U.S. DEPARTMENT OF COMMERCE
National Technical information Service
Evaluation of Pollution Potential of
Proposed Hampton Roads Energy Company
Refinery, Portsmouth, Virginia
Pacific Environmental Sgrvices, Inc, Sonta Monica, Calif
Prepared for
Environmental Protection Agency, Research Triangle Park, N C
Nov 76
-------
BIBLIOGRAPHIC DATA
1. Ufport No.
3. Kcc ipicnt'^ Accession No
P6-267' 63?
4. Title and Subtitle
Evaluation of Pollution Potential of Proposed
Hampton Roads Energy Company Refinery, Portsmouth, Virginia,
5. Report Dare
November 1976
7. Author(s)
Leslie E. Norton, and Karl Kuecitke.
8. l-VHcrminp Organi7ation K< pt.
No.
rVrtormmp Organization Name ;iiid Address
Pacific Environmental Services, Inc.
Santa Monica, California
10. Project,''! ask Aork t.'nil
It. Contract/Grant No.
12. Sponsunn£ Organisation Name .ind Address
Environmental Protection Agency
Research Triangle Park, N.C.
Office of Air Quality Planning and Standards
13. T>pe of Report & Period
Coveted
14.
i 5. Supplementary Notes
(PC All/HF A01)
16. Abstracts
This report is an evaluation of air pollution potential associated witli the
proposed Hampton Roads Energy Company Petroleum Refinery to be located at
Portsmouth, Virginia, and was prepared for the U.S. Environmental Protection
Agency. Emission estimates in this report reflect the use of best available
control technology for both fugitive and point sources. The report analyzes parti-
culate, sulfur oxides, nitrogen oxides, and hydrocarbon emissions. The estimate of
emissions from the refinery complex includes the unloading of crude tankers, and
the loading of product barges at the marine facilities. Calculations by the
refinery contractor (Foster Wheeler Energy Corporation) are reviewed.
17. Key 'iords and Document Analysis. 17o. Dcscnptirs
Air pollution control
''Refineries
Management plannina
Sulfur
Chemical plants
Process charting
Design criteria
Numerical analysis
Sampling
Petroleum industry
Inplant processes
Equipment specifications
Sources
Sludge
Hydrocarbons
Plant location
Crude oil
Particles
17l>. Iden'.ificTs/Opcn-Kndrd Terms
liampton Roads Energy Company Petroleum Refinery
Air pollution sampling
Oil water separators
Portsmouth(Virginia)
Point sources
17c. C.OSATI Field Group 7A
Sulfur oxides
Nitrogen oxides
Hydrogen
Catalysis
Industrial wastes
Combustion products
Distillation
Catalytic chamber process
Merox unit
IS. Availability Statement
National Technical Information Service
Springfield, VA 22161
19. S,..ur,ty ( U* (1 SlS
Rcpor;)'
20. sev uiity i I.ISS ( 1 hi -
2 I . No. (>! (
245
PC AH/MF A01
uv AN-I AND i M-SI o
!H1S ! OKM
Kl i'Ki>!>
-------
.»=»
PB
y\ EPA-450/3-76-037
; ] November 1976
EVALUATION
\ POTENTIAL
OF PROPOSED
OADS ENERGY
COMPANY REFINERY,
yrrH, V:
"">».
^> - >,'&>.
"'-.''W.
^^-..: ;AV,
"vi *: .ft.
''V.
F:
rc-r:-vr of ;
I.-.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and \V'asto Mana«jrnuint
Office of Air Ouality Plannin" ami Standards
Research Triangle Park. North Carolina 277 { ?
ia^i
(EPRC'OUCtD Of
NATIONAL TECHNICAL
INFORMATION SERVICE
L: s tTtPlVCHT Of COWWIPCE
I'1?, SOI UU '" J-ltl
-------
£PA-J.50/3-76-037
EVALUATION OF POLLUTION
P
Tl
i * Jt-J-S.*. JL )
by
Leslie E. Norton and Karl Lucdtke
Pacific Environmental Services, Inc.
1930 14th Slrofl
Santa Monica, Cnlifotnia
Contract No. 68-02-1378
Task No. 26
EPA Projcrl Officer: Da\i(J ^. Markwordt
Prepared for
ENVIRONMENTAL PROTECTION \t;ENO
Officf of -\ir and Vk a-te \lanagrmcnl
Office of Air Quality Planninft and Slaiidaril.-;
Rest-arch Trianple ['ark. \ortS Carolina CT711
Novcmlier I9V6
-------
This report is issued by the Environment.-); Pro'ection Agency to report
technical data of interest to a limited number ol readers. Copies are
available free of charge to Federal employees, current contractors and
grantees, and nonprofit organizations in limited quantities - from thr-
Library Services Cilice (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 tc the Environmental Protection Agency
by Pacific Environmental Services, Inc. , 1930 14th Street, Santa Monica,
California 93404 , in fulfillment of Contract No. 68-02-1378, Task No. 26.
The contents of this report nre reproduced here) n as received from
Pacific Environmental Services, Inc. The opinions, findings, and
conclusions expressed are those of the author and not necessarily
those of the Environmental Protection Agency. Mention of company
or piroduct names is not to lie considered as an endorsement by the
Environmental Protection Agency.
Publication No. EPA-450/3-76-037
11
-------
ABSTRACT
This report is an evaluation of air pollution potential associated
with the proposed Hampton Roads Cnergy Company Petroleum Refinery
to be located at Portsmouth, Virginia, and was prepared for the
U.S. Environmental Protection Agency. Emission estimates in this
report reflect the use of best available control technology for
both fugitive, and point sources. The report analyzes particulate,
sulfur oxides, nitrogen oxides, and hydrocarbon emissions. The
estimate of. emissions from the refinery complex includes the un-
loading of crude tankers, and the loading of product barges at
the marine facilities. Calculations by the refinery contractor
(Foster Wheeler Energy Corporation) are reviewed.
iii
-------
TABLE OF CO ML NTS
J5e_c_ti_oji Page
I. INTRODUCTION 1
II. EMISSION SOURCES 7
A. REFINERY EQUIPMENT 7
1. Pumps 7
2. Compressors 11
3. Flanges 13
4. Blinds 15
5. Valves... 16
6. In-Line Spares 20
7. Maintenance Practices.. 21
8. Combustion Units 23
B. CRUDE UNIT 27
C. HYDROGEN PLANT 30
D. NAPHTHA AND DISTILLATE HYDRODESULFURISERS 33
E. RESIDUUM HYDKODESUlFURIZER 35
F. DEBUTANIZER (STABILIZER) AND NAPHTHA SPLITTER... 38
G. CATALYTIC REFORMING I. .. 41
H. DEPENTANIZER 42
I. ISOMERIZER UNIT 45
J. LPG PLANT 48
K. AMINE TREATING 51
L. SULFUR PLANT 53
M, STEAM GENERATOR 61
N. HREC MISCELLANEOUS COMPRESSORS AND PUMPS 61
C. EMERGENCY FLARE SYSTEM, SLOWDOWN, STARTUPS,
SHUTDOWNS 62
P. COOLING TOVERS 64
Q. OIL-WATER SEPARATION 66
R. SOUR WATER STRIPPER 72
S. SLUDGE INCINERATOR 72
1. Fluidized Bed Incinerator;; , 73
?. Multihearth Units 74
iv
-------
TABLE OF CONTENTS (continued)
Section Page
T. CRUDE OIL UNLOADING 77
U. PRODUCT LOADING , 82
1. Liquid Petroleum Gases (Propane and
Butane) 84
2. Tank Truck Loading Rack 84
3. Pipeline 84
4. Marine Barge Loading 85
V. STORAGE TANKS 90
III. EMISSION ESTIMATE COMPARISONS .' 97
A. PARTICULATES 97
B. SULFUR OXIDES 100
C. NITROGEN OXIDES (NO ) 102
D. HYDROCARBONS 105
IV. SUMMARY OF BACT 109
A. COMBUSTION UNITS 109
1. Particulates 109
2, Sulfur Dioxide 110
3. Nitrogen Oxides 110
it. Hydrocarbons 110
B. SULFUR RECOVERY TAIL-GAS CLEANUP Ill
C. FLARING Ill
D. SLUDGE INCINERATION 112
E. MARINE TERMINAL OPERATION 112
1. Crude Unloading 112
2. Barge Loading 112
F. STOOGE TANKS 113
G. OIL-WATER SEPARATION 113
H. REFINERY EQUIPMENT 113
1. Relief Valves 113
2. Pipeline Valves, Control Valves and
Flanges 114
3. Pump and Compressor Seals 114
a. Rotating Shafts 114
b. Reciprocating Shafts 114
4. Blind Changing 114
5. Sampling 115
-------
TABLE OF CONTENTS' ( conuinued)
Section Page
SUMMARY 116
A. BEAVON-STRETFORD TAIL--:AS CLLANUP SYSTEM 116
B. MARINE TERMINAL 116
C. NITPOGEN OXIDES FROM PROCESS HEATERS 118
BIBLIOGRAPHY , 119
ACKNOWLEDGEMENT 121
APPENDICES
1. CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE CRUDE DISTILLATION UNIT A-l.l
2. CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE HYDROGEN PLANT A-2.1
3. CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM TKE NAPHTHA HYDRODESUL-
FURIZER A-3.1
4. CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE DISTILLATE HYDRODE-
SULFURIZER A-4.1
5. CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE RESIDUUM HYDRODES'JL-
TURIZER A-5.1
6. CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE DEBUTANIZER (STABILIZER) A-6.1
7. CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE NAPHTHA SPLITTER A-7.1
8. CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE CATALYTIC SPLITTER A-8.1
9. CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE DZPENTANIZER A-9. i
10. CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THS EMISSIONS FROM THE ISOMERIZER UNIT A-10.1
11. CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE LPQ PLANT A-ll.l
12. CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE OPTIONAL MEROX UNIT A-12.1
vi
-------
TABLE OF CONTENTS ( c on ; i :iu t
-------
TABLE OF CONTEXTS (continued)
LIST OF TABLES
Table
1. EMISSIONS FROM THE CRUDE UNIT 25
2. EMISSIONS FROM THE HYDROGEN PLANT 32
3. EMISSIONS FROM THE NAPHTHA HYURODESULFURIZER 34
4. EMISSIONS FPOM THE DISTILLATE HYDRODESULFURIZE1' 36
5. EMISSIONS FROM THE RESIDUUM HYDRODESULFURIZER 39
6. EMISSIONS FROM THE DEBUTANIZER (STABILIZER) 40
7. EMISSIONS FROM THE NAPHTHA SPLITTER (OPTIONAL) 41
8. EMISSIONS FROM THE CATALYTIC REFORMING UNIT t>k
9. EMISSIONS FROM THE DEPEKTANIZER 45
10. EMISSIONS FROM THE ISOMERIZER UNIT 49
11. EMISSIONS FROM THE LPG PLANT 51
12, EMISSIONS FROM THE OPTIONAL MEROX UNIT 51
13. EMISSIONS FROM THE SULFUR PLANT 60
14. EMISSIONS FROM THE STEAM GENERATORS 61
15. MISCELLANEOUS HREC EQUIPMENT EMISSIONS 62
16. EMISSIONS FROM THE FLARING SYSTEM 65
17. EMISSION FROM OIL-WATER SEPARATION 71
18. EMISSIONS FROM SLUDGE INCINERATION 78
15. EMISSIONS DUE TO CRUDE OIL UNLOADING 83
20. CASE I - MARKETING SCHEME. 91
21. CASE II - MARKETING SCHEME 92
22. EMISSIONS FROM HYDROCARBON STORAGE 93
23. PUMPS, COMPRESSOR AND IN-LINE SPARES 94
24. PARTICULATE SOURCES 93
25. SULFUR OXIDES SOURCES 101
26. NITROGEN OXIDE (AS N02) SOURCES 103
27. HYDROCARBON SOURCES , 106
viii
-------
TABLE OF CONTENTS (continued)
LIST OF FIGURES
Figure
1. PROPOSED HREC REFINERY 2
2. DIAGRAM OF SIMPLE UNCOLLED PACK£D SEAL 9
3. DIAGRAM OF SIMPLE MECHANICAL SEAL 9
4. LABYRINTH SEAL 11
5. OIL FILM SEAL 12
6. FLANGES 14
7. RELIEF VALVE 17
8. TYPICAL RUPTURE DISC INSTALLATION' 19
9. LIGHT COMPONENTS SEPARATION 69
10. HEAVY COMPONENTS SEPARATION 70
11, MULTIPLE-U2ARTH SLUDGE INCINERATOR 75
12. REFRIGERATION VAPOR RECOVERY UNIT BY EDWARDS 88
ix
-------
SECTION I. INTRODUCTION
The purpose of this report is to evaluate the air pollution
potential of the proposed Hampton Roads Energy Company (HREC) Pe-
troleum Refinery to be constructed in Portsmouth, Virginia. This
study is based on the best information available on the plans and
intentions of the developers, and is therefore necessarily limited
to discussions of likely process units and equipment and to formu-
lations of emission estimates based upon units "typical" of the
type anticipated. HREC has not completed detailed engineering de-
sign specifications, and appears uncertain as to how crude oil will
be received and marketed at this time. For this reason it is not
possible to describe and analyze in detail the specific units and
operations to be ultimately constructed at the refinery.
A representative of HREC has reported to PES that it is re-
luctant to invest the three to four million dollars required for
the engineering design units uatil they are confident that the pro-
ject will be approved by the Environmental Protection Agency (EPA).
HREC is also reluctant to sign a long-term contract committing
themselves to receipt or delivery by a specified date while the
construction is still in doubt. The shipping and receiving of
materials into the refinery, therefore, are limited to general as-
sumptions to allow calculations of emissions at various points.
All such assumptions necessitated by the absence of information are
carefully detailed in the text.
The refinery site is located along the west bank of the Eliza-
beth River due west of the City of Norfolk. The HREC facility is
whit is termed a "grassroots" refinery. The t<_rra grassroots per-
tains to the fact *.hat ti.is refinery is being built from the ground
up. There are currently no buildings or structures on the proposed
site. The refinery site consists of 623 acres of undeveloped land.
The planned refinery will immediately occupy 470 of the 673 i"-res>
-1-
-------
with processing units, tank farms, and storage, administration and
other buildings. A marine terminal is to be operated in conjunction
with the refinery. The main unloading platform will be offshore
approximately 2,700 feet. The proposed product barge terminal, con-
sisting of two piers, will be located about 600 feet offshore. A
vapor recovery system is planned lor the barge terminal. The re-
finery- has been designed to operate at a capacity of 175,000 bar-
rels of crude oil per day. The refinery site has additional space
to allow evj?ntual expansion of the refinery capacity to 250,000 bar-
rels per day.
The complex will be managed under two legal names. HREC will
own and operate the refinery processing equipnent and the inter-
mediate storage of products such as atmospheric gas oil and residuum.
The unloading and loading terminals and all crude oil and product
storage will be owned and operated by a company called Security
Marine Terminal (SMT).
Crude oil will enter the refinery at the SMT unloading facili-
ties by ocean-going tanker. The crude oil wi].l be unloaded into 11
crude oil storage tanks located on SOT property, and will be trans-
ferred by pipe tne to the HREC processing units.
A flow diagram of the proposed HR£C facilities is shown in
Figure 1. The crude oil is charged to the atmospheric distillation
column at the rate of 175,000 barrels per day. Aftei being passed
through a two-stage- electrical desalter, the crude oil is fraction-
ated in the distillation column into six main cuts. The overhead
cut consists of naphtha and all lighter fractions. The non-conden-
sibles from this stream is routed to the refinery fuel gas system
via the LPG plant,, and condensiblt: portions of this cut is sent tc
a naphtha hydrodesulxurizer^ In this desulfurizer the naphtha feed-
stocks are treated with hydrogen for removal of sulfur compounds.
The gases generated in this process are nmine rtripped for removal
of H2S, and this hydrogen-rich gas is routed to the sulfur plant.
-2-
-------
i-=3!BfJKr^ii1^3tss:-iSi=
0-XJ
ftf O
iitJ
2
^
83
T3
il
_j*-.fKsJ
"spray**?
1
L
T«f» *^C
I'X^/
fils^'1"
iPr^-
?₯'^
*,fr;-Ji
[^Z
*°° ~
i
& ffff|~ I atvsfi,
5?
V
!! ' " t
-((.
UCitCfi «H.
«»«*'«'«/*«.
'";/.'; »; rr'
Figure 1. PROPOSED HREC REFINERY
-------
The treated liquid pro :uct is sent to the debutanizer and the ex-
tracted butane is routed to a liquid petroleum gas (LPG) plant.
The deNutanized naphtna is sent to a naphtha splitter for furthor
fractionation, while the overheads are routed to the isomerization
unit for preparation of gasoline. Isor.erization upgrades the oc-
tane numbers of pentane. and hexane by rearranging the molecular
structure of the hydrocarbons from straight-ch.iin compounds to
branched-chain compounds.
The bottoms from the uaphtha splitter are routed to a cata-
lytic reformer vmich produces aro
-------
after being anine stripped or H^S, i., sem Lo the LPG plant. Vile1
naphtha is routed to the distillate hydrodesulfurizer and finally
enters the gasoline processing lines. Tii" dct>ul f uri::ej residuum
contains approximate} v 0.? percent sulfur by weif.hu and is marketed
as No. 5 fuel oil. \ portion of this fuel oil is sent to sr.oragt:
to be used as refinery fuel oil.
The refinery has many potential sources of emissions. Ail
of the riajor process units are operated under extremes of tempera-
ture and pressure. Leaks from such a system nust be kept to a
minimum. In addition to resulting in losses of naterial, a leak
can cause a deterioration of the process's desired operating en-
viror-aent which results in higher operating costs. Emissions also
result froro the transfer of niaterials about the refinery. Pumps,
seals, flanges and pressure relief valves are all hydrocarbon emis-
sion sources which are difficult to quantify. Other major sources
of hydrocarbon emissions are storage tanks, the marine terminals
(loading and unloading), truck or tank car loading racks (primarily
LPG racks), and cil-vater separation and treating facilities.
Combustion contaminants (particulates, sulfur oxides, ni-
trogen oxides and hydrocarbons) are produced in the various pro-
cess heaters and steam generating facilities and the emergency
flares.
The remainder of this report presents a detailed description
of the sources of emissions at the refinery and a discussion of
the work completed as well as the work prepared for this report.
Section II provides a discussion of each unit, Identifying sources
of emissions. A comparison of the emissions calculated by Foster
Wheeler Lnergy Corporation (FWEC) and Pacific Environmental Ser-
vices, Inc. (PES) appear as Section III. Section III also discusses
discrepancies which appear between calculations performed by FWEC
and PES. Section IV presents a qualitative view of the proposed
-5-
-------
refinery, stressing the types of equipment proposed and the PES
opinion of best available control technology. Section V compares
and quantifies the emission potentials of various control strate-
gies which nay result.
-6-
-------
SECTION II. LM lab ION SOURCES
There are many sources of emission at .a refinery. Tlic first
section discusses die various typ ?s of. refinery equLpnenl w'licli vi ] 1
be comnon. to the proposed Droce~!i jnit:-. The following sections
discuss each unit in detail, explaining the process, and identify-
ing and quantifying emissions from that unit.
A. REFT.N^PY EQUIPMEOT
3. Pumps
A refinery uses many different types of pumps to raove fluids.
Pumps vary in capacity up to 100,000 gallons per minute and in
pressure differential up to 30,000 pounds per square inch. A
pump is designed to perform a specific function, and is limited to
a rather narrow range of operation above, and below the design con-
dition. For this reason, hundreds of vumps of different styles and
modes of operation are used. The types most often used in refiner-
ies fall into two broad categories: centrifugal and positive dis-
placement devices. Centrifugal devices can be classified as volute,
axial, and turbine pumps. In general, centrifugal fluid-transport
devices have the following characteristics: (1) discharge is rel-
atively free of pulsation; (2) the mechanical design lends itself
to high through-puts, which means that capacity limitations are
rarely a problem; (3) they are capable of efficient performance
over a wide range of pressures i.nd capacities even at a constant-
speed operation; and (4) discharge pressure is a function of fluid
density.
Positive displacement devices are characterized as recipro-
cating piston, plunger, diaphragm, rotary vane, and gear pumps.
The large variety of displacement-type fluid-transport devices
makes it difficult to list characteristics common to each. However,
for most types it is correct to state that: (1) they are adaptable
-7-
-------
to high-pressure operation; (2) discharge of riany will pulsate un-
less an auxiliary damping system is employed; (3) mechanical con-
siderations limit maximum through-puts; and (-4) thsy are capable
of efficient performance at extremely low-volume through-put rates.
Asphalts and oth^r semi-solids, such as distillation column
residuum, are handled with reciprocating st
-------
"1VWi
vvv /
~^Z7
Figure 2. DIAGRAM OF SIMPLE UXCOOLED PACKED SEAL
SET StetIS
Jftliit -Ml (1C
u-f:» feus
Fitt uu UUSE uati
Figure .1. DIAGRAM OF SIMPLE MECHANICAL SEAL
One ring Is stationary while the other is attached to the shaft
and rotates with it. A spring and the action of fluid pressure.
keep the two faces in contact. Lubrication of the wearing faces
is effected by a thin film of the raterial being pumped. The
wearing faces are precisely finished to ensure perfectly flat
surfaces. Materials used in the manufacture of the sealing rings
are many and varied. The basic design of the mechanical seal has
not been altered significantly over the years, however, recent
-------
technological advancer, have been made in tho matt-rials used in
seals. Today, single mechanical seals are capable of replacing
double mechanical seals without a reduction in control efficiency
due to new plastic and synthetic rubber materials which have
solved many of the static staling problems encountered with older
materials.
With modifications, the mechanical seal can be considered to
be the best available control technology (BACT) for a centrifugal
pump. All centrifugal pumps in hydrocarbon service at the HREC
refinery vill be equipped with mechanical seals. Centrifugal
pumps handling water will be equipped with packed seals. A reci-
procating, punp is planned for the emergency pump from the flare
knockout drum and will be equipped with double outboard seals. An
outboard seal utilises packing, and is situated on the exterior
of the pumping mechanists. Since the packing is soft and flexible,
its operating lifetime is shorter thar. a mechanical seal. Al-
though there are different designs of packed seals, all have in
common the concept of a piston and cylinder, or a ring and gland
holding the packing material. As pressure is applied to the piston
or packing ring, the packing material is compressed along the pis-
ton shaft longitudinal axis. This applied pressure forces the
packing material against the seal housing and the moving shaft.
As more pressure is applied to the packing material, it develops
a pressure greater than that of the pumping fluid resulting in no
leakage. Using this principle, a double outboard seal is two
sets of seals combined in one housing. The existence of a barrier
fluid for improvement of the sealing process would be unlikely.
Due to the softness and flexibility of the sealing mechanism any
fluid in the zon^ between the two outboard ssals would have a ten-
dency to escape.
-10-
-------
2. Compressors
The only type of compressors planned for the HREC refinery
are centrifugal units, with the possible exceptions of a recipro-
cating unit compressing air and a reciprocating compressor on the
hydrodesulfurization units handling hydrogen-rich streams. These
reciprocating compressors would be equipped with ring packing and
any escaping gases would bleed off to the flare.
The low pressure centrifugal compressors will be equipped
with a labyrinth seal used to prevent loss of gas (see Figure 4).
The seal consists of a number of restrictions and openings through
which the escaping gas must flow. The labyrinth seal is normally
vented at soTie midpoint and bled back to a lower pressure stage or
to the compressor suction.
Figure 4. LABYRINTH SEAL
-11-
-------
The majority of the centrifugal compressors will hove oil
film seals. An oil film seal is a modification of the mechanical
seal which is so successful that it needs very little attention.
Ir is constructed like a mechanical seal but the wearing faces are
held apart while th? machine is running (Figure 5). The reason
there is no wear is that the oil pumped between sealing faces does
the actual sealing. The oil pump is driven by the compressor
shaft at the other end of the machine. Faces come together and
hold almost a perfect vacuum when the machine is shut down.
Oil from pump
Sffl if.y
iKfite floating
'^fl rinf
ring
feet \ 5
^-. OuHr tlooting ftal rinf. J
Figure 5. OIL FILM SEAL
The operation of an oil film seal .is as follows: when a
compressor starts, oil is pumped through internal passages in the
compressor. Oil flows through a filter and to the inside of the
doughnut-shaped seal bellows. Oil pressure expands the bellows,
moving the stationary sealing seat against spring pressure to the
stop position on the seal housing. The oil in the bellows then
passes through drilled passages in a stationary seal to an irregu-
-A
'.j
-12-
Uw^»^sa^^,,s,,^^-.^^rti»*yifcvi.jJJ-.i.'.J-^;1:
-------
!ar oil groove in the face of a stationary sealing sert. The
oil llous across an outer section 01 the stationary scaling lace and
into the space between th.2 shaft and the bellows assembly. Flow
from this space is restricted by an inner floating rin^. The flow
continues across the inner section of the sealing face and through
clearances betv/een the rotating sealing-seat-hub and the stationary
sealing seat. This flow is restricted by the close clearance of
the floating ring between these two parts. It is also restricted
by atmospheric or outer floating rings between the peal housing
cover and the rotating seal hub.
The pressure-regulating valve in the oil system remains closed
on starting. Thus, full capacity of the oil pump is assured for
initial oil supply. During operation, the oil passing between ti.e
seal faces seals the shaft against invard leakage of air so metal-
to-metal contact is not needed to do the sealing.
In spite of the use of oil seals, Foster Wheeler anticipates a
gas leakage rate of approximately 50 SCFM/compressor from the drain
trap. This gas is expected to be composed of 30 to 30 percent hy-
drogen. If the compressor is involved with non-contaminated ser-
vice (without sulfurous compounds) it will bleed directly to the
flare system. Contaminated units will be amina stripped prior to
venting to the flare system.
Some of the compressors will handle material which must remain
oil free (such as makeup unite handling hydrogen destined for the hy-
drodesulfurization units). Such a compressor vould be equipped with
a simple mechanical seal.
3. Flanges
Flanges provide a removable connection between pipe and
vessels or other items of equipment. Flanges are specified by
pressure rating and by facing. Pressure ratings used are: 150,
300, 400, 600, 900, 1500 and 2500 psi. The most common flange
facings are flat face, raised face, tongue and groove, and ring
-13-
-------
joint (see Figure 6), It is important that two opposing flanges
have the Fame rating and facing. If mismatched flanges are con-
nected together, there is a good possibility that the jo.int will
uarp and leak.
61V1ALL MALE.
4- pe^
to
! 5
RJ
JO/KJT
Figure C. FLANGES
-14-
-------
Foster Wheeler (TN'KC) has i nd u-.'tr <.<! tliat they will ex-rcije
excellent engineering practices when designing for the use of flanges
in a refinery. roster Ul.eoler1s basic policy is to minimize the use
of flanges. Each flange is placed where it can be it-ached by a work-
man for easy repl ice-meat of gasket!ng. FUT.C has installation and
operating procedures which minimize the scratching on the flange sur-
face. Each bolt is torqued to a designated specification.
Of the types of flanges commonly used, FWEC prefers the
ring joint because; it causes less problems. Many types of
Basketing are used, but in high pressure systems (with the
greatest emission potential), where copi,--g with temperature
and pressure fluctuations can be a problem, FWT.C uses what they
call "flexitalic" joints and gaskets. Flexitalic units combine
metal with asbestos to provide a flexible unit which is resistant
to temperature and pressure extremes. When a section is flanged,
the ring joint- consisting of an asbestos center surrounded by a
metal casing, is placed between the flanges DO that when bolting
occurs, the ring will expand to fill the space. These units are
Subject to deterioration. The asbestos wears out after awhile
and the temperature differentials cause the metal to flex. Proper
maintenance _nd inspection to ensure all units are tightened
can significantly reduce the emissions from .langes.
4. Blinds
A refinery will, on occasion, find it necessary to utilize
blinds. A blind is a flat soj.id piece of steel which can be inserted
in a flange r.o foim £ solid seal in a line- The insprtion of a blind
instead of the turning of a valve is necessitated ai times of repair
shutdown for reasons of safety and insurance against contamination
The presence cf a blind in a line eliminates the danger of an inad-
vertent cpenin;; of a valve which could result in ii.jury or contamina-
tion.
-13-
-------
Blind ch^rv'.lnr, is usually performed by n mechanic and requires
between 15 and 2O minutes for insertion. The recommended procedure
ii; as follows: tha bolts connect :ng tlie flanges are locisened and a
flange chisel nade out of brass or othei non-sparking material (to
reduce the danger of explosion) is driven into the connection, wedg-
i;ig a gap. When the blind is inserted and the wedge removed the
bolts are tightened, locking the blind firmly in place.
5. Valves
Valves are employed in every phase of the petroleum industry
where petroleum or petroleum product is transferred by piping from
one point to another. There is a great variety of valve designs
but, generally, valves may be classified by their application as
flow control or pressure relief.
Manual and automatic flow control valves are used to regulate
the flow of fluids through a system. Included under this classifica-
tion are the gate, globe, angle, plug and other common types of
valves. All of these valves tend to leak through the valve seat when
handling low viscosity fluids and/or when the pressure differential
across the valve is large. '-There a single valve is used to separate
a material that is considered a pollutant from the environment, the
outlet cf the valve is usually closed with a blind flange, a blind,
or a plug, to ensure that the material does not escape.
The second major source of valve leakage is the packing Chat
is used to seal the valve stem. This packed joint must slide along
the valve stem as the valve is opened and closed. A regularly
scheduled program of inspection to ensure that valve stem packings
are properly maintained will aid in preventing leakage to the at-
mosphere.
All control valves and connections will be held in place by
flanges. The stems of these valves will be-, pocked to reduce hydro-
carbon emissions. The type of packing used will depend upon the
operating temperature of the process. Up to 450°F, the valve stems
will use a self-lubricating packing. For operating temperatures
-16-
-------
above 450°F, the valve sterna will usr teflon asbestos or asbefl-js
with a grease seal.
Certain (non-specil'led) block valves in light ends service
will be equipped with a new vented packing which will vent to the
flare system.
Pressure relief valves (Figure 7) are used to prevent exces-
sive pressures from developing in process vessels and lines. These
valves may develop leaks because of the corrosive action of the
product or because of failure of the valve to reseat properly after
blowoff. The maintenance and operational difficulties caused by
the inaccessibility of many pressure relief valves may allow leak-
age to become substantial.
Figure 7. RELIEF VALVE
-17-
-------
Relief valves inay be insf.illi.-d for relief of gaseous or
liquid pressure. The liquiJ relief devices are normally installed
to relieve thermal expansion and are less inclined to leak than
vapor relief devices.
The most common type of relief valve is spring- loaded . In
a typical spring-loaded relief valve, fluid pressure is maintained
in the throat of the valve. If the operating pressure exceeds the
valve set oressure, the valve will open and relieve the system
pressure.
Most relief valves achieve a seal by forcing a inotsi plate
against a metal seat. These seats are durable, but tend to leak.
A more effective seal can be achieved by use of an "0" ring or
plastic seat where process conditions permit their use.
Relief valves are normally set to activate at a pressure that
is a maximum of 10 percent or 25 psi above the normal operating pres-
sure of a vessel. If the operating pressure is too close to the set
pressure, the valve will tend to lift off the seat during normal
operating cycles with resultant discharge of material through the
valve and possible damage to the valve seat. These fluctuations
in operation which cause this unseating can occur up to 50 or 60
times a minute. This occurrence is known as "chattering."
Most of the relief valves will vent to the flare system.
Due to the location and type of material handled (not specified),
it will not be practical to run flare system ducting to all re-
lief valves. These valves will be equipped with rupture discs.
The rupture disc or burst diaphragm (Figure 8) is a metal
diaphragm which is designed to rupture at a predetermined pressure.
By placing such a disc ahead of a relief valve, chattering and
its resultant emissions can be eliminated. This vapor- tight seal
may not remain in perfect condition. It is possible for the disc
-18-
-------
rz u ,=» r u
P I 50
Figure 8. TYPICAL RUPTURE DISC INSTALU.TION
-19-
-------
to develop a pinhole leak due to acmosphcric or internal corrosion.
If such a leak should develop, the product material would be con-
trolled only cy the relief valve. Pressure gauges are generally
used to indicate such a condition. Frequent inspections are
made by the operators and corrective maintenance performed as
soon as possible. Chattering valve;; must be checked for signs
of valve seat damage and rupture disc valves cust be viewed for
leaks.
FWEC states that a non-chattering valve is now available
fron manufacturers. In principle, such a valve is equipped with
a bypass system. After the valve relieves once, all other surges
vent via the bypass system. It will then become necessary to find
somewhere to vent this bypass system. This system would not be
of any additional benefit for the valves situated away from the
flare system ducting. Such a valve would have to have a rupture
disc in any event. Therefore, it is felt that installation of
these new valves will not cause an improvement in emisr on con-
trol over the planned system. Inadequate information is available
to compare the increased initial cost of non-chattering valves with
i
the higher maintenance costs associated with chattering valves.
6. In- Line Spares
In the operation of a refinery, it is extremely important
that processes operate as much of the time as is possible. In
order to minimize down-time caused by equipment malfunctions, the
refinery is usually designed with in-line spares. An evaluation
of the number of in-line spares can provide a qualitative assess-
ment of the hydrocarbon potential from malfunction.
The lack of a spare on an important unit, such as the crude
oil pump, ^ould cause operation to continue during a leakage con-
dition resulting in more hydrocarbon emissions than would tesult
if this pump had an in-line spare. Existence of this spare would
allow switching of the product line with minimal disturbance to
the procpos operations while the normal pump is repaired. In de-
-20-
-------
t._rmining which refinery equipment requires in-line spares, the
important consider.ition to be made id whether the loss ot that
particular pump or compressor would cause an immediate shut-down
of the associated operating unit. In the case of the affirmative,
the equipment would require a spare. A prime example of equipment
needing an in-line spare is a unit feed pump. In-line spare.9 for
pumps handling product and side-streams are not as essential since
a refinery has more flexibility in the handling of thr; products
than in the feed.
There are several practices which can be utilized to minimize
emissions resulting from switching to an in-line spare. A recom-
mended procedure is to switch operation from the norr..al purcp to the
spare once a "nek and operate the spare for approximately 24 hours.
This ensures refinery personnel tht the spare will be operable in
the case of emergency switchover. This procedure also allows raaizi-
tenance crews to perform any work necessary on the normal pump.
It is a recommended practice that if the normal operating pump is
electric that the rpare be a steam turbine operated unit with a
small amount of steam constantly circulating lo maintain the pump
at a heated temperature. The use of a steam turbine spcre ensures
continuous operation in the case of a powrj.r outage.
Table 23 on page 94 at the end of this section itemizes each
pump and compressor and whether or not £. spare for this unit in
anticipated.
7. Maintenance Practices
In a modern day refinery, management makes ever/ effort to
keep losses from faulty equipment at an absolute minimum. The
two primary reasons are economics and ecology. Crude prices
three to six times the previous levels have forced all enployees
to concentrate on keeping these raw materials or refined products
from escaping.
-21-
-------
The best approach to detection of operational losses is
frequent inspection by qualified personnel. Supervisory personnel,
such as operating day foremen and shift foremen are constantly
in touch with the crews that operate the plants. They establish
the rules and schedules by which the crews make routine inspec-
tions. The crews are constantly in circulation over the entire
plant each 24 hours. As an example, the day foreman will leave
an order that each ttll-tale sampler on each product cooler (oil-
water heat exVxangers) must be checK»d twice each shift. This is
c?one by tapping a bleeder line from the water outlet and drawing
a sample into a container which is set aside. In the event that
there is an oil leak through the tubes into the water, oil will
appear on the surface after a few moments. Ac ion will then fol-
low for removal of the unit from service and making the necessary
tests and repairs. Other in-plant lines, valves, pumps and equip-
ment not functioning properly or leaking are corrected immediately,
if possible. If the unit operator is unable to correct the problem,
it is reported to the head operator who informs the day foreman for
inclusion of the malfunction on the repair list for the next shut-
down.
For example, if a packing gland on a puirp is leaking badly,.
the unit operator would report this to the head operator as quickly
as possible. He obtains a mechanic who reports immediately. The
pump is switched to the spare and the leaking gland fixed. If
the pump cannot be removed from service, the day foreman will
evaluate whether the plant has to be shut down immediately or
is able to operate until a scheduled shutdown occurs.
The number of leaks that occur which cannot be repaired
immediately or within a few hours are relatively few in number
and small in magnitude.
-22-
-------
Standard procedures in almost every refinery these days is
to maintain a "leak detective" and d team of water and air pol-
lution specialists. These people work full time tracing sources
of odors, excessive oil in the separators, and othf=r indications
that there is a leak or leaks which nay not be immediately ob-
vious.
For pipelines off the premises, a line-walker is sometimes
utilized. He physically walks the ^ull length cf the line at
least once each day. Lines are not buried to any great extent in
modern refineries, if it is at all practical to avoid it.
In addition to the surveillance and control of hydrocarbons,
a refinery must be careful to control the emission of a very dan-
gerous and poisonous g.is produced during refinery processeshy-
drogen sulfide (H S). The HREC X'efinery will have area monitors
stationed about the premises with FUS analyzers (designed by
API) tied into an alarm system to aid in early detection of this
toxic gas.
8. Combustion Units i
\
The combustion units will He equipped with burner controls
in the form of excess oxygen contirol. Continuous opacity moni-
tors will be installed on every si:ack. The type of monitors
y planned will be laser sensing devices.
