EPA-450/3-76-038a
November 1976
BACKGROUND INFORMATION
ON HYDROCARBON
EMISSIONS FROM MARINE
TERMINAL OPERATIONS
VOLUME I: DISCUSSION
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
-------
BACKGROUND INFORMATION
ON HYDROCARBON EMISSIONS
FROM MARINE TERMINAL OPERATIONS
VOLUME I: DISCUSSION
C.E. Burklin. J.D. Colley. and M.L. Owen
Radian Corporation
8500 Shoal Creek Blvd.
P.O. Box 9948
Austin, Texas 78766
Contract No. 68-02-1319
Task No. 56
EPA Project Officer: William L. Polglase
Prepared for
ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
November 1976
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are
available free of charge to Federal employees, current contractors and
grantees, and nonprofit organizations - in limited quantities - from the
Library Services Office (MD-35) , Research Triangle Park, North Carolina
27711; or, for a fee, from the National Technical Information Service,
5285 Port Royal Road, Springfield, Virginia 22161.
This report was furnished to the Environmental Protection Agency by
Radian Corporation, 8500 Shoal Creek Blvd. , P.O. Box 9948, Austin,
Texas 78766, in fulfillment of Contract No. 68-02-1319, Task No. 56.
The contents of this report are reproduced herein as received from
Radian Corporation. The opinions, findings, and conclusions expressed
are those of the author and not necessarily those of the Environmental
Protection Agency. Mention of company or product names is not to be
considered as an endorsement by the Environmental Protection Agency.
Publication No. EPA-450/3-76-038a
11
-------
TABLE OF CONTENTS
VOLUME I
Page
1.0 INTRODUCTION 1
1.1 Ob j ectives 1
1. 2 Approach 1
1. 3 Report Contents , 3
2. 0 EXECUTIVE SUMMARY 5
2.1 Results 5
2.1.1 Background Information on Marine
Terminals in the Houston-Galveston
AQCR 5
2.1.2 Background Information on Marine
Terminals in the Los Angeles AQCR 9
2.1.3 Emissions 10
2.1.4 Emission Control Technology 12
2.1.5 Economics of Emission Control 13
2. 2 Conclusions '. 15
2 . 3 Recommendations 16
3.0 BACKGROUND INFORMATION ON MARINE TERMINALS 18
3.1 Relative Quantities of Crude Oil and Gasoline
Transported by Marine Terminals in the United
States 18
3.2 Marine Terminals Transferring Crude Oil and
Gasoline in the Houston-Galveston Instrastate
AQCR 20
3.2.1. Exxon Baytown Refinery Marine Terminal 24
3.2.1.1 Gasoline Loading System 24
3.2.1.2 Crude Oil Loading/Unloading
System 31
iii
-------
TABLE OF CONTENTS (Continued)
Page
3.2.2 Shell Deer Park Refinery Marine
Terminal 34
3.2.2.1 Gasoline Loading System 34
3.2.2.2 Crude Oil Unloading System... 36
3.2.3 AMOCO Texas City Refinery Marine
Terminal ,. 38
3.2.3.1 Gasoline Loading System 38
3.2.3.2 Crude Oil Unloading System... 40
3.2.4 ARCO Houston Refinery Marine Terminal. 43
3.2.4.1 Gasoline Loading System 43
3.2.4.2 Crude Oil Unloading System... 43
3.2.5 Texas City Refining Texas City Refinery
Marine Terminal 46
3.2.5.1 Gasoline Loading System 46
3.2.5.2 Crude Oil Unloading System... 49
3.2.6 Crown Central Houston Refinery Marine
Terminal 49
3.2.6.1 Crude Oil Unloading System... 49
3.2.7 Charter Oil Houston Refinery Marine
Terminal 51
3.2.7.1 Gasoline Unloading System.... 51
3.2.7.2 Crude Oil Unloading System... 51
3.2.8 Marathon Texas City Refinery Marine
Terminal 51
3.2.8.1 Gasoline Loading System 53
3.2.8.2 Crude Oil Unloading System... 53
3.3 Shipside Equipment and Transfer Procedures... 53
3.3.1 Crude Oil and Gasoline Loading of Ships 53
3.3.2 Crude Oil and Gasoline Loading Onto
Barges 58
IV
-------
TABLE OF CONTENTS (Continued)
Page
3.3.3 Crude Oil and Gasoline Unloading
from Tankers . .' 59
3.4 Quantities of Crude Oil and Gasoline Trans-
ferred in the Houston-Galveston Area 63
3.5 Projected Quantities of Crude Oil and Gasoline
Transferred in the Houston-Galveston Area
Through 1985 ; 67
3.6 Cruise History Information for Ships and
Barges Which Transferred Crude Oil or Gasoline
in the Houston-Galveston Area During 1975.... 73
3.6.1 Effects of Cruise History on Hydro-
carbon Emissions from Marine Loading
of Gasoline and Crude Oil 74
3.6.2 Types of Marine Vessels Used in Trans-
ferring Crude Oil and Gasoline in the
Eouston-Galveston Area 76
3.6.2.1 Marine Tankers . 76
3.6.2.2 Intercoastal Barge 77
3.6.2.3 Ocean Barge 77
3.6.3 Vessels Servicing Houston-Galveston
Marine Terminals 77
3.6.4 Hydrocarbon Emissions From a Gasoline
Tanker Cruise 80
3.6.5 Analysis of Tank Arrival Conditions for
Vessels Loading Gasoline and Crude Oil
in the Houston-Galveston Area 82
3.7 Marine Terminals Transferring Crude Oil and
Gasoline In the Metropolitan Los Angeles Area 84
v
-------
TABLE OF CONTENTS (Continued)
Page
3.7.1 Background Information on Marine
Terminals Transferring Crude Oil
and Gasoline in the Southern
California Area 84
3.7.1.1 Shoreside Equipment and Trans-
fer Procedures-Gasoline 36
3.7.1.2 Shoreside Equipment and Trans-
fer Procedures-Crude Oil 87
3.7.2 Shipside Equipment and Transfer Pro-
cedures for the Los Angeles AQCR 89
3.7.3 Quantities of Crude Oil and Gasoline
Transferred in the Los Angeles AQCR.. ... 90
3.7.4 Projected Unloading of Alaskan Crude
Oil in the Los Angeles AQCR 91
3.7.4.1 Port Site for Unloading Alaskan
Crude Oil in the LA AQCR 94
3.7.4.2 Types and Sizes of Tankers De-
livering Alaskan Crude Oil
from Valdez to the Los Angeles
AQCR 94
3.7.4.3 Projected Quantities of Alaskan
Crude Oil to be Unloaded in the
Los Angeles AQCR 97
3.7.4.4 Characteristics of Alaskan
Crude Oil 97
3.7.5 Similarities and Differences in Marine
Terminals Located in the Los Angeles
AQCR and the Houston-Calveston AQCR.... 100
3.7.5.1 Los Angeles County 100
3.7.5.2 Ventura County 102
VI
-------
TABLE OF CONTENTS (Continued)
Page
3.7.5.3 Santa Barbara County 103
4. 0 MARINE TERMINAL EMISSIONS «. . 105
4.1 Emission Characteristics 105
4.1.1 Source and Mechanism 105
4.1.2 Effects of Loading Rate 110
4.1.3 Effects of TVP 113
4.1.4 Effects of Cruise History 114
4.1.5 Composite Vapor Profiles 115
4.1.6 Chemical and Physical Properties 118
4. 2 Source Testing Results 126
4.2.1 Industry Testing 126
4.2.2 Radian Testing 131 '
4.2.3 Conclus ions 132
5 . 0 EMISSION CONTROL TECHNOLOGY 133
5 .1 Vapor Control Unit 134
5.1.1 Refrigeration 134
5.1.2 Absorption 139
5.1.3 Incineration 143
5.1.4 Alternative Vapor Recovery Units 147
5.1.5 Vapor Control Unit Installation 143
5.1.6 Inert ing 148 ~
5.1.7 Composite Vapor Profile 149
5.2 Shoreside Vapor Collection 151
5.2.1 Design 151
5.2.2 Efficiency 155
5.2.3 Cost 155-
5.2.4 Safety ' 155
5.2.5 State of Development 156
vzi
-------
TABLE OF CONTENTS (Continued)
Pa*
5.3 Shipside Vapor Collection 156
5.3.1 Design 157
5.3.2 Efficiency 159
5.3.3 Cost 159
5.3.4 Safety ; 160
5.3.5 Salient Considerations 160
5 .4 Alternative Control Strategies 161
5.4.1 Ullage Hatch Condensers 161
5.4.2 Ship Boiler Incineration 161
5.4.3 Foam 162
5.4.4 Product Cooling 162
5.4.5 Controlled Loading 162
6 . 0 ECONOMICS OF EMISSION CONTROLS 164
6.1 Establishment of Cases 164
6. 2 Methodology 170
6.3 Results 172
6.3.1 Base Case 172
6.3.2 Sensitivity to Cost Inputs 174
6.3.3 Sensitivity to Unit Size 176
6.3.4 Sensitivity to Vessel Mix 177
7. 0 TEST PLAN DEVELOPMENT 179
7.1 Obj ective 179
7. 2 Approach 181
7.2.1 Results Format 181
7.2.2 Parameters 182
7.2.3 Required Level of Sampling 184
7.2.4 Test Program - Instrumentation 185
7.2.5 Hydrocarbon Analysis 186
Vlll
-------
TABLE OF CONTENTS (Continued)
Page
7. 3 Sampling Procedure 192
7.3.1 Test Measurements 192
7.3.1.1 Vented Vapor Concentration
Profile 192
7.3.2 Recorded Information - Data'Sheets.... 197
BIBLIOGRAPHY
CONVERSION FACTORS
IX
-------
TABLE OF CONTENTS (Continued)
VOLUME II
Page
APPENDIX I
VESSELS TRANSPORTING CRUDE OIL AND GASOLINE
IN THE HOUSTON-GALVESTON AREA 1-1
APPENDIX II VAPOR CONTROL SYSTEM COST DATA.
II-l
APPENDIX III RESULTS FROM INDUSTRY TEST PROGRAMS .... III-l
APPENDIX IV INDUSTRY TEST DATA IV-1
APPENDIX V
RADIAN EMISSION TESTING RESULTS
V-l
APPENDIX VI RADIAN EMISSION TEST DATA AND TRIP REPORTS. VI-1
APPENDIX VII INDEPENDENT ANALYSIS OF VAPOR RECOVERY
SYSTEM COSTS
VII-1
-------
LIST OF TABLES
Pa^e
3.2-1 Marine Terminals Transferring Crude Oil or
Gasoline in the Houston-Calveston AQCR ..... 22
3.2-2 Gasoline Pumping System Lineups 26
3.2-3 Lines and Pumps for Marine Loading of Gasoline -
AMOCO Texas City , 38
3.2-4 Maximum Gasoline Loading Rate at Texas City
Refining"s Marine Docks 46
3.4-1 Quantity of Gasoline Loaded at Marine Terminals
in the Houston-Calves ton Area 64.
3.4-2 Reid Vapor Pressure of Gasolines Loaded at
Marine Terminals in the Houston-Calveston Area . 65
3.4-3 Quantity of Crude Oil Loaded at Marine Terminals
in the Houston-Calves ton Area 66
3.4-4 Quantity of Crude Oil Unloaded at Marine
Terminals in the Houston-Calves ton Area 68
3.4-5 Average RVP of Crude Oil Unloaded at Marine
Terminals in the Houston-Calves ton Area 69
3.5-1 Projected Quantities of Gasoline to be Loaded
at Marine Terminals in the Houston-Calves ton
Area Through 1985 70
-------
LIST OF TABLES (Continued)
3.5-2 Projected Quantities of Crude Oil Loaded at
Marine Terminals in the Houston-Calveston
Area Through 1985 71
3.5-3 Projected Quantities of Crude Oil Unloaded at
Marine Terminals in the Houston-Calves ton
Area Through 1985 72
3.6-1 Effect of Ship Cruise History on Arrival Hydro-
carbon Concentration Prior to Gasoline Loading . 75
3.6-2 Emission Factors for Gasoline and Crude Oil .
Loading by Tank Arrival Condition 33
3.7-1 Marine Terminals Transferring Crude Oil or
Gasoline in the Metropolitan Los Angeles AQCR. . 85
3.7-2 Quantity of Gasoline Loaded at Marine Terminals
in the Los Angeles AQCR 92
3.7-3 Quantity of Gasoline Unloaded at Marine Terminals
in the Los Angeles AQCR 92
3.7-4 Quantity <->f Crude Oil Loaded at Marine Terminals
in the Los Angeles AQCR 93
3.7-5 Quantity of Crude Oil Unloaded at Marine Terminals
in the Los Angeles AQCR 93
3.7-6 Projected Alaskan Crude Oil Tanker Fleet. ... 96
xii
-------
LIST OF TABLES (Continued)
Page
3.7-8 Composition of Vapor in Equilibrium with
North Slope Crude Oil 99
4.1-1 Chemical Composition of Gasoline Loading Vapors. 123
./
4.1-2 Composition of Vented Vapors, Vol. 70 Crude Oil
Loading Test, 5-8-76, Avila Terminal, Tanker:
Lion of California 124
4.1-3 Chemical Composition of Aviation Gasoline
Vapor ' 125
4.2-1 Summary of Petroleum Industry Testing Programs
on Marine Loading Emissions 127
4.2-2 Summary of Results Hydrocarbon Emissions from
Marine Loading Motor Gasoline 129
4.2-3 Summary of Results Hydrocarbon Emissions from
Marine Loading Aviation Gasoline 13Q
6.1-1 Statistics on the Proposed Houston-Calves ton
Vapor Recovery Systems 165
6.1-2 Summary of Cost Data for Marine Terminal Controls 167
6.1-3 Summary of Case Parameters 159
6.2-1 Results of Study on Vapor Recovery Economics . . 171
-------
LIST OF FIGURES
3.1-1 Transportation of Crude Oil, 1974 19
3.1-2 Transport of Gasoline 21
3.2-1 Location of Marine Terminals in the Houston-
Galveston AO.CR 23
3.2-2 Exxon Baytown Refinery Marine Terminal 25
3.2-3 Gasoline Loading Lines To Docks 1 and 2 27
3.2-4 Gasoline Line Manifolding at Dock 1 23
3.2-5 Gasoline Line Manifolding at Dock 2 29
3.2-6 Crude Oil Lines To/From Docks 2 and 5 32
2.2-7 Crude Line Manifolding at Docks 2 and 5 33
3.2-8 Shell Deer Park Manufacturing Complex Marine
Terminal 35
3.2-9 Gasoline Loading Lines to Shell's Marine Docks 37
3.2-10 AMOCO Texas City Refinery Marine Terminal 39
3.2-11 Gasoline Lines to AMOCO Texas City Marine
Terminal 41
xiv
-------
LIST OF FIGURES (Continued)
Page
3.2-12 Crude Oil Unloading Lines - AMOCO Texas City .. 42
3.2-13 ARCO Houston Refinery Marine Terminal 44
3.2-14 Gasoline Loading Lines To Docks A and B ARCO
Houston Marine Terminal ". 45
3.2-15 Crude Oil Loading Lines for ARCO Houston
47
3.2-16
3.2-17
3.2-18
3.3-1
3.3-2
3.3-3
3.3-4
3.7-1
4.1-1
4.1-2
Texas City Refining Marine Terminal
Crown Central Houston Refinery Marine Terminal.
Charter Oil Houston Refinery Marine Terminal . .
Tank Capacities and Manifold Arrangement of the
S . S . "Pasadena"
Grade A Cargo Tank Vent System
Single Skin Tank Barge
Grade B Cargo Tank Vent System
Offshore Terminal
Example Profile of Gasoline Loading Emissions..
Example Profile of Gasoline Ballasting Emissions
48
50
52
56
57
60
61
88
107
109
XV
-------
LIST OF FIGURES (Continued)
4.1-3 Effect of Initial Fill Rate on Vapor Blanket
Profile Ill
4.1-4 Hydrocarbon Profile Prior to Ballasting an
Empty Tank 115
4.1-5 Hydrocarbon Profile of a Ballasted Tank 115
4.1-6 Hydrocarbon Profile of an Empty Tank After
Ballast Discharge .- 115
4.1-7 Example Composite Vapor Profile for Loading
Sequential ' 119
4.1-8 Example Composite Vapor Profile for Simultaneous
Loading 119
4.1-9 Vapor Pressures of Crude Oil 121
4.1-10 Vapor Pressures of Gasolines and Finished
Petroleum Products 122
5.1-1 Refrigeration Vapor Recovery Unit 135
5.1-2 Absorption Vapor Recovery Units 140
5.1-3 Incineration Vapor Control Unit 144
5.1-4 Vapor Profiles of the Feed and Product of a
Vapor Recovery Unit 150
xvi
-------
LIST OF FIGURES (Continued)
5..2-1 Typical Vapor Collection System 153
5.3-1 Shipside Vapor Collection System • 158
7.3-1 Location of Sample Probe 193
7.3-2 Sample Points Relative to True Ullage (Concen-
tration) and Vapor Profile (Composition) 194
xvii
-------
ACKNOWLEDGEMENTS
The authors wish to acknowledge the assistance of those
individuals from various companies and organizations which made
this report complete. The program was conducted under the guidance
of William Polglase and Oscar Cabra, the Co-Project Officers.
Valuable information and guidance were received from other person-
nel within the EPA offices in Dallas and Research Triangle Park.
Special thanks also goes to those oil companies in the Houston-
Calves ton area who were responsible for providing Radian with
pertinent first-hand information. These companies were: Exxon
Company, USA; AMOCO Oil Company; Atlantic Richfield Company;
Shell Oil Company; Texas City Refining, Inc.; Marathon Oil
Company; and Crown Central Petroleum Corporation. Because of
the lengthy list of individuals contacted from these companies,
names cannot be presented. Other organizations and companies
to whom we are indebted are: the United States Coast Guard, the
American Petroleum Institute, Edwards Engineering Company, and
Texas City Terminal Railway Company.
xviii
-------
1.0 . INTRODUCTION
The loading and unloading of volatile hydrocarbon
liquids at marine terminals is known to be a source of hydro-
carbon emissions. However, until recently very little data have
been available on the sources, characteristics, and generation
rate of these emissions. This report presents the results of
a detailed study for EPA on hydrocarbon emissions from marine
terminal operations.
1.1 Objectives
The objectives of this program are to report emission
factors and to develop background information necessary for the
accurate assessment of hydrocarbon emissions from ship and barge
loading and unloading of gasoline and crude oil. Data gather-
ing focussed on the Metropolitan Houston-Calveston Intrastate
AQCR. Also, sufficient information was assembled to project
hydrocarbon emissions in the Metropolitan Los Angeles Intra-
state AQCR generated by the handling of gasoline and crudes,
including Alaskan North Slope crude.
1.2 Approach
The program objectives were accomplished through the
completion of the following elements:
1) Identification and' description of marine gaso-
line and crude oil loading and unloading
terminals in the Houston-Galveston area, based
upon industry data.
-------
2) Collection of statistical data from the petroleum
industry regarding daily and monthly loading
of gasoline and crude oil in the Houston-Calveston
area over a 12-month period.
3) Collection of cruisenhistory statistics from the
petroleum industry for ships and barges that have
loaded gasoline or unloaded gasoline and crude
oil in the Houston-Calveston area during the
past 12 months.
4) Review and evaluation of both foreign and domes-
tic information available which describes and/or
quantifies hydrocarbon emissions from ship and
barge loading, unloading, and transport.
5) Completion of limited Radian sampling studies
of hydrocarbon emissions from ship and barge
loading and unloading operations for the pur-
pose of verifying reported emission factors.
6) Review of industry available data on control
technology and associated costs applicable
to the control of hydrocarbon vapors generated
by the loading and unloading of petroleum pro-
ducts and crude oil into marine vessels.
7) Preparation of a test program based upon infor-
mation gathered in the course of this study to
statistically validate emission factors for
the loading and unloading of gasoline and crude
oil for ships and barges.
-2-
-------
Radian received statistics and information on the
marine terminal industry through contacting petroleum com-
panies involved in major marine terminal operations in the
Houston-Calveston and Los Angeles areas, and through review of
petroleum industry files maintained.by EPA. Marine terminal
site visits were also conducted by- Radian in the Houston-Calves -
ton area and in the Los Angeles area.
The original scope of work for this project did not
include addressing the safety problems associated with the ap-
plication of vapor recovery systems to gasoline loading of ships
and barges. However, during the course of the project it be-
came apparent that safety considerations are a very important
part of a complete discussion on emission control technology
for marine terminals. Because of this importance, Radian in-
cluded safety considerations in the marine terminal controls
discussion. This discussion is not intended to be a comprehen-
sive analysis of this subject, since the scope of the project
limited the detail of the analysis. Instead, the discussions
of safety include an analysis of the types of safety problems
encountered in the different control systems.
1.3 Report Contents
The results, conclusions and recommendations developed
from this study are presented in Section 2, the executive sum-
mary. Section 3 presents background information on marine
terminal facilities and operations. Included in Section 3 are
descriptions of marine terminals in the Houston-Galveston area,
statistics on the movement of gasoline and crude oil in both
the Houston-Galveston area and the Los Angeles area, and cruise
history statistics for ships and barges. Section 4 characterizes
hydrocarbon emissions from marine terminal operations and pre-
sents the results of emission testing conducted by the petroleum
-3-
-------
industry and Radian Corporation. Section 5 reviews hydrocarbon
emission control technology available for marine terminal opera-
tions from the aspects of principle of operation, efficiency,
safety, state of development, and cost. The economics associa-
ted with the implementation of marine terminal emission controls
are presented in Section 6. Section 7 develops a test plan for
establishing emission data in those areas where emission data
is inadequate.
-4-
-------
2.0 EXECUTIVE SUMMARY
This report is the first comprehensive examination of
marine terminals transferring crude oil and gasoline in the
Houston-Calveston AQCR. This section summarizes the results
from the specific 'areas examined, presents conclusions which
were drawn from these results, and provides a list of recommen-
dations for further work.
2.1 Results
The results of this report are summarized below.
They are organized according to the sections of the report
which discuss that specific area:
Background Information on Marine Terminals in
the Houston-Galveston AQCR
Background Information on Marine Terminals in the
Los Angeles AQCR
Emissions
Emission Control Technology
Economics of Emission Control
2.1.1 Background Information on Marine Terminals in the
Houston-Calveston AQCR
The results included in this section are based on in-
formation obtained from the petroleum industry and from site
visits made to marine terminals in the Houston-Calves ton area.
-5-
-------
Important facts included in this section are condensed below;
Seven marine terminals, all of which are
associated with refinery facilities,
accounted for virtually all of the gasoline
loaded onto ships and barges in 1975.
Approximately 82.5 million barrels of
gasoline (motor and aviation) were loaded
at seven marine terminals in the Houston-
Galveston area in 1975. The average RVP
ranged from a high of 13-14 psi in the
winter to a low of 9-10 psi in the summer.
Projections show that approximately the
same quantity of gasoline will be loaded
annually .in the Houston-Calveston area
through 1985. Therefore, no significant
growth in gasoline loading at marine
terminals in the area is foreseen through
1985.
Approximately 75 percent of the gasoline
transferred at marine terminals in the
area was loaded onto ships in 1975.
Eight marine terminals, each associated
with refinery complexes, accounted for
virtually all of the crude oil unloaded
from ships and barges in the Houston-
Galveston area in 1975.
-o-
-------
Available information indicates that approxi-
mately 160 million barrels of crude oil were
transported into seven of the marine terminals
which imported crude oil in the area in 1975.
«
Projections for the quantity of crude oil
imported by tanker into the Houston-
Galveston refinery terminals show a con-
tinuing increase through 1985. Should
the proposed SEADOCK offshore terminal be
completed in 1980 according to plan, the
quantity of crude unloaded in the harbor
will drop from approximately 360 million
barrels in 1979 to 240 million barrels in
1980 and to 190 million barrels in 1985.
Crude oil was loaded onto ships at one marine
terminal in the area. A total of 7.5 million
barrels were loaded there in 1975 most of
which was shipped to East Coast refineries.
Maximum gasoline loading rates at any one
dock at marine terminals in the Houston-
Galveston area ranged from 4,500 bbls/hr
to around 50,000 bbls/hr. Typical rates
were from 5,000-10,000 bbls/hr.
All vessels loading gasoline in the area keep
their ullage hatches open for visual inspection
of the cargo liquid level. Also, P/V (pressure/
vent) valves are manually opened during loading.
Crude oil shops also follow this practice when
they take on ballast after discharging their
cargo.
-7-
-------
Crude oil ships may ballast from 20 to 40 percent
of their cargo capacity before leaving the dock
where they unloaded. This operation may be com-
pleted at dock or continue as the vessel leaves
port, at the discretion of the ship's officers.
Intercoastal barges do not take on ballast;
however, the larger oceangoing barges do.
For Grade A tankers the vapors displaced from
the cargo tanks during loading and ballasting
are vented from two places - the ullage hatch
and the vent headers located 40 to 50 feet above
dock level. For Grade B tankers, the vapors
are vented from either the ullage hatch or the
P/V valve, located a few feet above deck level.
Ships arriving at marine terminals in the area
may have average hydrocarbon concentrations in
their cargo tanks varying from less than one
percent to greater than twenty percent (less
than 2 percent to greater than 40 percent of
•saturation). The ship's cruise history on
the previous voyage is the reason for this
difference.
Barges arriving to load gasoline usually have
not been cleaned during their prior trip.
They typically have high arrival vapor concen-
trations in their tanks.
-8-
-------
The parameters affecting the arrival vapor con-
centrations for ship's tanks are: 1) previous
prior cargo, 2) the type and extent of cleaning
performed during the return voyage, and 3) the
fraction of the tank ballasted.
Investigations indicate that for those tankers
loading gasoline 45 percent of ship's cargo
tanks arrive cleaned, 10 percent arrive bal-
lasted, and 45 percent arrive empty and undis-
turbed.
2.1.2 Background Information on Marine Terminals in the
Los Angeles AQCR
The results presented in this section are based on
information obtained from regulatory agency contacts, literature,
and onsite visits of marine terminals in the Los Angeles AQCR.
Results from the Southern California study are included in the
following material:
Eight marine terminals loaded approximately 9
million barrels of gasoline into ships in 1975.
Nine marine terminals unloaded approximately
15.7 million barrels of gasoline from ships
in 1975.
Approximately 18 million barrels of crude oil
were loaded into ships at ten terminals in 1975.
Slightly over 177 million barrels of crude oil
were unloaded from ships at twelve marine ter-
minals in 1975.
-9-
-------
Virtually no crude oil or gasoline is trans-
ported by barge.
Offshore as well as onshore marine terminals
are used for loading and unloading crude oil.
No cargo tank inerting is done in the LA AQCR,
and currently, few, if any, ships service the
area with the capability to inert.
A proposed project by SOHIO could unload
approximately 500,000 bpd of Alaskan crude
oil from tankers at Long Beach in the LA
AQCR by 1978.
The proposed Long Beach terminal could handle
tankers of up to 165,000 DWT.
Half of the projected Alaskan crude oil tanker
fleet of 26 ships will have inert gas systems
onboard.
A proposed project offshore of Santa Barbara
County will produce up to 120,000 barrels per
day of crude oil which will be loaded onto
tankers at an offshore terminal there.
2.1.3 Emissions
j
The results included in this section are based on data
from the petroleum industry and Radian verification and engineer-
ing analysis of this reported information. Some of the impor-
tant results from this analysis are included here:
-10-
I
-------
Hydrocarbons are emitted from marine vessel
cargo tanks during loading operations and
during ballasting operations following cargo
unloading operations.
Test data indicate that if the arrival hydro-
carbon concentration of a cargo tank to be
loaded with gasoline is in the explosive range,
it will remain in this range for about 807« of
the loading time. Gasoline ship cargo tanks
which have been ballasted or left uncleaned
on the return voyage may have arrival hydro-
carbon concentrations in the explosive range.
Explosive arrival concentrations occur less
often in those tanks which were cleaned on
the" return voyage. The above holds true for
the loading of volatile crude oils into ship
tanks also. However, loading gasoline and
volatile crudes onto intercoastal barges dis-
places vapors that are usually above the ex-
plosive range.
Industry data indicate that hydrocarbon
emissions (lb/103 gal) from loading motor
gasoline are approximately as follows:
Barges
Ships
Cleaned
Ballasted
Range
0-2.3
0.4-3
1.0
1.6
Ocean
Barges
Range
0-3
0.5-3
Avg
2.1
Range . Avg
UA* 1.2
NA**
Uncleaned 0.4-4 2.4 - 0.5-5 3.3 1.4-9 4.0
The results of the Radian testing substantiated
most of the above data.
* Unavailable
** Not Applicable
-11-
-------
r
Industry data indicate that hydrocarbon emissions
(lb/103 gal) from loading aviation gasoline are
approximately the following:
Cleaned Ships 0.5
Uncleaned Ships 2.4
Average Ships 1.3
Average Barge 4.3
Radian preliminary test results indicate
that hydrocarbon emissions from ballasting
crude tankers are in the range of 1 to 2
pounds per thousand gallons ballasted.
Parameters that lower emission factors
include low initial fill rates, low vapor
pressures for either the cargo being loaded
or the previous cargo, and tank cleaning or
ballasting prior to loading.
2.1.4 Emission Control Technology
Information provided by control equipment vendors and
petroleum companies was used in the emission control section.
Some of the results based on this information are included here:
The principal vapor control systems being con-
sidered in the Houston-Galveston area are based
on conveying hydrocarbon vapors generated onboard
the vessels to shoreside vapor control units for
recovery as liquid product or for disposal by
incineration.
