VOLATILE ORGANIC LIQUID STORAGE VESSELS
(Including Petroleum Liquid Storage Vessels)
BACKGROUND INFORMATION FOR
PROMULGATED STANDARDS OF PERFORMANCE
Emission Standards and Engineering Division
U. S. Environmental Protection Agency
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
JANUARY 1985
-------
TABLE OF CONTENTS
Section Page
1 SUMMARY 1-1
1.1 Summary of Changes Since Proposal 1-1
1.2 Summary of Impacts of the Promulgated Action 1-3
1.2.1 Alternatives to Promulgated Action 1-3
1.2.2 Environmental Impacts of Promulgated Action 1-3
1.2.3 Energy and Economic Impacts of the
Promulgated Action 1-3
1.2.4 Other Considerations 1-4
1.2.4.1 Irreversible and Irretrievable Commitment
of Resources 1-4
1.2.4.2 Environmental and Energy Impacts of Delayed
Standards 1-4
1.2.4.3 Urban and Community Impacts 1-4
2 SUMMARY OF PUBLIC COMMENTS 2-1
2.1 Selection of the Affected Facility 2-1
2.1.1 Vapor Pressure and Tank Size Cutoff 2-1
2.1.2 Vapor Pressure Determination 2-12
2.1.3 Special Exemptions 2-14
2.1.3.1 Horizontal and Underground Vessels 2-14
2.1.3.2 Facilities Producing Beverage Alcohol - 2-22
2.1.3.3 Vessels Located on the Outer Continental
Shelf (OCS) 2-22
2.1.3.4 "Slop Oil," Wastewater, and Waxy, Heavy
Crude Storage Vessels 2-23
2.1.3.5 Negligibly Photochemically Reactive Liquids . . 2-26
2.1.3.6 Production and Process Vessels 2-27
2.2 Emission Control Technology 2-30
2.2.1 External Floating Roof Vessels (EFR's) 2-30
2.2.2 Internal Floating Roof Vessels (IFR's) 2-31
2.2.3 Add-on Control Options 2-34
2.2.4 Column Fittings 2-38
2.2.5 Equivalency Determination 2-40
2.3 Recordkeeping, Reporting, and Inspection Requirements . . . 2-41
2.3.1 Recordkeeping and Reporting Requirements 2-41
2.3.2 External Floating Roof Vessel (EFR) Inspection
Requirements 2-43
2.3.3 Internal Floating Roof Vessel (IFR) Inspection ....
Requirements 2-44
2.3.3.1 Annual Visual Inspection 2-44
2.3.3.2 Ten-Year Inspection 2-46
2.3.4 Procedures for Vessels Found to be Out of Compliance 2-47
2.3.5 Notification of Refill 2-47
ii
-------
TABLE OF CONTENTS
(continued)
Section Page
2.4 Modification 2-48
2.5 Cost Effectiveness 2-50
2.5.1 Capital Recovery Factor 2-50
2.5.2 Product Recovery 2-50
2.5.3 Cost of Controls (Cost Effectiveness) 2-52
2.5.3.1 Add-on Controls 2-52
2.5.3.2 Floating Roof Vessels .... 2-53
2.6 Miscellaneous 2-56
i i i
-------
LIST OF TABLES
Table Page
2-1 List of Commeriters on Proposed Standards of Performance
for Volatile Organic Liquid Storage Vessels 2-2
2-2 Cost Effectiveness of BDT Controls for a 113-m3 Tank . . 2-10
2-3 Cost Effectiveness of a 95-Percent-Efficient Condenser
for Chloroform Storage at Chloroform Production
Facilities 2-18
2-4 Cost Effectiveness for 95-Percent-(Mass) Efficient
Condenser 2-19
2-5 Cost Effectiveness for 95-Percent-(Mass) Efficient
Condenser 2-20
2-6 Cost Effectiveness for 95-Percent-(Mass) Efficient
Condenser 2-21
2-7 Cost Effectiveness of BDT in Constant Level Tanks .... 2-29
2-8 Estimated Installed Cost of a Welded Contact Internal
Floating Roof with Secondary Seal 2-54
iv
-------
1. SUMMARY
On July 23, 1984, the Environmental Protection Agency (EPA) proposed
standards of performance for volatile organic liquid storage vessels
(including petroleum liquid storage vessels) (49 FR 29698) under authority
of Section 111 of the Clean Air Act. Public comments were requested in
the proposal in the Federal Register. A total of 15 comments from
industry, 5 from trade associations, and 3 from governmental agencies
were submitted during the comment period. The comments that were submitted,
along with responses to these comments, are summarized in the document.
The summary of comments and responses serves as the basis for the revisions
made to the standards between proposal and promulgation.
1.1 SUMMARY OF CHANGES SINCE PROPOSAL
In response to public comments, certain changes have been made in
the proposed standards. The more significant changes are summarized
below. All changes that have been made to the regulation are explained
fully in the responses to comments.
Two specific exemptions to the standards were made in response to
requests from commenters. An exemption was added for storage vessels at
retail gasoline service stations. The EPA did not intend to affect
vessels at gasoline service stations with these standards. Consequently,
no evaluation of the possible economic impact of these standards on
retail gasoline marketers was performed. These vessels are part of a
separate source category, and the decision as to whether to regulate
emissions from these vessels will be made when standards for that category
are developed. Refer to Section 2.1.3.1, Horizontal and Underground
Vessels, for further explanation of the change.
1-1
-------
Facilities producing beverage alcohol have been excluded from the
standards because they have been previously exempted from the source
category. Refer to Section 2.1.3.2, Facilities Producing Beverage
Alcohol, for further discussion of the change.
Two changes were made to the control technology requirement. The
major change is that the best demonstrated technology (BDT) for column
fittings has been expanded to allow the use of gasketed sliding covers
in addition to flexible fabric sleeve seals. The Agency determined that
the requirement for flexible fabric sleeve seals in the proposed standards
was unduely restrictive because these seals are not presently available
on noncontact decks. Refer to Section 2.2.4, Column Fittings, for
further discussion of the change.
The other change to the control technology requirements is a revision
to the flare exit velocity limitations. A recently completed EPA test
program has shown that flares can efficiently remove VOC's at higher
exit velocities when the net heating value of the gas being combusted is
high. Refer to Section 2.2.3, Add-on Control Options, for further
explanation of the change.
Two major changes were made to the monitoring requirements. Vessels
classified as "waste" tanks are subject to revised vapor pressure monitoring
requirements based on biannual physical testing rather than frequent
(perhaps daily) determinations to estimate the vapor pressure of a waste
mixture of indeterminate or variable composition. Refer to Section 2.1.3.4,
"Slop Oil," Wastewater, and Waxy, Heavy Crude Storage Vessels, for
further explanation of the change.
The other major change has been made to the annual inspection
requirement for internal floating roof seal systems. The Agency has
determined that it may not be possible to inspect these vessels without
emptying and degassing the vessel. The inspection requirement has been
revised to allow the owner or operator the option to equip the vessel
with a primary and a secondary seal and conduct an internal inspection
every 5 years. This option is considered equivalent to an annual visual
inspection of a single-seal system. Refer to Section 2.3.3.1, Annual
Visual Inspection, for further discussion.
1-2
-------
1.2 SUMMARY OF IMPACTS OF PROMULGATED ACTION
1.2.1 Alternatives to Promulgated Action
The regulatory alternatives are discussed in Chapter 6 of the
Volume I background information document (BID) for the proposed standards
(EPA-450/3-81-003a). These regulatory alternatives reflect the different
levels of emission control from which one is selected that represents
the best demonstrated technology, considering costs, nonair quality
health, and environmental and economic impacts for volatile organic
liquid storage vessels. Regulatory Alternative III has been revised to
expand the options for fitting controls. As discussed in Section 1.1,
the owner or operator has the option of installing gasketed sliding
covers instead of flexible fabric sleeve seals.
1-2.2 Environmental Impacts of Promulgated Action
The environmental impacts are discussed in Chapter 7 and Appendix D
of the Volume I BID. These impacts have remained unchanged since
proposal. The changes that have been made to the regulation are not of
sufficient magnitude to affect the environmental impact analysis.
Therefore, the Volume I BID now becomes the final Environmental Impact
Statement for the promulgated standards.
1.2.3 Energy and Economic Impacts of the Promulgated Action
The energy impacts of the proposed standards were evaluated in
Chapter 7 of the Volume I BID. It was determined that the control
technologies that are the bases for the regulatory alternatives do not
increase the power or other energy requirements of the VOL storage
vessels. Therefore, no energy impacts are attributed to the proposed or
final standards.
The economic impacts of the proposed standards were evaluated in
Chapter 9 of the Volume I BID. Since proposal, the cost of the liquid-
mounted primary seal has been revised upward to $98.40/meter and the
expected lifetime of the seals revised downward to 10 years (refer to
Section 2.5.3.2, Floating Roof Vessels). The cost of fitting controls
has also been revised (refer to Section 2.2.4, Column Fittings, and
Section 2.5.3.2, Floating Roof Vessels). These changes have been incor-
1-3
-------
porated into the analysis of the fifth year economic impact of the
standards. The fifth year impact still yields a net credit.
1.2.4 Other Considerations
1.2.4.1 Irreversible and Irretrievable Commitment of Resources.
The Volume I BID concluded in Chapter 7 that no long-term environmental
losses would result from the regulatory alternatives. The regulatory
alternatives do not preclude the development of future control options
that would be beneficial to the environment. The analysis of the final
standards remains unchanged in this respect.
1.2.4.2 Environmental and Energy Impacts of Delayed Standards. As
discussed in Chapter 7 of the Volume I BID, the only environmental
impact associated with a delay in proposing and promulgating the
standards would be an increase in VOC emissions from storage tanks
attributable to the construction of new tanks.
1.2.4.3 Urban and Community Impacts. There are no urban and
community impacts attributable to the proposed or promulgated standards.
1-4
-------
2. SUMMARY OF PUBLIC COMMENTS
A list of commenters, their affiliations, and the EPA docket entry
number assigned to each comment are shown in Table 2-1. Twenty-three
letters containing comments on the proposed standards for volatile
organic liquid (VOL) storage vessels and the background information
document (BID) for the proposed standards were received. Significant
comments have been organized into the following six categories:
1. Selection of Affected Facility;
2. Emission Control Technology;
3. Recordkeeping, Reporting, and Inspection Requirements;
4. Modification;
5. Cost Effectiveness; and
6. Miscellaneous.
The comment letters often contained several comments. Each comment is
addressed separately, and the comraenter is identified by the appropriate
docket number.
2.1 SELECTION OF AFFECTED FACILITY
2.1.1 Vapor Pressure and Tank Size Cutoff
Comment: Two commenters (IV-D-18, IV-D-21) said that the proposed
regulation should be consistent with the existing Subparts K and Ka and
limit application of controls to vessels over 151 m3 (=40,000 gal) in
capacity that store liquids with true vapor pressures above 10.4 kPa
(si.5 psia). One commenter (IV-D-15) suggested the 10.4 kPa (1.5 psia)
cutoff be maintained to achieve consistency with State Implementation
Plans (SIP's). Two commenters (IV-D-15, IV-0-18) stated that no data
are presented that show a significant reduction in emissions from vessels
storing liquids with true vapor pressures between 3.5 and 10.4 kPa (0.5
2-1
-------
TABLE 2-1. LIST OF COMMENTERS ON PROPOSED STANDARDS OF
PERFORMANCE FOR VOLATILE ORGANIC LIQUID STORAGE VESSELS
Docket item No.a Commenter/affiliation
IV-D-1 Mr. Richard E. Grusnick
Air Division
Alabama Department of Environmental
Management
1751 Federal Drive
Montgomery, Ala. 36109
IV-D-2 Mr. D. E. Park
Director, Corporate Environmental Affairs
Ethyl Corporation
P.O. Box 341
Baton Rouge, La. 70821
IV-D-3 Mr. Bruce Blanchard
Director, Environmental Project Review
U.S. Department of the Interior
Office of the Secretary
Washington, D.C. 20240
IV-D-4 Mr. Fin Johnson
North Carolina Department of Natural
Resources and Community Development
P.O. Box 27687
Raleigh, N.C. 27611
IV-D-5 Mr. Robert F. Brothers
Director, Regulatory Affairs
Eastman Kodak Company
343 State Street
Rochester, N.Y. 14650
IV-D-6 Mr. Peter W. McCallum
Senior Corporate Environmental Scientist
The Standard Oil Company (Ohio)
Midland Building
Cleveland, Ohio 44115
IV-D-7 Mr. J. K. Walters
Director, Measurement Coordination
American Petroleum Institute
1220 L Street, N.W.
Washington, D.C. 20005
(continued)
2-2
-------
TABLE 2-1. (continued)
Docket item No.a Commenter/affi 1iation
IV-D-8 Mr. W. T. Danker
Manager, Environmental Programs
Environment, Safety, Fire and Health
Chevron U.S.A. Inc.
P.O. Box 7643
San Francisco, Calif. 94120
IV-D-9 Mr. A. H. Nickolaus
Chairman, CTG Subcommittee
Air Conservation Committee
Texas Chemical Council
1000 Brazos, Suite 200
Austin, Tex. 78701
IV-D-10 Mr. M. E. Miller, Jr.
Manager, Environmental Engineering Unit
R. J. Reynolds Tobacco Company
Winston-Salem, N.C. 27102
IV-D-11 Mr. John J. Moon
Manager, Environment and Consumer Protection
Phillips Petroleum Company
Bartlesville, Okla. 74004
IV-D-12 Dr. Geraldine V. Cox
Vice President
Technical Director
Chemical Manufacturers Association
2501 M Street, N.W.
Washington, D.C. 20037
IV-D-13 Mr. Ronald F. Black
Environmental Specialist
Rohm and Haas Company
Engineering Division
P.O. Box 584
Bristol, Pa. 19007
IV-D-14 Mr. Lawrence B. Gotlieb
Assistant General Counsel
Distilled Spirits Council of the United
States, Inc.