V i
Combustion units will consist of steam generating units
and process1 heaters, each capable of firing a combination of
oil and gas. Steam requirements for the refinery are not precisely
known but are expected to be approximately 500,000 pounds per
hour. Steam will be generated in tnrce steam generators, each
capable of supplying one-half of thu total demand. The existence
of 50 peuent standby boiler capaciry is a recommended refinery
practice to ensure a more continuous operation. Two levels of
-------
steam have been selected for the refinery. Medium pressure super-
heated steam will be generated at 250 psig and 500°F. Low pres-
sure steam from turbine exhaust will be at 50 psig. Low pressure
steam requirements in excess of the turbine will be. supplied by
letting down 250 psig steam to 50 F3ig-
The formation of oxides of nitrogen is a by-product of the
normal combustion processes than occur in process heaters and
hoilers in a refinery. There are seven known oxides of nitrogen:
N00, NO, N-0,, N00, N'O. , N00C, and NO,. Some of these are very
*. 2 J Z 2 4 z J J
unstable at room temperature or below. There are only two of the
above which are important from an air pollution standpoint: NO
and NO^. Any mixture of nitrogen and oxygen subjected to tempera-
£,
tures over 2800°F up to about 4000°F will produce NO according to
the reaction:
N2 + 0, + 43,000 calories -^-2 NO
NO will decompose back to N- and 0_ if cooling back to below
2800°F is done at a rate slower than 30,000°F per second. However,
in most combustion situations, the cooling rate will be much faster
than this and the NO will remain fixed.
The stean generating units at HREC will make use of a tech-
nique called two-stage combustion to reduce nitrogen oxides emis-
sions. Two-stage combustion consists of bypassing a percentage of
the combustion air from the burner to injection ports in the bulk
gas region downstream of the primary chamber. The ports are pop-
ularly called "NO ports." A portion of combustion air is ad-
mitted at the burner at; usual, but the remainder is added downstream.
Usually 90 percent of the necessary theoretical air is admitted
with the fuel. This results in a reduction in the amount of oxy-
gen available near the burner for fuel air mixing, hence the amount of
oxygen available fur fuel bound nitrogen is also reduced. The
-24-
-------
fin.'l not oxidized ir the prinary r.oiii- is burned in the bulk g.is
zo'.ie where temperatures are lower due to heat loss to the oon-
bustion wall and macrosc.-.l.? mixing with the cooler gases in
the fire box. The net '-05. >!r is a slower rate of cooling which
helps inhibit the f:.xing of ::he XO.
Fostev Wheeler rppresentatives stated that while two-stage
combustion is available for a boiler, the technology and engineer-
ing is not available for process heaters. The examiners were
curious as to why a two-stage combustion system had not been de-
veloped for a process heater since the sarae concepts of combus-
tion, consisting of a fire box and tubes, are much the same as in
a boiler.
Several industrial process heater manufacturing sources were
contacted to find out if there are any limiting factors preventing
the two-stage combustion concept being used in process heaters.
The sources (who have requested that they remain anonymous) stated
that in the past, most process heaters were designed primarily to
heat the process material to within desired specifications, and
that fuel economy and pollution control were often lower priority
considerations. Until recently, there was insufficient incen-
tive produced by environmental agencies to warrant NO reduction
specifications in process heater design fro^ industrial customers.
With the advent of stricter emission standards, however, indus-
trial customers have now begun to request lower NO emission level
combustion devices. Burner and process heater manufacturers are
responding to these customer requests and are in the preliminary
design stages of systems which will accomplish this goal.
Most sources of information agreed that two-stage combustion
could be applied to a process heater although problems may be
cavoed by the fact that the physical size and location of the unit
and the extra ducting needed to support the combustion process
could not always be accommodated. No one had knowledge of any
unit currently on the market.
-------
One source contacted mentioned thai John Zink was current-
ly experimenting with a NO reduction program for process heaters.
Although specific details were not available, the approach uses
a fuel-rich primary combustion zone and ducting which recirculat.es
waste gas from the bulk gas region of the fire box. The recir-
culated gas is vented through a specially designed burner.
Preliminary results of the experiments show a significai t
decrease in NO concentrations in the flue gas. Although Johr
Zink has been hesitant to explain further at r.his time, unoffi-
cial figures released have indicated that NO emissions of 0.2
6 x 6
pounds,'10 BTU for liquid fuel and 0.1 pound/10 BTU for gaseous
fuels could be expected.
A problem which has been encountered with the two-stage com-
bustion approach stems from the fuel-rich primary conbustion zone.
When this mixture is used in a fire box, there is an excess of
carbon present as soot. Since these heaters do not have soot blow-
ing capabilities, chere is a tendency for the carbon to accumulate
on the. tubes, changing the designed heat transfer coefficient. As
the carbon collects, hot-spots are formed on the outside of the
tube walls. These hot-spots eventually burn through the tube
necessitating a process heater overhaul.
In addition to two-stage combustion there are other means of
reducing NO_ emissions. (1) Reduction of excess air will inhibit
the production of NO . The less oxygen available near the burner,
the lower the possibility of conversion of fuel bound nitrogen.
(2) Flue gas recirculation and reduced air preneat help reduce
the peak temperature of the primary zone. Flue gas recirculation
also aids by reclrculating gases low in unformed nitrogen and
oxygen. Reduced air preheat is not popular because it causes a
reduction of the combustion efficiency and does not limit the
available oxygen.
-26-
-------
The nost popular alternate form of nitrogen oxide control
is of f-stoichiomet ric co:;ibust ion. This method of combustion uses
to advantage multiburner configurations in the fire box. No ex-
traneous ducting or modification is ro'juired since the procedure
involves running various burners on air only, and the remaining
burners fuel-rich. Sor.e fuel escapes contact (hence reaction/1
with the air in the primary zone, and is combusted in the cooler
bulk gas recirculation zone. Although the overall stoichiometry
(chemical coorectness) is maintained, the procedure effectively re-
duces the amount of fuel and air interacted close to the bu-ner face,
extending ins Lead a substantial portion of the mixing and burning
into the secondary zone. A problem which requires special con-
sideration is the fact that carbon rconcxicle emissions can increase
for some combinations of fuel-rich burners. The most effective
combination of air-only burners and fuel-rich burners can only be
determined by trial and error.
Each of the following refinery process units utilizes some
if not all of the refinery equipment or techniques already men-
tioned. Each subsection consists of a brief description of the
type of process involved followed by an inventory of each emission
point in that prccess. Etch subsection concludes with a table
summarizing the estimated emissions iron that process. All cal-
culations and assumptions used to develop the tables are discussed
in noted appendices at the rear of the report.
B. CRUDE UNIT
Crude oil enters the HREC refinery by way of the Security
Marine Terminal unloading facilities and storage tanks. HREC was
unsure of the exact crude oil characteristics at the time of the in-
formation gathering trip associated with this project. When FWEC
made their study, they worked under the assumption that the crude
oil to be handled would be a light Arabian weathered type with a
Reid vapor pressure of less than two, a sulfur content of 1.7 per-
-27-
-------
cent by weight, and a 3A.2° API gravity. A weathered crude has
been exposed to th~ amtsophere or low pressure systems for a suf-
ficient time to allow a portion of the g?s-;ou.=; hydrocarbon? dis-
solved in the crude oil to have been lost.
From tankage, the oil with a maximum arrival .temperature of
105°F, is preheated by exchange with the ho; crude unit products
to about 250-300°F, and then desalted. Desalting removes salt,
silt, sand, water, and other crude oil contaminants referred to as
B.S. and W (bottom sediments and water). Foster Wheeler plans on
installing a two-stage electrical desalter. In this type of unit
crude oil is mixed with water and chemicals and is passed through
electrical fields which cause, coalescence of the contaminant drop-
lets. The scur watur collected by the desalter is drained to the
sour water stripper. After treatment, the water Is returned as
desalter makeup water.
The desalted crude oil is passed through booster pumps which
transport the crude oil through more heat exchangers and through
the crude furnace. The temperature of the oil coming out of the fur-
nace is t>pically SqO-"/00°F. From the furnace, the crude oil pacses
to the crude distillation column. Light naphtha and other products
are captured in the overhead condenser. The non-condsnsibles are
routed to the refinery fuel gas system by way of an amJne
stripper for use in the combustion units or as "~eed to the hydro-
gen plant. The condensible overheads are routed to the naphtha hy-
drodesulfurizer. The four side-streams are: heavy naphtha, kero--
sene, light fuel oil and heavy fuel oil. The top three streams
go tc the distillate hydrodesulfurizer. The heavy fuel oil is
blended with distillation residuum and routed to the residue hydro-
desulfurizer.
The points of potential e.,iissior.r, include the crude heater,
(which will exit through Stack #3), relief valves, pumps, compres-
sors, control valves and flanges. Emissions from various units of
the crude distillation unit are presented in Table 1.
-28-
-------
Table 1. EMISSIONS FROM THE CRUDE UNIT (TPY)*
i
ho
"~~"-~- ^CONTAMI NANT
SOURCE ~~~~^- __
Crude Heater
Relief Valves
Control Valves &
Flanges
Pumps & Compressor
Blind Changing
Sampling
TOTAL
PARTICIPATES
Uncontrolled
116.2
116.2
Control led
11-S.2
116.2
SULFUR DIOXIDES I
Uncontrolled (Controlled
547.2
547.2
347.2
547.2
NITROGEN OXIDES
Radian
896.3
896. 3
AP-40
941.1
1
941.1
1 HYDROCARBON'S
Uncontroll ed
50. 1
31.9
81.4
13.7
u.9
,,
I
184.7
1
Cont rol led
50. 1
-0
HI. 4
2.1
-0
'")
1 34 2
L
*~ -- '-
*For
ilat ions and assumptions used to determine tne values used for this t;ibli.', sec Appendix i.
-------
C. HVDRIVEN PIANT
Hydrogen is an intermediate material in the refining opera-
tion. The HRHC refinery will engage in a great deal of hydrodesul-
farizing of intermediate products. This will require much more
hvdrogen than can be generated by such units as the catalytic re-
former.
The principle method of producing hydrogen is steam reform-
ing of some available hydrocarbon such as natural gas, refinery
gas, propane, butane, or naphtha. HREC plans on using refinery
fuel gas or naphtha. Hie basic reaction is:
CH4 + 2H...O -*- AH2 + C02
The feed material, probably refinrry gas, is first treated
for sulfur reir.oval. The sulfur in the gas is absorbed into a
zinc oxide catalyst. The sulfur-free gas is mixed with high-
pressure steam and pre-heated in the convection section of the
reformer furnace. The nixed gas flows down through a nickel, zinc,
or copper catalysts-filled tube where the basic reaction takes
place producing carbon nonoxide in addition to hydrogen and carbon
dioxide. The high temperature effluent gas from the reformer fur-
nace flows through the tubeside of a steam generator producing
high-pressure steam. Additional high-pressure steam is generated
in the convection section of the furnace. This steam is consumed
in the reforming reaction. More than three-quarters of the total
hydrogen is produced in the reforming reaction. The remaining
hydrogen is produced by the shift conversion of carbon monoxide
to carbon dioxide. The catalytic conversion occurs in two
stages, one at high temperature and the second at low temperature.
The combination of shift converters will achieve 98 percent conver-
sion of the total carbon ^onoxides to carbon dioxide and hydrogen.
Reaction heat is removed between the two stages of shift conversion
-30-
-------
bv p,':nerating l°w pi'ossiirv. steam. The j;.as temperature beLwet.i iv;e
catalyst beds is also lovereil to provide more favorable conditions
for the shift reaction.
The raw hvdrogen jay from the shift converter is further
cooled before nixint; with a monoi-thano] ami ne (MEA) solution in an
exchanger. The heat nf reaction bet.veen carbon dioxide and MEA is
largely rcnoved in this precontact stage ahead of the amine ab-
sorbtr. The remaining carbon dioxide is removed frcn the gas
strewn by reaction in the absorber. The heat for MEA regeneration
is supplied partly from the process gas and partly fron low pres-
sure steam. An MEA reclaimer is provided to maintain a clean ac-
tive solution. Gt.her tneans of absorbing carbon dioxide niay be
used instead of MEA; these include Sulfinol and activated hot
carbonate.
The points of potential emissions as quantified in Table 2
include the hydrogen furnace which exits through Stack til and re-
lief valves which are vented to the flare gas system. There are no
pumps in hydrocarbon service other than methane. There is a pos-
sibility that a compressor will be involved to handle the incoming
refinery fuel gas and compress it from 50 to 400 psi. This compres-
sor would be equipped with oil film seals or flare vent seals de-
pending upon the type of compressor chosen. The oil film seal would
be desirable if the compressor is a centrifugal unit and flare vent
selas would be BACT for a reciprocating unit.
Base upon similar units in operation at other refineries, it
is anticipated that the catalyst beds will have to be replaced
every three to five years. During replacement, the spent catalyst
is oxidized to the metallic oxide state. In the oxide state, the
metal is insoluble in water and can be disposed of in a sanitary
land fill. In order to transfer this material, it must be extreme-
ly dry so that it will flow freely. This will constitute a fugitive
dust problem while transfer is made. Due to the infrequ?r:cy of the
-31-
-------
Table 2. EMISSIONS FROM THE HYDROGEN PLANT (TPY)*
i
u;
^~ -CONTAMINANT
SOURCE - ^_
Hydrogen Heater
Compressor
ReJitf Valves
Blind Changes
Sampling
Control Valves &
Flanges
TOTAL
1
PAKTICULATES
Uncontrolled
186.2
186.2
Control led
186.2
,
186.2
SULFUR DIOXIDES
Uncontrolled
827.4
827.4
Controlled
827. A
827.4
NITROGEN OXIDES H HYDROCARBONS
Padiai' ~j AP-40
]..3r5.4
1465.7
|.
1355.4 I 1465.7
~" - j
Uncontrol led ConLroJ 1 r
75.4
1.6
31.9
Q. 9
6.7
8!.'.
197.9
75.4
-0
-0
-0
. -
o
11
8 ! . 4
156.8
*For calculations and assumptions
-------
o;-or;iLion, these enis^ions are not considered to be significant.
D. NAPHTHA AND DISTIU-VfE HVDHODESL'LFL'RIZERS
Hydrodesulfurination (HDS, processes are used to remove sulfur
from liquid petroleun fractions. Sorae nitrogen removal and satura-
tion of olefin bonds may also occur. Sulfur removal is accom-
plished jy reacting the sulfur containing compounds with hydrogen
(created in the hydrogen plant) in the presence of a catalyst to
form hydrogen sulfide. The hydrogen sulfide is extracted in an
amine stripper.
The naphtha and distillate oil is heated in an exchanger
and tr.ixed with a hydrogen rich gas stream. The mixed feed is
heated in a fired heater and passed through a cobalt molybdenum
catalyst bed, where the hydrogen reacts with sulfur and nitrogen
compounds. reactor effluent is cooled and a small quantity of
wate1' is added to absorb ammonia compounds. The liquids are
separated from the vapors, and the liquid phase is withdrawn.
All of the condensible material from the top of the crude dis-
tillation column is treated in the naphtha hydrodesulfuri-
zer. The saturated H S gas is treated in a moncethanolamine (MEA)
scrubber before the gas is sent to the centralized refinery fuel gas
system. The H-S and MEA are separated and the H-S travals to the
sulf-ir plant. The gas; travels by convection and no pumping
should prove necessary. The created naphtha is routed to the de-
butanizer. Emissions from the proposed naphtha hydrodesulfurization
unit are tabulated in Table 3.
The distillate hydrodesulfurizer operates in much the sa^ne
tnanr.er. The three lightest side-cuts from the crude unit, heavy-
naphtha, kerosene, and light gas oil, are sent to intermediate
storage. Only one of the three products is processed through the
unit at any one time. One of the results of running naphtha through
-33-
-------
Table 3. EMISSIONS FROM THE NAPHTHA HYDRODESULFURIZER (TPY)*
-^CONTAMINANT
SOURCE ~~~ - ._
Naphtha HDS Heater
Relief Valves
Control Valves &
Flanges
Blind Changing
Sampling
Pumps & Compressors
TOTAL
PART I Clll ATKS
Uncontrolled
63.0
63. 0
Controlled
63.0
6J.U
SULFUR PTOXIDKS
Uncontrolled
280.2
- 280.2
!. -
Controlled
280.2
280.2
NITROGEN OXIDKS
R.ad ian
460.4
460.4,
AP-40
446.4
_ .
446.4
u HYDROCARBONS
lUnrotitrol led i Con I ml led
25.5
25. J
31.9 -0
t
v i - -
\
1
81.4
0.9
i">. 7
3.7
1
1 5(,. 1
.
81.4
. .
-0
( 1
2. >}
I"1 9. i
*For calculations and assumptions Ubuci to determine values used for this table, see Appendix 3.
-------
tni unit is the production of a lighter ''wild" naphtha which is col-
lected in a product stiipper and sent as feed ro the debur.mizer.
The heavier kerosenes ;;r.d gas oils Leave the unit to be marketed.
The kerosene has become Jet A--1 fuel and the gas oils are blended
to fern a
-------
Table 4. EMISSIONS FROM THE DISTILLATE HYDRODESULKURIZER (TPY)*
-..CONTAMINANT
SOURCE "~~ _
Distillate HDS
Heater
Relief Valves
Control Valves &
Flanges
Blind Changing
Sampling
Pumps & Compressors
TOTAL
PARTICULARS
Uncontrolled
31.4
31.4
Controlled
31.4
31.4
SULFUR DIOXIDES
Uncontrolled
139.3
139.3
Controllec
139.3
1
139.3
NITROGEN OXIDES
Radian
228.4
"
AP-40
192.0
I
r "--| i
1 228. 4 192.0
HYDROCARBONS
Uncontrolled
12.7
31.9
81.4
0.9
ft. 7
1
137.1
c
Control 1 od
12.7
-0
81..',
-{;
-0
O.S
" " "
94.9
.
*For calculations and assumptions used to determine value.s used for this L-lile, see Appendix 4
-------
Still
Heavy Bottoms
i50 r+_Bo 1.1 ons
i!i eh-Pressure -ov:-Pressure Fraction.-tor
Separator Separator
HDS
Application: Upgrades high sulfur atmospheric residua
to low sulfur fuel oils and minor amounts of low sulfur
naphtha and middle distillate (oplion.il). Modified ver-
sion can be used along with dcsulfurization of crude tower
overheads to pioduce !ow sulfur export crude. Products of
Types Til and IV can be directly cat cracked with per-
formance and catalyst requirement similar to virgin gas
oils; Type II product must be vacuum flashed to provide
suitable cat cracker feed. Alternate designs (Types II,
III and IV) provide increasingly deep desulfurization.
Description (Types II, III and IV) : Fresh filtered feed,
from desalted crude, is heated together with makeup
hydrogen and recycle gas, and mixture is charged to the
reactor section. Hydrogen-rich gas is flashed from reactor
effluent in a high-pressure separator. Heater, reactor sec-
tion, and separator arc usually duplicated in two trains.
Separator gas is purified prior to recycling to reacior sec-
lion; and separator liquid passes through a low-pressure
separator to remove ll.S and fuel gas, then on to the
fractionator for separation of naphtha, middle distillate
(optionalJ, and fuel oil.
Operating conditions: High pressure is used to mini-
mise colce production and assure economically attrac-
tive run lengths. In general, temperatures arc in the range
of 650 to 850° F, and space velocity 0.2 to 2 vol/iir/vol.
C.ttalyst is discarded and replaced on an annvial or semi-
annual sch-dulc. Feed is normally C50° F* reduced crude,
but other cutpoints may be used.
Commercial installations: Four for Kuwait residua.
Economics: Processing 650° F*
Basis: 50.000 bpsd, 2 cycies/vr:
Kuwait reduced crude.
HUi i*-«
Invt-vime.it: (U.S. Gu^f Coiit). $ p lb., 7X'° F). It.
Steam ( \i Ib., >M'd), a>.
Fuel rclt»^(^, .V, P.tu
Waicr, tfolme (?0* Frue),]
Co"idcnvjir (d'^rratcd), (til.
Oit.iN«t co-.t: rr \ti'"bbl
f»l.
u
473
1.8
Jl.9
26
76
"t.3
6.1
III
629
5.1
M.6
41
B4
1.7
14.4
IV
730
7.0
45.5
B4
515
1.7
57.S
Yields: Average for run:
Ko*ail
Gratitv 'API
^u'fyr *^
Ni -+ V, ppm
FIDS T^pc
PrMJutt V.eM
c'i-r«. »t T
f'., 17%'K Nap.'i'k*. vol. 7^
3;5-U5«j K DtttilUt^ TcL *^
f.Vl't J- I uctOl) «-ul *"r
f bcm 11 . Coa*ttm(rfi«i.: icf/bbl. . .
lurl 0,1
C/r^tii) 'API
V.T Crulcr Frctf C^>-il(iO*F)
*i'»Mtv "ATI
f '*r Jt*H. (fVn ) *"J
\i + V. P-,
V«c. Ta-rr bol'9m» ' t !0""1")
C.r«\lt) "ATI
.-..ilur. ;..
tbV « ,-l.n . .
< , In r..,, i .;
II
3.1
O.S
l.fl
< 4
(4 6
iio
1.0
24
UC5
0 fit
111
13
29
K.OiH 1
;i
16.6
3 8
CO
III
37
1.1
3.1
97.}
710
'ojo
J3.S
U '. 4
0.1
131
071
>JO
:s:
t
IV
3.t
14
i.t
V9.9
64{/
23 1
0.10
j- 4
0.10
-------
oil is blended again with i:2 fuel oil to rifike a /'5 fuel oil witli
0.3 percent sulfur by weight.
Overheads generated include saturated sour gas and wild
naphtha. The wild naphtha is sont to the distillate hydrodesul-
furizer for further treatment. The.overhead gas is amine strippec
using diethanolamine (DEA) for 1US removal. The H S is transported
to sulfur recovery and the hydrocarbon gas is routed to the LPG
plant.
The somces of potential pollution from the unit include the
process heater (which exits through Stack #1), relief valves, con-
trol valves and flanges, pumps and 'compressors, blind changing and
sampling. Emissions from this unit are reported in Table 5.
F. DEBUTANIZER (STABILIZER) AND NAPHTHA SPLITTER
The debutanizer is ^ocated downstream of the naphtha hydro-
desulfurizer. The feedstock consists ^?f naphtha from the HDS unit
and the wild naphthas generated in the distillate and residuum
hydrodesulfurizers. The feedstocks enter the unit at about mid-
point in the column. The material is allowed to stabilize with the
stocks lighter than butane leaving overhead to be routed to the
LFG plant. The butane and heavier materials which form the bottoms
from the debutanizer may be routed to a naphtha splitter. The
purpose of the naphtha splitter is to separate additional isopen-
tane for feed to the isomerization unit. The bottoms from the
splitter are sent to the catalytic reformer. FWEC is still uncer-
tain as to whether or not the splitter will be necessary.
These units do not have heaters associated with them. Re-
flux from the bottoms of each column, which are being routed back
and recharged to the systems, are passed through reboilers which
consist of heat exchangers jacketed with steam generated in another
part of the refinery.
-38-
-------
Table 5. EMISSIONS FROM THE RESIDUUM HYDRODESULFURIZER (TPY)*
i
U)
" CONTAMINANT
SOURCE - ~__
Residual HDS
Heaters
Relief Valves
Control Valves &
Flanges
Blind Changing
Sampling
Pumps & Compressors
TOTAL
F '.RTICULATES
Uncont lolled
77.4
77.4
Controlled
77.4
77.4
SULFUR DIOXIDES
Uncontrolled Controlled
343.9
343.9
343.9
343.9
NITROGEN OXIDES
Radian
563.4
563.4
AP-40 I
580.1
580.1
HYDROCARBONS
Uncontrolled jControltcd
31.3 31.:
1
i
J I . V W
S 1 . 4 81.*
i
0.') -U
1
67 0
12.2 I . ;
164.4 114.2
_j
*For calculations and assumptions used to determine values used for this table, see Appendix 5.
-------
Sources of emissions are relief valves, reflux yun'ps, con-
trol valves and flanges, blind changing, and sampling. All process
rlows other than reflux will result without pumps or compressors
due to the operational pressures and the lightness of the materials
involved. Table 6 presents the emissions from the proposed debu-
tanizer while Table 7 reports the emissions from the proposed
naphtha splitter.
Table 6. EMISSIONS FROM THE DEBUTANTZER (STABILIZER) (TPY)*
SOURCE
Relief Valves
Control Valves & Flanges
Blind Changing
Sampling
Reflux Pump
TOTAL
HYDROCARBONS
Uncontrolled
31.9
81.4
0.9
6.7
0.9
121.8
Controlled
-0
81. -
-0
-0
0.5
81.9
*For calculations and assumptions used to determine values
used for this table, see Appendix 6.
-40-
-------
Tible 7. EMISSIONS PROM THE NAPHTHA SPLITTER (OPTIONAL) (TPY)*
SOURCE
Relief Valves
Control Calves & Flanges
Blind Changing
Sampling
Reflux Pump
TOTAL
[_ HYDROCARBONS
! Uncontrolled
31 . 9
81.4
0.9
6.7
0.9
121.8
Control led
--0
81. i
-0
-0
0.5
81.9
*For calculations and assumptions used to determine values
used for tnis table, see Appendix 7.
G. CATALYTIC REFORMING
Catalytic reforming will be used by HREC to economically
upgrade low octane naphthas to produce premium quality motor fuels.
The reactions involved in the process results in the oroduc-
tion of hydrogen gas which can be used to treat products in the
naphtha hydrodesulfurizer and in the isoraerization unit.
The naphtha feedstock (bottoms from the naphtha splitter)
prior to entering the reformer, is hydrotreated for removal of
essentially all of the sulfur (1 ;jpm) which would act as a poison
to the platinum catalyst in the reforming unit. The sulfur leaves
the unit as H_S and is sent to the central refinery amine system.
The sulfur fret naphtha charge and hydrogen rich recycle gas are
heated in a furnace to the reactor inlet temperature and then passed
through fixed catalyst beds in a series of reactors. As tae ma-
jor reactions are endotheraic, the gas temperature drops across
each reactor end furnaces are required to reheat the gas between
the reactors. The effluent gas is condensed and r.eparated into
a liquid stream and hydrogen rich gas. The liquid is processed
-41-
-------
h the stabilizer ard withdrawal as finished, reform?.te. A
portion of the gas is recycled and the renainjc'ir of the gas travels
to the previously indicated destination.
Although the unit planned is a P.henii oiming or similar unit
(see the following, page for process specifications) which has a
fixed catalyst bed, continuously generating beds have been intro-
duced (such as Powerf orrring). Catalyst regeneration is required
because coke is ,'eposited on the catalyst surface during normal op-
eration causing a reduction in catalyst activity. During regen-
eration, the coke is burned off the catalyst under carefully con-
trolled conditions. A continuous regeneration process has a
larger particulate and carbon monoxide emission potential than a
fixed bed operation has.
Foster Wheeler anticipates that the HREC fixed bed reforming
unit will be steam-aired to decoke the catalyst beds once a year.
Gases generated during this operation will be vented to the flare.
The catalyst beds will be replaced every four to six years. This
replacement will have a fugitive particulate potential but due to
the infrequency of the operation it is considered to be insignificant.
Sources of potential pollution froia the unit are the refor-
mer heaters which, nince the number of reactors and furnaces are
known, are considered as one unit (all will exit, through Stack #2),
relief valves, control valves and flanges, pumps, blind changing
and sampling. There are no compressors associated with this unit.
The proposed catalytic reforming unit will have emissions as reported
in Table 8.
H. DEPEN7ANIZER
The depentanizer operates in much the same manner as the de-
butanizer discussed earlier. Loco.ted downstream of the catalytic
reforming unit, the depentanizer feed stocks are debutanizer bottoms
reformats. After stabilization the isopentane overheads are routed
as feedstock to the isomarization unit. The bottoms are blended
with products from the LPG plant and isomerization units to form
various grades of gasolines.
-------
React.,-rs
Pro. '-I, t
'-'i.-p.-i .vilor St.Thili7.t-r
Not HviJrv .-on to ReriniTv
Hydrogen to Kapr.tV
To Light
EnHs.
Recovery
vHI^+ » i
S^^r Reforraatel
Rheniforming
Application: To convert low-octane naphthas to high
yields uf high octane gasoline blendstock or aromatics
plant chargestock.
Charge: Hydrcfinetl straighlrun and cracked naphthas
and hydrocracked naphthas.
Products: Sustained high run-average yields of Cs pins
rcfomute and hydrogen are maintained throughout the
operating c\ c!c permitting 93 to 103 Research octane num-
ber clear severity even with highly pan-ffinic feedstock.
; There is essentiaiK n"> dropoff in either liquid yield or
! hydrogen yield as temperatures reach :he levels where
' hydrocracking noimally occurs with platinum catalyst.
Description: r.henifonninq is a regenerative, fi\-cd-!x-d
catalytic reforming process -which employs a bimetallic
catalyst. Special o]K-rating techniques permit maintenance
of high acti*. ity throu<;Iiout each run and return to fresh
activity after each catalyst regeneration. No commercial
, Uhenifornirr ha> "\cr changed out catalvsi aftn its initial
I filling. The resistance to fouling of rhenium-platinum
icat.ily.st jx'iinits ! >w pressure oj>.?ration (100 to 200 psig
gat last n.irtor outletj. Low jm.'Mirv increases yields of
jaro'.iaU'1 r.tphtha product anil hydrogen. 1 he incic.ised
| resistance to f.uilingaUo pro\ ides for i-N|Wision of cxisiing
plants usi ig higher space velocities and lower recycle
ratios. Converted ur.its arc operating with 1 N/IIC r.ilios of
3 lo 3.J ai.d long cycles lirlwirn regeneration.
Yields: Rheniforniing \iclds for severe reforming of a
paraffinic Arabian naphtha and a naphthenic Isocracked
naphtha'arc as follows:
Foilinc Range, *F
Compo'it on, LV%
Piraflir.i
Naplnhrnet
Aromalict
SuHur, ppm
Nitrogen, ppm
Reactor Outlet Prtinre.
Arablar. Naphtha
1K)-310
686
2J.«
en
< i.c
< 0.5
J2.6
55.5
11.9
< 1.0
< OJ
20il
Produrt Yields
Hvdroiten, jcf,-bbl Iced
Ci-Ci. s
15.7
->B.1
19"
1 ^
1:10
?80
73.0
' 99
31.7
a.i
678
?6
tag
:«9
Ii9
16
H30
100
64.1
100
Z7.4
2 2
7U4
1 5
1 0
:. i
19.6
9.«
Commercial installations:
strcam. 1 !i others are -.indcr e
Rh: uiformcis arc now on-
iiiiTrinn and construction.
Total capacity of the 07 units is :nnrc than 700.000 bpd.
Reference: Gould, C. D. ct. :i!.. "L'ltra Ixiw Picssvire
Klifiiifoimiiif; Gives High Yields." ])res->nted Ix-forc Na-
tional I'c-troleum Rrfmcrs Association, Mian'i, 1 l.i., March
31, 1074.
Licensor: Chevron Research Co.
-43-
Scplcinlirr 1971-
>R'K-.ARH'iN
-------
Table 8. EMISSIONS FROM THE CATALYTIC REFORMING UNIT (TPY)*
- _QQNTAMINANT
SOURCE ->___
Reformer Heaters
Relief Valves
Control Valves &
Flanges
Slind Changing
Sampling
Pumps
TOTAL
PART icu LATHS;
Uncontrolled
127.1
127.1
Controlled
127.1
127.1
SULFUR DIOXIDES
Uncontrolled [Controlled
564.6
564.6
564.6
564.6
NITROGEN OXID!iSl[ HYtWOCAKKONS |
Radian
925.2
924.2
AP-40 llUnconlrollodjControlJ^d j
961.9
96]. 9
50.7
31. 9
81.4
n. 9
6. 7
I .7
173. 3
.
1
50. 7 i
-0
«i.4 :
i
1
n
i
i
-o i
\-- -4
i
i
0.4 '
i :; 3 . n
*For calculations and assumptions used to determine values used for this table, see Appendix 8.
-------
Pu..'1'S are ust-i ':> tc.uv-i^rL t!:C oveihc i.is ;,c-;i hoc.to'js from rl;G
urit. Othoi potential sources of emissions will be relief valves,
control valves c.nd flanges, blind changing, and sampling. The potential
er.issions from the proposed depenv.,inizer unit are listed in Table 9.
Table 9. EMISSIONS FROM THE DKPENTANIZER (TPY)*
SOURCE
Relief Valves
Control Valves & Flanges
Blind Changing
Sampling
Pumps
TOTAL
HYDROCARBONS
Uncontrolled
31.9
81.4
0.9
6.7
1.7
122.6
Controlled
-0
81.4
-0
-0
0.9
82.3
*For calculations and assumptions used to determine values
for this table, see Appendix 9.
I. ISOMERIZER USIT
Isomerization is used to upgrade normal paraffins (straight-
chain hydrocarbons) to isoparaffins (branched chain)1. HREC will
apply the process to mixtures of pentane and hexane to improvr.'
their cctane ratings.
Foster Wheeler indicated that the type of unit planned for in-
stallation can be assumed to be a Hysomer process (see page 46).
The method employs a molecular -sieve (zeolite) catalyst and a fixed
bed. Since sulfur is also a catalyst poison in this process, the
feed is first mixed with hydrogen (provided by the catalytic refor-
mer) to suppress unwanted reactions. Since the feed usually con-
tains a mixture of normal and isopentane, the mixture is first
separated into its normal and iso components in a deisopentanizer.
The overhead product, isopentane, ia sent directly to blending. The
bottom product is mixed with hydrogen and heated by indirect heat
exchange with reactor effluent. The feed stream is further heated
-45-
-------
Mysomer
Application: To upgrade the octane-level of light gaso-
line fractions, predominantly consisting of pentanes and/or
hcxanes.
Charge: Hydrotre^ted light straight-run gasoline (160C F
EP* or other low-octane Cj/C* hydrocarbon streams.
Description: The vapor-phase, fixed-bed hydroisomeriza-
tion process uses a rugged, dual function catalyst consist-
ing of a noble metal on a molecular sieve zeolite base.
Feedstock specifications are not stringent. In Cj/C6 feed-
stocks up to 35 ppm sulfur, water of saturation and several
percent CT's and aromatics can be tolerated. At least 15%
naphthenes can be ri -u.iy processed, chemical hydrogen
consumption is lo-v (0.1-0.3% wt), and die make gas of a
catalytic reformer is suitable without treating or drying.
The catalyst is highly stable and regenerable.
The hydroisomerization unit can to a large extent be in-
tegrated with a catalytic reformer, resulting in substantial
equipment savings. In closely integrating (he Hysomer
piocess with an Union Carbide IsoSiv unit for iso/nonnal
paraffin separation to give a TIP system (total isomcriza-
tion process), a high octane product virtually free of
normal paraffins can be obtained. TIP feedstock speci-
ficntiors arc tiie same as for Hysomer.
Operating conditions, Hysomer:
Temperature, °K 450-550
Pressure, psi? 200-500
Space velocity, voi./vol./hr. 1-3
Hi-h\drocnrKon rool? ratio l-'l
Commercial insmllctions: Two units in operation in
uliich one is inter rated with catalytic reforming. Two
TIP systems m-ccr design/construction, others in en-
gineering and design.
Yields: Typical properties of
product (commercial results).
Gravity, 'API
ASTM distillation, *F
IBP
FBP
Research cx-tane, clear
C.* yield, wt. %
Composition, wt. %
Butanes
Isopcnlane
Normal pentine
2.2-Dimethvlbutane
2,3-Dimethylbutane
S-MethylpenUnes
n-Hexane
Cyclic CJC,
Economics:
feed and isomerized
Feed
90
91
153
73.2
0.7
29.3
-44.6
0.6
1.8
13.9
6.7
Hysomer
91
__
82.1
97.4
1.8
49.6
25.1
5.0
2.2
11.3
2.9
TIP
51
90.7
96.8
2.8
72.0
2.0
5.5
2.5
134
<0-I
Hysomer
TIP
Invcst-nenl (Basis: 8,000 bpsd feed,
including stabli^er. mid-1974 costs), 5
per bpsci capacity .................... 200 375
Catalyst, including roble rnctal ant) adsorbent:
Pint charge, S per bpsd capaci'y ...... 85 !-M)
Replacement, approx. «^, of hrst cost .... 55 65
Expected life, yr ............. ..... 5 5
Typical requirements, incl. stab., unit per bbl feed:
F.lcctririty, twh .................... 1.5 3.6
Fuel, M Hlu ........................ 40 150
Stc.im, low prcs., Ib ................. 33 45
Hydrogen, scf ...................... 85 95
Water," coolinj, gal ................... 50 430
Reference: Chetn. /:n». Progress, Vol. 67, No. 4, April
1971, pp. 65-70.
Licensor: Union Carbide Corp. for process developed by
the Royal Dutch/Shell group of ccmpanies.
-46-
Scptcmbcr 1974
PROCESSING
-------
in 'he isomerization heater and th«.n ;-."-.scd over the <.-uta]ysl in
the reactor. Reactor effluent is cooled, i'irst by exchange with
the feed and then in a water or a.r copied effluent condenser.
Tha condenser is followed by a separator in which the liquid md
gas phases are separated.