-12-
-------
Proposed vapor control systems are projected to
have the capability of removing non-tnethane hydro-
carbons from gasoline loading vapors to a con-
centration lower than 5 volume percent gasoline
vapors. If high methane concentrations are pre-
sent, as have been observed in some industry tests,
controlled emissions may exceed 5 volume percent
because conventional vapor recovery units are
ineffective on methane vapors.
Several potential safety problems exist with
proposed marine terminal vapor control systems.
Since the safety problems have not been com-
pletely resolved, the costs associated with
installing the necessary safety controls may be
inadequately defined.
2.1.5 Economics of Emission Control
The results presented below in this section were
based upon information provided by the oil industry and control
equipment vendors, and upon information developed in a detailed
Radian cost study:
The Radian cost study projects the initial
capital cost of the shoreside portion of a
vapor control system to be $80 per barrel
per hour of marine terminal loading capacity.
Projections from industry data are that the
initial capital cost of vessel modifications
will be $325,000 per ship and $68,000 per
barge.
-13-
-------
Based on industry data, vapor control unit operat-
ing costs including maintenance and utilities are
projected to be $15 per thousand barrels loaded.
When amortized over a 15 year equipment life at
12% interest, the total annualized cost of an
average vapor control system is projected to be
$33 per bph of marine terminal capacity.
Based on the annualized cost, the typical vapor
control system proposed for the Houston-Calveston
area is projected to exhibit a cost effectiveness
of $2900 per ton of hydrocarbons recovered, and
is projected to exhibit an economic impact of
$0.07 per barrel of gasoline loaded.
As a result of a Radian cost sensitivity analysis,
the cost effectiveness of vapor control systems
is reduced significantly when 1) the required
control system size is proportionally reduced,
2) a marine terminal loads large quantities of
gasoline onto barges, 3) the number of vessels
requiring modification is small, 4) the cost of
a given control system size can be reduced, or
5) the hydrocarbon concentration in the recovered
vapors is high. These parameters can potentially
reduce the cost effectiveness to $2000/ton of
hydrocarbon recovered and reduce economic impact
to $0.06 per barrel of gasoline loaded.
-14-
-------
The sensitivity analysis also indicates that the
cost effectiveness and economic impact are signi-
ficantly higher for small terminals requiring
disproportionately larger and more expensive vapor
recovery systems to handle infrequent but large
tanker loadings. For such situations the cost
effectiveness and economic impact could be twice
as high as the typical vapor recovery system
values. These parameters can potentially increase
the cost effectiveness to $9500/ton of hydro-
carbon recovered and increase economic impact to
$0.22 per barrel of gasoline loaded.
2.2 Conclusions
1. In addition to loading operations, hydrocarbon
emissions are also generated at marine terminals from ballasting
operations subsequent to gasoline and crude oil unloading. In-
transit hydrocarbon emissions occur at sea from diurnal tank
breathing, from tank cleaning activities, and from inerting in
the case of the ships in Alaskan Crude oil tanker fleet which
have inert gas systems.
2. Cruise history, a very important parameter
affecting marine terminal emission factors, has been demon-
strated to vary significantly from one company to another
company. Other parameters such as extent of ballasting, RVP,
loading time, etc., create variations in emission rates and
make the development of accurate emission factors difficult.
Emission factors developed to date do not accurately account
for the effect of all the loading variables. Also, accurate
emission factors do not exist for crude oil loading of for the
ballasting of gasoline or crude oil ships.
-15-
-------
3. Emission control technology for marine loading of
gasoline is a young technology which is faced with a unique set
of problems, one of which is safety. The principles involved
in marine vapor collection and control are well understood.
Applicable control technologies have been developed in other
fields and primarily need to be refined with respect to marine
loading operations.
4. Based on Radian cost evaluation, the projected
cost for installing marine terminal controls in the Houston-
Calves ton area averages $2900 per ton of hydrocarbons controlled.
Wide variations from $2000 to $9500 per ton may exist in the
cost of emission controls due to the cost of necessary safety
equipment and to individual marine terminal characteristics.
5. Due to the emission control problems created by
the presence of methane and to the possibility of regulations
based on non-methane hydrocarbons, it is appropriate to use
methane distinguishing analytical procedures in marine terminal
emission testing programs.
2.3 Recommendations
As a result of this project, several areas have been
identified as needing further work to more completely investi-
gate the hydrocarbon emissions from gasoline ship and barge
loading and crude oil ship ballasting. The following tasks are
recommended to more completely characterize the emissions:
1. A study to obtain detailed information
from ship's logs as to the actual break-
down of tank arrival conditions due to
cruise history for ships servicing the
-16-
-------
Houston-Calveston area and the Los Angeles
area.
2. A sampling program designed to produce.
accurate emission factors for gasoline '
loading into dirty ship tanks, clean barge
tanks, and unclean barge tanks. The program
should also develop factors for crude oil
loading onto ships and barges and crude oil
and gasoline ship ballasting.
3. An investigation of the safety aspects
involved in the application of control tech-
nology to gasoline marine loading emissions.
4. An investigation of the potential impact
on hydrocarbon emissions from marine
terminal operations due to cargo tank
inerting and purging.
-17-
-------
3.0 BACKGROUND INFORMATION ON MARINE TERMINALS
Many refineries are located on navigable waters and
operate marine terminals for transferring crude oil, gasoline
and other products by marine vessel. These facilities may vary
from terminal to terminal, differing in size, material trans-
ferred, type of vessel handled, loading/unloading rate, and
layout.
/
This section of the report presents information which
identifies and describes marine terminals which transfer crude
oil and gasoline in the Houston-Calveston area. Included in
this presentation are descriptions of shoreside and shipside
equipment and operating methods as well as cruise history statis-
tics for the vessels which service these Houston-Galveston
terminals. The majority of this information was obtained from
the owners of the terminals and the rest from EPA Region VI.
A brief section showing the relationship of the marine
terminal to the rest of the gasoline and crude oil transportation
system is given first. It is intended to provide a perspective
of the role that the marine terminal serves in the crude oil
transportation and gasoline marketing systems. •
3.1 Relative Quantities of Crude Oil and Gasoline
Transported by Marine Terminals in the United States
A general flow diagram which shows the relative daily
rates of the crude oil transported to U.S. refineries by pipeline,
tanker/barge, and truck/tank car is presented in Figure 3.1-1.
Based on 1974 data, Figure 3.1-1 shows that almost 28 percent of
the crude oil transported to U.S. refineries is carried at one
time by tankers or barges. Based on a total 1974 U.S. refinery
-18-
-------
DOMESTIC PRODUCTION
8.68
IMPORTS
3.47
PIPELINE-7.54
RAIL AND TANK TRUCK-0.20
BARGE AND TANKER-0.94
PIPELINE-1.02
MARINE TANKER-2.45
REFINERY
STORAGE
FIGURE 3.1-1
TRANSPORTATION OF CRUDE OIL, 1974
(RATES IN MILLIONS OF BARRELS PER DAY)
SOURCE: AM-186
-------
demand of 12.15 million barrels per day, marine tankers and
barges transported approximately 3.4 million barrels of petroleum
daily (Ref. 1).
Figure 3.1-2 is a general flow diagram which shows the
relative daily rates of the gasoline which is transported from
U.S. refineries by pipeline, tanker/barge, and truck/tank car.
The diagram indicates that about nine percent of the gasoline
shipped from U.S. refineries leaves by tanker or barge. .Based
on 1974 data, this amounts to roughly 570,000 barrels per day of
gasoline loaded into tankers and barges (Ref. 29).
3.2 Marine Terminals Transferring Crude Oil and Gasoline
in the Houston-Calveston Intrastate AQCR
The objective of this section is to identify and
describe those marine terminals in the Houston-Galveston Intra-
state AQCR which transfer crude oil and gasoline to and from
marine vessels. Table 3.2-1 lists those terminals for which
sufficient information was obtained to identify them as terminals
transferring crude oil and/or gasoline. Figure 3.2-1 shows the
general location of the terminals in Table 3.2-1.
Although there are several other marine terminals in
the area which transfer crude oil and/or gasoline, insufficient
information was obtained on them for inclusion in this report.
Their transfer operations are small enough, though, that their
omission will not greatly effect the accuracy of the report.
The owners of the terminals are Amerada Hess Corp., Pak Tank Co.,
and Adams Co.
A description of each of the marine terminals in
Table 3.2-1 follows. This description includes the layout
of the dockside facilities and the dockside operational methods
-20-
-------
I
ho
REFINERY
STORAGE
TOTAL
GASOLINE
6401
6357
MOTOR
GASOLINE
TANKER/BARGE 563
PIPELINE 4848
TANK CARS/TRUCKS 946
44
AVIATION
GASOLINE
TANKER/BARGE 10
PIPELINE 12
TANK CARS/TRUCKS 22.
FIGURE 3.1-2 TRANSPORT OF GASOLINE
(RATES IN THOUSANDS OF BARRELS PER DAY, 1974)
-------
Exxon- Bay town
Shell-Deer Park
AMOCO-Texas City
Marathon-Texas City
ARCQ-Pasadena
Charter -Houston
TABLE 3.2-1
MARINE TERMINALS TRANSFERRING CRUDE OIL OR GASOLINE
IN THE HOUSTON-GALVESTON AQCR
Terminal
Gasoline
Crude Oil
Ships
Barges
Ships
Barges
Loaded Unloaded Loaded Unloaaed LoadedUnloaded Loaded Unloaded
Texas City Refining-
Texas City
Crown-Deer Park
-------
to
CO
© ©X ' ©
CHARTER ARCO CROWN
TCR/MARATHON
AMOCO
FIGURE 3.2-1
LOCATION OF MARINE TERMINALS
IN THE HOUSTON-GALVESTON AQCR
-------
used for loading and unloading ships and barges with crude oil
and gasoline. No gasoline unloading operations were identified
in the area, and the Exxon terminal was the only one loading
crude oil.
3.2.1 Exxon Baytown Refinery Marine Terminal
The Exxon marine dock facility is located on the east
side of the Houston ship channel, the terminal consists of three
separate docks as shown in Figure 3.2-2. Each dock has two
berths, and each berth can handle one tanker or a minimum of
two barges. The following information was obtained by corres-
pondence between Radian and Exxon (Refs. 14,15).
3.2.1.1 Gasoline Loading System
Docks 1 and 2 handle all of the gasoline loading onto
marine vessels. From storage the gasoline is pumped to the dock
area through one or more of five loading lines by various combi-
nations of seven pumps. The different types of lineups that can
be made with the gasoline pumping system is shown in Table 3.2-2.
The gasoline storage tanks, located upstream of the
pumps, are gauged manually to determine how much gasoline is
loaded, and flow meters are also located on each of the five
dock loading lines. The meters are part of a computer controlled
blending operation.
The arrangement of the five gasoline loading lines to
Dock 1 and Dock 2 is shown in Figure 3.2-3. Each loading line
is capable of pumping gasoline to either dock. Figures 3.2-4 and
3.2-5 present the manifolding arrangement of the gasoline loading
lines at Dock 1 and Dock 2, respectively. Each circled pipe
intersection indicates that a valved interconnection exists
between the pair of crossing pipes.
-24-
-------
FIGURE 3.2-2
EXXON BAYTOWN REFINERY
MARINE TERMINAL
TO GALVESTON BAY
\
-25-
-------
TABLE 3.2-2
GASOLINE PUMPING SYSTEM LINEUPS
Nominal
Capacity, Products
Pump MB/Houra Pump Location Handledb
P-l 8 Pump Slab 137 Leaded Mogas,
Avgas
P-2 5 Pump Slab 137 Leaded Mogas,
Avgas
i
Ni
CT>
' P-3 10 Pump Slab 137 Leaded Mogas,
Avgas
P-16 10 Pump Slab 194 Leaded Mogas
P-17 10 Pump Slab 194 Leaded Mogas
P-18 10 Pump Slab 194 Leaded Mogas
P-19 10 Pump Slab 194 Unleaded Mogas
Dock Lines
Used
7, 18, 53
7, 18, 53
7, 18, 53
7, 12, 53
7, 12, 53
7, 12, 53
10
a M = thousands
b
Mogas = motor gasoline, Avgas = aviation gasoline
-------
TO
DOCK
53
18
10
FROM GASOLINE PUMPS
7 10 18 53 12
_/• \
Kx« \
12
18
10
TO
FIGURE 3.2-3 GASOLINE LOADING LINES TO DOCKS 1 AND 2
-------
D.L 7
D.L. 10
D.L. 12
D.L. 18
i
Ni
CO
I I
L.S. 1
L.S. 2
J
L.S. 3
L.S. 4
BERTH #1
BERTH *2
KEY
D.Lr DOCK LINE
L.S.- LOADING SPOT
VALVED INTERCONNECTIONS
_j_- 12" LOADING ARM
FIGURE 3.2-4 GASOLINE LINE MANIFOLDING AT DOCK 1
-------
BERTH #4
2468
^
VO
I
in. In
KEY
D.L- DOCK LINE
- VALVED INTERCONNECTIONS
lh
'
1357
BERTH #3
P.L.53
.D.L.18
;f
A
)
.......... r* ^
L
P=I v
J \^
J
J \
\
O e>B^L-7
\J ' "" \l7 -_--.-^^_ -.
o—o
D.L.12
O—6
D.L.10
CJ
&x
FIGURE 3.2-5 GASOLINE LINE MANIFOLDING AT DOCK 2,
-------
At Dock 1 eight-inch diameter cargo loading hoses are
used to transfer gasoline from the shoreside manifolds to the
vessel's manifold. These hoses are bolted to the specified vessel
and dock flanges for a particular cargo transfer. Booms are used
to lift the heavy hoses into place- for making the required connec-
tions and to position the hose so that the line does not twist
during loading. Each hose can handle a maximum of"8,000 bph.
At Dock 2 metal hydraulic loading arms provide a permanent
connection to the shoreside manifold. There are four 12-inch diam-
eter loading arms and one 8-inch arm at each berth. Each of these
arms has a special type of swivel joint that allows the arm to
pivot as needed during the loading operation. Each loading arm
is attached to the vessel manifold by a flange on the end of the
arm. Adaptors are kept at each berth so that loading arms can
be used with barges as well as tankers. Each 12-inch arm can
handle transfer rates of over 13,000 bph.
The gasoline loading rate at Dock 1 can vary from 5,000
bph to 40,000 bph. The actual rate for a ship varies widely,
depending upon such factors as the number of pumps and loading
lines being used, the number and size of vessel tanks being filled,
the types and sequence of materials loaded and the loading prac-
tices used by crew members on the vessel. The maximum loading
rate for a single barge is not expected to exceed 8,000 bph.
For Dock 2 the gasoline loading rate can vary from
5,000 bph to 50,000 bph. The actual loading rate depends upon
the factors mentioned above. Barges are not expected to be
loaded at a rate higher than 8,000 bph.
-30-
-------
3.2.1.2 Crude Oil Loading/Unloading System
Crude oil loading and unloading of marine vessels occurs
at Dock 2 and Dock 5. For loading, crude oil is pumped to the
docks through one or more of three lines by various combinations
of three pumps. Figure 3.2-6 shows how these three lines, numbered
43, 44, and 67 run between the docks, the pumps, and the refinery
storage area. Two of the three crude oil pumps which are located
on Pump Slab 91 have capacities of 10,000 bph while the other
has a capacity of 12,500 bph. As with gasoline, the quantity
of crude loaded is determined by gauging shoreside tank levels.
The manifolding that exists at Dock 2 and Dock 5 for
the crude oil lines is shown in Figure 3.2-7. The circled pipe
intersections indicate that tho.se lines are connected. Eight-
inch cargo hoses are used to transfer the crude oil between the
shore and the vessel at Dock 2. At Dock 5, both hoses and metal
hydraulic loading arms are used. Both Dock 2 and Dock 5 have
booms for positioning the cargo hoses.
, Typical tanker crude oil loading rates at Docks 2 and
5 vary for the same reasons cited in the gasoline loading system
discussion. The maximum and minimum rates are 32,500 bph and
10,000 bph, based on the capacity of the pumps at Pump Slab 91.
An average rate of 14,000 bph is the maximum for single,barges
loading crude. Average rates for a tanker loading crude can
be as high as 30,000 bph.
For crude oil unloading, tankers and barges use the
same facilities at Dock 2 and Dock 5 as described for crude oil
loading shown in Figure 3.2-7. The crude oil is pumped using
the vessel's pumps through Line No. 105 (see Figure 3.2-6) to the
primary refinery crude storage tanks or through some combination
-31-
-------
I
to
N5
I
67
44
43
TO/FROM
DOCK 5
TO/FROM
DOCK 2
105
X
62
61
PUMP
SLAB
91
TO REFINERY
CRUDE TANKAGE
TO/FROM REFINERY
CRUDE TANKAGE
LINES TO/FROM
DOCK AREA
CRUDE TANKAGE
FIGURE 3.2-6 CRUDE OIL LINES TO/FROM DOCKS 2 AND 5
.
-------
10
o
-j
•
o
CO
CO
i
BERTH 6
N.
ID
-t
_i
d
CO
BERTH 5
in
o
(O
_J
Q
O
BERTH 4
BERTH 3
DOCK 5
II Cb—ll
DOCK 2
KEY
—1| 8" CARGO HOSE CONNECTION
—-4 12" LOADING ARM
CJ>— VALVED INTERCONNECTION
D.L- DOCK LINE FIGURE 3.2-7 CRUDE LINE MANIFOLDING AT DOCKS 2 AND 5.
-------
of Lines 43, 44, and 67 to a number of small crude storage tanks
in the dock area. Should these smaller tanks have insufficient
capacity for the quantity of unloaded crude, the three crude oil
pumps at Pump Slab 91 can be used to transfer crude from the dock
tanks to the primary crude storage tanks at the same time the
vessel is unloading. The shoreside tank levels are manually
gauged to determine the quantity of crude oil unloaded from the
tanker or barge.
Crude oil unloading rates vary considerably from vessel
to vessel, depending primarily upon the number and size of the
vessel's pumps. Larger tankers with high capacity pumps unload
at average rates up to 40,000 bph. Peak rates for these tankers
can be as high as 60,000 bph early in the unloading cycle when
the ship's pumps have relatively high suction pressure and low
discharge pressure due to the high tank level on the ship and low
tank level on shore. The average unloading rate for barges
may be as high as 11,000 bph. Relatively little crude oil is
shipped into the refinery by barge.
3.2.2 Shell Deer Park Refinery Marine Terminal
The Shell marine facility is located in a slip on the
south side of the Houston Ship Channel. There are four docks
at the terminal located as shown in Figure 3.2-8. Each dock can
handle one tanker or two barges." The following information was
obtained by correspondence with Shell.
3.2.2.1 Gasoline Loading System
Gasoline is loaded at any of the four docks onto marine
tankers or barges. From storage the gasoline is pumped to the
dock area through one of several systems. The piping system
-34-
-------
HOUSTON SHIP CHANNEL
SHELL MARINE
TERMINAL
FIGURE 3.2-8
SHELL DEER PARK MANUFACTURING
COMPLEX MARINE TERMINAL
-------
for the pumping of gasoline from storage to the four docks is
shown in Figure 3.2-9. Each dock can handle Regular, Unleaded,
and Premium Shell motor gasoline.
At each dock cargo loading hoses are used to transfer
gasoline from the shoreside manifolds to the vessel's manifold.
The hoses are bolted to the specified vessel and dock flanges
for a particular cargo transfer operation. Booms are used to
lift the heavy hoses into place for making the connections and to
position the hose so that kinks do not develop during loading.
The loading pumps located in the refinery tank farm
vary in capacity from 3500 to 5500 bph. The reported maximum
rate at which gasoline can be loaded across any one dock is
25,000 bph. The maximum loading rate for all docks loading
simultaneously is also 25,000 bph. The actual rate for a ship
or barge varies widely, depending upon such factors as the number
of pumps and lines being used,'the number and size of vessel
tanks being filled, the types and sequence of materials loaded
and the loading practices of the vessel's crew.
3.2.2.2 Crude Oil Unloading System
Crude oil is received at the Shell refining complex
at Dock 1 or Dock 4. The crude oil is pumped, using the vessel's
pumps, through either of two available 24-inch lines. This
arrangement allows two vessels to discharge crude simultaneously
at the terminal. The two 24-inch lines are sized hydraulically
for 40,000 bph maximum flow rate when used in parallel. Average
flow rates generally run from 70 to 80% of the ship's maximum
pump rate. Construction is currently underway at the terminal
for the addition of a fifth dock which will be employed solely
for receiving crude oil. Expected date of completion is 1977.
-36-
-------
STORAGE TANKS
LOADING PUMPS
REGULAR
GASOLINE
NORTH 1 1
GASOLINE y y
SOUTH i J.
GASOLINEf T
BLENDER J "-
UNLEAOEOl
UNLEADED
REGULAR
[""""""
REGULARl
PREMIUM]
i
RFGIII AP
BLENDING STOCKS
1 I Y i
T T 7 T
1 1 1 1
hl=h
h' h
j | i 1 1
r"*^
J xZx
1 "1 1 '
\^j
~] "I 1 L
r
|
r
|
r
1
i
/" ~\
L J
^™^
i
f \
XL/
f \
H
r
(
r ~\
r
i r
\^J
1 '
r
r.^
I
,
i
r
i
DOCK
DOCK
" 2
UNLEADED
REGULAR . _QC1,
PREMIUM OOCK
REGULAR '
FIGURE 3.2-9
GASOLINE LOADING LINES TO SHELL'S
MARINE DOCKS
-37-
-------
3.2.3 AMOCO Texas City Refinery Marine Terminal
The AMOCO marine facility consists of 6 docks located
on' the west side of the Texas City Harbor as shown in Figure 3.2-10,
Each dock has one berth. Dock No. 30 is used exclusively for
handling barges, Nos. 31 and 32 are used for both ships and barges,
No. 33 is for barges only, and Nos. 32A and 40 are used exclusively
for handling crude oil from ships. Information for this section
was supplied to Radian from AMOCO Oil Co. (Refs. 2,3).
3.2.3.1 Gasoline Loading System
Docks 30, 31, and 32 handle the loading of gasoline to
ships and barges. From storage the gasoline is pumped to the
docks using one of four available lines. Table 3.2-3 identifies
the lines and pumps typically used for loading gasoline.
TABLE 3.2-3
LINES AND PUMPS FOR MARINE LOADING OF
GASOLINE - AMOCO TEXAS CITY
Line Number Pumps Nominal Cap.
9 F-62A or F-62B 5,000
9A F-62A or F-62B 5,000
9B L-l, L-2, or L-3 4,500
804 L-l, L-2, or L-3 3,850
-38-
-------
I
co
VO
FIGURE 3.2-10
AMOCO TEXAS CITY REFINERY
MARINE TERMINAL
INDUSTRIAL SHIP CANAL
-------
The arrangement of the gasoline lines to Docks 30, 31,
and 32 is shown in Figure 3.2-11. At each dock, hoses are used
for loading the gasoline cargo from the dock manifold to the ship
manifold. These hoses are bolted to the specified vessel and dock
flanges for a particular cargo transfer.
Figure 3.2-11 shows that the maximum rate gasoline can
be loaded onto vessels at "Docks 31 and 32 is 18,350 bph. The
maximum rate barges at Dock 30 may be loaded is 4,500 bph. The
actual loading rate at Docks 31 and 32 varies widely, depending
upon the lines and pumps used, the number and size of the vessel
tanks being filled, the types and sequence of materials loaded
and the loading practices of the crew. The maximum loading
rate for barges is usually less than that for ships. The amount
of gasoline loaded is determined by measuring the ship tank ull-
ages and/or gauging the onshore product storage tanks.
3.2.3.2 Crude Oil Unloading System
Crude oil can be unloaded at Docks 31, 32, 32A, 33, and
40. However, Dock 33 generally handles distillates. Nos. 32A
and 40 are employed solely for receiving crude oil from tankers.
The crude oil is pumped from the ship cargo tanks to onshore
storage using lines 800, 801, and 805. Figure 3.2-12 shows the
arrangement of these lines to the different docks. There are two.
booster pumps, one each for lines 800 and 801. There is no boos-
ter pump for line 805. The vessel pumps are used primarily for
unloading the crude from the cargo tanks. Although actual unload-
ing rates for each vessel vary considerably with the number and
size of the vessel pumps, the maximum unloading rate for Docks
31, 32, and 33 is 10,000 bph. Docks 32A and 40 have individual
maximum unloading rates of 28,000 bph.
-40-
-------
-p-
h-1
I
LINE 9 5,000 BBLS/HR
FROM
REFINERY STORAGE
LINE 9A 5,000 BBLS/HR
LINE OB 4.500 BBLS/HR
LINE 804 3,850 BBLS/HR
FIGURE 3.2-11
GASOLINE LINES TO AMOCO TEXAS CITY
MARINE TERMINAL
DOCK
32
DOCK
31
DOCK
30
-------
I
•p-
K5
I
LINE 800 5.000 BBLS/HR
TO REFINERY
STORAGE
LINE 801 5,000 BBLS/HR
FIGURE 3.2-12
CRUDE OIL UNLOADING LINES -
AMOCO TEXAS CITY
DOCK 31
TO REFINERY
STORAGE
DOCK 33
LINE 805 28,000 BBLS/HR
D.OCK 32
DOCK 32A
TO MARATHON OIL CO.
DOCK 40
-------
3.2.4 ARCO Houston Refinery Marine Terminal
The ARCO refinery marine facility is located on the
south side of the Houston Ship Channel as presented in Figure
3.2-13. Dock A is used solely for barges and Dock 13 for ships.
Information for this section was supplied to Radian by ARCO
(Refs. 5,6).
3.2.4.1 Gasoline Loading System
Figure 3.2-14 is a diagram of the piping which transfers
gasoline from ARCO's tank farm to the two docks. From the tank
farm gasoline is pumped by one of three pumps to the dock area
through one or more of three loading lines.
The gasoline loading rate at Dock A can vary from 1,050
bph to a maximum of 11,050 bph. The actual rate varies, depending
upon the number of pumps and loading lines used, the number and
size of the barge tanks being filled, and the loading practices
of the barge tankmen.
The gasoline loading rate at Dock B can vary from 5,000
bph to a maximum of 20,000 bph. The actual loading rate varies
with the same factors mentioned above.
3.2.4.2 Crude Oil Unloading System
Crude oil can be unloaded at all three of ARCO's docks.
Dock A can handle only barges, while Dock B can handle both ships
and barges and Dock C can handle light ships and barges.
An unloading facility may be built in the future at
Bayport, Texas for crude oil destined'for ARCO's Houston refinery.
Use of this facility would substantially reduce the crude oil
transfers at the existing docks.
-------
N
ARCO REFINERY
FIGURE 3.2-13
ARCO HOUSTON REFINERY MARINE TERMINAL
-------
TANK
FARM
CLEAR, 1050 B/H
SUPREME 10,000 B/H
ARCO 10.000 B/H
SUPREME 5,000 B/H
ARCO 5.000 B/H
CLEAR, 1050 B/H
SUPREME 10,000 B/H
ARCO 10,000 B/H
DOCK A
BARGES
DOCK B
SHIPS
FIGURE 3.2-14
GASOLINE LOADING LINES
TO DOCKS A AND B
ARCO HOUSTON MARINE TERMINAL
-45-
-------
Figure 3.2-15 is a diagram of the crude lines which
transfer the crude oil from each dock to the refinery tank farm.
Hydraulic arms and cargo hoses are used to connect ship to shore.
Metering is done by gauging of onshore tanks.
3.2.5 Texas City Refining Texas City Refinery Marine
Terminal
Texas City Refining shares its marine terminal docks
with Marathon Oil Co. The facility consists of 5 docks located
on the west side of the Texas City Harbor as shown in Figure 3.2-16
Each dock has one berth. Dock No. 3 handles barges only. The
others can handle tankers or barges. Texas City Refining provided
Radian with part of the information presented in this section
(Ref. 25). The rest of the data came from EPA Region VI (Ref. 26).
3.2.5.1 Gasoline Loading System
The loading of gasoline onto barges and ships is
accomplished at three docks, Nos. 2, 4, and 5. TCR has five
pumps which transfer the gasoline to the dock's. Their capacity
ranges from 2,800 bph to 7,000 bph. Table 3.2-4 indicates
the maximum gasoline loading rate at each dock. The gasoline
loaded is metered by gauging of onshore tanks and/or ship tank
ullage.
TABLE 3.2-4
MAXIMUM GASOLINE LOADING RATE AT TEXAS CITY
REFINING'S MARINE DOCKS
Dock No. Maximum Rate (bph)
2 ^ 12,000
4 ' 12,000
5 20,000
-45-
-------
2,000 BPH
•
DOCK A
5-°00 BPH . -. . BARGES
5,000 BPH
7,000 BPH
DOCK B
25.000 BPH
SHIPS AND BARGES
25.000 BPH DOCK C
LIGHT SHIPS
AND BARGES
FIGURE 3.2-15 CRUDE OIL LOADING LINES FOR
ARCO HOUSTON MARINE TERMINAL
-47-
-------
00
I
INDUSTRIAL SHIP CANAL
FIGURE 3.2-16 TEXAS CITY REFINING MARINE TERMINAL
-------
At docks 2, 4, and 5, 8-inch loading hoses are used
for connecting the dock manifolds to the ship manifolds.
3.2.5.2 Crude Oil Unloading System
Very little information was obtained on TCR's crude
oil unloading system. However, it is assumed that all five of
the docks shared with Marathon are capable of unloading crude
oil. The maximum rate at which crude was unloaded from a ship
in 1975 was 16,500 bph. The ship's pumps are .used to pump the
crude to TCR's storage tanks. Metering is done by tank gauging.