1250 Eye Street, N.W., Suite 900
Washington, D.C. 20005
. (continued)
2-3
-------
TABLE 2-1. (continued)
Docket item No.a Coramenter/affi1iation
IV-D-15 Mr. Tom E. Lingafeller
Manager, Environmental Regulatory Activities
for Air
Dow Chemical U.S.A.
2030 Willard H. Dow Center
Midland, Mich. 48640
IV-D-16 Mr. Phillip L. Youngblood
Director, Air Programs
Conoco Inc.
Suite 2136
P.O. Box 2197
Houston, Tex. 77252
IV-D-17 Mr. Walter Roy Quanstrom
General Manager
Environmental Affairs and Safety
Department
Standard Oil Company (Indiana)
200 East Randolf Drive
Chicago, 111. 60601
IV-D-18 Mr. John Prokop
President and General Counsel
Independent Liquid Terminals Association
1133 15th Street, N.W.
Suite 204
Washington, D.C. 20005
IV-D-19 Mr. Kirk F. Sniff
Manager, Environmental Section
Legal Department
The Southland Corp.
P.O. Box 719
Dallas, Tex. 75221
IV-D-20 Mr. Joseph J. Zlogar
Environmental Engineer
Northern Petrochemical Company
P.O. Box 459
Morris, 111. 60450
(continued)
2-4
-------
TABLE 2-1. (continued)
Docket item No.a Commenter/affi1iation
IV-D-21 Mr. L. G. Lund
Manager of Regulatory Affairs
Health, Environment and Safety
Himont U.S.A., Inc.
P.O. Box 1687
Lake Charles, La. 70602
IV-D-22 Mr. L. K. Arehart
Supervisor, Regulatory Analysis
Health and Environmental Affairs Department
Diamond Shamrock Corp.
717 North Harwood Street
Dallas, Tex. 75201
IV-D-23 Mr. Alan T. Roy
Allied Corp.
Fibers Division
Margaret and Bermuda Streets
Philadelphia, Pa. 19137
The docket No. for this project is A-80-51. Dockets are on file at
EPA Headquarters in Washington, D.C. and at the Office of Air Quality
Planning and Standards in Durham, N.C.
2-5
-------
and 1.5 psia). One commenter (IV-D-15) said that no evidence is presented
showing that emissions from these vessels contribute to ozone formation.
Response: The Agency considers the emission reduction attributable
to the standards for tanks storing liquids in this vapor pressure range
to be significant and a contributing factor to ozone formation. There-
fore, consistency with some SIP's or the previous standards (K and Ka)
is not germane. Baseline emissions from fixed roof tanks (both affected
and nonaffected) storing liquids with vapor pressures between 3.5 and
10.4 kPa would total about 135,000 Mg (149,000 tons) in 1988 without
these standards. The emission reduction obtained by best demonstrated
technology (BDT) controls for an estimated 2,000 new fixed roof tanks
storing liquids in this vapor pressure range is estimated to be about
20,000 Mg (22,000 tons) in 1988. The Agency has determined that no
changes are necessary in the final standards on these bases.
Comment: One commenter (IV-D-21) maintained that control of vessels
storing liquids having vapor pressures between 3.5 and 10.4 kPa (0.5 and
1.5 psia) would not contribute greatly to the reduction in emissions and
are less cost effective to control. Another commenter (IV-D-18) said
that the emission reduction attributed to vessels of this size class is
overstated at "for hire" terminals because of the low turnover rate
(approximately 5 per year) on these vessels. This commenter suggested
that floating roof vessels average at least 10 annual turnovers and
fixed roof vessels average at least 50 annual turnovers before becoming
subject to the control requirement.
Response: The Agency evaluated the cost effectiveness of BDT
controls for a typical chemical industry tank (a volume of 606 m3
[160,000 gal], diameter and height of 9.2 m [30 ft]. Tanks with this
volume associated with the.chemical industry typically turnover 60 times
per year. However, in this case, the analysis was conducted assuming
10 turnovers per year to evaluate the cost effectiveness of controls at
a low turnover rate. A molecular weight of 80, a vapor pressure of
6.9 kPa (1.0 psia), and a product value of $360/Mg were also assumed.
The cost effectiveness of BDT is about $1,140/Mg for this case. As
discussed in the response to the previous comment, the emission reduction
achieved by the control of these vessels is significant, and the cost
2-6
-------
effectiveness of the controls on vessels in this vapor pressure range is
reasonable even with lower turnovers.
The commenter (IV-D-18) correctly noted that the Agency's analysis
of tanks in this size range was based on a turnover rate of 50. Because
the number of turnovers does play a role in the cost effectiveness of
BDT controls for fixed-roof tanks, the Agency examined the impact on the
cost effectiveness of BDT controls of low turnover rates in this vapor
pressure range (3.5 to 10.4 kPa). As the number of turnovers decreases,
fixed roof tank emissions will decrease, the emission reduction obtained
by BDT will decrease, and therefore, the cost effectiveness of BOT will
increase.
The average volume of a tank at a "for-hire" terminal is about
3,300 m3 (871,000 gal). The analysis assumed a diameter of about 18.4 m
(60 ft), a height of about 12.2 m (40 ft), a mid-range vapor pressure
(6.9 kPa or 1.0 psia), a molecular weight of 80, and a product value of
$36Q/Mg. Built as a fixed-roof tank, this vessel would emit 8.5 and
6.7 Mg/yr at 5 and 2.5 turnovers per year, respectively. The emission
reductions obtained by constructing a BOT internal floating roof tank in
place of a fixed-roof tank are about 7.0 and 5.2 Mg/yr at 5 and 2.5 turn-
overs, respectively; and the associated cost-effectiveness values are
$490/Mg and $790/Mg at 5 and 2.5 turnovers, respectively.
It should be noted that cost effectiveness is not a measure of the
economic impact of the standards to individual owners. Rather, it is a
measure of the overall cost efficiency for various classes of sources
subject to the standard. The Agency recognizes that there will be
variations in cost-effectiveness values among individual facilities
within a class. Where it is practical to do so, without affecting the
objectivity and enforceability of the standards, the Agency has considered
limiting the scope of the standards to preclude subclasses which have
unreasonable cost-effectiveness values. Nevertheless, variation in
cost-effectiveness values among individual facilities does remain. This
is expected and is not unreasonable.
An exemption based on annual turnovers is not possible without
affecting the objectivity and enforceability of the standards. The
number of turnovers is not constant from year to year and cannot be
2-7
-------
predicted with certainty at the time the vessel is built or reconstructed.
As such, any standards designed to exempt individual vessels which may
have low turnover rates would be impractical both from an enforcement
perspective and from the owner's perspective.
The Agency has concluded that, even in cases of low turnovers, the
control of vessels storing liquids with vapor pressure between 3.5 and
10.4 kPa is reasonable. As discussed above, a cutoff based on turnovers
is not practical even for those instances where cost-effectiveness
values are high. Therefore, because the overall cost of the standards
produces a net credit and because an exemption for the subclass of low
turnover vessels is not practical, no changes have been made to the
proposed cutoffs in these final standards.
Comment: One commenter (IV-D-12) requested that EPA reevaluate the
inclusion of small volume (75 to 151 m3 [20,000 to 40,000 gal]) vessels
by using a range of annual turnovers. According to the commenter, low
turnover rates (less than 10 per year) for these vessels are not cost
effective. Another commenter (IV-D-18) said that the turnover rate for
21,000- to 40,000-gallon vessels in the for-hi re terminal industry is as
low as 2.5 to 4 times per year. This commenter said that EPA's selection
of higher turnover rates results in overstated overall emission reduction
and cost-effectiveness values. A third commenter (IV-D-9) requested
that justification and data be given to support EPA's selection of the
turnover rate (50 per year) used in determining size cutoffs.
Response: Data on turnovers are not available for tanks at terminals
(6 percent of the tank population) or petroleum refineries (approximately
56 percent of the tank population). These data are available for chemical
industry tanks (38 percent of the tank population) in the 75 to 151 m3
size range. This size tank in the chemical industry averages 260 turnovers
per year. Assuming as a worst case that the remaining 62 percent of the
tanks only have four turnovers per year, the average number of turnovers
for a 75-m3 tank would be about 100. Therefore, 50 turnovers per year
is a conservative analytical basis for determining the cutoff.
Although tanks at terminals represent only 5.9 percent of the total
VOL tank population, the Agency examined the cost effectiveness of BDT
controls in a 113 m3 (30,000 gallon) tank for four specific liquids, and
2-8
-------
at two turnover rates (5 and 10 per year). Typical vessels located at
terminals turnover five times per year,according to one industry source.
The results of these analyses are presented in Table 2-2. The cost
effectiveness at five turnovers annually ranged from $930/Mg to $2,430/Mg
and averaged $1,530/Mg. While BDT for these low turnover vessels is
less cost effective than that for vessels with higher turnovers, this
cost effectiveness for certain individual tanks is judged to be reasonable.
The actual cost effectiveness of BDT controls is dependent upon tank-
specific parameters (diameter, height, and volume) and product-specific
parameters (vapor pressure, molecular weight, and chemical formulation)
that cannot be predicted, and may be higher or lower than those presented
in Table 2-2. Because of these variable parameters, an objective exemption
that would exempt only those vessels that always have low turnovers
would be complex and impractical. Therefore, because the average cost
effectiveness is reasonable even at lower turnovers, no changes in the
cutoffs have been made in the final rule.
Comment: One commenter (IV-D-12) suggested that a range of diurnal
temperatures and product molecular weights be used in determining the
cutoff for vessels less than 151 m3 (40,000 gal) in capacity. Commenters
(IV-D-9, IV-D-12) said that the choice of an 11.1°C (20°F) diurnal
temperature change does not represent the norm and advised that an 8.3°C
(15°F) change would be more realistic. Two commenters (IV-D-12, IV-D-18)
noted that the use of a single molecular weight results in overly-
generalized analysis.
Response: Storage facilities located in coastal areas generally
experience lower average diurnal temperature changes than vessels located
further inland. However, an 11.18C (20°F) change represents the national
norm. The average diurnal temperature change for the States of California,
Texas, Louisiana, and New Jersey was calculated to be 10°C (18°F), which
indicates that even in coastal States with a high proportion of the
nationwide storage population, the average diurnal temperature change is
greater than that suggested by the commenters.
The cost effectiveness of a 75 m3 (20,000 gal) tank storing a
typical VOL was examined for sensitivity to use of a 8.3°C (15°F) diurnal
temperature change. The selection of a diurnal temperature change value
2-9
-------
TABLE 2-2. COST EFFECTIVENESS OF BDT CONTROLS FOR A 113-m3 TANK
Compound
Cost effectiveness at
5 turnovers/yr,
$/Mg
Cost effectiveness at
10 turnovers/yr,
$/Mg
n-Pentane
1,580
360
Cyclopentane
1,390
220
Isoprene
2,430
740
Ethyl Ether
930
Credit
Average
1,530
280
2-10
-------
affects the magnitude of breathing losses on the order of 14 percent.
Breathing losses are typically small (approximately 14 percent in this
case) and the impact of changing the temperature factor on the cost
effectiveness of control is small. The cost effectiveness of storing a
typical VOL was a credit when an 11.1°C diurnal temperature change was
assumed, and was also a credit when an 8.36C temperature change was
assumed. Because emission rates are not affected to an excessive extent
by differing diurnal temperature changes and considering the impractic-
ability of basing a national standard on differing local temperature
conditions, no change has been made in the final standards.
The Agency has adequately considered a range of molecular weights
in developing the standards. The responses to other comments on the
proposed standards utilize analyses based on a variety of product molecular
weights. The average cost effectiveness of controls is reasonable under
a wide range of storage situations, including varying molecular weights
and product prices.
Comment: Three commenters (IV-D-7, IV-D-8, IV-D-12) noted that the
American Petroleum Institute (API) is revising Bulletin No. 2518 that
provides the basis for estimates of evaporation loss from fixed roof
tanks. Two commenters (IV-D-7, IV-D-8) also noted that API is conducting
a test program to evaluate evaporation loss from fittings in external
floating roof tanks. According to one commenter (IV-D-7), the new
estimates of emissions from fixed roof vessels could alter the baseline
emission level. The other commenter (IV-D-12) noted that the new fixed
roof emission estimates will include the effects of losses from tanks
with pressure vacuum vents. Because the existing estimates do not
include this variable, the commenter recommended that EPA delay making a
determination on small size cutoffs until the revised estimates are
available. Another commenter (IV-D-8) requested that EPA await the
results of both emission estimate revisions before finalizing the proposed
standards because of possible changes in the baseline emission level.
Response: The Agency considered the benefits of delaying promulgation
of the proposed standards until the results from the API test program
are available. A delay would allow the development of external floating
roof fitting loss equations, which would quantify more precisely the
2-11
-------
emission reduction achievable with the use of best demonstrated technology
(BDT). However, if it is assumed that the emission reduction achieved
by controlling emissions from external floating roof fittings would be
comparable to the reduction achieved with internal floating roof fittings,
this reduction would be very small in comparison to the overall emission
reduction achieved by the standard. Therefore, the benefits of proceeding
with final standards for external floating roof tanks outweigh the
benefits of delaying the standards.
The fixed roof test program has the potential to modify the breathing
loss equations and to quantify the emission reduction obtained by using
vents. However, breathing losses are generally small (approximately
15 percent in the typical storage vessel that is 606 m3 [160,000 gal]),
and a reduction in breathing losses would not make a significant difference
in the cost effectiveness of internal floating roofs as a control device.
The Agency also considered the fact that research test programs of
this nature have uncertain completion dates and that there is no guarantee
that the program will be completed in a timely manner. A proposal
without subsequent, timely promulgation leaves the industry in an uncertain
position with respect to how affected facilities should achieve compliance.
The uncertainty also has a negative impact with respect to State air
programs and the issuance of permits.
Therefore, for the reasons noted above, the Agency has determined
that there are no benefits to be obtained from delaying the standards
and has decided to proceed. It should be noted, however, that data from
a completed test program will be incorporated into the next review of
this NSPS and into the next revision of the EPA Publication AP-42.