The gas frora the separator, primarily hydrogen, is recycled
to the reactor by lueans of a compressor (a motor-driven recipro-
cating unit). The liquid phase contains dissolved hydrogen and
other gases, which are removed in a stabilizer. The stabilizer
is a distillation column in which all liquid reflux is returned
to the column. Cas from the stabilizer accumulator (including any
H S generated) is sent to the fuel gas system by way of the aniine
scrubbing system. Stabilizer bottoms ara a mixture of iso and un-
i
converted normal peutane which are routed to the deisopentanizer
for reseparation. The normal penta.ie is reprocessed through the
isomerizer.
As with the catalytic reformer, the catalyst beds in the
isomerization unit gradually accumulate coke which rr.ust be removed
to maintain activity levels. Since the catalyst bed is fixed, the
regeneration process is performed in place. After the unit has been
shutdown, the reactor is depressurized to the refinery flare gas
system through the relief valves. Inert gas (probably nitrogen)
is used to free the reactor of combustible gas. The catalyst is
brought to temperature by recycling inert gas through the heater
and reactor, and carbon is burned off the catalyst by adding a
controlled anount of air to the circulating gas stream. Since the
stream contains products of incomplete combustion (carbon monoxide)
the off g;)s is usually incinerated in the isomeriz^cion heater.
The regeneration process occurs very infrequently, perhaps once
a year, and should not be considered to be a significant emission source.
A similar insignificant source of fugitive particulate emissions
-47-
-------
will occur every two to three years when catalyst replacement oc-
curs. Sources of more constant potential emissions are the iso-
merizaticn heater (which exits through Stack #2), relief valves,
control valves and flanges, blind changing, sampling and pumps
and compressors. Presented in Table 10 are the potential emis-
sions from the proposed isomerization unit.
The KREC isomerization process should be carefully examined
to determine if organic chlorira is added to the feed. In some
versions of the isopentane isomerization process, an organic
chloride is added to the feed tc increase the catalyst activity.
This chloride eventually appeals in the vapor streams as hydrogen
chloride. Most of the hydrogen chloride is recycled in the pro-
cess, but some is eliminated with the gases traveling to the fuel
gas system. In such cases, this strean will have to be treated
in a caustic scrubber to remove t^ - hydrogen chloride before the
gas is burned as fuel.
J. LPG PLANT
/
After the overhead stream from the crude unit is desuifurized
in the naphtha hydrodesulfurizer unit, it goes to a debutariizer
plant »hich separates the stream into the "bottoms" which
consist of pentane and heavier fractions, while the butanes and
lighter fractions compose the overhead. The overhead stream then goes
to the LPG plant where it is further separated into a butane stream,
a propane stream and fuel gas containing a snail quantity of FUS.
The butane and propane streams go to storage for marketing or blending.
The overhead is further desuifurized or sweetened in the sulfur plant
and goes either to the refinery fuel gas system for fuel or as feed for
the hydrogen manufacturing unit.
The points of potential emissions are the pumps and compres-
sors, relit-.f valves, piping, flanges and valves. Blinds in LPG plants
are usually of the Hamer type which can be changed almost
instantaneously without loss. This valve has an intagral hand-
-------
Table 10. EMISSIONS FROM THE ISOMER12ER UNIT (TPY)*
i
.£>
^^ -^CONTAMINANT
SOURCE ~"~^ - ~^_
Isomerization
Heater
Relief Valves
Central Valves &
Flanges
Blind Changing
Sampling """""
Pumps & Compressors
TOTAL
PARTICULATES
Uncontrolled
21.90
21.90
Controlled
21.90
21.90
SULFUR DIOXIDES iNITKOGEN OXIDES
Uncontrol led
95.6
95.6
Control led l|r
-------
v/neel to release -he pressure on a rubber-gasketed doub]e spec-
tacle blind. One side is solid and the other is ring-shaped for
use during normal operations. When the pressure is released, the
blind is merely slid across to the other position 'ind the pres-
sure reapplied by the hand-wheel. These blinds are relatively
expensive, but redeem their cost by being situated only in posi-
tions where frequent changes are required such as pumping mani-
folds. These blinds can be chJnged by the operators without
tools. Also the mechanic's time normally required for a conven-
tional blind is eliminated. Cocks are normally used instead of
valves in LPG service also.
The two-stage cencrifugal compressor will have mechanical
seals vented to the flare. All pumps will be conventional reci-
procating pump: with packing glands. The examiners consider a
packing gland to be the beat current control technology for a
reciprocating pun-p. however, recent literature has pointed up
the possibility cf application of the mechanical seal to various
pumps. The September 1976 issue of Hydrocarbon Processing, page
431 announced the existence of a new off-the-shelf mechanical
seal which covers 90 percent of all pump applications. They can
be interchanged with nearly all existing pump seals or packing
without requiring re-engineering. The r.anufacturtr is. A.W.
Chesterton Co., Middlesex Industrial Park, Stoneham, Massachusetts
02180.
To date examiners have been unable to obtain literature
describing these seals and do not Kr>ow whether they have poten-
tial applicability to the HREC refinery.
It is not known at this time whether or not a Merox unit
will be required. If so, the purpose of this unit is to remove
apy remaining inercaptars from the LPG by converting them to di-
sulfides. Potential losses fro:n this unit ar?. reported in
Tables 11 and 12.
-50-
-------
Table 11. EMISSIONS FROM THE LPf PLANT (TPY)*
SOURCE
1
Relief Valves
Control Valves & Flanges
Blind Changing
Sampling
Pumps & Compressors i
TOTAL
r
imROC.-UBOXS
L'n ecu t roil lei
! 31.9
31.4
Ne?,.
6.7
11.4
131.4
\~
Cnnrrol led
-0
81.4 1
-o I
-o ;
0.,
89.6
*For calculations and assumptions used to determine val-
ues used for this table, see Appendix 11.
Table 12. EMISSIONS FROM THE OPTIONAL MEROX UNIT (TPY)*
SOURCE
Pumps
TOTAL
HYDROCARBONS
Uncontrolled
1.7
1.7
Controlled
0.1
0.1
*For calculations and assumptions used to determine values
for this table, see Appendix 12.
K. AMINE TREATING
Because Foster Wheeler has not performed the detailed en-
gineering design work for tha HREC refinery, it is impossible to
judge whether or not the amine system planned will be adequate
to handle fluctuations in the percentage of sulfur in the in-
coming crude oil. As a result, this discussion must remain
general in nature.
-------
An estimated 270,000 metric tons per annum (MTPA) of sour
gas will be routed to amine units. In addition to the aiiine
scrubber associated with each of the three hydroJesulfurizers,
Foster Wheeler will have a central amine stripper to handle
the rich amine solution from the amine scrubbers.
In an amine unit, the hjdrogen sulfide is removed by ab-
sorption in an alkaline solution. The central unit, the naphtha
hydrodesulf u'-' zer and the distillate hydrodesulfurizer will use
monoethanolamine (MEA) as the alkaline material while the residuum
hydrodesulfurizer system will us° diethanolamine (DEA). These
materials are chosen so that the chemical bond formed during ab-
sorption can be broten by heating. The hydrogen sulfide is
stripped from the t.eated solution which, after cooling, is ready
for reuse.
The absorption medium is a 10 to 20 weight percent of MEA
(or 20 to 30 weight percent of DEA) in water. The H S is re-
moved from tha sour gas by contacting with the solution in the
countercurreni: absorber. The sour gas enters the bottom of the
column and cool, lean amine (containing no H.S) enters at the top.
Treated gas leaves the top of the absorber and passes to the re-
finery fuel gas system. Foster Wheeler estimates this refinery
fuel gas will contain less thau five grains of sulfur per cubic
foot. Rich amine (containing H«S) is used to cool the lean amine
and is fed to the top of the. stripper. Steam used for stripping
the rich amine is generated by boiling the stripper bottoms in the
reboiler. HjS and steam leave the top of the stripper column and
the steam is condensed. Condensate and H-S are separated in the
acid gas separator and the condensate is pumped back to the strip-
per as reflux. The H-S leaves the condenser as overheads and is
routed to the sulfur plant. Hot lean amine from the stripper re-
boiler is cooled and filtered before returning to the absorber.
The H.S does not have to be pumped to the sulfur plant but
travels due to the pressure gradient. Rich amine is not pumped,
but lean aoine is. Each unit has a reflux pump.
-52-
-------
H^S is a highly corrosive material. Over the years, tie
need for better corrosion protection had led to the development of
a complete scientific discipline developed aroano the corrosion
engineering field. Studies using such devices as corrosion spools
and reference points have led to published data establishing the
relative abilities of different materials to handle corrosive ma-
terial.
Upsets in an amine unit of a nature extreme enough to re-
quire more than a standard shutdown of the atnine stripper or
one of the three hydrodesulfurizer amine scrubbers is extremely
unlikely. However, should such an event occur, and the sour gas flow
is not able to be bypassed to another unit, it will be flared. The
repairs to the amine unit can be quickly accomplished since the
units are generally built with screwed piping which can easily be re-
placed with minimal down time.
There are no pollutant emission points associated with this
unit. The process is fully enclosed without any relief valves.
Flanging is not used and there are no fired heaters.
L. SULFUR PLANT
The HREC refinery will utilize the Claus process to convert
hydrogen sulfide to elemental sulfur. The unit planned for this
refinery will have two trains, each with three reactors having a
total conversion efficiency of 96 percent. The total capacity of
the unit will be 400 long tons per calendar day. Each train will be
designed to handle between 55 to 60 percent of this load during
normal operations. During shutdown of one of the trains, the units
would have a maximum capability of 75 percent of the peak load.
During these periods, the refinery would be forced to reduce the
crude oil throughout to a level which could be handled by one train.
-53-
-------
In the Glaus reaction, hydrogen sulfide is converi.ed to elemen-
tal sulfur in two steps according to the following reactions:
H2S + 3/2 00 »~S02 + H20
2H2S + S02 »-2H20 + 3S
In the first step, H?S is partially burned to SO with air. The
H_S/SO- mixture is then reacted over a catalyst to proouee sulfur
and water. This reaction is known as the shift conversion
and is carried out in two or three stages with sulfur removal af-
ter each stage. The design of a sulfur recovery plant depends
upon the inlet H_,S concentration. If the concentration of H»S in
the feed is high, a "straight-through" process is used. In the
straight-through configuration, all of the H~S and air are fed
tJ the burrer (boiler). If the H-S concentration in the feed is low,
a "split-flow" or "sulfur recycle" process is used. Foster
Wheeler does not know what the H.S concentration will be. In
the "split-flow" process, a portion of the feed is burned com-
pletely to S0_ and combined with the remainder of the feed to
provide the proper H^S/SO.ratio for the shift conversion. In
the "sulfur recycle" process, the product sulfur is recycled to
the burner to support combustion. A fourth type of sulfur re-
covery process uses the "direct oxidation" approach. This con-
figuration, which is for very lean feeds, eliminates the burner
by feeding the acid gas/air mixture directly to a catalytic bur-
ner.
Most sulfur plants in refineries are the "straight-through"
or "split-flow" type. The straight-through process is used when
the inlet stream to the Glaus unit contains at least 50 percent
of H_S. The split-flow process is used for the leaner H,S streams.
While Foster Wheeler does not know what the H-S concentration will
be, the inlet gas can be expected to be fairly rich allowing the
use of the straight-through process since the refinery will use
amine stripping. The acid-gas stream, typically containing
-54-
-------
H9S iSU to 93 percent), CO,, (2 t,. it) p.-r^nt), water (5 to 10
percent), and ninor amounts of hydrocarbons (0.5 to 2 percent),
is fed to an inlet separator where any entrained liqu.f.d is re-
moved. The acid gas and air are fee) to rhe sulfur boiler. Fuel
gas lines are provided to assist in planr startup. Boiler feed
water is fed to the .sulfur boiler to generate low-pressure steam.
The sulfur boiler usually contains three tube passes. A portion
of the gases is diverted from the boiler after two passes to
provide preheat for the shift converter feed streams. In some
plants processing leaner H_S streams, an auxiliary burner is pro-
vided to furnish preheat to the reactors. Liquid sulfur is sep-
arated from the boiler effluent gases and is sent to the sulfur
pit for storage. The gases are mixed with the reheat stream and
fed to the first-stage shift converter. The effluent from the
reactor is passed through the condenser. Sulfur is separated and
sent to the sulfur pit and the gases are fed to''the second-stage
converter. If the sulfur plant operates at projected conversion
efficiency, 4 percent of the sulfur in the feed will remain in
the tail gas. Incomplete conversion occurs because the basic
Claus reaction is reversible and limited by chemical equilibrium.
The sulfur is converted to. SO- In the tail gas incinerator.
In some sulfur plants, the .-.ulfur boiler and condenser are
combined into a single unit. The catalytic converters are often
combined into a single horizontal or vertical vessel. Combining
the units in this manner permits many smaller plants to be shop
fabricated and skid-mounted.
The acid-gas feed to the sulfur plant is on pressure con-
trol. The plant is designed to accept acid-gas flows as they de-
velop in the'refinery and a plant can be designed to operate at
as low a value as 25 percent of capacity to allow for future de-
mand. The ratio of acid gas to air is controlled by a special
ratic flow controller. To set the proper ratio, the acid gas
must b° analyzed regularly to determine the H^S concentration.
-55-
-------
The proper H0S/air ratio is critical to the operation of the
plant since the concentration of S09 in the incinerator stack
gases increases if the improper ratio is set. Automatic
stream analyzers can be used to set the proper ratio on a nearly
continuous basis. The volume of reheat streams is controlled to
set the inlet temperatures to the shift converters. If these
temperatures are allowed to fall, the reaction is incomplete
and sulfur recovery will drop. Incinerator stack temperatuie is
controlled by setting fuel consumption and air bypass. A high
temperature is required to ensure complete combustion of sulfur
compounds to S0_.
The tail gases from a typical Glaus sulfur recovery
unit contains about one-third water, 5 percent to 15 percent
CCL, 2 to 4 percent sulfur compounds of all kinds, and the
balance nitrogen. The percentages of sulfur compounds expressed
in parts per million, is roughly 20,000 to 40,000 ppm. Most
control agencies now expect or limit the concentration of these
emissions to 250 ppm or less. In order to reduce the sulfur
content of the tail gas to an acceptable level, it becomes nec-
essary to further treat this effluent in a "tail-gas clean-up
unit."
The tail-gss clean-up unit" selected to be employed by HREC
and described below is known as the Beavon Sulfur Removal Pro-
cess or a Beavon-Stretfor unit.
The Stretford Process has been used in Great Britain for
many y-ars to recover hydrogen sulfide from natural gas and
convert It to sulfur. The feed gas is passed through an absorp-
tion tower which removes the H S. The absorbent is an organic
liquid which also serves to oxidize the dissolved H.S to sulfur.
The sulfur is removed from the liquid by filtration, Jnd the sol-
vent is regenerated by air oxidation. Very high conversions
of H-S to sulfur are possible with this process. Since the
Stretford Process is not suitable for use with feed gases con-
-56-
-------
taining ?00, modification had to bo a.-velopej. Tnis nod if i cat ion is
called the Beavon Gullur Rewal Process.
Tin.-. adv.-mtiim?s of L I- Stroifoia ,;LOC...-SS are ti'iat
it functions well at atmospheric pressure, is unaffected bv i_he
carbon dioxi-!e prescMit, produces high-quality sulfur while puri-
fying the gas to a very high degree and does not re-quire special
operator skills-. Several United States engineering firms nave
produced operational units of this processParsons and Pritchard
primarily. Of the two, Foster Wheeler feels that the Parsons
design is the better and PES agrees. A Pritchard unit w.is in-
stalled at a Southern California refinery. Problems in the water
cooling and condensing sect:'on caused a chronic plugging problem
which resulted In frequent down time.
The Beavon Stretford unit Dimply hydroger.ates the sulfur
compounds, 302, ^°^» ^7 to ^2^ unc*er moderate temperature and
pressure conditions using a cobalt-molybdate catalyst. After the
reactor, the hydrogenated stream is cooled, water is condensed out,
and vapor, containing H~S, is ready for processing to eliminate the
sulfide. It would be desirabl to return this lUS to the Claus
plant feed but, unfortunately, the stream contains so much CO^ that
cannot be easily removed that that the build-up of inert gas in the
Claus unit could not be tolerated.
Since tne H S concentration is about iO,000 ppn> and must
ultimately be reduced to 1 ppm, a Stretford section is added.
The H_S stream is directed into a column and contacted with
sodium carbonate to convert it to sodium hydrosulfide. This is
oxidized tc sulfur by sodium vanadate. Subsequently, vanadium is
oxidized'back to the penta valtnt state by blowing in air with
sodium anthraquinone disulfcna';e working as an oxidation catalyst.
Sulfur particles ere finely divide^ and appear as a froth which
is skimned, filtered and returned to the Claus plant to be in-
-57-
-------
eluded in the elemental sulfur product. The tail gas will now
contain less than 250 ppra SO and 10 ppin of ^,,8. Overall sul-
fur recovery efficiency is expected to be 99.9 percent.
An important consideration to be examined is whether or not
the sulfur unit planned by Foster Wheeler is sized ]arge enough
to bandle the crude oil. The total capacity of the unit will be
400 long tons/calendar day. The refinery throughput is expected
to be 175,000 barrels of crude oil/calendar day with a sulfur
content of 1.7 percent by weight. The crude oil has an API gravity
of 34.2 which corresponds to approximately 7.12 pounds/gallon.
The nmount of sulfur entering the refinery each calendar day
would be:
(175,000 barrels/day)(42 gallons/barrel)(7.12 pounds of
crude oil/gallon)(0.017 pounds of sulfur/oound of crude oil)
= 889,644 pounds of sulfur/day e/iterinj, tne refinery
Using the most conservative estimates o; other sources of
emissions, it is assumed that all other pro: acts will be sulfur
free. The only points at which sulfur leaves the refinery is as
#5 fuel oil (product and refinery fuel) and iv> the sulfur plant
(refinery fuel gas is assumed to be sulf ur-f rer.1.
(76,930 barrels/day) (42 gallons/barrel) (/ . :>d6 pounds of
crude oil/gallon)(0.003 pounds of sulfur/pound of crude oil)
= 73,532 pounds of sulfur/day
Thio leaves 816,112 pounds/day of 3ulf-jr which must be recovered
by the Glaus unit. A Claus unit with a conversion efficiency of
96 percent capable of recovering 400 long tons/calendar day of
sulfur is capable of receiving:
(400 long tons/day)(2240 pounds/long ton) = 933,333 pounds/day
(.96) of sulfur
This demonstrates that the sulfur plant as planned is capable of
handling the anticipated crude oil. Another point which must be
studied is the possibility of HREC having to run higher sulfur
-58-
-------
crude oil stocks through ihe i-?fin< cy. Depending on bow firn a
contract HREC has niade for their crude, they may have to run
different crudes. If the .mine and hydrodesulfurization units are
sized large enough to ha.idle the changes, how high a sulfur con-
tent can the sulfur plant handle? Working bacl-^ards through the
steps just taken yields:
(933,333 pounds/day of sulfur) + (73,532 pounds/clay of sul-
fur in #5 Fuel oil) = 1,006,865 pounds/sulfur day entering
the refinery
(1,006,865 pounds of sulfur/day)
(175,000 barrels/day)(42 gallons/barrel)(7.12 pounds of crude/galIon)
= 0.0192 pounds of sulfur per pound of crude
In other words, the sulfur plant could handle up to a 1.92 percent
sulfur crude oil.
The sulfur plant uncontrolled and controlled emission estimates
are shown in Table 13. All of these emissions exit through Stack #1,
-59-
-------
Table 13. EMISSIONS FROM THE SULFUR PLANT (TPY)*
-CONTAMINANT
SOURCE "~ -^- __
Sulfur Burner
Sulfur Plant Incin-
erator Uncontrolled
Sulfur Plant Sulfur
Removal Process
TOTAL
P ARTICULATES
Uncontrolled
12.1
12.1
Controlled
12.1
12.1
SULFUR DIOXIDES
Uncontrolled
5J.8
11,480
11533.8
INITROGKN OXIDIiS
Controlled!) Radian
5'i.8
NA
287.0**
340.8
88.4
88.4
AP-40
62.9
62.9
HYDROCARBONS
'Uncontrolled
4.6
4.6
Controlled
4.6
....
.
4.6
1
*For calculations and assumptions used Lo determine values used for this table, see Appe-'dix 13.
^Emissions are calculated as S02. In actuality, emissions will be as carbonyl sulfide (COS).
NA = Not Applicable
-------
STEAM
As mentioned previously, the steam requirements for the re-
finery will be generated from three generators, each capable of
supplying SO percent of the refinery needs. The boiler. fecdwater
makeup will be supplied by process water.
The combustion contaminants from the steam generators will
exit through Stack #3 and are shown in Table 14.
Table 14. EMISSIONS FROM THE STEAM GENERATORS (TPY)*
P \RTICULATES
132.6
SLLFUR OXIDES
589.2
. , (
NITROGEN OXIDES 1
Radian 1 AP-40
965.1
986.9
i f
i
HYDROCARBON'S
53.7
*For calculations and assumptions used to determine values used
for this table, see Appendix 14.
N. HREC MISCELLANEOUS COMPRESSORS AND PUMPS
There will be a centrifugal compressor which will operate
on the fuel gas systeai. This compressor will be equipped with
simple mechanical seals which will vent any escaping gases to the
flare system. This system will also have a centrifugal pump
equipped with a simple mechanical seal. The fuel gas produced
by the process units will be amine scrubbed and then fed to a
collecting main and passed to a central knockout vessel. From
this vessel the mixed fuel gas stream will be delivered to the
process units. In order to transfer the refinery fuel oil to its
necessary destination, a motor-driven pump is supplied. The fuel
oil will be stored in tanks in the utility area near the boilar
house.
All of the HREC refinery products will hi.ve final storage
before marketing at the SMT facilities adjacent to the refinery
site. Each of the products has a pump to transport it to stor-
age. This is a total of seven pumps.
-61-
-------
Several othe1- systems exist at the refinery which are not
important from an air pollution standpoint but are shown for
completeness. Two compressors, one turbine and one motor-driven,
with one receiver and one dryer will be provided to supply plant
and instrument air. Each compressor will deliver the required
quantity of instrument air and a limited amount of plant air.
One corapressoi will be in normal operation while the other compres-
sor will be on automatic start on pressure control.
During a unit shutdown, when more plant air will be required,
both machines will be in operation.
The HREC plant will maintain a diesel generator to act as
an emergency electrical system. Since the unit will only be used
in cases of emergency, emissions were not calculated.
Emissions from the various pumps and compressors described above
will be in the form of fugitive hydrocarbons and are listed in Table 15.
Table 15. MISCELLANEOUS HREC EQUIPMENT EMISSIONS,(TPY)*
1 .
UNIT AND SERVICE
Refinery Fuel Gas Compressor
Refinery Fuel Gas Pump
Fuel Oil Pump
Seven Product Punps
TOTAL
HYDROCARBONS
Uncontrolled
1.9
1.9
0.7
6.1
10.8
Controlled
1.9
1.9.
-0
2.9
6.7
*For calculations and assumptions used to determine values used
for this table, see Appendix 15.
0. EMERGENCY FLARE SYSTEM, SLOWDOWN, STARTUPS, SHUTDOWNS
It is not possible at present to operate a refinery con-
tinuously without having to shutdown and start up units or with-
out havi-ig a certain number of emergencies requiring the release
-62-
-------
of large quanuitii-s 5t gases or '.thout ijenerating quantities
of waste gases during normal operation. In ordei.- to dispose of
these gases safely with a minimum quantity of pollutants entering
the atmosphere, the fades must be incinerr.ted in on emergency
flare systeia. A typical flare system consists of tlie collecting
lines or blowdown lines from ths various units. All pressure
vessels, some pump seals and some pressure relief valves are
vented to this blowdown system along with any source of waste gas
that requires incineration.
Between the unit and the flare, the gases pans through
a-knockout pot to eliminate any condensate. The condensate
is usually puir.ped to a slop oil tank which, after settling,
is pumped back to the crude unit for reprocessing. In
the proposed HREC system, there will be two flares of differing
types. One will be a ground type w'li ch if. low and enclosed in
a louvered shroud to eliminate noise and light. The other is
an elevated smokeless type.
A water seal drum is placed ahead of each flare. The drum
associated with the ground flare regulates the incoming gas pres-
sure to allow combustion up to a given flow rate. This procedure
will take care of normally generated waste gas flows. Iii the
event of a necessity for "dumping" a unit in an emergency, the
amount of gas in excess of the ground flare capacity will go to
the elevated flare. In the event of an upset in the sour gas
system, this gas would be dumped directly to thi_ elevated flare
to ensure that contamination of the normal refinery generated gas
does not occur. The sour gas will be routed through a separate
stainless steel valve and flange system to the burner.
Each flare requires steam to inspirate sufficient air to
burn the gases completely and smokelessly. An automatic demand
system provides the proper quantity of steam. A manually
operated valve can add additional steam, if necessary, to improve
-63-
-------
combustion conditions and reduce, a smoky condition that might
temporarily arise. Each of the flares will be on a continuous
pilot fueled by refinery fuel gas.
In a normal start-up procedure for a fired unit, the sys-
tem is filled with a cold, non-volatile stock such as gas oil
which is circulated throughout the system to check proper operation
of all equipment and detect any possible leaks that might have
been overlooked. The feed is then started to displace the
gas-oil and temperatures are gradually brought up to the normal
operating point. Start-ups will rarely create an emergency or
any great loss of gases to the blowdown system. On a normal
shutdown where the standard procedures can be followed, only
slightly more gas is released to the flares. However, emergency
shutdowns caused by fires, power outa^ , cooling water failures,
etc. can release tremendous quantities of gas to the flares.
These are not predictable and may never occur. However., the
flare system is designed for handling the worst case and even in
an emergency, the amount of hydrocarbons to the atmosphere will
be minimal. The ground level flare will be constructed to be
capable of handling up to one million pounds of material an hour
during an emergency. The combustion zone in such a flare is
very large, having a diameter in excess of 2rJ feet (the authors
have observed a ground level flare in Los Angeles with a diameter
of approximately 50 feet).
Table 16 shows emission estimates due to piloting and upsets.
P. COOLIN'G TOWERS
HREC will use air as the primary cooling medium. The oil
will be routed by pipe through the cooling towers. The towers
are designed with the oil-bearing tubes situated atop a four-
bladed motor-driven fan. Ambient air is driven past the
-64-
-------
Table 16. EMISSIONS FROM THE FLARING SYSTEM (TPY)*
-^CONTAMINANT
OP ERAT ION "
Piloting
Upsets**
TOTAL
PARTICIPATES
-0
0.15
0.15
SULFUR DIOXIDES
0.5
35.00
35.5
NITRO<:F,N OXIDES
0.2
2.00
2.2
HYDROCARBONS
-0
0.2
0.2
*For calculations and assumptions used to determine values used in this table, see Appendix
16.
**These upsets are associated with startup and shutdown.
-------
tubes, cooling tho material. Air cooling is less efficient than
water cooling but is desirable from an air pollution standpoint
because it is not a source of hydrocarbon emissions as i= water
cooling. Emissions result whe.n water is used due to the fact
that salts in the circulating water cause accelerated corrosion
of the heat exchanger tubes. Leaks in the tubes cause the
water to become contaminated with oil which is emitted from
the water cooling tower. In order to reduce exchanger tube
problems, all refining tube seals will be welded.
Foster Wheeler has calculated that air ccoling can handle
all of the refinery cooling requirements as long as the ambient
air temperature remains at or below 90°F. During the summer
months, the daytime temperatures will exceed 90°F and the water
cooling system will have to be used. It has been estimated that
the hydrocarbon emissions which will result from the use of the
water cooling system will be three tons/year (see Appendix 17).
Q. OIL-WATER SEPARATION
Refineries generate significant amounts of process water
which has been in contact with oil and chemicals. In addition,
rainfall which has fallen on the refinery site may become con-
taminated with these same products. These wastewater streams
require extensive treatment before discharge into a body of
water.
The HREC refinery will be located in an area of relatively
high rainfall. As a result, Foster Wheeler has developed a sys-
tem to handle surges. The process drains from each of the major
process units will all be underground and closed. In order to
fully enclose the sewers, HREC should be encouraged to install
water traps on the drains. During dry weather, process water will
drain to the low total dissolved solids (LTDS) syster. The sewers
will transport the process water containing approximately 500 ppm
of oil at an average rate of 230 gallons/minute to a divertcr.
After passing through a trash rake, the process water enters the
diverter.
-66-
-------
The diverter is a box approxinjtely ten feet square and
eight feet deep and is divided into two ciu'ir.ibers bv an underflow
baffle. During normal operations the baffle does not become
important, since the vater flows directly to the bottom of the
chambers where a pump transfers the water directly to the final
separator.
luring a severe storm (five to six inches of rain in 24
hours) rainwater will fall in t'..e process unit areas and enter
the process sewer system. The amount of water entering the LTDS
system and the diverter could reacli 500 gallons/minute. Since
the final separator pump can only handle 250 gallons/minute, in
such cases the diverter capacity will be exceeded and large
amounts of water will build up in the unit. An overflow pond
system is located downstream of the underflow baffle to handle
these situations. The water in the diverter is forced under the
baffle as the level increases. Because the oil has a lower
density than water, the majority of the oil will remain in the
skimming section instead of moving under the baffle. This oil
will be collected by a Brill rope skimmer. The rope is made of
a permeable material which continuously rakes up the oil. As
a result, Foster Wheeler believes the overflow water reaching
the pond will only have 150 ppm of oil. Once the storm has
passed and th« process water rate has been reduced to normal,
the storm water diverted to the overflow pond will be pumped back
to the diverter for processing through the final separator.
Rain falling in the tank farm will be impounded within the
spill dikes for dispersal at a controlled rate. If testing shows
the impounded water to be clean and free of oil, the water will
be drained through a network of ditches and sewers to the LTDS
wastewater holding basin. If the impounded water shows traces of
oil contamination, it will be drained to the LTDS sewer system at
a controlled rate for treatment.
-67-
-------
Clean storm water runoff from open areas are noi subject
to oil contamination. This water will travel thr-jugh a
series of ditches and storm sewers to the LTfiS created water
holding basin for use as fresh water make-up.
The second sewage and treatment system, termed the high
total dissolved solids (HTDS) system,serves such units as boiler
blowdown water, cooling tower blowdown, desalter water and
other streams which contain a large amount of dissolved minerals
and solids. There is no diverter on this sewer system. After
passing through a simple trap, this water enters another final
separator at the rate of 342 gallons/minute. FWEC is still un-
sure as to whether there will be two fins] separators or only
one.
The final separators will be corrugated plate intercep-
tors (CPI) with a fixed roof covering. Corrugated plate oil
separators were developed in the years 1964 through 1966 in the
Netherlands. The development was a joint effort by the Royal
Dutch Shell Oil Compar.y and Pielkenrood-Vinitex. The heart of
a corrugated plate separator is the corrugated plate pack. In
it a number of corrugated plateseach 0.05 inches thickare
mounted parallel to each other at a distance of three-fourths
of an inch. FO:T ease of handling during installation, the
plates are contained iri a box like structure. The plate pack
is installed at an ai-.gle of 45° in a process tank. The fluid
flow to be separated passes between the closely spaced plates
in the pack. Laminar flow conditions are established while the
fluid passes from par.k inlet to pack outlet. This is an ideal
condition for gravity separation of the disimilar components
in the fluid.
The light components in the fluid being treated flow up-
wards into the tops of the corrugations and rise to the high
end of the plate pack. From there they rise to the top of the
process tank where they are removed. The heavy components of the
-68-
-------
fluid will settle in the bottoms of the corrugations and slide
to the low end of the plate pack. From iliere they fall i; to
the bottom of the process tank to be discharged intermittently
through a blow-off valve.
The flow of fluid is in a downward direction through the
pack, when removal of light components is the principal objec-
tive of the separation process (see Figure 9). VJlien heavy
components predominate and they need to be removed, the fluid
flow is upwards through the pack (see Figure 10).
_LIGHT
COMPONENTS
HEAVY
^COMPONENTS
Figure 9. LIGHT COMPONENTS SEPARATION
-69-
-------
INFLUENT
EFFLUENT
HEAVY
COMPONENTS
Figure 10. HDWY COMPONENTS SEPARATION
Advantages of this sytem ove.r an API oil skimmer art: (1)
lower operating and maintenance costs; (2) smaller space re-
quirements because liquid detention time in the CPI is only three
to ten minutes instead of the one to six hours necessary in a
conventional settling device; (3) because the unit is so compact,
a fixed roof can easily and effectively be installed; and (4)
the relative low cost of such a unit allows installation on
separate process streams instead of routing to one central unit.
The effluent from the CPIs will flow into air flotation
units where fine droplets of oil still remaining and finely
divided solids are removed. These particles, being lighter than
water, do not settle out easily.
-70-
-------
In t Ue flotation pr.'ress a colUxlial flocculanf and .iiv ur-
Jer pressure are injei-tcJ into the var, c ew.u or. The stream is tnen
fed to a clarifier through a back pressure valve that reduces
the pressure to atmospheric. The dissolved air is suddenly re-
leased in the form of tiny bubbles that carry the particles of
oil and solids to the surface where they are skiTjaed off by
mechanical flight scrapers.
Leaving the air flotation unit, the water will enter a
biological oxidation unit. In this unit active biological
sludge, with mechanical aeration, rapidly consumes the organic
substances in the water. The effluent from the biological oxi-
dation unit flows to a holding basin for storage prior to being
recycled to cooling water makeup and to fresh water treating
for reuse.
Table 17 shows the emissions attributable to the waste
water facilities. Sources of emissions to be considered are
the LTDS diverter, overflow pond and final separator; and the
HTDS final senarator. The process sewers are all underground
and closed, and do not constitute a problem. The air flotation
systems will be under pressure and do not cause emissions.
Table 17. EMISSIONS FROM OIL-WATER SEPARATION (TPY)*
UNIT
LTDS System
HTDS System
TOTAL
HYDROCARBONS
Uncontrolled
328.5
449.4
777.9
Controlled
20,0
18. C
"< 0
.
I
*For calculations and assumptions used to detir;nJr.^ values
r.sed for this tabli?, see Appendix 18.
-------
R. ?;: Iijl WATF.K STRIPPER
"Sour water'1 is Che refinery parlance for water or conden-
sate beating or containing sulfur compounds such as hydrogen
sulfide or nercaptans. This water comes from the crude dis-
tillation unit and ir.any other small sources such as knockout
drums in the flare system.
The sulfur-bearing water is heated by a steam reboiler to
a temperature at which the sulfur compounds are volatized and
leave as overhead and the "stripped water" goes out as bottoms.
This water is suitable for use in the desalter unit for crude
distallation feed preparation or can be routed directly to the
process water disposal system. In the. HREC refinery the stripped
water will be pumped back to the crude unit and used in the
crude desalter unit. The overhead passes through an amine scrub-
ber and the V S, when stripped, forms part of the acid gas feed
to the Claus unit.
Foster Wheeler is still undecided whether they will design
two sour water strippers, one operating and one spare, or use one.
stripper and a feed surge drum.
There ere no sources of emissions associated with this unit.
A possible surge drum would be enclosed. The on-off pump which
transports tl e recovered water back to the desalter is not a
source of emissions.
S. SLUDGE 1NCINERATOP
Primary sludges are generated by removal cf suspended solids
fror process wastowater by gravity separation in the CPI and air
flotation units. These insoluble solids consist of such things
os sand and grave.!, bits of iron rust, bits of metal, carbonaceous
material, small particles of asphalt and similar particulates.
-72-
-------
Secondary sludges are composed primarily of water and organic mat-
te1' with little ash, sulfur or hydrocarbons.
Foster Wheeler has stated that while several new systems for
treating sludges without incineration are being examined for
possible application, the besc currently available technique for
disposal of this material is some type of incinerator.
Use of an incinerator eliminates objectionable odor-produc-
ii.g materials from the sludge before it is sent to a land-fill area.
The selection of a sludge incinerator must take into account the
types and quantities of sludge materials to be processed as well
as the emission constraints which will be imposed on the flue
gases from line unit. The two major types of sludge incinerators
which are currently use 1 in refinery applications are: (1) f.uid-
.ir.ed jed units; and (2) multihearth units.
1. Fluidized Bed Incinerators
The use of fluid bed incinerators has proven an effec-
tive technique for sulfite sludge disposal in the paoer industry
for many yearc. Only recently, however, has this process been
applied to refinery wastes. The first such unit was constructed
at the Mandan, North Dakota, refinery of American Oil Company
(AMOCO) i" 1969. This initial unit was constructed to liandle
oily API separator sludge bottoms, tank bottonis, stable emulsions
and spent caustic solutions. Spent caustics and oily sludges are
fed to separate steam-heated vessels where they are kept in a
riore or less homogeneous fluid state by mixers. A blwvjr intro-
duces air through a bed of hot sand and fuel is admitted until
the bed temperature reaches 1300°F. When primary sludge, is burned,
the fuel will be oil. Secondary sludge will use refinery fuel gas.
Oily sludges and spent caustics are fed to the bed at such a rate
to maintain a constant bed temperature. Sludge having a heating
value of 57,000 BTU per gallon proved to be adequate .it the
-73-
-------
Mandan installation. As the sludge is burned, solids accumulate
in the fluid bed while gaseous products discharge through the
gas-cleaning system. Any sodium oxides which are formed react
with combustion products to form sodium carbonate and sodium sul-
fate, both of which stay in the bed. Ash is periodically with-
drawn from the bed as it accumulates. Ash solids are inert and
are well suited to land fill.