3.2.6 Crown Central Houston Refinery Marine Terminal
Crown Central Petroleum Company's marine facility is
located on the south side of the Houston Ship Channel as shown
in Figure 3.2-17. It consists of only one dock. No gasoline
was loaded at this dock in 1975. It was used to receive crude
oil and to periodically ship out various refinery products. Data
in this section are based on correspondence between Crown and
Radian (Ref. 11).
3.2.6.1 Crude Oil Unloading System
Both ships and barges unload crude oil at the dock.
The crude is discharged using the vessel's pumps into three
8-inch cargo hoses which connect the ship manifold to the dock
manifold. From the dock one 18-inch pipe and one 10-inch pipe
transfer the oil to refinery storage. A normal range for the
discharge rate is 9,000 to 15,000 bph. Metering is accomplished
by onshore gauging.
-49-
-------
N
HOUSTON SHIP
i
Ol
o
CROWN CENTRAL REFINERY
FIGURE 3.2-17
CROWN CENTRAL HOUSTON REFINERY MARINE TERMINAL
-------
3.2.7 Charter Oil Houston Refinery Marine Terminal
Charter Oil Company's marine facility is located on
the south side of the Houston Ship Channel as pictured in
Figure 3.2-18. It consists of three docks only two of which
transfer crude oil and gasoline. The Traweek Dock handles both
ships and barges, while Dock No. 4 handles only barges. Dock
No. 3 transfers chemicals. The information in this section was
obtained from EPA Region VI files (Ref. 10).
3.2.7.1 Gasoline Loading System
No information was provided about the exact arrangement
of lines and pumps from refinery storage" to the docks at Charter's
terminal. However, it is known that barges can load gasoline
at #4 Dock at a rate of 4,500 bph and ships can load at 7,500 bph
at the Traweek Dock. Eight-inch cargo hoses are used to transfer
the liquid.
3.2.7.2 Crude Oil Unloading System
No information was provided concerning the crude oil
unloading operations or the dock equipment. It is assumed, how-
ever, that the Traweek Dock would be used, and the discharge
rate would depend upon the specific ship unloading.
3.2.8 Marathon Texas City Refinery Marine Terminal
The Marathon Oil Company Marine Terminal consists of
five marine docks shared with Texas City Refining and one shared
with AMOCO. Figure 3.2-12 shows Dock 40 which Marathon shares
with AMOCO, and Figure 3.2-16 shows the five docks which Marathon
shares with Texas City Refining. The following information was
obtained partially from Marathon correspondence with Radian and
partially from EPA Region VI (Ref. 19, 20).
-51-
-------
CHARTER REFINERY
FIGURE 3.2-18
CHARTER OIL HOUSTON REFINERY MARINE TERMINAL
-------
3.2.8.1 Gasoline Loading System
Relatively little information was obtained on the set-
up of docks, pumps, and lines used for transferring gasoline
from Marathon's refinery to ships and barges at its terminal.
It is known, however, that three docks are used fo.r loading
gasoline onto marine vessels and that the typical loading rate
is 5,500 bph. As mentioned before, all docks except No. 3 can
handle both ships and barges. No. 3 handles only barges.
3.2.8.2 Crude Oil Unloading System
As with the gasoline loading system, relatively little
information concerning the crude unloading system was obtained.
The only information available pertains to Dock 40 which Marathon
shares with AMOCO. This dock is dedicated to unloading crude
from tankers. The actual unloading rate varies from tanker to
tanker depending on the number and size of the vessel pumps.
3.3 Shipside Equipment and Transfer Procedures
This section describes ship and barge equipment used
in transferring gasoline or crude oil in the Houston-Galveston
area. Typical procedures used to load and unload gasoline and
crude oil from marine vessels are also described.
3.3.1 Crude Oil and Gasoline Loading of Ships
In the Houston-Galveston area the loading of crude oil
and gasoline involves similar techniques and equipment since the
same docks and similar size tankers are involved. The following
material relates the typical sequence of steps involved in
loading a tanker with gasoline.
-53-
-------
This hypothetical case involves a Grade A tanker in
dedicated gasoline service from a refinery terminal in the
Houston-Galveston area to the East Coast. As the ship nears
the refinery dock, it discharges into the Channel the clean
ballast water it has onboard. With the help of several tugs it
is piloted into position at the dock and made fast to the shore
moorings with its heavy docking lines. Crew members and shore
personnel next connect the ship's slop line' to the shore slop
line. Any oily ballast water onboard the ship is pumped through
this line to the refinery for treatment before it can be dis-
charged. Depending upon the amount of dirty ballast on board,
this operation may take 10 or more hours to complete. Once the
deballasting of the cargo tanks is complete, they are stripped,
using small stripper lines located in the bottom of each tank.
This operation removes the small amount of ballast the larger
cargo pumping lines, cannot remove.
After the deballasting is finished, cargo loading
hoses are used to connect the ship with the shore in preparation
for receiving product. A specific loading pattern and loading
sequence for the tanks is determined by the ship's officers.
Improper loading patterns can cause the vessel to be improperly
trimmed or even rupture if the stresses are sufficiently high.
A diagram of the ships cargo tanks and the product each will
carry is shown in Figure 3.3-1. Flexible hoses will be attached
to the proper shore and ship flanges. After each tank has been
visually inspected and okayed, the deck officer advises the
shoreside operators that the ship is ready to accept cargo.
Before the shoreside loading pump is turned on, the product is
usually allowed to drain from the shoreside tank through the
loading line and into the vessel's tank (or tanks). This is
done to insure that flow has been established and that the cargo
lineup is correct.
-54-
-------
Once verification has been made that the lineup to a
tank is correct, the crew advises the shoreside operators to
turn on the loading pump. The displaced vapors are usually
vented through the ullage cap located atop the cargo tank hatch.
The term "ullage" refers to the distance between the cargo
liquid level and the rim of the ullage cap. A ^sketch of a
Grade A tanker cargo vent system is shown in Figure 3.3-2. The
vapors may be vented out the P/V valve to the stack if the ullage
cap is closed. Each cargo tank P/V valve is manually lifted off
its seat during the loading operation to insure that a faulty
valve does not cause overpressurization of the tank. Periodic
checks of the ullage gauges of the tanks are made as they fill
with gasoline. Typically, several tanks are being filled at once,
Loading may be interrupted from time to time to correct trim
on the vessel. For those situations in which three tanks across
are being filled simultaneously with the same grade of gasoline,
a special loading sequence is usually followed. The level of
the center tank and the two wing tanks is allowed to reach an
ullage of perhaps 15 to 20 feet. Then the flow to the center
tank is shut off and the two wing tanks are brought up, one
level slightly behind the other. Usually two to four members
of the crew are responsible for bringing the product level up
to the final ullage called "topping off". They do this with
calibrated sticks about five to six feet long which resemble
crosses. These sticks are inserted like a dipstick into the
tank from the' ullage cap and the ullage read directly from the
stick. When the product reaches the desired final ullage the
flow to that tank is shut off. Then the other wing tank is
topped off soon after. Following this the flow is resumed into
the center tank until it is topped off. This procedure is used
for safety reasons. The wing tanks have a smaller volume than
-55-
-------
FLANGES }
CO
o
a
CO
•?
\~
O
J-
i
• £
0
o
T-
c
J" 6
i
Ui
ON
I
10
PORT
4,833 BBLS
10
CENTER
11,026
BBLS
10
STARBOARD
4,833 BBLS
8P
11,801 BBLS
90
11,047
BBLS
OC
11,075
BBLS
OS
11,801 BBLS
7P
5,991 BBLS
70
11,103
BBLS
7S
5,991 BBLS
5P
11,997 BBLS
60
11,131
BBLS
60
11,158
BBLS
5S
11,997 BBLS
4P
5,791 BBLS
40
11.131
BBLS
4S
5,971 BBLS
2P
11,841 BBLS
30
11,103
BBLS
20
11,047
BBLS
2S
11,841 BBLS
1P
6.024 BBLS
1C
13,736
BBLS
1S
6.024 BBLS
FIGURE 3.3-1
TANK CAPACITIES AND MANIFOLD
ARRANGEMENT OF THE S.S. "PASADENA
-------
I
Ul
FIGURE 3.3-2
GRADE A CARGO TANK VENT SYSTEM
8" PIPE
-SPILL VALVE (SECTION VIEW)
ULLAGE CAP
P/V VALVE (SECTION VIEW)
CARGO
TANK HATCH
m/ Vin
1 psan 1
UJ llj
,-TANKER DECK
f
n J
n i
i .
u
VENT
STACK SCREEN
STAINLESS STEEL
WIRE CLOTH
CD
I
in
-------
the center tank. . Should any problem occur during the topping
off of a wing tank, flow can be quickly and easily diverted
into the center tank which has plenty of available space. Another
reason for this sequence is that it is more difficult to top off
three tanks in a short time than it is to first finish the two
wing tanks and then the center.
For the topping off of the final cargo tank loaded
with gasoline, the crew keeps in touch with the shoreside
operators with walkie-talkies. A crew member notifies the
operators the instant they should shut off the loading pumps
of that grade of gasoline to complete the product transfer.
Then the loading lines are disconnected; the ullage caps are
sealed shut; the P/V valves are returned to their operating
position; and the crew readies the ship for departure.
3.3.2 Crude Oil and Gasoline Loading Onto Barges
The loading of gasoline onto barges in the Houston-
Calves ton area is a common practice. However, very little crude
oil is loaded onto barges in the area. The loading procedures
are similar, however, for the same reasons that loading ships
with gasoline and crude oil is similar.
Barges differ from ships in that they do not take on
ballast after unloading. Empty barges are returned by tugboat
to the terminal where they are to load their next product.
Usually no cleaning is performed on the cargo tanks because barges
lack cleaning facilities and convenient disposal methods for the
cleanings, and for these reasons they remain in a single product
service. This is true until it is sent to drydock for repairs.
-58-
-------
In this case the barge tanks are cleaned by removing all
hydrocarbon vapors so that regularly schedules maintenance
on its equipment can be performed. Following this work, the
barge would be free to switch cargo service.
For loading gasoline or crude oil the barge is moved
into position at the marine dock by a tugboat and then tied up
using its mooring ropes. Cargo hoses or hydraulic arms, if
they are available, are attached to the barge's cargo loading
header and to the shore manifold. A diagram of a barge's tanks
and piping is shown on Figure 3.3-3.
The barge is filled in much the same manner as is a
ship. Usually, only one person is available to monitor loading
operations on the barge, though. Barge tanks require more fre-
quent monitoring because the loading rate is generally higher
relative to tank size on barges than on tankers. Topping off is
completed in the same manner on barges as on ships. Observa-
tions on the product level are made by direct sighting through
an ullage cap. The tank ventilation system on barges resembles
Grade B cargo tank vent systems. Figure 3.3-4 shows a sketch
of this type of ventilation system.
3.3.3 Crude Oil and Gasoline Unloading from Tankers
The unloading of gasoline from tankers rarely, if
ever, occurs in the Houston-Galveston area. The unloading of
gasoline would be similar to that of crude oil unloading since
similar size tankers are involved. Large quantities of crude
oil are imported into Houston-Galveston area refineries by tanker,
The majority of these ships fly foreign flags. A description of
the equipment and the procedure used for unloading crude from
tankers follows.
-59-
-------
o
i
1
o
1
0
in
0
in
CM
o
i
in
CM
' ... " ' ' a t ii
7-6 52-6 49-6
0 NO. 4P
2770 BBLS :
PUMP ENG'INE
j p&s i '
\ JT C3H>#
ClARGO PUMP ^
' p&s r
1 rfi
'NO. 4S DISCH
2^67>->BBLs HEADE!
M
H
R
! NO. 3P or
^| 2671 BBLS |
10!" LOAD. HE'ADER j
! [
, 52-6
I NO. 2P o
2742 BBLS
i •
^
iffo^TC ~~ ~~~~ ~"~ — — — — — fif — — — — — -— — —
MD'Cit NO. 3S~ "i
V^i V 2665 BBLS !
*) , lO^LONG. ^HEADER |
HO1!1 LOAD. CONN. 1
1 0<
1 1
/ // / /-
52-6 23-6
NO. 1 P o'
2557 BBLS |
2" SOUND'ING^ PLUG }
rv-jj-*^'011 LOAD. CONN.i VOID
1° NO. 2S
-------
I
cr>
ULLAGE CAP
CARGO TANK HATCH
P/V VALVE
(SECTION VIEW)
FIGURE 3.3-4 GRADE B CARGO TANK VENT SYSTEM
-------
After the ship is docked at the terminal where it will
discharge its load, dock and ship personnel connect the shore
and ship manifolds using cargo hoses or hydraulic arms. Then
the ship's main cargo pumps are used to discharge the crude.
These pumps vary in number and capacity for different tankers.
The tanks are unloaded from the bottom, just as they are loaded.
During unloading the P/V valves are manually opened and the
ullage caps are opened. There are two cargo tank vent systems
for tankers. Figure.3.3-2 shows the system for a Grade A ship
and Figure 3.3-4'shows the system for a Grade B ship.
Once the main cargo pumps have removed all the oil
they can, they are switched off. The smaller stripper pumps
and lines are used to remove the remaining crude oil from each
tank. This procedure is called stripping. Each cargo tank's
strippings are pumped to a designated cargo tank usually located
aft. When all the tanks have been stripped, the main cargo
pump for the tank holding the strippings is used to pump them
ashore. This completes the unloading operation.
Before the tanker departs, however, it must take on
some ballast to make it seaworthy. A ballast diagram drawn by
one of the ship's officers determines which tanks will be
ballasted. The sea valves to these tanks are opened allowing
water to flow in. The displaced vapors are vented through the
ullage cap and P/V valve which is still open.
Ships reportedly may ballast anywhere from 20 to 40
percent of their cargo capacity before leaving the dock, depend-
ing upon the ship officer's orders. Should weather conditions
dictate it, more ballast may be taken on while the ship is at
sea. The level of ballast in those tanks ballasted is usually
brought up fairly high to minimize the danger of the ship's
-62-
-------
developing severe rolling in bad weather due to the sloshing of
the ballast in its tanks.
Assuming all ballasting is done in port, after it is
completed, the ullage caps are closed; the P/V valves are re-
turned to 'their normal position; and the ship readies for departure,
3.4 Quantities of Crude Oil and Gasoline Transferred in
the Houston-Calves ton Area
This section provides monthly information concerning
the quantities of crude oil and gasoline transferred at marine
terminals in the Houston-Calveston area. In all but one case
the information was obtained directly from those oil companies
in the area transferring either crude oil or gasoline or both
to marine vessels.- Estimates from the Texas Air Control Board
were used for the terminal owned by Amerada Hess. The list of
companies is not complete since a few of the small terminal
operators in the area did not provide data on their transfers.
However, their totals are'practicably insignificant compared to
the larger terminals (Ref. 3,6,10,11,15,19,23,25).
A summary of the quantity of gasoline (motor and
aviation) loaded at marine terminals in the Houston-Calveston
area is presented in Table 3.4-1. Assuming that Charter Oil and
Amerada Hess loaded approximately the same amount of gasoline
in 1975 as in 1974, then the total amount loaded for the area
is roughly 82.5 million barrels for 1975. Table 3.4-2 shows
the variation of the Reid Vapor Pressure (RVP) of the gasoline
loaded at the terminals by month for 1975.
Table 3.4-3 presents data on the amount of crude oil
loaded in the Houston-Calveston area; Exxon in Baytown was the
only terminal located which loaded crude oil. In 1975 Exxon
loaded about 7.5 million barrels of crude.
-63-
-------
TABLE 3.4-1
Exxon
She 1.1
AMOCO
Mnr.-itliou
AHCO
Aiiicr;ir.
!,739 1.737 2,777 2,445
.575 1,033 862 1
NA NA NA
NA NA NA
274 476 80
184 271 679 1
NA NA NA
NA NA NA
.843
NA
NA
306
.021
NA
NA
May
2,1)15
1,094
NA
NA
582
247
NA
NA
June July
2, 946 3,583
1,757 1.519
NA NA
NA NA
499 479
389 65
NA NA
NA NA
Jhbls)
Aug. Sept. Oct.
3,274 2,910 3,129
1.669 1,367 1.368
NA NA NA
NA ' NA NA
304 69 143
427 278 318
NA NA NA
NA NA NA
Nov. Due. Tutiil
2.456 2.721 31.514
1.36.3 1.145 I6.VJ4
NA NA 19, S60
NA NA H, 328
511 270 3.993
22 32 3,913
NA NA 4 1 0
NA NA 2.595
NA - Not Av.-iU.-ihle
-------
Marine Terminal
Exxon
AMOCO
Shell
TABLE 3.4-2
REID VAPOR PRESSURE OF GASOLINES LOADED AT
MARINE TERMINALS IN THE IIOUSTON-CALVESTON AREA
RVP (psi)
Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec.
U.I 12.4 11.6 11.0 10.5 10.0 10.1 10.3 11.0 12.3 13.2 13.2
13.1 13.1 13.1 10.1 10.1 10.1 10.1 10.1 10.1 13.1 13.1 13.1
12.9 13.1 12.6 10.8 10.4 10.0 9.8 9.7 10.1 11.4 12.9 13.0
i
C7>
Ul
Marathon
ARCO
Texas City Ret.
Amerada lless
Charter
11.5 11.5 11.5 9.5 9.5 9.5 9.5
NA
NA
9.5
9.5 - 11.5 11.5 . 11.5
10.8 11.1 10.4 9.9 9.4 9.5 9.5 9.3 9.4 10.1 11.2 11.7
14.5 14.5 12.5 12.5 12.5 11.5 11.5 11.5 12.5 12.5 12.5 14.5
NA
NA NA
NA
NA
11.0 11.0 11.0 9.0 9.0 9.0 9.0
NA
9.0
NA
NA
NA
NA
9.0 11.0 11.0 11.0
NA - Not Available
-------
TABLE 3.4-3
QUANTITY OF CRUDE OIL LOADED AT MARINE TERMINALS
IN THE HOUSTON-CALVESTON AREA
Marine Terminal
Year Jan.
Feb.
Quantity Loaded (103bbls)
Mar.
Apr.
May
June
July
Aug.
Scpc.
Oct.
Nov.
Dec.
Total
Exxon
1975
649 1.773 1,363
773
304
645 1,070 346 554 7,477
o-.
I
-------
Table 3.4-4 lists the marine terminals which unloaded
crude oil in 1975 along with the monthly and yearly quantities
unloaded. Although this list down not include all of the marine
importers of crude oil in the area, it accounts for over 90
percent of the crude oil imported by tanker/barge. The total
crude oil unloaded as shown in Table 3.4-4 for 1975 is roughly
160 million barrels. The average RVP of this crude is listed
by marine terminal in Table 3.4-5.
3.5 Projected Quantities of Crude Oil and Gasoline
Transferred in the Houston-Calveston Area Through 1985
This section presents projections of the quantities
of crude oil and gasoline to be loaded and unloaded at marine
terminals in the Houston-Calveston area through the year 1935.
Data was obtained from those refineries located in the area
which transfer crude oil or gasoline (Rev. 3,6,15,19,23,25).
Projections presented in Table 3.5-1 of the quantities
of gasoline to be loaded for each year through 1985 are those
made by the owners of the marine terminals.
Table 3.5-2 presents projections of the amounts of
crude oil to be loaded at marine terminals in the area through
1985.
Table 3.5-3 presents the projected quantities of
crude oil to be unloaded at marine terminals through 1985.
-67-
-------
TABLE 3.It-ft
QUANTITY OF CRUDE OIL
UNLOADED
AT MARINE TERMINALS
IN THE HOUSTON-CALVESTON AREA
Marine Terminal Year Jan. Feb. Mar. APr- May
Exxon 1975 3,997 4.286 1.942 2,749 1,980
AMOCO 1975 NA NA NA NA NA
Shell 1975 3,195 2.040 1,690 1,310 871
1 ASCO 1975 1.289 666 1.003 1.576 1.990
Co
' Marathon 1975 NA NA NA NA NA
Texas City Ref. 1975 1.358 992 916 1,303 1,421
Charter ND ND ND ND ND
Crown NA NA NA NA NA
Quantity
June
2,767
NA
347
1.289
NA
1,223
ND
NA
(103 bbls)
July Aug. Sept. Oct.
3.445 4.791 3,347 4,196
NA NA NA NA
1,751 2,465 2,763 2.890
1.758 1.492 1.475 1.134
NA NA NA NA
1,123 1,439 750 1,182
ND ND ND ND
NA NA NA NA
Nov. Dec. Total
5,380 5.012 42.892
NA NA 37,330
2.497 2.516 24,335
1,353 1,277 16.302
NA NA 10.210
1.270 1,386 14,364
ND ND ND
NA NA 15,000
NA - Not Available
ND - No Data
-------
TABLE 3.4-5
AVERAGE RVP OF CRUDE OIL UNLOADED
AT MARINE TERMINALS IN THE
HOUSTON-GALVESTON AREA
Marine Terminal Average RVP
Exxon 4.4
AMOCO 4.0 -
Shell 5.1
ARCO 3.8
Marathon 2.0
Texas City Ref. ND
Charter ND
Crown 3.7
ND - No Data
-69-
-------
TABLE 3.5-1
PROJECTED QUANTITIES OF GASOLINE TO BE LOADED AT MARINE
TERMINALS IN THE HOUSTON-GALVESTON AREA THROUGH 1985
Terminal
Quantity (103bbl)
1976 1977 1978 1979 1980 1981 1982 1983 1984 1985
Exxon
45,625 31,025 32,850 36,500 ND
ND
ND
ND 36,500
AMOCO
Shell
Marathon
Texas City Ref.
19,560 19,560 19,560 19,560 19,560 19,560 19,560 19,560 19,560 19,560
16,590 16,590 16,590 16,590 16,590 16,590 16,590 16,590 16,590 16,590
8,330 8,330 8,330 8,330 8,330 8,330 8,330 8,330 8,330 8,330
3,600 3,850 4,300 4,650 4,950 5,250 5,550 5,875 5,875 5,875
ARCO
500 0
0
ND - No Data
-------
TABLE 3.5-2
PROJECTED QUANTITIES OF CRUDE OIL LOADED
AT MARINE TERMINALS IN THE HOUSTON-GALVESTON AREA THROUGH 1985
Terminal
Quantity (103bbl)
1977 1978 1979 1980 1981 1982 1983 1984
1985
Exxon
1,825 0
9,125 ND
ND
ND
ND
ND - No Data
-------
TABLE 3.5-3
PROJECTED QUANTITIES OF CRUDE OIL UNLOADED AT
MARINE TERMINALS IN THE HOUSTON-GALVESTON AREA THROUGH 1985
Terminal
1976 1977 1978 1979 1980 1981 1982 1983 1984 1985
Exxon
105,850 116,800 127,750 80.3001 ND
ND
NO
ND 18.2501
AMOCO
73,000 73,000 73,000 73,000 73,000 73,000 73,000 73,000 73,000 73,000
Shell
36,500 58,400 65,700 73,000 7.3002 10.9502 10.9502 10.9502 14.6002 14.6005
' ARCO
25,000 53,000 47,000 40,000 40,000 40,000 40,000 40,000 40,000 40,000
Marathon
10,210 10,210 10,210 10,210 10,210 10,210 10,210 10,210 10,210 10,210
Texas City Ref,
20,100 20,300 27,000 31,600 33,000 33,200 35,600 37,000 37,000 37,000
'Assuming Seadock is built
2Assuming Seadock is built. However, if not, the quantities will be 76,650; 80,300; 80,300; 80,300; 83,950;
and 83,950 for the years 1980, 1981, 1982, 1983, 1984, and 1985 respectively. Quantities in 103bbls.
-------
3.6 Cruise History Information for Ships and Barges
Which Transferred Crude Oil or Gasoline in the
Houston-Calveston Area during 1975
The loading of gasoline or crude oil into cargo tank
compartments of marine vessels displaces air which contains vary-
ing concentrations of hydrocarbons. Two factors contribute to
the total emission of hydrocarbons from the vessel's tanks. One
of these is termed the "arrival" component. This portion of the
loss is attributed to the hydrocarbons which are present in the
vessel's empty cargo compartments from its prior voyage. The
other portion is termed the "generated" component, originating
from the hydrocarbons which evaporate from the surface of the
liquid being loaded.
\
The hydrocarbon concentration in the arrival component
is a function of the prior cruise history of the vessel being
loaded and, depending upon the vessel's prior cruise, this compon-
ent may or may not contribute significantly to the overall loading
emission.
This section presents information concerning cruise
histories of the marine vessels which transported crude oil and
gasoline to and from the Houston-Calveston area. The effects of
a vessel's cruise history upon its arrival hydrocarbon concen-
tration are investigated. Descriptions of the types of marine
vessels used to transport crude oil and gasoline in the area are
given. Also, vessels which service the various refineries in the
area are identified. Finally, a breakdown of the emissions which
occur during each portion of a vessel's journey in transporting
gasoline, and a specific analysis of the tank arrival conditions
and associated hydrocarbon concentrations as a function of cruise
history are presented for the Houston-Calveston area.
-73-
-------
3.6.1 Effects of Cruise History on Hydrocarbon Emissions
from Marine Loading of Gasoline and Crude Oil
The results of Radian's sampling study verified the
fact that a vessel's cruise history has a significant effect on
its emissions from loading gasoline. The differences in average
arrival hydrocarbon concentrations ranged from less than 1 percent
for tanks which had been cleaned to over 20 percent for uncleaned
tanks. Ballasted tanks generally showed arrival concentrations
averaging less than 10 percent for a limited number of tests.
Industry data supports the trends observed during the
Radian sampling study. A program was conducted in 1975 by Exxon
at their Baytown refinery which provided the most comprehensive
information to date on ship and barge cruise history (Ref.15).
The primary objective of the Exxon study was to develop accurate
emission factors for the loading of gasoline onto marine vessels.
Data was collected on the arrival condition and prior cruise
history of each tank sampled. From this data three categories
based on the cruise history of the tank were established for
cargo tank arrival hydrocarbon concentrations. Exxon discovered
that defining hydrocarbon emission estimates on these categories
improved the accoracy of the predictions. For tankers, the cate-
gories were defined as follows:
Cleaned - Cargo tanks which had been cleaned
in some manner during the previous
trip or tanks which had previously
carried nonvolatile products.
Dirty ballasted - Tanks which arrived con-
taining dirty ballast.
-74-
-------
Empty and undisturbed - Tanks which had con-
tained gasoline on the
previous voyage and
arrived empty and
undisturbed.
Table 3.6-1 shows the range of values Exxon measured „
within each category. The results shown in Table 3.6-1 are based
on data taken on 70 ship tanks. These results reflect Exxon's
test results and may vary from other company averages due to
different ship operation characteristics. The reasons for the
range of values within a category are differences in 1) the amount
and volatility of the previous product left in the tank after un-
loading, 2) the type and extent of cleaning used during the return
voyage, and 3) the fraction of the tank used for ballast.
TABLE 3.6-1
EFFECT OF SHIP CRUISE HISTORY ON ARRIVAL
HYDROCARBON CONCENTRATION PRIOR TO GASOLINE LOADING
Tank Arrival Average Arrival Typical Range of Arrival
Condition HC Concentration (Vol %) HC Concentration (Vol %)
Cleaned 2.5 0-5.0
Dirty Ballasted 5.0 2.0-8.0
Empty and Undisturbed 8.0 2.5 -13.5
Contacts made by Radian with marine industry personnel
concerning the arrival condition of intercoastal barges indicated
that the barges usually arrive with empty and undisturbed tanks.
This is because barges do not take on ballast, and the lack of
available manpower onboard prevents cleaning of the cargo tanks.
Exxon's study confirmed this observation. They found that all
barges sampled at their docks arrived with empty, uncleaned tanks.
-75-
-------
3.6.2 Types of Marine Vessels Used in Transferring Crude
Oil and Gasoline in the Houston-Galveston Area
There are three distinct classes of vessels used for
transporting crude oil or gasoline into and out of the Houston-
Galveston area. The most common is the marine tanker. The
other two are the intercoastal barge, and the oceangoing inte-
grated tug-barge.
3.6.2.1 Marine Tankers
Marine tankers transport over 90 percent of the gaso-
line and crude oil shipped in the Houston-Galveston area. They
vary in capacity from around 20,000 Dead Weight Tons (DWT) to
over 75,000 DWT. The Ship Channel's depth of 40 feet limits
their size due to their draft requirements. No data was obtained
on whether or not any tankers are lightered in the Houston-
Galveston area. Lightering could effect emissions because cargo
is transferred from a large tanker to a barge or smaller tanker
outside of the port for transport into terminals in the area.
The number of cargo tanks and their capacities are de-
pendent upon the individual ship design, size and service. The
smaller tankers may have as many tanks as those ships with capaci-
ties 2 to 3 times greater. Clean product tankers generally have
from 24 to 33 tanks on board, but crude oil tankers have about
half that many, generally. The tank capacities may range from
5,000 to 10,000 barrels for the smaller tankers and from 5,000 to
over 30,000 barrels for the larger tankers. Tank depths vary from
about 40 feet for the smaller tankers to over 55 feet for the
larger ones.
-76-
-------
3.6.2.2 Intercoastal Barge
The intercoastal barge is a relatively small vessel
used for transferring gasoline and crude oil over the inland
waterways. Their capacity is usually around 20,000 barrels and
they typically have 8 to 10 cargo compartments of approximately
equal size. Barges are commonly transported in pairs by tug
boats. Tank depths for barges generally vary from 10 to 12 feet.
3.6.2.3 Ocean Barge
The ocean barge is a large vessel designed to be inte-
grated with a tug for power. The ocean barges are few in number.
Their DWT capacity is in the range of that for average size tankers,
but they typically have fewer cargo tanks. Tank depths are usually
shallower than that for tankers, ranging from 30 to 40 feet.