2.1.2 Vapor Pressure Determination
Comment: Two commenters (IV-D-5, IV-D-9) suggested that EPA use
the annual average temperature instead of the maximum monthly average
temperature to determine maximum true vapor pressure. The commenters
said that this would be consistent with EPA's procedure for analyzing
emissions and cost effectiveness. A third commenter (IV-D-18) requested
clarification regarding which temperature should be used. The commenter
also questioned the fairness of requiring controls on vessels containing
2-12
-------
liquids that may exceed the vapor pressure threshold only one day per
year or less.
Response: The EPA agrees that affected liquids may have vapor
pressures that are below the cutoffs for much of the year, but also
notes that nonaffected liquids may have true vapor pressures above the
cutoffs for portions of the year such as daylight hours during summer
months. In developing the applicability provision, EPA realized that
basing applicability on maximum instantaneous vapor pressure would
result in the broadest applicability and, therefore, the largest emission
reduction. This approach could cause planning problems for the industry
because they might not be able to adequately predict which vessels would
be affected. Because industry may not be able to account for particularly
hot days adequately, the instantaneous vapor pressure was rejected as
the basis of applicability.
The next applicability format examined was an annual average vapor
pressure. This format has the advantage that it is in line with the
emission calculation methodologies used in the BID. However, vapor
pressures of VOL's are higher in the warmer, summer months, when ambient
ozone levels are highest. If applicability were based on the annual
average vapor pressure, vessels would not come under the standards even
though they were storing liquids with true vapor pressures greater than
3.5 kPa. These vessels would then emit significant quantities of VOC
during the summer when ambient ozone levels are highest. Therefore, EPA
decided to examine a shorter time frame that would broaden the applic-
ability of the standards, particularly during the summer.
An applicability based on maximum monthly average vapor pressure
was selected because this would have a broader applicability than annual
averages without the planning problems associated with an applicability
based on instantaneous vapor pressure and would base applicability on
the contribution to VOC emissions when ozone levels are highest. The
commenters are correct in noting that as the vapor pressure of the
stored liquid decreases, the cost effectiveness of BDT controls increases.
The cost effectiveness of BDT controls for a typical chemical storage
tank with an average annual vapor pressure of 3.5 kPa (0.5 psia) would
vary from about $250/Mg to $600/Mg depending on the assumed product
2-13
-------
value ($360/Mg and $695/Mg, respectively). However, if the maximum
monthly average vapor pressure exceeds the cutoff and the annual average
vapor pressure is 1.75 kPa (0.25 psia), the cost effectiveness of BDT
controls for this tank would range from $1,100/Mg to $1,400 Mg. This is
judged to be reasonable and, therefore, the unit time basis of vapor
pressure determination in the promulgated standards remains unchanged
from that in the proposed standards.
2.1.3 Special Exemptions
2.1.3.1 Horizontal and Underground Vessels. The fact that horizontal
and underground storage vessels were not specifically excluded from the
proposed standards prompted several comments.
Comment: Two commenters (IV-D-1, IV-D-16) requested that underground
storage vessels at gasoline service stations be exempted from the rulemaking.
One commenter (IV-D-1) stated that it would be an unnecessary recordkeeping
burden for both the operators of smaller affected vessels not subject to
the control requirements and the regulatory agencies that would have to
keep the records. Both commenters stated their belief that it was not
EPA's intent to include underground storage vessels at gasoline service
stations as affected facilities.
Response: Commenters are correct in their assertion that EPA did
not intend to affect vessels at gasoline service stations with these
standards. Consequently, no evaluation of the possible economic impact
of these standards on retail gasoline marketers was performed. Emissions
from retail gasoline marketers are part of the gasoline marketing source
category, as well as part of the VOL storage category. The decision as
to whether to regulate emissions from these vessels will be made when
standards for the gasoline marketing source category are developed. The
Agency has decided to specifically exempt storage vessels at retail
gasoline service stations from the final standards.
Comment: Commenters (IV-D-7, IV-D-16) suggested that underground
storage vessels be excluded when the volume of liquid added to and taken
from the tank in a year does not exceed twice the volume of the vessel.
A provision of this nature is included in Subparts K and Ka, and the
commenters requested that it also be included in Subpart Kb.
2-14
-------
Response: It is impracticable to control emissions from underground
tanks with internal floating roofs. Emissions from these tanks can be
controlled with vapor recovery or disposal systems designed and operated
in compliance with these standards. While the cost effectiveness of
vapor recovery or disposal systems in these isolated instances may be
high, cost effectiveness is not a measure of economic impact. Rather,
it is a measure of cost efficiency in individual cases. The average
cost effectiveness of these standards is a net credit. Additionally,
above ground internal or external floating roof tanks equipped with the
controls required by the standards could be constructed in lieu of
underground storage vessels. Therefore, the Agency has decided not to
include this exemption in the promulgated standards. It should be noted
that the exemption does continue for vessels constructed after the date
of proposal for Subpart K but prior to the proposal of Subpart Kb.
Comment: One commenter (IV-D-5) contended that the BID does not
support the inclusion of underground storage vessels, particularly those
smaller than 100 m3 (26,000 gal) in capacity. The commenter also noted
that emissions from these vessels are excessively costly to control.
According to the commenter, installations storing VOL for use in manu-
facturing operations may need one or two horizontal, underground tanks
as large as 95 m3 (25,000 gal) to store material received from railroad
tank cars that have capacities up to 75 m3 (20,000 gal). The commenter
recommended that an exemption be included in the proposed standard for
underground storage vessels with capacities less than 100 m3.
Response: Further discussions with the commenter revealed that the
commenter's primary concern was that some manufacturing plants may not
have adequate space to install aboveground tanks. According to Factory
Mutual Research, adequate spacing of tanks is necessary to reduce the
possibility of the spread of fire from the tank initially involved to
exposed structures or adjacent tanks. For example, a 75-m3 (20,000-gal)
tank would have to be placed at least 7.5 to 15 m (25 to 50 ft) away
from the buildings depending on the flamability of the liquid being
stored. The minimum tank-to-tank spacing is one-half the diameter of
the largest tank; 4.3 m (14 ft) in the case of adjacent 75 m3 tanks. In
2-15
-------
contrast, underground tanks may be located as close as 1.5 m (5 ft) to
building foundations and 0.6 m (2 ft) from other tanks and pipelines.
Above ground tanks are a proven and safe method of storage, and
there appear to be no technical reasons why above ground tanks could not
be installed in place of underground tanks when space permits. Further-
more, it is expected that space would not be a problem at new plants
because they can be designed to allow sufficient space for above ground
tanks. However, EPA cannot predict which existing facilities may have
spacing problems in installing new above ground tanks. In cases where
space is a problem, the owner or operator may install underground tanks
equipped with add-on controls allowed by § 60.112b(a)(3).
The cost effectiveness of controlling emissions from a 113-m3-capacity
underground tank is about $2,100/Mg. This assumes the tank undergoes
10 turnovers per year, which is typical for the commenter's industry.
While this is higher than the cost effectiveness of BDT control (i.e.,
floating roof) in comparable tanks ($280/Mg), it is not necessarily
unreasonable because the overall cost of the standard is a net credit.
The EPA concluded that the standards contain sufficient flexibility to
be achievable, and therefore, the final standards will not be revised to
include a blanket exemption for underground storage vessels because
adequate substitutes (above ground vessels or underground vessels equipped
with add-on controls) exist.
Comment: One commenter (IV-D-22) noted that horizontal tanks are
used widely in the synthetic organic chemical manufacturing industry
(SOCMI). The commenter said that because floating roofs cannot be used
in these tanks, there is a problem in applying the proposed standards to
them.
Response: The standards provide three fundamentally different
methods of compliance:
1. External floating roof tanks equipped with liquid-mounted or
shoe primary seals, and a rim-mounted secondary seal;
2. Internal floating roof tanks equipped with liquid-mounted
primary seals or vapor-mounted primary and secondary seals; and gasketed
fittings; or
2-16
-------
3. A 95-percent-effective vapor control system.
Horizontal tanks are typically small (volumes rarely exceed 113 m3
(30,000 gallons), and because external floating roof tanks are rarely
smaller than 492 m3 (130,000 gal), these horizontal tanks could not be
constructed as new external floating roof tanks. However, the other
options allowed by the standard are suitable for vessels in the size
range of horizontal tanks. In subsequent discussion, the coimnenter
agreed that vertical tanks equipped with internal floating roofs could
be used in place of horizontal tanks; although in some instances, such
as separation processes, horizontal tanks were advantageous.
Additional information was obtained from the State of Texas on the
issue. Texas requires equipment similar to BDT (internal floating
roofs) for all new storage vessels with capacities of 95 m3 (25,000 gal)
or greater storing liquids with vapor pressures of 3.5 kPa (0.5 psia) or
greater and, thus, currently requires controls on vessels of concern to
the commenter. Texas Air Control Board (TACB) personnel have stated
that, in their permitting experience, there are very few circumstances
in which the tanks must be horizontal. If a horizontal tank is used,
the TACB generally requires add-on control systems (carbon adsorption or
thermal oxidation). Previous studies of storage in the chemical industry
indicate that add-on control systems are cost effective (less than
$1,000/Mg) for tanks with volumes less than 151 m3 (40,000 gal) storing
liquids with high vapor pressures. For example, as shown in Table 2-3,
the average cost effectiveness of a 95-percent-efficient condenser for
chloroform storage at chloroform production facilities is $630/Mg. This
issue was further analysed by examining the cost effectiveness of
controlling a 113 m3 (30,000 gallon) horizontal tank, as a function of
turnover rate and filling rate. As shown in Tables 2-4 through 2-6 the
average cost effectiveness ranged from $62Q/Mg to $1,4G0/Mg. This range
is judged to be reasonable. The standards are achievable and, in many
cases, are cost effective even if add-on controls are adopted. Therefore,
no exemption for these vessels has been incorporated into the final
rule.
2-17
-------
TABLE 2-3. COST EFFECTIVENESS OF 95-PERCENT-EFFICIENT CONDENSER
FOR CHLOROFORM STORAGE AT CHLOROFORM PRODUCTION FACILITIES
(July 1982 $/Mg)
Plant
Location
Cost
effectiveness
$/Mg
Diamond Shamrock
Belle, W. Va.
2,800
Dow
Freeport, Tex.
L3,300
Dow
Plaquemine, La.
a
Li nden
Moundsville, W. Va.
1,370
Stauffer
Louisville, Ky.
On standby
Vulcan
Geismar, La.
—
Vulcan
Wichita, Kans.
(230)b
Average
630
Indicates no control required.
( ) Indicates a net credit due to chloroform recovery.
2-18
-------
TABLE 2-4. COST EFFECTIVENESS FOR 95-PERCENT-
(MASS) EFFICIENT CONDENSER3
Total Cost
Vapor Installed annual- Emission effec
pressure capital ized Product reduc- tive-
Compound
at 70°F,
psia
cost, Dec.
1982 $D
cost.
$/yr
value,
$/Mg
tion,
Mg/yr
ness
$/Mg
n-Pentane
8.433 v
15,800
5,100
449d
4.1
780
Cyclopentane
5.240
11,400
3,620
550e
2.3
1,000
Isoprene
9.668
17,100
5,570
530f
4.6
670
Ethyl Ether
8.702
15,800
5,100
1,0149
4.3
160
Average
620
Basis: 30,000 gallon horizontal tank; 10 turnovers/yr: filling rate =
ulOO gal/min.
Based on Appendix C of Organic Chemical Manufacturing Volume 3;
EPA-450/3-80-025. T - -20°F line was used for costing.
Based on 6 percent maintenance, 16.3 percent capital recovery, 5 percent
miscellaneous capital costs. Electricity was averaged nationally at
$78.65/1,000 kWh. The condenser was operated 12 hours/day and was
destimated to use 1.5 kW/ton refrigerant.
Quoted December 21, 1984, by Ashland Chemical Company in December 1984
dollars.
Quoted December 21, 1984, by Phillips Chemical Company in December 1984
fdollars.
^Quoted December 21, 1984, by Research Triangle Institute in 1982 dollars.
gQuoted from Oecember 17, 1984, Chemical Marketing Reporter.
2-19
-------
TABLE 2-5. COST EFFECTIVENESS FOR 95-PERCENT-
(MASS) EFFICIENT CONDENSING SYSTEMS
Compound
Vapor
pressure
at 70°F,
psia
Installed
capital
cost, Dec.
1982 $D
Total
annual-
ized
cost.
$/yr
Product
value,
$/Mg
Emission
reduc-
tion,
Mg/yr
Cost
effec-
tive-
ness
$/Mg
n-Pentane
8.433
15,800
5,100
449d
3.2
1,160
Cyclopentane
5.240
11,400
3,620
550e
1.7
1,546
Isoprene
9.668
17,100
5,570
530f
3.6
1,000
Ethyl Ether
8.702
15,800
5,100
1,0149
3.4
510
Average
980
aBasis: 30,000 gallon horizontal tank; 5 turnovers/yr; filling rate =
.100 gal/mi n.
Based on Appendix C of Organic Chemical Manufacturing Volume 3;
EPA-450/3-80-025. T = -20°F line was used for costing.
cBased on 6 percent maintenance, 16.3 percent capital recovery, 5 percent
miscellaneous capital costs. Electricity was averaged nationally at
$78.65/1,000 kWh. The condenser was operated 12 hours/day and was
.estimated to use 1.5 kW/ton refrigerant.
Quoted December 21, 1984, by Ashland Chemical Company in December 1984
dollars.
Quoted December 21, 1984, by Phillips Chemical Company in December 1984
fdollars.
Quoted December 21, 1984, by Research Triangle Institute in December 1982
dollars.
gFrom December 17, 1984, Chemical Marketing Reporter.
