollutant Discharge Elimination
System (NPDES), 13, 25, 63, 69, 70, 74,
122
neutralization, 16, 17, 19, 55
non-conventional pollutants, 58
once-through cooling, 3, 20
Ontario, 14, 28, 48, 54, 55, 56, 57
oxidation ponds, 3
Permit Compliance System (PCS), 13, 26, 28,
58, 63, 64, 65, 69, 74, 75, 77, 78, 79, 80,
81, 88, 91, 93, 95, 99, 100, 103, 104
petrochemicals, 2, 8
polishing, 3, 7, 15, 21, 23, 24, 26
pollutant loadings, 58, 69
pollutant loads, 100
precipitation, 22
pretreatment, 4, 14, 18, 39, 55
Pretreatment Standards for Existing Sources
(PSES), 1, 4, 39
process operations, 5, 15, 30
regeneration, 14, 17, 46, 47, 51, 53, 54
Resource Conservation and Recovery Act,
(RCRA) 35
risk, 90, 91, 97, 98
sediment contamination, 102, 103
solvent refining, 7
sour water strippers, 2, 15, 16, 17
steam stripping, 6, 16
subcategories, 1, 2
topping, 2
Toxic Release Inventory (TRI), 58, 64, 83, 85,
86, 87, 88, 91, 93, 95, 99, 100, 103
toxic weighting factors, 97
wastewater investment, 119
water use, 13, 18,30,34, 54
123
-------
whole effluent toxicity testing (WET), 100,101, 102
124
------- |