Sludge water (as steam), gaseous products of combustion,
and suspended fine ash particles pass through a process cyclone
which either returns solids to the bed or discharges solids with
the excess bed material. From the cyclone, gases are passed
through a water scrubber. Stack gases to the atmosphere consist
primarily of water vapor, nitrogen, oxygen, carbon dioxide, a
few tenths of a part per million of sulfur dioxide and less than
C.03 grains per day SCF of particulate matter.
The advantages of the fluid bed incinerators systems are:
(1) they are relatively low polluting and odor-free, with less
than 0.03 grains particulate per dry SCF; solids formed are
inert; (2) the units are simple and relatively inexpensive to
operate if the sludge has sufficient heating value; and (3) many
types uf sludges can be processed including waste oils and bio-
logical treating sludges.
2. Multihcarth Units
The most sxtensively used system i§ the multiple-hearth fur-
nacu. After preliminary diverting to reduce the sludge moisture
content from its original 90-plus percent to about 30 to 50 per-
cent, drying and combustion are accomplished in the roultiple-
hearr.h incinerator, shown in Figure 11. The nultiple-hearth fur-
nace consists of a circular steel shell surrounding a number of
solid refractory hearths and a central rotating chaft to which
rabble ar.ns are attached. Each hearth has ar. opening thsc allows
-74-
-------
COOLING AIR DISCHARGE
FLOATING DAMPER
SLUDGE INLET
FLUE GASES OUT
RABBLE ARM
AT EACH HEARTH
DRYING ZONE
COMCUSTiON
ZONE
COOLING ZONE
ASH
DISCHARGEjj-
A
CO.\A BUST ION
AIR RETURN
- RABBLE ARM
DRIVE
\
COOLING AIR FAN
Figure 11. MULTIPLE-HEARXII SLUDGE INCINERATOR
75-
-------
sludge to be dropped to the next lower hearth "iny or all of
the hearth stages have oil-or gas-fired burners suppl. addi-
tional heat to the furnace. The rotating central ift and rabble
arms break up the large sludge particles to 'nducc ra; id d com-
plete combustion.
Intermediate hearths provide a high-temperature zone (1CT1
to 1800°F), where, combustion of the fixed carbon takes place. 'i.,e
bottom hearths of the furnace serve as a cooling zone (600°F),
from which the exhaust gases rise to the top of the unit and then
are ducted to a scrubber. A minimum of 50 percent exces: air is
required to burn the sludge properly. The fly ash slurry and ash
from the incinerator are discharged through a hopper and trans-
ported to a landfill or lagoon.
As shown in Figure 11, a separate air system cools the
central shaft. A forced-draft cooling air fan supplies air to the
bottom of the shaft. The cooling air is heated as it passes
through a cooling air discharge duct, separate from the incinera-
tor flue gas stack. I
Particulate emissions are most effectively controlled by a
wet scrubber. Venturi and impingement t>pes of scrubbers have
enabled incinerators to successfully meet emission standards.
In the venturi scrubber, the particulate-laden gas passes
through a duct throat, where high velocities of 60 to 180 m/jec
(200 to 600 fps) are attained at pressure drops oJ 50 to 75 cm
(20 to 30 inches) water gage. Coarst water spray, injected into
the duct throat at the rate of .68 to 1.36 liter? per cubic meter
of gas (five to ten gallons pei 1000 c'Voic feet cf gas) atomizes
and impacts with the particulate.
In impingement towers, the large partijulates are. : ssr.oved
by impingement on wet surfaces and contact with water spray in
an area below the filter bed. The gas containing the regaining
-76-
-------
psrticulates then passes upward through a bed of spheres. These
pariiculates arc subjected to increased velocities in the inter-
stices of the bed, which results in their impingement upon the
surfaces of the spheres.
Continuously running sewage sludge incinerators require
little or no auxiliary fuel to attain complete burnout of the
sludge. Many incinerators are, however, shutdown during the
weekend and must be restarted. To avoid excessive particulate
emissions during start-up, the hearths of these units must be
preheated with gas- or oil-fired heat before sludge is added.
Table 18 estimates the emissions from the multiple-hearth
unit. Insufficient, information was known about the exit gas from
the incinerator to apply the fluid bed particulate emission fac-
tor.
T. CRUDE OIL UNLOADING
As discussed earlier in this report, the crude oil to be pro-
cessed through the refinery will arrive at the Security Marine Ter-
minal by ocean-going tanker. The terminal will have two berths,
each capable of handling an 85,000 dead weight tons (DWT) tanker.
The refinery crude oil requirements will be 175,000 barrels/calendar
di; '.
(175,000 barrels/day)(42 gallons/barrel)(7.12 pounds/galJon)
(2240 pt-unds/dead weight ton)
- 23,363 dead weight tons/day of crude oil
\
Tankers vary in size from old-style units of 28,000 up to 80,000
(DWT) capacity into supertankers of up to 250,000 DWT.
The newer tankers and supertankers contain many air and water
pollution control measures which are not available or. the older tan-
kers. These include segregated ballast and inert gas blanketing. Re
grattably, no existing port facility in the United States is equipped
-77-
-------
Table 18. EMISSIONS FROM SLUDGE INCINERATION (TPY)*
"^--CONTAMINANT
UNI'i ~ ~~-^_
Mu^-'e-Hearth
Uni t
PARTICULATES
nncontrolled
59.3
«
Controlled
1.S
SULFUR DIOXIDES
Uncontrolled
0.6
Controlled
0.5
NITROGEN OXIDES
Uncontrolled
3.6
Controlled
3.0
HYDROCARBONS
Uncontrolled Controlled
1
i
0.9 1 0.6
*For calculations and assumptions used to determine values used for this table, see Appendix 19. .
-------
to handle supertankers. Current prr> {.Lees call for the supertankers
moving out of port and having boryes brought out to unload the prod-
uct. This is a process called Lightering. In light of these facts,
it is unlikely that the Security Marine Terminal will be capable of
handling supertankers. Instead, more conventional types of tankers
will be used. For the purpose of this discussion, assume that 80,000
DWT tankers will call at the terminal 107 tim^s a year. An 80,000
DW1 tanker was selected because tanker registration data did not in-
dicate the existence oc an 85,000 DWT unit.
The emissions from a crude tanker in berth will result from
boiler stack contaminants, venting of the hold during ballasting and
possible hold purging.
After maneuvering and hotelling of the tanker, the crude oil
is pumped out of the hold. Thit, process requires operation of the
ship's boilers. As the liquid leaves the hold, it can be displaced
by either air or combustion gases from the boilers. The use of com-
bustion gases is a form of inerting and is the preferred method.
Inerting reduces the oxygen concentration in the cargo hold to a
value below the flamm.-'bility limit for a hydrocarbon-oxygen mix-
ture. By operating the boilers while in port, sufficient flue gas
can be generated to be cooled and directed into the cargo areas.
However, boiler flue gas contains S0» which is corrosive. A scrubber
is thus installed for SO- removal which can be accomplished effi-
ciently using seawater as the absorbent.
If proper unloading procedures for the cargo are used,
there will be minimal, if any, release through the pressure vents
in the top of the cargo hold. Even in the event of sucn a venting,
the gases vented would consist primarily of the inert gas instead
of hydrocarbons. Until equilibrium is reached in the hold, a
vertical stratification of gases exis :r. in the cargo hold. The
hydrocarbons tend to remain at the bottom next to the liquid inter-
face while the top of the hold at the vent consists of the inert
-79-
-------
gases. If ballasting is performed within several hours of unload-
ing, equilibrium will not have been reached.
The purging process efficiently flushes the empty oil tank
with scrubbed combustion gas in a short time. For one cargo tank
a complete atmospheric change can be accomplished in approximately
80 minutes, with a complete tanker requiring between 8 and 12 hours.
A normal purging operation encompasses three complete atmosphere
changes and hence requires about four hours per tank.
Once a tanker is in operation and the cargo hold contains an
inert gas and hydrocarbon mixture, there are two conditions sug-
gesting a need for purging the hold: (1) to restore a satisfac-
tory inert atmosphere in the hold; (2) to reduce the hydrocar-
bon concentration before personnel enter the cargo compartment
for inspection or maintenance; (3) to gas free after tank cleaning
prior to entering drydock; and (4) to gas free tank cleaning if a
special cargo is to be back loaded.
Under all conditions, purging results in a substantial emis-
sion of hydrocarbons. Estimates of the maximum emissions from a
purge are approximately 60,000 pounds/visit for a 100 percent
purge of all the holds of a 120,000 DWT tanker. Estimates are not
provided for an 80,000 DWT tanker. Available information states
that no United States tanker of this weight class is equipped with
an inerting systen and therefore are incapable of being purged.
There should be no reason to purge during berth in the harbor un-
less in an extreme emergency and every effort should be made to
ensure that the practice is not used. Due to the infrequency of
the operation, the effects of purge at berth vail not be shown.
Ballast is added to the ship during or after discharge of the
cargo to assure safe operation of the tanker. It is common prac-
tice to ballast tankers to approximately 35 percent cf their dead
weight tonnage for open sea travel. This is usually done i~> two
stages with 10 percent to 20 percent ballasted in berth and the re-
-80-
-------
Tnainder when_the tanker is undervav. However, in times of severe
winds or high seas, all ballasting may be completed at berth.
New ships contain segregated ballast compartments to avoid con-
tamination of the water. In order to reduce the amount of con-
tamination which can result from unsegregated ballasting systems
and to reduce the amount of onshore ballast storage necessary, a
technique called "load on top" was developed a few years back.
A conversation with Mr. Jim Dailey of Standard Oil of California
in El Segundo described the procedure as follows. By the time
crude oil arrives in port, the hold has been stratified into ba-
sically three layers. On top is the crude oil, below which is the
bottom sediment and water (BS&W) and on the bottom is the water.
The crude oil is pumped out of the tanker hold through a crude oil
exit near the tank hold bottom until the BS&W level is reached.
This level is noticeable due to a color change. The crude oil
pumping stops and a bottoms pump draws off the water from the hold
bottom until the BS&W "color" again appears, at which time all
pumping from that hold stops. After each cf the tanker holds have
been emptied in this manner, the BS&W from each hold is concen-
trated in one hold and the necessary extra ballast is added. Be-
fore reloading of crude oil takes place, all ballast water which
by now has again stratified in the bottom of the hold is removed
except for the BS&W. Crude oil is then "loaded on top" of the
BS&W through the crude oil exit. A degree of mixing takes place
causing a portion of the BS&W to remain in suspension during trans-
it and leave with the oil. The crude oil tankage is equipped with
agitators to keep it in suspension. The BS&W is then-removed in
the crude desalter.
Another source of emissions from tankers is venting. Unlike
purging which is a purposeful voiding of che vapors in the hold,
venting (or breathing) is an unavoidable loss due to pressure
fluctuations, and in the case of a non-inerting system can result
-81-
-------
in hydrocarbon losses. Venting is not a problem on an inerting
ship. The pressure vents en such a vessel are set at high levels
to ensure that the conbustion gases do not escape. For the pur-
poses of this discussion, consideration is only given to emissions
resulting from unloading of crude oil. Tne amount _,f material
emitted from the hold would be a function of wind, temperature,
pressure fluctuations, hydrocarbon vapor buildup, noveiaent of the
ship and the amount of leakage fron the cargo spaces to the at-
mosphere.
Another minor source of emissions is the refueling o5 the
tanker.
In the report prepared by PES discussing the inpact of che
unloading of Alaskan crude oil at California ports, no problems
were discovered associated with the unloading of sour (greater
than 1 percent) crude.
For the figures in Table 19, the controlled condition was
taken to be a tanker with between 15 and 18 percent segregated
ballast with inert gss blanketing which did not practice purg-
ing and does not vent hydrocarbons. The uncontrolled case will
include a non-segregated ballast requiring the t iking on of 20
percent of ballast in port and no inert gas blanketing resulting
in the venting of hold gas containing 5 percent hydrocarbon vapors.
Emission factors used for this study were generated by PES in a
report Air Quality Analysis of the Unloading el Alaskan Crude Oil
at California Ports, R. J. Bryan and others for EPA, Office of
Air Quality Planning and Standards, Pesearch Triangle Park,
North Carolina.
U. PRODUCT LOADING
HREC is equally as unsure of their buyers as they are of
their sources of crude oil. KREC is reluctant to sign a contract
-82-
-------
Table 19. EMISSIONS DUE TO CRUDE OIL UNLOADING
SHORT TONS PER YEAR (TPY)*
^~ .^CONTAMINANT
OPERATION _.
Holler Operation
Refueling
Venting
Ballasting
TOTAL
PARTICULATES
Uncontrolled
28.0
23.0
Controlled
25.5
25.5
SULFUR DIOXIDES
Uncontrolled
291.6
291.6
Controlled
265.8
265.8
NITROGEN OXIDES
Uncontrolled
126.5
J26.'i
Controlled
115.3
~ ~
115.3
[
HYDROCARBONS
Uncontroll ed
3.9
Nog.
53.4
105.8
163. 1
Control 1«
3.:
Me ,; .
-0
o
. ..
i : t.
I
LO
*For calculi-Lions and assumptions u;ied to d^iermin-s values used in this table, see Appendix 20.
**If the inerted tanker practiced 50 percent purging at every visit, the emissions would use to 534.3 tons/year
(see Appendix 20).
Uncontrolled - non-inerting tanker using a non-segregated ballast.
Controlled - tanker containing between 15 ana 18 percent segregated ballast and inert gas blanketing.
-------
guaranteeing delivery ,^f a product commencin" on a certaii, day
until they ;ire certain of a starting date for the refinery. Jev-
eral possibilities exist which are discussed below. The reader
should be advised that except for the I.PG loading rack, all other
discussion of marketing schemes and associated emissions are hy-
pothetical.
1. Liquid Pet.rcleum Gases (Propane and Butane)
LPG gases will be distributed through a tank car loading
rack. This loading rack will be a fully t-^closed system. Such
systems connect thr; vapor return line overhead to the tan.c car.
Filling displaces these vapors back to the storage !^iu?et which
is a high pressure tank preventing venting of the vapors. The
recovered vapors gradually condenre back to the liquid sr.r.te.
The on^y source of emissic J will V°. vapors trapped in tho Bill-
ing and return lines after shutoff. These vapois cannot s& cal-
culated but are thought to represent an insignificant amount.
2. Tank Truck Loading Rack
Wl:ile .:ot specified in the plans, F.REC and FWEC do not deny
the possibility tba a truck loading ra.k may prove necessary.
However, preliminary marketing studipj do not anticipate this re-
quirement to be very high. If a truck loading rack prove? neces-jry
it will be designed with a maximum capacity throughput of less
than 20,000 gallons per day which would exempt it from any anti-
cipated rule restrictions. Therefore, it would operate uncon-
trolled. Vapor 'recovery systems exist for units even this small.
Small bulk plants with similar throughputs utilize the vapor
balance control technique.
3- Pipeline
Several of the products may leave by pipeline. Emissions
from such a source are relatively small, primarily being restricted
to pump seals and valves.
-84-
-------
4. Marine Barge ],oau_|ng,
FWEC and HREC aiv confident that most, if not all, of the
products will be marketed r.'irougli the Security Marine Terminal.
The marine terminal loading piers will ha designed to handle four
barges at any one time. Since tankers will not be in port at
all times, these two berths will also be capable of handling a
product barge. The tankers could be used to transport products
but for this study, it has been assumed that barges will handle
all products.
Water ballast which may be carried by the incoming product
barges will be emptied into onshore treating facilities before tak-
ing on cargo. This ballast water will be pumped into the ballast
water tanks where sufficient time will be provided for oil-water
separation. The separated oil layer will be pumped to the crude
tanks for processing. The water layer will be pumped to the bal-
last water treating system for treatment. The treated ballast
water will be stored in tanks for re-use as ballast water for the
outgoing crude tankers. '
Tha loading of volatile materials into barges can. produce
significant quantities of hydrocarbon emissions, so studies have
been made into the feasibility of controlling these emissions.
The most promising method developed thus far for the volatile
loading operation is submerged fill loading combined with the use
of a vapor recovery system. This technique has been proven in
utilization on truck loading systems, but the physical logistics
of the barge loading situation presents a different set of prob-
lems which must be solved.
In order to use a vapor recovery system for a barge or ship,
modifications must be made to the loading apparatus on the vessel
itself, closed gauging systems must be installed, and vapor re-
covery equipment must be constructed for the on-shore docking fa-
cility. Also, a piping system must be constructed to transfer the
-65-
-------
product frcm on-shove storage to the vessel and transport dis-
placed vapors back to the recovery system.
Rather simple changes must be performed to modify the hatch
on the cargo tank to make it acceptable to the vapor recovery
system. Modifications need not be made to the loading arras them-
selves, assuming that submerged fill loading practices through
internal piping are followed and a good seal is maintained on the
loading hatch.
A large number of systems have been tested and proven for
on-shore vapor recovery systems. The most popular are refrigera-
tion and absorption svstems. Absorbing mediums can be gasoline
or lean oil fractions. Certain refrigeration systems may require
precondensers to remove most of the moisture and dual condenser
systems so that when one condenser is defrosting, the other will
be in service. For nafety reasons, it is normal practice to
build two or more parallel systems for multiple-berth docking fa-
cilities. This type of system could prevent an in-line explo-
sion from propagating through tha vapor recovery system and into
another vessel which was tied in at the dock.
Care should be. exercised in the construction of the piping
systems to place flame arresters in sufficient quantity and to
avoid oil or product spillage through leaks.
The major problems associated vith the use of vapor recovery
systems involve potential safety hazards, large initial construc-
tion costs and the possibility that high vapor collection effi-
ciencies will not be maintained.
Potential safety probleius are always present in a system
handling volatile materials. Any proposed vapor recovery systen
for a marine terminal would have to be approved by the Coast
Guard before being constructed or operated. The main safety
-86-
-------
features would have to include flame arresters, overfill protec-
tion equipment, and a system of spill retention apparatus.
Costs for construe', ion and operation of a narine vapor re-
covery system to handle a 23,000 tons/hour loading capacity were
given in 1976 by an oil rompany spokesman as follows:
On-shore facility construction - $10,000-12,000
Barge modifications - $50,000-100, 000/barge
Operating costs - $2,000.000/year
Other problems with r.he use of marine vapor recovery systems
are those which reduce the total efficiencies of the systems.
For instance, depending on the barge mix which is serviced by the
port, some vessels may not be equipped with the proper gauging
and piping systems for use of the vapor re':;very units and may
emit hydrocarbon vapors uncontrolled. Other methods of loading
which could be utilized on these vessels might include variation
of loading rates and "short" loading practices. Vapor recovery
system efficiencies may also be reduced due to the necessity of
open gauging practices on vessels equipped for vapoi. conrrol.
FWEC estimated the planned vapor recovery system to be 90
percent efficient for the collection of hydrocarbon vapors. This
is a conservative estimate. One of the recovery systems being
examined for possible use is a refrigeration system.
The straight refrigeration system is based on the conden-
sation of gasolir.e vapors by refrigeration at atmospheric pres-
sure. Figure 12 shows the flow scheme of such a system. Vapors
displaced from the tanker cr barge enter a horizontal fin-tubr>
condenser where they are cooled to -100°F and condensed. Because
vapors are treated on demand, no vapor holder is required. Con-
densate is withdrawn from the condenser bottom and the remaining
air, containing only a small amount of hydrocarbon, is vented from
the condenser top. The vented gases emitted are very cold and may
require heating to minimize the explosive potential from vapors
settling in the surrounding are?;. Cooling for the condenser coils
-87-
-------
Condcnwo Liquid
3[bon Drain
I
00
00
A - AIR COOLCO CONDCNSER. HIGH STAGE
D - HIGH STAGE COMPUCSSOR
C -HIGH TEMt-tRATL'ME EVAPOHATOB
«nd LOW TCMI'tnATUnC CONDENSER
O - LOW r.TAGF COMPHt.'^GOH
E - LOW TEMPEHATUME EVAPORATOR
F - BRINE I'DMP
G - COLO BRINE STOI'iAGE RCf.ERVOlR
H - OEmOST UP.INC ini'J CXPAI\)SION CHAMBER
J -DEFROST PUMP
K -COOLANT PUMP
L-VAPOR CONDENSER
M-ELECTRIC WATER CONTROL VALVE
N - POSITIVE OlSPLACEt.'.ENT METERING PUMP
roncoNOCN^r.o HVDROCARDONS
P - FLOAT VALVE
Figure
RKFRIGERATION VAPOR RECOVERY UNIT BY EDWARDS
-------
is supplied by a methyl chloride reservoir. A uwo-stage refrig-
eration unit is used to refrigerate tV.> stored brine solution to
between -iQ5°F and -i:S°F.
The straight refrigeration syste.ia is based on the conden-
sation of gasoline vapors by refrigeration at atmospheric pres-
sure. Figure 12 shows the flow scheme of such a syster.. Vapors
displaced from the terminal enter a horizontal fin-t.ube condenser
where they are cooled to -100"F and condensed. Because vapors
are treated on demand, no vapor holder is required. Conoansate
is withdrawn from Che condenser bottom and the remaining air,
containing only s small amount cf hydrocarbon, is vented from the
condenser top. Cooling for the condenser coils is supplied by a
luf.thyl chloride reservoir. A two-stage refr\geration unit is
u<;ed to refrigerate tbe stored brine solution to between -lOS'F
ar.d -125°F.
The vapor recovery efficiency of rerrigeration systems is
aijain dependent on the hydrocarbon concentration of the inlet
v.apors. Field tests of a unit with a condenser temperature of
-100°F indicate the outlet, hydrocarbon concentration is relatively
fixed by the condenser temperature at 0.6 percent to 2.6 percent
by volume. Typical hydrocarbon recoveries are 93 percent to 97
percent with recoveries reaching 99 percent for saturated inlet
vapors. In spite of these numbers, evaluators used the FWEC
value of 90 percent to allow for deterioration or operating ir-
regularities of the system.
In order to estimate the emissions associated with the mar-
keting of HREC products, several different schemes must be examined.
Investigators decided to compare a complete marine barge loading
scheme with a nixed scheme. Case I is a mixed scheme which con-
sists of a 20,000 gallon/day truck loading rack for gasoline
-89-
-------
which operates uncontrolled, A pipeline distribution system is
assumed to handle 40 percent of each of the products with the
other 60 percent being handled through the marine termini]. Case
II assumes all products leave by marine barge. Emissions based
on these two cases are presented in Tables 20 and 21.
Barges can vary in size a great deal. The very large ocean-
going barges are 150 feet wide and 450 feet long while other small
barges are only 36 fe
-------
Table 20. CASE I - MARKETING SCHEME (TPY)*
" - -__TONTAMINAN1
UNIT ' "-"- __
Tfuck Loading Rack
Pipeline (40%)
Marine Barge Load-
ing (60%)
Tugboat Operations
TOTAL
PARTICIPATES
34.2
34.2
SULFUR DIOXIDES
25.7
25.7
NITROOKN OXIDKS
24.0
24.0
Uncontrolled
45.3
4.2
530.5**
15.4
595.4
HYDROCARBONS
Submerged KIT!
15.0
4.2
530. 5
15.4
565.1
V.ipor Rfi'ovi
15.0
1.1
53.0***
1 5 . i
84.5
*For calculations and assumptions us.-'d to determine values used in this table, see Appendix 22.
**Uses clean barge emission factor for gasoline.
**AValue would be 96.7 TPY if gasoline barges are cleaned at berth.
-------
Table 21. CASE II- MARKETING SCHEME (TPY)*
"^ -^£ONTAMI NANT
UNIT ' __.
Gasoline
Jet A-l Fuel
//2 Fuel Oil
05 Fuel Oil
Tugboat Operations
TOTAL
PAR71UCLATES
57.2
57.2
SULFUR DIOXIDE
42.9
42.9
NITROGEN OXIDES
40.1
40.1
HYDROCARBONS
Clean Barges
333.6
465.5
61.7
27.6
.
25.7
914.1
Vapor Recovery
33.4
46.5
6.2
2.8
25.7
114.7**
ro
I
*For calculations anc! assumptions used to determine values used in this table, see Appendix 21,
**Value would be 187 TPY if gasoline barges are cleaned at berth.
-------
Table 22. EMISSIONS FROM HYDROCARBON STORAGE (TFY)*
TANK NUMBER
1-11
12-24
25-28
29-32
33-34
35-38
39-45
46-47
48-49
50-52
53-55
56-57
58-59
60-62
Subtotal
Miscellaneous
PRODUCT STORED
Crude Oil
Propane LPG
Butane LPB
Gasoline
Jet Fuel
No. 2 Fuel Oil
No. 5 Fuel Oil
N.-phtha
Light Naphtha
Gas Oil
Atmospheric Residuum
Refinery Fuel Oil
Slop Oil
Ballast Water
TYPE OF P.OOF
Flont ing
Pressure Tank
Pressure Tank
Floating
Floating
Fixed
Fixed
Floating
Floating
Fixed
Fixed
Fixed
Fixed
Fixed
LOCATION
SMT
SMT
SMT
SMT
SMT
SMT
SMT
HREC
HREC
HREC
HREC
HREC
HREC
SMT
Tank Farm and Off-Sites
TOTAL
ESTIMATED
EMISSIONS
Uncontrolled** Controlled
17T3
NC
NC
'JO/: 3
474
142.5
92.4
421
1339
34.5
67.0
9.5
27.8
40.9
7406.7
87.5
7497. 1
lhfj.2
NC
NC
106.5
23.0
142.5
92.4
14.5
15.7
3'.. 5
t,7.0
".5 !
27.8
40.9
742.6
36.8
782.3
1
*For calculations and assumptions used to determine the values used in this table, see
Appendix 22.
**Uncontrolled emissions for storage tanks were taken to mean storage in a fixed roof
tank.
NC = Not calculated. Since these are pressurized tanks, the only emissions will be from
the transfer of material and these emissions are calculated through the pump seal.
-------
Table 23. PUMPS, COMPRESSORS AND liv-LINE SPARES
UNIT AND SERVICE
IN-LINE SPARE
CRUDE UNIT
Crude feed pump to desalter
Crude feed pump to heater
Residuum pump
Heavy gas oil pump I
Light gas oil pump|
Kerosene pump I
Heavy naphtha pumpf
Reflux drum to collector pump
Condensed overheads to stabili2ing pump
Gasoline to storage pump
Stabilizer reflux pump
Intermediate reflux pump
BotLoias reflux pump
Water pump #1
Water pump #2
Water punp #3
Overhead non-condensibles compressor
HYDROGEN PLANT
Caustic scrubber circulating pump
Fuel gas compressor
NAPHTHA HYDRODESULFURIZER
Charge pump
Bottoms pump
Overheads pump
Recycle and make-up compressor
DISTILLATE HYDRODESULFURIZER
Naphtha circulation pump]
Charge pump I
Stripper bottoms pump
Stripper reflux pump
Recycle and make-up compressor
RESIDUUM HYDRODESULFURIZER
i V
.Residuum pump from distillation
Residuum feed pump #
Residuum feed pump #2
Residuum feed pump //3
Yes
Yes
Yes
Common Spare
Common Spare
Yes
No
Yes
Yes
No
No
No
; No
No
No
No
No
Yes
Yes
Yes
Yes
Common Spare
Yes
Yes
Yes
Yes
Common Spare
-94-
-------
Table 23. PUMPS, COMPRESSOR A:;D IN-I IKE STARTS
(cont i :K:L-
-------
Table 23. PUMPS, COMPRESSOR AM)
(continued)
IN-LINE SPARES
JJNIT_AND SERVICE
MEROX UNIT (OPTIONAL)
Charge pump
Product pump
AMi;,'E SYSTEMS
1 lean amine pump
4 reflux pumps
SULFUR PLANT
Tail-gas cleanup air compressor
HREC MISCELLANEOUS EQUIPMENT
Refinery fuel gas compressor
Refinery fuel gas punp
Fuel oil pump
Gasoliue product pump #11
Gasoline product pump #2 I
LPG product pump
Jet fuel product pump I
#2 fuel oil product pump)
*5 fuel oil product purp #1
#5 fuel oil product punp it2
EMERGENCY FLARE SYSTEM
Flare knockout drum pump (centrifugal)
WATER COOLING SYSTEM
Water circulation pump
LTDS WATER PUMP
STRIPPED SOUR WATER PUMP
6 TANK FARM PUMPS
62 TANK PUMPS
I IN-LINE SPARE
Unl.nown
Unknown
No
No
Yes
Yes
Yes
Common Spare
Yes
Common Spare
Common Spare
Yes
(reciprocating)
Yes
(PES assumption
Unk nown
No
No
No
-------
SECTION III. EMISSION KSTI MATE -COMPARISONS
In order to obtain the necessary authorities and permits to
construct the HREC refinery from the Commonwealth of Virginia,
Foster Wheeler developed an estimate jf the emissions which could
be expected from the operation of this refinery. Foster Wheeler's
calculations primarily used AP-42 emission factors. PES used AP-42
only when no other ir.ore suitable emission factors could be found.
The AP-42 values are based upon an average of measurements made of
existing units both new and old.
In the following portions of this section each of the four
criteria pollutants which have been quantified are discussed.
A. PARTICIPATES
A summary of all particulate sources ar.d their emissions are
shown in Table 24. For combustion units, the PES values are
slightly higher thin those prepared by FWEC. Values calculated
for the ^lare are considered to be relatively insignificant and
are not included.
FWEC feels that the AP-42 emission factor of 840 pounds/10
barrels oil burned is too high. They argue that modern combus-
tion methods allow more complete combustion and an associated re-
duction in the amount of condensible hydrocarbons vhich would leave
as particulates. FWEC argues that the e-mission factor should there-
fore be more of a reflection of the ash content of the specific
fuel oil burned and PES agrees.
Foster Wheeler stated that the ash conten*. of the desulfur-
ized fuel oil to be burned will be 0.013 percent. FWEC estimated
that the ash content in the oil constituted 11.7 percent of the
total particulates. Based upon this value, FWEC calculated an
-97-
-------
Table 24. PARTICULATE SOURCES (TPY)
UNIT . j
1
STACK ill
Residual HDS Heater
Hydrogen Plant Heater
Sulfur Plant Heater
STACK #2
Naphtha HDS Heater
Distillate HDS Heater
Reformer Heater
Isomerizer Hear.er
STACK #3
Crude Distillation Heater
Utilities (Boilers)
SLUDGE INCINERATOR
CRUDE UNLOADING
PRODUCT LOADING
TOTAL
FWFC
Uncontrolled
NC
NC
HC
NC
NC,
NC
NC
NC
NC
NC
NC
NC
NC
Controlled
69.985
161.252
10.567
54.581
27.189
110.49
19.002
106.671
114.812
10.00*
NC
N'C
684.208
PES
Uncontrolled
77.4
186.2
12.1
i>3.0
3.1. . 4
127.1
21.90
1
116.2
132.6
59.3
28.0
57.2
912.4
Control led
77.4
186.2
12.1
63.0
31.4
127.1
21.90
116.2
132.6
1.8
25.5
57.2
852.4
*Estimated by the Commonwealth of Virginia
NC = Not Calculated
-08-
-------
emission factor of 353.99 pounds/10 barrels. These values ap-
pear in Table 24 under Che F*'EC controlled column for Stacks //I,
//2, and #3.
In investigating the Foster Wheeler value, PES learned that
FWEC had test result data which supported the theory that a cor-
relation existed between Lhe fuel oil ash content and total par-
ticulate emissions. Total particulate emissions consist of nn-
burned carbon and inorganic materials (ash) which existed in the
oil before combustion. Neasureroent or stack gases from combus-
tion units demonstrated to Foster Wheeler that the contribution
of ash to the total particulate emission rate varied from 10 to
50 percent, depending on the ash content of the oil. Higher ash
contents cause the ash contributing percentage to increase. The
fuel gas which HREC anticipates using has an ash content which when
burned contributes only 10 percent to the total particulate rate.*
Foster Wheeler's use of the value, 11.7 percent was based
upon the rounding-off of calculated totals and working backward.
PES decided to use the value of 10 percent. Use of the 10 percent
value will result in calculations indicating the highest emission
values which can be expected under the indicated conditions. PES
used the value FWEC estimated for the net heating value of hydrodesul-
furized fuel oil of 17,500 BTU/pound. This is a lower value than
is normally associated with tj fuel oil and will result in higher
fuel oil usages.
The #5 fuel oil weighs 7.586 pounds/gallon (based upon an
API gravity of 24.2) and contains 0.013 percent ash.
106
.,- enn , , = 57.14 pounds of fuel oil necessary to produce
(17,500 BTU/pound) ,
10 BTU
*The Foster Wheeler data have been evaluated by the Commonwealth
of Virginia and EPA, Research Triangle Park, and will be incor-
porated into new emission factors.
-99- '
-------
(57.14 pounds of oil/10 BTU)(0.00013 pounds of ash per pourd of oil)
= 0.0074 pounds of ash/106 BTU
Since the ash represents 10 percent of the total parciculate emis-
sions:
Total particulate emission factor =
(O.Q074 pounds of ash/106 BTU) n _, , . 6
^ L Q-£Q = 0.074 pounds/10 BTU
The Virginia State Air Pollution Control Board estimated the
particulate emissions from the sludge incinerator as 10 tons/year.
The PES values of 59 tons/year uncontrolled and 2 tons/year controlled
were obtained by applying the AP-42 emission factor for a multi-
ple-hearth incinerator to the proposed amounts of sludge generated.
Comparable calculations for a fluid bed sludge incinerator were
not possible since the process is too new to have had emission
factors developed. Grain loading after a scrubber is expected to
be 0.03 grains/DSCF.
When ocean-going tankers are in port to discharge crude oil,
the ship's boilers will burn a considerable amount of fuel oil
for the purposes of discharging crude and taking on ballast, inert-
ing, hotelling and maneuvering. The uncontrolled value in Table
24 for crude unloading reflects all boiler contaminants exiting
the stack. .The controlled va_je represents the routing of 15
percent of the flue gas into the cargo holds for inerting pur-
poses.
B. SULFUR OXIDES
A summary of all sulfur dioxide sources and their estimated
emissions are shown in Table 25. PES values for combustion sources
are slightly lower than the estimates made by FWEC, even though
Foster Wheeler considered the sulfur content of the refinery fuel
gas to be negligible and used the emission factor of 6720 (% sul-
fur) pounds/10 barrels of oil burned appearing in AP-42.
-100-
-------
Table 25. Sl'LFUR OXIDES SOURCES CITY)
UNIT
STACK If I
Residual HDS Heater
Hydrogen Plant Heater
Sulfur Plant Heater
Sulfur Plant Tail Gas
STACK #2
Naphtha HDS Heater
Distillate HDS Heater
Reformer Heater
Isotnerizer Heater
STACK #3
Crude Distillation Heater
Utilities (Boilers)
FLARE
CRUDE UNLOADING
PRODUCT LOADING
TOTAL
FWtC
Uncontrolled
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
Controlled
355.7
856.3
56.1
624.6
259.8
144.4
584.9
100.9
566.5
609.7
35.0***
NC
NC
NC
PES
Uncontrolled (Controlled
343.9
828.1
53.8
11,480*
280.2
139.3
564.4
95.6
547.2
589.2
35.5
291.6
42.9
15,291.0
343.9
828.1
53.8
287.0**
280.2
139.3
564.4
95.6
547.2
589.2
35.5
265.8
42.9
4072.2
*This value represents 96 percent conversion efficiency.
**The figure is based upon t'ne designer's claims of 99.9 percent sulfur con-
version efficiency. Emissions are primarily as COS and CS2 and actual
operating efficiency may be more on the order of 99.8 percent.
***Estimated by the Commonwealth of Virginia.
NC = Not Calculated
-101-
-------
PES calculated values by usinp, mass bal.incos. All
sulfur in fuels was assumed to leave the stack as SO^. Fuel spe-
cifications were obtained from Foster Wheeler's submittal data
tr> the CoMnonwe.alth of Virginia. The values used were: 82.87
percent heat ^uty, 132,753 BTU/gallon, 7.586 pounds/gallon, and
0.3 percent sulfur by weight for oil; and 17.13 percent heat duty.
930 BTU/cubic foot, 0.051A pounds/cubic foot, and 0.025 percent
sulfur by weight for gas.
The Foster Wheeler value of 624.6 tons/year of SO emissions
from the sulfur plant tail gas is based on a conversion efficiency
of sulfur in a Beavon-Stretford process of 99.8 percent. PES used
the designer's efficiency estimate of 99.9 percent but concedes
that the unit built may not achieve this efficiency.
Table 25 shows a minor discrepancy between FWEC and PES for
emissions from a flare. The PES value is slightly higher due
to an estimate made for the burning of refinery fuel gas in the
pilot.
PES calculated an estimation of emissions associated with
the unloading of crude oil. These emissions result from the
operation of the ship's boilers during the unloading process.
The sulfur dioxide emissions from the sludge incinerator
wer'j considered to be insignificant and are not included in Table
25.
C. NITROGEN OXIDES (N02)
A summary of the major sources of nitrogen oxide at the re-
finery appear in Table 26. The table shows the results of Foster
Wheeler calculations and several PES comparisons. The last column
is titled "preferred." After corn-paring Foster Wheeler and PES
values for the same unit, a decision was made of which value best
represented the emissions from that unit. That value appears in
the "preferred" column.