3.6.3 Vessels Servicing Houston-Calveston Marine Terminals
Most of the marine terminals in the Houston-Calves ton
area utilize ships, intercoastal barges, and ocean barges for their
marine shipments of gasoline. Crude oil is received primarily by
ship with a small amount entering by barge. Very little, if any,
gasoline is imported by marine vessel into the Houston-Calveston
area. The crude oil which is loaded in the area is transported
in tankers.
Appendix I contains information supplied by owners of the
larger marine terminals in the Houston-Calveston area concerning
the marine tankers which visited their docks to transfer crude oil
or gasoline. The responses were not consistent with respect to
the type of information presented. Data on vessel names, DWT,
ownership, service, quantity loaded in 1975, number of cargo tanks,
and number of visits in 1975 were obtained in different responses.
-77-
-------
Very little information was obtained on the specific barges that
transferred gasoline and crude oil in 1975 in the Houston-
Calves ton area.
The following discussions summarize by company the
data presented in Appendix I.
it
Exxon Terminal
Almost all of the crude oil unloaded at the Exxon Bay-
town marine terminal was transported by marine tanker. Less than
one percent of the crude was imported by barge. Forty-four of
the fifty ships which transported crude oil into the terminal
were foreign, four were Exxon ships, and two had unknown ower-
ship. Their capacities ranged from 20,000 DWT to 78,000 DWT.
Average- size was about 50,000 DWT (Ref. 15).
Of the ships loading gasoline at the Exxon terminal,
ten were owned by Exxon, and twenty-two were owned by some other
U.S. shipping company. The Exxon ships loaded about 74 percent
of the gasoline, while the other tankers loaded about 7 percent.
Barges loaded the remaining 19 percent. All Exxon ships were in
mul tiple service.
Four Exxon-owned ships were used for transporting the
crude oil that was loaded at the Baytown terminal.
No data was obtained on the specific barges that trans-
ferred crude oil and gasoline at the Baytown refinery.
Shell Terminal
The majority of the gasoline which was loaded onto ships
in 1975 was transported by U.S. tankers under long-term charter
to Shell Oil Company. Because Shell Oil is not a U.S. company,
-78-
-------
it cannot own the ships it uses for transporting products from
its Deer Park facility. Under the Jones Act, vessels engaged in
transporting goods between U.S. ports must be under U.S. registry.
A vessel cannot be under U.W. registry unless it is owned by a
U.S. citizen or company. All of the tankers transporting gasoline
from the Deer Park terminal carry other products such as fuel oil
and heating .oil as well. The sizes of these tankers range from
20,000 DWT to about 45,000 DWT (Rev. 23).
Very little information was presented on the ships
which unload crude oil at Shell's refinery in Deer Park.
AMOCO Terminal
In 1975 fifteen tankers loaded the gasoline transported
from AMOCO's Texas City refinery. Three ships were owned by AMOCO,
and twelve were owned by other U.S. corporations. The bulk of the
gasoline transported by tanker was handled by the three AMOCO ships.
Also, three ocean-going barges and sixteen intercoastal barges
loaded gasoline at the AMOCO terminal in 1975. None of the barges
were owned by AMOCO.
A total of thirty-one ships called at AMOCO's terminal
to unload crude oil in 1975. Most of them made more than one
trip to the terminal during the year. Twenty-seven tankers were
foreign-owned, and the other four were U.S. company owned.
ARCO Terminal
Eight tankers were used to load gasoline at ARCO's ter-
minal in 1975. Three of these ships were owned or company chartered
by ARCO, two were time or trip chartered, and the remaining three
were not controlled at all by ARCO. Also, nine intercoastal barges
loaded gasoline at the terminal in 1975. Only one was owned by ARCO.
-79-
-------
A total of twenty-four tankers were used to transport
crude oil to the refinery terminal in 1975. Three of these were
owned or company chartered by ARCO (Ref. 6).
3.6.4 Hydrocarbon Emissions From a Gasoline Tanker Cruise
»
This section presents estimates of the hydrocarbons
emitted from a hypothetical cruise of a tanker transporting
gasoline. Estimates for the loading, unloading, and transit
losses are calculated. The selection of conditions for the
cruise, material transferred, and size of the tanker is based
upon operations observed by Radian to be typical of the marine
transport industry in the Houston-Calveston area.
The length of time for a round trip cruise was chosen
as twelve days: 1 day (24 hours) for loading, 5 days for a one
way trip, 1 day for unloading, and 5 days for the return cruise.
This length of time is typical for a tanker travelling to the
East Coast from Houston, Texas. A 50,000 DWT capacity ship was
chosen as typical. The ship is to be transporting 350,000 barrels
of motor gasoline with an RVP of 10.0 as a typical cargo.
Loading Loss
For this case the emission factor used to calculate the
loading loss is 1.2 pounds per 1000 gallons of gasoline loaded.
The factor is taken from Table 4.2-2 in this report and its origin
is discussed in Section 4.2. It represents the statistical average
factor for a ship loading gasoline in the Houston-Calveston area.
Based on the quantity loaded, the loading emission is 18,000 pounds
Transit Loss
The ship is in transit with a full cargo load for a
period of five days. Using the estimate in the EPA's Compilation
-80-
-------
of Air Pollution Emission Factors for marine vessel transit loss,
the emission can be estimated. The factor is 3.6 pounds of hydro-
carbon per week per 1000 gallons transported. Based on this factor
the transit loss is 53,000 pounds. This factor originated from
API Bulletin Number 2514 published in 1959 and it is based on a
limited amount of data. API states that the transit loss is
significant but more data is needed to verify this.
Unloading Loss
Following the discharge of its gasoline cargo, the
ship must take on ballast to make it seaworthy. A typical quantity
of ballast taken on for ships is 25 percent of its cargo capacity.
The ballasting operation displaces the vapors which have evaporated
into each tank during cargo unloading. Preliminary dat* reported
in Section 4.2 on ballasting losses show the emission factor for
crude oil ships to be in the range of 1 to 2 pounds per thousand
gallons of ballast. Using a conservative factor of 1.0 pounds per
thousand gallons of ballast for this gasoline ship's ballasting
operation, the unloading loss is 4,000 pounds of hydrocarbon.
Return Transit Loss
On the return trip several operations may occur which
could cause significant hydrocarbon emissions. These include:
1) leaving cargo tank hatches open to the atmosphere; 2) reballast-
ing, which involves discharging the dirty ballast and taking on
clean ballast; 3) tank "breathing losses"; and 4) tank cleaning
operations such as gas-freeing. No data is available to allow
the potential emissions to be calculated, but they could be
significant.
Conclusion
Although limited information is available to confirm the
emissions discussed above, a single gasoline ship cruise may emit
-81-
-------
over 75,000 pounds of hydrocarbons. The major contributor to
this total is the transit loss with the return transit loss being
potentially significant.
3.6.5 Analysis of Tank Arrival Conditions for Vessels Loading
Gasoline and Crude Oil in the Houston-Calveston Area
In Section 3.6.1 an Exxon study was discussed concerning
the effect of vessel cruise history upon arrival hydrocarbon con-
centration. The result of the study is presented in Table 3.6-2.
The results show that all of the intercoastal barges
which load gasoline arrive with tanks that are empty and undis-
turbed. They also show that 45 percent of the tanker volume
loading gasoline enters cleaned, 10 percent ballasted, and 45
percent empty and undisturbed. For ships loading crude oil, 25
percent of the tanker volume enters clean and 75 percent enters
empty and undisturbed. The Exxon results for crude oil loading
onto ships are not based on as much information as are the gasoline
loading results.
While this breakdown should be a reasonable represen-
tation of Exxon loading conditions it is not necessarily the case
for all the other refineries in the Houston-Calveston area which
load gasoline. At least two exceptions are Shell Oil in Deer Park
and ARCO in Houston. Not enough data is available on the other
companies in the area to draw firm conclusions on arrival conditions
Shell Oil is different in that its ships often arrive with all tanks
cleaned so that if a last minute product loading change is made, the
ship will be ready to accept any cargo. ARCO on the other hand,
usually loads gasoline into ship holds which are essentially vapor
free because they previously carried a low-volatile product.
ARCO loads only small quantities of gasoline by tanker, thus
back-to-back gasoline service by a tanker is rare.
-82-
-------
TABLE 3.6-2
EMISSION FACTORS FOR GASOLINE AND CRUDE OIL
LOADING BY TANK ARRIVAL CONDITION
Vessel Type
Arrival Condition
Percent of Vessel Volumes
Observed with Tank
Arrival Condition
i
00
LO
Intercoastal
Barge-Gasoline
TVP=6.0psia
Ship-Gasoline
TVP=6.0psia
Empty and Undisturbed
Cleaned
Ballasted
Empty and Undisturbed
100%
45
10
45
Ship-Crude Oil
TVP=4.5psia
Cleaned
Ballasted
Empty and Undisturbed
25
0
75
-------
Therefore, it appears as though the results for the
Exxon study when applied to the entire Houston-Galveston area
may slightly overestimate the emissions from gasoline loading.
3.7 Marine Terminals Transferring Crude Oil and Gasoline
In the Metropolitan Los Angeles Area
This section discusses the types of marine terminals
located in the Metropolitan Los Angeles AQCR which transfer
crude oil and/or gasoline. The objectives of the section are to
present sufficient information on the marine terminals to assess
their impact on hydrocarbon emissions in the Southern California
.area, and to compare the marine terminal operations in the
Houston-Galveston area to the Southern California operations.
The objectives were accomplished by gathering information on
Southern California operations from refulatory agency contacts,
literature, and on-site visits. The results of the Southern
California study are presented in the following sections.
3.7.1 Background Information On Marine Terminals Transferring
Crude Oil and Gasoline in the Southern California Area
There are eighteen identified marine terminals in the
Metropolitan Los Angeles Intrastate AQCR which transferred crude
oil and/or motor gasoline in 1975 (Rev. 7). Of the nine terminals
which load or unload motor gasoline all but one of them are located
near Los Angeles. Table 3.7-1 presents a breakdown of the marine
terminals located in the three counties .in the Los Angeles AQCR
and the number of terminals transferring crude oil and/or gasoline.
Table 3.7-1 shows that several of the terminals in the Los Angeles
area load and unload both gasoline and crude oil. There are a
total of 12 terminals in Los Angeles County, 4 in Santa Barbara
County, and 2 in Ventura County which transfer crude oil and/or
gasoline. The following sections discuss the similarities and
differences in the shoreside equipment and transfer procedures
-84-
-------
TABLE 3.7-1
i
CO
Ul
County
Los Angeles
Santa Barbara
Ventura
MARINE TERMINALS TRANSFERRING CRUDE OIL OR
GASOLINE IN THE METROPOLITAN LOS ANGELES AQCR
Number of Terminals Gasoline
Transferring Crude Oil No. of Terminals No. of Terminals
and/or Gasoline Loading Gasoline Unloading Gasoline
12 8 8
4 0 1
2 00
Crude Oil
No. of Terminals Number of
Loading Crude Oil Unloading
5 12
3 0
2 0
Terminals
Crude Oil
Source: Ref. 7,17,22.
-------
between Los Angeles AQCR terminals and Houston-Galveston AQCR
terminals.
3.7.1.1 Shoreside Equipment And Transfer Procedures-Gasoline
The equipment and transfer procedures used for gasoline
loading and unloading at terminals in the LA AQCR are similar to
those found in the Houston-Galveston AQCR. Inspection by Radian
of three refinery associated terminals provided most of the
information for this section. The facilities for transferring
gasoline at the terminals inspected is similar to the equipment
typically found at Houston terminals. The docks are constructed
of concrete, and two of the three terminals inspected were equipped
with metal hydraulic loading arms.
Gasoline loading rates at the LA terminals are approxi-
mately the same as for the Houston terminals. The average loading
rate for gasoline at two of the terminals toured was about 10,000
bph. For the other terminal the average gasoline loading rate
was 3,000 bph. No information was obtained for the unloading
rate of gasoline. However, the unloading rate for gasoline from
tankers is dependent primarily upon the ship cargo pump capacity,
and the capacity of the ship to shore connector. Larger tankers
with high capacity pumps are capable of unloading at average
rates of up to 40,000 bph. Smaller tankers (20,000-30,000 DWT)
may unload at average rates of only 10,000 to 15,000 bph. Speci-
fic information on individual ship to shore connector capacities
for the LA terminals was not obtained.
-86-
-------
3.7.1.2 Shoreside Equipment and Transfer Procedures -
Crude Oil
Important differences exist between the Los Angeles AQCR
and the Houston AQCR in relation to the loading and unloading of
crude oil at marine terminals. There are basically two types of
terminals transferring crude oil in-the LA AQCR. The onshore
terminal whose equipment and operational procedures are similar to
those observed in the Houston AQCR. The other is the offshore
terminal which is significantly different in terms of equipment,
operational procedures, and emission control alternatives.
The onshore terminals which transfer crude oil are operated
much the same as the ones in the Houston AQCR. Several of the
terminals are dedicated solely to transferring crude oil. The
terminals which are proposed to handle the Alaskan crude oil are
discussed in a separate section, Section 3.7.4.
The offshore terminals in the LA AQCR evolved because of
the deep waters which exist close to land. This affords ships easy
access to much of the coast in the LA AQCR.
The terminals consist of four or five buoys firmly anchored
to the ocean floor and an underwater pipeline from a shore storage
facility to the buoy area. With the aid of a tugboat the tanker will
move to the anchored bouys and then hoist the flexible free end of
the transfer line up to the ship deck where it is connected to the
cargo tank manifolding. Figure 3.7-1 illustrates a typical offshore
terminal. After the transfer line is connected, the crude oil or
gasoline is either loaded onto or unloaded from the ship.
-87-
-------
SHORE
TRANSFER
LINE
ANCHOR BUOY
TANKER
ANCHOR BUOY
ANCHOR BUOY
ANCHOR BUOY
ANCHOR BUOY
STORAGE
TRANSFER LINE
FIGURE 3.7-1. OFFSHORE TERMINAL
-88-
-------
Because of the lack of a platform or support structure,
the offshore terminal poses a problem in its potential for hydro-
carbon emission control by vapor recovery systems. Insufficient
space and safety considerations hamper installation of a system
onboard the ship itself. Instead, emission control alternatives
would be procedure or process modification types. For loading
operations these may include prior tank cleaning, slow initial
fill rate, and light loading of the cargo. For unloading and
subsequent ballasting, an alternative for ships which must ballast
would be requiring the vessel to take on minimum ballast at the
terminal and finish ballasting when farther out at sea. Alterna-
tives to the above options would be the construction of a plat-
form to support control equipment or the construction of pipelines
to markets, either of which would be very expensive.
3.7.2 Shipside Equipment and Transfer Procedures For The
Los Angeles"AQCR
General information was obtained on the shipside equip-
ment and transfer procedures for vessels transferring crude oil
and gasoline in the Los Angeles AQCR. The information indicated that,
for the most part, the equipment and transfer procedures for the
vessels servicing this area are similar to that seen in Houston-
Galveston. There are two exceptions. First, the information
suggests that very little gasoline and no crude oil are transported
via barge. Presumably this is because of the lack of navigable
intercoastal canals and rivers in California. The second exception
noted was the closed loading technique used by all the ships owned
by one oil company. Evidently, company policy prevents any
vapors from being displaced at deck levels during loading of a
liquid cargo. All tanks are close loaded with the displaced vapors
being vented at a mast height roughly forty to fifty feet above
deck level. Cargo tank level minitors are used during loading.
-89-
-------
No Visual inspection of the tank level occurs throughout loading
unless there is a problem.
Some information on cruise history of tankers loading
gasoline in the Los Angeles AQCR was obtained. It indicated
that the tankers usually transport the gasoline to the Washington-
Oregon region and occasionally to the Sari Francisco Bay Area.
This is a shorter distance than that travelled by the tankers
leaving the H-G AQCR with gasoline. Their destination is often
the East Coast which may be two to three times further than the
distance from LA to the Washington or Oregon ports. Also, the
available information indicated that tankers load gasoline less
frequently at LA terminals than at H-G terminals. This is a result
of the larger quantity of gasoline (approximately 10 times larger)
loaded in the H-G area than in the LA area. Thus, tankers in the
LA AQCR loading gasoline are less likely to carry back-to-back
gasoline loads, an occurence often seen in the H-G area at the
more active terminals. This difference could effectively reduce
hydrocarbon emissions from the loading of gasoline into tankers
in the LA AQCR, assuming low volatility products such as fuel
oil or heating oil are carried on the trips the tankers make
between gasoline service. See Table 4.2-2 and the discussion in
Section 4.1.4 on'the effects of cruise history on emissions.
3.7.3 Quantities of Crude Oil and Gasoline Transferred In
The Los Angeles AQCR
Data were obtained on the quantity of crude oil and
gasoline transferred at marine terminals in the Los Angeles AQCR
for the year 1975. The information was obtained from the Southern
California APCD for Los Angeles County, from the Ventura County
APCD for Ventura County, and from the Santa Barbara County APCD
for Santa Barbara County (Ref. 7,17,22).
-90-
-------
Tables 3.7-2 through 3.7-5 summarize estimates of the
quantities of gasoline and crude oil loaded and unloaded in the
Los Angeles AQCR. Table 3.7-2 presents the quantity of gasoline
loaded by county. It shows that a total of 8,912,000 barrels of
motor gasoline were loaded onto marine vessels in the LA AQCR in
1975. Table 3.7-3.shows that 15,650,000 barrels of motor gasoline
were unloaded at marine terminals in the LA AQCR. Tables 3.7-4
shows that 17,950,000 barrels of crude oil were loaded onto ships
in 1975 in the LA AQCR and Table 3.7-5 shows that 177,318,000
barrels of crude oil were unloaded from ships in 1975.
Information on the projections of the Alaskan crude oil
unloaded from ships in the LA AQCR is discussed in the following
section.
3.7.4 Projected Unloading of Alaskan Crude Oil In The
Los Angeles AQCR
This section presents information on the variables
affecting the hydrocarbon emissions from the potential unloading
of Alaskan North Slope crude oil in the Metropolitan Los Angeles
Intrastate AQCR in the future. Data were obtained on the port
site, types and sizes of tankers, oil delivery quantities and
frequencies, and the North slope oil characteristics. The pri-
mary source of this information was the report Air Quality Analysis
of the Unloading of Alaskan Crude Oil At California Ports
(Nov. 1976), by Pacific Environmental Services, Inc. (Ref. 9).
This section addresses only the areas mentioned above and does
not examine the emissions potential from the support facilities
such as tug assistance, crude oil storage tanks, and the ship's
boilers.
-91-
-------
TABLE 3.7-2
QUANTITY OF GASOLINE LOADED AT
MARINE TERMINALS IN THE
LOS ANGELES AQCR
County
Los Angeles
Ventura
Santa Barbara
Total
Year
1975
1975
1975
Quantity
8,916
0
__0_
8,916
(103 bbls)
Source: Ref. 7,17,22
TABLE 3.7-3
QUANTITY OF GASOLINE UNLOADED AT
MARINE TERMINALS IN THE
LOS ANGELES AQCR
County
Los Angeles
Ventura
Santa Barbara
Total
Year
1975
1975
1975
Quantity
14,950
0
700
15,650
(103 bbls)
Source: Ref. 7,17,22
-92-
-------
TABLE 3.7-4
QUANTITY OF CRUDE OIL LOADED AT
MARINE TERMINALS IN THE
LOS ANGELES AQCR
County
Los Angeles \
Ventura
Santa Barbara
Total
Year
1975
1975
1975
Quantity
7,211
6,804
3,935
17,950
(103 bbls)
Source: Ref. 7,17,22
TABLE 3.7-5
QUANTITY OF CRUDE OIL UNLOADED AT
MARINE TERMINALS IN THE
LOS ANGELES AQCR
County
Los Angeles
Ventura
Santa Barbara
Total'
Year
1975
1975
1975
Quantity
177,318
0
0
177,318
riO3 bbls)
Source: Ref. 7,17,22
-93-
-------
3.7.4.1 Port Site for Unloading Alaskan Crude Oil In The LA AQCR
The crude oil produced in North Slope fields in Alaska
will probably be transported to the Continental U.S. via ocean-
going oil tankers. Numerous projects have been proposed to handle
the tankers and to pipeline the crude to the existing refining
capacity in the U.S. West Coast refining demand for the Alaskan
crude is estimated to account for 400,000 to 800,000 barrels per
day (bpd). In 1978-1980 when Prudhoe Bay production reaches full
capacity (1.2 million bpd) there is expected to be a surplus of
400,000 to 800,000 bpd of Alaskan crude (Ref. 16).
Standard Oil Company (Ohio) proposes to combine about
800 miles of existing natural-gas pipeline with newly constructed
pipeline to move Alaskan crude from Long Beach, California to
Midland, Texas. This project could deliver some 500,000 bpd of
crude oil to Texas where it could be pipelined through existing
networks with direct access to almost two-thirds of the nation's
refining capacity (Ref. 16). The receiving port SOHIO plans to
build in Long Beach will be a three-berth terminal. The terminal
will handle tankers of up to 165,000 dwt and 55 ft. draft (Ref. 30)
Several oil refineries in the Los Angeles AQCR are
planning to bring in Alaskan crude oil via tankers. These refin-
eries are owned by ARCO, Shell, and Mobil. The crude oil will be
used as refinery feedstock. No information on the size of these
operations is known, but considerably lesser quantities of Alaskan
oil will be received at the three refinery terminals combined than
at the SOHIO terminal.
3.7.4.2 Types and Sizes of Tankers Delivering Alaskan Crude
Oil from Valdez to The 'Los Angeles AQCR
Sufficient tanker tonnage exists to transport Alaskan
oil from Valdez Alaska to the lower West Coast states. The total
. -94-
-------
tonnage of the ships is in the range of 2,696,000 to 3,166,000 DWT
according to the Pacific Environmental Services report. The final
number depends on Exxon's plans to utilize its tanker fleet and
the possible construction of two 165,000 DWT tankers for SOHIO.
ARCO also is planning to build two 150,000 DWT tankers but is
uncertain as to delivery date or usage (Ref. 9).
Table 3.7-6 summarizes information on the tankers which
are projected to be used for transporting Alaskan crude. It
shows that all tankers for which information was obtained have
some amount of segregated ballast. Reportedly, a quantity of
ballast of from 15 to 20 percent of tonnage capacity is generally
sufficient to move a tanker out of port. Fully segregated ballast
(approximately 35 percent of cargo tonnage) is planned for the
ships currently under construction. The possibility exists that
some ships without segregated ballast may be used to transport
Alaskan crude to the Long Beach port facility. The quantity of
crude that they transport is not expected to be significant
(Ref. 9).
Most of the newer ships, as can be seen in Table 3.7-6,
will have inert gas systems for deoxygenation of their cargo
tanks. The inerting system is used to provide an inert gas
blanket over the crude oil during transport from Valdez and to
purge the empty cargo tank following discharge of the crude oil.
Information available on the SOHIO tankers which are to be equipped
with inerting systems indicates that it will require a minimum of
approximately 8 hours to provide a volume of gas from the inerting
system equivalent to the total volume of the cargo holds (Ref. 9).
Purging of empty cargo tanks results in a substantial
emission of hydrocarbons. However, SOHIO indicates that there
will be no purging of the cargo holds of their ships equipped with
inerting systems within the Southern California Air Basin (Rev. 9).
-95-
-------
TABLE 3.7-6
PROJECTED ALASKAN CRUDE
Owner
ARCO
CHEVRON
EXXON
MOBIL
SHELL
SOHIO
Dead Weight
Tonnage
120,000
150,000
70,000
70,000
76,000
130,000
188,000
80,000
120,000
165,000
Number of Anticipated Unloading
Ships Rate (bbls/hr)
3
2
3
2
3
1
2
2
2
6
67,000
Unknown
Unknown
25,000
Unknown
70,000
106,000
50,900
76,400
102,000
OIL TANKER FLEET
Segregated Ballast
(% of DWT)
20
Unknown
Unknown
20
Unknown
20
35
15-18
35
35.6
Inert Gas
System
No
Yes
No
No
No
Yes
Yes
No
Yes
Yes
Availability
Now
1977-79
Now
Now
Now
Now
1977-78
Now
1976-77
1977-78
Source: Ref. 9
-------
3.7.4.3 Projected Quantities of Alaskan Crude Oil To Be
Unloaded In The Los Angeles AQCR
The SOHIO project will be capable of unloading up to
700,000 barrels per day of Alaskan crude at the Long Beach termi-
nal by mid-1978 if permit applications are approved by the second
quarter of 1977. A pipeline from Los Angeles to Midland, Texas
would be capable of handling 500,000 barrels per day of crude.
The extra 200,000 barrels per day would allow for future deliver-
ies in the Los Angeles area (Ref. 27).
The SOHIO project is presently the only proposal for
unloading large quantities of Alaskan crude oil in the LA AQCR.
Several refineries in the AQCR have tankers presently available
or under construction for moving crude from Valdez to Los Angeles
for processing. The refineries are ARCO, Shell, and Mobil. Infor-
mation on the quantity of Alaskan crude to be unloaded at these
refineries was not obtained, but the total quantity will pro-
bably be less than that for the SOHIO project.
SOHIO states that some 400 tanker visits are expected
annually at the Long Beach terminal. Presumably this is at the
maximum delivery rate of 700,000 barrels per day.
3.7.4.4 Characteristics of Alaskan Crude Oil
Information on the properties of the Alaskan crude oil
to be transported to.the LA AQCR was obtained from the Pacific
Environmental Services report referenced earlier (Ref. 9). The
properties are given below:
-97-
-------
Sulfur Content:
Density:
Specific Gravity:
Vapor Pressure:
Temperature of
Delivered Crude:
Molecular Weight
of Crude Oil
1 percent by weight
312.7 Ib/barrel at 60eF
0.893
Reported estimates of the vapor
pressure of North Slope crude
range between 7.4 psia as a
winter minimum in Southern
California to 9.6 as a summer
maximum.
It is reported that the temper-
ature of the crude oil in the
tankers after loading in Valdez
during the first year of operation
will be 45 to 60°F. Later, the
temperature will rise to 80-97°F.
For maximum flow through the pipe-
line the oil will be heated to
reduce its viscosity, and the
temperature of the crude will
approach 140°F at Valdez.
Based on data supplied by SOHIO
the average molecular weight of
the vapor in equilibrium with
North Slope crude is 51. This
assumes that the light ends are
not lost during storage or
transit. Table 3.7-8 presents
the composition of the vapor in
equilibrium with North Slope
crude oil.
-98-
-------
TABLE 3.7-8.
COMPOSITION OF VAPOR IN EQUILIBRIUM
WITH NORTH SLOPE CRUDE OIL
Carbon Dioxide
Water
Methane
Ethane
Propane
i- Butane
n-Butane
i-Pentane
n-Pentane
Hexane Plus (As Cc)
b
MOL %
3.90
5.30
9.90
12.60
20.10
8.60
19.90
6.60
7.60
5.50
100.00
% HC
•_
--
10.9
13.9
22.1
9.5
21.9
7.3
8.4
6.1
100.1
Molecular
WT.
„_
-_
16
30
44
58
58
72
72
-90
% x M.W./100
— _
__
1.74
4.17
9.72
5.51
12.70
5.26
6.05
5.49
50.64
Avg. M.W.
Source: Ref. 9
-------
3.7.5 Similarities And Differences In Marine Terminals
Located In The Los Angeles AQCR And The Hbuston-
Galveston AQCR
#
This section examines the similarities and differences
in marine terminals which transfer or which are projected to trans-
fer crude oil and/or gasoline in the Los Angeles AQCR and the
Houston-Calveston AQCR. The comparisons are based upon data Radian
obtained in the course of completing the Houston-Calveston phase
and the Los Angeles phase of this project. A broad data base was
established by Radian on the Houston-Calveston marine terminals.
Because of time limitations, the Los Angeles study was not as
detailed.
This section also points out those areas where additional
information on the marine terminals in the Los Angeles AQCR would
be valuable for assessing their impact on hydrocarbon emissions.
Similarities and differences are discussed first for the marine
terminals in Los Angeles County, then for Ventura County, and
finally for Santa Barbara County. Areas requiring additional
information are pointed out within each of the three discussions.
3.7.5.1 Los Angeles County
Los Angeles County marine terminal operations.include
loading and unloading of both crude oil and gasoline from tankers.
There are several differences existing between Los Angeles County
and the Houston-Calveston area with respect to marine terminals:
little or no usage of barges to transport gasoline
or crude oil at Los Angeles County terminals
the unloading of gasoline at Los Angeles County
terminals (15,650 x 103 barrels in 1975) versus
no unloading of gasoline in the Houston-Calveston
area
-100-
-------
a smaller quantity of gasoline loaded at Los
Angeles County terminals (8,916 x 103 barrels
in 1975) versus the amount of gasoline loaded
in the Houston-Calveston area (82,500 x 103
barrels in 1975)
the- use of offshore marine terminals in Los
Angeles County
differences in the cruise histories between
ships loading gasoline on the LA AQCR and
ships loading gasoline on the H-G AQCR.
There are some similarities between marine terminals in the
two areas:
almost the same quantity of crude oil
unloaded at Los Angeles County terminals
(177,320 x 103 barrels in 1975) as in the
Houston-Galveston area (160,000 x 103 bar-
rels in 1975)
similar equipment and techniques used for
loading and unloading crude oil and gasoline
except for the closed loading technique prac-
ticed by one company.
• no inerting of cargo tanks at either port,
however, if Alaskan crude is unloaded at
Long Beach, there will be tankers entering
the LA AQCR with inerting capabilities.
Several areas were noted where additional information
is needed to accurately determine the impact of gasoline and
crude oil transfers on hydrocarbon emissions at marine terminals
in Los Angeles County:
-101-
-------
analysis of the cruise history of those ships
which load crude oil and gasoline in LA County
analysis of the composition of the hydrocarbon
vapors vented from cargo tanks during loading
or ballasting operations.