2-20
-------
TABLE 2-6. COST EFFECTIVENESS FOR 95-PERCENT-
(MASS) EFFICIENT CONDENSING SYSTEMS
Vapor
pressure
at 70°F,
Compound psia
n-Pentane 8.433
Cyclopentane 5.240
Isoprene 9.668
Ethyl Ether 8.702
Average
Total
Installed annual-
capital ized
cost, Dec. costA
1982 $° $/yr
25,700 8,600
19,100 6,280
28,100 9,470
25,700 8,600
Cost
Emission effec-
Product reduc- tive-
value, tion, ness
$/Mg Mg/yr $/Mg
449d 4.1 1,650
550e 2.3 2,200
530f 4.6 1,510
1,014® 4.3 970
1,400
aBasis: 30,000 gallon horizontal tank; 10 turnovers/vr: filling rate =
h200 gal/min.
Based on Appendix C of Organic Chemical Hanufacturing Volume 3;
EPA-450/3-80-025. T - -20°F line was used for costing.
Based on 6 percent maintenance, 16.3 percent capital recovery, 5 percent
miscellaneous capital costs. Electricity was averaged nationally at
$78.65/1,000 kWh. The condenser was operated 12 hours/day and was
.estimated to use 1.5 kW/ton refrigerant.
Quoted December 21, 1984, by Ashland Chemical Company in December 1984
dollars.
Quoted December 21, 1984, by Phillips Chemical Company in December 1984
fdollars.
Quoted December 21, 1984, by Research Triangle Institute in December 1982
dollars.
gFrom December 17, 1984, Chemical Marketing Reporter.
2-21
-------
2.1.3.2 Facilities Producing Beverage Alcohol.
Comment: One commenter (IV-D-14) requested that the EPA grant an
exemption from the proposed standards for vessels used to store non-
industrial, distilled beverage alcohol. The commenter requested the
exemption for the following reasons: (1) producers of distilled spirits
are insignificant sources of VOC emissions; (2) the suggested control
technology would either be extremely damaging to the product as a food
item or would be proscribed by existing federal regulations; and (3) the
costs and other problems that would result from implementation of the
proposed standards would violate Executive Order 12291.
Response: The Agency concurs with the commenter that the proposed
control technologies required by these standards could contaminate
beverage alcohol resulting in a product with little or no market value.
Also, because beverage alcohol is not part of the 50CMI source category,
t
nor is it a petroleum liquid, storage vessels containing beverage alcohol
are exempt from the final standards. However, any storage vessels that
are used to store nonbeverage, fermented products are subject to the
standards if they are found to be affected facilities.
2.1.3.3 Vessels located on the Outer Continental Shelf (PCS).
Comment: One commenter (IV-D-3) noted that the U.S. Department of
the Interior, not EPA, is authorized to regulate air emissions from oil
and gas operations on the OCS. It was recommended that VOC and petroleum
liquid storage vessels located on the OCS be exempted as affected
facilities for this reason.
Response: According to the Outer Continental Shelf Lands Act
Amendments, the Secretary of Interior is authorized to issue air emission
regulations for storage vessels located on the OCS. However, the Agency
does not construe this authority as exclusive jurisdiction. It is also
noted in the OCS Act that "the Secretary shall cooperate with relevant
departments and agencies of the Federal Government and of the Affected
States." Under the authority of the Clean Air Act, the Agency is required
to control air emissions from new, modified, or reconstructed sources.
Because of this authority the Agency has decided not to exempt specifically
vessels located on the OCS from the final standards.
2-22
-------
In any event, the final standards do exempt vessels that store
liquids prior to custody transfer as defined in § 60.111b, unless the
vessels have a design capacity greater than 1,589.7 m3 (420,000 gal). A
study of offshore oil and gas production facilities indicated that if a
platform has storage vessels, those vessels store liquids that have not
undergone custody transfer. Therefore, the Agency believes DCS vessels
are exempt from the standards via the custody transfer provision.
2.1.3.4 "Sloo Oil." Wastewater, and Waxy, Heavy Crude Oil Storage
Vessels.
Comment: One commenter (IV-D-6) requested an exemption to the
recordkeeping requirements for tanks used to store a mixture of different
products ("slop oil"). The commenter said that the constantly changing
nature of the products and the associated vapor pressure in these vessels
would necessitate physical testing to determine vapor pressure as required
in the proposed standards. According to the commenter, these vessels
represent a small portion of the vessels at any one facility, and little
benefit, if any, would be gained by including them in the vapor pressure
recordkeeping requirements.
Response: The purpose of the vapor pressure determination is to
distinguish between the three possible classes of VOL's that are of
concern:
1. Those liquids with vapor pressures greater than or equal to a
control cutoff (27.6 kPa for vessels with capacities of 75 ¦» or greater
and 3.5 kPa for vessels with capacities of 151 ¦» or greater);
2 Those liquids that are exempt from all vapor pressure record-
keeping provisions (§ 60.116(c)) of the standards (less than 15 kPa for
vessels with capacities between 75 and 151 •>, and less than 1.75 kPa
for vessels with capacities of 151 i»3 or greater); and
3 Those liquids for which monitoring, but not emission control,
is required (between 15 and 27.6 kPa for vessels with capacities ranging
from 75 to 151 m* and between 1.75 to 3.5 kPa for vessels with capacities
of 151 m3 or larger).
For most chemical and petroleum products, the class to which a
given liquid belongs will be obvious, for instance, the vapor pressure
2-23
-------
of No. 2 fuel oil will not exceed 1.75 kPa, and, therefore, vessels
storing this liquid would be exempt from all provisions of the standards.
However, waste tanks with constantly changing mixtures pose a
different issue. While a range of possible vapor pressures will be
known, constant minor fluctuations in composition will prevent the
determination of the actual vapor pressure without extensive (perhaps
daily) testing. However, these fluctuations generally are not so large
that under normal operating conditions, large daily changes
in vapor pressure would be expected. Extensive testing of these liquids
would be unduly burdensome to industry without providing a correspond-
ing benefit. Therefore, EPA sought an alternative that would preserve
the intent of the requirement without being unreasonably burdensome.
Prior to construction of the vessel, the range of likely liquid
compositions will be known, as will the maximum monthly average storage
temperature. Given these, it is possible to estimate the vapor pressure
of the mixture by Raoult's law:
Pt ¦ 2 Vn
where P^. = the total vapor pressure
Pn = the vapor pressure component
Xn = the mole fraction of a component
As with all other liquids, if the anticipated liquid composition with
the highest vapor pressure is below the monitoring cutoffs, the vessel
would be exempt from the standards, and no additional monitoring is
needed.
For these types of anticipated liquids, the provisions for monitoring
have been changed from those proposed. If the anticipated liquid composi-
tion is above the cutoff for monitoring but below the cutoff for controls,
the standards require an initial physical test of the vapor pressure and
a physical test at least once every 6 months thereafter. This testing
is not costly (less than $100) and would serve the intent of the proposed
standards without being burdensome. Records of the results will be
kept, but reports will be required only in the event the vapor pressure
of the stored liquid exceeds the threshold for controls.
2-24
-------
Comment: One commenter (IV-D-16) requested that wastewater holding
vessels be exempted as affected facilities. The commenter noted that
use of holding vessels to retain wastewater after the organic liquids
have been removed in an oil-water separator is common practice in the
petroleum industry. According to the commenter, these holding vessels
contain liquids with low vapor pressures such that operational monitoring
and emission controls are inappropriate.
Response: The commenter has raised two issues. The first is the
control of wastewater tanks. The volume and vapor pressure cutoffs have
been selected so that controls are cost effective, and the control of
vessels and stored liquids meeting these criteria is reasonable regardless
of liquid type. Therefore, no exemption for wastewater tanks has been
adopted.
The second issue is the vapor pressure monitoring requirements for
these tanks. The previous comment and response deal with revised require-
ments for waste tanks. These revised requirements are also reasonable
for wastewater tanks and would reduce the recordkeeping burden for these
tanks. Therefore, the revised monitoring requirements will apply to all
waste tanks, including wastewater tanks.
Comment: One commenter (IV-D-7) requested that vessels storing
waxy, heavy crude oils be exempt from secondary seal requirements.
According to the commenter, secondary seals are inappropriate for use
with waxy, heavy crude oils due to potential operational and safety
concerns.
Response: Tank vendors were contacted for further information on
this issue. The Agency has determined that the storage of heavy, waxy
crudes is a significant problem for EFR's equipped with secondary seals.
The main problem is that solid wax adheres to the shell as the deck is
lowered. The wax melts as the shell wall heats during the day, and runs
down the wall, over the secondary seal, and onto the deck. The wax is
flammable and presents a fire hazard, plugs drains, and fouls fire
protection equipment. It can also adhere the secondary seal to the
shell, thereby causing damage to the seal when the deck position changes.
The relatively stiff secondary seal can also scrape solid wax off the
2*25
-------
shell and throw it onto the deck as the deck Is raised. As the wax
melts, it can cause the problems mentioned above.
However, IFR's can be used to store heavy, waxy crudes. According
to one vendor, IFR seals are softer and do not scrape wax off the shell.
In addition, IFR's may be equipped with roof-mounted fire protection
systems that will not be fouled by wax on the deck. Another vendor said
that heavy, waxy crudes are commonly stored in IFR's equipped with steam
coils and an insulation jacket. This system keeps the wax in solution
and eliminates the problem of wax buildup on the tank wall or seal
system. Because these standards are achievable for heavy, waxy crudes
by using IFR's, the Agency has determined that no exemption for these
liquids will be included in the standards.
2.1.3.5 Negligibly Photochemically Reactive Liquids.
Comment: Two commenters (IV-D-13, IV-D-15) requested that
negligibly photochemically reactive liquids such as 1,1,1-trichloroethane
and methylene chloride be exempted from the proposed standards. One
commenter (IV-D-18) requested that these chemicals be listed as
negligibly photochemically reactive. One commenter (IV-D-15) claimed
that previous VOC standards have exempted these liquids. Another
commenter (IV-D-23) recommended that proposed standards list all VOC
compounds to be included and exempt VOC's that are negligibly photo-
chemically reactive. The commenter said that the inclusion of negligibly
photochemically reactive liquids could increase the control costs. One
commenter (IV-D-3) suggested that a listing of those compounds that are
defined as volatile organic compounds be referenced in the text.
Response: Previously, the EPA has determined that the following
11 compounds are negligibly photochemically reactive:
1. Methane;
2. Ethane;
3. Methylene chloride (dichloromethane);
4. Methyl chloroform (1,1,1-trichloroethane);
5. Trichlorofluoromethane;
6. Dichlorodifluoromethane;
7. Chlorodifluoromethane;
8. Trifluoromethane;
2-26
-------
9. Trichlorotrifluoroethane;
10. Dichlorotetrafluoroethane; and
IX. Chloropentafluoroethane.
Because these compounds do not significantly contribute to the
formation of ozone, the Agency agrees that the control of vessels storing
these compounds would not reduce ambient ozone levels and has, therefore,
exempted vessels that store only these compounds from the final rule.
Because this list of compounds will change from time to time as research
continues, no list of exempt compounds is included in the final rule.
Rather the approach that has been taken is to exempt each compound that
has previously been declared negligibly photocheaically reactive in
previous Federal Register notices.
Regarding the request that a list of VGL's and/or VOC's be provided
as part of the final standard, it should be noted that it is the Agency's
position that all organic compounds are photocheaically reactive and,
therefore, potentially subject to this NSPS, until such time as they are
declared negligibly photochemically reactive. In essence, the commenters
are requesting that a list of all organic compounds except those
determined to be negligibly photchemically reactive be provided as part
of the final standards. The Agency sees no reason to add such a list to
the final standards and feels that the provisions determining
applicability of the NSPS are adequate without it. Therefore, no such
list has been incorporated into the final rule.
2.1.3.6 Production and Process Vessels.
Comment: One commenter (IV-D-23) requested that production and
process vessels having an intermediate function, not raw material or
product storage, be exempted from the proposed standards. The commenter
said that estimates of working losses for these vessels were incorrectly
based on "total throughput." The commenter said that "net throughput"
is a more realistic measure of turnovers. The commenter stated that the
control technology may not be cost effective for production and process
vessels. The commenter recommended that EPA reevaluate the standards
using net throughput.
Response: The EPA agrees that total throughput (tank volume divided
into annual liquid throughput) may not accurately reflect the actual
2-27
-------
change in liquid level (net throughput), which is an underlying mechanism
of working losses. The commenter provides an example of a 75 m3 tank
that would undergo 689 turnovers per year as measured by total throughput,
but only 87 turnovers per year as measured by net throughput. For the
specific tank cited by the commenter, working losses would be 10.5 Mg/yr
as calculated with total throughput versus 3.2 Mg/yr as calculated with
net throughput.
In evaluating this issue, the EPA first examined the cost effec-
tiveness of controlling the sample tank cited by the commenter. Because
the commenter did not fully specify the necessary tank parameters, the
emission reduction obtained by BDT controls was evaluated for working
losses, based on net throughput, and was assumed to be 90 percent. To
be conservative, a welded steel deck with Teflon®, liquid-mounted,
primary seals was costed as the control technology. Tank diameter was
assumed to be 4.5 m (15 ft) and product value was assumed to be $360/Mg.
The calculated cost effectiveness for controlling this tank is about
$93Q/Mg. In the case cited by the commenter, the cost effectiveness of
controls is still reasonable even though the use of net throughput
reduces estimated working losses to 30 percent of the losses based on
total throughput.
In previous studies by the EPA, model plants were developed for
storage associated with selected chemical process. Some of these models
contain "constant level" tanks (tanks with high total throughputs but
low net throughputs). These tanks were evaluated for control, and the
results are presented in Table 2-7. The average cost effectiveness was
found to be $354/Mg. Although there are instances where the cost-
effectiveness value is very high, the average cost effectiveness of
controlling constant level tanks is reasonable. These costs are also
representative of production and process tanks that are operated typically
as constant level tanks. Therefore, the final standards do not provide
an exemption for constant level tanks.
2-28
-------
TABLE 2-7 COST EFFECTIVENESS OF BDT IN CONSTANT LEVEL TANKS
Product
Tank
volume, m3
Vapor
pressure, kPa
Turn-
overs
Emission
reduction,
Mg/yr
Product
value,
$/Mg
Cost
effec-
tiveness,
$/Mg
Acetic acid
1,200
4.447 @ 40°C
2
1.453
551
9,395^
Acetic acid
455
4.447 @ 40°C
6
0.5782
551
17,195°
1,2-di chloroethane
1,140
11.36 @ 27ttC
6
5.252
375
1.145*
1,2-di chloroethane
3,800
11.36 @ 27°C
12
27.69
375
!4°j
Methanol
1,890
16.92 @ 27°C
6
3.911
180
959
Methanol
757
11.81 @ 206C
6
1.175
180
2,551<|
Cumene
3,785
6.275 @ 71.1°C
6
11.86
507
18d
Benzene
230
7.702 @ 20°C
6
0.3044
364
3,781 d
Cyclohexane
213
70.67 @ 68°C
6
6.846
441
(credit).
Ett^} aery late
6,941
4.435 @ 21.4°C
6.08
13.07
1,334
(credit)
Styrene
900
3.448 @ 50°C
6
1.584
772
l,^16d
Ethanol
4,396
5.978 @ 20.3°C
6.08
5.051
603
723d
Acetone
378.5
28.84 g 26.7°C
6
2.206
507
549 d
Acrylonitrile
2,500
16.41 ° 27*C
6
8.282
1,003
(credit)j
Crude product lf
757
5.377 @ 16°C
6
19.177
400
(credit}
Crude product 2
378
2.62 @ 38°C
6
6.20
400
553d
AVERAGES
1,856
6.14
7.165
534
3549
^Prices from December 3, 1984, Chemical MarketincuReporter.
Based on stainless steel welded deck with Teflon seal.
.Based on mild steel welded contact deck with liquid-mounted seal.
Based on aluminum noncontact deck with liquid-mounted seal.
^Assumed product composition (1/3 each) crude methylene chloride, chloroform, carbon tetrachloride.
Assumed for carbon tetrachloride.
qA * .• _ 1 net annual costs
aAveraae cost effectiveness = ^—-*—> t—r-»—
s vccod z emission reduction
-------
2.2 EMISSION CONTROL TECHNOLOGY
2-2.1 External Floating Roof Vessels (EFR's)
Comment: Two commenters (IV-0-18, IV-D-21) noted that hydroscopic
VOL's cannot be stored in EFR's because of the potential contamination
of the liquids by water.
Response: The Agency agrees that not every liquid covered by the
standard is appropriate for storage in external floating roof tanks.
Such liquids may be stored in internal floating roof tanks (ventilated
or nonventilated) or in fixed roof tanks equipped with vapor recovery or
disposal systems.
Comment: One commenter (IV-D-18) stated that industry experience
with secondary seals is limited and that there is uncertainty regarding
the effective life of the seals. The commenter also requested that EPA
acknowledge a potential safety hazard from the formation of a vapor
space between the primary and secondary seal.
Response: The commenter is correct in stating that there may be
some uncertainty in the lifetime of secondary seals; the actual lifetime
may be longer or shorter than the 10 years estimated at proposal.
Comments received on the draft Control of Volatile Organic Compound
Emissions from Volatile Organic Liquid Storage in Floating and Fixed-Roof
Tanks (CTG), August 1983 suggested that vapor-mounted primary seals on
an internal floating roof would have a lifetime of 10 years or more in
the more severe chemical services. Because the construction of these
seal systems is similar to that of secondary seals used on external
floating roof tanks, it is reasonable to assume a 10-year life on secondary
seals for external floating roof tanks as an average.
The Agency has determined that adequate technology exists and is
commonly employed to operate external floating roof tanks with double
seals in a safe manner. Although specific data are not available for
EFR's equipped with double-seal systems, the fire and explosion hazard
is greatly reduced by use of floating roofs in comparison to fixed roof
tanks because the floating roof eliminates vapor space. The reduced
hazard is substantiated by Factory Mutual Research fire loss experience
for a 13-year period from 1962 to 1974. Fixed roof tanks were involved
2-30
-------
in 53 percent of all losses; whereas, floating roof tanks were involved
in only 34 percent. Significantly, 47 percent of fixed roof tanks were
totally destroyed, and an additional 50 percent suffered major damage.
Only 12 percent of floating roof tanks were totally destroyed, and
36 percent suffered roof, ring, or shell damage. Most fires in floating
roof tanks were extinguished by portable foam or water hose streams
before serious damage occurred. Based on this, the Agency concludes
that there are no safety hazards associated with BOT beyond those commonly
accepted by industry-
2.2.2 Internal Floating Roof Vessels (IFR's)
Coranenters representing the chemical industry were concerned that
IFR's and som liquid-mounted primary seals are not an appropriate or
proven, safe control technology for the range of chemicals stored.
Comment: Three co-enters (IV-D-9, IV-D-15, IV-D-22) noted that
vented IFR's may not be the best choice of storage vessel in cases where
the stored liquid must be protected from moisture or oxygen. One commenter
(IV-0^22) cited chemical products such as chlorinated solvents that may
be contaminated by exposure to moist ambient air. One counter (IV-D-15)
noted that vented IFR's would greatly enhance the cost of inert gas
pads.
Response- These comments are based on the premise that the internal
floating roof tank required in § 60.112b(a)(l> must be vented. The
Agency agrees that ventilating tanks storing liquids which must be
protected fro. contact with anient air is not wise. Neither tta proposed
nor the promulgated standards require the interna, floating roof tank to
be vented Internal floating roof tanks may be ventilated or nonventilated,
padded or'unpadded, according to the preference of the owner or operator,
without affecting the compliance status of the tank. Therefore, the
. . finatina roof will impose no additional
requirement for an internal floating roo **
. x- uiamc cafetv problems or gas padding costs over normal
contamination problems, sarety .
industry practice. No changes to the standards have been made.
Comment: Three commented (IV-D-9, IV-D-15, IV-D-22) stated that
use of~the_floating roof Itself is incompatible with storage of highly
4 n.nwc According to two of the commentsrs (IV-D-15, IV-D-22),
corrosive liquids. Accoramy u n •*». * *•
• „ Hamanp the vessel may either be lined with plastics,
to prevent corrosion damage,
2-31
-------
fluoropolymers, or synthetic materials, or it may be constructed with
fiberglass reinforced plastic. These commenters state that such
materials are unable to withstand the abrasion which is incipient in the
operation of floating roofs. One commenter (IV-D-15) recommended that
an alternate control technology be defined or that lined tanks be
exempted.
Response: Regarding the abrasion of the tank liner, internal
floating roof seals are typically made of soft materials and are softer
than common liners. The seals do not exert much compressive force
against the tank sidewall. The anticipated point of wear would be the
seal and not the tank liner. Internal floating roofs have been installed
in lined tanks and have operated properly without excessive wear to the
liner. Therefore, no exemption for lined tanks has been incorporated
into the final standards. [The response concerning fiberglass reinforced
plastic tanks is being developed.]
Comment: One commenter (IV-D-9) questioned whether ventilation in
IFR's is adequate to prevent an explosion hazard. Another commenter
(IV-D-21) said that the proposed standards should recognize that the use
of IFR's could promote the formation of explosive vapor mixtures above
the floating roof. Another commenter (IV-D-22) noted that, by design,
the product stored in an IFR is isolated from any roof-mounted deluge
system, thus reducing the probability of early control of any fire that
occurs.
Response: The Agency has determined that the proposed standards do
not pose a safety hazard. A representative of the Texas Chemical Council
(IV-D-9) has stated that his company's (I. E. Dupont) safety personnel
had reviewed the proposed standards and did not believe that the required
controls would pose a hazard. Data from vendors indicate that in tests
on a noncontact internal floating roof in vented tanks storing a wide
variety of products, the lower explosive limit was never reached. While
the vapor space above the deck in a vented IFR may become explosive
during periods of rapid fill, the vapor space in unblanketed fixed roof
tanks will become explosive during emptying when air is drawn into the
tank. Based on these facts, properly vented IFR's should pose no unusual
hazard over unblanketed fixed roof tanks.
2-32
-------
As stated previously, there is no requirefllent that IFR's be vented.
Vessels equipped with internal floating roofs nay be vented or nonvented,
and blanketed or unblanketed.
Regarding isolation from any roof-mounted deluge system, vendors of
internal floating roofs and Factory Mutual Research Standards have
stated that foam distribution systems have been used successfully against
fires in internal floating roof tanks. Foaming the deck closes off the
oxygen supply, so that any vapor space under the deck will quickly be
deficient in the oxygen necessary to support a fire.
Comment: Two commenters (IY-D-9, IV-0-15) questioned the accuracy
of the API emission equations for IFR's. One commenter (IV-D-9) discussed
the accuracy of the equations when applied to vented vessels smaller
than 20 feet in diameter, which are more common in the chemical industry.
The commenter said that the equations were more accurate for the larger
vessels typically found in the petroleum industry. The commenter requested
that EPA consider the nonvented IFR because the air flow pattern through
the vents and, therefore, the wind speed would not be a factor, and
emissions should be reduced.
One commenter (IV-0-15) questioned the overall applicability to the
chemical industry of emission estimates based on data developed by the
petroleum liquid storage Industry. The co-enter said that emissions
from chemical tanks, especially those that are gas blanketed, will be
less than the emissions predicted by the equations.
Response- Overall, the emission test data on internal floating
roof tanks indicate that there is no significant dependence on wind
speed Wind-Induced ventilation removes vapors that have already escaped
the space confined by the deck, seals, or fitting covers, and it does
not appear to cause these emissions. It 1s expected that the different
ventilation patterns in smaller tanks would not play a significant role
in generating emissions, especially if vapor space is minimized by
employing liquid-mounted seals, or pressure differentials are prevented
by using tight fitting vapor-mounted primary seals and secondary seals.
Therefore, the EPA believes that the current emission equations are
adequate to support the selection of BOT for both small and large tanks.
2-33
-------
2.2.3 Add-on Control Options
Comment: Two commenters (IV-D-13, IV-D-22) discussed alternate
control options. One commenter (IV-D-13) asked if use of control devices
such as carbon adsorption, refrigerated condensers, and thermal oxidizers
are allowed if they achieve 95-percent emission reductions. Another
commenter (IV-D-22) noted that an alternate control option, vent gas
condensation, is commonly used and that its efficiency may be enhanced
through the use of product cooling and tank insulation. The commenter
requested that the EPA consider allowing this control option.
Response: Any type of closed vent system and control device such
as carbon adsorbers, vent gas condensers, and thermal oxidation units
that is designed and operated to reduce inlet VOC emissions by 95 percent
or greater is in compliance with § 60.112b.
Product cooling and tank insulation could be used to enhance the
effectiveness of condensation systems. The standards are based on the
vapor pressure of the liquid at its storage temperature. Therefore, the
use of this strategy is not prohibited by the standards.
Comment: One commenter (IV-D-9) requested that EPA allow use of
pressure/vacuum (conservation) vents. The commenter said that in vessels
with zero turnovers, a conservation vent is more effective than a vented
IFR. Another commenter (IV-D-22) stated that conservation vents, coupled
with tank insulation and product cooling, provide a reasonable level of
control in many circumstances.
Response: Conservation vents may have some effect in controlling
breathing losses from fixed roof vessels. However, the ability of a
conservation vent to reduce emissions in the chemical and petroleum
liquid industries is not well established. Additionally, VOC emissions
due to working losses significantly outweigh emissions due to breathing
losses, particularly as the number of tank turnovers increases.
As an example, the model tank used in the Volume I BID (volume of
606 m3 [160,000 gal], diameter of 9.1 meters [30 ft], 50 turnovers per
year, vapor pressure of 6.9 kPa [1 psia], and molecular weight of 80)
has total emissions of 6.22 Mg/yr as a fixed roof tank. Only 0.88 Mg/yr
result from breathing losses. Therefore, if a conservation vent reduced
breathing losses to zero, the emission reduction would total about
2-34
-------
14 percent. This is in sharp contrast to the 96-percent reduction
obtained by equipping the tank with an internal floating roof.
The scenario of zero turnovers presented by the coramenter is
unrealistic. Tanks are not permanent storage facilities for chemical
products and, therefore, do not have zero throughput. Tanks in the
chemical industry average about 60 turnovers per year in the 600 in3 size
range and about 150 turnovers in the 151 ra* (40,000 gal) size range.
Therefore, the final standard does not allow the use of conservation
vents in lieu of the specified controls.
The combination of conservation vents and product cooling could be,
in specific instances, an equivalent control technology allowed by
§ 60.114b. However, it would be the obligation of the applicant to
demonstrate that the proposed controls are equivalent to those promulgated
in § 60.112b.
The test program that is currently being conducted by API will
examine the emission equations for fixed roof vessels. The results of
this test oroarara will be used to evaluate the potential contribution of
conservatiorTvents to an mission Action program in the next review
of the standards and may be useful in supporting an equivalency determine-
tion.
Comment: One co-enter (IV-D-23) discussed the fugitive emission
monitorirtg~requirement for closed vent systems. The counter questioned
the cost effectiveness of the fugitive emission require«,nt for two
reasons: (1) the fugitive emission level is low and (2) a product
recovery credit cannot be claimed because the "recovered" material ends
up in the waste stream. The Commenter requested that all references to
. • .itrt.
-------
the intent of EPA to apply the standards to an existing facility of this
type.
2. The requirement referred to as the SOCMI equipment leak standard,
40 CFR Part 60, Subpart W, § 60.485(b), should be subject to all the
limitations and exceptions contained in Subpart. VV.
3. The commenter noted that the VOL standards are more restrictive
than the SOCMI standards with respect to leaks from valves and pumps.
The commenter said that the 500 ppm cutoff in the proposed standards is
more restrictive than the 10,000 ppm cutoff in the SOCMI standards. The
commenter requested that a 10,000 ppm cutoff be adopted for valves and
pumps in the proposed VOL standards to maintain consistency with the
SOCMI standards.