-102-
-------
Table 26. NITROGEN OXIDE (AS N02) SOURCES (TPY)
o
V
UNIT
STACK //I
Residuum HDS Heater
Hydrogen Plant Heater
Sulfur Plant Heater
STACK #2
Naphtha HDS Heater
Distillate HDS Boater
Reformer Header
Isomerizer Heater
STACK //3
Crude Distillation Heater
SUBTOTAL
UTILITIES (BOILERS)
CRUDE UNLOADING
PRODUCT LOADING
TOTALS
FWEC
Uncontrolled
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
Controlled
i
3868.5**
567.0
NC
NC
4453.5
PES
Radian*
563.4
1355.4
88. 4
460.4
228.4
925.2
159.6
896.3
4677.1
965.1
NC
NC
5642.2
AP-40
580.1
1465.7
62.9
446.4
192.0
961.9
125.3
941.1
4775.4
986.9
NC
NC
5762.3
Other
NC
NC
NC
we
NC
NC
NC
NC
NC
115.3
40.1
Pretcrred
563.4
1355.4
88.4
459.0
228.4
925.2
]59.''
896.3
4675.7
567.0
115.3
40.1
5398.1
*The Radian factors used are the same as the factors which appear in AP-42.
**Total for all process heaters.
NC - Not Calculated.
-------
Foster Wheeler did not calculate the emissions fron each
process heater. They stated that it was possible to obtain 0.39
pounds of NO /10 BTU. This factor was applied to the total heat
X
input associated with the operation of the heaters. This results
in the value of 3868.5 tons/year which appears iti Table 26. PES
was not convinced that the FWEC emission factor properly repre-
sented the higher NO emissions which are associated with the burn-
ing of fuel oil. To test the Foster Wheeler factor, PES cal-
culated values for NO using both the Radian (AP-42) emission fac-
tor for each unit according to its fuel usage and the AP-40 emis-
sion factor for NO based upon the heat input for each fuel burned
in each unit. The NO emission totals using these emission factors
were 4677.1 and 4775.^ tons/year respectively. The closeness of
these figures compared to the Foster Wheeler values convinced PES
that the F^'EC numbers were too low. The Radian numbers were chosen
over AP-4? due to the fact that finding factors on a log-log dia-
S'-^m is less accurate than applying an emission factor.
Evidence is available to show that combustion techniques
are on the market that will reduce NO emissions from a steam
x
generator to the ranges claimed by Foster Wheeler. The emission
factors from Radian and AP-40 were developed before these prac-
tices were introduced and do not reflect the improvement. There-
fore, the preferred value was the Foster Wheeler number.
The NO emission rate shown in the table for crude unloading
X
represents the inerting of combustion gases.
NO values for flaring and sludge incineration are relatively
small and are not included in Table 26.
-104-
-------
D. HYDROCARBONS
Table 27 summarizes all hydrocarbon sources and their emis-
sions. The controlled and uncontrolled values for each process
unit appear horizontally. The verticil columi"3 total the poten-
tial and actual emissions for each type of refinery equipment
and process. The FWEC total, including their estimate of marine
terminal losses, is 2750.9 tons/year. The PES total is 2152.6
tons/year.
The primary difference between the Foster Wheeler and PES
values lie in emissions from what are commonly termed fugitive
sources. Foster Wheeler totalled the AP-42 emission factors fov
vapor recovery and flaring, pipeline valves and flanges, vessel
relief valves, pump and compressor seals, and miscellaneous to
arrive at a value of 76 pounds/10 barrels refinery capacity.
They then reasoned that this factor applied to a refinery contain-
ing 18 process units. The proposed HREC refinery will have only
11 of these units, so FWEC modified the factor by multiplying by
11/18.
Foster Wheeler carefully noted in their submittal 'to Virginia
that even though AP-42 emission factors were used, they did not
believe that these factors were correct. PES believes that many
of the values appearing in AP-42 for refinery emissions do not re-
flect modern techniques and equipment. AP-42 bases emission fac-
tors for these sources on refinery throughput. The number of re-
lief valves, flanges, etc., is not a function of refinery through-
put but a function of the number of units in operation at a re-
finery. Gaseous emission factors for losses through valves and
flanges should reflect the number OL valves and flanges. This num-
ber should be more or less proportional to the number of unit pro-
cesses to which the hydrocarbons are subjected. A refinery of
150,000 barrels/day which merely distills crude oil, sweetens it
-105-
-------
Table 27. HYDROCARBON SOURCES (TPY)
SO'. *.CT. ANT) TTP2
lUslduun HOS
Cydrog'Ji Pl«ot
S.'ifur Recovery
Sjpi.tr.-. :-CS
01»tllUt< HOS
R*-;orirt
Mi-rux Plar.t
D.'tut nizet
Napiitl~.d Spll*t«r
Drp^ntanlrer
.~U j.'el I jneou* URLC
Equl pment
Tl.re
Sludge Lnc ider^cer
Har In** TenalMl
i/uilrji
Unload Ing
G-'olln* Tower
SLblCT.U.
Tan>*aga
1-11 Cru4« Oil
012-24 Propane
»25-28 !ut«ne
T9-32 Caiolln*
»33-34 Jet fuel
»35-33 ?2 Fuel 01)
»:9-<.5 P5 Fuel Oil
»*6-47 K»ihf.ha
l.'.*-^ lUht Ncphth*
*W-'>2 C»i Oil
»5J-55 Atrv-jphftlc
Fttr) Iduun
t'A-'i! Reflr-ry fu«l
Oil
;5S-^1 Slop Oil
Idb-til BJlli-'t W«t»r
Ki»rol l«n»;^ij t Tank
in- t ore-sue
SlsroTAJ.
TU7A1 PtS VAU'ES
REFINT.RY IJ>I1IP>1F.NT
CO.".a-JSTIOS PROTESSL'S
Uncontrolled
31.3
75.4
4.6
25.5
12.7
50.7
8.9
50.1
53.7
0.2
0.9
25.7
4.1
343.8
-0-
?43.8
~ontrolled
31.3
75.4
4.6
2J.5
12.7
50.7
8.9
50.1
53.7
0.2
0.6 _
jT?
3 5
342.9
-o-
343.9
RrLKF VALVES I Blt.'.'D Cll.viOII.r.
ncontrollcd
31.9
31.9
31.9
31.9
31.9
31.9
11.9
11.9
-0-
31.9
31.9
J1.9
350.*
-0-
350.9
Controlled UncuntrDllt-tl
-0-
-0-
-0-
-0-
-0-
-0-
-0-
-o-
-0-
-0-
-0-
-0-
-0-
.
-0-
-0-
0.9
0.9
0.9
0.9
0.9
0.9
0.9
0.0
-0-
0.9
0.9
0.9
5.0
!
_ -v.
9.0
Cont rol Jert
-0-
-0-
-0-
-0-
-0-
-0-
-0-
-0-
-0-
-n-
-0-
-0-
-0-
_.
-0-
svii-i i:;i,
I'nconr rol led
6.7
(..7
6.7
*,7
6.7
6.7
6.7
6.7
-0-
6.7
6.7
6.7
n, j
-0-
[pont rol led
-0-
-0-
-0-
-0-
-0-
-0-
-0-
-0-
-n-
-0-
-0-
-0-
-0-
p'j>irs t, .-(.MrRissuhj
L'nt onr r^t led
12.2
1.6
3.7
3.5
1.7
1.7
:3.7
11.4
1.'
0.9
0.9
1.7
10.8
«.5
7'(.0 '
-0- ! 79.0
-o- 1 ;<,4,s
Con* Tf 1 l*d
1.5
-0
2.9
0.8
0.9
0.9
2. 7
6.2
0.1
C. 5
0.5
0.9
6.7
26.6
3»>.»
J6.8
or.iM'L VALVIS ii:.,-i;.'.fs| si«''i:"
l:ir. ntt"!!.^
61.4
M.4
01.4
ei t,
8i.4
fl!.4
81.4" "
ni 4
-0
HI. 4
81.4
fll.4
:::::
6') j.i
F.5
a.i
"ntrollf.l
81.4
«!.«.
01. 'i
tl.'.
SI. 4
PI. 4
81.4
81.4
SI.-
Bi.i
81.4
.
l-n.:<-n»r..llv(!l<...r.tr -Ut
IS'.. 4
19.'. 9
'.',
150.1
1)7. 1
173.1
131.5
IK.',. 7
53.
131.4
1. 7
121.S
121 .8
Ml.h
10.8
0.2
0.1
914.1
i', ). 1
3.0
777,9
f,'«i.4 j l',"4 9
f
-Q-
65.* 1 9'.'J.5 i^Vi
. - , » I
i;n
NC
sc
3048
47'.
142. i
97.4
4?i
13)9
U.5
67. n
9. i
27.8
40 9
,M
""J
U.J
I','.. 8
<*
IC9.B
"..<)
1)1. ')
41 ./
134.7
»')'»,
',.1
Jl V
SI. 9
82. 3
6.;
0.2
0.6
1 i . . :
1. '-
l.T
1" 0
1 J'«..t
i. 2
-(^
-V
in'., j
TH. O
9/ ',
l»-5
15. 7
14. S
h7.0
t. >
n.K
40.9
E:.J_
;«: j
iH'.'JUi J_XI;L~L_
1 ' '~^ ^_^__ ^_^_________ ' -1 *ij^*j__
- Not CalcuUtfl
-------
and sends it to storage will have far fewer valves and flanges
than a 150,000 barrel/day fully integrated refinery which switches
these streams to many different parts of the refinery and many
varied process units such as fluid catalytic cracking, alkyla-
tion, isomerization and asphalting. Yet, application of the AP-42
emission factors to each of these refineries would generate iden-
tical values for both.
After review of the derivation of the AP-42 and Radian emis-
sion factors for control valves end flanges, blind chaining, re-
lief valves, and sampling, and determination that better emission
estimates for these sources were not possible, PES applied the
Radian revised emission factors without modification. The emission
factor for each fugitive source was divided equally among the 10
major units planned and the optional naphtha splitter at t!'e HREC
refinery t;> be consistent with the tabular format of Section II.
Pump and compressor seal emissions can also be categorized
as fugitive sources. Information was provided to PES detailing
each pump and compressor and its intended service. The Air Pol-
lution Engineering Manual, Second Edition (EPA AP-40) contains
emission factors for seals which are based upon the vapor pres-
sure of the material being handled. PES felt these factors were
more representative of emissions than enissJ.on factors of pounds/
day/seal or pounds/refinery throughput which are used in AP-42
and Radian reports.
PES estimates of emissions from the cooling tower appearing
in Table 27 were based on the FWEC view that water cooling will
be necessary only in the three hottest months of the year when
the air becomes too warm to rely solely on air cooling.
Foster Wheeler incorporated the emissions from the process
drains and oil-water separation into the general emission factor.
PES divided the streams into two systems, the Low Total Dissolved
-107-
-------
Solids and High Total Dissolved Solids systems. Throughputs for
each system were known and emissions were estimated for the
effect of operating the diverter.
The procedure used by Foster Wheeler to determine the value
of 176.5 tons/year of hydrocarbons from the marine terminal is
not known. PES divided the emissions into unloading and loading
values. Crude oil unloading potentials include venting, ballast-
ing and the burning of oil in the tanker's boiler. The poten-
tials from these sources appear in the uncontrolled subtotal for
unloading in Table 27. The controlled value represents segregated
ballast and inerting.
For loading of products, PES used the Radian factors for
clean barges. The effi ct of cleaning the barges in berth does not
appear in Table 21. This procedure would result in an increase
in hydrocarbon losses through the vapor recovery system by 72.3
tons/year.
Foster Wheeler calculated storage losses fro~ tanks using
the API equations as they are presented in AP-->2. PES used the
modified equations developed by Radian which use an adjustment
i
factor. The data used in the equations were the values used by
Foster Wheeler. Tho atmospheric residuum tanks (#53-55) are
heated to maintain this highly viscous material at 250°F. PES
had to calculate a vapor pressure for this liquid (see Appendix
21).
Table 27 shows a heading titled miscellaneous tank farms
and off-sites. PES assumed each tank to have one pump and five
valves.
-108-
-------
SECTION IV. SUMMARY OF HACT
The total emissions of each pollutar.C as summarized ay PL'S
in Section III reflect, in general, t'.e application of best ava-li-
able control technology (BACT). It is quite |>ossib]e that these
values nay still oe high for certain units and that utilization of
other techniques or technological breakthroughs between the rime
of this report and construction may make available reductions not
discussed here. One prablen with BACT is chc accessibility of in-
formation and the changing nature of technology. Techniques may
exist but. unless they have appeared in literature or the examiner
has been, informed of their existence, they will not appear in an
evaluation.
The major difference between the Foster Wheeler calculations
and PES" values are from fugitive hydrocarbon emission sources. It
is felt by PES that technological advances in these areas have out-
distanced studies of their improved emission potential and that
the large difference in values only points up the need for further
study in these areas.
The rest of this section briefly summarizes what PES con-
siders to be BACT for each process or type of equipment.
A. COMBUSTION UNITS
1. Particulates
If equipment is carefully designed for good fuel economy, com-
plete combustion can be accomplished. Tills will minimize the
amount of unburned hydrocarbons emitted. Instrumentation to indi-
cate C0_ and/or excess oxygen should be required to monitor the
combustion efficiency. Minimization of unburned hydrocarbons will
result in fuel oil particulates consisting primarily of ash.
No effort has been made to install particulate BACT on the
stacks. While the stacks will have in-stack opacity meters, these
act primarily as a combustion monitoring device. No correlation
-109-
-------
can be made at this time between grain loading and opacity, but
it is possible that such a relationship oiay be developed. Mult .-
clones are currently utilized on stacks b irning residual fuel oil
on the East Coast. Such units, as well as electrostatic precipi-
tators and scrubbers, ATP not shelf items but must be designed to
fit the specific application depending upon what the parameters
dictate.
2. Sulfur Dioxide
The combustion of low sulfur treated oil and gas is -he best
method of controlling SO,,. A residual fuel oil with 0.3 percent
sulfur is one of the cleanest fuels of its type currently produced.
The content of scrubbed refinery gas (0.025 percent) is an accept-
able sulfur value.
3. Nitrogen Oxides
t*£S considers two-stage tombustion and off-stoichiometric
combustion to be equivalent control measures which currently exist
for application to boilers which can reduce NO concentrations LO
less than 200 ppro.
PES agrees with Foster Wheeler that NO control devices for
process heaters are not currently on the market. Studies are
currently underway to produce a process heater two-stage cf ims-
cion system. The increased product tube deterioration associated
with this process may delay its Introduction onto the market.
PES has learned that o:f-stoichiometiic combustiri is theoretical-
ly possible in a multiple burner heater and this avenue should be
studied.
A. Hydrocarbons
Since conder.sible hydrocarbons are considered to be parti-
culate matter, combustion hydrocarbon emissions consist of unburned
-110-
-------
gaseous material. Combustion conti'olt, Lo ensure complete combus-
tion should all but eliminate these emissions.
B. SL'LFl'R RECOVERY TAIL-GAS CLEANUP
PES believes that the Parsons engineered Beavon-Stretford
process or tho SCOT process are the best available sulfur recovery
tail-gas cleanup units. The SCOT process is licensed in the Un5ted
States by Shell Development Company. The process consists of two
steps that involve reducing sulfur compounds from the Claus unit
to H_S, followed by recover)' of the H_S in an amine-type system
that utilizes ^ selective solvent. The H_S rich gas is then re-
cycled through the Claus unit to recover the additional sulfur as
a usable produce. High down-times associated with plugging have
been a problem in the past for thd Pritchard designed system.
Careful study should be made before installation of such a process.
Depending on whether you are reading the designer's or the
installer's (FWEC) efficiency estimates, the collection efficiency
of the Beavon-Stretford process may vary from 99.8 to 99.9 per-
cent. Foster Wheeler is confident they car ensure a 99.8 percent
figure but will not guarantee the higher figure of 99.9 percent.
The difference between these values is approximately 287 tons/
year of SO, .
C. FLARING
The use of a refinary fuel gas system is both an economically
sound process and a procedure which gre.icly reduces flaring emis-
sions. The use of automatic steam injection helps to improve com-
bustion efficiency and reduce opacity problems. A cennnon pro-
cedure which gives added insurance that opacity problems will not re-
sult is the use of continuous television monitoring of flare stacks.
-Ill-
-------
D. SLUDGE INCINERATION
Should sludge incineration prove necessary, either the mul-
tiple-hearth or fluid bed type unit should be used. The multiple-
hearth is the established system and adoption of such a system to cope
with Lhe specified qualities of the HREC sludges should prove less
diffic.uit than with the newer fluid bed system.
The fluid bed system is more desirable from an air pollution
standpoint, but due to the newness of the system, it is difficult
to predict if down-time may be higher, resulting in other problems.
The best control system would be a venturi or impinger
scrubber of about 12 co 16 inches of water pressure drop, assuming
the particle size of the effluent particulate matter is in the
normal expected range.
E. MARINE TERMINAL OPERATIONS
1. Crude Unloading
Desirable procedures for tanker crude oil unloading include
the use of the load on top procedure to reduce on-^shore ballast
requirements, segregated ballasting, and inert (combustion) gas
blanketing of the hold. Purging should not be allowed in-port.
The burning of a better grade of fuel oil in the tanker boilers
would improve combustion contaminants enissions. For this, it is
necessary to have segregated fuel oil compartments.
2. Barge Loading
A vapor recovery system with a collection efficiency of 90
percent is well within the range of current technology. In fact,
values of between 93 and 97 percent can be expected. Careful
consideration of the barge mix should be made. A uniform barge
type should be chosen if possible to ensure that the vapor re-
-112-
-------
covery hose .onnectici will fit the barge hatch connection. If an
irregular mix should exisc, and the connection not be adaptable
to all hatch covers, the effect of the vapor recovery system will
be reduced.
F. STORAGE TANKS
Control technology for storage tanks is spe< ified in the
standards of performance for new stationary sources, 39 FR 9311 -
Requirement for Petroleum Liquid Storage Vessels - March 8, 197A.
The EPA, ESED, is presently reviewing the requited technology to
determine if modifications in the recommended control technology
are necessary.
G. OIL-WATER SEPARATION
The use of a corrugated plate interceptor (CPI) and an air
flotation system should prove equivalent to any API system. The
CPI is growing in popularity due to its smaller area and cost
requirements. A CPI can be placed on individual streams elimin-
I
ating the need for a final separator.
One drawback is that a CPI requires a fixed cover, which in-
curs the greatest explosion potential.
The diverter on the LTDS system should include a floating
roof system consisting of plastic balls or the equivalents to reduce
the vapor space on the liquid surface.
H. REFINERY EQUIPMENT
1. Relief Valves
The use of a rupture disc or ventins ^° the fl^re for any
valve with over a one-and-a-half-inch diameter is the best tech-
-113-
-------
2. Pipeline Valves, C on troj^aj^vgj^and^ Flanges
Emissions from these sources are a function of the position-
ing and proper preventive maintenance, surveillance and house-
keeping. The best emission control technique for these sources
is a well-trained and adequately staffed inspection crew, usually
the operators themselves.
3. Pump and Compressor Seals
a. Rotating Shafts
An oil-film mechanical seal is the best control technology.
It improves upon the simple mechanical seal by filling the avail-
able space with oil to reduce emission possibilities.
b. Reciprocating Shafts
Current evidence indicates that mechanical seals will not
work on a reciprocating shaft. The best procedure for controlling
emissions from these sources is still the use of packing glands
and a good maintenance program. Certain shaftswill allow ti.o use
of a fitted g^and which can vent to the flare. This technique
is relatively new and should be further studied for adaptability
to other reciprocating shafts.
4. Blind Changing
Blind changing is an on-going operation that is expensive
as well as wasteful, and produces a certain amount of hydrocarbons
to the atmosphere during occurrence. The best available control
technology is a permanent quick-change blind such as the Hamer
blind, described in Section II of this report. These blinds are
especially advantageous in pumping manifolds or other places re-
quiring frequent changing. They also result in a minimal loss of
hydrocarbons. The additional cost ca-, be recovered in the re-
-------
duced need for mechanics to change conventional blinds. Operators
already on duty can rapidly change this type of blind.
'5. Sampling
Efficient sample coolers will reduce sampling losses at the
unit site. Proper disposal facilities will reduce sampling losses
ir the lab.
-115-
-------
SECTION V. SUMMARY
The values which PES has calculated for the emissions from
the HREC refinery are based upon the use of the best available
control technology. Several of these control efficiencies re-
quire the operation of equipment at conditions that may not be
possible in actual application at this refinery. Other values
may require expensive adaptation of new techniques to existing
equipment and may prove financially jnrealistic. PES feels
that an appropriate final .statement for this report is a summa-
tion of these sources and other emission estimates which may re-
sult.
A. BEAVON-STRETFORD TAIL-GAS CLEANUP SYSTEM
It is difficult to predict the efficiency of a cleanup sys-
tem without a detailed analysis of the gas stream it must treat.
The expected efficiency of the unit will probably range between
99.8 to 99.95 percent. PES used the value of 99.9 percent which
is the efficiency claimed by the inventor D.K. Beavon. This cor-
responds to emissions of 287 tons/year SO . The actual emis-
sions from the HREC refinery tail gas system may be as high as 574
tons/year (99.8 percent conversion) or as low as 144 tons/year
(99.95 percent conversion efficiency). Another point to con-
sider is that these estimates are sulfur emissions calculated as
SO-. In actuality, most sulfurous emissions from the unit will
be as COS and CS_ and their environmental impact should be judged
on this basis.
B. MARINE TERMINAL
The emissions which PES predicted for the marine terminal
may prove to be lower than the actual. It is impossible to judge
until the actual vessels HREC plans on using are studied. PES
-116-
-------
has stated that the best technique for crudr- oil unloading from
tankers is the use of a ship equipped with a segregated ballast
and combustion gas inerting. In the course of a study PES per-
formed to determine the impact of importing Alaskan crude oil
through Long Beach, California, the consensus of industrial
sources contacted was that no 80,000 DWT tanker was currently
equipped with combustion gas inert ing capabilities. It is pos-
sible to convert a ship to inerting but it is an expensive opera-
tion. If HREC uses 80,000 DWT tankers without inerting systems,
the hydrocarbon losses from venting will cause an increase in re-
finery emissions of 56 tons/year (see Table 19). If HREC chooses
to install inorting systems, the tanker has the capability of
purging, which has a large hydrocarbon emission potential. The
practice of a 50 percent purge of each tanker hole! while in port
would contribute approximately 535 tons/year to the atmosphere.
The value of 89.0 tons/year of hydrocarbons from the barge
terminal vapor recovery systec assumes that all barges loaded
will be able to be connected to vapor return lines. If 10 per-
cent of the barges loading each product are not able to utilize
the vapor recovery system, the emissions could increase to 178.0
tons/year if no other measures are utilized such as "short" load-
ing. The value of 89.0 tons/year also assumes that each barge
will be "clean" before arriving at the refinery. This practice
results in the lowest emissions from the refinery grounds. This
is not the lowest overall emission value possible with the clean-
ing operation. Barges are cleaned (washed of contaminants) en-
route between points to minimize time in port. This is the most
economical method of operation. However, cleaning enroute means
that the hydrocarbons generated during the process escape uncon-
trolled. If barges were required to clean at the HREC terminal,
they could utilize the vapor recovery system. This would cause
an increase in HREC emissions of 72.3 tons/year. However,
traded cff against this in-port emission increase would be the
-117-
-------
elimination of approximately 700 tons/year of hydrocarbons
emitted during cleaning at sea. An important factor to be
considered in this argument is that longer residence time at
the terminal will require the operation of a larger number of
barges.
C. NITROGEN OXIDES FROM PP.OCESS HEATERS
t
PES is convinced that the two-stage combustion process is
not currently applicable to a process heater. A form of NO
conti'ol is available if the process heater is equipped with
multiple, burners. By careful study of each heater's charac-
teristics, it may prove possible to utilize the off-stoichio-
luetric combustion process to reduce NO formation. With this
x
in nind, HREC should be encouraged to purchase multiple burner
units wherever possible. A 25 percent reduction in the formation
of NO from process heaters would reduce NO emissions by over
1000 tonb/year.
-118-
-------
BIBLIOGRAPHY
1. A Program to Invef-igate Various Factors in Refinery Siting.
Radian Corporation, Austin, Texas. Prepared for the Environ-
mental Protection Agency, Washington, D.C. February 15, 1974.
2. A Study of Vapor Control Methods for Gasoline Marketing Opera-
tions, Volume I - Industry Survey and Control Techniques.
Final Report. Radian Corporation, Austin, Texas. Prepared
for the Environmental Protection Agency, Research Triangle
Park, North Carolina, under Conf^dct Number EPA f8-02-1319.
April 1975.
3. Air Pollution Engineering Manual; Second Edition. Publication
Number AP-40. May 1973
4. Air Quality Analysis of the Unloading of Alaskan Crude Oil at
California Ports. Pacific Environmental Services, Inc., Santa
Monica, California. Prepared for the U.S. Environmental Pro-
tection Agency, Office of Air Quality Planning and Standards,
Research Trp.ingle Park, North Carolina under Contract Number
EPA 68-02-1405. July 1976.
5. Beavon, David. Add-On Process Slashes Claus Tailgas Pollution.
Chemical Engineering, p. 71-73, December 13, 1971.
6. Cheremisinoff, Paul N. et. al. Sludge Handling and Disposal:
A Special Report. Pollution Engineering, p. 22-23, January
1976. ,'
7. Compilation of Air Pollutant Emission Factors, Second Edition.
Publication Number AP-42. February 1976.
8. Crude Tanker Pollution Abatement. A Position Paper by Exxon
Corporation, April 1976.
9. Emissions of Oxides of Nitrogen from Stationary Sources in Los
Angeles County, Report No. 2. A Joint District, Federal, State
and Industry Project. September 1960.
10. Evaporation Loss from Fixed-Roof Tanks. API Bulletin 2518.
June 1962.
11. Evaporation Loss from Floating-Roof Tanks. API Bulletin 2517.
February 1962.
12. Field Surveillance and Enforcement Guide for Petroleum Re-
fineries. The Ben Holt Co., Pasadena, California. Prepared
for U.S. Environmental Protection Agency, Office of Alr and
Waste Kanagemiv.it, Research Triangle Park, North Carolina,
under Contract Number EPA 68-02-0645.
-119-
-------
13. Inspection Manual for the Enforcement of New Source Performance
Standards: Sewage Sludge Incinerators. rEDCo-r.nvironmental
Specialists, Inc., Cincinnati, Ohio. Prepared for U.S. Environ-
mental Protection Agency, Division of Stationary Source Enforce-
ment, Washington, D.C. under Contract Number EPA 68-02-1073.
January 1975.
IM. Mallatt, R. C., J. F. Grutsch and H. E. Simons. Incinerator
Sludge and Caustic, Hydrocarbon Processing, p. 321-122, May 1970.
15. 1974 Refining Processes Handbook. Reprinted from the September
1974 issue of Hydrocarbon Processing. Copyright 1974 by Gulf
Publishing Company, Houston, Texas.
16. Nitrogen Oxide Abatement Technology in Japan - 1973. Processes
Research, Inc., Cincinnati, Ohio. Prepared for U.S. Environ-
mental- Protection Agency, Office of Research and Monitoring,
Washington, D.C. under Contract Number EPA 68-02-0242. June
1973.
17. Pielkenroad Corrugated Plate Separator Packs. International
literature distributed by the Pielkenroad Separator Company,
Houston, Texas.
18. Plant Energy Systems. By the editors of Power. McGraw-Hill
Book Company 1967.
/
19. Revision of Evaporative Hydrocarbon Emission Factors. Radian
Corporation, Austin, 1>xas. Prepared for the U.S. Environ-
mental Protection A^ncy, Office of Air Quality Planning and
Standards, Research Triangle Park, North Carolina. June 15,
1976.
20. Samuelsor., G. Scott. The Combustion Aspects of Air Pollution.
Advances in Environmental Scie-nce and Technology. Edited by
J. N. Pitts and R. L. Metcalf. John Wiley and Sons, Inc.
Publishers - Volume 5, p. 219-323, Copyright 1975.
21. Siddiqi, Aziz A., Larry D. Killion, John W. Tenini and James
T. Adans, Jr. Tests Quantify Emissions from Ship Loadings.
Reprinted from Hydrocarbon Processing, April 1976.
-120-
-------
ACKNOWLEDGEMENT
A large number of people provided the authors a greot deal of guid-
ance, technical assistance and constructive criticism. The authors
would like to express their gratitude to these individuals, without
whom this report would not have been possible.
DAVID MARKWORDT - U.S. Environmental Protection Agency, Re-
search Triangle Park, North Carolina
JOHN HOWELL - U. S. Environmental Protection Agency, Region
III, Philadelphia, Pennsylvania
RICHARD KEPPLER - U.S. Environmental Protection Agency, Re-
gion I, Boston, Massachusetts
RICHARD E. RAYFORD - Foster Wheeler Energy Corporation,
Livingston, New Jersey
ANDREW A. KUTLER - Foster Wheeler Energy Corporation, Liv-
ingston, New Jersey
FRANK ISAACSON - Foster Wheeler Energy Corporation, Livings-
ton, New Jersey
ROBERT E. PORTERFIELD - Hampton Roads Energy Company, Ports-
mouth, Virginia
J.W. DAILY - Standard Oil Company of California, Western
Operations, Inc., El Segumlv-, California
L.R. RAY - Standard Oil Company of California, Wester Op-
erations, Inc., El Segundo, California
HARRY OUTFIELD - South Coast Air Pollution Control District,
Los Angeles, California
ROBERT MURRAY - South Coast Air Pollution Control District,
Los Angeles, California
-121-
-------
GENERAL ASSUMPTIONS
All combustion units were assumed to operate at maximum
rated capacity of heat input. All combustion units were also
assumed to burn a cotr.bination of 17.13 percent refinery fuel gas
and 82.87 percent #5 fuel oil simultaneously. At this point of
the design, Foster Wheeler is not certain which units, if any,
will utilize a larger portion of the fuel gas.
The refinery was assumed to operate 24 hours a day, 347
days per year. This value was determined by the following cal-
culation:
2,622, 74.:,000 gallons/year of crude throughout
(Foster Wheeler submiital to Virginia Air Board)
(2,682,742.OOP gallons/year)
(42 gallons/barrel)(18i,000 barrels/stream day)
347 stream days/year
i
All through the appendices reference will be made to emis-
sion factors from the "Radian report." This reference corresponds
to the report titled: Revision of Evaporative Hydrocarbon Fmission
Factors, draft of final report, prepared by Radian Corporation for
EPA Office of Air Quality Planning and Standards, dated June 15,
1976.
-------
APPENDIX 1
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE CRUDE DISTILLATION UNIT
A-l.l
-------
CRUDE HEATER
455 x 10 BTU/hour rate capacity
(455 x ID6 BTU/hour)(0.8287) = 377 x 106 BTU/hour for
(377 x 106 BTU/hour for oil) ,,K,
tnr.755 Bill/gallon) = ^
(455 x 10° BTU/hour)(0.1713) = 78 x 1Q6 BTU/hour for refinery gas
(78 x 10 BTU/hour for refinery gas.) 6
(930 BTU/cubic ~^ = °-°839 X 10 cublc
Parti-iulate Emissions
For oil: (377 x 10 BTU/hour) (.0.07- pounds/10 BTU)* = 27.90 pounds/hour
For gas: use emission factor of 20 pounds/10 cubic feet (AP-42, pa0^
9.1-3)
(20 pounds/10 cubic feet)(O.OS39 x 10 cubic feet/hour) =
1.68 pounds/hour
[(27.90 -i- 1.6S) pounds/hour](24 hours/day)(347 days/year)
(2000 pounds/tor;
123.17 tons/vear
Sulfur Dioxide Emissions
For oil: sulfur content of f5 fuel oil is 0.3 percent by weight
(2840 gallons/hour)(7.586 pounds/gallon)(0.003) =
64.63 pounds/hour of sulfur
Assuming complete oxidation of sulfur to SO
64.63 pounds/ hour of sulfur compounds to 129.26 pounds/
hour of SO
(129.26 pounds/hour of S02)(24 Inurs/day)(347 days/year)
(2000 Bounds/ton)
538.24 tons of SO /year
*Derived by PES on pages 99-100.
A-1.2
-------
For gas: Amount of fuel, gas = 83,900 SCFH
= (83,900 SCF/hour)(0.0514 pounds/SCF)
= 4312.46 pounds/hour
Amount of sulfur in fuel gas
00"")
= (4312.46 pounds/hour) -yr^' weight percent*
= 1.08 pounds/hour
1.08 pounds/hour of sulfur correspond to 2.16 pounds/hour S02
therefore, SO emission =
(2.]6 pounds/hour)(24 hours/day)(347 days/year)
(2000 pounds/ton)
8.98 tons/year
Total SO emissions = 538.24 + 8.98 = 547.22 tons/year
* Z-."^7-..'"" "" .' ' -"_".!.-'_'
Nitrogen Oxides (N02)
3 3
For oil: (69 pounds/10 gallons burned)(2.840 x 10 gallons/hour)
s/t
For gas: 230 pounds/10 cubic feet of gas burned (same as AP-42,
1. Eased on Radian emission factors (same as AP-42, page 9.1-3)
o
(69 pounds/10 gal]
195.96 pounds/hour
230 pounds/^
page 9.1-3)
(230 pounds/106 cubic feet)(0.0839 x 10 cubic feet/hour) =
19.30 pounds/hour
[(195.96 + 19.30 pounds/hour](24 hours/d^y)(3A7 days/year)
(2000 pounds/ton)
896.34 tons/year
*Foster Wheeler value submitted to the Commonwealth of Virginia.
A-1.3
-------
2. AP-40 values - values for heat inputs of oil and gas are
taken directly from tne following graph: .
It. CCO
1.000
103
£
a
§
10
fc«
t 0
II
1 Cl
1
1
/
/ /
n
/
i_
/
_~ +-
-ti-T!
: > i
' 1 i
i
4-f-!-!- 1
I
\
!
i
i/
I
1
^
f
H-
;
j
t
1 i hU
j J
1
\
l rT"r~
/
r
/
V
j
/i
/
i
t r
U !
1 i
t 1-
/
T^° " ~y":
X
j/] .
r
0s 10s
.JJ
1
i
zf"-
1 1
1 J~tr
-f-*- - * 11-" T "i '*-r
' ! ! '
-J
'
\ ,
ZI r
i
L.
-
V
; i y
' 'J~
jr y
y i x
I/
V ' 1
, i T-; j4-« yi-i -j/' t M 1
1 {
-4-1 y*~"*
y^
f
\
f
! !/ i^ ""
vt -y*
1 /
i"/"
i
i
1
^
-v>-
r
A
-i
"n
^
r
i
1 ~r^t(
i
- - «
i
, 1 i ! | j ' ' » '1
j
I
»
1
.
= NO - Xfij C»UUUTfcP. >S «2>
ISPUT IKClUDES CKOn KEIT IK
FUEL HEiT COMliKEO IK TIC
HEITIO CCK3U3TIOK «!3
S 6RHPH if'LIES CNIT 10 COI-
TION PBJCfSSES TtKIKG H«C£ «T
KEtR 5K£ tliOSPHEDE OF (BSOLUTE
ssuaE.
.^_/. _. ; t -
r~
-!--
_^.,
a
*«-
b»
u.
:::
m
i
i
4-i
^
r
+
l
1-
i
10! ' I0a ' . iO* 13
»tt««CE RATE OF HEM INPU1 TO t WIT III > CIVEN Cl»S: CF COiSUSllW CUUIftCNT. |tu hr
A-1.4
-------
For oil: the value for 377 x 10 BTU/hour heat input is 2K) pounds/
hour
For gas: the value for 73 x 10 ETU/hour .;ear input is 16 pounds/
hour
[(210 + It) pounds/hour](24 hours May)(347 days/year)
(2000 pounds/ton) ' = J^il
Hydrocarbon Emissions
For oil: use emission factor from AP--42, page 9.1-3
140 pounds/103 barrels burned - 3.33 pounds/103 gallons burned
(3.33 pounds/10 gallons burned)(2.84 x 103 gallons/hour) =
9.5 pounds/hour
For gas: use emission factor from AP-42, page 9.1-3
30 pounds/106 cubic feet
(30 pounds/106 cubic feet) (0.0839 x 106 cubic feet/hour) "-
2.52 pounds/hour
[(9.5 + 2.52) pounds/hour](24 hours/day)(347 days/year)
(2000 pounds/ton) = l^-05 tons/year
Uncontrolled Emissions
3N
Used emission factor of 11 pounds/10 barrels of refinery capacity
(both Radian and AP-42)
(11 pounds/10 barrels of refinery capacity) _ 1.0 pounds/10 barrels
(11 units/refinery) per unit
o 3
(1.00"pounds/10 barrels)(184 x 10 barrels/day) = 184.00 pounds/day
(184.00 pounds/day)(347 days/year) H 31.92 tons/year
(2000 pounds/ton)
A-1.5
-------
Controlled Emissions
All relief valves from the crude distil lafion unit will vent to
the flare system. Therefore, the emissions will be essentially
zero.
CONTROL VALVES AND FIANCES
Due to the quality of the information available, it was not pos-
sible to accurately assess the number of t jnges and vlaves the
unit will have. Since emissions are a functior of housekeeping,
inspection and maintenance schedules, procedures which cannot be
quantified, it was decided to present the worst possible case
and have uncontrolled and controlled emissions the same.