3.7.5.2 Ventura County
There are two marine terminals in Ventura County.
They both are used almost solely for loading crude oil, and
both are located offshore. Combined they loaded at a total of
6,800 x 103 barrels of crude oil onto tankers in 1975.
There are very few similiarities between these two
terminals and the terminal in the Houston-Calveston area which
loads crude oil. The Ventura and Houston-Calveston terminals
loaded roughly the same quantities of crude in 1975. However,
the terminal in the Houston-Calveston area loads at a higher rate
(10,000 bph minimum, 32,500 bph maximum) versus about 7,500 bph
for the ventura terminals. As mentioned earlier, a unique hydro-
carbon emission control problem exists in Ventura because of their
location offshore.
Radian attended a hydrocarbon emission sampling run
conducted by Chevron Research personnel onboard a tanker loading
crude oil at the more active of the two Ventura terminals. It
was learned that usually a tanker only picks up a partial load
of crude oil at the terminal. Also, there are wide variations
in vapor pressure of the crude loaded from one tanker visit to
the next. This is because there are several crudes produced
from different fields in the area having different vapor
pressures.
-102-
-------
Further information on the terminals and their hydro-
carbon emissions will result when the Western Oil and Gas
Association sponsored emission testing program conducted by
Chevron Research is completed. This program is comprehensive
and covers all aspects of the hydrocarbon emission problem
occuring at Ventura.
3.7.5.3 Santa Barbara County
There are four marine terminals located in Santa
Barbara County. Three of them load crude oil onto tankers, and
one unloads gasoline from tankers. The ships which unload gaso-
line reportedly do not ballast following discharge of the cargo
and, thus, do not vent hydrocarbons. The terminals loading
crude oil are located offshore. In 1975 they combined to load
a total of 700,000 barrels of crude oil.
Projects are proposed which could increase the produc-
tion of crude oil in and near Santa Barbara waters. Information
indicates that as much as 120,000 barrels per day or approxima-
tely 44 million barrels per year of crude oil may be produced
and then loaded at offshore terminals near Santa Barbara County.
Very little information was available on the terminals
or the crude oil being loaded in this area. To accurately
determine the hydrocarbon emission impact on Santa Barbara County,
more information is necessary on the existing and proposed marine
terminals:
location of each terminal and quantity of
crude loaded or projected to be loaded there
-103-
-------
analysis of the cruise history of the ships
being used to load the crude
analysis of the composition of the vapors
being vented from current loading operations
analysis of the composition of the vapor in
equilibrium with the crude oil to be loaded
at the proposed terminals.
-104-
-------
4.0 MARINE TERMINAL EMISSIONS
The source and mechanism of hydrocarbon emissions from
marine terminal transfers of gasoline and crude oil are well un-
derstood. However, so many factors affect the magnitude of gaso-
line and crude oil transfer emissions that it becomes a very
involved task to develop adequate emission factors or correlations
for estimating the hydrocarbon emissions. Section 4.1 presents
the general nature and characteristics of marine transfer emis-
sions. Source testing data collected by Radian and by the petro-
leum industry concerning marine transfer emissions are presented
in Section 4.2.
4.1 Emission Characteristics
4.1.1 Source and Mechanism
Hydrocarbon emissions are generated at marine terminals
when volatile hydrocarbon products are either loaded onto or
unloaded from ships and barges.
Loading Emissions
Loading emissions are attributable to the displacement
to the atmosphere of hydrocarbon vapors residing in empty vessel
tanks by volatile hydrocarbon liquids being loaded into the
vessel tanks. Loading emissions can be separated into the
arrival component and the generated component. The arrival com-
ponent of loading emissions consists of hydrocarbon vapors left
in the empty cargo tanks from previous cargoes. The generated
component of loading emissions consists of hydrocarbon vapors
-105-
-------
generated in the cargo tanks as hydrocarbon liquids are being
loaded.
The arrival component of loading emissions is directly
dependent on the true vapor pressure of the previous cargo, the
unloading rate of the previous cargo, and the cruise history of
the cargo tank on the return voyage. The cruise history of a
cargo tank may include heel washing, ballasting, butterworthing,
vapor freeing, or no action at all. Temperature gradients,
vessel motion, and long elapse times contribute to the well
mixing of empty cargo tanks, resulting in almost uniform vapor
concentrations in the arrival component. The arrival component
for vessels loading gasoline characteristically range from 0 vol %
to 20 vol 70 hydrocarbons, but can exceed 50 vol 70.
The generated component of loading emissions is -pro-
duced by the evaporation of hydrocarbon liquid being loaded into
the vessel tank. The quantity of hydrocarbons evaporated is
dependent on both the true vapor pressure of the hydrocarbons
and the loading practices. The loading practice which has the
greatest impact on the generated component is the loading rate.
An example profile of gasoline concentrations in a
vessel tank during loading is presented in Figure 4.1-1. As
indicated in the figure, the hydrocarbons present throughout
most of the vessel tank vapor space are contributed by the arrival
vapor component and the concentration is almost uniform. There
is a sharp rise in hydrocarbon vapor concentration just above
the liquid surface. This is the generated component. The gener-
ated component, also called a vapor blanket, is attributable to
evaporation of the hydrocarbon liquid.
-106-
-------
From Figure 4.1-1 it is apparent that for large vessels
with 55 foot ullages, the average hydrocarbon concentration of
vapors vented during loading operations is primarily dependent
on the arrival component. For smaller vessels such as barges
with 12 foot ullages the average hydrocarbon concentration in
the vented loading vapors is dependent on both the generated com-
ponent and the arrival component.
Unloading Emissions
Unloading emissions are hydrocarbon emissions displaced
during ballasting operations at the unloading dock subsequent
to unloading a volatile hydrocarbon liquid such as gasoline
or crude oil. During the unloading of a volatile hydrocarbon
liquid, air drawn into the emptying tank absorbs hydrocarbons
evaporating from the liquid surface. The greater part of the
hydrocarbon vapors normally lies along the liquid surface in a
vapor blanket. However, throughout the unloading operation,
hydrocarbon liquid clinging to the vessel walls will continue
to evaporate and to contribute to the hydrocarbon concentration
in the upper levels of the emptying vessel tank. Figure 4.1-2
presents a hypothetical profile of gasoline vapor concentrations
in a vessel tank during ballasting. If significant temperature
gradients are present, they will create convection currents which
in time will disrupt the vapor blanket and promote a homogeneous
hydrocarbon vapor concentration throughout the tank.
Before sailing, an empty marine vessel must take on
ballast water to maintain trim and stability. Normally, on
vessels that are not fitted with segregated ballast tanks, this
water is pumped into the empty cargo tanks. As ballast water
enters cargo tanks, it displaces the residual hydrocarbon vapors
to the atmosphere generating the so termed, "unloading emissions"
-108-
-------
I
M
O
I
V)
cc
o
Q.
CO
z
o
m
cc
<
o
o
cc
o
FEET ABOVE LIQUID SURFACE
FIGURE 4.1-2 EXAMPLE PROFILE OF GASOLINE BALLASTING EMISSIONS
-------
4.1.2 Effects of Loading Rate
The initial loading rate, bulk loading rate, and final
loading rate all noticeably affect marine loading emissions.
However, the influence of other parameters makes it difficult to
quantify loading rate impacts. This section qualitatively pre-
sents the effects of loading rates on marine loading emissions.
Initial Fill Rate
There is a significant degree of splashing and liquid
turbulence as cargoes are first pumped .into empty vessel tanks.
This splashing and turbulence results in rapid hydrocarbon
evaporation and the formation of a vapor blanket. By reducing
the initial velocity of cargoes entering empty tanks, it is
possible to reduce the turbulence associated with initial tank
filling and, consequently, to reduce the size and concentration
of the vapor blanket. Figure 4.1-3 presents the results of
AMOCO Oil Company tests on the effect of slow loading the first
foot of gasoline cargo tanks. The curves in Figure 4.1-3 indi-
cate a 50% to 607o reduction in vapor blanket size by using slow
initial loading rates. For a clean gasoline cargo tank with a
very small arrival vapor component, a reduction in the vapor
blanket size is very significant.
Bulk Fill Rate
Normally, the vapor blanket profile is established by
the initial filling rate and undergoes very little change through-
out the loading sequence. The bulk loading rate normally has
very little effect on the vapor blanket because of the relatively
slow diffusion rate of hydrocarbon vapors in air. However,
if the bulk loading rate is very slow, or is interrupted by ship
personnel, the vapor blanket profile can change appreciably.
-110-
-------
OT
tr
O
a.
CO
z
o
ffl
DC
O
O
DC
Q
>-
X
40-
3Q.
20-
10-
NORMAL INITIAL FILL RATE. NORMAL FILL RATE FOR BALANCE OF FILL
SLOW INITIAL FILL RATE, NORMAL FILL RATE FOR BALANCE OF FILL
rl
FEET ABOVE LIQUID SURFACE
FIGURE 4.1-3
EFFECT OF INITIAL FILL RATE ON VAPOR BLANKET PROFILE
(AMOCO ILLINOIS - NOV. 6, 1974)
REFERENCE 2
-------
Marine loading emission tests conducted by Atlantic Richfield
indicated that lowering the bulk loading rate of a gasoline
tanker from 3300 barrels per hour to 450 barrels per hour
raised the average hydrocarbon emission rate from 2 vol % to
5.7 vol 7o. The emissions were almost tripled. (Ref. 6)
Final Fill Rate '
As the hydrocarbon level in a marine vessel tank
approaches the tank roof, .the action of vapors flowing towards
the ullage cap vent begins to disrupt the quiescent vapor
blanket. Disruption of the vapor blanket results in noticeably
higher hydrocarbon concentrations in the vented vapor. AMOCO
test results from slow final loadings indicate that, although
not as significant as slow initial loading, slow final loading
can lower the quantity of hydrocarbon emissions from marine
vessel loading of volatile hydrocarbon liquids. (Ref. 4)
Short Loading
Displacement of the high hydrocarbon concentration
generated vapor blanket during the final stages of loading gaso-
line or volatile crudes into a cargo tank causes a significant
part of the total hydrocarbon emissions from that tank. By
stopping the loading of the cargo tank short the vapor blanket
can be partially or totally kept within the tank. The depth of
the blanket usually varies from 6 to 8 feet (see Figure 4.1-1).
Therefore, to keep most of the blanket from being displaced,
loading must be stopped about six feet from deck level.
The effect of short loading on the emissions from
gasoline loading onto a ship can be estimated .from the numbers
in Table 4.2-2. The contribution of the vapor blanket displace-
ment to total tank emissions is approximately the same as the
-112-
-------
emissions from loading into a clean-vapor free tank. In this
case only the vapor generated or evaporated from the liquid
surface during loading (which forms the vapor blanket) contri-
butes to the hydrocarbon emissions. Therefore, from 0.5 to 1.5
pounds of hydrocarbons per thousand gallons loaded represent the
quantity of hydrocarbons emitted when the vapor .blanket is dis-
placed. From Table 4.2-2 it can be seen that emissions theoreti-
cally may be reduced from 10070 to 2070 for tankers loading gasoline,
depending upon concentrations of hydrocarbons in arriving tanks.
There is a consequence of short loading which may ad-
versely affect its ability to reduce emissions from loading gaso-
line on a tanker on successive trips. This problem becomes ap-
parent when one loading/unloading cycle for a tank is examined.
First, assume a tank is short loaded with a six foot space left
from deck level to liquid level. During the time required for
transport of the cargo to 'its destination the space left above
the liquid will likely become saturated with gasoline vapors.
As the cargo is unloaded the vapors become diluted with air to
a lower concentration. When the ship returns for a new load,
the vapor blanket which was not displaced during the previous
loading now manifests itself as the arrival component -of the
emissions and will be displaced as the tank is refilled. There-
fore, unless the vapors are vented from the tank during the
return voyage the hydrocarbon emissions will not be effectively
reduced.
4.1.3 Effects of TVP
The true vapor pressure (TVP) of a hydrocarbon liquid
has a marked impact on the hydrocarbon content of its loading
and unloading emissions. TVP is an indicator of a liquid's
•volatility and is a function of the liquid's Reid Vapor Pressure
-113-
-------
and of the liquid's temperature. Compounds with high TVP exhibit
high evaporation rates and, consequently, contain high hydro-
carbon concentrations in their loading and ballasting vapors.
Section 4.1.6 on the Chemistry of Emissions presents information
on the vapor pressures of gasolines and crude oils during loading
operations. Marine loading tests conducted by Atlantic Richfield
in the winter of 1974-1975 indicated that under the same loading
conditions, a gasoline with a TVP of 6.6 psia generated emissions
with a hydrocarbon content of 2.1 vol %, whereas a gasoline with
a TVP of 8.0 psia generated emissions with a hydrocarbon content
of 2.6 vol %. The TVP of gasoline can easily range from 6 psia
to 10 psia, and the TVP of crude oils can range from 0 psia to
greater than 15 psia. (Ref. 5)
4.1.4 Effects of Cruise History
The cruise history of a marine vessel includes all of
the activities which a cargo tank experiences during the voyage
prior to a loading or unloading operation. Examples of signifi-
cant cruise history activities are ballasting, heel washing,
butterworthing, gasfreeing, and split-deliveries. Cruise history
impacts marine transfer emissions by directly affecting the
arrival vapor component. Barges normally do not have significant
cruise histories because they rarely take on ballast and do not
have the manpower to clean cargo tanks.
Ballasting
Ballasting is the act of partially filling empty cargo
tanks with water to maintain a ship's stability and trim. Figures
4.1-4, 4.1-5, and 4.1-6 present sample hydrocarbon vapor profiles
for empty gasoline cargo tanks prior to ballasting, for ballasted
gasoline cargo tanks, and for gasoline cargo tanks after ballast
-114-
-------
FIGURE 4.1-4 HYDROCARBON PROFILE PRIOR TO BALLASTING
o
H
CC
2 15-
til
Ul
o
z -
o *
o _, '0-
z o
o > •
o
« 5-
o
0
GC
O o
>
Z 0
0
c
H .5-
mm
UJ
0
o 2 I0_
0 _j I0
z >
0 2
03
< s"
o
0
• u
I C
z
g
oc
UJ
0
2 S
S -i '°-
s!
03
c 5 -
(J
0
(T
0 0
z
AN EMPTY TANK
„ ^
y'
/
/
f
1
I
/
I
I
1
f
V
' i • j ' i ' | • i ' J i i i | 1 i i i i i ,;
10 20 30 40 50
ULLAGE (FT)
FIGURE 4.1-5 HYDROCARBON PROFILE OF A BALLASTED TANK
BALLAST WATER SURFACE
^\
\
/^ ' — * BALLAST WATER '-~-
/
/ -
/
/
/
/
.
f
r S
) <0 20 30 40 50
ULLAGE (FT)
FIGURE 4.1-6 HYDROCARBON PROFILE OF AN EMPTY TANK
AFTER BALLAST DISCHARGE
x^
/
/
/ ^
f I • [ 1 • ' J • . • J 1 1 . 1 . • . 1 . 1 )
0 10 20 30 40 50
ULLAGE (FT)
-115-
-------
discharge. (Rev. 2) As Figure 4.1-4 indicates, prior to
ballasting, empty cargo tanks normally contain an almost homo-
geneous concentration of residual hydrocarbon vapors. When
ballast water is taken into the empty tank, Figure 4.105 indi-
cates that hydrocarbon vapors are vented but that the remaining
vapors not displaced retain their original hydrocarbon concentra-
tion. Upon arrival at a loading dock, a ship discharges its
ballast water and draws fresh air into the tank. The fresh air
dilutes the arrival vapor concentration and lowers the effective
arrival vapor concentration by an amount proportional to the
volume of ballast used (Figure 4.1-6). Although ballasting
practices vary quite a bit, individual tanks are ballasted about
8070 and the total vessel is ballasted approximately 4070.
Consequently, ballasting potentially lowers individual tank
arrival component by 8070 and lowers the total ship arrival
component by 4070.
Heel Washing and Butterworthing
The heel of a cargo tank is the residual puddles of
hydrocarbon liquids remaining in cargo tanks after emptying.
These residual liquids will eventually evaporate and contribute
to the arrival component of subsequent vessel-filling vapors.
By washing out this heel with water, AMOCO Oil Company found
that they were able to reduce the hydrocarbon emissions 'from
subsequent filling operations from 5.7 vol 7» to 2.7 vol 70 hydro-
carbons. Butterworthing is the washing down of tank walls in
addition to washing-out tank heels. Butterworthing also reduces
loading emissions by reducing the arrival component concentration,
The hydrocarbon liquids washed from the tanks are stored in a
slops tank for disposal onshore. (Ref. 2)
-116-
-------
Gasfreeing
Heel washing and butterworthirig lower arrival vapor
components by removing residual hydrocarbon liquids from tank
walls and bottoms before they evaporate. However, these two
techniques do not affect hydrocarbon vapors which have already
formed. Marine vessels' can purge the hydrocarbon vapors from
empty and ballasted tanks during the voyage by several gasfreeing
techniques which include air blowing and removal of ullage dome
covers. A combination of tank washing and gasfreeing will
effectively remove the arrival component of loading emissions.
Split-Transfers
Sometimes a tanker will deliver its cargo to two dif-
ferent ports, or will pick up cargoes from two different ports.
These split-transfers increase the quantity of hydrocarbons
emitted from loading and unloading operations. After the partial
delivery of a volatile hydrocarbon product, the partially filled
cargo tanks contain large volumes of air into which hydrocarbons
evaporate from the liquid surface during the balance of the ship's
voyage. These hydrocarbons will be vented to the atmosphere
during the' ballasting operations following the discharge of the
remaining cargo or when the tanker loads a new cargo. In a like
manner, when tankers are engaged in multiple cargo pick-ups, the
vapor space in partially filled cargo tanks increases in hydro-
carbon concentration from evaporation of volatile vapors which
are subsequently vented to the atmosphere when the balance of
the cargo tank is filled.
Previous Cargo
The previous cargo conveyed by a tanker also has a direct
impact on the arrival component of loading emissions. Cargo tanks
-117-
-------
which carried nonvolatile liquids on the previous voyage normally
return with vapor spaces which are essentially vapor free. EXXON
Oil Company tests conducted in Baytown indicated that the arrival
component of empty uncleaned cargo tanks which had previously
conveyed fuel oil ranged from 0 vol 7«, to 1 vol % hydrocarbons.
Cargo tanks with the same .cruise history except for the fact that
they had previously conveyed gasoline, exhibited hydrocarbon con-
centrations in the arrival vapors which ranged from 4 vol % to
30 vol 70 and averaged 7 vol %. (Ref. 12)
4.1.5 Composite Vapor Profiles
The hydrocarbon vapor profile experienced by vapor
control equipment is a composite of the vapor profiles of each
cargo tank venting into the vapor collection system. The com-
posite vapor profile is of a different form than the individual
profiles and is very dependent upon loading procedures. Figures
4.1-7 and 4.1-8 present two example composite vapor profiles.
The profile in Figure 4.1-7 represents the composite vapor profile
of a vessel which is loading each cargo tank separately. The
profile in Figure 4.1-8 represents the composite vapor profile of
a vessel which is loading several tanks simultaneously and tops-
off all of the tanks near the end of the loading period. Most of
the loading operations attended by Radian were more closely
represented by Figure 4.1-8. However, a wide range of loading
procedures are practiced on marine vessels depending on the ship
loading officer and dock situations.
4.1.6 Chemical and Physical Properties
This section presents some of the chemical and physical
properties of gasoline and crude oil vapors which are emitted to
the atmosphere during loading and unloading operations.
-118-
-------
CO
z
o
03
c
o
o
e
a
LOADING TIME
FIGURE 4.1-7
EXAMPLE COMPOS IT
VAPOR PROFILE FOR
LOADING SEQUENTIAL
en
z
o
a
e
<
o
O
ce
o
END OF
LOADING
LOADING TIME
FIGURE 4.1-8
EXAMPLE COMPOSIT
VAPOR PROFILE FOR
SIMULTANEOUS LOADING
-119-
-------
RVP
The Reid vapor pressures (RVP) of gasoline loaded in
the Houston-Calveston area range from 9.5 psia in the summer to
13.6 psia in the winter, see Table 3.4-2. Section 3 of this
report lists the RVP and quantities of gasolines loaded each
month. The RVP of crude oils unloaded in the Houston-Calveston
area normally range from 2 psia to 7 psia but may fall on either
side of this range. Alaskan crudes which will be unloaded on
the West Coast in the future are projected to have an RVP in
the range of 14 psia to 15 psia.
The true vapor pressure of a liquid is a function of
its RVP and its temperature. The nomographs presented in Figures
4.1r9 and 4.1-10 correlate the true vapor pressures of crude oils
However, since the TVP-RVP relationship for crude oil is very
sensitive to the composition and amounts of light hydrocarbons
in the crude oil, Figure 4.1-9 is not always accurate and actual
crudes will vary widely from this nomograph.
Chemical Composition
Tables 4.1-1, 4.1-2, and 4.1-3 present the chemical
compositions of hydrocarbon vapors emitted by motor gasoline,
crude oil, and aviation gasoline marine loading operations. As
.indicated by Table 4.1-1, gasoline vapors can exhibit a wide
range of compositions. ARCO has measured some high concentra-
tions of methane in their gasoline loading vapors, while Shell
measured only trace concentrations. The presence of methane in
gasoline loading emissions presents a problem from the stand-
point that most vapor control systems are ineffective on methane.
Table 4.1-1 also indicates the wide range of molecular weights
which gasoline vapors can be expected to exhibit.
-120-
-------
TRUE VAPOR PRESSURE IN POUNDS PER SQUARE INCH ABSOLUTE
III | I I I I |l|l|l|l|l |l| I | I | I | I |I'II|HM| I I I I |MM|M I I | I I I I I
c
33
m
i
-------
I— 0.20
— 0.30
— 0.40
,20-
Ul
3
•J
0
CO
5
z
Ul
"
3
a
s
Ul
(L
O
Z
3
»
Z
—
Ul
oc
3
82 -
XVtf\ 9 " 3
H
Ul
2
Z
Ul
S
<
u.
Ul
Ul
0
• Z
*™
'Van A 3
— 4.0O flVM 12 30-i
TjfXV "
— 'fir •*/ ~
••ftv- 16 :
— »-00 -017' • „„ :
- W^18 40~
- -UV^ 20 2
- flC//
*
H
<
(2
Ul
0.
Ul
— 6.00 ij ^ -5
1 ,.oo • f M •
E • 3
r- 8.00 "|
= 20-=
— 9.00
_ Sa SLOPE OF THE ASTM DISTILLATION CURVE AT :
— 10.0 10 PER CENT EVAPORATED « ~
;
i
110 OE4 f AT 13 PER CENT' MINUS OES P AT 3 PER CENT 10-5
- 10 -
— 12.0 I0 z
j
i
4
I" I3'° IN THE ABSENCE OF DISTILLATION DATA THE FOLLOW- fl_§
— 14.0 INS AVERAGE VALUE OF S MAY BE USED :
I" IS'0 MOTOR GASOLINE 3
— 16.0 AVIATION GASOLINE 2
— |7.0 LIGHT NAPHTHA (9-1* LS RVP) 3.3
Z_ |8>0 . NAPHTHA (2-8 LB RVP) 2.3
— 1 9.0
— 20.0
— 21.0
— 22.0
— 23.0
=- 24.0
FIGURE 4.1-10
VAPOR PRESSURES OF GASOLINES
AND FINISHED PETROLEUM PRODUCTS
-122-
-------
TABLE 4.1-1
CHEMICAL COMPOSITION OF GASOLINE LOADING VAPORS
Shell Oil Company.
Ship-Valley Forge
10/19/1974
Atlantic Richfield Oil Company3
Ship-ARCO Enterprise
11/13/1974
Compound Vol. 70
Ci + C2 0.02
C3 0.02
Cii paraffins 2.20
C^ olefins 0.16
C5 paraffins 1.01
Cs olefins 0.06
C6 paraffins 0.19
C6 olefins
benzene
C7 paraffins 0.04
C7 olefins
toluene 0.15
Cs paraffins 0.13
C8 olefins
C8 aromatics 0.02
Air 96.00
Compound
C1
C2
C3. paraffin
C3 olefin
Vol.
n-C^
Ci,-61efin
iso-Cs
n-C5
Cs-olefin
n-C6
C6-paraffins
C6+
3.31
0.07
0.20
0.00
0.47
2.59
0.13
63
27
0.24
0.08
0.57
0.23
2,
1,
Air
88.21
Molecular
Weight of
Hydrocarbons
66.9
Molecular
Weight of
Hydrocarbons
52.99
Reference 19
Reference 5
-123-
-------
TABLE 4.1-2
-P-
I
Cargo Tank
Final True Ullage, Ft
Tank Depth, Ft
Oil loaded, bbl
True Ullage, Ft
C2
C3
iC5
nC5
C6
C7
C8
C9
Molecular Weight
Total Sample
% Air
% C02
% Hydrocarbon
COMPOSITION OF VENTED VAPORS. VOL. %
CRUDE OIL LOADING TEST, 5-8-76,
AVILA TERMINAL, TANKER: LION OF CALIFORNIA
Hydrocarbon Fraction Only
3 Port 3 Center
1.9 1.6
39.3 40.2
3911 8706
29.1 9.0
24.17 2.23
3.72 0:39
14.38 3.25
5.27 3.48
22.08 25.10
10.44 23.63
8.58 20.35
7.50 13.60
4.59 6.88
0.84 0.67
0.32 0.37
0.08 0.05
0.03
53.5 70.1
97.26 95.35
0.07 0.06
2.67 4.59
3 Starboard
1.9
39.4
3911
16.4
16.76
3.20
13.85
5.46
21'. 21
11.66
9.84
10.19
6.52
0.75
0.51
0.05
57^-6
96.75
0.07
3.18
Temperature, °F
Tank Vapor
Ambient •
Dew Point
66
56
•13
77
55
4
68
55
-8
Reference 22
-------
TABLE 4.1-3
CHEMICAL COMPOSITION OF AVIATION GASOLINE VAPOR
Vapor Concentration (Weight Percent) at 80 F°
Component (AvGas 80)
n-Butane 3.170
Isopentane 62.711
n-Pentane 5.724
2-3 Dimethyl- Butane 3.261
2 Methyl Pentane 2.073-
Cyclopentane 0.207
3 Methyl Pentane 0.824
Hexane 0 . 165
Cyclohexane 0.000
C7 and Heavier 21.865
Total 100.000
(AvGas 100)
2.517
62.118
5.238
3.168
2.064
0.212
0.865
0.000
0.000
23.818
100.000
(AvGas 115)
4.071
62.366
5.396
2.687
1.666
0.217
0.644
0.000
0.087
22.866
100.000
Average
3.253
62.398
5.453
3.039
1.935
0.212
0.778
0.055
0.029
22.848
100.000
a
These compositions were calculated by Exxon Company using actual
av-gas compositions and equilibrium flash calculations.
Reference 12
-125-
-------
The chemical compositions and molecular weights of
crude oil vapors will vary over a much broader range than those
of gasoline vapors. Crude oil vapors will range in molecular
weight from 45 to 100 pounds per pound-mole.
Explosive Range
i
The explosive range of hydrocarbon product is defined
as the range of hydrocarbon concentrations in air for which the
gaseous mixture will support combustion. For gasoline vapors
whose major components are n-butane, n-pentane, and iso-pentane,
the explosive range in air is 1.4 vol % to 8.4 vol 7o hydrocarbons.
If the concentration of hydrocarbons in gasoline vapors falls
below 1.4%, there are insufficient hydrocarbons in the vapors to
support combustion. If the concentration of hydrocarbons -is above
8.4%, there is insufficient oxygen present in the vapors to sup-
port combustion. For those gasoline vapors which contain signi-
ficant amounts of lighter components, especially methane and ethane,
both the lower and upper explosive limits in air are increased.
From ballasted or uncleaned tanks, the vented vapors
may be on the explosive range for up to 80%, of the gasoline
tanker loading operation. (Rev. 13).
4.2 Source Testing Results
4.2.1 Industry Testing
The petroleum industry has been very involved in test
programs to quantify the hydrocarbon emissions from gasoline
and crude oil transfer operations at marine terminals. Table
4.2-1 summarizes the test programs which have been conducted
by the petroleum industry. The industry programs have included
motor gasoline, aviation gasoline, and crude oil loading onto
-126
-------
TabU 4.2-1
I
I-1
to
Coaifaay Tvpaa of Harlna Taetloa. Loral Ion Data Kalmt of Taatl
ballaallng anlealono Onion Oil Tamlnal (icata ara ongoing)
for cri*da oil and Cctty Oil Tarnlnal
natural gaaolloa California
avgaa and cruda loading
Uiarg la.. Iran
AM r lean notor gaaollaa loading predominantly In 1*14-1914
Fatrolaun Houaton-Calvaatm
InattlHta aran (
loading •( tankera April 1911
•M Cnfaalon Fact or a
Ibat cailaaluna fron loading
crude and not gaaolloa ara 0.9
lu 1.0 Iba/lOOO galluoa
Caaullwe Loading
taafcttr - gee free
tanknr - ballaaied
taufcer - twcltanrd
ayerag* Eaiion leaker
occaw barge * gea frea
oceau barga - Muclaanad
Aviation Caaollua Loading. %
tanker - gan frra
tanker * wnclran(av-gaa frav.)
tanker - uncleaia(ew> gaa prav.)
barga II. JJ vol 1(4.21 U/«gal)
Ualehtad Averaga Dock 1,1 In/ng
.14 vo
.94 vn
1 .24 vo
.4) vo (1.4? t»/a*al)
.01 vo
14.40 va
11. M vo 1(2.44 Jb/ntal)
1.4) vo 1
4.4i vo I
10.44 vo X
4.13 vo t(l.l) U/ng*D
al
Alao Itava n TV* dependent cnrralatlon -
cltan lanknra 1.) Ik/ngal
tmclcanad bargea J.I l»/ngal
Caaollna Loading on Tanker
feet load, lo- JVC. clean
feel load, ewd Trr.claan
alow load, high Trr. clean
1.1 vol t(0.4 Ib/ngal)
•2.4 vol I<0.5 !»/•«•!>
4.2 vol t(0.9 I»/n«al)
Prl»rllr »olor UbllUB. »l
(••alia* 1 o»4 lag
cruJ* v«r»< unlo*dlM T«u* City. T.
l/li/H lo I/11/1I «0~M lull
J/H/H lo «/)/)> t Ml*
Itllllk
r-tff.Ux
Octofc«r 1*14
Itll
J-10 taata
. . . .