Another commenter (IV-D-13) requested further information on the
basis for using 500 ppm above background as the level of no detectable
emissions for closed vent systems. The commenter questioned whether
readings at this level are consistent with current instrument capabili-
ties and whether the requirement is consistent with leak rates set under
fugitive emission rates.
Response: The SOCMI fugitive standards were promulgated on October 18,
1983. The Agency determined at that time that the control of fugitive
emissions is cost effective.
The commenter (IV-D-23) questioned the applicability of the SOCMI
standards in circumstances where a new, modified, or reconstructed
storage vessel is controlled by an existing closed vent system and
control device. If the owner or operator chooses to control a new source
through the use of a closed vent system and control device, the control
equipment must be operated and maintained to conform to the requirements
of the standards. In this case, these requirements include meeting the
fugitive emission leak standards as specified in § 60.112b(a)(3)(i). A
new fixed roof storage vessel is not considered to be in compliance with
these standards when it is linked to an existing closed vent system and
control device unless the existing equipment is in compliance with
§ 60.112b(a)(3)(i). Therefore, the Agency has determined that no changes
are necessary in the final standards. It should be noted that, in order
to maintain consistency between the standards, the limitations and
2-36
-------
exceptions contained in Subpart W also apply to the fugitive leak
requirements for Subpart Kb.
The comment that a 500 ppm above background cutoff for allowable
detectable emissions is more restrictive than the cutoff level allowed
under the SOCMI standards is erroneous. The SOCMI standards require
that closed vent systems shall be designed and operated with no detectable
emissions indicated by an instrument reading of less than 500 ppm above
background and visual inspections. Any valves and pimps on the closed
vent systems would be subject to the 10,000 ppm cutoff per the requirements
in the SOCMI standards for valves and pumps. The Agency has determined
that adequate instrumentation exists to determine emissions below 500 ppm.
Comment: One commenter (IV-D-9) recommended modification of the
requirement for a 95-percent reduction in emissions when a closed vent
system and control device are used. The commenter recommended that, in
circumstances where vented IFR's cannot be used, the control device
should instead be designed and operated to achieve an emission reduction
equivalent to that calculated for the affected storage vessels by the
formulas in AP-42 and API Bulletin No. 2519 or by a reduction of 95 percent,
whichever is lower.
Response: The three fundamentally different techniques for compliance
allowed by these standards are not equivalent in terms of emission rate.
In the process of developing the requirements for internal floating
roofs, each component of the roof was evaluated for emission reduction
and cost. Thus, a BDT for internal floating roofs was developed. The
calculated emission reduction obtained by a BDT internal floating roof
varies from about 50 percent up to 99 percent over a fixed roof tank
depending on the number of turnovers, vapor pressure, and molecular
weight of the stored material. The reduction obtained in typical tanks
is about 95 percent.
Properly designed and operated vapor control systems can reduce V0C
emissions by 95 percent. This is similar to the reduction typically
achieved by the BDT internal floating roof. If vapor recovery systems
are to be used, they should be designed and operated to be as efficient
as possible. Installation of vapor control systems designed to function
at low efficiencies would increase missions unnecessarily. Therefore,
2-37
-------
the final standards do not incorporate the provision requested by the
commenter.
Comment: While one commenter (IV-D-2) stated support for the use
of flares as a control device, three other commenters (IV-D-6, IV-D-9,
IV-D-12) stated that the flare exit velocity limitations are unduly
restrictive and suggested that they be reviewed in light of the latest
information from the Chemical Manufacturers Association and EPA. Two
commenters (IV-D-9, IV-D-12) said that the velocity specifications are
identical to those proposed in the SOCMI equipment leak NSPS, and suggested
that both the SOCMI and the proposed VOL standards be revised to encompass
a consistent set of limitations based on a recently completed study
showing 98 percent or better destruction efficiencies at velocities
greater than the existing velocity limitation. One commenter (IV-D-9)
suggested a public comment period on flare operation limitations.
Response: The flare exit velocity limitations have been reviewed
by EPA in the time since the standards were proposed. The data obtained
by an EPA test program showed that VOC destruction efficiencies of
98 percent or better are achievable using higher exit velocities when
the net heating value of the gas being combusted is greater than
37.3 MJ/scm (1,000 Btu/scf). Both the VOL standards and the SOCMI
equipment leak NSPS have been revised to reflect this information. The
final standards limit flare exit velocity of steam-assisted and non-
assisted flares to 18.3 m/s (60 ft/s) unless the net heating value of
the gas being combusted is greater than 37.3 MJ/scm (1,000 Btu/scf). In
this latter case, exit velocities may be between 18.3 m/s and 122 m/s
(400 ft/s). The revised standards also permit the owners to operate the
flare at a theoretically determined maximum exit velocity so long as it
is less than 122 m/s.
2.2.4 Column Fittings
Comment: Three commenters (IV-D-7, IV-0-8, IV-D-20) found the
required use of flexible fabric sleeve seals on column penetrations to
be restrictive and recommended that EPA also allow the use of gasketed
sliding covers. They noted that flexible fabric sleeve seals are a
fitting design unique to a single manufacturer and are not generally
available. They also noted an insignificant difference in overall
2-38
-------
emission reduction (0.1 to 0.2 percent) when flexible fabric sleeve
seals are used in place of gasketed sliding covers. Two of the
coraroenters (IV-D-7, IV-D-8) stated that the use of "built-up" columns,
which are currently in wide-spread use, is disallowed under the proposed
standards because sleeve seals can only be used with pipe columns. One
commenter (IV-D-20) discussed the potential for damage and maintenance
repair problems with use of sleeve seals which could result in lengthy
downtime.
One commenter (IV-D-7) pointed to an apparent miscalculation in the
BID resulting in a 50-percent overestimatron in the emission reduction
associated with controlled versus uncontrolled fittings. According to
the commenter, a miscalculation would place an undue emphasis on
maximizing fitting controls and, therefore, would restrict column and
seal designs that should be allowable.
Two commenters (IV-D-7, IV-D-18) said that slotted gauge/sample
poles are frequently used and recommended that they be allowed in the
final rule.
Response: Flexible fabric sleeve seals are currently available
only on contact decks. It is not the intent of the Agency to prohibit
the use of noncontact decks with this fitting requirement. While the
annualized cost of redesigning a noncontact deck to allow the adoption
of flexible fabric sleeve seals is not known, it is possible to estimate
an upper limit for that cost beyond which the cost effectiveness of the
flexible fabric sleeve seal would be unreasonable.
The total emission reduction achieved for storage of a typical VOL
is 0.007 Mg/yr/column. In order not to exceed a reasonable cost effec-
tiveness, the maximum capital cost may not exceed $33/column and the net
annualized cost may not exceed $7/yr/column. The Agency has determined
that it is highly unlikely that noncontact decks could be redesigned and
flexible fabric sleeve seals installed below these costs. Alternatively,
gasketed sliding covers are widely available and may be employed on both
contact and noncontact decks. Therefore, the Agency has decided to
revise the proposed regulations to allow the use of either flexible
fabric sleeve seals or gasketed sliding covers.
2-39
-------
The EPA has reevaluated the costs to control emissions from fittings.
These controls are estimated to cost $500 (4 percent of the cost of the
deck) for a typical vessel with a diameter of 9.1 m (30 ft). The cost
effectiveness of controlling emissions from fittings on this vessel
storing a typical VOL is $990/Mg. The capital cost of controlling
fitting emissions on a typical 30.5-m (100-ft) diameter vessel storing
gasoline is $1,320, and the cost effectiveness is $345/Mg. The Agency
has determined that these values are reasonable, and therefore, no
changes have been made to the requirements for fitting controls.
The standards will not be revised to allow the use of slotted
gauge/sample poles. The BDT for sample pipes and wells is the slit
fabric seal with 10 percent open area. Slit fabric seals are widely
available at no additional cost and achieve a greater emission reduction
than slotted gauge/sample poles and are included in the above evaluation.
2.2.5 Equivalency Determination
Comment: Two commenters noted the possibility of construction
delays resulting from a determination that a proposed technology is
equivalent to the technologies allowed in the proposed standards. One
commenter (IV-D-7) requested that EPA be required to make an equivalency
determination within 90 days of receipt of a request rather than allowing
the unspecified period in the proposed standards. A second commenter
(IV-D-22) requested that State agencies operating with approved State
Implementation Plans be allowed to make the equivalency determination.
Response: No changes will be made in the language regarding equiva-
lency determinations. This NSPS is an equipment standard (§ 111(h) of
the Clean Air Act) because it is impracticable to measure emissions.
Any future requests for equivalency determinations would most likely be
made for a new type of roof or seal. A straightforward emissions measure-
ment is not possible in these cases, and it is likely that the determination
could involve complex modeling, the evaluation of new test methodologies,
or other time consuming tasks. A 90-day time period is not sufficient
to evaluate the proposed equipment, propose a determination, respond to
public comments, and promulgate a final determination. The amount of
time needed to make a determination can vary greatly depending on the
complexity of the equivalency determination request. Therefore, the
2-40
-------
time in which the Agency is to make tfie determination is left unspecified
in the final standards.
No changes have been made in the standards to allow the States to
make equivalency determinations. Even when States have been delegated
NSPS authority, they have not been granted the authority to make
equivalency determinations because these involve proposal and promulgation
rulemakings that can only be undertaken by EPA. The reason for this
policy is that different States could come to opposite conclusions on
the equivalency of a new technology. This would be both confusing and
contrary to the intent of the Clean Air Act to promulgate uniform,
national performance standards for new sources.
2.3 RECORDKEEPING, REPORTING, AND INSPECTION REQUIREMENTS
2.3.1 Recordkeeping and Reporting Requirements
Comment: Several commenters requested changes in the provisions
requiring operators to maintain records showing the dimensions and
capacity of storage vessels larger than or equal to 40 m3 (10,000 gal).
Three commenters (IV-D-2, IV-0-10, IV-D-16) said that the recordkeeping
provision is without purpose or benefit because vessels that are less
than 75 m3 (20,000 gal) are not subject to the control requirements.
One commenter (IV-0-10) said that the inclusion of any sort of require-
ment on vessels down to 40 m3 in capacity would generate both demand for
performance standards for these vessels and more permitting requirements
at the State or local level. One commenter (IV-D-4) suggested that the
regulation only apply to storage vessels with a capacity greater than
70 m3 (18,500 gal), while two others (IV-D-10, IV-0-16) suggested that
the cutoff for recordkeeping and control be the same level (75 m3).
Response: The Agency has required the owner/operator of tanks
between 40 m3 and 75 m3 to retain a record of the size of the tank to
aid in enforcement of the standards. The records are necessary for
enforcement because they remove any ambiguity in identifying vessels
that are close to the control cutoff. According to one commenter
(IV-D-10), the recordkeeping requirement is not burdensome because such
records are routinely maintained. The comment that an NSPS would encourage
further regulation is irrelevant. The fact that EPA has asked owners of
2-41
-------
a category of storage vessels to keep a record of the size of the vessel
is not a basis for State or local agencies to extend control requirements
to that category.
Comment: One commenter (IV-D-13) stated that the reporting require-
ments are burdensome and time consuming for both the regulated community
and the Agency. The commenter recommended that the reporting requirements
be replaced with a requirement that the regulated community maintain
reasonable records of performance.
Response: The Agency has determined that the reporting requirements
are reasonable and do not place an undue burden on the regulated community
or the Agency. Reports are submitted only in the event of a system
failure and given the expected low probability of system failure, the
reports will not be an undue burden. The reporting requirement is
essential in enforcing the standards.
Comment: One commenter (IV-D-23) said that the requirement that an
operator of a vessel equipped with a closed vent system (other than a
flare) submit an operating plan for approval to the Administrator is
burdensome. The commenter noted that under the General Provisions
(§ 60.8), a performance test is already required within 60 to 180 days
of start-up to verify the efficiency of the system. The commenter said
that the testing requirement alone should be sufficient to demonstrate
compliance and that submittal of an operating plan for approval should
not be required.
Response: It was not the intent of the Agency to require a perfor-
mance test for these standards. Emissions from fixed roof tanks are
variable and are often at rates that are too low to measure. When
liquid is entering a vessel, the liquid surface rises, forcing vapors
above the liquid surface out of the vessel. While this is occurring,
the vapor flow rate and the emissions are large. When liquid is exiting
the vessel, the liquid surface falls, and the resulting pressure differen-
tial sucks air or a blanketing material into the vessel. During these
operations, vapor flows into the storage vessel resulting in no atmospheric
emissions. When the liquid level is held constant, pressure differentials
resulting from diurnal temperature variations expel vapors at very low
flow rates at intermittent times during the cycle.
2-42
-------
Certain components of uncontrolled emissions have been measured in
very specialized tests conducted by the EPA and industry. Total emissions
have not been measured, however, and to do so would require that the
operation of the vessel be strictly controlled during the testing period.
Because of methodology problems, it may not be possible to measure both
the flow rate and the concentration simultaneously. This would cast
doubt on the accuracy of the measurement. In any case, testing would be
extremely expensive for individual compliance determinations. For these
reasons, it was concluded that it was impracticable to measure the
emissions exiting the tank.
For the same reasons, it would be impracticable to measure the
emissions captured by the closed vent system or entering the control
device. Therefore, it was concluded that reduction efficiency measure-
ments are not feasible for closed vent systems and control devices, and
the format of the standards for closed vent systems and control devices
is an equipment standard. Therefore, no performance test is required.
Operational requirements, which consist, among other things, of
inspection, repair, and work practice requirements, are necessary to
ensure the proper operation and integrity of control equipment meeting
the standards Submittal of an operating plan is essential to demon-
^"iance with the standi, and the r*,U1r«„t Has been
retained in the final standards. The retirement will be modified,
however, to specifically exempt storage vessels fro. the General Provision
requirement for a performance test (S 60.8) for the reasons discussed
above.
2.3.2 Cvt.rnal Floats Boof Vessel (EHO Inspection Requirements
Comment: One commenter (IV-D-11) said that annual measurements of
secondary~seal gaps on external EFR's are excessive and that these
measurements should be conducted every 5 years to coincide with measure-
ment of the primary seal gap.
Response- The time and resources required to conduct these inspec-
tions are minimal. The only way to ensure seal integrity and compliance
with the standards is through these Inspections. Therefore, the annual
inspection requirement has been retained in the final standards.
2-43
-------
Comment: Two commenters (IV-D-7, IV-0-18) requested that EPA allow
the initial seal gap measurements to occur during the hydrostatic testing
of the vessels or within 60 days (IV-D-7) or 6 months (IV-0-18) of the
initial fill.
Response: The Agency has revised the final standards to allow the
seal gap measurements to occur during hydrostatic testing. Measurements
conducted during the hydrostatic testing would reveal any seal gaps that
would be discovered during an inspection after the initial fill. The
60-day time frame to conduct the inspection after the initial fill is
reasonable considering the inspection takes less than a day. Therefore,
the time frame after the initial fill has not been changed from the
proposed standards.
2.3.3 Internal Floating Roof Vessel (IFR) Inspection Requirements
2.3.3.1 Annual Visual Inspection.
Comment: One commenter (IV-D-9) said that an annual visual inspection
of IFR seals precludes the use of nonvented IFR's because of the excessive
time, materials, and manpower required to inspect the vessels. The
commenter said that the inspection would also violate internal company
safety rules. The commenter suggested that IFR's with primary and
secondary seals be inspected internally at 5-year intervals. If EPA
were to approve a 5-year inspection interval, the commenter further
proposed that it be considered equivalent to an annual inspection of a
single-seal system. The commenter calculated that overall emission
rates due to seal failure are equivalent under the two options.
Response: After evaluating this issue, the Agency has determined
that it may not be possible to inspect all IFR's without emptying and
degassing the vessel. The Agency evaluated the commenter's proposed
revision and has decided to revise the standards. If the operator
equips the vessel with a primary and a secondary seal and conducts an
internal inspection every 5 years, then the controls are considered
equivalent to a single-seal system and annual visual inspection. Under
the double-seal system option, the addition of a secondary seal will
reduce emissions beyond the emission reduction achieved by a single-seal
system, thus offsetting the risk of increased emissions due to seal
failure. In any case, seal failure rates are generally quite low, and a
2-44
-------
major failure (such as a deck sinking) would be evident to the operator
even in the absence of an annual inspection because of a loss of material.
A worst case analysis of the possible impact this suggestion would
have on emissions was performed. Even if 10 percent of the tanks equipped
with primary and secondary seals experienced total failure of one seal,
average emissions would be 4 percent lower than if all tanks had been
equipped with BOT.
Comment: Three commenters (IV-0-9, IV-D-15, IV-D-17) discussed a
potential safety hazard in conducting an annual visual inspection in all
IFR's. Two commenters (IV-D-15, IV-D-17) suggested as an alternative to
the inspection requirement that VOC emissions be monitored annually from
a small fitting on the roof. They said that if monitoring indicates a
significant increase in emissions, an internal inspection would be
warranted to find and correct the problem. One commenter (IV-D-17)
suggested that this alternative replace the annual visual inspection
while the other commenter (IV-D-15) suggested that it be considered
equivalent to the visual inspection. Another commenter (IV-D-12) recom-
mended that, where safety problems can be caused by the requirements of
an annual visual inspection, visual inspection should be required when
degassing occurs or at intervals not greater than 10 years.
Response: The Agency agrees that in particular tanks, the annual
visual inspection may pose a problem. Therefore, as noted in the previous
response, the requirement has been altered. Because tanks in the chemical
industry are typically cleaned and degassed once every 5 years under
current practices, the revised requirements will alleviate any safety
problem with the annual visual inspection.
The alternative suggested by the commenters has not been incorporated
into the final standards. There are no data on which to select a hydro-
carbon concentration that would indicate a problem with the control
equipment. The hydrocarbon concentration measured at the roof fitting
would be heavily dependent upon recent tank operations (e.g., filling,
emptying, or static level) and liquid level. The Agency is not aware of
any method by which an annual concentration measurement could be used to
establish the condition of the control equipment.
2-45
-------
Comment: One commenter (IV-D-9) said that the time (1 hour) estimated
for the annual inspection is greatly underestimated. The commenter
requested a reference for the estimate.
Response: Based on site visits by EPA personnel, the Agency deter-
mined that 1 hour is a reasonable estimate for the time required to
conduct an annual visual inspection. Personnel from the Texas Air
Control Board said that the inspection takes 4 hours, which includes
time for a file search and time for writing the report. However, even
using the 4-hour estimate, the cost of the annual inspection is not
burdensome; therefore, the costs based on the time estimate will not be
changed.
2.3.3.2 Ten-Year Inspection.
Comment: Two commenters (IV-D-7, IV-D-20) said that the requirement
that IFR's be emptied and degassed at least once every 10 years is
unreasonable for facilities with only one IFR or where acceptable alter-
nate tankage is not available. One commenter (IV-D-7) requested that
pipeline tank stations with only one IFR be exempt from these require-
ments or that an operationally compatible method to empty and degas the
tanks be allowed.
Response: As stated in the previous response, tanks in the chemical
industry are typically emptied, degassed, and cleaned every 5 years.
Benzene storage vessels are typically emptied, degassed, and cleaned on
10-year intervals. The Agency has determined that it is typical industry
practice to clean tanks on a regular basis and that the 10-year internal
inspection requirement is not an undue burden. The Agency has determined
that no special provisions are necessary for pipeline tanks. In many
instances, alternate tankage such as an EFR will likely be available.
In any case> it is not unreasonable to require a facility with only one
IFR to conduct a planned, internal inspection every 10 years.
Further discussion with one of the commenters (IV-D-7) revealed
that existing tanks were of concern. This NSPS will not affect existing
storage vessels; therefore, no changes have been made as a result of
these comments.
2-46
-------
2-3-4 Procedures for Vessels Found to be Out of Compliance
Comment: Several commenters said that the 30-day allowance for
repairing or emptying storage vessels found to be out of compliance is
unreasonable. One commenter (IV-D-20) said that the provision would
necessitate the installation of two small tanks rather than a single
large tank to provide the flexibility to transfer material from a
vessel in need of repair. Another commenter (IV-D-5) noted that the
provision would be a problem in the event a facility found that several
vessels were simultaneously out of compliance. One commenter (IV-D-19)
suggested that a 45-day allowance would give the operator sufficient
time to order, receive, and install new equipment without having to
request an extension. Four other commenters (IV-D-6, IV-0-7, IV-O-12
IV-D-20) suggested a 90-day allowance as a reasonable time to repair a
vessel or remove it from service.
Response: Discussion with storage vessel manufacturers indicated
that a 30 day allowance for repairing or exempting storage vessels in
conjunction with the option of requesting a 30-day extension is reasonable
from the supplier's viewpoint. However, In the event that special
materials not normally kept in stock (such as Teflon® seals) were required
this time would probably be insufficient. The Agency has decided to
revise the proposed standards to provide a 45-day allowance to accom-
modate these circumstances. A 30-day extension may still be requested
if repairs are likely to exceed the initial allowance.
2.3.5 Notification of Refill
Comment: Two commenters (IV-D-19, IV-D-23) said that the requirement
to notify EPA 30 days prior to refilling storage vessels after conducting
a planned inspection is unnecessary and may delay the refilling of
needed storage capacity. One commenter (IV-D-19) said that inspection
prior to refilling should be a recordkeeping function. Another commenter
(IV-D-5) stated that the intent of the requirement was to provide 30 days
notice to EPA prior to refilling the vessels. Two commenters (IV-D-5,
IV-D-23) suggested that the 30-day notification period begin prior to
the inspection.
Response: The requirement is intended to provide 30 days notice to
EPA prior to refilling the vessels, and not to delay the start of the
2-47
-------
notification period until after the vessel has been emptied. The require-
ment has been revised to allow the notification period to begin prior to
emptying and inspecting the storage vessel.
Comment: Two commenters (IV-D-22, IV-D-23) discussed the 7-day
notification period in the event of an unplanned inspection. One commenter
(IV-D-22) said there would be the possibility of excessive downtime and
recommended that this requirement be deleted. If it is retained, the
commenter suggested a 2- to 3-day notification period. The commenter
also said that EPA should be required to provide a telephone contact on
weekends and holidays to enable the period to begin at any time. The
other commenter (IV-D-23) noted the possible delay in returning to
normal operation resulting from the 7-day requirement and suggested
shortening the notification period to 8 business hours in the event of
an unplanned inspection.
Response: The Agency has determined that a 7-day notification
period in the event of an unplanned inspection is reasonable. Discussions
with a representative of a tank service company revealed that it takes
from 2 days to 2 weeks to empty, degas, clean, and inspect a tank.
Factors such as increasing tank size, heavy sludge deposits, worker and
equipment access to the tanks, and the need to replace seals can all
increase the length of time needed to perform the work beyond the minimum.
Furthermore, the ability of the service company to respond rapidly to an
emergency call depends on the availability of workers and equipment.
Therefore, the Agency has decided that a 7-day notification period will
not result in excessive downtime throughout the industry. Furthermore,
the 7-day period is necessary from the Agency's perspective to allow
adequate time for the mobilization of the necessary manpower and resources
to inspect the vessel. In the event the owner/operator makes the Agency
aware of a special problem on a vessel requiring immediate refill, the
Agency will make every effort to respond as quickly as possible.
2.4 MODIFICATION
Comment: Five commenters (IV-D-2, IV-D-13, IV-D-18, IV-D-22,
IV-D-23) requested clarification of the modification provision. The
commenters suggested excluding the changing of liquids in storage vessels
2-48
-------
from constituting a modification if the existing facility was originally
designed to accommodate the alternate liquids. Two commenters (IV-D-18,
IV-D-22) suggested that this be accomplished by exempting as affected
facilities vessels meeting these conditions. One commenter (IV-D-13)
suggested the proposed standards clearly state that little can be done
to a storage vessel to qualify as a modification.
Response: The commenters are correct that a change of the stored
liquid will not be a modification to an existing vessel. If a vessel
constructed prior to July 23, 1984 (date of proposal of these standards),
was storing a non-VOL or a VOL with a vapor pressure below the cutoffs
and the vessel contents are changed to an affected VOL, the vessel would
not be considered an affected facility under Subpart Kb. However, if a
vessel constructed after proposal undergoes a similar change, the vessel
could fall under the control requirement provisions of Subpart Kb.
Because this distinction is clear in § 60.14 of the General Provisions,
no changes are incorporated into the final rule.
Comment: One commenter (IV-G-22) suggested that vessels that
suffer from catastrophic failure should be exempt from the proposed
standards if they are reconstructed "in-kind" and preexisting controls
achieved at least an 85-percent emission reduction. The commenter also
stated that existing vessels that are modified should be exempt if they
are operated with controls that are at least 85-percent efficient.
Response: The provisions for modification and reconstruction are
clearly defined by §§ 60.14 and 60.15 of the General Provisions. They
do not allow "in-kind" reconstruction or modification, even in the event
of a catastrophic failure. The intent of Section 111 of the Clean Air
Act is to reduce emissions as older facilities are modified or replaced
with new facilities by requiring that the modified or new facilities
include BDT. In the case of a facility which is not completely new, the
Agency has made provisions that reconstructed facilities will also be
required to include BOT. An allowance for "in-kind" replacement would
be inconsistent with the Act's intent. Additionally, the cost effec-
tiveness of controls for a modified or reconstructed facility would be
comparable to that of a new facility. This cost effectiveness is
2-49
-------
reasonable (a net credit on average). Therefore, the proposed standards
have not been revised as the commenter suggests.
2.5 COST EFFECTIVENESS
2.5.1 Capital Recovery Factor
Comment: One commenter (IV-D-8) stated that the capital recovery
factors used in the cost analysis are not representative of the ones
used in the petroleum industry. According to the commenter, the analysis
should have been based on "equity financing" rather than "debt financing."
The commenter said that the use of equity financing would double the
capital recovery factors and increase the costs of the control options.
Response: Capital recovery factors for regulatory cost and cost-
effectiveness analyses are traditionally based upon a 10-percent real
interest rate and the physical lifetimes of control equipment. Such
estimates are intended to reflect the cost of regulation to society.
As noted in Chapter 9 of the Volume I BID, industry financial
characteristics were considered in the economic impact analysis through
the calculation of an industry-specific cost of capital, which incorpo-
rates all types of financing. This cost of capital estimate is an
average of the costs of debt, equity, and preferred stock, weighted by
the percentage of funds generated by each type of financing. The cost
of capital estimate was based on a sample of 100 firms, some of which
were petroleum companies. The cost of capital estimate reflects the
types of financing employed by both firms that operate primarily in the
chemical industry and firms that operate primarily in the petroleum
industry.
2.5.2 Product Recovery
Comment: Three commenters questioned the product recovery credits
used in the analysis. One commenter (IV-D-8) said that the assumed
product value is about double the product value commonly assumed in the
petroleum industry. The commenter concluded that the economic analysis
overstates the credit for product recovery and makes the cost-effective-
ness values for controls on the petroleum industry appear more attractive
than they really are.
2-50
-------
Another commenter (IV-D-15) requested that the choice of product
recovery values be clarified. The comenter noted that values of $360/Mg
($400/ton) and $695/Mg ($770/ton) were noted in the preamble to the
proposed standards while values of $460/Mg ($510/ton) were used in the
background information document (BID).
A third commenter (IV-Q-IS) said that no product recovery credit is
available to the "for hire" terminals.
Response: The Agency reexamined the cost-effectiveness values for
controls on petroleum liquid storage vessels. A lower product value,
$220/Mg, was assumed. Based on information developed for the draft
Control Techniques Guideline document, this product value is reasonable.
The cost effectiveness of a typical petroleum liquid storage vessel
(30 ra [100 ft] in diameter) was calculated for a low vapor pressure
crude oil and for a higher vapor pressure gasoline. Assuming 12 turnovers
per year on the crude vessel, the cost effectiveness is $794/Mg. The
cost effectiveness on the gasoline storage vessel is $1,Q14/Mg. The
cost effectiveness of control of these vessels is considered reasonable
even when a lower product value is assumed.
The product values in the preamble reflect a worst case cost for
chemicals ($360/Mg) and an average price of chemicals between 3.5 and
10.4 kPa (0.5 and 1.5 psia) ($694/Mg), respectively. ' These prices were
used to set the control cutoffs ($360/Mg) and to evaluate the impacts of
the standards on a typical tank storing a low pressure liquid ($694/Mg).
The value in the BID ($46G/Mg) represents a nationwide average of retail
prices of high volume chemicals and was used to evaluate the overall
impacts of the standards on the chemical industry.
While a direct product recovery credit is not available at "for
hire" terminals, an indirect credit is available because less product is
lost as the result of emission controls, and the company can reasonably
charge more for storage in the tanks which have lower losses.
Comment: One commenter (IV-0-15) stated that the cost calculations
show that as larger vessels are considered, the product recovery credit
becomes greater than the cost of control. The commenter questioned
whether the proposed standards are necessary in this case.
2-51
-------
Response: In some cases, the cost of controlling VOC emissions
yields a net credit because of the value of the recovered product, and,
in such cases, industry may adopt the control independent of any regulation.
However, there is no assurance or experience to show that this will
occur. Moreoever, there are other cases where control of storage vessels,
while cost effective, has a net cost to the owner. In these cases, in
the absence of an NSPS, BDT would probably not be installed by the
facility. In addition, it may be possible to install control technology
that would yield a net credit to the owner but would not be as efficient
as the best demonstrated technology. This would occur where the overall
cost is a credit and where there is an incremental cost between the
differing levels of control. In such situations, an NSPS is necessary
to ensure that BDT is installed in all cases. Even when the BDT is
installed, the NSPS ensures proper operation and maintenance of the
equipment, thus ensuring the maximum emission reduction. For these
reasons, and in order to implement uniform, nationwide standards, the
Agency has decided to apply the NSPS to all affected facilities.
2.5.3 Cost of Controls (Cost Effectiveness)
2.5.3.1 Add-On Controls.
Comment: One commenter (IV-D-2) stated that control of a 151-m3
(40,000-gal) vessel storing liquid with a vapor pressure of 3.5 kPa
(0.51 psia) would have an unreasonable cost effectiveness of at least
$8,800/Mg ($8,000/ton) if the VOC's were ducted to an existing control
device. The commenter noted that if the costs of a new control device
were considered, the cost-effectiveness value would likely be an order
of magnitude higher. Another commenter (IV-D-5) said that no considera-
tion was given to the incremental cost of utilizing carbon adsorption,
incineration, or other similar control devices when considering individual
vessels.
Response: The owner/operator does not have to install a closed
vent system and control device. Internal floating roofs can be built at
a reasonable cost and are applicable to a wide variety of storage situa-
tions. These are the reasons why this technology was selected as BDT.
For example, the cost effectiveness of using an internal floating roof
to control the tank described by the commenter (151 m3 in capacity and
2-52
-------
storing a liquid with a vapor pressure of 3.5 kPa) is $1,94G/Mg, assuming
50 turnovers/year. However, these standards allow the owner or operator
to control emissions with a closed vent system and control device that
is 95-percent efficient because these devices achieve the same emission
reduction as BDT. As noted in previous responses, these systems are
also widely used and can have a reasonable cost effectiveness when
controlling small tanks storing high pressure liquids or whan utilized
on a pi ant-wide basis.
2.5.3.2 Floating Roof Vessels.
Comment: One commenter (IV-D-7) questioned the assumption that
controlled fittings costs are negligible.
Response: The Agency agrees that the cost of the control requirements
for fittings is greater than zero, but still maintains that it is small
in comparison to the overall cost of the IFR. Estimates from vendors
indicate that controlled fittings increase the installed capital cost of
the IFR between 1 and 4 percent. In a lG-m (33-ft) diameter tank, this is
about $500 if controlled fittings increase the cost by 4 percent.
Previous responses have discussed the cost effectiveness of
controlling fittings, and have demonstrated that the cost effectiveness
of the revised requirements is reasonable.
Comment: One commenter (IV-D-8) stated that the estimated costs
for IFR's are understated. The commenter compared his estimate of
$40,000 to install an internal floating roof in a 15-m (49-ft) diameter
vessel to the $17,120 assumed in the analysis. The commenter requested
that all factors and assumptions used in estimating installed costs be
stated.
Another commenter (IV-0-7) requested that EPA reevaluate the
cost-effectiveness values for IFR's and EFR's and any regulatory
decisions resulting from them. The commenter's estimates for these
values were significantly higher than those made by EPA. The commenter
also said that it was not possible to replicate EPA's estimates using
the information supplied on estimated installed capital costs of IFR's.
Response: Although the commenter did not provide all of the
assumptions used in his cost estimate, it is likely that the $40,000
estimate is based on welded steel deck. As Table 2-8 shows, the
2-53
-------
TABLE 2-8. ESTIMATED INSTALLED COST OF A WELDED CONTACT INTERNAL
FLOATING ROOF WITH SECONDARY SEAL
(Fourth-quarter 1982 dollars)
Tank
Roof
diameter, m
cost, $
5
15,900
10
30,000
15
44,000
20
58,100
25
72,100
30
86,100
The basic cost of the roof and primary seal is
estimated from the equation: cost ($1,000) =
1.91 + 2.54D; where D equals the tank diameter in
meters with the correlation coefficient r2 = 0.883.
The additional cost of a secondary seal is estimated
based on the factor, $85 per linear meter of circum-
ference. The secondary seal cost is the average
price of 13 seals from 8 different vendors.
2-54
-------
coramenter's estimate is comparable to the installed costs of welded
steel decks that were developed in the Volume I BID. With one exception,
the Agency believes that the costs presented in the Volume I BID are
reasonable and accurate estimates of the costs to control tanks. Two
comments received on the draft CTG stated that the costs used in that
document accurately reflected the commenters1 actual experience in
retrofitting tanks. The costs used in the CTG were developed from the
costs presented in Chapter 8 of the Volume I BID. The primary assumption
used in the costs is that the vessel is empty, clean, and otherwise
ready for the IFR.
One change that has been made in the costs since proposal 1s in
regard to the cost and lifetime of equipping a noncontact deck with a
liquid-mounted primary seal. Information received in responses to the
CTG indicates that the $2.60/1inear meter seal cost had been under-
estimated. Also, the lifetime of this seal system has been revised
downward from 20 years to 10 years. However, the cost effectiveness of
BDT is still reasonable. The cost effectiveness of controls will range
from about $545/Mg (assuming the product has no value) to net credits
(if the product value exceeds $545/Mg) for a model tank with the
following parameters:
1. Volume = 606 tn3 (150,000 gal);
2. Diameter = 9.1 m (30 ft);
3. Vapor pressure - 6.9 kPa (1 psia);
4. Molecular weight 31 80 lb/lb mole; and
5. Turnover rate = 50 per year.
These values are reasonable, and no changes have been made to the
controls required by § 60.112b as a result of these changes in costs.
The estimates for the emission reduction obtained by fittings
should have been 0.894 Mg/yr and not the 0.1 Mg/yr presented in the
Volume I BID. The cost effectiveness of the revised fitting requirements
is presented in Section 2.2.4 of this document.
Comment: One commenter (IV-D-15) said that the costs of lined
tanks or pad systems, which are required on many S0CMI tanks are not
represented by the cost estimates, which are based on general tanks
holding petroleum liquids.
2-55
-------
Response: The cost of tank liners or pad systems for a tank do not
affect the costs of controls. These costs are part of the tank, and the
tank would be so equipped both with and without the controls. Therefore,
these costs are part of both the controlled and uncontrolled case and do
not affect the incremental costs of control.
2.6 MISCELLANEOUS
Comment: One commenter (IV-D-17) said that it may be difficult for
an operator to determine whether or not a vessel is covered by the
standards and, if so, what provisions are applicable. The commenter
recommended that EPA redraft the proposed standards. Another commenter
(IV-D-5) recommended that EPA supplement the text on these standards
with a table summarizing the control requirements by tank size and vapor
pressure. The commenter said that this would avoid potential confusion
in determining applicability of the standards.
Response: The Agency has determined that the format for determining
applicability in the proposed standards is adequate, and no changes have
been made to the final standards. Any storage vessel owner/operator
with specific questions regarding applicability may contact EPA or the
permitting authority.
Comment: Two commenters (IV-D-7, IV-D-8) said that EPA's analysis
in the BID is based solely on tanks in the chemical industry, and does
not reflect tank populations, VOC properties and costs, equipment costs,
or control cost effectiveness associated with petroleum liquids.
Response: As stated in several previous responses in this document,
the Agency has determined that controlling emissions from petroleum
liquid storage vessels is cost effective.
The costs presented in Chapter 8 of the Volume I BID reflect the
costs of controls for petroleum liquid storage vessels. Tank population
and the emission reduction obtained by the regulatory alternatives are
presented in Appendix D of the Volume I BID. Therefore, it is concluded
that BDT is appropriate for petroleum liquid storage vessels, and no
changes have been made to the final standards as a result of this comment.
Comment: One commenter (IV-Q-4) suggested that numbers be rounded
in the proposed standards. Two commenters (IV-D-5, IV-D-9) requested
2-56
-------
that English unit equivalents be shown in parentheses in the proposed
standards.
Response: Numbers have been rounded in the final standards where
appropriate. However, in cases where rounding would undermine historical
precedents for control cutoffs, values have been retained as written.
English unit equivalents have been included, for the industry's con-
venience, in the Preamble to the final standards. English unit equivalents
have not been added to the regulation, however. Such a revision, because
of conversion factors, would modify the control cutoffs, and would be
counterproductive to a timely promulgation of the standards.
Comment: One commenter (IV-D-7) noted a typographical error in
§ 60.111b(f). The commenter said that "nonvolatile" should be changed
to "volatile."
Response: The comroenter was correct in noting a typographical
error in § 60.111b(f). The correction has been made in the final standards.
Comment: One commenter (IV-0-7) requested that the proposed standards
include definitions that are found in the preamble for terms such as
"VOC" and "petroleum liquid."
Response: The term "VOC" has been defined in § 60.2 of the General
Provisions of the Clean Air Act. The terra "petroleum" has been defined
in the final standards in § 60.111b(g), and the Agency considers this a
synonymous definition for "petroleum liquid." Therefore, no changes
have been made in the final standards.
Comment: One commenter (IV-0-7) suggested that floating roof
fittings be defined per page 22 of API Bulletin No. 2519. The commenter
said that this will avoid inconsistency with API Bulletin No. 2519 and
recognize that an automatic bleeder vent is the same as a vacuum breaker.
Response: The commenter was correct in noting inconsistency in the
floating roof fitting definition. The Agency recognizes that automatic
bleeder vents are the same as vacuum breaker vents, and the correction
has been made in the final standards.
Comment: One commenter (IV-D-7) had the following comments on
§ 60.112b(a)(l) and (2):
1. The section should distinguish between contact and noncontact
roofs. The comraenter also said that deck penetration on contact roofs
2-57
-------
in IFR's and EFR's should not extend into the liquid as is required in
these sections for all floating decks regardless of type.
2. The wording "except for contact decks" should be added to
§ 60.112b(a)(l)(iii) to make §§ 60.112b(a)(l) and (2) consistent
throughout.
3. The wording "column wells" should be deleted in
§ 60.112b(b)(l)(iv) and "leg sleeves" should be added for completeness.
4. Section 60.112b(a)(l)(v) should be revised to read . .at
all tiroes when the roof is floating, except when the roof is being
floated off or landed on the roof by supports. ..." for correctness
and consistency with proposed § 60.112b(a)(2)(ii).
5. The word "form" should be changed to "foam" in the fourth line
of § 60.112b(a)(2)(i)(A).
Response: The requirement that deck penetrations on contact decks
extend to the liquid has been deleted. The Agency recognizes that this
requirement is not necessary because there is no vapor space under the
deck and all related changes have been made in the final standards. The
typographical errors noted by the commenter have been changed in the
final standards.
Comment: One commenter (IV-D-13) said that if there is no reason
not to allow the use of control options defined in §§ 60.112b(a)(l),
(2), and (3) for tanks storing high vapor pressure liquids defined in
§ 60.112(b), then § 60.112b(a) can be rewritten as follows:
"The owner or operator of each storage vessel with a design capacity
greater than or equal to 151 m3 containing VOL that, as stored, has a
maximum true vapor pressure equal to or greater than 3.5 kPa or a design
capacity greater than or equal to 75 m3 but less than 151 m3 containing
VOL that, as stored, has a maximum true vapor pressure equal to or
greater than 27.5 kPa, shall equip each storage vessel with one of the
following . . ." and § 6Q.112b(b) can then be deleted.
Response: The modification suggested by the commenter is not
equivalent in meaning to the original language. The suggested wording
would delete the control requirement for pressure vessels equipped with
closed vent systems and for control devices storing high vapor pressure
2-58
-------
liquids (76.6 kPa [11.1 psia] or greater). Therefore, the Agency has
not revised the standards.
Comment: One commenter (IV-D-4) suggested that "external floating
roof" be defined in § 60.111b rather than in § 60.112b(a)(2).
Response: The definitions in § 60.1116 are genera] in nature and
apply to the subpart as a whole. The term "external floating roof" is
best described in the section (§ 60.112b(a)(2)) describing control
techniques, and no changes have been made in the final standards.
Comment: One commenter (IV-D-7) made several minor editorial
comments on the Volume I BID.
Response: None of the changes suggested by the commenter will have
any impact on the selection or form of the final standards. Therefore,
the Volume I BID is considered a final document, and no changes have
been made.
2-59
------- |