Emission factor used was 28 pounds/10 barrels refinery capacity
from Radian and AP-42
3 3
(28 pounds/10 barrels of refinery capacity) _ 2.55 pounds/ 10 bar-
(11 units/ref dnery) rels per unit
(2.55 pounds/103 barrels) (184 x 103 barrels/day) = 469.2 pounds/day
, ,
(2000 pounds/ton)
BLIND CHANGING
Uncontrolled Emissions
Use emission factor of 0.3 pounds/10 barrels refinery capacity
(Radian report).
n 2
(0.3 pounds/10 barrels of refinery capacity) = 0.027 pounds/10 bar-
(11 units/refinery) = rels per unit
(0.027 pounds/103 barrels unit capacity)(184 x 10 barrels/day) =
4.97 pounds/day
(4.97 pounds/day)(347 days/year) = 0.86 tons/vgar
(2000 pounds/ton) =
Controlled Emissions
Good refim-.ry procedures such as line flushing, use of "line" blinds
and blind insulation with gate valves will reduce emissions to a neg-
ligible quantity.
A-1.6
-------
SAMPLING
Uncontrol.' ed Emissions
Use emission factor of 2.3 pounds/10 barrels refinery capacity
(Radian report).
3 3
(2.3 pounds/10 bairels of refinery capacity) _ 0.209 pounds/10" bar-
(11 units refinery) ~ rels per unit
3 "}
(0.209 pounds/10 barrels of unit capcity)(184 x 10 barrels/d.i> ) =
38.46 pounds/day
(38.A6 pounds/day)(347 days/year) ,
(2000 pounds/ton)^= 6'67 tons/year
Controlled Emissions
Good refinery practices such as avoidance of excessive sample purg-
ing and flushing sample purge to the sump will reduce emissions to
a negligible quantity.
PUMPS AND COMPRESSORS
Controlled Emissions
The following table lists all pumps and compressors planned for
crude unit. Column one notes the planned service for each piece
of equipment. Column two shows the reid vapor pressure of each
material. Some of these values were provided by Foster Wheeler.
Other estimates were made based upon the API gravity of the noted
stream. Still others, such as reflux, are the best guess. Column
three shows an estimation of emissions based upon the vapor pre-
sure (source: AP-40, page 686).
(15.6 poun;Wday)(347 days/year) _ .-««./--,,.
" LJ /L-T ' "" I x *- * ' -* HJliD / y cot
(2000 pounds/tor.) - *
Uncontrolled Emissions
Used the emission factors of: 5 pounds/day for each pump seal
9 pounds/day for each compressor sea3.
The crude unit operates 14 pumps in hydrocarbon usuage, each with
one seal. The unit also operates one compressor in hydrocarbon
service with one seal.
*Supplied by the Radian report.
A-1.7
-------
CRUDE UNIT PUMPS AND COMPRESSORS
r
UNIT AND SERVICE
Crude feed to desalter
Crude feed to heater
Residuum
Residuum spare*
Heavy gas oil
Light gas oil
Kerosene
Heavy naphtha
Reflux drum to collector
Condensed overheads to stabilizer
Gasoline to storage
Stabilizer reflux
Intermediate reflux
Bottoms reflux
Overhead non-condensible compressor
Three water service pumps
TOTAL
REID VAPOR
PRESSURE
3.5
3.5
0.007
O.OC7
0.02
1.75
1.75
1.75
5.2
7.0
7.0
7.0
1.75
1.75
20.0
0
POUNDS/DAY
0.3
0.3
0.0
0.0
0.0
0.3
0.3
0.3
2.7
2.7
2.7
2.7
0.3
i 0.3
2.7
0.0
15.6
*The residuum spare is included because this pump is in service at
all times. This pump must remain heated tooperating temperature
at all times so that if it should take over service, the residuum
will not stick, a problem which could occur if the pump were 'j.-.ld.
A-1.8
-------
(5 pounds/day-seal)(14 seals/unit) = 70 pounds/day for pumps
(9 pounds/day-seal)(1 seal/unit) = 9 pounds/day for compressor
[(70+ 9) pounds/davl(347 days/year) .
j^ -,-rr 7^T "\ ' " '' 13.7 tons/vear
(20uO pounds/ton) r-^z^^-^
A-1.9
-------
APPENDIX 2
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE HYDROGEN PLANT
A-2.1
-------
HYDROGEN HEATER
688 x 10 BTU/hour ra^ed capacity
(688 x 106 BTU/hour)(0.8287) = 570 x 106 BTU/hour for oil
(570 x 10 BTU/hour) .
(132,753 BTU/gallon) = A294 gallons/hour of_oil
(688 x 106 BTU/hour)(0.1713) = 118 x ]Q6 BTU/hour for refinery gas
(118 x 10 BTU/hour) n m i n6 u-, * /,
(930 BTU/cubic foot) = ^-127 x 10 cubic feet/hour
Particulate Emissions
For oil: (570 x 10b BTU/hour)(0.074 pounds/106 BTU)* =
42.18 pounds/hour
For gas: use emission factor of 20 pounds/10 cubic feet
(20 pounds/10 cubic feet)'0.127 x 10 cubic feet/hour)
2.54 pounds/hour
[(42.18 + 2.54) pounds/hour 1_(24 hours/day) (3A7 d.-;ys/year) =
(2000 pounds/ton)
186.21 tons/year
Sulfur Dioxide Emissions
For oil: sulfur content of #5 fuel oil is 0.3 percent by weight
(4294 gallons/hour)(7.586 pounds/gallon)(0.003) =
97.72 pounds/hour of sulfur
97.72 pounds/hour of sulfur corresponds to 195.44pounds/
hour of S02
(195.44pounds/hour of S02>(24 hours/day)(347 days/year) =
(2000 pounds/ton)
813.81 tons/year
For gas: Amount of sulfur gas = 127,000 SCF/hour
= (127,000 SCF/hour)(.0514 pounds/SCF)
= 6527.8 pounds/hour
*Derived by PES on pages 99-100.
A-2.2
-------
Amount of sulfur in fuul gas
^ (65.27. P. pounds/hour) (0.025/100)*
= ">nunds/hour
pounds/hour of sulfur correspond to 3.26 pounds/hour SO.
therefore, S00 emission =
(3.26 pounds/hour j (24 hocrs/dav) (3A7 days/year)
(2000" --- ' - 13'57
Total PO emission = 813.81 + 13.57 = 827. 38 tons /year
Nitrogen Oxides (N02>
1. Based on Radian emission factors
For oil: (69 pounds/103 gallons burned) (4. 294 x 10 gallons/hour) =
296.29 pounds/hour
For gas: (230 pounds/106 cubic feet) (0.127 x 106 cubic feet/hour) =
29.21 pounds/hour
[(296.29+ 29.21) pounds/hour] (24 hours/day) (347 days/year) =
(2000 pounds/ ton)
1355.38 tons/year
2. Based on AP-40 values from the chart in Section A-l
For oil: the value for 570 x 10 BTU/hour heat input is 330 pounds/
hour
For gas; the value for 118 x 10 BTU/hour heat Input is 22 pounds/
hour
[(330 + 22) pounds/hour 1(24 hours/day)(347 days/year) _ Ufi5 ?3 tons/
(2000 pounds/ton) -
Hydrocarbon Emissions
For oil: use emission factor of 3.33 pounds/10 gallons burned
(3.33 pounds/103 gallons burned)(A.294 x 103 gallons/hour) =
14.3 pounds/hour
*Weight percent estimated by Foster Wheeler.
A-2.3
-------
For gas: use emission :.uctnr of 30 pounds/10 cubic feet burned
(30 pounds/10 cubic feet) (0.127 x 106 cubic f-jet/hour) =
3.81 pounds/hour
[(14.3 + 3.81) pounds/hour] (24 hours/day) (347 days/vear) -,? -, i
-" .-_,-,,. j; r -Li-i 2~^=i- = 75.41 tons/year
(2000 pounds/ton)
RELIEF VALVES
Uncontrolled Emissions
Use emission factor developed in Appendix 1 of 1.0 pounds/10 barrels
per unit.
3 3
(1.0 pounds/10 barrels)(184 x 10 barrels/day) = 184.0 pounds/day
(184.0 pounds/day)(347 days/year) , tons/year
(2000 pounds/ton) ~J
Controlled Emissions
All relief valves from the hydrogen plant will be routed to the flare
system. Therefore, the emissions will be essentially zero.
PUMPS ANP COMPRESSORS
The hydrogen plant anticipates having only one compressor servicing
hydrocarbon gas (methane). All pumps will be in non-hydrocarbon
service and are not mentioned. All emissions of hydrogen and
oethane from seals will be vented to the flare so the actual emis-
sions will be essertially zero.
Emissions from the one compressor seal have a potential of 9 pounds/
day for each seal.
(9 pounds/day-seal)(1 seal/unit) = 9 pounds/day for the unit
(9 pounds/day)(347 days/year) = tons/year
(2000 pounds/ton) -= J
A-2. 4
-------
BLIND CHANGING
Uncontrolled Emissions
Use emission factor 0.027 pounds/unit from Appendix 1.
CO. 027 pounds/10 barrels per unit) (184 x 10 barrel b/Jay) (347 dr. ys/year)
(2000 pounds/con)
0.86 tons/year
Controlled Emissions
Good refinery procedures such as line flushing, use of "line" and blind
insulation with gate valves will reduce '^missions to a negligible quan-
tity.
\
CONTROL VALVES AND FLANGES
Controlled and uncontrolled emissions are the sane (see Appendix 1).
Use the emission factor of 2.55 pounds/10-^ barrels per unit.
o -i
(2.55 pounds/10 barrels)(184 x 10 barrels/day)(347 days/year)
(2000 pounds/ton)
81.41 tons/year
SAMPLING
Uncontrolled Emissions
3
Use emission factor of 0.209 pounds/10 barrels
(0.209 pounds/103 barrels)(184 x 1Q3 barrels/day)(347 days/year)
(2000 pounds/ton)
6.67 tons/year
Controlled Emissions
Good refinery practices such as avoidance of excessive sample purg-
ing and flushing sample purge to the sump will reduce emissions to
a negligible quantity
A-2.5
-------
APPENDIX 3
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE NAPHTHA HYDRODESULFURIZER
A-3.1
-------
NAPHTHAJHDS HEATER
233 x 10 BTU/hour rated capacity
(233 x 106 BTU/hour) (0.3287) = 193 x 1C6 Bill/he ir for oil
(193 x 106 BTU/hour) ,,., ., ,, , . n
^=^- T^rr, r; \ = 14:>4 gallons/ hour of oil
(132,755 BT'J/gallou) § .
(213 y. 106 BTU/hour) (0.1713) = 40 x 10^ BTU/hour for refinery gas
(40 x _-> BTU/hour) 6
(530BTU/:ubic foot) = °-043 X 10 Cubtc
Particulatc Emissions
For oil; (193 x 106 BTU/hour) (0.074 pounds/106 BTU)* = 14.23 pounds/hour
For gas: (20 pounds/10 cubic feet) (0.043 x 10 cubic feet/hour) =
0.86 pounds/hour
[(14.28+0.86) pounds/hour] (24 hours/day) (34/ days/year)
(2000 pounds/ton)
63-04 tons/vear
Sulfur Dioxide Emissions
For oil: sulfur content of #5 fuel oil is 0.3 percent by weight
(1454 gallons/hour of oil)(7.586 pounds/gaJIon)(O.P'>3) -
33.09 pound?/hour of sulfur
33.09 pounds/hour of sulfur corresponds to 66.18 pounds/
hour of SC>2 (assuming complete oxidation of sulfur to SC>2)
(66.18 pounds/hour of SOj) (24 hours/day)(347 days/year) _
(2000 pounds/ton)
275.57 tons/vear
Fcr gas: Amount of sulfur = 43,000 SCF/hour x 0.0514 pounds/SCF x
(0.025/100)**
=0.55 pounds/hour
0.55 pounds/hour sulfur correspond to 1.1 pounds/hour SO?
*l>erived by PCS on pages 99-100.
**Weight percent estimated by Foster Wheeler.
A-3.2
-------
emission =
(1.1 pounds /hcu ir) (24 hours/ d ay_) ( 347 days /year)
(YOOO pounds/'ton) A'6 tons/year
Total SO emission -- 27S.57+ 4.6= 230.17 tons/year
Nitrogen Oxides
1. Based on Radian emission factors
3 3
For oil: (69 pounds/10 gallons burned) (1 . 459 > 10 gallons/hour) -
100.67 pounds/hour
For gas: (230 pounds/10 cubic feet) (0.043 x : ri° cubic feet/hour) =
9.89 pounds/hour
[(100.67 + 9.89) pounds/hour] (24 hours/day) (347 days/year) ,,,-, ,
- r^nnr. -- j 77 - C - - " - = 460.37 tons/year
(2000 pounds/ton) ^
2. Based on AP-40 values from the chart in Section A-l
For oil: the value for 193 x 10 BTU/hour heat input is 100 pounds/
hour
For gas: the value for 40 x 10 BTU/hour heat input is 7.2 pounds/
hour
[(100 + 7.2) pounds/hour] (24 hours/day) (347 days/year) _
(2000 pounds/ton)
^46.38 tons/year
Hydrocarbon Emissions
For oil: (3.33 pounds/10 gallons burned) (1.454 gallons/hour) =
4.84 pounds/hour
For gas: (30 pounds /106 cubic feet) (0.043 x 10 cubic feet/hour) =
1.29 pounds /hour
.84 + 1.29) pouPds/hou-](24 hours/day) (347 days/year) =
(2000 pounds /ton)
25.53 tons/year
A-3.3
-------
RELIEF VALVES
Uncontrolled Emissions
Use emission factor developed in Appendix 1 of 1.0 pounds/103 barrels
per unit.
3 3
(1.0 pounds/10 barre]_sj_Qjj4_jc__l_p__jbarrpj_s/_d_ay ) (347 -day s /y earj)
(2000 pounds/ton)
31.92 tons/year
J£p v;-.~1 led Emissions
It is assumed that all relief valvas from the naphtha hydrodesulfurizcr
will vent to the flare system. Therefore, t.ie emissions will be
essentially zero.
CONTROL VALVES AND FLANGES
Controlled and uncontrolled emissions are the same (Appendix 1).
;)(184 x 103 barre
(2000 pounds/ton)
(2.55 pounds/103 barrels)(184 x 1Q3 barrels/day)(347days/year)
81.41 tons/year
BLIND CHANGING
Uncontrolled Emissions
e]
(2000 pounds/ton)
0.027 pounds/103 barrels)(ISA x 10 barrels/day)(347 days/year)
0.86
Controlled Emissions
n.ood refinerv procedures such as line flushing, use of "line" blinds,
and blind iHSu3^t3on with gate valves will reduce emissions to a
negligible quantity.
:' MPLING
Uncontrolled Emissions
1 3
(0.209 pounds/10 barrels)(184 x 10 barrels/day) (347 days/year) =
(2000 pounds/ton)
6.67 tons/year
A-3.4
-------
Controlled Emissions
Good refinery practices such as avoidance of excessive sampling purging
and flushing sample purge to the sump will reduce emissions to a
negligible quantity.
PUMPS AND COMPRESSORS
Controlled Emissions
For explanation of the follov.\ug table, see Section A-l.
NAPHTHA HYDRODESULFURIZER PUMPS AND COMPRESSORS
UNIT AND SERVICE
Charge Pump
Bottoms Pump
Overheads Pump
Recycle & Make-Up Compressors (2)*
TOTAL
REID VAPOR
PRESSURE
10
7
26
Unknown
POUNDS /DAY
2.7
2.7
11.1
16.5
*Two compressors consist of two units, each with 60% capacity.
In normal operation, both would be in service at reduced load.
Gases consist of methane (3%) and hydrogen (97%). For the
purpose of this report have considered it to be only hydrogen.
(16.5 pounds/day) (347 days/year)
(2000 pounds/ton)
2.86 tons/year
Uncontrolled Emissions
Use the emission factor of 5 pounds/day for each pump 3eal except the
overheads pump which is considered to be 11.1 pounds/hour
(5 pounds/day-seal) (2 seals) = 10 pounds/day
(11.1 pounds/day-seal) (1 seal) = 11.1 pounds/day
A-3.5
-------
J(10 + 11.1) pounds/day](347 days/year) = JLji6_tgps/yeag
(2000 pounds/ton)
Since the compressors handle hydrogen, the hydrocarbon potential is
zero.
-------
APPENDIX 4
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE DISTILLATE HYDROPESULFURIZER
A-4.1
-------
DISTILLATE EPS HEATER
116 x 10 BTU/hour rated capacity
(116 x 106 BTU/hour) (0.8287) = 96 x 10° BTU/hour for oil
(96 x 106 BTU/hour) _^
(132,755 =
(116 x 10 BTU/hour) (0.1713) = 20 x 106 BTU/hour for refinery gas
(20 x 106 BTU/hour) ,, _.,., 106 . . .. ,,
(930 BTU/cubic foot) = 0-0215 x 10 cubic feet/hour
Particulate Emissions
For oil: (96 x 10 BTU/hour)(0.074 pounds/10 BTU)*= 7.10 pounds/hour
For gas: (20 pounds/10 cubic feet)(0.0215 x 10 cubic feet/hour) =
0.43 pounds/hour
1(7.10 + 0.43) pounds/hour](24 hours/day)(347 days/year) _
(2000 pounds/ton) " -31-33 tons/year
Sulfur Dioxide Emissions
For oil: sulfur content of //5 fuel oil is 0.3 percent by weight
(723 gallons/hour)(7.586 pounds/gallon)(0.003) -
16.45 pounds/hour
16.45 pounds/hour of sulfur corresponds to 32.90 pounds/hour
of S02
(32.90 pounds of S02/hcur)(24 hours/day)(347 days/year)
"(2000 pounds/ten)
137.00 tons/year
For gas: Amount of sulfur = (21,500 SCF/hour)(0.0514 pounds/SCF)
(0.025/100)**
= 0.28 pounds/hour
0.28 pounds/hour of sulfur correspond to 0.55 pounds/hour SO-
*Derived by PES on pages 99-100.
**Weight percent estimated by Foster Wheeler.
A-4.2
-------
S0? emission =
(0.55 pounds/hour) L?!_-_-.^_-.Lj_=^J!_x~ .. -;-, , -Q,,
(2000 pounds/ton) ~ ""= 2'3 tons/year
Total SO emission = 137.00 + 2.3 = .j3Jv.J_ipns/year
Nitrogen Oxides (NOo)
1. Based on Radian emission factors.
3 3
For oil: (69 pounds/10 gallons burned)(o.723 x 10 gallons/hour) =
49.89 pounds/hour
For gas: (230 pour.ds/10 cubic feet)(0.0215 x 10 cubic feet/nour) =
4.95 pounds/hour
[(A9.89 + 4.95) pounds/hour](24 hours/day)(347 days/year) =
(2000 pounds/ton)
228.35 tons/vear
2. Based on AP-40 values for the chart in Section A-l.
For oil: the value for 96 x 10 BTU/hour heat input is 43 pounds/hour
For gas: the value for 20 x 10 BTU/hour heat input is 3.1 pounds/hour
JJA3 + 3.1) pounds/hour](24 hours/day)(347 days/year) ^ m % tons/vear
(2000 pounds/ton) ' - y
Hydrocarbon Emissions
For oil: (3.33 pounds/10 gallons burned)(0.723 x 10 gallons/hour) =
2.41 pounds/hour
For gas: (30 pounds/106 cubic feet)(0.0215 x 10 cubic feet/hour) =
0.65 pounds/hour
[(2.41 + 0.65) pounds/hour](24 hours/day)(347 days/year) = ^^ tons/year
(2000 pounds/ton) - ' ' -
A-4.3
-------
Uncontrolled Emijsionjs
3 3
(1.0 pounds/10 barrels)(184 x 10 barrels/day)(347 days/year^
(2000 pounds/ton)
31.92 tons/y-;ar
Controlled Emissions
All relief valves from the distillate hydrodesulfurizer will be
equipped with a rupture disk or vent to the flare system. In
either case, the emissions will be essentially zero.
CONTROL VALVES AND FLANGES
Controlled and uncontrolled emissions are the same (see Appendix 1).
ils) (184 x 103 bar]
(2000 pounds/ton)
(2.55 pounds/103 barrels)(184 x 1Q3 barrels/day)(347 days/year) _
81.41 tnn s /year
BLIND CHANGING
Uncontrolled Emissions
(U.027 poands/103 barrels)(184 x 1Q3 barrels/day)(347 days/year) =
(2000 pounds/ton)
O.S6 tons/year
Controlled Emissions
Good Tfifir.eey procedures such as line flushing, U3e of "line blinds,
and blind ins-jlacicn with gate valves will reduce emission to a neg-
ligible quantity
SAMPLING
Uncontrolled Emissions
-. }
JO. 2Qj pounds/10 bp;rr*ls)_(] 84 _x_JLC_^_barrels/day) (347 days/year)
(2000 pour.QS/ton)
6.67 tons/year
-------
Controlled Emissions
Good refinery practices such as avoidance o£ excessive sampling,
purging and flushing sample purge to the sump will reduce emis-
sions to a negligible quantity.
PUMPS AND COMPRESSORS
Controlled Emissions
For a general explanation of the table on page A-4.6, see Section A-l.
The distillate HDS unit is a block process handling each of the
three materials separately. From the refinery block diagram, the
following average throughputs are known:
Naphtha - 21,578 BFCD
Kerosene - 17,535 BPCD
Gas Oil - 17,745 BPCD
56,858 BPCD
Each of the materials are handled the following percentage of the
time:
Naphtha - 38%
Kerosene - 30.8%
Gas Oil - 31.2%
I
For naphtha service, the emission total will be:
10.8 pounds/day)(347 days/year)(0.38) = tons/vear
(2000 pounds/ton) ' '
For kerosene service, the emission total will be:
(1.2 pounds/day)(347 days/year)(0.308) = 0.06 tons/year
(2000 pounds/ton) *
For gas oil service, the emission total will be zero
Uncontrolled Emissions
Use the emission factor of 5 pounds/day for each pump seal.
(5 pounds/day-seal)(A seals) = 20 pouads/day
A-4.5
-------
DISTILLATE HYDRODESULFURIZEt1. POMPS AND COMPRESSORS
UNIT ANb SERVICE
Charge Pump
Naphtha Circulation Pump
Stripper Boltems Pump
Stripper Reflux Pump
Recycle & Make-up Compressors (2)*
i _
REID VAPOR PRESSURE
Naphtha
5.2
5.2
5.2
5.2
Unknown
Kcroso.ne
1.75
1.75
1.75
1.75
Unknown
Gas Oil
0.07
0.07
0.07
0.07
Unknown
POUNDS/ HAY
Naphtha
2.7
2.7
2.7
2.7
0.0
Kerosene
0.3
0.3
0.3
0-. 3
0.0
Gss Oil
0
0
0
0
0
*The compressors consist of two separate units, each with 60 percent capacity. In normal opera-
tion, both would be at reduced load. Cases consist of methane (3 percent) snd hydrogen (97 per-
cent). For the purpose of this report, have considered the compressors to handle cn.ly hydrogen.
-------
i20_r_qunds/day) (347 days/year), M 3^? tons/yf_ar
(2000 pounds/ton) =-
Since the compressors handle hydrogen, the hydrocarbon potential is
zero.
A-4.7
-------
APPENDIX 5
CALCULATIONS AKD ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE RESIDUUM HYDRODESULFURIZER
A-5.1
-------
RESIDUA!. HDS HEATER
286 x 10 BTU/hour rated capacity
(286 x 106 BTU/hour) (0.8287) = 237 x 10° llTU/hour for oil
(237 :; 106 BTU/hour)
7I32T55 BTU/gallon> = -15 8^1 Ions/hour, of_^l
(286 x 10 BTU/'-.our) (0.1713) = 49 x 106 BTU/hour for refinery gas
(49 x 10 BTU/hour) _nn._^ 106 . . . .,_
coir, n-n!/^,,K-i^ ^:««^^ ~ U.L.J// x 11, cubic feet/hour
Particulate Emissions
For oil: (237 x 10 BTU/hour)(0.074 pounds/106 BTU)* = 17.54 pounds/hour
For gas: (20 pounds/10 cubic feet)(0.0527 x 106 cubic feet/hour) =
1.05 pounds/hour
[(17.54 + 1.05) rounds/hour](24 hours/d?y)(347 days/year) _
(2000 pounds/ton)
77.41 tons/year
Sulfur Dioxide Emissions
For oil: sulfur content Oi." #5 fuel oil is 0.3 percent by vreight
(.1785 gallons/hour) (7. 586 pounds/gal? ons) (0.003) =
A0.62 pounds/hour
4r!.62 pounds/hour of sulfur corresponds to 81.24 pounds of
so2
(81.24 pounds of S02/hour)(24 hours/day)(347 days/year)
(2000 pounds/ton)
338.28 tons/year
For gas: Amount of sulfur = (52,700 SCF/hour)(0.0514 pounds/SCF)
(0.025/100)**
= 0.68 pounds/hour
0.68 pounds/hour sulfur correspond to 1.35 pounds/hour SO-
*Derived by PES on pages 99-100.
**Weight percent estimated by Foster Wheeler.
A-5.2
-------
SO emission =
(1.35 pounds/hour) (24 hours/day) (347 uays_/y_eajr)_
(2000 pounJs/ton)
5.64 tons/year
Total SO emission = 338.^3 + 5.64 = _34_3.92 tons/year
Nitrogen Oxides
1. Based on Radian emission factors.
For oil: (69 pounds/10 gallons burned)(1.785 gallons/hour) =
123.17 pounds/hour
For gas: (230 pounds/10 cubic feet) (0.0527 x 10 cubic feet/hour)
12.12 pounds/hour
[(123.17 + 12.12) pounds/hour] (24 hours/day)(347 days/year)
(2000 pounds/ton)
563.35 tons/year
2. Based on AP-4Q values for the chart in Section A-l.
For oil: the value of 237 x 10 BTU/hour heat input is 130 pounds/
hour
For gas: the value of \9 x 10 BlU/hour heat input is 9.3 pounds/
hour
1(130+ 9.3) pounds/hour] (24 hours/day)(347 days/year) =
(2000 pounds/ton)
580.05 tons/year
Hydrocarbon Emissions
3 3
For oil: (3.33 pounds/10 gallons burned) (1. 785 x 10 gallons/hour)
5.94 pounds/hour
For gas: (30 pounds/10 cubic feet) (0.0527 x 10 cubic feet/hour) =
1.58 pounds/ho'ir
[(5. 94 +1.58) Pounds/ho-.rH24 hours/day) (347 d^gAy^r), ?1 31 tons/year
(2000 pounds /ton) =_ . . .-
A-5.3
-------
RELIEF VALVES
Uncontrolled Emissions
(1.00 pounds/IP __b_a_rr_^lsj (184__x_10.^Jlarr_el_s/_d?v_)_(34_7_day^A'_ear)
(2000 pounds/ton) " "" ~
31.92 tons/vear
Controllcd Emissions
All relief valves from the resiuual hydrodesulfurizer will be
equined with a rupture disk or vent to the flare system. In either
case, the emissions will be essentially zero.
CONTROL VALVES AND FLANGES
Controlled and uncontrolled emissions are the same (see Appendix 1).
els)(184 x 103 hai
(2000 pounds/ton)
(2.55 pounds/103 barrels)(184 x 103 barrels/day) (347 days/year) _
81.41 tons/year
BLIND CHANGING
Uncontrolled Emissions
«
(2000 pounds/ton)
(0.027 pounds/103 barrels)(184 x 1Q3 barrels/day)(347 days/yea^l =
0.86 tons/year
Controlled Emissions
Good refinery procedures such as line flushing use of "line" blinds
and blind insulation with gate vlaves will reduce emissions to a
negligible quantity.
-------
SAMPLING
Uncontrolled Emissions
(0. 209,pounds/10 barre^sJJJ184_j<__10^jLa^_o_l_s/.-inv ) (347 days/year)
(2030 pounds/ton)
tcns/vear
Controlled Emissions
Good refinery practices such as avoidance of excessive sampling
purging ai:d flushing sample purge to the sump will reduce emis-
sions to a negligible quantity.
PUMPS AND COMPRESSORS
Controlled Emissions
For explanation of the following table, see Section A-l.
RESIDUAL HYDRODESULFURIZER PUMPS n..JD COMPRESSORS
UNITS AND SERVICE
Residuum Pump from distillation
Residuum Feed Pump #1
Residuum Feed Pump // 2
Residuum Feed Pump //3
Wild Naphtha Pump #1
Train //I Residuum Pump
Train #2 Residuum Pump
Train #3 Residuum Pump
Wild Naphtha Pump It 2
Wild Naphtha Pump #3
Fractionator Bottoms (Residuum Pump)
#5 Fuel Oil Pump
Fractionator Overhead Pump
Multi-Stage Reciprocating Make-up
Compressor #1*
Multi-Stage Reciprocating Make-up
Compressor #2*
Single- Stage Centrifugal Recycle
Compressc r*
Two-Stage Reciprocating Low Pres-
sure Recycle Compressor*
TOTAL
REID VAPOR PRESSURE
0.007
0.007
0.007
0.007
10.0
0.007
0.007
0.007
10.0
10.0
0.007
O.OC85
1.75
Unknown
Unknown
Unknown
Unknown
POUNDS /DAY
0.0
0.0
0.0
0.0
2.7
0.0
0.0
0.0
2.7
2.7
0.0
0.0
0.3
0.0
0.0
0.0
0.0
8.7
cent methane. As a result, the hydrocarbon emissions fron these
units are considered to be negligible.
A-5.5
-------
(8.7 pounds/day)(347 days/year)
(2000 pounds/ton) =
Uncontrolled Emissions
Use the emission factor of 5 pounds/day for each pump seal. No
estimate is made for compressor emissions. Since the material
handled is primarily hydrogen, there is no hydrocarbon potential.
(.5 pounds/day - seal) (13 seals) (347 days/year)
(2000 pounds/ton) ' ' _ll-3 tons/gear
A-5.6
-------
APPENDIX 6
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE DEBUTANIZER (STABILIZER)
A-6.1
-------
RELIEF VALVES
Uncontrolled Emissions
3 3
(1.00 pounds/10 barrels)(184 x 10 barrels/day)(347 days/year)
(2000 pounds/ton)
31.92 tons/year
Controlled Emissions
All relief valves from the debutanizer will be equipped with a rupture
disk or vent to the flare system. In either case, the eiuissions will
be essentially zero.
CONTROL VALVES AND FLANGES
Controlled and uncontrolled emissions are the same (see Appendix 1).
i (ISA x 303 barrels
(2000 pounds/ton)
3 3
(2.55 pounds/10 barrels)(ISA x 30 barrels/day)(347 days/year) _
81.41 tons/year
BLIND CHANGING
Uncontrolled Emissions
(0.027 pounds/103 barrels)(184 x 1Q3 barrels/day)(347 days/year) =
(2000 pounds/ton)
0.86 tons/year
Controlled Emissions
Good refinery procedures such as line flushing use of "line blinds,"
and blind insulation with gate valves will reduce emissions to a neg-
ligible quantity.
SAMPLING
Uncontrolled Emissions
o 3
(0.209 DnunHc/i_r>J bnr-clc)(134 x 10 barrels/day)(347 days/year)
(2000 pounds/ton)
6.67 tons/year
A-6.2
-------
Control led Emls s ions
T'jdJ refinery practices such as avoidance of excessive sampling purg-
ing an^ar) = 0./|7 ton? /year
(2000 pound s/torj - ^~ - -- -
A-6.3
-------
APPENDIX 7
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE NAPHTHA SPLITTER
A-7.1
-------
RELIEF VALVES.
Uncontrolled Emissions
3 3
(1.00 pqunds/IO barre] s)_(184 x _10_ barrel s_/_dav Li_3^
(2000 pounds"/ton) "
31.92 tons/year
Controlled Emissions
All relief valves from the naphtha sp. Ltter wj.ll be equipped with
a rupture disk or vent to the flare system. In either case, the
emissions will be essentially aero.
CONTROL VALVES AND FLANGES
Controlled and uncontrolled emissions are the same (see Appendix 1).
3 3
(2.55 pounds/10 barrels)(184 x 20 barrels/day)(3^7 days/year)
(2COO pounds/ton'/
81.41 tons/vear
BLIND CHANGING
Uncontrolled Emissions
3 3
(0.027 pounds/10 havre.ls) (184 x 10 t arrelg/dayj (347 days/year)
(20CK3 pounds /ton)
0.86 tons/year
Controlled Emissions
Good refinery procedures £uch as line Mushing, use of "line" blinds,
and blind insulation with £/>te valves vill reduce emissions to a neg-
ligible quantity.
-------
SAMPLING
llr-.'-ontrolled Emissions
.(P.: 20° pounds/10J barrels) (184 x 10 barrels/day) (347 days/year)
(2000 pounds/ton)
6. f>7 _->. ~ '-/car
REFLUX PUMP
iJnc"nt rolled Emissions
(5 pounds/day - saal)(1 seal) = 5 pounds/day
(: po. 7ic'q/day) (347 days/year) __ .
-b---cr-^^;. . J .. -*- = 0.87 tons/year
(pnf.^ Dounas/day) -^
Controlled ET ' ssions
ir.hs refli"-. being pumped to have a Reid vapor pressure of 6.
This value corresponds to an emission factor of 2.7 pounds/day (see
Appendix 1).
(2.7 pounds/day) (347 days/year) = tons/year
(2000 pounds/ton) ^
A-7.3
-------
APPENDIX 8
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE CATALYTIC REFORMING UNIT
A-8.1
-------
CATALYTIC REFORMER HEATERS
470 x 10 BTU/hour rated capacity
(470 x 10 BTU/hour) (0.8287) = 389 x 1Q6 BTU/hour for oil
(389 x 10 BTU/hour)
(132,755 =
(470 x 106 BTU/hour) (0.1713) = 81 x 106 BTU/hour for refinery gas
(81 x I"!6 BTU/hour) . A0-.. . .6 v. c ,,_
(930~BTU/ cubic foot) = °'°87i x 10 cubic f^t/hour
Particulate Emissions
For oil: (389 x 10 BTU/hour)(0.074 pounds/106 BTU)* = 28.79 pounds/hour
For gas: (20 pounds/10 cubic feet)(0.0871 x 10 cubic feet/year) =
1.74 pounds/hour
[(28.79 + ].74)pounds/hour](24 hours/day) (347 days/year)
snr\r\s\ ,/ -.
(2000 pounds/ton)
Sulfide Emissions
For oil: sulfur content of //5 fuel oil is 0.3 percent by weight
(2930 gallons of oil/hour)(7.586 pounds/gallon)(0.003) =
66.68 pounds/hour of sulfur
66.68 pounds/hour of sulfur corresponds to 133.36 pounds/hour
of S02
(133.36 pounds/hour of S02)(24 hours/day)(347 days/year)
(2000 pounds/ton)
355.31 tons/year
For gas: Amount of sulfur = (87,100 SCF/hour)(0.0514 pounds/SCF)
(0.025/100)**
= 1.12 pounds/hour
1.12 pounds/hour of sulfur correspond to 2.24 pounds/hour 30^
*Deiived by PES on pages 99-100.
**WeiRht percent estimated by Foster Wheeler.
A-8.2
-------
SO emission =
(2.24 pounds/hour)(24 hours/day)(347 days/year) ,
(2000 pounds/ton) = 9'32 tons/year
Total SO emission = 555.31 + 9.32 = 564.63 tons/year
Nitrogen Oxides (N02)
1. Based on Radian emission factors.
3 3
For oil: (69 pounds/10 gallons burned)(2.930 x jo eallons/hour) =
202.17
For gas: (230 pounds/10 cubic feet)(0.0871 x 10 cubic feet/hour) =
20.03 pounds/hour
[(202.17+ 20.03) pounds/hour](2A hours/day)(347 days/year)
(2000 pounds/ton)
92i 24 tons/year
2. B. °d on AP-40 values from the chart in Section A-l.
For oil. the value for 389 x 10 BTU/hour heat input is 215 pounds/hour
For gas: the value for 81 x 10 BTU/hour heat inpvt is 16 pounds/hour
1(215 + 16) pounds/hour] (24 hours/day) (347 days/year) , _
_»_s c *-"-. *-*-* « - 9ol.oo
(2000 pjunds/ton)
Hydrocarbon Emissions
For oil: (3.33 pounds/103 gallons burned)(2.930 x 10 gallons/hour) =
9.76 pounds/hour
For gas: (30 pounds/106 cubic feet)(0,0871 x 106 cubic feet/hour) -
2.61 pounds/hour
[(9.7642.61) pounds/hour](24 hours/day)(347 days/year) = $1 ^ tons/year
(2000 pounds/ton) .. ' y :
A-8.3
-------
RELIEF VAIA'ES
Uncontrolled Emissiois
_q._0p pounds/10 barrels) (ISA _x _-.0^_ barrels/d ay) (347 days/year)
(2000 pounds/ton) "' ~~"~
31.92 tons/year
Controlled Emissions
All relief valves from the catalytic reformer will probably vent to
the flare system. Therefore, the emissions will be essentially zero.
CONTROL VALVES AND FLANGES
Controlled and uncontrolled emissions are the same (sse Appendix 1).
(2.55 pounds/10' barrels)(184 x 1Q3 barrels/day)(347 days/year) =
(2000 pounds/ton)
81.41 tons/year
BLIND CHANGING
Uncontrolled Emissions
(0.027 pounds/10 barrels)(184 x 10 barrels/day)(347 days/year)
(2000 pounds/ton)
0.86 tons/vear
Controlled Emissions
Good refinery procedures such as line flushing, use of "li^e" blinds,
and blind insulation with gate valves will reduce emissions to a
negligible quantity.