•low load. hl(h TVf. oart claao 4.9 vol 1(1.1 U /«..))
av( A«X> 1 anker 1.9 vol 1(0.14 !»/•«• 1)
ewn* d«««lopc4
won* ••vtlapad
AHOCO d!4 alata that avaraga
••laalona lor AHOCO attla laaa
lfca» 10.1 vol I
MOCM d*v«loa>*d
awn* dev*lop«4
Co—«.!»
• J«MI kcglMi
(liat lk«f k«
. 1C I*
flBMI CO«'»CI«4 .BtVIMlV. I..II.I Vlllt
Tk. ATI (••pMMltra Lo<> COM!IK.
•II .»»II«M. MtlM l• ff««««t«< !• >••!. !»• Mill* <•!•
g.lh.r.4 (ov 4.v«lap*.nt •< fl.cl.ffB.
•M<"'< ralMlo* l.cl.r. .i«...l«4 I. Ik. «••«.
-------
tankers, barges, and ocean barges. Well over 200 cargo tanks
were sampled in these programs. Although not reported in this
study, the petroleum and chemical industries have also conducted
a limited number of tests on hydrocarbon emissions from petro-
chemical loading operations. The petroleum industry tests
were primarily conducted between 1974 and 1975 in the Houston-
Galveston area. Tests have also been conducted on the West
Coast and in the Great Lakes area.
Tables 4.2-2 and 4.2-3 summarize the results of the
petroleum industry emission testing programs. The individual
industry results are presented in Appendix III and test data
from the industry test programs are presented in Appendix IV.
The range of emissions presented in Tables 4.2-2 and
4.2-3 represent the range in values reported by the petroleum
industry. The average value reported in the tables is the average
of all values reported by the petroleum industry. The "average
condition" reported in the tables represent the average emission
factor reported by industry for their individual dock and cruise
history situations.
Tables 4.2-2 and 4.2-3 indicate that the hydrocarbon
emission rates from aviation gasoline transfers are very simi-
lar to those for motor gasoline transfers. The average tanker
loaded in the Houston-Galveston area appears to have been cleaned
or ballasted and has a hydrocarbon emission rate of approximately
1.2 pounds per thousand gallons transferred. The average ocean
barge loaded in the Houston-Galveston area has a higher hydro-
carbon emission rate of approximately 2.7 pounds per thousand
gallons transferred. The average barge loaded in the Houston-
Galveston area is neither cleaned nor ballasted and has a hydro-
carbon emission rate of approximately 4 pounds per thousand
gallons transferred.
-128-
-------
TABLE 4.2-2
SUMMARY OF RESULTS
HYDROCARBON EMISSIONS FROM MARINE LOADING
MOTOR GASOLINE
hydrocarbon emissions Ib/tngal
range average
Tankers
clean-vapor free . 0.50-1.50 1.0
ballasted 1.62 1.6a
uncleaned 2.38-2.50 2.4
average condition 0.84-1.47 1.2
Ocean Barges
' clean-vapor free 1.3a
ballasted 2.1a
uncleaned 3.3a
-7 a
average condition 2./
Standard Barges
clean-vapor free 1.2a
uncleaned 3.80-4.14 4.0
average condition 4.1a
a
Exxon Company was the only supplier of information in these
categories.
-129-
-------
TABLE 4.2-3
SUMMARY OF RESULTS
HYDROCARBON EMISSIONS FROM MARINE LOADING
AVIATION GASOLINE
Emissions
Ib/mgal
Tankers
clean-gas free 0.45a
uncleaned aviation previous cargo 1.83a
uncleaned motor gasoline previous cargo 2.92a
average tanker (Exxon) 1.47a
average tanker (Military) 1.13a
Barges
o
average condition . 4.25
a
Exxon Company was the only supplier of information in these
categories. (Reference 13)
-130-
-------
As the industry test data indicates, the cruise history
of a marine vessel greatly impacts its loading emissions.
Various company docks will exhibit variations in their loading
emission rates due to variations which exist in each company's
cruise history policies.
4.2.2 Radian Testing
Radian Corporation conducted a limited sampling pro-
gram for the purpose of gathering data to use in the verification
of industry reported emission factors. The test data gathered by
Radian are contained in Appendix VI. Results of these tests are
presented graphically in Appendix V.
The following observations were made based on the
Radian test data.
• Hydrocarbon emissions from ballasting cargo tanks
which were previously filled with a 3 RVP to 6 RVP crude oil
will range from 5 to 10 vol % and average approximately 7 vol 7«
(1.4 Ibs/m gal).
• Hydrocarbon emissions from ballasting cargo tanks
which were only partially filled with crude oil or which were
used to collect strippings are generally much greater than
emissions from ballasting tanks which were completely filled.
This is due to evaporation into a larger vapor space.
a The vapor blanket above the surface of gasoline
being loaded into cargo tanks is less than 5 ft. thick and nor-
mally ranges from 2 ft. to 3 ft. thick.
-131-
-------
• The arrival components of ship and ocean barge
cargo tanks were found to be the following:
cleaned tanks 0-2 vol 70
ballasted tanks 2-10 vol %
uncleaned tanks -21 vol 70 (single sample) .
• The total emissions from loading gasoline onto
ships and ocean barges were found to be as follows:
cleaned tanks 3-5 vol % (=0.6-1 Ib/m gal)
ballasted tanks 9-13 vol %(=1.8-2.7 Ib/m gal)
uncleaned tanks =25 vol °/0 (=5.1 Ib/m gal)
Note: The uncleaned tank value is a single test
point.
• Limited barge data indicated barge emissions are in
the range of 27 vol "I* hydrocarbons or approximately 5.5 Ibs/m gal,
4.2.3 Conclusions
Radian test data indicate that the industry emission
factors presented in Tables 4.2-2 and 4.2-3 accurately represent
the hydrocarbon emissions from marine transfer operations. The
Radian emission data deviated somewhat from industry data for
loading uncleaned ships and loading barges; however, in each
case, the Radian values were based on single data points. It
may be necessary to conduct additional testing programs in these
areas to verify industry emission factors.
Radian test results also indicate that hydrocarbon
emissions from ballasting crude tankers are in the range of 1
to 2 pounds per thousand gallons ballasted.
-132-
-------
5.0 EMISSION CONTROL TECHNOLOGY
Emission control technology for marine loading of
gasoline is a young technology which is faced with a unique set
of problems. Although vapor recovery has been used on marine
loading operations in the petroleum and chemical industries for
several years, the marine loading operations currently being
controlled are closed systems with no air present, which
have much slower loading rates, and involves products handled
in relatively small quantities. The presence of air in vented
hydrocarbon vapors presents a potential explosion hazard when
in specific concentration ranges. The cumulative loading rates
for a single gasoline tanker can be as high as 50,000 barrels
per hour, which is equivalent to 4700 scfm of displaced vapors.
Although gasoline vapor control technology has also been applied
to tanktruck loading operations, truck terminal vapor control
technology is not directly applicable to marine terminals because
the flow rates are much smaller than those experienced in marine
loading, and because the hydrocarbon vapor concentrations in
tanktruck loading vapors are above'the explosive range.
The safety and design problems associated with vapor
control technology for marine loading of gasoline are not thought
to be technically insurmountable. The principles involved in
marine vapor collection and control are well understood.
The principle vapor control systems being considered
in the Houston-Calveston area are based on conveying hydrocarbon
vapors generated onboard the vessels to shoreside vapor control
units for recovery as liquid product or for disposal by incinera-
tion. Shoreside vapor control systems can be divided into three
-133-
-------
components; the vapor control unit, the shoreside vapor collection
system, and the shipside vapor collection system. These three
components of shoreside vapor control systems are discussed in
depth in Sections 5.1, 5.2 and 5.3, respectively. Alternative
strategies for vapor emission control are presented in Section
5.4.
5.1 Vapor Control Unit
A key component in each vapor control system is the
vapor control unit which serves the function of reducing the
hydrocarbon content of marine loading vapors to an acceptable
level. This section discusses the major vapor control units
individually from the aspects of principle of operation, effi-
ciency, safety, cost, and salient considerations.
5.1.1 Refrigeration
Principle of Operation
The refrigeration vapor recovery system recovers the
hydrocarbon content of gasoline loading vapors by condensation
at cryogenic temperatures and atmospheric pressure. The flow
diagram of a refrigeration vapor recovery system is presented
in Figure 5.1-1,
The vapors collected from gasoline loading operations
are routed into one of two identical vapor processing trains.
The gasoline vapors are first cooled in a dehydrator to a
temperature of 25°F to 35°F by direct contact with finned tube
cooling coils, A major portion of the moisture and a small
portion of the heavier hydrocarbons in the vapors are condensed
in the dehydrator. Vapors flow from the dehydrator to the
condenser where they are cooled to a teir.perature of -80°F to
-134-
-------
I
GASOLINE
i VAPORS
u>
Ul
'
^^-w/iitH iu uiarua/M-
OIL - ,
WATER ^ ^
^ CONUbNStl) **
SEPARATOR pnnp,,nT
OIL/WATER , \STORAGE^/
CONDENSATE CONDENSED
PRODUCT
DEHYDRATOR CONDENSER VENTED
26»F TO 36»F -BO«F TO -IOO»F AIR
I •!
i ' i
i It
COOLANT 2 8TAQE
STORAGE REFRIGERATION
; |:
DEHYDRATOR CONDENSER VENTED
25*F TO 36»F -80»F TO -tOO»F
CONDENSED
PRODUCT
WATCH B» CONDENSED ^ PR°DIJ
WATCR ^ PRODUCT ^"RETUR
SEPARATION I.STORAGEJ TO RE
OIL/WATER
PRODUCT
RETURNED TO
REFINERY
FIGURE 5.1-1 REFRIGERATION VAPOR RECOVERY UNIT
-------
-100°F. Most of the remaining hydrocarbons in the gasoline
vapors are condensed in the condenser unless methane is present
in the vapors. Condenser temperatures of -100°F will not con-
dense methane, and it will be vented from the condenser along
with the treated air.
Water and oil condensates collected in the dehydrator
are separated in a gravity oil-water separator. The recovered
water is routed to waste water treatment, and the recovered prod-
uct goes to product storage. Condensed product from the con-
denser also goes to product storage. The product recovered by
the refrigeration vapor recovery unit is primarily composed
of propane, butanes, and pentanes. Because of their high vapor
pressures, these recovered products must be stored and handled
in closed pressurized systems. Recovered product is pumped
regularly from product storage back to the refinery.
Refrigeration for the dehydrator and condenser is
supplied by a two-stage refrigeration unit. A large volume of
cold coolant is stored in the system to supply instantaneous
cooling capacity on demand.
Two vapor processing trains are used so that one train
can be defrosted without interruption of vapor processing.
The cold air vented from the condenser will create a
moisture plum.e in high humidity weather. To avoid the formation
of plumes, some designs incorporate air reheat using waste heat
from the refrigeration unit.
-136-
-------
Efficiency
The vapor recovery efficiency of the refrigeration
vapor recovery unit is directly dependent on the chemical
composition of the hydrocarbon vapors, the concentration of the
hydrocarbon vapors, and the operating temperature of the conden-
ser. Some companies have identified methane in their marine
loading vapors. Very little methane will condense at the con-
denser temperatures. Refrigeration vapor recovery units in ser-
vice at tanktruck loading terminals are consistently reducing
the hydrocarbon content of gasoline vapors to levels of 3% to
47o by volume. Information from the manufacturer indicates that,
excluding methane, the hydrocarbon content of gasoline vapors
can be reduced consistently to levels of less than 570 with con-
denser temperatures of -100°F.
Cost
Itemized costs estimates for the purchase and installa-
tion of proposed refrigeration vapor recovery units at existing
terminals are presented in Appendix II. Estimates for the bare
unit cost of a refrigeration unit range from $300,000 to $470,000
per 10,000 bbl/hr of gasoline loading capacity. Estimates for
the installed cost of refrigeration vapor recovery units range
from $350,000 to $1,300,000 per 10,000 bbl/hr of gasoline loading
capacity. The average installed cost reported by the petroleum
industry is $900,000 per 10,000 bbl/hr of gasoline loading capa-
city. In a separate cost analysis conducted by Radian (Appendix
VII) the average installed cost for just the unit was estimated
to be $600,000 per 10,000 bbl/hr capacity. Installed costs in-
clude such items as foundations, site work, utility hook-up, and
instrumentation. Operating costs excluding taxes and depreciation
range from .$11 to $25 per thousand barrels of throughput.
-137-
-------
Safety
The refrigeration vapor recovery unit is considered
by many to be the safest among the three vapor control units
being considered by industry. The low vapor velocities experi-
enced in the dehydrator and condenser reduce the chance of static
charge build-up. All moving and electrical parts are contained
in the refrigeration unit and can be positioned well away from
the dehydrator and condenser. The extremely cold vapor tempera-
tures also make ignition more difficult. However, the lower
explosive limit of a gas is reduced at cryogenic temperatures.
This, in effect, widens the range of explosive limits for the gas.
The vapors both entering and exiting the refrigeration
vapor recovery unit may be in the explosive range. The positioning
of flame arresters and/or water seal drums on the vapor feed line
and vent line of the refrigeration vapor recovery unit may aid in
preventing flame fronts from propagating to or from the unit.
However, such flame control devices have not been tested for
this type of service; therefore, data on their performance is not
available. Careful engineering and design must be used in devel-
oping a safe and reliable refrigeration vapor recovery unit.
Refrigeration vapor recovery units have been operating
very successfully at pipeline terminals for approximately two
years. These units, though, have not dealt with the wide ranges
of vapor compositions and concentrations to be handled by vapor
recovery units in marine terminal application. Marine terminal
application will require a ten- to fifteen-fold scale-up in
size. No major problems are foreseen with the equipment scale-up,
but initial systems will likely be oversized for insurance and
will require considerable trial-and-error testing to achieve
reliable operation. One of the initial operating problems
-138-
-------
requiring the testing will be the need for the unit to operate
15-20 hours continuously with minimal pressure drop.
5.1.2 Absorption
Principles of Operation t
The absorption vapor recovery unit recovers hydrocarbon
from gasoline vapors by absorption into a lean oil stream-from
the refinery. Figure 5,1-2 presents two lean ail absorption
systems under consideration for use in the Houston-Calves ton area.
One proposed lean oil absorption system operates at
low pressure and ambient temperature. -Gasoline loading vapors
are compressed to 20 psig by a blower or liquid ring compressor.
The compressed vapors then contact a lean oil stream in a packed
bed absorber where hydrocarbons in the vapor are absorbed by
the lean oil. Projected lean oil flow rates are 500 bbl/hr of
lean oil per 1000 acfm of vapor. Purified air is vented from
the absorber unices methane is present in the vapors. Lean oil
will not absorb methane, and it will pass through the column and
be vented with the treated air.
The lean oil source is cat-cracker feed from the
refinery. A one-month supply of lean oil is stored on the vapor
recovery site. Rich oil laden with light hydrocarbons absorbed
in the absorber is recycled to the lean oil storage tanks
until the vapor pressure of the lean oil is too high for effective
absorption. At such a time, the enriched oil is returned to the
refinery and the lean oil storage tanks are replenished with fresh
cat-cracker feed.
-139-
-------
VENTED
AIR
LEAN OIL
STORAGE
GASOLINE
LOADING
VAPORS
SATURATOR
LEAN OIL
BLOWER OR
COMPRESSOR
RICH OIL
RICH OIL
RETURNED
TO REFINERY
LEAN OIL
FROM REFINERY
-P-
O
I
GASOLINE
LOADING -
VAPORS
EDUCTOR
AIR— *-T——'J— +- VENTED
AIR
LEAN OIL
LEAN OIL
STORAGE
RICH OIL
TO REFINERY
S— 1-"^ LEAh
RICH OIL
LEAN OIL
FROM REFINERY
FIGURE 5.1-2 ABSORPTION VAPOR RECOVERY UNITS
-------
In order to avoid some of the potential safety problems
which have been associated with blowers and compressors, a second
type of lean oil absorption system has been proposed which
utilizes air eductors as the primary motive force. Because the
use of air eductors results in a slight vacuum in the lean oil
absorber, effective absorption requires the use of low tempera-
tures. The lean oil is cooled to 40°F by refrigeration. Lean
oil flow rates for the refrigerated absorption vapor recovery
unit have been estimated at 200 bbl/hr per 1000 acfm.
Some absorber designs call for constant lean oil flow
rates which are in excess of the flow rate required for effective
hydrocarbon recovery at maximum loading rates. Other absorber
designs include instrumentation and control systems for regulating
lean oil flow rates with demand to conserve energy.
Efficiency
The. hydrocarbon removal efficiency of absorption vapor
recovery systems depends directly on the ratio of the lean oil
flow rate to the vapor flow rate. Normally, the hydrocarbon
content of gasoline loading vapors can be reduced to less than
5 volume percent with reasonable lean oil to vapor ratios. Some
petroleum companies have identified methane in gasoline loading
vapors. Lean oil absorbers have very little impact on methane,
and methane is expected to pass unaffected through absorption
units.
Cost
Itemized cost estimates supplied by industry for retro-
fitting absorption vapor recovery units are presented in Appendix
II. Estimated installed costs for absorption systems range from
-141-
-------
$200,000 per 10,000 bbl/hr to $1,000,000 per 10,000 bbl/hr and
average $600,000 per 10,000 bbl/hr. A separate cost analysis
conducted by Radian (Appendix VII) estimated the installed cost
of absorption vapor recovery systems at approximately $600,000
per 10,000 bbl/hr capacity. Vapor collection systems are not
included in this cost. The utilities.component of the annual
operating cost was estimated at approximately $12 per thousand
barrels of gasoline loaded.
Safety
Some concern has been expressed for the safety of lean
oil absorption systems as with all vapor recovery systems. In
general, safety concerns center around the equipment used to
force gasoline loading vapors through the absorber. Blowers
and fans have been documented as the source of static electrical
discharges leading to coal mine explosions. Misaligned bearings
and impeller shafts can develop hot spots which serve as ignition
sources. Safety problems have also been identified with eductors.
Under certain conditions, eductors, like blowers, can also build
up static charges.
Enrichment systems which raise the hydrocarbon content
above the explosive range have been suggested in some designs,
but these put an increased load on recovery systems and result
in higher operating utility costs. It should also be noted that
enrichment systems protect only the portion of the system which
is between the enrichment system and the vapor control unit.
Safety is not an insurmountable problem with absorption
systems, but it does require careful engineering and design.
-142-
-------
State of Development
Several lean oil absorption systems are in service at
pipeline terminals for the control of tank truck loading emis-
sions. Althougn most of the units are working well, some units
are experiencing problems in maintaining required removal
efficiencies. The lean oil flow rate necessary for the required
hydrocarbon removal is higher than the design flow rate.
Marine terminal absorption units will require a ten
to fifteen fold scale-up from the largest pipeline terminal
absorption unit now in operation. This large scale-up factor
may produce minor problems in initial marine loading absorption
units. The pipeline terminal units do not handle the wide variety
of vapor concentrations and compositions as the marine terminal
units will be subject to. This may add to scale-up problems.
5.1.3 Incineration
Principle of Operation
The incineration vapor control unit reduces the hydro-
carbon content of gasoline loading vapors by combusting the
hydrocarbons to carbon dioxide and water. The flow diagram
of a typical incineration vapor control unit is shown in
Figure 5.1-3.
Gasoline vapors from marine loading operations are
drawn through a flame arrester and a saturator. In the saturator
gasoline vapors are saturated with hydrocarbons by contact with
recirculating gasoline product. Some designs merely call for the
enrichment of gasoline vapors beyond the upper explosive level.
Some systems also call for propane to be used instead of gasoline
as the enrichment source.
-143-
-------
GASOLINE
LOADING ~
VAPORS
KNOCK-OUT
DRUM
ENRICHED
VAPORS
BLOWER
SATURATOR
FLAME
ARRESTOR
RECYCLE
GASOLINE
INCINERATOR
OR FLARE
( D
WATER
SEAL
CONDENSATES
-GASOLINE
COMBUSTION AIR
FIGURE 5.1-3 INCINERATION VAPOR CONTROL UNIT
-------
The enriched or saturated vapor is conveyed by blower
to a knockout drum where condensates are allowed to settle.
From the knockout drum the gasoline vapors enter an incinerator
or flare after passing through a water seal. In the incinerator
or flare, combustion air is mixed with the gasoline vapors to
bring them back into the combustion range. Subsequently, the
vapors are combusted.
Incineration units require sophisticated instrumenta-
tion and control systems to maintain proper fuel to air ratios
for all gasoline vapor flow rates and concentrations. Most
incineration unit designs use oxygen analyzers, hydrocarbon
analyzers and temperature sensors to control saturator operation
and combustion air flow rates,
Efficiency
The efficiency of incineration units for control of
gasoline loading vapors is well over 997« based on the hydrocarbon
vapors vented from the ship. Hydrocarbon concentrations in the
flue gas vented from incineration units are well below 1%.
Cost
Complete cost data is not available on incineration
vapor control systems because, tentatively, none of the companies
in the Houston-Calveston area have plans to install such systems.
Rough cost estimates indicate, however, that the installed cost
of an incineration system at an existing terminal range from
$300,000 to $400,000 per 10,000 bbl/hr of gasoline loading
capacity.
-145-
-------
Safety
The presence of a flame in incineration vapor control
units is viewed by some as a major safety problem. Although
the saturator prevents the possibility of a flame propagating
from the incinerator back through.the system, a malfunctioning
saturator could allow flame propagation back through the system
to the ship or barge, causing a major explosion. Water seals
at the incinerator or flare base are designed to protect against
such occurrences, but their effectiveness has not been established.
The"presence of a blower in the incineration system is
not expected to present a safety problem because it is positioned
downstream of the saturator. Safety problems associated with
incineration requires careful engineering and design to develop
a safe reliable unit.
State of Development
Incinerators and flares have been used in the petroleum
and chemical industry for quite some time. The major components
of these systems are generally considered well-developed techno-
logy. Incineration systems have also been used for the control
of gasoline vapors from tanktruck loading at pipeline terminals.
However, marine terminal incineration systems present
unique problems which require saturators and sophisticated
instrumentation. The effectiveness of this equipment remains
to be demonstrated.
-146-
-------
5.1.4 Alternative Vapor Recovery Units
There are two other commonly used vapor recovery units
which are not currently being considered for the Houston-Galveston
area primarily because of safety reasons.
The first of these units is the compression-refrigeration-
cohdensation (CRC) unit which recovers hydrocarbon vapors by con-
densation at low temperature and moderate pressure. The moderate
pressures used in CRC systems are supplied by multi-stage com-
pressors. Although a saturator is used in the system, the multi-
stage compression, of hydrocarbons in the presence of oxygen is
considered by many to be a significant safety risk.
A second vapor recovery unit which is not currently
being considered for use on marine loading emissions in the
Houston-Galveston area is carbon adsorption. In a carbon adsorp-
tion unit gasoline loading vapors are passed through a carbon
bed, and the hydrocarbon constituents in the vapors are removed
by adsorption onto the carbon. Adsorption is an exothermic
reaction which could possibly serve as an ignition source in marine
loading vapor control systems.
As the carbon bed becomes saturated with hydrocarbons,
its removal efficiency drops, and the bed must be reclaimed.
Normally, carbon beds are reclaimed by using steam stripping and
a vapor recovery unit. The need for a vapor recovery unit in
addition to the carbon adsorption unit results in high capital
costs for carbon adsorption systems.
-147-
-------
5.1.5 Vapor Control Unit Installation
The installation of a vapor control unit is a quite
involved process. The units must be placed on a firm foundation.
Designs for the Houston-Calveston area generally include con-
struction of pier supported concrete foundations for the vapor
control unit.
Different vapor control units may require utilities
including compressed air, water, electricity, lean oil, steam,
enrichment gas, and wastewater sewers. These utilities must be
run out to the unit site and hooked up.
Additional items which must be installed in conjunction
with vapor control units include rsAis, control houses, sumps,
and fire protection equipment. Operators and maintenance people
must also be provided for the vapor control unit.
5.1.6 Inerting
Inerting systems replace the oxygen rich air residing
in the vapor space of cargo tanks with inert, oxygen deficient
flue gas. The flue gas is obtained from either the ship exhaust
or from the exhaust of a special fuel burner. Prior to their
use in the cargo tanks, the inert exhaust gases are passed through
a water scrubber where they are cooled and cleaned. Currently
the very large crude carriers (VLCC's) utilize inerting systems.
Inerting their cargo tanks is considered by the marine industry
to be the safest way of operating them.
Vapor inerting has been proposed by some as a possible
means of solving the safety problems which potentially exist
with the collection and processing of explosive vapor mixtures.
-148-
-------
However, most petroleum companies have expressed strong doubts
about the effectiveness of inerting systems.
i
If operating properly, inert gasoline vapors are collected
and transferred to shore where they can be processed with a very
minimal chance of an explosion or fire. Inerting systems are much
less energy intensive than either vapor dilution systems or vapor
saturation systems.
Some disadvantages of using inerting systems include
(1) potential errosion problems due to sulfuric and sulfurous
acids generated from combusting fuel oils containing sulfur,
(2) added personnel and equipment expenses, (3) extreme hazard
potential if operating problems introduce hot exhaust into cargo
tanks, (4) the false sense of security created by inerting sys-
tems, (5) potential cargo contamination with CO* in the inert
gas, and (6) increased hydrocarbon emissions.
Despite these disadvantages, inert gas systems on
board the tanker or on the dock hold some potential to reduce
the safety risks involved in marine terminal VRU.
5.1.7 Composite Vapor Profile
The composite vapor profile discussed in Section 4.1.5
has a significant impact on the efficiency of vapor recovery units.
During portions of a clean vessel loading, the hydrocarbon concen-
tration in the composite vapors is likely to be below 5 vol 7».
This phenomena is depicted in Figure 5.1-4. Vapor recovery units
are designed primarily to reduce the hydrocarbon content of con-
centrated vapors to a concentration of less than 5 vol 70. This
situation complicates attempts to calculate the net reduction
of hydrocarbons achieved by controlling marine facilities loading
relatively clean vessels.
-149-
-------
co
c
0
.£>
CO
U
O
•a
s
**
o
50-
45-
40-
35-
30-
25-
20-
«5 -
10-
•
PROFILE OF
VAPOR FEED TO Q POTENTIAL PROFILE OF
RECOVERY UNIT 7 VAPORS VENTED FROM THE
V VAPOR RECOVERY UNIT
\
I
1
1 er^ 5% VAPOR CONTROL LEVEL
1 _^
JV *^:^.._
t 2 3
Time, hrs.
FIGURE 5.1-4 VAPOR PROFILES OF THE FEED AND PRODUCT
OF A VAPOR RECOVERY UNIT
-150-
-------
The principle of vapor recovery is further complicated
with refrigeration systems which may retain small quantities of
condensed hydrocarbons on its condensation coils. Subsequent
loading of a clean cargo tank with an arrival hydrocarbon component
of less than 5 vol 7o may result in re-evaporation of residual
hydrocarbons on the condensation coils into the vented vapors.
However, the re-evaporation cannot exceed the equilibrium concen-
tration of the vapor recovery unit which is conservatively set
at some point below 4 vol 70.
It should also be noted that although re-evaporation
may take place, with a properly operating vapor recovery unit,
there will be a net recovery of hydrocarbons, and the hydrocarbon
concentration in the vapors vented from the recovery unit will
never exceed 5 vol 7».
With respect to absorption control units a similar
problem might exist. Clean vapors may be capable of stripping
some hydrocarbons from the lean oil.
5.2 Shoreside Vapor Collection
Shoreside vapor control systems include a vapor
collection system which conveys the gasoline loading vapors
collected onboard the ship to the shoreside vapor control unit.
This section presents the design, cost, and safety aspects of
several shoreside vapor collection systems proposed for the
Houston-Calveston area.
5.2.1 Design
The function of shoreside vapor collection systems
is to safely and efficiently convey gasoline vapors collected
-151-
-------
onboard marine vessels to the shoreside vapor control unit.
Figure 5.2-1 presents the flow diagram of a typical vapor collection
system.
The ship to shore connection which conveys the collected
gasoline vapors to shore can be made by either a flexible rubber
hose or a hinged loading arm. Flexible hoses and loading arms
are both currently used at marine loading terminals to load gaso-
line and will be compatible with current marine equipment. Projected
hose and loading arm diameters range from 8 inches to 16 inches.
Following the ship to shore connection, most vapor col-
lection system designs incorporate either flame arresters or water
seal drums which may inhibit the propagation of an explosion from
either the vessel to the shore or from the shore to the vessel.
Their effectiveness, as discussed earlier, is questionable.
After passing through a flame arrestor or water seal
drum, the gasoline vapors are conveyed by large diameter piping
to the vapor control unit. Several docks may be manifolded into
a single vapor collection line, or each dock may use a separate
vapor collection line to convey vapors to the central vapor con-
trol unit. Some vapor collection system designs also specify
the use of separate vapor control units on each dock. If the
vapor collection line traverses a long distance or is exposed
to potential hazards, a second flame arrestor or water seal drum
may be installed immediately before the vapor control unit. If
shipside precautions have not been taken against possible tank
overfills, an overflow sump may be needed on the dock for the
purpose of catching overflows and keeping the vapor collection
line free of liquid. In the case of a spill the sump could
pose a substantial fire hazard.
-152-
-------
TANKER
Oi
CO
i
GASOLINE
VAPORS
WATER
SEAL DRUM
VAPOR
CONTROL
UNIT
/OVERFLOWN
V SUMP )
RECOVERED
PRODUCT
RETURNED TO
REFINERY
ON SITE
PRODUCT
STORAGE
FIGURE 5.2-1 TYPICAL VAPOR COLLECTION SYSTEM
-------
A major distinguishing characteristic of vapor
collection systems is their means of inducing vapor flow. Vapor
flow may be induced by displacement or by vacuum. In displace-
ment systems, gasoline vapors are forced out of the ship or
barge tank and through the collection system by the gasoline
entering the tank. Displacement vapor collection systems require
that there be very little pressure drop in the system to impede
the flow of vapors. . The displacement system also requires that
» t
vessels use closed gauging systems in that the opening of
gauging hatches would release vapors from the tanks to the atmos-
phere and prevent their recovery. Liquid overfill of a cargo
tank can cause structural damage to the ship and poses a potential
for water pollution.
Vacuum collection systems utilize a blower or eductor
to draw vapors from the vessel tanks and through the vapor col-
lection system. Steam, air, and solvent eductors have been pro-
posed. Vacuum .systems allow the use of open-hatch gauging
because the vessel tank will be at a negative pressure, prevent-
ing the possible escape of gasoline vapors. The chance of tank
overfills and gasoline spills is considered more remote when open
gauging is used. However, the presence of an eductor or a blower
in the vacuum collection system presents the potential safety
problem of static charge build-up. Both means of moving vapors
have been associated with igniting explosions.
Different types of vapor recovery units have different
magnitudes of pressure drop associated with them. For instance,
refrigeration systems have relatively small pressure drops while
the absorption and incineration units have relatively high drops.
This makes vacuum collection systems necessary for absorption
and incineration systems, and thus allows for open-hatch loading.
However, closed-hatch loading with resulting cost and safety ad-
vantages is feasible with refrigeration systems.
-154-
-------
5.2.2 Efficiency
In a properly functioning vapor collection system, no
vapors will be lost from the system. All vapors will be con-
veyed from the marine vessel to the vapor control unit.
5.2.3 Cost
Itemized cost estimates provided by the petroleum in-
dustry for the installation of vapor collection systems at exist-
ing terminals are presented in Appendix II. Projected cost esti-
mates range widely from $100,000 to $2,000,000 per 10,000 bbl/hr
of gasoline loading capacity. The averaged projected cost was
approximately $1,000,000 per 10,000 bbl/hr gasoline loading
capacity. In an independent cost study Radian Corporation esti-
mated the cost of the vapor collection system to be approximately
$200,000 per 10,000 bbl/hr loading capacity.
5.2.4 Safety
The gasoline vapors conveyed through the vapor collection
system will be in the explosive range during a significant portion
of ship and barge loading operations. A fire or explosion on
either the ship or the vapor control unit could be.spread through
the vapor collection system to other ships, barges, or vapor con-
trol units. A leak in the vapor collection piping would also
present a safety problem.
It has been suggested that flame arresters and water
seal drums be positioned at the junctions between the vapor
collection system and marine vessels and between the vapor
collection system and vapor control units. These safety pre-
cautions would prevent the vapor collection system from becoming
a mechanism for spreading fire. Water seal drums can also be
-155-
-------
equipped with internal chemical fire extinguishers. However,
flame arrestors and water seal drums reportedly have not been
tested for this magnitude of flow and variability of conditions.
Blowers and eductors have been cited as sources of
static electrical discharges and present a potential safety
problem. The positioning of blowers downstream from saturators
greatly reduces these potential safety problems.
As an added safety precaution, some control system
designs isolate each ship and barge by providing individual
vapor control units for each vessel being loaded.
Although the presence of explosives and combustible
vapors in vapor collection systems presents many potential
safety problems, preliminary investigations indicate that such
technology exists to construct safe vapor collection systems.
However, the designs are preliminary and the costs may be sub-
ject to increases.
5.2.5 State of Development
The technology for designing a vapor collection system
is well developed. Significant questions remain as to how much
safety equipment is required to construct a safe system. It is
anticipated that initial systems may be overdesigned until marine
vapor control technology is refined.
5.3 Shipside Vapor Collection
The shipside vapor collection system is a network of
vapor piping installed on ships and barges for the purpose of
collecting and transporting to shore vapors generated by gasoline
-156-
-------
loading operations. This section presents the design, cost,
safety, and state of development for ship-side vapor collection
technology.
5.3.1 Design
* -^•
The flow diagram for a typical ship-side vapor collection
system is presented in Figure 5.3-1. Vapors generated by gaso-
line entering the tank are collected in the ullage dome and
conveyed through vapor collection lines to a vapor collection
header. All loading vapors are routed to the vapor collection
header which then transports the vapors to the shoreside vapor
control system. Block valves are provided on each vapor collec-
tion line such that individual tanks loading nonvolatile products
can be isolated from the vapor control system. The loading of
heavy oils in clean tanks produces insignificant emissions which
need not be processed by the vapor control unit. Some collection
systems also call for spill valves on each ullage hatch which
will drain spills and prevent liquid from clogging the vapor
collection header.
Vapor flow into the vapor collection system can be
induced either by displacement forces or by vacuum forces. In
displacement systems, vapors residing in the cargo tank are dis-
placed into the vapor collection system by the gasoline entering
the cargo tank. All tank openings other than the vapor collection
line must remain closed throughout the tank loading. The
ullage cap cannot be opened for tank gauging. In vacuum systems
a compressor or blower onshore draws a vacuum in the vapor
collection system, which in turn, pulls vapors from the vessel
tanks and conveys them to the vapor control unit. Because the
vacuum system creates a slight vacuum in the vessel tanks, ullage
hatches can be opened for tank gauging without loss of vapor.
-157-
-------
TO
ON-SHORE
VAPOR
CONTROL
SYSTEM
LIAGE HATCH
SHIP'S
HULL
SHIP'S
WING
TANK
VAPOR COLLECTION LINE
VAPOR
COLLECTION
HEADER
FIGURE 5.3-1 SHIP-SIDE VAPOR COLLECTION SYSTEM
-158-
-------
Regardless of size and construction materials, ship
and barge tanks are not capable of withstanding large pressures
or vacuums without warping. Normally, ships and barges are
rated for a -0.5 psig vacuum. Most ships are rated for a pressure
of 2.0 to 2.5 psig and most barges are rated for a pressure of
1.5 psig. To insure against over pressuring or drawing excess
vacuums, a pressure/vacuum valve is positioned in each vapor
collection line. The pressure/vacuum valve is designed to release
excessive pressures or vacuums which may occur in malfunctioning
vapor collection systems before ship or barge damage occurs.
5.3.2 Efficiency
The vapor collection efficiency of the ship-side vapor
collection system is expected to be 100%. All hydrocarbon vapors
generated during gasoline loading will be conveyed to the onshore
vapor control system.
5.3.3 Cost
Projected costs for the installation and operation of
vapor collection equipment onboard ships are presented in Appendix
II. The estimated installation costs for ships generally ranged
from $300,000 to $350,000 per vessel. The estimated installation
costs for shallow barges ranged from $50,000 to $85,000 per
vessel. The single projected installation cost for deep-draft
ocean barges was $150,000 per vessel. In addition to the installa-
tion costs, operating costs have also been projected for maintaining
onboard collection equipment and accounting for the extra labor
and time required to load controlled ships. The projected operating
costs for ships is $110,000 per ship-year and for barges is
$36,000 per barge-year.
-159-
-------
5.3.4 Safety
The potential for loading accidents onboard ships is
increased by the use of vapor collection systems. However,
there are safety devices which can be installed to greatly
reduce the potential of accidents. A system of pressure/vacuum
valves can be installed to guard against the formation of exces-
sive pressures or vacuums in the cargo tanks. Routine mainte-
nance of pressure/vacuum valves will insure their proper
performance. The risk of contaminating other cargoes by tank
overflows into the vapor piping is minimized by the use of over-
flow valves. Equipping each tank with two level gauges and a
high level alarm gives added protection against tank overfills.
Ships can be isolated from shore-side fires and explosions by
positioning water seal drums in the vapor collection piping
between the ship and shore-side vapor control equipment,
5.3.5 Salient Considerations
The installation of vapor collection systems on all
ships loading gasoline will require a significant effort to
insure that all ships and shore-side facilities install compatible
systems. Ship-to-shore connections must be compatible with each
other. Problems may also arise if ships designed for vacuum
vapor collection attempt to load at terminals designed for the
displacement system, and vice-versa.
' The time required to retrofit a ship or barge with
vapor collection equipment is approximately two weeks. Most
vessels are scheduled for dry docking and repairs on a two-year
cycle. Therefore, it is estimated that two years will be re-
quired for the complete retrofitting of all ships with vapor
collection equipment within a ship's normal maintenance schedule.
-160-
-------
5.4 Alternative Control Strategies
Several alternative control strategies have been pro-
posed for controlling gasoline loading vapors. At this time,
however, none of these control strategies look favorable.
5.4.1 Ullage Hatch Condensers
Ullage hatch condensers are a set of refrigeration
coils located under the ullage dome and designed to condense
hydrocarbon compounds from gasoline vapors vented through the
ullage hatch. The refrigeration coils would be maintained at
a temperature of -100°F by a refrigeration unit located either
onboard the ship or on the dock. The need for conveying ex-
plosive mixtures onto shore for recovery is eliminated by the
use of ullage hatch condensers. Drawbacks to the use of ullage
hatch condensers include (1) the high cost of retrofitting each
ullage hatch of each barge and ship with condensing units, (2)
the restricted access of ship personnel to the vessel tanks
caused by ullage hatch condensers, (3) the questionable efficiency
of ullage hatch condensers because of the short contact time
between fast flowing gasoline vapors and condenser coils, and
(4) difficulty in defrosting the coils while loading the tanks.
5.4.2 Ship Boiler Incineration
Ship boiler incineration systems would incinerate
gasoline loading vapors in much the same manner as shoreside
incineration units described in Section 5.1.3. Explosive
vapors would not need to be transported to shore for processing
with the ship boiler incineration system. However, the risk of
combusting explosive vapors in the ship's boiler is considered
very high. Loading rates would also have to be greatly reduced
because of the limited capacity of ship boiler systems.
-161-
-------
5.4.3 Foam
Foam systems have been developed which cover liquid
surfaces and effectively reduce hydrocarbon evaporation. In
the proposed foam system, a shallow layer of foam would be
placed in each tank before filling. As the gasoline cargo is
pumped into the tanks, the foam floats on its surface, preventing
the evaporation of hydrocarbons into the vented vapors. Although
foam systems were effective in reducing hydrocarbon vapors, the
foam left an intolerable scum in the gasoline product.
5.4.4 Product Cooling
Systems based on the reduction of hydrocarbon vapors
by cooling the product prior to loading have also been investi-
gated. Gasoline would have to be cooled to 0°F in order to
reduce the hydrocarbon concentration in loading vapors to 570.
The cooling load required to lower the temperature of gasoline
to 0°F is cost-prohibitive.
5.4.5 Controlled Loading
One proposed control technique involves the combina-
tion of ship cleaning and loading rate control. While at sea,
the ship or barge would thoroughly clean and devapor each cargo
tank to be loaded with gasoline. Effective tank cleaning can be
accomplished by double ballasting or by butterworthing. When
loading gasoline, the first three feet would be loaded slowly
so as to minimize gasoline vapor generation. The balance of the
tank would be rapidly loaded and loading would be terminated
three feet from the tank roof, thus, preventing the venting of
the vapor blanket which forms above the gasoline surface.
Baffles can be placed in the tank to prevent sloshing, thus main-
taining ship stability.
-162-
-------
The efficiency of the controlled loading system has
not been determined, but the system is potentially capable of
lowering the hydrocarbon content of gasoline loading vapors
to a maximum of 4-7 vol 7».
-163-
-------
6.0 ECONOMICS OF EMISSION CONTROLS
The economics of installing hydrocarbon emission con-
trols on existing marine terminals in the Houston-Calveston area
i
are very difficult to establish. Each marine terminal involved
is confronted with unique problems. There are also wide varia-
tions in the estimated costs associated with marine terminal
controls because currently there are no marine terminals equip-
ped with gasoline loading controls. The unique arrangements
required to control gasoline vapors from marine loading prohibit
the direct translation of cost data from truck loading control
technology to marine loading control technology.
Section 6.1 presents the cost evaluation techniques
used by Radian to evaluate the"cost effectiveness of marine
terminal vapor control systems and to assess the sensitivity of
control costs to unique situations in the Houston-Calveston area.
Section 6.2 discusses the results of Radian's cost effectiveness
and sensitivity study.
6.1 Establishment of Cases
The wide variety of marine terminal operations in the
Houston-Calveston area combined with the wide range of projected
control system costs greatly complicate the task of evaluating
the cost effectiveness and economic impact of installing marine
terminal controls. In an attempt to simplify this task, it was
decided to develop cost evaluations for several typical situations
which might occur in the Houston-Calveston area.
Table 6.1-1 presents statistics on the vapor recovery
systems proposed by the six major shippers of motor gasoline in
the Houston-Galveston area. Marathon Oil Company and Texas City
Refining Inc. share dock facilities and propose to construct a
-164-
-------
TABLE 6.1-1
STATISTICS ON THE PROPOSED HOUSTON-GALVESTON VAPOR
Petroleum
Company
Exxon
AMOCO
£ She11
Ui
1 Marathon/TCR
ARCO
1975 Marine
Facility Throughput
106 bbl/yr
33.5
20
16.6
12.3
3.9
Projected
Recovery Unit
Size
10 3 bph
50
18
25
30
16
No. of Vessels
to be Modified
by Company
Ship
9
3
7
0
2
s Barges
0
0
0
0
RECOVERY SYSTEMS
Volume of Gasoline
Transported
106 bbl/yr
Ships
27.3
12
16.6
4.5
3.5
Barges
6.2
8
0
7.8
0.4
Refinery ,
Crude Capacity
1975
103 bbl/cd
390
333
294
150
213
Note: 1. AMOCO has 35 barges on charter which will pass modification costs on to AMOCO as a charter
price increase.
-------
jointly owned vapor recovery system. Projected recovery unit
sizes range from 16 to 50 thousand barrels-per-hour vessel loading
rate. The ratio of vapor recovery unit size to marine terminal
throughput ranges from 1 to 4 bph per 103bbl/yr. The number
of company owned vessels which must be modified varies greatly
in the H-G area. Many companies charter or contract shipping vessels
and will not be required to put forth the initial capital outlay
to modify marine vessels. There is also a wide variation in the
split between the volume of gasoline transported by tanker and
the volume of gasoline transported by barge among the H-G marine
terminals. From 070 to 6070 of a company's marine transported
gasoline may be transported by barges.
In addition to variations in the statistics on proposed
vapor control systems, there are also wide variations in the esti-
mated costs for marine loading vapor control systems. Table 6.1-2
summarizes projected cost data for the installation and operation
of marine loading vapor control systems. Installed capital costs
for all shoreside portions of marine loading vapor control systems
ranged from 0.5 to 3.5 million dollars per 10,000 bph capacity on
comparable units. The average of the shoreside system capital
costs supplied by industry appeared to be.2 million dollars per
10,000 bph capacity. An independent Radian cost study estimated
the total shore side system capital costs to be $800,000 per
1.0,000 bph capacity. The variations in projected ship modification
costs ranged from 0.15 to 1.0 million dollars per ship, and ave-
raged approximately 0.35 million dollars per ship. Estimates of
barge modification costs more closely centered around $67,000
per barge. Annual operating costs reported in Table 6.1-2 com-
prise maintenance and utility costs. The annual operating costs
for, the shoreside portion of vapor recovery systems are projected
to range from $11 to $25 per thousand barrels transferred. The
average projected cost was $15 per thousand barrels. Only one
source projected operating costs for the onboard portion of the
vapor collection system, and they were $110,000 per ship and
$36,000 per barge.
-166--
-------
TABLE 6.1-2
SUMMARY OF COST DATA FOR MARINE TERMINAL CONTROLS
Installed Capital Costa
Vapor Recovery Units (lO^/lQ1* bph)
Vapor Collection System (lO^/lO*1 bph)
Ship Modifications (103$/ship)
Barge Modifications (103$/barge)
Ocean Barge Modifications (103$/barge)
Annual Operating Costs
Vapor Recovery Systems ($/103 bbl)
Vapor Collection Systems
Ship Collection Systems ($/ship)
Barge Collection Systems ($/barge)
. Ranee
11 - 25
neg
Average
15
neg
110,000*
36,000*
Radian Study
300 -
100 -
150 -
50 -
_
1300
2000
1000
85
800
1000
325
67.5
150*
600
200
-
-
_
* Only one value was gjven in these cost categories
-------
Based upon the information presented in Table 6.1-1
and Table 6.1-2, Radian selected ten unique cases for which to
develop cost information. Table 6.1-3 presents these ten cost
cases. These.ten cases incorporate the variations expected among
the vapor control systems proposed for marine terminals in the
Houston-Galveston area. Case 1 represents the median among the
marine terminals and proposed control systems and, therefore,
should also represent the aggregate of the Houston-Galveston
marine terminals.
(
Six parameters are varied among the ten cost cases.
The ratio of recovery unit size to terminal throughput is varied
from 1 to 4xl03bph/106bpy'. The projected capital costs of the
shores ide systems range from $0.80 to $3. OxlOVlO^bpd. Ship and
barge modification costs range from $325,000 and $68,000 per
vessel, respectively, to $650,000 and $136,000 per vessel, respectively.
Annual operating costs range from $15 to $25 per barrel transferred.
The operating costs projected by one company for onboard control
equipment were omitted because of their uncertainty. The volume
ratio of gasoline transported by ships/barges ranges from equal
volumes transported by ships and barges to all gasoline trans-
ported by ships.
The vessel modification optimization efficiency reported
in Table 6.1-3 is an attempt to account for all of the ships and
barges which will require modification. Although most barges
and many ships loading gasoline in the Houston-Galveston area
are not owned by the petroleum companies, the costs for their
modification will be passed on to the refineries either directly
or indirectly. The task of determining how many vessels will
require modification per company is complicated by the fact that
many vessels call at more than one Houston-Galveston terminal
during the year. Data from Exxon and Shell indicate that a typical
ship transports 5xl06bbl/yr, and a typical barge transports 0.4xl06bbl/yr
-168-
-------
TABLE 6.1-3
Case
1
2
3
, *
5 5
VO
1 6
7
8
9
10
Ratio of
Recovery Unit
Size to Terminal
Throughput
(103bph/106bpy)
2
2
2
1
4
2
2
2
2
2
Capital Cost
of Shoreside
System
(106$/10,000 bph)
0.8
2.0
3.0
0.8
3.0
0.8
0.8
0.8
0.8
0.8
SUMMARY OF CASE
Capital Cost
of Vessel
Modification
(103$/ship :103$/barge)
325:68
325:68
325:68
325:68
325:68
325:68
"325:68
325:68
325:68
650:136
PARAMETERS
Annual Operating
Cost of Recovery
Unit
($/103bbl)
15
15
15
15
15
15
15
15
25
15
Volume
Ratio of Product
Transported by
Ships/Barges
3
3 •
3
3
3
1
all ships
3
3
3
Efficiency in
Vessel Modification
Optimization
(%)
50
50
50
50
50
50
50
75
50
50
-------
These estimates were combined with the total volumes of gasoline
transported by ship and by barge and the vessel modification
optimization efficiency to provide a projected number of ships
and barges which must be modified.
6.2 Methodology
The methodology used to calculate the cost and effective-
ness of each case is discussed in this section. Table 6.2-1 presents
the results of these calculations.
The cost and effectiveness- calculations used to evaluate
each cost case are based upon a terminal throughput of 10xl06bbl/yr
of gasoline. Standardizing marine terminal throughput will not
impact the relative cost effectiveness of the cost cases; yet it
will put all results on a common basis.
All annualized capital costs developed for the vapor
control systems are based upon a service life of 15 years and
an annual interest rate of 12%.- Although various portions of
the vapor control systems have shorter service lives, it is
believed that the major capital components of vapor recovery
systems do have a service life of 15 years. It is also true that
prime interest rates are lower than 12%; however, the before tax
investment potential of petroleum industry capital is likely to
be in the range of 1270 or better. The annualized cost factor
was calculated to be 0.147.
The quantity of gasoline recovered by application of
vapor controls on marine loading of gasoline is assumed to be
0.5 lb/103gal transferred for ships and assumed to be 3.0 lb/103
gal transferred for barges. These two values were based on the
following information. Uncontrolled hydrocarbon emissions from
-170-
-------
TABLE 6.2-1
RESULTS OF STUDY ON VAPOR RECOVERY ECONOMICS
Cose
Case
1
2
3
4
5
6
7
8
9
10
Capacity
Marine of Vapor
Terminal Recovery
Throughput Unit
(10*bhl/yr) (103bbl/hr)
10
10
10
10
10
10
10
10
10
10
20
20
20
10
40
20
20
20
20
20
Shore-
side
Capital
Cost
1.6
4
6
0.8
12
1.6
1.6
1.6
1.6
1.6
An nu -
ulized
Shore-
side
Cost
(io}$)
235
588
882
118
1764
235
235
235
235
235
Annua 1
Operating
Cost of
Recovery
Unit
0.22
0.07
0.06
0.06
O.OH
0.09
-------
vessel loading in the Houston-Calves ton area may average 1.3
lbs/103gal transferred for ships and 4 lbs/103 gal transferred
for barges. Controlled hydrocarbon emissions for ships will be
below the 1 lb/103 gal proposed by regulation because the arrival
vapors of many clean ships contain less than 0.1 lb/103gal.
Therefore, the controlled hydrocarbon concentration for ships
was assumed to be 0.8 lb/103gal. This phenomena is discussed
further in Section 4.1.5. The controlled hydrocarbon emissions
from barge loading, were assumed to be the maximum allowed by the
proposed regulation, or 1.0 lb/103gal throughput.
The number of ships and barges which require modification
was calculated from the volumes of gasoline transported by ship
and by barge assuming that a typical tanker transports 5xl06bbl
per year, and a typical barge transports 0.4xl06bbl per year.
These two factors were derived from data supplied by Exxon and
Shell on their marine vessel movements. The number of ships and
barges calculated from the factors was divided by a scheduling
efficiency factor which accounts for extra vessels requiring
modification because of scheduling inefficiencies.
6.3 Results
The cost effectiveness and economic impact of applying
marine terminal vapor controls can be evaluated by studying the
results of the ten cost cases which are summarized in Table 6.2-1.
6.3.1 Base Case
Cost case 1 is the base case and represents what appears
to be both the typical vapor recovery system and the aggregate
vapor recovery technology which will be applied in the Houston-
Galveston area. Input data for the 10 million barrel per year
gasoline terminal in case 1 include the following:
-172-
-------
• 20,000 barrel per hour gasoline loading rate.
• 1.6 million dollar capital cost for shore side
equipment (based on results of a Radian cost study).
• 7.5 million barrels per year of gasoline is trans-
ported on 3 ships.
• 2.5 million barrels per year of gasoline is trans-
ported on 13 barges.
o 462 thousand pounds of gasoline per year is
recovered by the recovery unit.
• Vessel modification costs (based on industry data)
are 325 thousand dollars per ship and 68 thousand
dollars per barge.
Based on construction cost estimates developed in the
Radian cost study (Appendix VII), the total capital investment
of the base case vapor recovery system is approximately 3.5
million dollars or 0.35 dollars per yearly barrel of marine
terminal capacity. The annual capital cost of the vapor re-
covery system amortized over the expected equipment life is 514
thousand dollars; the annual operating cost is approximately
150 thousand dollars. These data yield a total annual cost for
the case 1 vapor recovery system of approximately 0.7 million
dollars. The net cost effectiveness of the case 1 vapor recovery
unit is estimated to be 2900 per ton of hydrocarbon recovered.
The net economic impact of the case 1 vapor recovery unit is
estimated to be 0.07 per barrel of gasoline loaded at the
terminal.
-173-
-------
6.3.2 Sensitivity to Cost Inputs
A wide range was observed in the capital cost esti-
mates provided by industry for marine terminal vapor recovery
systems. Industry cost estimates for total shoreside capital
costs ranged from a low of approximately $500,000 per 10,000
bbl/yr loading capacity to a high of approximately $3,500,000
per 10,000•bbl/yr loading capacity. The Radian vapor recovery
system cost study estimated the total shoreside capital costs
to be approximately $800,000 per 10,000 bbl/yr. The predominantly
higher petroleum industry cost estimates are assumed to incor-
porate high contingency factors to account for possible scale-up
problems associated with the construction of the first marine
vapor recovery systems. Cost Case 1 utilizing the Radian capital
cost estimate also represents the lower range of industry cost
estimates. Case 2 and Case 3 were designed to investigate the
sensitivity of the cost effectiveness and economic impacts pro-
jected in Case 1 to the higher capital costs reported by the
petroleum industry. In Case 2 the shoreside vapor recovery
system capital costs were estimated to be 2 million dollars
per 10,000 bph of capacity or 4 million dollars. This figure
represents the median range of industry-reported capital costs
for vapor recovery systems in the Houston-Calves ton area.
Table 6.2-1 indicates that the cost effectiveness of Case 2
would be $4,400 per ton of hydrocarbons recovered, and the
economic impact of Case 2 would be $0.10 per barrel of gasoline
loaded.
In Case 3 the shoreside vapor recovery system capital
costs were estimated to be 3 million dollars per 10,000 bph of
gasoline loading capaicty, or 6 million dollars. This case
represents the upper range of capital costs reported for the
proposed vapor recovery systems in the Houston-Calveston area.
-174-
-------
The cost effectiveness of case 3 is estimated to be $5700 per ton
of hydrocarbons recovered and the economic impact is estimated
to be $0.13 per barrel of gasoline loaded.
Case 10 investigated the effect of doubling the cost
of vessel modifications. Ship modification costs were inputted
as 650 thousand dollars per vessel and barge modification costs
were inputted as 136 thousand dollars per vessel. Table 6.2-1
indicates that the cost effectiveness of Case 10 is $4100 per
ton of hydrocarbons recovered and the economic impact is $0.09
per barrel of gasoline loaded. Although most cost data indicate
that vessel modification costs will probably be much closer to
the base case cost than the case 10 cost, it is apparent that
changes in vessel modification costs have as large an impact on
cost effectiveness and economic impact as changes in shoreside
capital cost.
Case 9 investigates the impact of a vapor recovery
system operating cost at $25 per thousand barrels instead of
$15 per thousand barrels. This case represents the upper range
of projected operating costs reported by the petroleum industry.
The results of Case 9 indicate that higher operating costs would
raise the cost effectiveness of the base case to $3300 per ton
of hydrocarbons recovered and would raise the economic impact
of the base case to $0.08 per barrel of gasoline loaded.
From the results of the cost cases investigating sen-
sitivity to cost parameters it can be concluded that higher shore-
side capital costs or higher ship modification costs as estimated
by the petroleum industry could potentially raise the base case
cost effectiveness by $1500 to $2800 per ton of hydrocarbons re-
covered and could potentially raise the base case economic impact
by $0.03 to $0.06 per barrel of gasoline loaded. Unlike the im-
pact of changes in modification and capital costs, higher operating
-175-
-------
costs only raise the cost effectiveness of the base case by $400
per ton of hydrocarbons recovered and raise the economic impact
of the base case by $0.01 per barrel of gasoline loaded. The
cost effectiveness and economic impact of vapor recovery systems
are much less sensitive to operating cost than they are to
capital cost and vessel modification costs.
6.3.3 Sensitivity to Unit Size
The size of the vapor recovery system relative to the
marine terminal throughput is a very important factor. The
normal rate at which a single ship is loaded represents a mini-
mum capacity for vapor recovery units. This minimum capacity
would be required even if no more than one ship were loaded per
month. Under these conditions a vapor recovery unit size to
terminal size ratio of 4x103bph/106bpy could occur in the Houston-
Galveston area. The cost to capacity ratio is also likely to be
higher for smaller vapor recovery systems. Case 5 investigates
the impact of a unit to terminal size ratio of 4x103bpy/106bpy
.and a shoreside capital cost of $3x106/10,000 bph. The cost
effectiveness of Case 5 is $9500 per ton of hydrocarbons recovered,
and the economic impact is $0.22 per barrel of gasoline loaded.
The dramatic increases in cost effectiveness and economic impact are
in part attributable to the fact that the increased annualized
cost of Case 5 is not offset by any increase in hydrocarbon recovery.
In a similar manner, larger dock facilities may be able
to install a smaller vapor recovery unit relative to their yearly
gasoline throughput. Case 4 investigates the cost effectivenes
and economic impact of installing a vapor recovery system with
a recovery unit size to terminal throughput ratio of Ixl03bph/10s
bpy. The resulting cost effectiveness of Case 4 is $2400/ton of
hydrocarbons recovered and the economic impact of Case 4 is
$0.06/barrel of gasoline loaded.
-176-
-------
From the results of Case 4 and Case 5 it can be concluded
that a high volume marine terminal installing vapor recovery systems
may experience a cost effectiveness and economic impact lower than
the base case for the Houston-Calveston area. It is also apparent
that-a low volume marine terminal installing a vapor recovery system
may experience a cost effectiveness and economic impact much higher
than the base case for the Houston-Calveston area.