SAMPLING
Uncontrolled Emissions
re_J
(2000 pounds/ton)
(0.209 pounds/103 barrels)(184 x 10 barrels/day)(3A7 days/year) =
6.67 tons/vear
A-8.4
-------
Controlled ''mssions
Good refinery practices such as avoidance of excessive sampling
purging and flushiij. sarapie purge to the stifnp will reduce emis-
sions to a negligible quantity.
PUMPS /tNP COMPRESSORS
Controlled Emissionc
For an explanation of the following table, 3".e Section A-l.
CATALYTIC REFORMER PlIMPS
UNIT AND SERVICE
Charge Pump
Separator Liquid Pump
TOTAL
REID VAPOR PRESSURE
5.2
5.2
POUNDS /DAY
2.7
2.7
5.4
(5.4 pounds/day)(347 days/year)
(2000 pounds/ton)
0.94 "cms/year
Uncontrolled Emissions
(5 pounds/day-seal)(2 seals) = 10 pounds/day
(10 pounds/day)(347 days/year) , ^ tons/year
(2000 pounds/ton) J
A-8.5
-------
APPENDIX 9
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE DEPENTANIZER
A-9.1
-------
I'ncontrol 1 «. J _ Lmissions
l
(2000
3 3
(1.00 pounds/10 barrels) (184 x 10 barrels /cj y ) ( 3 '< 7 d ay s/ ye a r )
'"
31.92 tons /year
Controlled Emissions
All relief valves from thu depentanizer will be equipned with a
rupture disk or vent to the flare system. In either case, the
emissions will be essentially zero.
CONTROL VALVES AND FLANGES
Controlled and uncontrolled emissions are the same (see Appendix 1)
(2.55 pounds/10"
81.41 tons/year
3 3
(2.55 pounds/10 barrels') (184 x 10 barrels/day)(3A7 days/year) _
(2000 pounds/ton)
BLIND CHANGING
Uncontrolled Emissions
3 3
(0.027 pounds/10 barrel?)(ISA x 10 barrels/day)(347 days/year)
(2000 pounds/ton)
<. 86 tons/year
Controlled Emissions
Good refinery procedures such as line flushing, use of "line" blinds,
and blind insulation with gate valves will reduce emissions to a
negligible qvantity.
SAMPLING
Uncontrolled Emissions
(0.209 pounds/10 1 barrels)(184 x 10 bar re Is/day) (347 days/year) _
(2000 pounds/ton)
6.67 tons/vear
A-5.2
-------
Controlled Emissions
Good refinery practice." such as avoidance of excessive sampling
purging and flushing sample purge to the sump will reduce emissions
to a negligible quantity.
PUMPS
Controlled Emissions
For an explanation of the following table, see Section A-l.
DEPENTANIZER PUMPS
UNIT AND SERVICE
Overheads pump
Bottoms pump
TOTAL
REID VAPOR
PRESSURE
7.0
5.2
POUNDS/DAY
2.7
2.7
5.4
(5.4 pounds/day) (3 A7 days/year), _ 0.94 tons/year
(2000 pounds/ton) J
Uncontrolled Emissions
(5 pounds/day seal)(2 seals) = 10 pounds/day
(10 pounds/day)(347 days/year) = x ?4 tons/year
(2000 pounds/ton) '
A-9.3
-------
APPE1IDIX 10
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE THE
EMISSIONS FROM THE ISOMERIZER UNIT
A-10.1
-------
TSOMERIZER HEATER
81 x 106 BTU/hour rated capacity
(81 x 106 BTU/hour) (0.8287) = 67 x 106 BTU/hour for oil
(67 x 106 BTU/hour) 1n ., .
(132.7S5 BTU/gallon) = 505 8allo"s/hour o. oil
(81 x 106 BTU/hour) (0.1713) = 14 x 106 BTU/hour for refinery gas
(14 x 106 BTU/hour) _ n mc. 6 K- t /i,
(930 BTU/cubic feet) " 0'°151 X 10 cublc f«t/hour
Particulate Emissions
For oil: (67 x 106 BTU/hour) (0.074 pounds/lO^ BTU)* = 4-96 pounds/hour
For gas: (20 pounds/106 cubic feet) (0.0151 x 106 cubic feet/year) =
0.30 pounds/hour
[(4.96 + 0.30) pounds/hour] (24 hours/day) (347 days/year)
(2000 pounds/ton) 21.90 tons/gear
Sulfur Dioxide Emissions
For oil: sulfur content of #5 fuel oil is 0.3 percent by weight
(505 gallons/hour) (7.586 pounds/gallon) (0.003) =
11.49 pounds/hour of sulfur
11.49 pounds/hour of sulfur corresponds to 22.58 pounds/hour
of S02
(22.58 pounds/hour of S02) (24 hours/day) (3-4? days/year) _
(2000 pounds/ton)
94.02 tons/year
For gas: Amount of sulfur = (15,100 SCF/hour) (Q.0514 pounds/SCF)
(0.025/100)**
= 0.19 pounds/hour
0.19 pounds/hour sulfur corresponds to 0.39 pounds/hour S02
*Dervied by PES on pages 99-100.
**Weight percent estimated by Foster Wheeler.
A-10.2
-------
SC>2 emission
_(CK3V ^OLrtWnour) (J4 hours /da;>-;_ ( 3ft_;_d_a_y_s_/j; eaO_
(2000 pounds /ton) "
1.62 t ins /year
To;al S02 emission = 94.02 1.62 = 95.64 tons/ <><> r
Nitrogen Oxides (N02)
1, Bascvl on Radian emission factors
For oil: (69 pounds /103 gallons burned) (0.505 gaUoi, ''hi-jr) --
34. S5 pounds /hour
For gas: (230 pounds/106 cubic feet) (0.0151 x 106 cubii foe: /hoar ) =
3.47 pounds/hour
[(34.85 + 3.A7) pounds /hour] (24 hours /day) (347 days/year)
C2000 pounds/ton)
159.56 tons/year
2. Based on AP-iO values fron the chart in Section A-l
For oil: the value for 67 x 106 BTU/hour heat input is 28 po^:»di./hoi r
For gas: the value for 14 x 10^ BTU/hour heat input is 2.1 pounds /h"ir
[(28 +2.1) pounds /hour] (24 hours/day) (347 days/year)
2000 pounds /ton)
125.34 t.
Hydrocarbon Emissions
For oil: (3.33 pounds/103 gallons burned) (0-505 x 103 gallons /ho\ r)
1.68 pounds /hour
For gas: (30 pounds/106 cubic feet) (0.0151 x 106 cubic feet/ho<.r) =
0.45 pounds /hour
[(1.68 + 0.45) pounds /h our] (24 hours/day) (347 days/year) =
(2000 pounds /ton)
8.87 tons/year
A-10.3
-------
RELIEF VALVES
Uncontrolled Emissions
(1.00 pounds/10 barrels) (184 x 10 barrels/day)(347 days/year)
(2000 pounds/ton)
31.92 tons/year
Controlled Emissions
All relief valves from thj isomerizer will probably vent to the
flare system. Therefore, the emissions will he esentially zero.
CONTROL VALVES AND FLANGES
Controlled and uncontrolled emissions are t:.e same (see Appendix 1),
;) (184 x 103 barre!
(2000 pounds/ton)
(2.55 pounis'103 barrels)(184 x 103 barrels/day)Q47 days/year)
81.41 tonj/year
BLIND CHANGING
Uncontrolled
(0.027 pounds/103 barrels)(184 x 1Q3 barrels/day)(347 days/year) =
(2000 pounds/ton)
0.86 tc.is/year
Controlled Emissions
Good refinery procedures such as line flushing, use of "line" blinds,
and blind insulation with gate valves '.fill reduce emissions to a neg-
ligible quantity.
SAMPLING
Uncontrolled Emissions
> o
£0.209 pounds/10 barrels)(184 x 10 barrels/day)(347 days/year)
(2000 pounds/ton)
6.67 tons/year
A-10.4
-------
Controlled Emissions
Good refinery practices suc>. as avoidance of excessive sampling
purging and flushing sample purge to the sump will reduce
emissions to a negligible quantity.
PUMPS AND COMPRESSORS
Controlled Emissions
For an explanation of cne following table, see Section A-l.
ISOMERIZER PUMPS AND COMPRESSORS
UNIT AND SERVICE
Charge Pump
Stablizer Overheads Pump
iiakt-Up Compressor*
TOTAL
REID VAPOR
PRESSURE
7.0
7.0
0.0
POUNDS /DAY
2.7
2.7
0.0
5.4
*?4 tons/
(2000 pounds/ton) J
A-10.5
-------
APPENDIX 11
CALCULATIONS VND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE LPG PLANT
A-13.1
-------
RELIEF VALVES
Uncontrolled Emissions
3 3
(1.00 pounds/10 barrels) (184 x _1_0 barrels/day) (3n7 days/year)
(2000 pounds/ton)
31.92 tons/year
Controlled Emissions
All relief valves from the LPG unit will be equipped with a rupture
disk or vent to the flare system. In either case, the emissions
will be essentially zero.
CONTROL COCKS AND FLANGES
It is assumed that the cocks involved will emit in the same raanner
as a valve. Controlled and uncontrolled emissions are the same
(see Appendix 1).
(2.55 pounds/10 barrels)(184 x 1Q3 barrels/day)(347 days/year) =
(2000 pounds/ton)
81.41 tons/year
BLIND CHANGING
Due to the use of Hareer blinds, the potential and actual emissions
will be negligible.
SAMPLING
Uncontrolled Emissions
3 3
(0.209 pounds/10 barrels)(ISA x 10 barrels/day)(347 days/year) _
(2000 pounds/ton)
6.67 tons/year
Controlled Emissions
Good refinery practices such as avoidance of excessive sampling
purging and flushing sample purge to the sump will reduce emissions
to a negligible quantity.
A-11.2
-------
PJJMTS AN))
Controlled Emissions
For expalnation of the following table, see section A-l.
LPG PLANT PUMPS AND COMPRESSORS
UNIT AND SERVICE
Drum level control pump*
Inter-stage drum pump
Aftercooler pump
Absorber bottoms pump
Debutanizer overheads pump
Depropanizer overheads pump
Overhead 2-stage antifugal
compresson**
TOTAL
REID VAI'OR
PRESSURE
26
26
20
3.5
26
26
26
POUNDS /DAY
11.1
11.1
2.7
0.3
U.I
11.1
0.0
47.4
j
*Pump is an on-off unit, not in continuous service. Impossible
to determine amount of service so assumed it to operate continuous
**Emissions from compresson seals will be vented to the flare.
(47.A pounds/day)(3A7 days/year)
(2000 pounds/ton)
8.22 tons/year
Uncontrolled Emissions
Because the average uncontrolled factor for pump and compressor
seals are values lower i'han actual emission estimates, the actual
emission estimates for these units are used.
UNCONTROLLED EMISSIONS
Drum-level control pump
Inter-stage drum pump
After cooler pump
Absorber bottoms pump
Debutanizer overheads pump
Depropanizer overheads pump
Compressor
TOTAL
- 11.1
- 11.1
- 5.0
- 5.0
- 11.1
- 11.1
- 11.1
65.5
pounds/day
pounds/day
pounds /day
pounds/day
pounds/day
pounds/day
pounds/day
pounds/day
11.36 tons/year
A-11.3
-------
APPENDIX 12
CALCULATIONS AND ASSUMPTIONS USEE TO DETERMINE
THE EMISSIONS FROM THE OPTIONAL MF.ROX UNIT
A-12.1
-------
PUMPS
Controlled Emissions
For an explanation of the follovi-ig table, see sect Jon A-l
MEROX TUMPS
UNIT ANC SERVICE
Charge Pump
Product Tump
Total
REID VAPOR
PRESSURE
3.5
3.5
I
POUNDS /HOUR
0.3
0.3
0.6
LOJ_Igggk/Jay) W days/year) . tons/vear
(.2000 po'inds/ton) ==
Uncontrolled Enissions
(5 pounds/day seal) (2 seals) = 10 pounds/day
L.74 tons/vear
(10 pounds/dayU3A7 days/year)
(2000 pounds/ton)
A-12.2
-------
APPENDIX 13
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE SULFUR PLA4T
A-13.1
-------
SULFUR BURNER
45 x 10 BTU/hour rated capacity
(45 x 106 BTU/hour) (0.8287) = 37 x 1Q6 BTU/liour for oil
(37 x 10 BTU/hour) ., ,, ..
(132,755 BTU/galloa) = 279 gallons/hour for_oil
(45 x 10 BTU/hour)(0.1713) = 8 x 106 BTU/hour for refinery gas
(8 x 106 BTU/hour) n nno, ..6 , . c .,
/mn P-T.IT/r~ c \ = 0.0086 x 10 cubic feet/hour
(930 BTU/cubic foot)
Particulate Emissions
For oil: (37 x 106 BTU/hour)(0.074 pounds/106 BTU)* = 2.74 pounds/hour
For gas: (20 pounds/10 cubic feet)(0.0086 x 10 cubic feet/hour) =*
0.17 pounds/hour
J(2.74 -f- Q.17) pounds/hourK24 hours/day) (3A7 days/year) _ , . - ,
-i-i. i' ,-... ^-. 7 *-* ^ = 12.12 tons/vear
(2000 pounds/ton) ,
Sulfut Dioxide Emissions
For oil: sulfur content of #5 fuel oil is 0.3 percent by weight
(279 gallons/hour)(7.586 pounds/gallon)(0.003) = 6.35 pounds/
hour of sulfur
6.35 pounds/hour of sulfur corresponds to 12.70 pounds/hour
of S02
(12.70 pounds/hour of S02)(24 hours/day)(347 days/year)
-. -,_ ._ - - .- . _, . ....... B
(2000 pounds/ton)
52.88 tons/year
For gas: Amount of sulfur « (8,600 SCF/hour) (0.05K pou.ids/SCF)
(0.025/100)**
= 0.11 pounds/hour
0.11 pounds/hour sulfur correspond to 0.22 pounds/hour SO
*Derived by PES on pages 99-100.
**Vei£lit percent estimated by Foster Wheeler.
A-13.2
-------
SO,, emissions =
(0.22 pounds/hour) (24 hours/_day_)_^3A7_davs/^earj_
(2000 pounds/ ton)
0.92 tons/year
Total S00 emission = 52.88+ 0.92 = 53.8 tone/yaar
Nitrogen Oxides (N
1. Based on Radian emission factors.
3 3
For oil: (69 pounds/10 gallons burned) (0.279 x 10 gallons/hour) =
19.25 pounds/hour
For gas: (230 pounds/IP/ cubic feet) (0.0086 x 10 cubic feet/hour) =
1.98 pounds /hour
[(19.25 + 1.98) pounds/hour] (24 hours/day) (3A7 days/year) ^
(2000 pounds /ton)
88. AQ tons/year
2. Based on AP-40 values for the chart in Section A-l.
£
For oil: the value for 37 x 10 BTU/hour heat input is 14 pounds/
hour
For gas: the value for 8 x 10 BTU/hour heat input is 1.1 pounds/
hour
+ 1.1) pounds/hour] (24 hours/day) (347 days/year) ,, ca ,.,/..,.,
f if\/\r\ j /^ \ O^_« OO COvtS / \ Go
(2000 pounds/ ton)
Hydrocarbon Emissions
For oil: (3.33 pounds/103 gallons burned) (0.279 x 103 gallons/hour)
0.93 pounds/hour
For gas: (20 pounds/106 cubic feet)(0.0036 x 10 cubic feet/hour)
0.26 pounds/hour
[(0.93 + 0.26) pounds/hour](24 hours/day)(347 days/year) = A>% tons/
(2000 pounds/ton) J:-
A-13.3
-------
SULFUR PLANT INCINERATOR
Uncontrolled Emissions
1. Based upon material balance from Foster Wheeler. Foster Wheeler
estimates that337 long tons of sulfur will be recovered each
calendar day. If the conversion efficiency is 96 percent (a
reasonable value for a three-stage reactor), then the other 4
percent which is not converted will exit through the stack.
This aaount could be:
I (337 long tons of suJfur/day)(2240 pounds/long ton)
[ 0.96
31,453 pounds/day of sulfur
It is assumed that all of the sulfur leaving the tail gas in-
cinerator was as S02, then 31,453 pounds of sulfur/day corres-
ponds to 62,906 pounds/day of S02-
($2,906 pounds/dav of S02)(365 days/year) = . ,&Q ,
(2000 pounds/ton)
On initial examination, this appears to be a very h'igh number,
so a comparison can be made using the AP-42 emission factor
for a 96 percent efficient three-stage conversion process.
2. Using AP-42 emission factor, page 5.18-2 of 167 pounds/ton of
sulfur produced.
(167 pounds/tons of sulfur) (337 tons of sulfur/calendar day) (365 days/year)
(2000 pounds/ton)
» 10,271 tons/year
This value confirms the reasonableness of the first number.
Controlled Eirissiors
Foster Wheeler predicts that emissions from the tail-gas cleanup
system will be approximately 200 ppm of sulfur compoxmds expressed
as S02- This will include 50 ppm of carbonyl sulfide (COS) and
carbon disulfide (CS_), and 10 ppra of H2S. These values corres-
pond to predicted results published by David K. Beavron, the
designer, in 1971 (Chemical Engineering. December 13, 1971).
Since the exit gas flow rate and temperature are unknown, these
numbers, in themselves, are inadequate ro provide an estimate in
terms of tons/year.
A-13.4
-------
Mr. Beavon states that cl.is treatment of a tail-gas incinerator
contaminants can result i-i salfur control of 99.9 percent. This
value corresponds to the published value in AP-^2 and is used to
estimate controlled emissians.
(337 long tons of sulfur/day) (2240 pounds/long ton)
0.96
786.33 pounds/day of sulfur eir.ltted
786.33 pounds/day of sulfur corresponds to 1572.66 pounds/day of SO,
(1572.66 pounds /day) (365 days/year). = 287.01 tons/y
(2000 pounds /ton) -- '
ear
A-13.5
-------
APPENDIX 14
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE THE
EMISSIONS FROM THE STEAM GENERATING PLANT
A-14.1
-------
STEAM GENERATING BOILERS
490 x 10 BTU/hour rated capcity
(490 x 106 BTU/hour)(0.8287) = 406 x 10& BTU/hour for oil
(406 x 106 BTU/hour)
(132,755 BTU/gall^T = ^8 gallons/hour of oil
(490 x 106 BTU/hour)(0.1713) = 84 x 106 BTU/hour for rffinery gas
(84 x 106 BTU/ho-jr) 6 , . , .,
(930 BTU/cubic foot) 0-0903 x 10 cubic feet/hour
Particulate Emissions
For oil: (406 x 10 BTU/hour)(0.074 pounds/106 BTU)* = 30.04 pounds/hour
For gas: (20 pounds/10 cubic feet)(0.0903 x 10 cubic feet/year) =
1.81 pounds/hour
[(30.04 + 1.81) pounds/hour](24 hours/day)(347 jays/year) _
(2000 pounds/ton)
132.62 tons/year
Sulfur Dioxide Emissions
For oil: sulfur content of #5 fuel oil is 0.3 percent by weight
(3058 gallons/hour)(7.586 pounds/gallon)(0.003) =
69.59 pounds/hour of sulfur
69.59 pounds/hour of sulfur corresponds to 139.18 pounds/
hour of SO
(139.18 pounds/hour" of S02)(24 hours/day)(347 days/year)
(2000 pounds/ton)
579.55 tens/year
For gas: Amount of sulfur = (90,300 SCF/hour)(0.0514 pounds/SCF)
(0.025/100)**
= 1.16 pounds/hour
*Derived by PES on pages 99-100.
**Weight percent estimated by Foster Wheeler.
A-14.2
-------
1.16 pounds/hour sulfur correspond to 2.32 pounds/haur SO
SO emission =
(2.32 pounds/hour)(24 hours/day)(347 days/year) . ..
(2000 pounds/ton) ^ = 9'6b
Total S02 emission = 579,55 + 9.66 = 589.21 tons/year
Nitrogen Oxides (N02)
1. Based on Radian emission factors.
For oil: (69 pounds/10 gallons burned)(3.058 x 10 gallons/hour) =
211.00 pounds/hour
For gas: (230 pounds/10 cubic feet)(0.0903 x 10 cubic feet/hour) «
20.77 pounds/hour
[(211,00+ 20.77) pounds/hour](24 hours/day)(347 days/year) =
(2000 pounds/ton)
965.09 tons/year
2. Based on AP-40 values for the chart in Section A-l.
For oil: the value for 406 x 10 BTU/hour heat input is 220 pounds/
hour
For gas: the value for 84 x 10 BTU/hour heat input is 17 pounds/
hour
[(220 + 17) pounds/hour](24 hours/day)(347 days/year) = gg6 8? tons/year
(2000 pounds/ton) ' ^
Hydrocarbon Emissions
3 3
For oil: (3.33 pounds/10 gallons burned)(3.058 x 10 gallons/hour) -
10.18 pounds/hour
For gas: (30 pounds/10 cubic feet)(0.0903 x 10 cubic feet/hour) =
2.71 pounds/hour
[(10.18+2.71) pounds/hour](24 hours/day)(347 days/year) = ?3>67 tons/year
(2000 pounds/ton) . f '
A-14.3
-------
APPENDIX 15
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE THE
EMISSIONS FROM MISCELLANEOUS HREC EQUIPMENT
-------
REFINERY FUEL GAS CO^-RESSOR
Uncontrolled Emissions
(11.1 pounds/day-seal) (1 seal) 11.1 pounds/day
Controlled Emissions
Assume the pressure of the primarily methane stream en the seal to
be 26 RVP. The er.issicn value which corresponds to this value is
11.1 pounds-'hour (see Appendix; 1).
(11.1 pounds/hour) (3^7 cavs/vear) , . ,
,OArvn :; r = 1.93 tons/year
(2000 pounds/ton) 3
REFINERY FUEL GAS FUMF
Uncontrolled Egissic/is
(11.1 pounds/day-seal) (1 seal) = 11.1 pounds/day
(11.1 pounds/day) (3w dcys/yearj = l^ tons/ r
(20UO pour.as/ton) J
Controlled Emissions
(11.1 pounds/day) (347 days/year) , 1>93 tons/
(2000 pounds/ton) '
FUEL OIL PUMP
Uncontrolled Emissions
_(5 pounds/day) (3^7 days/year) = 0-87 Eons/year
(2000 pounds/day) ^
Controlled Uraissior.s
The vapor pressure of the product pumped will be 0.008 which will
emit negligible hydrocarbons.
A-15.2
-------
7 PRODUCT PUMPS
Uncontrolled Emissions
(5 pounds/day-seal) (7 seals) = 35 pounds/day
(35 pounds/day) (347 da;Ts/year)
(2000 pounds/ton)
= 6.07 tons/year
Controlled Emissions
The seven product pumps are shown below:
UNIT
LPG Pump
Gasoline Putnp //I
Gasoline Pump #2
Jet Fuel Punp
$2 Fuel Oil Pump
#5 Fuel Oil Pump #1
#5 Fuel Oil Pump 02
TOTAL
REID VAPOR
PRESSURE
26
7
7
1.75
0.07
0.008
0.008
POUND" /DAY
11.1
2.7
2.7
0.3
0.0
0.0
0.0
16.8
J16.8 pounds/day) (347 days/year) = 2 91 tor.s/vear
(2000 pounds/ton) 7
A-15.3
-------
APPENDIX 16
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE THE
EMISSIONS FROM THE EMERGENCY FLARE SYSTEM
A-16.1
-------
PILOTING
The ground level flare will burn pilot gas at the rate of 3,000
SCFH.
(3000 cubic ^eet/hour)(2A hours/day)(355 days/yea.) = 26. 2 i x 1Q6
cubic feet/year
The elevated flare will burn pilot gas at the rate of 1,200 SCFH.
(1200 cubic feet/hour)(24 hours/day)(365 days/year) = 10.512 x 1Q6
cubic feet/year
Particulates and Hydrocarbon Emissions
Each flare is equipped with steam injection to improve tne combus-
tion efficiency. Proper operation will ensure that all hydrocar-
bons entering the burner will be destroyed. Since the ash con-
tent of the fuel gas is negligible, the psrticulates and hydrocar-
bon emissions are considered to be insignificant.
Nitrogen Oxides (N02)
The emission factor used is 0.014 pounds/10 BTU of pilot gas
burned from Emissions of Oxides of Nitrogen Stationary Sources
in Los Angeles County, Report //2, September 1960.
(36.792 x 106 cubic feet/year)(930 BTU/cubic foot) = 34216.56 x 10
BTU/yp.ar
(34216.56 x 106 BTU/year)(0.014 pounds/106 BTU) B Q ^ tons/year
(2000 pounds/ton)
Sulfur Dioxide Emissions
(36.792 x 106 cubic feet/year)(0.0514 pounds/cubic foot)(0.00025)
= 472.78 pounds/year of sulfur which corresponds to 945.56 pounds/
year of S0_.
(945.56 pounds/year) _ ^ tons/year
(2000 pounds/ton) : . /
A-16.2
-------
V'-.fc investigators could not develop any strategies of their oun
-t- calculate the frequencies or duration of upsets associated
yith startups and shutdowns. Therefore, the estimates shown
-.ere are the values predicted by Foster Wheeler.
Tr.e amount of time that the ground level flare will operate dur-
ing routine start-up and shutdown operations is estimated to be
approximately 50 hours per year. It must be remembered that dur-
ing initial shake down and corrective operations occuring during
the first year of operation the number of shutdowns and start-
ups will be greater than the normal shown here. The total numbei
of times the ground level flare will be in use during normal re-
finery operations is estimated to be about 30 times per year,
the operations varying in length from about one-half hour to
at>out eight hours. The f-lue gas rate will vary, depending on
operation, from less than 40,000 pounds/hoUi. to about one million
pounds/hour. The estimated operations and emissions are sum-
marized in the table on the following page. A. second table fol-
lows showing flare loads from each unit during these periods.
It is impossible to estimate emissions resulting from emergency
dumping to tht elevated flare. Emissions could be very high for
short periods of time. However, these conditions occuv very sel-
dom and have, not been considered.
A-16.3
-------
..ut DC-AU 60 15 10
St-.rl V;>
Deiu-iT-: !.
Puvf,e
C a t ,- ] / L- v P. c\: o ,1 c- r a t i t n
Mstillat.^ Jlyilro-.lesr "'"'.>-;: c?r :
Stfirt I"-.
De;n ;:::;.'!-;
Purge
Ca ta lyt - ."^ c-'r,ev.i- . ~ t ic">
j\o."iJ'.'iTi l-vt'.rcd o.'vi] t^ri^cr!
St.iri Lr.
i)cp;''.-s:-u'"e
Tur^c
Catalyst P.cgcncrc'- icr,
tt'butanizer-Shu'i. uovu
Naphtha Gp] -Itter-S!;ul; Bovn
Catalytic Refomer:
Start Up
Deprcssure
Purge
Catalyst Regeneration
Dcpcntanizor-Shut Dov>i
Isomcriiit-r :
Start Up
Fl f-> T"* ro*"r*iii*r*
j'irj L.L.I»>L*' t_
Pur^c-.
Catalyst Ro^oneral ion
LPC I'lnnt:
*J(Jpl t*l?SIIl t?
}Ji > t- r» r\
t , L } ,L.
Reproduced from I^'B
best a-ailable copy. \&£?
80 30
10 13
3 4
20
:^ 67
4,000 420
16b 100
4 S 30
20
A, 210 570
x-* »/-,*-. ^O/*"*
Oi , '*l,A -L , -'"V
/. c < , Tin
*4 ? 'j J. / U
3AO 50
532 . 100
6?.,5i>2 2,160
10
8
72
22
6
7
107
15
67
26
8
i /.
j 'i
115"
2,123 10
_530 __3_
2 ,"6^3 3 ^ 3
A-1 fi It
n 4- U *T
3
5
1 '
2
1.1
18
8
3
1
30
o r\
J ^1
1 4
8
'lOA
7
e
6
?
1
I
10
12
4
3
1
o
^
To
8
/
in
\
/*
7
o
3
r&
27
I'.:
4
?.
45
I?'1
". R
J o
d
]/
lii
10
10
10
3
2
1
IG"
is
6
5
2
i
J
T6
12
3
15
-------
3r,.
A-16.5
-------
0-30
r. o
Si-
S"r. Unit and Frequency
3 3 SfcH'J-Lli;-! Ko/Ycrir
'tvj S:n:t i'o'Ti 1/2
::?.-- htha ;.D3:
Si;Tt I'p 1/1
!;oprc.~.:-.'irc 1/2
?r.HY.ro:;.:vvc 2
i'u"r° 2
l""-- R'-"fn- /(
Cac. i'cf or-.. or:
C c , .» 11. 1
.^W'«Li-^>kj J.
C ^ ~> r o s P \ i r c 1
Cat . Ke^eil. 1
r 1 ;i :." " '
VT ''!'' I (f !)^
Pollnt.-irt. J-'.-'i- ''.;.(..>. ."') '. : ,7 , .-,-
1/2 10 .1-', 3 r- ;.:>. -,:."i:i '. '. ".
1 20 2 3 '4,^ 1 3 ,'>.') '.. . >'. ' "i . .-':
3 l,33/« 1-.Q ,' 9 2'i,0/fi,201 f/. °i '.-'"'i ' '')
7 -S3 3D /, 6 I2.r"-.,00') 0. '!'
1 -',5 TO 3 M .'i,'-';. .:!'.; 0." ! ' ; , -'-
1 - .'3 l ?. /.,)'<;. />'.(; (.:: .. ; ; : . ' fi
/* 7.075 7.30 1.0 15 /.3,2'..),ri ,n ''..'.' ' 1 :, :
2 17.0 '.:} 3 3 '.'.; '.':,:')) C..-P »'; i. ..
.1 70 y, r 3 ".,!;::, vo >' j.. :"' .. i
1 133 23 2 .» 3, .',.:!.: ...:'.,
2 36 3 r> 13.. '?:, C".^ 0.0:7.?'.
1 P..' : '! " '' i TM ';.r " ' ! ' !
1/2 - I'.'. L /. .-4...v,.:"r, o.'i;°"! : ' '.'
1 - 7 1 A 1 ,7! 1 , '-K.'j 0.'il-'-7 : ,. . ,
?',.'.t i)o-n
Xr.ph:h.-. Sj litter:
"'."It OO'.,T»
1/2 1/2
1/2 1
30
'0 /.O.fj'v'i /i"0 C. O.1' -3
36 ^6,^6:,6CO ;).o:iv^
-------
Unit and
Operation
Depentauizer :
Shut Covm
St.irt Up
Dc;;resmre
Pur je
Car. Ro^en.
LPG Plant :
Depress '.ire
Fui-.-i
;iyr.ro£o.u Unit:
Start Up
Shut Down
Frequency
No/Year
1
1
1
1
1
1/2
1/2
1/2
1/2
Durntion
Hours
1-3/4
2
1
1/2
1
O
1/2
8
8
v;-:v ,k
sn? t:o.. ni . f
9 7
33 2
26 3
1" 2
14 2
2,123 10 G
530 3 2
J.J4 U
114 12
i"'il | r)-
' ' ' ^ ' ".^
10
-
3
3
3
12
3
1 «^
1".
"" Flow
.' i :
17,379
3 , 7 '. 0
7..-.0''.
^ . f. :
4.27')
"i 2 , 000
12..V.1
27, ISO
27,;.TJ
;<;:tc
:'il .-
,700
» -''''
. >"'()
, P ao
,oOO
.';'JO
.son
,3(^0
,';'.'()
o.o: c'.'
-------
APPENDIX 17
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM THE WATER COOLING SYSTEM
A-17.1
-------
COOLING TOWER
The only emissions will b? in the form of hydrocarbons. Since
the system will only operate during the hot summer months, it
has been assumed that operation would be three months/year.
Foster Wheeler estimated the water circulation rate as being 8,000
gallons/mir.ute during operation.
The emission factor used was 6 pounds/10 gallons of ccoling water
(AP-42, page 9.1-5).
SO min/t
(2GCO pounds/ton)
(0.008x10 gal/tain)(60 min/hr)(24 hr/day)(3A7 days/yr)(.25)(6 lb/106
3.00 tons/year
A-17.2
-------
APPENDIX 18
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE THE
EMISSIONS FROM THE OIL-WATER SEPARATION UNITS
A-18.1
-------
LTDS SYSTEM
Uncontrolled Emissions
Use emission factor of 5 pounds/10 gallons waste vater (Radian).
3
(5 pounds/10 gallons of waste water) (2iO gallons/oinute)(60 minutes/hour)
= 69 pounds/hour
(69 pounds/hour)(24 hours/day)(365 days/year) _
(2000 pounds/ton) = 302.22 tons/year
Controlled Emissions
Use emission factor of 0.2 pounds/10 gallons waste water (Radian).
3 3
(0.2 lb/10 gal, waste water)(0.23x10 gal/min)(60 min/hr)(24 hr/day)(365 day'yr>
(2000 pounds/ton)
= 12.09 tons/year
Another source of emissions is the diverter and its associated over-
flow pond. All tank farm water which is felt to be contaminated will
be run throagh this system in addition to process drain overflow. It
is very difficult to predict how often it will rain and how much water
will be in the overflow pond at any one time. The normal process
water flow is 230 gallons/minute through the diverter pump. Since
the maximum flow through this pump is 250 gallons/minute, the maxi-
mum rate at which overflow water can be reworked through the system
is 20 gallons/minute. It must be assumed that, the overflow system
and diverter pump have been sized so that at ti?e very worst, 20 gallons/
minute would maintain equilibrium. The uncontrolled potential from
this system if equilibrium were being maintained ,*euld then be:
o o
(5 lb/10 gal)(0.02x10 gal/min)(60 min/hr)(2A hr/day)(365 days/yr) ^
(2000 pounds/ton)
26.28 tons/year
This emission factor is for normal process water. The overflow from
the diverter will pass through the underflow baffle and the oil wiH
be collected by a brill rope skimmer. This is predicted to effect
a hydrocarbon reduction of the water of from 500 ppm to 150 ppm, or
70 percent. This 350 ppm of oil is to be collected in the diverter
by the skimmer. In order to ensure that this 350 ppm of recovered
oil is not transformed into air pollution before the skimmer can
collect it, an additional control measure should be considered. A
simple practice which is used in chrome plating tanks is to cover
A-18.2
-------
the surface with plastic floating balls. This provides a very
flexible floating roof on the divcrtcr surface. The only vapor
space exists in the gaps between the balls. The roof is also
flexible enough to allow the skimmer to work around it and will
not be effected by the storm water surges. Addition of this system
could effect a 70 percent reduction in the process water. If the
tank farn storm water is considered to be as dilute as the. diver-
ter overflow water, then the actual emissions from this system
can be estimated as being:
(26.28 tons/year)(1 - 0.70) = 7.88 tons/year
HTDS SYSTEM
Since there is no diverter on this system, the only source of emis-
sions will be the final separator.
Uncontrolled Emissions
3 ^
(5 lb/10 gallons waste water)(.342 x 10" gallons/minute)(60 minutes/hour)
= 102.6
(102.6 pounds/hour)(24 hours/day)(365 Pays/year) = 449.39 tons/year
(2000 pounds/ton) -
Controlled Emissions
]_
(2000 pounds/ten)
CO.2 Ib/hr)(0.342 x 103 gal/min)(60 cdn/hr)(24 hrs/day)(365 days/yr)
17.98 tons/year
A-18.3
-------
APPENDIX 19
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE THE
I
EMISSIONS FROM THE SLUDGE INCINERATOR
A-19.1
-------
MULTIP T.E-HEARTH INCINERATOR
Amounts of primary sludge recovered have been estimated to be:
Low TDS system: 25 pounds/hour
High TDS system: 100 pounds/hour
Ballast water: 60 pounds/hour
TOTAL: 185 pounds/hour
Amounts of secondary sludge have been estimated to be 100 pounds/hour
The input to the incinerator could average 285 pounds/hour if the unit
is continuous.
(285 pounds/hour) (24 hours/day) Q47 days/year,) 11Q, ,. ,
,_-.: -. ' ' "- = 1186.74 tons/year
(2000 pounds/ton.)
Particulate Emissions
Uncontrolled Emissions: Use emission factors from AP-42, page 2.5-2.
(100 pounds/ton) (1186.74 tons/year) = 59 _, tons/vear
(2000 pounds/ton) y
Controlled Emissions:
(3 pounds/ton) (1186. 74 tons/year) = 1>7g ton,;/year
(2000 pounds/ton)
Sulfur Dioxide Emissions
Use emission factors from AP-42, page 2.5-2.
Uncontrolled Emissions:
(1 pound/ton)(1186.74 tons/year) = Q_59 tons/vear
(2000 pounds/ton) '.'
Controlled Emissions:
(0.8 pounds/ton) (n_g_6^1^_Lgi^/^£ail = O.A7 tons/year
(2000 pounds/ton) - --^=-
A-19.2
-------
Nitrogen Oxides (NC^) Emissions
Use emission factors frora AP-A2, page 2.5-2.
Uncontrolled Emissions:
(6 pounds/ton) (1186. 74 tons/year) . _,
- (2000 pounds/ton) " " J'56
Controlled Enissior.s:
(5 pounds/ton)(1186.74 tons/year) 0_ ,
UOOO pounds/ton) ' ' 2'97 tons/yegr
Hydrocarbon Emissions
Use CTrission factors from AP-^»2, page 2.5-2.
Uncontrolled Emissions:
(1.5 pounds/ton) (1186.74 tons/year) _
. _ , . ' . "" U» O
(2000 pounds/ton)
Controlled Emissions;
(1 pound/ton)(1186.74 tons/year) _ 0_59 tons/year
(2000 pounds/ton) .
A-19.3
-------
APPEMI/IX 20
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
THE EMISSIONS FROM CRUDE OIL UNLOADING
A-20.1
-------
BOILER OPERATIONS
542.27 barrels of fue] are used per visit
(542.27 barrels/visit)(107 visits/year) = 58,023 barrels/year of fuel
burned
Particulate Emissions
Uncontrolled (Nor.-Inerting) Emissions: use emission factor of 0.966
pounds/barrel of fuel burned.^
(0.966 pounds/barrel)(58.023 barrels/year) _ ,
/^f\nf\ j / \ ~~ zo.UJ tons/year
(2000 pounds/ton) /.
Controlled (Inerting) Emissions: PES estimated in the Alaskan crude
oij. project that 59 percent of the fuel usc.d per visit was associated
with the discharging of crtide oil. During discharge, approximately
15 percent of the combustion gages generated are routed to the tanker
holds for inerting purposes.
(28.03 tons/year)[1 - (0.59)(0.15)] = 25.55 tons/year
Sulfur Dioxide Emissions
2
Assume fuel oil used to be #6 fuel oil with 1.5 percent sulfur.
Emission factor used is 6.7 (percent sulfur) Bounds/barrel of fuel
oil burned (Esso).
Uncontrolled E-nissions;
[6.7(1.5) Bounds/barrel1(38.^L barrftls/yearl = 291>57 tons/year
(2000 p^nds/ton) - ' :
Cortrolled Emission?:
(291.57 Eons//rar;[l - (0.59)(0.15)] - 265.77 tons/gear
^sco Research and ^.iglnee "ing Company. Suivey of Ship Discharges,
Report on Contract No. v> 1-35049 to Maritime Adroinistraticv.. Office
of Research anu Development, July 1974.
2Ass trap Lion mad; by Federal Energy Administration for West Coast
traffic.
A-20.2
-------
Nitrogen Oxides (NOo) Emissions
F-nission factor used is 4.36 pounds/barrr1 of fuel oil burned (Esso).
Uncontrolled Emissions:
(4.36 pounds/barrel)(58,023 barrels/vear)
(2000 pounds/ton) ' * -126'A9 tons/year
Controlled Emissions:
(126.i9 tons/year)[1 - (0.59)(0.15)] - 115.30 tons/year
Hydrocarbon Emissions
Use enissien factcr of 0.13-^ pounds/barrel of fuel burned (Esso).
Uncontrolled EEissicns:
JO.134 p3unds/barrel)(58.023 barrels/year), 3.89 tons/year
(2000 pounds/ton) .
Controlled Emissions (combustion contaminants routed to the tanker hold):
(3.89 tons/year)[1 - (0.59)(0.15)] - _3.55 tons/year
REFUELIXG
Uncontrolled and controlled situations are the same. Use emission fac-
tor of 0.61 pounds/vi*it.
(0.61 pounds/visit) (107 visits/year) Q Q3 tons/.rear
(2000 pounds/ton)
VENTING
Uncontrolled (Non-Inerting) System for 5 Percent Hydrocarbon Vapors;
(998 pounds/visit)(107 visits/year) 53.39 tons/year
(2000 pounds/ton) - .. '~
A-20.3
-------
Controlled (Inerting) System: no emissions of hydrocarbon vapors.
S\LLASTINC
Emission factor assumes 20 percent of the ballast by weight would be
added in port.
Uncontrolled (Non-Segrega ted) Emissions:
(1978 pounds/visit)(107 visits/year) _nr Q0 ,
/..nnn J/ \ " 105.82 tons/year
(2000 pounds/ton) ;-1
Controlled (Segregated) Emissions: if a segregated ballast of be-
tv?een 15 anc 18 percent is used, the hydrocarbon emissions drop to
essentially zero.
PIERCING
Tljis process is only possible on a tanker capable of inerting. The
possibility of converting an existing non-inerting tanker to inerting
cztpa oil! ties is considered. The emission factor used pror-r-tes exist-
irig factors and assumes that 50 percent of the tanker hold gas (con-
ts-ining 5 percent hydrocarbons) is purged to the ambient air to re-
dujce the hold atmosphere to below the lower explosive limit on each
vlpit.
(9?87 pounds/visit)(107 visits/year) _ 53A.3Q tons/year
( 2000 pounds/ton) . .' -
A-20.4
-------
APPENDIX 21
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE THE
EMISSIONS FROM MARKETING FACILITIES
A-21.1
-------
AMOUNTS OF PRODUCTS FROM THE HREC REFINERY
30,000 barrels/calendar day of regular gasoline = 1,260,000 gallons/day
6,273 barrels/calendar day of premium gasoline. = 263,466 gallons/day
21,324 barrels/calendar day of Jet A-l fuel = 895,608 gallons/day
34,787 barrels/calendar day of £2 fuel oil = 1,461,054 gallons/day
66,723 barrels/calendar day of #5 fuel oil = 2,802,366 gallons/day
CASE I: 20,000 gallons/day of gasoline will leave by truck loading rack
40 percent of all redlining products will leave by pipeline:
gascliae: 601,386 gallons/day
Jet A-l fuel: 358,243 gallons/day
#2 fuel oil: 584,422 gallons/day
#5 feel oil: 1,120,946 gallons/day
The remaining material will leave by barge:
gasoline: 902,080 gallons/day
Jet A-l fuel: 537,365 gallons/day
02 fuel oil: 876,632 gallons/day
#5 fuel oil: 1,681,420 gallons/day
PIPELINE EMISSION;
Assume each produce pumped has one pump and five valves associated with
transferring and switching into the line.
Controlled Pump Seal Emissions
Gasoline pump seal: 2.7 pounds/day
Jet A-l fuel pump seal: 0.3 pounds/day
#2 fuel oil putip seal: 0.0 pouziCs/day
#5 fuel oil punp seal: 0.0 pounds/day
TOTAL: 3.0 pounds/day
(3.0 pounds/day)(365 days/year) = p 55 tons/year
(2000 pounds/ton)
Uncontrolled Purp Seal Emissions
(5 pounds/seal-day)(4 seals) = 20 pounds/day
I
A-21.2
-------
(20 pounds/dav)(365 Javs/vear)
" = 3-05 tons'tear
Uncontrolled Pipeline Valves Emissions
Use Radian emission factor. (Consider controlled emissions to be
the same.)
(0.15 pounds/day-valve)(20 valves)(365 d?v/vear)
(2000 pounds/ton) ' ~~^' = °'55 tons/year
GASOLINE TRUCK LOADING RACK.
Splash Loading
Use general AP-42 enission factor and not equation due to lack of
information.
(20 x 10 gallons/day loaded)(12. A pounds/103 gallons)(365 days/year)
(2000 pounds/con)
" 45.26 tons/year
Submerged Fill
3 3
(20 x 10 gallons/day loaded)(A.I pounds/10 gallons)(365 days/year)
(2000 pounds/ton)
" 16.97 tons/vear
MARINE BARGE LOADING
Use Radian equation of L » 12.46 SPM for all products except gasoline
T
which uses straight Radian emission factors. For the » ;ose of this
examination, it will be assumed that barges are clean^. i transit.
Uncontrolled Emissions (includes submerged fill and bottom loading):
Gasoline:
jl.2 pounds/103 gal Ions)(902.OS x J_0_jagllons/day)(347 days/year)
(2000 pounds/ton)
= 197.56 tons/year
A-21.3
-------
Jet A-l Fuel: equaticu is 12. 46(0 . 20) (1 .0) (80)
--- ' ' -- * - = 2.848 pounds/10
gallons
(2. 848- pounds/10 gal !ons\(537_._3_65_x_10 __ ?£jjj>ns/d_ay ) Qblj^l? 6leili)
(2000 pounds/ton) ~~~ ' '"
279.30 tons/vear
92 Fuel Oil; equation is 12.46(0.20)(0.05)(130) . .... , ., 3
^~^ = 0.2314 pour.ds/10 gallons
3 °
(0.2314 pounds/10 gallons)(876.652 x ID"' gallons/day)(365 days/year) =
(2000 pounds/ton)
37.02 tons/vear
#5 Fuel Oil; equation is 12.46(0.20) (0.008) (190) .. .... , ,...3
!'r^' -^ * = 0.054 pounds/10 gallons
(0.054 pounds/103 gallons)(1681.42 x 103 gallons/day)(365 days/year) =
(2000 pounds/ton)
16.57 tons/year
Vapor Recovery Emissions
Assumed 90 percent collection efficiency.
(530.45 tons/year)(1 - 0.90) = 53.045 tous/year
Tugboat Emissions
The barge has an estimated volume of 40* x 150' x 10' = 60,000 cubic feet
17.48 gallons/cubic foot)(60,OOP cubic feet/barBe) . 1Q 6g6 barrelE/barge
(42 gallons/barrel)
If the barges are assumed to average being 95 percent filled with prod-
uct, then each barge will transport 10.15 barrels.
A-21.4
-------
Case I considers 3,997,497 gallons/day leaving by barge
_ (3,997.497 gallons/day) _
(10,152 barrels/barge) (42 gallons/barrel) J/i barees
(9.375 barges /day) (365 days/year) = 3422 barges/year
Since a tugboat is required to move each barge, within five tniles of
the refinery each barge will necessitate the use of ten tugboat
(3422 barges/year) (10 tufeboat miles/barge) = 34,220 tup^oat miles/year
Particulates:
Use emission factor of 2 pounds/tugboat mile, AP-42, Tables 3.2,3-2.
(2 pounds/tugboat mile) (34, 220 tugboat miles/year) ,. ,
* - ftnnn - 7"* - \ -- - - ~ 34. 22 tons/year
(2000 pounds/ ton)
Sulfur Dioxide:
Use emission factor of 3 (percent sulfur) pounds/tugboat mils and
0.5 percent sulfur content
I
[3(0.5) pounds/ tug boat milej(34,220 tugboat miles/year) = 9r>67 tons/year
(2000 pounds/ ton) =lV -
Hydrocarbons;
Use emission factor of 0.9 pounds/tugboat mile, AP-42, Tables 3.2.3-2
(0.9 pounds/tugboat mile) (34. 220 tugboat miles/year) _ 15>AQ tons/year
(2000 pounds/ton) = -
N02:
Use emission factor of 1.4 pounds/ tugboat mile, AP-42, Tables 3.2.3-2
(1.4 pounds /tugboat mile) (34, 220 tugboat miles/year) = 23.95 tons/year
(2000 pounds/ ton) -
A-21.5
-------
CASE II:
All products leaving by barge.
Uncontrolled Emissions (includes submerged fill and bottom loading):
3 3
(1.2 pounds/10 gallons)*(1523.A66 y. 10 gallons/day)(365 days/year)
(2000 pour.ds/ton)
333.64 tons/year
Jet Fuel;
3 3
(2.848 pounds/10 gallons)(895.608 x 10 gallons/day)(365 days/year)
(2000 pounds/ton)
A65.50 tons/year
C/2 Fuel Oil:
;ds/103 gall
(2000 pour.ds/ton)
(0.2314 pounds/103 gallons)(1461.054 x 1Q3 gallons/day)(365 days/year)
61.70 tons/year
#6 Fuel Oil;
(0.054 pounds/10 gallons)(2802.366 x 10 gallons/day)(365 days/year)
(2000 pounds/ton)
27.62 tons/year
Controlled Emissions (based upon use of a 90 percnet efficient vapor);
Gasoline:
(33.64 tons/year)(l - 0.9) - 33.A tons/year
Jet Fuel:
(465.50 tons/year)(1 - 0.9) = 46.6 tcns/^ear
*Emission factor for clean barges.
A-21.6
-------
12 Fuel Oil;
(16.70 tons/year)(l - 0.9) =6.2 tons/vear
#6 Fuel Oil:
(27.62 tons/year)(1 - 0.9) = 2.8 tons/vear
Tugboat Emissions
Case II considers 6,682,494 gallons/day leaving by barge
(6,682,494 gallons/day)(365 days/year)
(10,152 barrels/barge) (42 gallons/barrel)
(5721 barges/year)(10 tugboat miles/barge) = 57,210 tugboat miles/year
All emission factors used come from AP-42, Tables 3.2.3-2
Particulates:
_C2 pounds/tugboat mile) (57,210 tugboat miles/vear) ..., .,, ,
,or.^r T~7I\ " = 57.21 tons/year
(2000 pounds/ton) " .-=_
Sulfur Dioxide:
[3(0.5) pounds/tugboat milel(57,210 tugboat miles/year) , ni _ ,
i c ° J-1r B = 42.91 tons/year
(2000 pounds/ton) J
Hydrocarbons;
(0.9 pounds/tugboat mile)(57.210 tugboat miles/year) = ?4 ton /
(2000 pounds/ton) *
(l.A pounds/tugboat mile)(57.210 tugboat miles/year) = AQ Q5 tons/ ar
(2000 pounds/ton) . . /
A-21.7
-------
APPENDIX 22
CALCULATIONS AND ASSUMPTIONS USED TO DETERMINE
EMISSIONS FROM THE STORAGE TANK FACILITIES
A-22.1
-------
CALlTLATIOIx OF STO!v\G£ TANK LOSSES
A. FLOAT IN." nOU? T.-j;K?
The r ;_ loi:iri equations are used to calculate standing storage
losses and v:: ;'-irrr: losses from floating roof tanks. These equa-
tions are trie rociiied version of the API equation after introduction
of a (O.C8 - 'v) facrer as recommenced by API in API bulletion 2523.
Stan-ir.c. S:cr^.ge Losses
S = 0.00921 CO 'D^-5 h^TIp]0'7 (vj'? (KC) (Kj (Kp) (Kj Ibs/day
for tanks < 150 ft diameter
S = 0.00921 0:) (3 x v'150;
14.7-P
for tanks >150 ft diameter
Wit'ndravl Losses (for gasoline only)'
LWD = * >w t"F/ lb/1000 gal throughput
*Rli.ian stated that this equation was derived from gasoline data and
' its' applicability to other stored liquids ±s uncertain. Therefore
PEJ did not use this equation for other liquids.
A-22.2
-------
NOMENCLATURE
P = True vapor pressure at bulk liquid temperature, psia
D = Tank diameter, Ft
H = Average vapor space height, including correction for roof volume, Ft
AT = Average daily ambient: temperature change, °F
F = Paint factor, fixed roof tank
C = Adjustment factor for tanks smaller than 20 feet in d'aroeter
V « Capacity of tanks, bbls
W = d = vapor density, Ib/gal
K = Characterization factor depending on type of product stored and tank
M = Vapor molecular weight, Ib/lb-mole
Bg = Breathing losses of gasoline
m = Factor dependent on liquid stored
K = Turnover factor = (180+N/6N)
N = Number of turnovers per year (ratio of annual throughput to tank capacity)
K *= Factor dependent on tank construction
V = Average wind velocity, miles/hour
K = Seal factor
s
K = Paint factor for color of shell and roof, floating roof tanks
P
L - Working losses; lb/1000 gal throughput
S « Standing storage losses, Ib/D x 10 gal capacity; or; bbls/yr; or Ibs/day
L - Withdrawl losses; floating roof tank, bb?^/10 bbl throughput; or
lb/1000 gal throughput
C_ *» lank construction factor
* for D >150; use D ^150 instead of D
A-22.3
-------
Tanks 1 to 11
Product stored: Crude oil
Capacity of each tank: 22050 x 10 gallons
Total throughput: 2,682,742 x 10 gallons/yr (total of 11 tanks)
P- 2.0 psia; Vw = 11.025 mile/hr;D = 264 ft
H= 50.0 Ib/lb mole, K = 0.045, K = 1.0, K = 1.0, K = 0.84
t s p c
Standing Storage Losses
. T p -|0.7 0.7
0.00921 (M) (D x/150) |]rSrp w c s P c
L ' J
10.7
. f o o "I0>7 o 7
0.00921 x 50 x 264 x/150 \u -^ -Q- (11.025) x 0.045 x 1.0 x 1.0 x 0.84 Ibs/day
82.81 Ibs/day
\.
total Standing Storage Losses (all 11 tanks)
82.81 (Ibs/day) x 365 (D/yr) x -~ (T/lb) x 11 = 166.24 T/yr
A-22.4
-------
29-32 (Total 4 tanks)
Product: Gasoline
Capacity /Tank: 4220 x 10* gallons
Throughput: 566,663 x 10 gallons/yr (total of 4 tanks)
P = 5.2 psia, Vw = 11.025 mile/hr, D = 134 ft
M = 68 Ib/lb mole, K. = 0.045, K = 1.0, K = 0.9, K =- 1.0
C S p C
Standing Storage Losses . .
1 5 I 5 2 I0'7 07
S = 0.00921 x 68 x (134)J-':> x L^^-J x (11.025)0'7 x 0.04£ x 1.0 x 0.9 x 1.0
= 138.5 ILs/day per tank
- 138.5 (Ibs/day) x 365 (day/yr) x - (T/lbs) x 4 (No. of tanks)
= 101.2 T/yr
Wi' ndrawl Losses
(22.4)(5.6) x 0.02 / _ Ib _ ] ,,, ,,- ,^3 (feal thruputj
Lwd ' * - 134 - \1QOQ gal thruputj X 566'663 x 10 ^ yr j
= 10,609 Ibs/yr
Lwd - 5.3 T/yr
Total Losses: S + L , = (101.2 + 5.2) T/yr
U'Q
- 106.5 T/yr
A-22.5
-------
Tanks 33-34 (Total 2 tanks)
Product: Jet fuel
Capacity/tank = 13,649 x 103 gallons
Throughput = 326,897 x 10 gallons/yr (for total of 2 tanks)
P = 1.0 psia; Vw = 11.025 nile/hr; D = 220 ft
M = 80 Ib/lb mole; K = 0.045; K = 1.0; K »= 1.0; K =1.0
L- S D O
Standing Storage Losses
.
S = 0.00921 x 80 x 220 x /ISO" x L° J ' x(11.025)°'7 x 0.045 x 1.0 x 1.0 x 1.0
' 0. 7
'
i.7-1.0^
= 76.74 Ibs/day per tank
= 76.74 (ILs/day) x 365 (day/yr) * 2i5oO (T/lbs) x 2 (tanks)
S - 28.0 T/yr
» Tanks 46-47 (Total 2 tanks)
Product: Regular naphtha
Capacity/Tank: 5,064 x 10 gallons
Throughput: 457,430 x 10 gallons/yr (total for 2 tanks)
P - 1.0 psia, Vw = 11.025nile/hr; D = 134 ft
M « 80 Ib/lb mole; K - 0.045; K - 1.0; K - 0.9; K » 1.0
Standing Storage Losses
S = 0.00921 x 80 x (134)1'5 x f-^-ylJI ' x (11.025)0'7 x 0.045 x 1.0 x 0.9 x 1.0
= 39.8 Ibs/day per tank *
= 39.8 (Ibs/day) x 365 (D/yr) x -£~ (T/]b) x 2 (tanks)
- 14.5 T/yr
A-22.6
-------
0 Tanks 48-49 (Total 2 tanks)
Product: Light i aphtha
Capacity:
2,105 x 10 gallons each tank
^
Throughput: 177,912 x 10 gallous/yr
P - 3.5 psla; V = 11.025 ;nile/hr; D = 80 ft
M - 68 Ib/lb mole; K = 0.045; K = 1.0; K = 0.9; K = 1.0
it s p c
Standing Storage Losses
1 1 5 / 3 5 \°'7 07
S - 0.0092 x &8 >: (80) x I j^-yjrj (11.025) x 1.0 x 0.9 x 1.0 x (0.045)
,
= 43.1 Ibs/clUy per tank
= 43.1 (Ib/
- 15.7 T/yr
43.1 (Ib/dAy) x 365 (days/yr) x ~ (T/lb) x 2 (tanks)
A-22.7
-------
B. FIXED ROOF TANKS
Using Radian Equations:
B = Breathing losses = 0.000221 x M x h~ J " x D1<73x H°-51xAT°'5x Fp ? C
Ibs/day
LW = Working losses = 0.024 x M x P x KR x K lbs/1000 gal throughput
180
Tanks 35-38 (lotal of 4 tanks)
Product: No. 2 fuel oil
Capacity /tank: 13,649 x 10 pal Ions
Throughput: 533,285 x 103 galloas/yr (total for 4 tanks)
No. of turnovers/yr = N = 10.0
P - 0.05 psia; D = 220 ft, H = 24 f t , AT = 16.1 °F, F = 1.0, C = 1, K = 1.0,
M = 130 Ibs/lb mole °
Breathing Losses
O Aft
B = 0.000221 x ISOi":0:? -J " x (220)1'73x (24)°-51x (16.1)°'5x 1.0 x 1.0
\14. 7-0.05 /
* 138.2 Ibs/day per tank
» 138.2 (Ib/day) x 365 (day/yr) x ^ (T/lb) x 4 (tanks)
- 100.9 T/yr
Working Losses
K - 1.0
L^ - 0.024 x 130 x 0.05 x 1.0 x l.o(1Qj g^ thruput) " 533,285 x 103 (gals/yr)
- 83,192 Ibs/yr
- 41.6 T/yr
Total Losses = B + LU - (100. P + 41.6) T/yr
- 142.5 T/yr
A-22.8
-------
Tanks 39-45 (Total of 7 tanks)
Product: No. 5 fuel oil
Capacity/tank: 10,660 x 10" gallons
Throughput: 1,^22,864 x 10 gallons/yr (Total for 7 tanks)
No. of turnovers/yr = N = 13
P = 0.008 psia, D = 180 ft, H = 28 ft, AT = 16.1°F, F = 1.33,* C = 1.0,
K = 1.0, M - 190 Ib/lb mole
Breathing Losses
B = 0.000221 x 190 I Y4~fi§^§08) x (180)1-73x (28)°'51x (16.1)°<5x 1.33 x 1.0 x 1.0
- 59.0 lb&/day per tank
= 59.0 (Ibs/day) x 365 (day/yr) x -^^ (T/lb) x 7 (tanks)
= 75.4 T/yr
Working Losses
Kn = 1.0
(» !
lb \i,022,864 x 10 (
10 gals thruput/
= 34,329.0 Ibs/yr
- 17.2 T/yr
Total Losses = B + L = (75.4 + 17.2) T/yr
w
- 92.4 T/yr
*Use colored tank^ to aid in keeping the liquid pour-able.
A-22.9
-------
Tanks 50-52 (Total of 3 tanks)
Product: Gas oil
Capacity/tank: 7,644 x 10"' gallons
T
Throughput: 540,842 x 10' -.lions/year (For total cf 3 tanks)
No. of turnovers/yr = N = Ii4.j
P = 0.0152 psla; D = 180 ft, H = 20 ft, AT = 16.1 °F, F = 1.0, C = 1.0
K =1.0, M = 130 Ib/lb mole
Breathing Losses
B = 0.000221 x 130. .:;---- x (180)1-73x (20)°'51x (16.1)°'5x 1.0 x 1.0 x 1.0
I I'-t . / U .Ui_>^f
= 39.^ Ibs/doy per tank
= 39.5 (Ib/day) x 365 (day/yr) x ^QQ (T/lb) x 3 (tanks)
= 21.7 T/yr
Working Losses
= 1'° , , f ^
= 0.02A x 130 x 0.0152 x 1.0 x 1.01 x 540,842 x 10:
yiO gals rhruput/
» 25,649 Ibs/yr
- 12.8 T/yr (Total for 3 tanks)
Total Losses = B + LW - (21.7+12-8) T/yr
- 34.5 T/yr
gals
A-22.10
-------
« Tanks 53-55 (Total of 3 tanks)
Product: Residuum (atmospheric) kept at 250°F by heating
Capacity/tank: 18,951 x 103 gallons
Throughput: 1,203,200 x 10 gallons/yr (For total of 3 tanks)
P = 0.00632 psia,* D = 240 ft, H = 28 ft, AT = 16.1°F, Fp =1.46, C = 1.0,
KC = 1.0, M = 190 Ib/lb mole
Breathing Losses
B = 0.0002r.l x 1901 7lo°oo63.->| x (28)°'5'x (240)1-73x (16.1)°<5x 1.46 x 1.0 x 1.0
\ " ' "I
= ?0.7 ]li-/day per tank
= 90.7 (Ib/day) x 365 (D/yr) x -~^ (T/lb) x 3 (tanks)
- 49.7 T/yr
Working Losses
L = 0.024 x 190 x 0.00632 x 1.0 x l.ofr 1 x 1,203,200 x 10
'* I ^ f\ *
gals
10 gals chruput/ V yr /
= 34,675 Ibs/yr (Total for 3 tanks)
* 17.3 T/yr
Total Losses = B + LW = (49.7 + 17.3) T/yr
= 67.0 T/yr
*Losses are quite high as compared to reported data by Foster Wheeler bec-iuse
Foster Wheeler used vapor pressure 0.0005 psia which is at 60*F, while actual
storage condition is at 250°v which when correlated to vapor pressure yields
a value of 0.00632 psia. (See following p?ge.)
A-22.11
-------
VAPOR PRESS, or ATMOSPHERIC RESIDUUM
room temperatu
a temporal or*.
Vapor pressure
is not known.
data eorresror.
vapor press jr=
100°F. These
a general ec-_i
'""- :'' -c;'-'jr rili ht-' stored in tanks 33-55. Since at
' --- :s r.iunly viscous, these tanks will be kept at
-- -" r ease movement into and out of the tanks.
c:ua :or atmospheric residuum at this temperature
-:-. trie absence of this information, vapor pressure
:-n»: t; residual oil No. 6 is used. However, the
-at?. :or N'o. 6 residual oil is only available up to
ai-es ;re plotted as Log P versus 1/T according to
VP
- B.'I*
The above
in the abser. ;e
'"yp = '-'i-cr pressure, ATM
~ = ;K
=cu2*irn is a very generalized equation. However,
c-r data, it is used for the purpose of this report.
7.ESIDUAL OIL No. 6
t,°F
40
50
60
70
80
90
100
200
250
300
283.0
288.6
294.1
299.7
305.2
310.8
366.0
394.0
422.0
1/T
0.00360
0.00353
0.00347
0.00340
0.00333
0.00328
0.00322
0.00273
0.00254
0.00237
Pw=psi.
0.00002
0.00003
0.00004
0.00006
0.00009
0.00013
0.00019
When this data is plotted (see following graph):
B
'V
0.00000136
0.00000204
0.00000272
0.00000408
0.00000612
0.00000884
0 00001290
P., at 250CF i.e. at 0.00254 is 0.0043 ATM
*- 0.0043 (ATM) x 14.7 (psia/ATM)
= 0.00632 psig
i
*Reld, Sherwood: Properties of Gases and Liquids, Second Edition, McGraw Hill, p.114,
^ A-22.12
Reproduced from |r£ '
besl available copy. %ry
-------
.00
.000:-
.
:
J-
-.
?
)0
s
t
5
s.
4_
3_
'.
: :-'; f~
':-
: -. : :
_ . ...
_
- - :
--'-"--'
-.-_-_---
-i_- :
--
-
"-..:
_:.
i
r:,. . ,|
1 ,-
-r f-:-.
: r'"^
zr_ -r=t :
- ;.-- - .
----- --I;-
Jr^=i~.
T ' - :-
'
r~.^.~^.
1 ;-.
[
1 !
-^- l.-!
-=:- - -
ir?=r :-
--_: -
- -
7-: :i -i
r- -;----
i^ .:_- :
- * '-
- -
-"i 1 ;
i--.f:-i
---i.r-;-
; : -i
1
.. . 1
-
.- -^i;:
:- .:.-:
:-::;r "'
-H-~ -;.r:
: ; .-*-.
- -
-------
- . > 1 .~
!
- -J-,---
r
( ; ~
L -.-.
---:- 1=T
"r_T±_--
ir" -t.
- i j i j-
i
I
' ! 1 '
.;-- .?_ L
: -:- -.;_
_^j i :_
: ;. i
- ' : 'i -.-
' : ----I' i-
-."- ; . : i-
'- -i i
I -i L-l
. . j 1 j
i i i
,.
.: J
.
- : " '
'-'.'.'
. -^ r -:
- .-:_--
r-*-.
_._
.
r- -. ~
_1_
;; i-^.--.
-
-^=T-
. '
i ' i
I !
1 '
---.--;. \'
----- ----
---.-..
:.-: :'--
-E'..: -:
. _.
-H ;
;- . ^
. , . ._
' ' I I
i i : :
1 ;
.
' _i
-
-
- .
'_ ___
-
-
- --
.- r
, ._ ..^
-j^~-
-i- | i
1
i 1 ! i
1 '
. :- -- -"
r . :
t-- .::- :
;.; : - '
;: } -.: : -
.. i .;.-..
., . :-:
1 , .- - 4_.
.! i :.'!
.- -
-
'----
.. . ._
" i
^ *
--. - \
i
. -. ;
- - - i
-. x
-----
~^- - - --- ---
.-...- _
1; ^~-.^-~-~
-
-
___
-
..-. -_ L
-:. - ---: .
i-- T"
- - :
- . _J
__- - -.;
- :
' - . --
>
'
\- . '- .
.....
1
7
---, H
- , _
- -. I
5
. . - - -
P^
.. .,-
~
__,.
;.:.:, r-^
- -.--
.:: - -_ : r
-
. J
. . .J.
., .. n
... :
:
- .. ; .
T. . .. -1 .- -:-:.
_ ' j- J
. -_^'-
_' | . .
L --i""" "I
-. ,_.
1
-:- . ,=-= ;
= . . '
pv£p:r.
I . -.I:-^:,,T_,
^=r::i^:j=--i
L
t _ .
r; ) ~
9 '
i J
1
-'-,- - - .
-. S4>. -. | ; - :-
- : r -: r - -
:-- T H- .-: : :
~::^:4 -~-r":
. u .
-::---;_:-
".;--'4'- i) ; -
-;-.'- ^ '-; i
1 ' : . j 1
}
; i
' - I
:!-.-
.
_+_-.
-------
|
__ . L..
-
. . "- .
'
- - 1
i
----- .T-q:.. -----
"--(-"
._.-_- ; -
L"--- :
-- f" -"
1.
!
4
. .-_--. : - - ._-- - i
-. -- . .* -
.... 7 - --
. _
.
-:_-_- -?-. ;^r.
- -. .- ------ -
- --.- --. _r~- . ::-
h -MrD:::--.!
;.-: .."i"|l ;
-r-rrl- H r
. : .. : i
.....
. -: i. ; ;.-.-..
. . 1 . 1
.... j
-
' ' ~"~
- - ' '-
- . '.- '
-
_ . . .
- -
---------
-: - -
_- -_,
- --
'-'---
L.
. .
.
-. '~
r. -
j.
--
".
.." _- ,
- ^:
--;
'I
- -
--~
-.:;-:.-:-
1 . .- . .. :
- .. . 1 ... : -
; . ;-
'
j
: i
- i
: 1
i
*
- ! . I
t
. r -I- J : =_- 1-_=;
._ ._. - ___
1
... : . ---
----- .(.._-.
- -
L . -
: -
1 J
|
*~
... .:
._
' ' - - - - -
r-
- ;:.-'^4
- * .1
1
t
. ..... ...
i
- . - .. .-.=.-_-
-- -;^r-^= ^
1 :":-r;M::.::l;f
r -;..-!- ^z^---
. . \~.-'--1 d."
3 . - ' T--f^51l-
- - - " . . rr-I i J_
-- i:: . : '
:;_t_:i_ii
--}_L_i_L-
L^_..7_
i- --j-_. _^_
.0000
0.0000 0.001 0.002 0.003
I/I
0.004
0.005
A-22.13
-------
Tanks 56-57 (Total of 2 tanks)
Product: Refinery fuel oil
Capacity/tank: 2,350 x 10 gallons
Throughput: 156,472 x 10 gallons/yr (For total of 2 tanks)
P = 0.008 psia. D = 100 ft, H = 20 ft, AT = 16.1°F, Fp = 1.33, C = 1.0,
KC = 1.0, K = 190 Ib/lb mole
Turnovers/yr = N 23
Breathing Losses
B = 0.000221 x 190|T7-~7r!n«l ' * (100)1>73x (20)°'51.x (16.1)°'5x 1.33 x 1.0 x 1.0
114 . 7-0. 008 /
= 18.0 Ibs/day per tank
= 18.0 (Ibs/day) x 365 (day/yr) x - (T/lb) x 2 (tanks)
=6.6 T/yr
Kn = 1.0
Working Losses
i^- 1 ir llfi LT) v in^ J
L = 0.024 x 190 x 0.008 x 1.0 x 1.0 = x 156,472 x 10 .
w ylO gals thruput/ \
= 5,708 Ibs./yr
= 2.9 T/yr (Total for 2 tanks)
Total Losses = B + L = (6.6 + 2.9) T/yr
w
= 9.5 T/yr
A-22.14
-------
Tanks 58-59 (Total of 2 tanks)
Product: Slop oil
Capacity/tank: 2,350 x 10 gallons
Throughput: 156,472 x 10 gallons/yr (For total of 2 tanks)
Turnovers/yr = 33
P = 0.05 psia, D = 100 ft, H = 20 ft, AT = 16.1°F, Fp = 1.33, C = 1.0,
K = 1.0, M = 130 Ib/lb mole
c
Breaching Losses
B = 0.000221 x ISOL^, 7'o505) * (100)1-73x (20)°'51x (16.1)°'5x 1.33 x 1.0 x 1.0
= 42.8 Ibs/day per tank
= 42.8 (Ibs/day) x 365 (day/yr) x ~^ (T/lb) x 2 (tanks)
= 15.6 T/yr
Working Losses
Kn"1"° / \ / \
L = 0'.-024 x 130 x 0.05 x 1.0 x 1.0)~ 1 x 156,472 x 10 r~]
w \iO gals thruput/ \ y /
= 24,410 lbs/yr (Total for 2 tanK.s)
=12.2 T/yr
Total Losses = B + L^ = (15.6 + 12.2) T/yr
= 27.8 T/yr
A-22.15
-------
Tanks 60-62 (Total of 3 tanks)
Product: Ballastic water (assumed there was a layer of kerosene
middle oilon top of water at all times).
3
Capacity/tank: 1,504 x 10 gallons
3
Throughput: 520,400 x 10 gallons/yr (Total for 3 tanks)
Turnovers/yr = N = 115.3
Data for kerosene are used for emission calculations.
P = 0.05 psia, D = 100 ft, H = 20 ft, AT = 16.1°F, Fp = 1.33, C = 1.0,
K =1.0, M = 130 Ib/lb mole
Breathing Losses
B = 0.000221 x 130(7-. °'°5nd x (100)1<73x (20)°'51x (16.1)°-5x 1.33 x 1.0 x 1.0
114.7-0.05/
= 42.82 lb&/day per tank
= 42.82 (Ibs/day) x 365 (days/yr) ^~ (T/lb) x 3 (tanks)
= 23.4 T/yr
Working Losses
Kn = 0.43 (from API Bulletin 2518, Figure 11, for N = 115.3)
or
_ 180 + 115.3 = - ,-
n 6 x 115.3 = / \ / V
lb i 3
L = 0.024 x 130 x 0.05 x 0.43 x l.Olr 1 x 520,400 x 10
w \10 gals thruput/
= 34,908 Ibs/yr
«= 17.5 T/yr (Total for 3 tanks)
Total Losses = B + LW = (23.4 + 17.5) T/yr
- 40.9 T/yr
A-22.16
-------
MISCELLANEOUS TANK FARM AND OFF-SITES
PUMP SERVICE
Tank Field Pumps
Jet fuel pump
2 Gasoline pumps
#2 fuel oil pump
2 Crude oil pumps
Tank Pumps
11 Crude oil pumps
12 Propane LPC pumps
4 Butane LPG pumps
4 Gasoline pumps
2 Jet fuel pumps
4 #2 fuel oil pumps
7 #5 fuel oil pumps
2 Naphtha pumps
2 light naphtha pumps
3 Gas oil pumps
3 Atmospheric residuum pumps
2 Refinery fuel oil pumps
2 Slop oil pumps
3 Ballast water pumps
Reid Vapor
Pressure
1.75
7.0
0.07 ;
Hydrocarbons
Uncontrolled
5
10
5
3.5 10
3.5
26
26
7
1.75
0.07
0.008
1.75
3.50
1.75
0.007
0.008
0.05
1.75
TOTAL
55
133.2
44.4
20
10
20
35
10
10
15
15
10
10
15
432.6
in !bs/day
Controlled
0.3
5.4
0
0.6
3.3
133.2
44.4
10.8
0.6
0.0
0.0
0.6
0.6
0.9
0.0
0.0
0.0
0.9
201.6
A-22.17
-------
Uncontrolled Pump Seal Emissions
(432.6 pounds/dav) (365 days/vear) -,0 . ,
/->r.rvn ^/ \ '' = 78.05 tons/year
(2000 pounds/ton) J
Controllea Punp Seal Emissions
(201.6 pounds/day)(365 days/vear) = 36.79 tons/year
(2000 pounds/ton) '^=
Valves
(5 valves/tank)(62 tanks) = 310 valves associated with tanks.
Uncontrolled: use Radian emission factor.
(0.15 pounds/day-valve)(310 valves)(365 days/year) _ g ^ tons/vear
(2000 pounds/ton) -
Controlled Emissions
Negligible
A-22.18
-------
------- |