6.3.4 Sensitivity to Vessel Mix
Three cases were established to investigate the sensi-
tivity of cost effectiveness and economic impact to the relative
volumes of gasoline loaded onto barges and ships and to the number
of ships and barges requiring modification. Because fewer hydro-
carbon vapors are generated and, consequently, recovered during
ship loading than during barge loading, it is felt that cost
effectiveness is highly, dependent on the mix of ships and barges
loading at the terminal. Case 6 investigates the impact of loading
equal volumes of gasoline onto ships and barges. Case 7 investi-
gates the impact of loading all gasoline onto ships only. For
Case 6 loading equal volumes of gasoline onto ships and barges
reduces the cost effectiveness of marine terminal controls of the
base case to $1900 per ton of hydrocarbons recovered while the
economic impact of controls remains relatively the same at $0.07
per barrel of gasoline loaded. For Case 7 loading all gasoline
onto ships raises the cost effectiveness of marine terminal controls
for the base case to $5500 per ton of hydrocarbons recovered, while
the economic impact of controls remains relatively the same at
$0.06 per barrel of gasoline loaded.
The base case conservatively assumed that petroleum
companies and shipping companies would find it necessary to modify
twice the optimum number of ships and barges required to transport
-177-
-------
their yearly gasoline shipments. Case 8 investigates the cost
effectiveness and economic impact of more efficient shipping
operations which require that only 1.5 times the optimum number
of vessels be modified. In Case 8 two ships and nine barges
are modified as opposed to three ships and thirteen barges as
required in the base case. The cost effectiveness of Case 8 is
$2500 per ton of hydrocarbons recovered and the economic impact
of Case 8 is $0.06.
The results from Cases 6, 7, and 8 indicate that the
ratio of gasoline loaded onto ships to gasoline loaded onto barges
has a much greater impact on vapor control economics than does a
change in the number of ships and barges requiring modification.
The primary impact of loading ratios between ships and barges is
attributable to their impact on the volume of hydrocarbons
recovered and, consequently, to the cost effectiveness. The
effect of loading ratio to the economic impact of vapor controls
is minor. The effect of reducing the number of vessels requiring
modification is small and applies equally to the cost effectiveness
and to the economic impact.
-178-
-------
7.0 TEST PLAN DEVELOPMENT
The results of this program to develop background
information on marine terminal emissions and emission control
technology indicate several areas where further emission testing
is warranted. Section 7 develops a test plan for establishing.
emission data in those areas where emission data is inadequate.
7.1 Objective
As part of this program, Radian conducted a series of
emission tests to verify emission data which have been reported by
the petroleum industry. The results of the emission tests
indicate that industry reported values for the hydrocarbon
emissions from loading gasoline onto clean ship tanks and
ballasted ship tanks adequately represent marine loading emissions
in the Houston-Calveston area. The accuracy of petroleum industry
values for these emissions is within the range of sampling
accuracy and within the range of variations introduced by fluctua-
tions in undefined parameters.
Insufficient data was collected in the emission tests
to verify reported emission data for gasoline loading into
uncleaned ship tanks, uncleaned barge tanks, and cleaned barge
tanks. Although large quantities of industry data are available
for these emission categories, a program of spot testing is
required to establish their accuracy independently. A second
area of extensive industry data requiring verification is the
chemical composition of vented vapors. Wide ranges have been
reported for the volume of methane present in the vented vapors.
Methane is very difficult to control with conventional vapor
recovery units and may present a major problem to some terminal
facilities.
-179-
-------
Studies on available emission data have also identified
the following areas where practically no data is available on marine
terminal emissions:
crude loading - ships and barges
crude ballasting - ships
gasoline ballasting - ships
chemical and fuel loading - ships and barges
chemical and fuel ballasting - ships
It has been apparent that a composite vapor profile for ships
loading gasoline and crude oil is important in assessing the
hydrocarbon reduction potential of vapor recovery systems pri-
marily treating ship loading vapors. Average emission factors
per cargo tank do not present as complete a picture of loading
emissions as does a plot of the composite vented vapor profile.
The major geographical areas of concern in this study
are the Houston-Galveston area and the West Coast. Gasoline
ballasting and chemical and fuel loading and unloading opera-
tions are not primary sources of hydrocarbon emissions in these
areas.
Based on this information, Radian developed the following
test plan to (1) quantify and characterize hydrocarbon emissions
from crude loading and crude ballasting operations, (2) verify
reported emissions from gasoline loading onto uncleaned ships
and barges, and (3) characterize the composite vapor profile
from operations loading gasoline onto ships.
-180-
-------
7.2 Approach
This section outlines the basic approach to a concise
point source testing program to effectively achieve the stated
test program objectives.
7.2.1 Results Format
The proposed test plan is designed to develop emission
data in the form of emission factors. Evaluation of emission
testing results indicate that emission correlations and equations
for marine terminal operations are difficult to develop and
generally do not predict emissions with significantly greater
accuracy. This is in part due to the inability of correlations
to account for the large number of parameters affecting the
emission rate.
The emission factors will be in units of pounds of
hydrocarbons per thousand gallons of gasoline and categorized
by arrival condition. Radian proposes the following arrival
condition categories:
Loading Operations
Clean Tanks - Cargo tanks with very low
arrival components attribut-
able to either the previous
cargo being non-volatile or
to.tank cleaning on the return
voyage.
Ballasted Tanks - Cargo tanks which carried
ballast water on the return
voyage and which did not under-
go any form of cleaning.
-131-
-------
Uncleaned Tanks - Cargo tanks which were
neither cleaned nor
ballasted on the return
trip.
• Ballasting Operations
Full Tanks - Cargo tanks arriving
completely filled
Partially Filled - Cargo tanks arriving
partially filled.
Stripping Tanks - Cargo tanks used to collect
strippings from other tanks
prior to pumping them to
shore.
The results of the tests to characterize the composite
vented vapor profile will be presented in the form of composite
vapor concentration-vs-time for the duration of the vessel loading
operation. This would be equivalent to a concentration-vs-time
profile for the vapor vent of a ship having all tank vents mani-
folded into a single vent.
7.2.2 Parameters
Emission factors for each tank will be obtained by
measuring the hydrocarbon concentration in the vented vapor
when the liquid level is at given ullages. From a plot of hydro-
carbon concentration-vs-ullage it is possible to obtain a volumetric
average hydrocarbon concentration for the tank filling operation
by integration of the plot.
-182-
-------
An emission factor is calculated from the volumetric
average hydrocarbon concentration using equation-(1).
1 + /Y-Y<
(1)
when F = emission factor (lbs/1000 gallons transferred)
Y0 = avg arrival vapor concentration (%)
YI = volumetric average vented hydrocarbon concentration (7o)
MW = molecular weight (Ib/lb-mole)
T = temperature of vented vapor (°R)
T /Y,-Yn \1
The term 1 +[—* &• l corrects for the expansion in vented vapor
due to evaporation of the product during loading operations.
When calculating ballasting emissions this term should be set
equal to 1.
The proposed method for obtaining an emission factor is
based upon ullage measurements instead of direct vapor volume
measurements-. Although direct vapor volume measurements are
potentially more accurate, gas meters and flow instruments large
enough to directly measure vapor flow are too bulky to be con-
veniently moved from ullage hatch to ullage hatch and from ship
to ship.
The parameters required to generate composite vented
vapor profiles are ullage levels-vs-real time, tank volume, and
cruise history for each cargo tank on the vessel. Real time
is meant to be actual clock time recorded, so that the rela-
tive stage of fill for each cargo tank will be known with
respect to the stage of fill for each of the other cargo tanks.
Combining this information with information on tank volumes,
cruise histories for each tank, and standard vapor concentration
-183-
-------
profiles for the cruise histories represented, it is possible to
develop a graph of hydrocarbon concentration in the composite
vented vapors-vs-time for the entire vessel loading or ballasting
operation.
*
7.2.3 Required Level of Sampling
The level of sampling required to produce meaningful
results is very difficult to predetermine. Decisions on how
many tests should be conducted should be re-evaluated as the
test program develops based upon the consistency of data collected.
When conducting tests to verify existing emission factors it will
suffice to test the vapors vented from three or more cargo tanks
on two or three vessels for each emission factor. When testing
for the purpose of establishing emission factors it is necessary
to collect more emission data than that is required to verify
existing emission factors. It will be advisable to test three or
more cargo tanks on a minimum of six vessels for each emission
factor.
When collecting vapor samples from gasoline loading
for determining chemical composition operations, it is important
to collect several vapor samples from cargo tanks in various stages
of the loading sequence. A set of gasoline vapor samples should
be collected for each company loading gasoline because of the
wide variations in vapor composition which have been observed
among companies. In sampling crude oil vessels several vapor
samples should again be collected from cargo tanks in various
stages of loading or ballasting. Sets of crude oil vapors should
be collected for each major category of crude transferred.
-184-
-------
The level of sampling required to establish meaningful
composite vented vapor profiles is unknown because the extent
to which these profiles vary among ships is undetermined. How-
ever, at this time it is estimated that three or four composite
vapor profile tests will provide an accurate indication of com-
posite vapor characteristics.
7.2.4 Test Program - Instrumentation
Gasoline vapors are composed principally of saturated
hydrocarbons. The components may typically range from Ci to Ci0
or even higher. Straight chain hydrocarbons as well as their
isomers tend to make the mixture even more complex. These com-
pounds are similar chemically, making component analysis of the
vapors difficult. Also, it has been proposed that methane be
excluded from current hydrocarbon regulations. Such regulations .
would require not only the determination of the total hydrocarbon
present but also the methane content. This section presents
information on the instrumentation available for measuring total
hydrocarbon concentration and for analyzing hydrocarbon components
In choosing the proper or best suited instrument for
obtaining hydrocarbon emission information from gasoline loading
of marine vessels, certain test parameters should be met by the
equipment. Some of these parameters are listed below:
• The equipment should not impair the time required
to sample. The data should be obtainable fairly
rapidly.
• With the result of the program being dependent to
a large part on the accuracy of the instrument, a
minimum accuracy of ± 10 percent should be achievable.
-185-
-------
The results should be highly reproducible by
• following tests.
The costs which are dependent on several factors,
should be kept to a minimum by considering the
following:
1) Initial equipment cost
2) Equipment reliability
3) Maintenance costs
4) Skill level required of operators
5) Data reduction time
6) Data interpretation
7.2.5 Hydrocarbon Analysis
The characterization of saturated hydrocarbons relies
on tests which measure physical properties such as boiling point,
thermal conductivity, infra-red absorption, etc. For this reason,
analysis of pure compounds is accomplished by a variety of test
methods. The analysis of mixtures of hydrocarbons in terms of
total hydrocarbon content also can be done without great difficulty.
However, the quantitative analysis of a sample (by individual com-
ponent) requires much more complicated testing.
The type of analysis required in sampling gasoline load-
ing emissions is directly associated with the degree of sophisti-
cation necessary in both equipment and operator skills. Should
regulations for marine gasoline loading be revised to exclude
methane, complicated test procedures would be needed.
The following discussions present information on the
current state-of-the-art of methods which are applicable to
direct hydrocarbon analysis.
-186-
-------
Combustible Gas Indicator
The combustible gas indicator is designed to measure
the total concentration of hydrocarbons up to a maximum detectable
concentration which corresponds to the Lower Explosive Limit
(LEL) of the hydrocarbon mixture. This limitation is due to
insufficient oxygen being present at higher hydrocarbon levels
to insure that complete combustion will take place. For this
reason direct application of this type of instrument is limited
to vapors whose concentrations-are below the LEL. Therefore, if
the stream being analyzed has a greater concentration of hydro-
carbon, the sample must be diluted to the proper concentration
range prior to analysis. This dilution step, though, tends to
lower the accuracy of this method due to the errors associated
with this additional handling step.
The hydrocarbon concentration is measured in this analyzer
by monitoring the output voltage of a balanced Wheatstone bridge
circuit. Part of the bridge circuit is a heated wire filament
which burns any combustible gas which enters as a sample. As the
temperature of the filament rises, its resistance also increases,
causing a change in the output voltage of the bridge. The
magnitude of the output voltage is proportional to the concentration
of hydrocarbon(s) in the sample.
Thermal Conductivity Meter
A thermal conductivity meter measures thermal conduc-
tivity of a sample relative to air by means of a heated filament.
The filament is part of a balanced Wheatstone bridge. It is
cooled by the hydrocarbons in the sample and causes a change in
the conductivity of the filament. This, in turn, changes the
electrical output of the circuit an amount .proportional to the
concentration of the hydrocarbon(s) in the sample.
-137-
-------
The meter must be calibrated with a known hydrocarbon
over the range of 0 to 100 percent. For best results the cali-
bration gas mixture should approximate the composition of the
stream to be sampled.
Infra-Red Analyzer
Infra-red analysis of saturated aliphatic hydrocarbons
shows a characteristic absorption peak in the infra-red region
at 3.4 microns. The particular stretching frequency of the carbon-
hydrogen bonds in the hydrocarbon molecules account for this.
However, the spectra of the homologous series Ci to Ca are so
similar that there is no practical way to distinquish these
compounds on the basis of infra-red absorption. Nonetheless,
infra-red spectrometers are very accurate instruments for analyz-
ing total hydrocarbon concentration in a stream.
There are two techniques available to quantitatively
determine the hydrocarbon concentration in a mixture by infra-red
analysis. For one method the spectrometer is set to detect the
percent absorption at 3.4 microns. Since the hydrocarbons absorb
energy in proportion to their concentration, the total hydrocarbon
concentration can be determined. This method can be used to mea-
sure concentrations on a continuous basis by feeding the instru-
ment cell with a constant sample flow. The second method is based
on a direct comparison between the sample and a known standard.
The standard (sealed in a cell) and the sample are both exposed to
the same infra-red source. Each cell will generate an electrical
signal whenever the gas in it absorbs a quantum of energy which
causes it to heat up and expand. The instrument is standardized
by placing identical gases in each cell, and the output is nulled
to zero. During testing a sample is drawn into the sample cell.
Then, both cells are irradiated; and any electrical output is an
indication of a compositional difference between the sample and
-138-
-------
the standard. By using known mixtures the instrument may be
calibrated to give a direct readout of the total hydrocarbon
concentration.
Gas Chromatography
With proper calibration the gas chromatograph is capable
of determining total hydrocarbon concentration and of quantitatively
measuring multi-component mixtures of hydrocarbons. This sampling
method requires expensive equipment and skilled personnel to assure
reliable results. The time required to process a sample is roughly
20 minutes but could be shorter in a routine situation. For this
reason it is not suitable for continuous stream analysis. It
would best serve in testing for the component analysis of the
vapors vented from a particular cargo tank.
Mass Spectrometry
A mass spectrometer can analyze a hydrocarbon sample
in terms of the individual components present as well as in terms
of the total hydrocarbon concentration. It operates by converting
molecules into ions and by separating these ions on the basis of
their mass/charge ratios. This instrument is probably the most
accurate hydrocarbon analyzer available. However, as with the gas
chromatograph, it does not lend itself to continuous applications;
the equipment is quite expensive; and a very high operator skill
level is required for reliable results.
Oxygen Analyzers
Measuring the oxygen concentration in a hydrocarbon
sample from a tank vent stream is an indirect method of determin-
ing total hydrocarbon content. The principle is simple, though
-189-
-------
From the oxygen measurement the amount of air in the sample can
be readily calculated, assuming the oxygen is present as a com-
ponent of air. However, the problem that exists in using oxygen
as a basis for calculating hydrocarbon content is that any error
in detection is magnified five times. Therefore, a great deal
of uncertainty exists using this technique if the hydrocarbon
content of a sample is low. For sample streams containing high
hydrocarbon content (low oxygen), though, this technique may be
more suitable since the analyzer can be run at a higher sensitivity,
thereby reducing hydrocarbon measurement error.
Currently there are four methods for quantitatively
measuring oxygen:
1) Selective absorption
2) Paramagnetic susceptibility
3) Polarographic analysis
4) lonization
Gas Densitometers
The gas densitometer technique calculates the hydro-
carbon loss based on the assumption that the difference between
the mass of vapor leaving a cargo tank and the mass of air
represents the mass emission of hydrocarbon. This technique also
assumes that the difference in the masses of air entering and
leaving the cargo tank is negligible over a reasonable period
of time.
Densitometers based on one of three principles are
available:
-190-
-------
1) Bouyancy
2) Centrifugal force
3) Mass vibration
Summary
Numerous methods are available for analyzing the hydro-
carbon concentration of vent streams from marine vessels loading
gasoline or crude oil and from tankers ballasting after unloading
gasoline or crude oil. All equipment must be handled properly
and correctly calibrated to assure accurate results. Only two
instruments, though, offer the ability to analyze for individual
hydrocarbon components in a vent stream - gas chromatography and
mass spectrometry.
Several instruments are portable and are reasonably
accurate in measuring total hydrocarbon concentrations, e.g.,
combustible gas indicators and thermal conductivity meters.
Some are more expensive than others, but they offer measurement
accuracy, e.g., infra-red spectrometers.
Besides accuracy and expense, another important con-
sideration for a testing apparatus is its safety. On board a
marine vessel a testing apparatus must meet stringent safety
specifications. All equipment and instrumentation used must be
labeled intrinsically safe or be suitable for use in Class 1,
Division 1 atmospheres.
-191-
-------
7.3 Sampling Procedure
This section outlines a procedure which could be used
for obtaining emission factors for marine vessels loading gasoline
and crude oil as well as for crude oil and gasoline ship ballasting
operations. The basic principle involved in'this procedure is to
record with an accurate hydrocarbon analyzer the hydrocarbon con-
centration of the vented vapors as a function of ullage for both
loading and ballasting. This information along with a tank's final
ullage and average hydrocarbon concentration just prior to loading
or ballasting can be used to generate reasonably accurate emission
factors.
7.3.1 Test Measurements
7.3.1.1 Vented Vapor Concentration Profile
Measurements during actual loading or ballasting
operations shall be recorded for the hydrocarbon concentrations
of the vented vapors using a suitable instrument. The probe
shall be positioned as shown in Figure 7.3-1.
A. Frequency Of Measurements
Ships - From maximum tank ullage to the 20 ft.
true ullage mark, measurements will be
recorded every 5 ft. From 20 ft. to 10
ft. true ullage, measurements are recorded
every 2 ft. From 10 ft. to the final
ullage, measurements are recorded every
1 ft. See Figure 7.3-2.
-192-
-------
TO
TOTAL
HYDROCARBON
ANALYZER
ULLAGE
TRUNK
ULLAGE TAPE
REFERENCE
POINT
ULLAGE VENT
TO PV
VALVE
END OF SAMPLE PROBE
IN PLANE OF DECK
TRUE ULLAGE REFERENCE POINT
FIGURE 7.3-1. LOCATION OF SAMPLE PROBE
-193-
-------
CONCENTRATION MEASURMENTS
#
•
_J
o
CO
O
O
UJ
z
UJ
O
00
cr
<
o
O
cc
Q
EVERY 5 FT. _,
EVERY 2 FT.
«
•
i
ARRIVAL VAPORS
®/-\
1
EVERY 1 FT.
c
4? /
o i
a /
^ /
et
//
7~
1
20 10
TRUE UL'LAGE, FT.
O SAMPLE POINTS
O SAMPLE MAY BE OMITTED ON DUPLICATED RUNS
NOTE: COMPOSITION SAMPLES SHOULD BETAKEN
AT THE INDICATED POSITIONS RELATIVE TO
PROFILE; ULLAGE VALUES SHOWN ARE
ILLUSTRATIVE ONLY
FIGURE 7.3-2 SAMPLE POINTS RELATIVE TO TRUE ULLAGE
(CONCENTRATION) AND VAPOR PROFILE (COMPOSITION)
-194-
-------
Barges - Since most barges have maximum true
ullages around 10 ft., measurements
will be recorded every 1 ft. to the
final ullage.
B. Information Recorded
The information to be recorded for each measure-
ment is time, true ullage, concentration, product
loaded, tank ID, and vented vapor temperature.
Emission factors can be calculated from this
information, however, more-data and information
are needed to categorize the emission factors
according to cruise history. The following
information will allow proper categorization
of the recorded emission data:
1. Previous cargo - type and RVP
2. Transit history
a. Transit time
b. Nature of tank cleaning for each tank
c. Ballast handling, including ullage or
percent of tanks ballasted.
C. Vented Vapor Composition
Information will also be obtained on the composition
of the vapors vented from crude oil and gasoline
loading and ballasting operations. This data will
not only allow a component analysis of the
emissions for reactivity classification, but
-195-
-------
it will also provide a check for the concentra-
tion measurements taken.
For crude oil and gasoline loading, samples will
be collected at six different points during the
loading. Figure 7.3-2 shows the sample points
with respect to the vented vapor concentration
profile. The locations are:
1) just after loading begins
2) midway through the horizontal leg
3) just before the inclined leg
4) just after the start of the inclined leg
5) midway along the incline
6) at the final ullage
If a number of samples are taken for the same test
conditions at the same refinery, compositional
vapor samples are needed for only two tanks.
Also, samples No. 2 and No. 6 above may be
eliminated after the first test for each test
condition.
The following information shall be attached to
a tag on each sample cylinder immediately after
sampling and also recorded on a separate data
sheet for record:
1) date and time
2) terminal name and location
3) tanker name
4) tank I.D.
5) true ullage .reading at time of sampling
-196-
-------
6) temperature and pressure in tank at time
of sample
7) ambient temperature
8) name of person sampling
For ballasting operations and barges loading
gasoline or crude oil, samples shall be taken
just after the' start, at the midpoint, and at
the final ullage. The same information listed
above shall be attached to each of the samples.
7.3.2 Recorded Information - Data Sheets
The information and data taken during a test run shall
be recorded on data sheets. Examples of a format which might
be used follows on the next few pages.
-197-
-------
TEST PROGRAM FOR MARINE EMISSIONS
SHORESIDE INFORMATION
DATA SHEET I
General Information: .
Date
Name of Vessel
Terminal
Location
Product(s) Loaded
Ambient Conditions:
Air Temperature
Weather Conditions
Prepared by:
-198-
-------
TEST PROGRAM FOR MARINE EMISSIONS
SHORESIDE INFORMATION
DATA SHEET II
General Information:
Date
Name of Vessel
Type ship ; barge
Number of cargo tanks
Vessel size (DWT)
Cruise History:
Transit Time
Do tanks have stripper
lines?
Open or closed hatches?
From ship log, sketch cargo tank layout below and show prior
cargo arrangement, and the type of cleaning for each tank and
which tanks were ballasted on the return trip:
Prepared by:
-199-
-------
TEST PROGRAM FOR MARINE EMISSIONS
RECORDED DATA
DATA SHEET III
Date:
Product Loaded
Cargo Tank No.
Approx. Loading Rate
Time
Ullage (ft)
True
Tape
Concentration
Temperature ( F)
Vapor Liq.
Composition
Sample(/)
-200-
-------
•BIBLIOGRAPHY
1. American Petroleum Institute, Basic Petroleum Data
Book, Petroleum Indus try Statistics, Washington, D.C.,
Oct. 1975.
2. Amoco File, EPA Region VI, Air and Hazardous Materials
Division, Air Programs Branch, Technical Support Section,
Dallas, Texas, 1976.
3. Amoco Correspondence, L. V. Durland, Refinery Manager,
Amoco Oil Company, Texas City, Texas, June 15, 1976.
4. Amoco Correspondence, J. G. Huddle, Coordinator, Air
and Water Conservation, Amoco Oil Company, Chicago,
Illinois, July 9, 1976.
5. Arco File, EPA Region VI, Air and Hazardous Materials
Division, Air Programs Branch, Technical Support
Section, Dallas, Texas, 1976.
6. Arco Correspondence, H. J. Grimes, Manager, Environmental
Engineering, Atlantic Richfield Company, Harvey, Illinois,
June 21, 1976.
7. Botros, M. , Private Communication, Marine Terminal
Operations Survey, Air Pollution Control District,
County of Los Angeles, Feb. 1976.
8. British Petroleum Correspondence, Gordon Wanless, British
Petroleum Company, New York City, New York, July, 1976.
-201-
-------
BIBLIOGRAPHY (Continued)
9. Bryan, R.J., et al., Air Quality Analysis of the Unloading
of Alaskan Crude Oil of California Ports, Final Report.
EPA Contract No. 68-02-1405, Task 10. Santa Monica, CA,
Pacific Environmental Services, Inc., Nov. 1976.
0
10. Charter File, EPA Region VI, Air and Hazardous Materials
Division, Air Programs Branch, Technical Support Section,
Dallas, Texas, 1976.
11. Crown Correspondence, W. L. Warnement, Manager Environ-
mental Engineering, Crown Central Petroleum Co.,
Houston, Texas, July 28, 1976.
12. Edwards Engineering Correspondence, Ray Edwards, President,
Edwards Engineering Corp., Pompton Plains, New Jersey, 1976
13. Environmental Protection Agency, Compilation of_ Air
Pollutant Emission Factors, 2nd ed. with supplements,
AP-42, Research Triangle Park, N.C., 1973.
14. Exxon File, EPA Region VI, Air and Hazardous Materials
Division, Air Programs Branch, Technical Support
Section, Dallas, Texas, 1976.
15. Exxon Correspondence, L. 0. Fuller, Supervisor of
Environmental Engineering, Exxon Company, Baytown,
Texas, June 14, 1976.
16. "Federal Energy Administration: Hands Off N. Tier Supply
Problem", Oil Gas J. 1976 (Aug. 9), 36.
-202-
-------
BIBLIOGRAPHY (Continued)
17. Hart, Lawrence, Private Communication, County of Santa
Barbara, Health Care Services, Air Pollution Control
District, Aug. 1976.
18. Kilgren, K. , A Program for the Measurement of_ Hydro-
carbon Emissions During Tanker Loading of Crude Oil in
Ventura County. Chevron Research Company, California,
April 15, 1976.
19. Marathon Correspondence, L. M. Echelberger, Environmental
Coordinator, Marathon Oil Company, Texas City, Texas,
April 22, 1976.
20. Marathon File, EPA Region VI, Air and Hazardous Materials
Division, Air Programs Branch, Technical Support Section,
Dallas, Texas, 1976.
21. Monsanto File, EPA Region VI, Air and Hazardous Materials
Division, Air Programs Branch, Technical Support Section,
Dallas, Texas, 1976.
22. Rauge, Jim, Private Communication, Ventura County APCD,
July 1976.
23. Shell Correspondence, R. V. Mattern, Superintendent,
Environmental Conservation, Shell Oil Company, Deer
Park, Texas, May 18, 1976.
24. Shell File, EPA Region VI, Air and Hazardous Materials
Division, Air Programs Branch, Technical Support
Section, Dallas, Texas, 1976.
-203-
-------
BIBLIOGRAPHY (Continued)
25. Texas City Refining Correspondence, P. D. Parks, Environ-
mental Coordinator, Texas City Refining Co., Texas City,
Texas, 1976.
26. Texas City Refining File, EPA Region VI, Air and Hazardous
Materials Division, Air Programs Branch, Technical Support
Section, Dallas, Texas, 1976.
27. i:Two Oil Lines from West Seen Needed", Oil Gas J. 1976
(Aug. 16), 58.
28. Union Oil Correspondence, R. Y. Salisbury, Senior
Environmental Engineer, Union Oil Company, Los Angeles,
California, July 8, 1976.
29. U. S. Bureau of Mines, Div. of Fuels Data, Crude
Petroleum, Petroleum Products, and Natural Gas Liquids:
1974, final summary, Washington, D.C., April 1976.
30. Wilson, Howard M., "Beefed-up Tanker Fleets Readied for
N. Slope Oil", Oil Gas J. 1976 (June 14), 23.
-204-
-------
CONVERSION FACTORS
The references used in developing this report generally
stated flows, capacities, weights, etc. in English measurement
units. The following table can be used to convert these measure-
ments to metric units.
To Convert From
To
Multiply .By
Ib
bbl
lb/103 bbl
scf
ton
gal
lb/103 gal
Ib/ton
Btu/bbl
ton
Btu
kg
1
kg/103l
Nm3
MT
1
kg/103l
kg/MT
kcal/1
kg
kcal
0.454
159.0
.002855
0.0283
0.9072
3.785
0.1199
0.5004
1.585
907.2
0.252
-205-
-------
TECHNICAL REPORT DATA
(Please read lusiructions on the reverse before completing)
REPORT NO.
EPA-U50/3-76-038a
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Volume I
Discussion - Background Information on Hydrocarbon
Emissions from Marine Terminal Operations
5. REPORT DATE
November 1976
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
C. E. Burklin, J. D. Colley and M. L. Owe.n
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Radian Corporation
8500 Shoal Creek Boulevard
P. 0. Box 9948
Austin, Texas 78766
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-1319, Task 56
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. Environmental Protection Agency
Research Triangle Park,
North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report presents results of a study to develop background information necessary
for the accurate assessment of hydrocarbon emissions from ship and barge loading and
unloading of gasoline and crude oil. The report assesses marine terminal facilities,
marine terminal operations, cruise history and product movement statistics, hydro-
carbon emission rates and characteristics, control technology state of the art, safety
considerations of marine terminal control technology and economics of controlling
marine terminal emissions. The report also includes the results of a detailed cost
analysis for a refrigeration and an absorption marine terminal vapor recovery system.
Data gathering activities focused on .the Houston-Calves ton area; however, information
was also assembled on hydrocarbon emissions from marine terminal operations in the
metropolitan Los Angeles area generated by handling of gasoline and crude oils,
including Alaskan north slope crude.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air Pollution
Control Equipment
Hydrocarbons
Marine Terminals
Ships & Barges
Gasoline Loading & Unloading
Crude Oil Loading & Unloading
Air Pollution Control
Mobile Sources
Hydrocarbon Emission
Control
Organic Vapors
13. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
205
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |