United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
March 1980
Air
VOC Fugitive Draft
Emissions in Synthetic EIS
Organic Chemicals
Manufacturing Industry
Background Information
for Proposed Standards
Preliminary
Draft
-------
NOTICE
This document has not been formally released by EPA and should not now be construed to represent
Agency policy. It is being circulated for comment on its technical accuracy and policy implications.
VOC Fugitive Emissions
in Synthetic Organic Chemicals
Manufacturing Industry
Background Information
for Proposed Standards
Emission Standards and Engineering Division
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air, Noise, and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carojina 27711
March 1980
-------
This report has been revised by the Emission Standards and
Engineering Division of the Office of Air Quality Planning and
Standards, EPA, and approved for publication. Mention of trade
names or commercial products is not intended to constitute
endorsement or recommendation for use. Copies of this report
are available through the Library Services Officer (MD-35),
U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina 27711, or from the National Technical
Information Services, 5285 Port Royal Road, Springfield,
Virginia 22161.
-------
METRIC CONVERSION TABLE
In keeping with U.S. Environmental Protection Agency policy, metric
units are used in this report. These units may be converted to common
English units by using the following conversion factors:
Metric Unit
m
m
m,
m"
Kg
Mg
Gg
GJ
GJ
J/g
Nm /sec
m/s
Metric Name
LENGTH
meter
meter
VOLUME
1i ters
cubic meters
cubic meters
HEIGHT
kilogram (106 grams)
megagram (10g grams)
gigagram (10 grams)
ENERGY
giga joule
gigajoule
joule per gram
VOLUMETRIC FLOW -.
normal cubic meters per second
SPEED
meters per second
Equivalent
English Unit
39.3700 in.
3.2810 ft.
0.2.642 U.S. gal
264.2 U.S. gal
6.29 Barrels (bbl)
2.2046 Ib.
1.1023 tons
1,102.3 tons
9.48 X 10 Btu
277.76 KWh
0.430 Btu/lb.
2242 SCFM (ft /min)
196.86 ft/min
Temperature in degrees Celcius (°C) can be converted to temperature
in degrees Farenheit (°F) by the following formula:
(°F) = 1.8 (°C) + 32
-------
TABLE OF CONTENTS
Page
LIST OF FIGURES x
LIST OF TABLES . .- vi
ABBREVIATIONS AND CONVERSION FACTORS . .
2. . INTRODUCTION ... 2-1
2.1 Authority for the Standards 2-1
2.2 Selection of Categories of Stationary Sources 2-5
2.3 Procedure for Development of Standards of
Performance 2-7
2.4 Consideration of Costs 2-8
2.5 Consideration of Environmental Impacts 2-10
2.6 Impact on Existing Source 2-11
2.7 Revision of Standards through Experience 2-12
3. DESCRIPTION OF FUGITIVE EMISSION SOURCES 3-1
3.1 Introduction and General Industry Information . 3-1
3.2 Fugitive Emission Definition and Potential 3-3
3.3 Baseline Control 3-17
3.4 References 3-21
'4. EMISSION CONTROL TECHNIQUES 4-1
4.1 Leak Detection and Repair Methods 4-1
4.2 Equipment Specifications 4-13
4.3 References. 4-24
5. MODIFICATION AND RECONSTRUCTION . 5-1
5.1 General Discussion of Modification
and Reconstruction 5-1
5.2 Applicability of Modification and Reconstruction
Provisions to the SOCMI .5-3
6. MODEL PROCESS UNITS AND REGULATORY ALTERNATIVES 6-1
6.1 Model Units 6-1
6.2 Regulatory Alternatives 6-4
6.3 References 6-7
-------
TABLE OF CONTENTS (cont.)
Page
7. ENVIRONMENTAL IMPACT 7-1
7.1 Impact on Atmospheric Emissions 7-1
7.2 Impact of Water Quality 7-8
7.3 Impact on Solid Waste 7-11
7.4 Enerqy Impact 7-12
7.5 Other Environmental Concerns 7-14
7.6 References 7-14
8. COST ANALYSIS 8-1
8.1 Cost Analysis of Regulatory Alternatives 8-1
8.2 Other Cost Considerations . 8-24
8.3 References 8-25
9. ECONOMIC ANALYSIS 9-1
9.1 Industry Profile 9-1
9.2 Economic Impact Analysis 9-16
9.3 Socio-Economic and Inflationary Impacts 9-40
9.4 References 9-41
APPENDIX C C-l
APPENDIX D D-l
APPENDIX E E-l
APPENDIX F F-l
-------
LIST OF TABLES
Page
3-1 Approximate Level of Uncontrolled Fugitive Emission
Factors in the Synthetic Organic Chemical Manufacturing
Industry (SOCMI) . . 3-18
4-1 Fraction of Total Mass Emissions From Various Source
Types That Would be Controlled by Different Action Levels 4-7
4-2 Estimated Occurrence and Recurrence Rate for Various
Monitoring Intervals 4-10
. 4-3 Percent of Mass Emissions Affected by Various Repair
Intervals 4-11
4-4 Average Emission Rates from Sources Above 10,000 ppmv
and at 1000 ppmv 4-11
4-5 Example of Control Efficiency Calculation 4-15
4-6 Impact of Monitoring Interval on Correction Factor Accounting
for Leak Occurrence/Recurrence (For Example Calculation) 4-15
4-7 Effectiveness of Equipment Modifications 4-23
6-1 Fugitive Emission Sources for Three Model Units 6-3
6-2 Regulatory Alternatives for NSPS for Fugitive Emission
Sources in the SOCMI 6-5
7-1 Emission Factors for Sources Controlled Under Regulatory
Alternative II 7-3
7-2 Emission Factors for Sources Controlled Under Regulatory
Alternative III 7-4
7-3 Emission Factors for Sources Controlled Under Regulatory
Alternative IV 7-5 .
7-4 Example Calculation of VOC Fugitive Emissions from Model
Unit A Under Regulatory Alternative II 7-6
7-5 Estimated Emissions and Emission Reduction on A Model
Unit Basis 7-7
7-6 Total VOC Fugitive Emissions from Affected Model Units
for Regulatory Alternatives 7-10
-------
LIST OF TABLES (cont.)
Page
8-1 Capital Cost Data 8-2
8-2 Capital Cost Estimates for New Model Units 8-3
8-3 Annual Monitoring and Leak Repair Labor Requirements for
Regulatory Alternative II 8-5
8-4 Annual Monitoring and Leak Repair Labor Requirements for
Regulatory Alternative III 8-6
8-5 Annual Monitoring, and Leak Repair Labor Requirements for
Regulatory Alternative IV 8-7
8-6 Derivation of Annualized Labor, Administrative, Maintenance,
and Capital Charges 8-8
8-7 Labor-Hour Requirements for Initial Leak Repair 8-11
8-8 Recovery Credits 8-12
8-9 Annualized Control Cost Estimated for Model Unit A 8-13
8-10 Annualized Control Cost Estimated for Model Unit B 8-14
8-11 Annualized Control Cost Estimated for Model Unit C 8-15
8-12 Cost Effectiveness for Model Units 8-16
8-13 Capital Cost Estimates for Modified/Reconstructed
Facilities 8-18
8-14 Annualized Control Cost Estimated for Modified/
Reconstructed Model Units Under Regulatory Alternative IV 8-19
8-15 Nationwide Costs for the Industry Under Regulatory
Alternative II 8-21
8-16 Nationwide Costs for the Industry Under Regulatory
Alternative III 8-22
8-17 Nationwide Costs for the Industry Under Regulatory
Alternative IV 8-23
8-18 Statutes That May be Applicable to SOCMI 8-25
9-1 Estimated Plant Capacity By State 9-3
-------
LIST OF. TABLES (cont.)
Page
9-2 Distribution of Plants by Capacity and Region . . .9-4
'
9-3 Distribution of Industry Capacity By Plant Size and Region .... 9-6
9-4 Production and Sales of Synthetic Organic Chemicals 9-7
9-5 SOCMI Resource Use 9-9 -
9-6 Industrial Organic Chemicals: U.S. Imports and Exports,
1966-77 9-10
9-7 Industrial Organic Chemicals: U.S. Trade, By Principal
Trading Partners, 1976 and 1977 9-13
9-8 Industrial Organic Chemicals: U.S. Imports For Consumption
By Principal Source, 1972-77 9-14
9-9 Industry Concentration 9-18
9-10 Estimated Cost of Capital for Firms in SOCMI 9-19
9-11 Average Rate of Return Impacts 9-28
9-12 Model Units Experiencing Significant Rate of Return Impacts
Under Full Cost Absorption 9-30
9-13 Average Price Impacts of Regulatory Alternatives. 9-32
9-14 Model Units Requiring Significant Price Increases to
Maintain Target Rates of Return 9-33
9-15 Investment Impacts of Regulatory Alternatives . . . ... . ... . . 9-36
9-16 Employment Impacts of Regulatory Alternatives 9-38
9-17 Model Unit and Industry Annualized Control Costs 9-39
C-l Frequency, of Leaks from Sources In Synthetic Organic
Chemical Plants C-4
C-2 Sampled Process Units from Nine Refineries During
Refinery Study . . £--5 ,
C-3 Leak Frequencies and Emission Factors from Fugitive Sources
in Petroleum Refineries C-7
vm
-------
LIST OF TABLES (cont.)
Page
C-4 Comparison of Leak Frequencies for Fugitive Emission
Sources in SOCMI Units and Petroleum Refineries C-8
C-5 Summary of Maintenance Study Results from Union Oil Company
Refinery in Rodeo, California C-10
C-6 Summary of Maintenance Study Results from the Shell Oil
Company Refinery in Martinez, California C-12
C-7 Summary of EPA Refinery Maintenance Study Results C-14
C-8 Unit D-Ethylene Unit Block Valve Repair C-15
El-1 SOCMI Chemicals Included in ITC Category Ratios E-3
El-2 Ratios Used To Weight ITC Data E-5
E2-1 Projections of Replacement Investment E-8
E3-1 Yields by Rating Class for Cost of Debt Funds, 1979 E-12
E3-2 Financial Data for 100 Firms in SOCMI E-14
-------
LIST OF FIGURES
3-1 General schematic of process levels that make up the organic
chemical industry ...... . ................. 3-2
3-2 Diagram of a simple packed seal ................. 3-4
3-3 Diagram of a basic single mechanical seal-.-., ....... ... 3-5
3-4 Diagram of a double mechanical seal (back-to-back arrangement) . 3-6
3-5 Diagram of a double mechanical seal (tandem arrangement) .... 3-6
3-6 Chempump canned-motor pump .................. . 3-7
3-7 Shriver mechanically actTFated-daap.hragm pump ......... t 3-8
""' "''"*' -- -f.^.
,3-8 Liquid-film compressor shaft seal. ...... . ; = \-., , ..... 3-9
3-9 Diagram of a gate valve ..................... 3-10::
3-10 Example of bellows seals ....... . . ........... 3-11
3-11 Diagrams of valves with diaphragm seals ..... . ....... 3-12
.3-12 Diagram of a spring-loaded relief valve ............. 3-13
'3-13 Cooling tower (cross-flow) ................... 3-14
3-14 Diagram of hydraulic seal for agitators .......... ... 3-15
3-15 Diagram of agitator lip seal ............ ...... 3-16
4-1 Cumulative distribution of total emissions by screening values -
valves - gas/vapor streams ................... 4-16
4-2 Cumulative distribution of sources by screening values -
valves - gas/vapor streams ................... 4-16
-------
2. INTRODUCTION
Standards of performance are proposed following a detailed investi-
gation of air pollution control methods available to the affected industry
and the impact of their costs on the industry. This document summarizes
the information obtained from such a study. Its purpose is to explain in
detail the background and basis of the proposed standards and to facilitate
analysis of the proposed standards by interested persons, including those
who may not be familiar with the many technical aspects of the industry.
To obtain additional copies of this document or the Federal Register
notice of proposed standards, write to EPA Library (MD-35), Research
Triangle Park, North Carolina, 27711. When ordering, specify the Back-
ground Information Document (BID), Volume 1: Proposed Standards of
Performance for the Synthetic Organic Chemical Manufacturing Industry.
2.1 AUTHORITY FOR THE STANDARDS
Standards of performance for new stationary sources are established
under Section 111 of the Clean Air Act (42 United States Code 7411), as
amended, hereafter referred to as the Act. Section 111 directs the
Administrator to establish standards of performance for any category of
new stationary source of air pollution which "...causes or contributes
significantly to health or welfare."
The term new source is defined as "any stationary source, the
construction or modification of which is commenced after the publication
of regulations (or, if earlier, proposed regulations) prescribing a
standard of performance under this section which will be applicable to
such source."
The term stationary source is further defined as "any building,
structure, facility or installation which emits or may emit any air
pollutant."
2-1
-------
The term standard of performance as applied to stationary sources
(other than fossil-fuel-fired sources) is defined as an "allowable emis-
sion limitation for such category of sources." The Act requires that
standards of performance for stationary sources reflect, "...the degree
of emission limitation achievable through the application of the best
technological system of continuous emission reduction. . . the Administra-
tor determines has been adequately demonstrated." In addition, for
stationary sources whose emissions result from fossil fuel combustion,
the standard must also include a percentage reduction in emissions. The
Act also provides that the cost of achieving the necessary emission
reduction, the non-air quality health and environmental impacts and the
energy requirements all be taken into account in establishing standards
of performance.
The term technological system of continuous emission reduction is
interpreted as either:
1) "a technological process for production or operation by any
source which is inherently low polluting or nonpolluting", or
2) "a technological system for continuous reduction of the
pollution generated by a source before such pollution is
emitted into the ambient air, including precombustion
cleaning or treatment of fuels."
If a standard of performance as defined above cannot be prescribed
or enforced, "the Administrator . . . may distinguish among classes,
types, and sizes within categories of new sources for the purposes of
establishing such standards." This allows certain types and sizes of
facilities to be exempted from compliance with a general standard, or to
have a different standard of performance specified. This might be done,
for example, to avoid extreme economic hardship on very small facilities.
Section 111 prescribes three steps to follow in establishing
standards of performance.
1) The Administrator must identify those categories of stationary
sources for which standards of performance will ultimately
be promulgated by listing them in the Federal Register. For
those categories of major stationary sources which have not
2-2
-------
already been listed, the following schedule has been specified
for the promulgation of standards: 25 percent by August 7,
1980, 75 percent by August 7, 1981, and 100 percent by August 7,
1982.
2) The regulations applicable to a category so listed must be
proposed by publication in the Federal Register within 120
days of its listing. This proposal provides interested persons
an opportunity to comment.
3) Within six months after proposal, the standard must be
promulgated, incorporating any alterations deemed necessary or
desirable.
It is further required that standards of performance be reviewed
every four years. If there has been significant change in the industry or
control technology, then the standard must be revised to reflect the new
condition.
Standards of performance do not guarantee protection of health or
welfare because they are not designed to achieve any specific air quality
levels. They are designed to reflect the degree of emission limitation
achievable through application of the best adequately demonstrated tech-
nological system of continuous emission reduction. In this application
the cost of achieving such emission reduction, non-air quality health and
environmental impacts, and energy requirements should be considered.
Congress had several reasons for including these requirements. First,
standards with a degree of uniformity are needed to avoid situations where
some states may attract industries by relaxing standards relative to other
states. Second, stringent standards enhance the potential for long-term
growth. Third, stringent standards may help achieve long-term cost savings
by avoiding the need for more expensive retrofitting when pollution
ceilings may be reduced in the future. Fourth, the standard-setting
process should create incentives for improved technology.
Promulgation of standards of performance does not prevent state or
local agencies from adopting more stringent emission limitations for the
same sources. States are free under Section 116 of the Act to establish
even more stringent emission limits than those established under Section
2-3
-------
Ill or those necessary to attain or maintain the National Ambient Air
Quality Standards (NAAQPS) under Section 110. Thus, new sources may in
some cases be subject to limitations more stringent than standards of
performance under Section 111, and prospective owners and operators of
new sources should be aware of this possibility in planning for such
facilities.
A similar situation may arise when a major emitting facility is to
be constructed in a geographic area which falls under the prevention of
significant deterioration of air quality provisions of Part C of the Act.
These provisions stipulate, among other things, that major emitting facil-
ities to be constructed in such areas are to be subject to best available
control technology. The term "Best Available Control Technology (BACT),
as defined in the Act, means "...an emission limitation based on the
maximum degree of reduction of each pollutant subject to regulation under
this Act emitted from or which results from any major emitting facility,
which the permitting authority, on a case-by-case basis, taking into
account energy, environmental, and economic impacts and other costs,
determines is achievable for such facility through application of pro-
duction processes and available methods, systems, and techniques,
including fuel cleaning or treatment or innovative fuel combustion tech-
niques for control of each such pollutant. In no event shall application
of "Best Available Control Technology1 result in emissions of any pollu-
tants which will exceed the emissions allowed by any applicable standard
established pursuant to Section 111 or 112 of this Act."
In addition, Section lll(h) authorizes the Administrator to grant
waivers of compliance to permit a source to use innovative continuous
emission control technology. In order to grant the waiver, the Adminis-
trator must find: (1) a substantial likelihood that the technology will
produce greater emission reductions than the standards require, or an
equivalent reduction at lower economic, energy or environmental costs;
(2) the proposed system has not been adequately demonstrated; (3) the
technology will not cause or contribute to an unreasonable risk to public
health, welfare or safety; (4) the governor of the state where the source
is located consents; and that (5) the waiver will not prevent the
2-4
-------
attainment or maintenance of any ambient standard. A waiver may have
conditions attached to assure the source will not prevent attainment of
any NAAQPS. Any such condition will have the force of a performance
standard. Finally, waivers have definite end dates and may be terminated
earlier if the system fails to perform as expected. In such a case, the
source may be given up to three years to meet the standards, with a
mandatory progress schedule.
2.2 SELECTION OF CATEGORIES OF STATIONARY SOURCES
Section 111 of the Act directs the Administrator to list categories
of stationary sources which have not been listed before. The Administrator
"...shall include a category of sources in such a list if in his judgment
it causes, or contributes significantly to, air pollution which may
reasonably be anticipated to endanger public health or welfare." Proposal
and promulgation of standards of performance are to follow while adhering
to the schedule referred to earlier.
Since passage of the Clean Air Amendments of 1970, considerable
attention has been given to the development of a system for assigning
priorities to various source categories. The approach specifies areas of
interest by considering the broad strategy of the Agency for implementing
the Clean Air Act. Often, these "areas" are actually pollutants which
are emitted by stationary sources. Source categories which emit these
pollutants were then evaluated and ranked by a process involving such
factors as: (1) the level of emission control (if any) already required
by state regulations; (2) estimated levels of control that might be
required by standards of performance for the source category; (3) pro-
jections of growth and replacement of existing facilities for the source
category; and (4) the estimated incremental amount of air pollution that
could be prevented, in a preselected future year, by standards of per-
formance for the source category. Sources for which new source performance
standards were promulgated or were under development during 1977 or earlier
were selected on these criteria.
The Act amendments of August 1977 establish specific criteria to be
used in determining priorities for all source categories not yet listed
by EPA. These are: (1) the quantity of air pollutant emissions which
2-5
-------
each such category will emit, or will be designed to emit; (2) the extent
to which each pollutant may reasonably be anticipated to endanger public
health or welfare; and (3) the mobility and competitive nature of each
category of sources and the consequent need for nationally applicable new
source standards of performance.
In some cases, it may not be feasible to immediately develop a
standard for a source category with a high priority. This might happen
when a program of research is needed to develop control techniques or
because techniques for sampling and measuring emissions may require refine-
ment. In the development of standards, differences in the time required
to complete the necessary investigation of different source categories
must also be considered. For example, substantially more time may be
necessary if numerous pollutants must be investigated from a single source
category. Further, even late in the development process the schedule for
completion of a standard may change. For example, inability to obtain
emission data from well-controlled sources in time to pursue the develop-
ment process in a systematic fashion may force a change in scheduling.
Nevertheless, priority ranking is, and will continue to be, used to
establish the order in which projects are initiated and resources
assigned.
After the source category has been chosen, determining the types of
facilities within the source category to which the standard will apply
must be decided. A source category may have several facilities that
cause air pollution and emissions from some of these facilities may be
insignificant or very expensive to control. Economic studies of the
source category and of applicable control technology may show that air
pollution control is better served by applying standards to the more
severe pollution sources. For this reason, and because there is no
adequately demonstrated system for controlling emissions from certain
facilities, standards often do not apply to all air pollutants emitted.
Thus, although a source cateogry may be selected to be covered by a
standard of performance, not all pollutants or facilities within that
source category may be covered by the standards.
2-6
-------
2.3 PROCEDURE FOR DEVELOPMENT OF STANDARDS OF PERFORMANCE
Standards of performance must: (1) realistically reflect the best
demonstrated control practice; (2) adequately consider the cost, and the
non-air quality health and environmental impacts and energy requirements
of such control; (3) be applicable to existing sources that are modified
or reconstructed as well as new installations; and (4) meet these condi-
tions for all variations of operating conditions being considered anywhere
in the country.
The objective of a program for development of standards is to identify
the best technological system of continuous emission reduction which has "
been adequately demonstrated. The legislative history of Section 111 on
what has been adequately demonstrated is not limited to systems that are
in actual routine use. The search may include a technical assessment of
control systems which have been adequately demonstrated but for which
there is limited operational experience. In most cases, determination of
the "...degree of emission reduction achievable..." is based on results
of tests of emissions from well controlled existing sources. At times,
this has required the investigation and measurement of emissions from
control systems found in other industrialized countries that have developed
more effective systems of control than those available in the United
States.
Since the best demonstrated systems of emission reduction may not be
in widespread use, the data base upon which standards are developed may be
somewhat limited. Test data on existing well-controlled sources are
obvious starting points in developing emission limits for new sources.
However, since the control of existing sources generally represents
retrofit technology or was originally designed to meet an existing state
or local regulation, new sources may be able to meet more stringent
emission standards. Accordingly, other infomation must be considered
before a judgment can be made as to the level at which the emission
standard should be set.
A process for the development of a standard has evolved which
considers the following:
2-7
-------
1. Emissions from existing well-controlled sources as measured.
2. Data on emissions from such sources are assessed with consi-
deration of such factors as: (a) how representative is the
tested source with regard to feedstock, operation, size, age,
etc.; (b) age and maintenance of the control equipment tested;
(c) design uncertainties of control equipment being considered;
and (d) the degree of uncertainty that new sources will be able
to achieve similar levels of control.
3. Information from pilot and prototype installations, guarantees
by vendors of control equipment, unconstructed but contracted
projects, foreign technology, and published literature are also
considered during the standard development process. This is
especially important for sources where "emerging" technology
appears to be a significant alternative.
4. Where possible, standards are developed which permit the use
of more than one control technique or licensed process.
5. Where possible, standards are developed to encourage or permit
the use of process modifications or new processes as a method
of control rather than "add-on" systems of air pollution control.
6. In appropriate cases, standards are developed to permit the
use of systems capable of controlling more than one pollutant.
7. Where appropriate, standards for visible emissions are developed
in conjunction with concentration/mass emission standards. The
opacity standard is established at a level that will require
proper operation and maintenance of the emission control system
installed to meet the concentration/mass standard on a day-to-day
basis. In some cases, however, it is not possible to develop
concentration/mass standards, such as with fugitive sources of
emissions. In these cases, only opacity standards may be
developed to limit emissions.
2.4 CONSIDERATION OF COSTS
Section 317 of the Act requires, among other things, an economic
impact assessment with respect to any standard of performance established
2-8
-------
under Section 111 of the Act. The assessment is required to contain an
analysis of:
1. The costs of compliance with the regulation and standard
including the extent to which the cost of compliance varies
depending on the effective date of the standard or regulation
and the development of less expensive or more efficient methods
of compliance;
2. The potential inflationary or recessionary effects of the
standard or regulation;
3. The effects of the standard or regulation on competition
. among small businesses;
4. The effects of the standard or regulation on consumer cost,
and
5. The effects of the standard or regulation on energy use.
Section 317 requires that the economic impact assessment be as
extensive as practical, taking into account the time and resources
available to EPA.
The economic impact of a proposed standard upon an industry is
usually addressed both in absolute terms and by comparison of the costs of
typical existing state control regulations with the control costs that
would be incurred as a result of complying with the standard. An incre-
mental approach is taken since both new and existing plants would be
required to comply with state regulations in the absence of a federal
standard of performance. This approach requires a detailed analysis of
the impact upon industry resulting from the cost differential that exists
between a standard of performance and the typical state standard.
The costs of controlling air pollutants are not the only costs con-
sidered in analyzing the economic impacts of the proposed standard. The
costs associated with the control of water pollutants and solid wastes
are also analyzed wherever possible.
A thorough study of the profitability and price-setting mechanism of
the industry is essential to the analysis so that an accurate estimate of
potential adverse economic impacts can be made. It is also essential to
know the capital requirements placed on plants in the absence of federal
standards of performance so that the additional capital requirements
2-9
-------
necessitated by these standards can be placed in the proper perspective.
Finally, it is necessary to recognize any constraints on capital availa-
bility with an industry, as this factor also influences the ability of
new plants to generate the capital required for installation of additional
control equipment needed to meet the standards of performance.
2.5 CONSIDERATION OF ENVIRONMENTAL IMPACTS
Section 102(2)(C) of the National Environmental Policy Act (NEPA) of
1969 required federal agencies to prepare detailed environmental impact
statements on proposals for legislation and other major federal actions
significantly affecting the quality of the human environment. The
objective of NEPA is to build into the decision-making process of federal
agencies a careful consideration of all environmental aspects of proposed
actions.
In a number of legal challanges to standards of performance for
various industries, the Federal Court of Appeals have held that environ-
mental impact statements need not be prepared by the Agency for proposed
actions under Section 111 of the Clean Air Act. Essentially, the Federal
Court of Appeals have determined that "...the best system of emission
reduction,...require(s) the Administrator to taken into account counter-
productive environmental effects of a proposed standard, as well as
economic costs to the industry..." On this basis, therefore, the Courts
"...established a narrow exemption from NEPA for EPA determination under
Section 111."
In addition to these judicial determinations, the Energy Supply and
Environmental Coordination Act (ESECA) of 1974 (PL-93-319) specifically
exempted proposed actions under the Clean Air Act from NEPA requirements.
According to Section 7(c)(l), "No action taken under the Clean Air Act
shall be deemed a major federal action significantly affecting the quality
of the human environment within the meaning of the National Environmental
Policy Act of 1969."
The Agency has concluded, however, that the preparation of environ-
mental impact statements could have beneficial effects on certain
regulatory actions. Consequently, while not legally required to do so by
Section 102(2)(C) of NEPA, environmental impact statements will be
2-10
-------
prepared for various regulatory actions, including standards of performance
developed under Section 111 of the Act. This voluntary preparation of
environmental impact statements, however, in no way legally subjects the
Agency to NEPA requirements.
To implement this policy, a separate section is included in this
document which is devoted solely to an analysis of the potential environ-
mental impacts associated with the proposed standards. Both adverse and
beneficial impacts in such areas as air and water pollution, increased
solid waste disposal, and increased energy consumption are identified and
discussed.
2.6 IMPACT ON EXISTING SOURCES
Section 111 of the Act defines a new source as "...any stationary
source, the construction or modification of which is commenced..." after
the proposed standards are published. An existing source becomes a new
source if the source is modified or reconstructed. Both modification and
reconstruction are defined in amendments to the general provisions of
Subpart A of 40 CFR Part 60 which were promulgated in the Federal Register
on December 16, 1975 (40 FR 58416). Any physical or operational change
to an existing facility which results in an increase in the emission rate
of any pollutant for which a standard applies is considered a modification.
Reconstruction, on the other hand, means the replacement of components of
an existing facility to the extent that the fixed capital cost exceeds 50
percent of the cost of constructing a comparable entirely new source and
that it be technically and economically feasible to meet the applicable
standards. In such cases, reconstruction is equivalent to new construc-
tion.
Promulgation of a standard of performance requires states to establish
standards of performance for existing sources in the same industry under
Section lll(d) of the Act if the standard for new sources limits emissions
of a designated pollutant (i.e., a pollutant; for which air quality criteria
have not been issued under Section 108 or which has not been listed as a
hazardous pollutant under Section 112). If a state does not act, EPA
must establish such standards. General provisions outlining procedures
for control of existing sources under Section lll(d) were promulgated on
November 17, 1975, as Subpart B of 40 CFR Part 60 140 FR 53340).
2-11
-------
2.7 REVISION OF STANDARDS THROUGH EXPERIENCE
Congress was aware that the level of air pollution control achievable
by any industry may improve with technological advances. Accordingly,
Section 111 of the Act provides that the Administrator "...shall, at
least every four years, review and, if appropriate, revise..." the
standards. Revisions are made to assure that the standards continue to
reflect the best systems that become available in the future. Such
revisions will not be retroactive but will apply to stationary sources
constructed or modified after the proposal of the revised standards.
2-12
-------
3. DESCRIPTION OF FUGITIVE EMISSION SOURCES
3.1 INTRODUCTION AND GENERAL INDUSTRY INFORMATION
3.1.1 Introduction
The primary purposes of this chapter are to define the synthetic
organic chemical manufacturing industry (SOCMI) and describe the potential
fugitive emission sources that are typically found in this industry. Where
possible, the leak rates of uncontrolled emissions from the various poten-
tial fugitive emission sources are quantified. Industrial practices and
state or local regulations, that currently reduce fugitive emissions from
the SOCMI are also briefly discussed in this chapter.
3.1.2 General Information
Organic chemicals are manufactured in a multi-leveled system of
chemical processes that is based on about ten feedstock chemicals which
are principally produced in petroleum refineries. These feedstocks then
proceed through one or more of the process levels and result in literally
thousands of intermediate or finished chemicals (see Figure 3-1).
Generally, each process level contains more chemicals than the preceding
level; the plants manufacturing the products are smaller than the plants
supplying the feedstock; and the volatilities of the products are lower
than the volatilities of the feedstocks. Because of the number and
diverse nature of the organic chemicals included in the multi-leveled
system, the synthetic organic chemical manufacturing industry (SOCMI)
is defined, for this study, to consist of 378 of the higher volume, higher
volatility intermediate and finished products. A list of the 378
chemicals is presented in Appendix F.
3-1
-------
RAW MATERIALS
(CRUDE OIL. CRUDE NATURAL GAS. ETC)
REFINERIES
CHEMICAL
FEEDSTOCK
PLANTS
CHEMICAL
FEEDSTOCKS
CHEMICAL
PLANTS
CHEMICAL
PRODUCTS
Figure 3-1. General schematic of process levels that make up
the organic chemical industry.
3-2
-------
Although there are organic chemical manufacturing plants in most
industrialized areas of the country, about 60 percent of the SOCMI volume
is produced in Texas and Louisiana. Each plant site may manufacture from
one to several organic chemicals using one or more processes. Although
most processes result in one basic product, some produce a family of
chemicals. Conversely, many chemicals are produced by more than one
process. Yearly, production quantities at each plant can range from a
few million to several billion kilograms.
3.2 FUGITIVE EMISSION DEFINITION AND POTENTIAL SOURCE DESCRIPTION
3.2.1 Definition
In this study, fugitive emissions in the SOCMI are considered to be
those volatile organic compound (VOC) emissions that result when process
fluid (either liquid or gaseous) leaks from plant equipment. Those VOC
emissions resulting from the transfer, storage, treatment, and/or disposal
of process wastes will be covered by other standards.
3.2.2 Potential Source Characterization and Description
There are many potential sources of fugitive emissions in a typical
synthetic organic chemical plant. The following sources will be con-
sidered in this chapter: pumps, compressors, in-line process valves,
pressure relief devices, open-ended valves, sampling connections, flanges,
agitators, and cooling towers. These potential sources are described below.
3.2.2.1 Pumps. Pumps are used extensively in the SOCMI for the
movement of organic liquids.1 The centrifugal pump is the most widely
used pump in the SOCMI; however, other types, such as the positive-
displacement, reciprocating and rotary action, and special canned and
diaphragm pumps, are also used in this industry. Chemicals transferred
by pumps can leak at the point of contact between the moving shaft and
stationary casing. Consequently, all pumps except the shaftless type
(canned-motor and diaphragm) require a seal at the point where the shaft
penetrates the housing in order to isolate the pump's interior from the
atmosphere.
Two generic types of seals, packed and mechanical, are currently in
use on pumps in the SOCMI. Packed seals can be used on both reciprocating
3-3
-------
and rotary action types of pumps. As Figure 3-2 shows, a packed seal
consists of a cavity ("stuffing box") in the pump casing filled with
special packing material that is compressed with a packing gland to form a
seal around the shaft. Lubrication is required to prevent the buildup of
frictional heat between the seal and shaft. The necessary lubrication is
provided by a lubricant that flows between the packing and the shaft.2
Deterioration of the packing will result in process liquid leaks.
box.
FLUID
E-kJO
1
_j
xixixjXTxixixr
^ SEAL FA
-teS
\
,/^C*\.**JQ
) 5HAfT
'tXIXIXIXJXiXiXI
/
0
^^-
POiMBLE.
LC.AK.
Figure 3-2. Diagram of a simple packed seal.3
Mechanical seals are limited in application to pumps with rotating
shafts and can be further categorized as single and double mechanical
seals. There are many variations to the basic design of mechanical seals,
but all have a lapped seal face between a stationary element and a
rotating seal ring. In a single mechanical seal application (Figure 3-3),
the rotating-seal ring and stationary element faces are lapped to a very
high degree of flatness to maintain contact throughout their entire
mutual surface area. As with a packed seal, the seal faces must be
lubricated to remove frictional heat; however, because of its construc-
tion, much less lubricant is needed.
A mechanical seal is not a leak-proof device. Depending on the
condition and flatness of the seal faces, the leakage rate can be quite
low (as small as a drop per minute) and the'flow is often not visually
3-4
-------
detectable. In order to minimize fugitive emissions due to seal leakage,
an auxiliary sealing device such as packing can be employed.**
PUMP
UA.VJD
RltOO,
KJSER.T PXCKJUfi
FCUID
EXlO
S.TA.TIOUARV
EUE.MEJJT
POSil&LE.
L.C.A.K.
Figure 3-3. Diagram of a basic single mechanical seal. 5
In a double mechanical seal application, two seals can be arranged
back-to-back or in tandem. In the back-to-back arrangement (Figure 3-4),
the two seals provide a closed cavity between them. A seal liquid, such
as water or seal oil, is circulated through the cavity. Because the seal
liquid surrounds the double seal and lubricates both sets of seal faces
in this arrangement, the heat transfer and seal life characteristics are
much better than those of the single seal. In order for the seal to
function, the seal liquid must be at a pressure greater than the operating
pressure of the stuffing box. As a result some seal liquid will leak
across the seal faces. Liquid leaking across the inboard face will enter
the stuffing box and mix with the process liquid. Seal liquid going
across the outboard face will exit to the atmosphere. Therefore, the seal
liquid must be compatible with the process liquid as well as with the
environment.6
In a tandem double mechanical seal arrangement (Figure 3-5), the
seals face the same direction. The secondary seal provides a backup for
the primary seal. A seal flush is used in the stuffing box to remove the
heat generated by friction. The cavity between the two seals is filled
3-5
-------
Stuffing-box
housing'^-,,.'
Pumped liquid
Inner mating ring-
Inner primary
ring
___-Gland plate
Shaft
Outer mating ring .
T Outer primary
.../_ ring
Figure 3-4. Diagram of a double mechanical seal
(back-to-back arrangement).7
Stuffing-box Bypass
housing / flush
Buffer liquid,
out
(top)
in
(bottom)
Pumped
liquid
Gland
plate
Outer
primary mating
ring ring
Shaft
Figure 3-5.
Diagram of a double mechanical seal
(tandem arrangement).8
3-6
-------
with a buffer or barrier liquid. However, the barrier liquid is at a
pressure lower than that in the stuffing box. Therefore, any leakage
will be from the stuffing box into the seal cavity containing the barrier
liquid. Since this liquid is routed to a closed reservoir, process liquid
that has leaked into the seal cavity will also be transferred to the
reservoir. At the reservoir, the process liquid could vaporize and be
emitted to the atmosphere. To ensure that VOC's do not leak from the
reservoir, the reservoir can be vented to a control device.9
Another type of pump that has been used in the chemical industry
is the shaftless pump which includes canned-motor and diaphragm pumps.
In canned-motor pumps (Figure 3-6) the cavity housing the motor rotor
and the pump casing are interconnected. As a result, the motor bearings
run in the process liquid and all seals are eliminated. Because the
process liquid is the bearing lubricant, abrasive solids cannot be
tolerated. Canned-motor pumps are being widely used for handling
organic solvents, organic heat transfer liquids, light oils, as well as
many toxic or hazardous liquids, or where leakage is an economic
problem.10
Figure 3-6. Chempump canned-motor pump.11
3-7
-------
Diaphragm pumps (see Figure 3-7) perform similarly to piston and
plunger pumps. However, the driving member is a flexible diaphragm
fabricated of metal, rubber, or plastic. The primary advantage of this
arrangement is the elimination of all packing and seals exposed to the
process liquid. This is an important asset when hazardous or toxic
liquids are handled.12
Suction
Figure 3-7. Shriver mechanically actuated diaphragm pump.
13
3.2.2.2 Compressors. Gas compressors used in the SOCMI are similar
to pumps in that they can be driven by rotary or reciprocating shafts.
Correspondingly, the same types of seals that are used on pumps are
used on compressors to isolate the process gas from the atmosphere.
As with pumps, these seals are likely to be the source of fugitive
emissions from compressors.
In addition to the mechanical seals that can be used on compressors,
centrifugal compressors can be equipped with liquid film seals (Figure
3-8). This seal is formed by a film of oil between the rotating shaft
and stationary gland. The seal oil exits the compressor from chambers
on both sides of the gland. The oil leaving the chamber on the process
3-8
-------
side is under pressure and contaminated with process gas. When the
contaminated oil is returned to the oil reservoir, process gas can be
released and emitted to the atmosphere.11* To eliminate the release of
VOC emissions from the oil reservoir, the reservoir can be vented to a
control device.
OIL. »M
IKJTERWAL,
GAS PRE.S5URE-
COM TAMI KJ ATE.D
OIL. OUT
TO R.£.SEJ=^VOIR
OIL. OUT
Figure 3-8. Liquid-film compressor shaft seal.15
3.2.2.3 Process Valves. One of the most common pieces of equipment
in organic chemical plants is the valve. The types of valves commonly
used are control, globe, gate, plug, ball, relief, and check valves. All
except the relief valve (to be discussed further below) and check valve
are activated by a valve stem, which may have either a rotational or
linear motion, depending on the specific design. This stem requires a
seal to isolate the process fluid inside the valve from the atmosphere as
illustrated by the diagram of a gate valve in Figure 3-9. The possibility
of a leak through this seal makes it a potential source of fugitive
3-9
-------
emissions. Since a check valve has no stem or subsequent packing gland,
it is not considered to be a potential source of fugitive emissions.
Sealing of the stem to prevent leakage can be achieved by packing
inside a packing gland or 0-ring seals. Valves that require the stem to
move in and out with or without rotation must utilize a packing gland.
Conventional packing glands are suited for a wide variety of packing
material; the most Common are various types of braided asbestos that
contain lubricants. Other packing materials include graphite, graphite-
impregnated fibers, and tetrafluorethylene; the packing material used
depends on the valve application and configuration.16 These conventional
packing glands can be used over a wide range of operating temperatures.
At high pressures these glands must be quite tight to attain a good seal.17
PACK-IMG GUAMD
VAUVE1
POiSlSUE.
(-2.AK. AFLEAS
Figure 3-9. Diagram of a gate valve.18
Elastomeric 0-rings are also used for sealing process valves. These
0-rings provide good sealing but are not suitable where there is sliding
motion through the packing gland. Those seals are rarely used in high
pressure service and operating temperatures are limited by the seal
material ,19
3-10
-------
Bellows seals are more effective for preventing process fluid leaks
than the conventional packing gland or any other gland-seal arrangement.20
This type of seal incorporates a formed metal bellows that makes a barrier
between the disc and body bonnet joint. An example of this seal is
presented in Figure 3-10. The bellows is the weak point of the system
and service life can be quite variable. Consequently, this type of seal
is normally backed up with a conventional packing gland and is often fitted
with a leak detector in case of failure.21
Disc
Bellows
Body
Bonnet
Figure 3-10. Example of bellows seals.
22
A diaphragm may be used to isolate the working parts of the valve and
the environment from the process liquid. Two types of valves which utilize
diaphragms are illustrated in Figures 3-11(a) and (b). As Figure 3-11(b)
shows, the diaphragm may also be used to control the flow of the process
fluid. In this design, a compressor component pushes the diaphragm toward
the valve bottom, throttling the flow. The diaphragm and compressor are
connected in a manner so that it is impossible for them to be separated
under normal working conditions. When the diaphragm reaches the valve
3-11
-------
bottom, it seats firmly against the bottom, forming a leak-proof seal.
This configuration is recommended for fluids containing solid particles
and for medium-pressure service. Depending on the diaphragm material,
this type of valve can be used at temperatures up to 205°C and in severe
acid solutions. If failure of the seal occurs, a valve employing a dia-
phragm seal can become a source of fugitive emissions.23
Diaphragm-'
Stem
Diaphragm
(a)
Figure 3-11. Diagrams of valves with diaphragm seals.21*
3.2.2.4 Pressure Relief Devices. Engineering codes require that
pressure-relieving devices or systems be used in applications where the
process pressure may exceed the maximum allowable working pressure of the
vessel. The most common type of pressure-relieving device used in the
SOCMI is the pressure relief valve (Figure 3-12). Typically, relief valves
are spring-loaded and designed to open when the process pressure exceeds a
set pressure, allowing the release of vapors or liquids until the system
pressure is reduced to its normal operating level. When the normal
3-12
-------
pressure is re-attained, the valve reseats, and a seal is again formed.25
The seal is a disk on a seat, and the possibility of a leak through this
seal makes the pressure relief valve a potential source of VOC fugitive
emissions. Two potential causes of leakage from relief valves are:
"simmering or popping", a condition due to the system pressure being
close to the set pressure of the valve, and improper reseating of the
valve after a relieving operation.26
Rupture disks are also common in the SOCMI. These disks are made of
a material that ruptures when a set pressure is exceeded, thus allowing
the system to depressurize. The advantage of a rupture disk is that the
disk seals tightly and does not allow any VOC's to escape from the system
under normal operation. However, when the disk does rupture, the system
depressurizes until atmospheric conditions are obtained; this could result
in an excessive loss of product or correspondingly an excessive release
of fugitive emissions.
SEAT
SPRI-UG
DISK.
WOZ.Z.UE
PROCESS SIDE.
Figure 3-12. Diagram of a spring-loaded relief valve.27
3-13
-------
3.2.2.5 Cooling Towers. Cooling towers (Figure 3-13) are found
in most SOCMI plants. The purpose of these towers is to cool the
plant's process cooling waters which have been heated while removing
heat from various process equipment (reactors, condensers, heat
exchangers). This cooling process is achieved by evaporation when the
process cooling water and air are contacted. Under normal operating
conditions, a cooling tower would not be considered a fugitive emission
source. However, if a leak occurs in the process equipment and if this
equipment is operating at a pressure greater than that of the cooling
water, organic chemicals can leak into the water. When the process water
is recirculated to the cooling tower, these chemicals can be released to
the atmosphere.
Figure 3-13. Cooling tower (cross-flow).28
3.2.2.6 Agitators. Agitators are commonly used in the SOCMI to
stir or blend chemicals, Like pumps and compressors, agitators may leak
organic chemicals at the point where the shaft penetrates the casing.
Consequently, seals are required to minimize fugitive emissions from
agitators. Four seal arrangements are commonly used with agitators; they
include: compression packing (packed seal), mechanical seals, hydraulic
3-14
-------
seals, and lip seals.29 Packed seals for agitators are very similar in
design and application to the packed seals for pumps (Section 3.2.2.1).
Although mechanical seals are more costly than the other three seal
arrangements, they offer a greatly reduced leakage rate to offset their
higher cost. The maintenance frequency of mechanical seals is, also, one-
half to one-fourth that of packed seals.30 In fact, at pressures greater
than 1135.8 kPa (150 psig), the leakage rate and maintenance frequency are
so superior that the use of packed seals on agitators is rare.31 As with
packed seals, the mechanical seals for agitators are similar to the design
and application of mechanical seals for pumps (Section 3.2.2.1).
The hydraulic seal (Figure 3-14) is the simplest and least used
agitator shaft-seal. In this type of seal, an annular cup attached to the
process vessel contains a liquid that is in contact with an inverted cup
attached to the rotating agitator shaft. The primary advantage of this
seal is that it is a non-contact seal. However, this seal is limited to
low temperatures and pressures and can only handle very small pressure
fluctuations. Organic chemicals may contaminate the seal liquid and then
be released into the atmosphere as fugitive emissions.32
Inverted cup.^
Annular cup
a. Hydraulic seal
Figure 3-14. Diagram of hydraulic seal for agitators.33
A lip seal (Figure 3-15) can be used on a top-entering agitator as a
dust or vapor seal. The sealing element is a spring-loaded elastomer;
subsequently, lip seals are relatively inexpensive and easy to install.
Once the seal has been installed the agitator shaft rotates in continuous
3-15
-------
contact with the lip seal. Pressure limits of the seal are 2 to 3 psi
because it operates without lubrication. Operating temperatures are
limited by the characteristics of the elastomer. Fugitive VOC emissions
could be released through this seal when this seal wears excessively or
the operating pressure surpasses the pressure limits of the seal.3k
Lip Seal
Figure 3-15. Diagram of agitator lip seal.
35
3.2.2.7 Open-Ended Valves or Lines. Some valves are installed in a
system so that they function with the downstream line open to the atmos-
phere. Examples are purge valves, drain valves, and vent valves. A
faulty valve seat, or incompletely closed valve would result in leakage
through the valve and fugitive VOC emissions to the atmosphere.
3.2.2.8 Sampling Connections. The operation of a process unit is
checked periodically by routine analyses of feedstocks and products. To
obtain representative samples for these analyses, sampling lines must
first be purged prior to sampling. The purged liquid or vapor is
sometimes drained onto the ground or into a sewer drain, where it can
evaporate and release VOC emissions to the atmosphere.
3.2.2.9 Flanges. Flanges are bolted, gasket-sealed junctions used
wherever pipe or other equipment such as vessels, pumps, valves, and heat
exchangers may require isolation or removal. Normally, flanges are
employed for pipe diameters of 50 mm or greater and are classified by
pressure and face type.
3-16
-------
Flanges may become fugitive emission sources when leakage occurs due
to improperly chosen gaskets or a poorly assembled flange. The primary
cause of flange leakage is due to thermal stress that piping or flanges in
some services undergo; this results in the deformation of the seal between
the flange faces.36
3.3 BASELINE CONTROL
There are presently no federal regulations that specifically reduce
emissions from synthetic organic chemical manufacturing plants. However,
some fugitive emission reduction is achieved by operating practices
currently followed by industry and applicable state or local regulations.
Because these practices and regulations only "incidentally" control
fugitive emissions, they are considered, in this study, to be the baseline
control level. The procedures, specific control techniques, and regula-
tions that make up the baseline control level are discussed below.
Fugitive emissions occurring under the baseline control level are
subsequently considered in this report to be uncontrolled emissions. Data
characterizing the uncontrolled levels of fugitive emissions in the SOCMI
are presently unavailable. However, data of this type have been obtained
for the refining industry. These data are presented in Table 3-1.
Because the operation of the various process equipment in the SOCMI is not
expected to differ greatly from the operation of the same equipment in
the refining industry, it is felt that the refinery fugitive emission data
can be used to approximate the levels of fugitive emissions in SOCMI.
3.3.1 Industrial Practices
The organic chemical industry has been primarily interested in leaks
that are large enough to be physically evident (leaks that can be seen,
heard, or smelled); such leaks are normally repaired to minimize the loss
of product and are, consequently, termed "easily detectable leaks". Fugi-
tive emissions, as they are considered in this report, are also the results
of leaks from process equipment but have considerablj^_small^r_emission
rates than "easily detectable leaks". In the past, SOCMI has generally not
monitored equipment for fugitive emissions nor repaired equipment on the
basis of reducing the level of fugitive emissions. Processes which have
emitted toxic or hazardous compounds have been .exceptions to this rule.
3-17
-------
TABLE 3-1. UNCONTROLLED FUGITIVE EMISSION FACTORS IN THE SYNTHETIC
ORGANIC CHEMICAL MANUFACTURING INDUSTRY (SOCMI)
Fugitive omission source
Uncontrolled emission
factor.9 kg/hr
Pumps
Light liquids
With packed
With
With
With
seals
single mechanical
double mechanical
no seals
d
Heavy Liquids
With packed seals
With single mechanical
With double mechanical
With no seals
Valves (in-line)
Gas .
Light liquid^
Heavy liquid
Safety/relief valves
Gas .
Light liquid.
Heavy liquid
Open-ended valves
Gas .
Light liquidjj
Heavy liquid
Flanges
Sampling connections
Compressors
Cooling towers
Agitators
seals
seals
seals
seals
0.12
0.12
0.12C
0.0
0.020
0.020.
0.020
0.0
0.021
0.010
0.0003
0.16
0.006
0.009
0.025
0.014
0.003
0.0003
0.015
0.44
13.6-1107*
NAf
aThese uncontrolled emission levels are based upon the refinery data presented
in reference 37.
bLight liquid is defined as a fluid with vapor pressure greater than 0.3
kPa at 20°C. This vapor pressure represents the split between kerosene and
naphtha and is based on data presented in reference 37. The averaqe vapor
pressure between these components is approximately 0.04 psi at 68 °F.
GAssumes the inner seal leaks at the same rate as single seal and that the
VOC is emitted from the seal oil degassing vent.
dHeavy liquid is defined as a fluid with vapor pressure less than 0.3 kPa at
20°C. This vapor pressure represents the split between kerosene and
naphtha and is based on data presented in reference 37. The average vapor
pressure between these components is approximately 0.04 psi at 68 °F.
eThese levels are based on cooling tower circulation rates that range from
0.05-3.66 m /sec (714-58,000 GPM). /Ref. 38.
NA = no data available.
3-18
-------
While SOCMI has been primarily concerned with easily detectable
leaks, certain equipment and procedures used in many organic chemical
plants may help to reduce fugitive VOC emissions. For instance, some
plants cap-off or use double block valves on the end of process lines;
either of these procedures will reduce fugitive emissions. In some plants,
relief valves are checked to see if the valve has reseated properly after
39
relieving. As previously mentioned, an improperly seated relief valve may
allow fugitive VOC emissions to occur. Rupture discs, which are commonly
.used in the SOCMI, also prevent fugitive VOC emissions. Some organic
chemical plants employ closed-loop sampling which may help to reduce fugi-
tive emissions.
The flaring of vapors vented from various vessels or equipment is
another technique which is used by some plants (particularly those produc-
ing toxic or hazardous chemicals) that will reduce fugitive emissions.
3.3.2 Existing Regulations
There are, presently, two types of regulations that impact fugitive
VOC emissions from organic chemical plants. The first type is to regulate
industrial operating practices on the basis of worker health and safety.
Because some aspects of these regulations deal with worker exposure to
process emissions, they may have some impact on fugitive VOC emissions.
The second type of regulations is regulations that were specifically de-
veloped to limit fugitive emissions.
3.3.2.1 Health and Safety Regulations. Several regulations have
been established under the direction of the Occupational Safety and Health
Administration and National Institute for Occupational Safety and Health
to: 1) limit the level of process emissions and 2) limit worker exposure
to process emissions. These regulations may result in a reduction in the
levels of fugitive VOC emissions.
In the vinyl chloride monomer and benzene industries, the safety
and health regulations are designed to limit the ambient VOC levels
to which workers may be exposed. Since these standards do not stipulate
how the allowable ambient levels should be achieved, workers can be
protected from high ambient VOC levels by: 1) a reduction in the fugitive
3-19
-------
VOC emissions or 2) the use of special equipment (such as personal
respirators) to isolate the worker from the emissions. This example
illustrates that the present health and safety regulations do not
mandate a reduction in fugitive VOC emissions, and any reduction in
fugitive emissions resulting from these regulations can be considered
to be "incidental". By contrast, fugitive emission regulations do
require the fugitive emissions to be reduced. »
3.3.2.2 Fugitive Emissions Regulations. Currently, there are no
federal fugitive emission regulations for the SOCMI. However, California
has established such regulations, and organic chemical plants in this
state must comply with the appropriate regulations.
California presently requires open-ended process lines to be capped-off
in order to minimize fugitive VOC emissions. This state also requires
relief valves to be vented to a flare system, monitored and maintained, or
a rupture disk to be used. In addition to these regulations, the South
Coast Air Quality Management District requires organic chemical plants
to vent fugitive emissions from compressor seals to a fired-heater or
flare system. The South Coast and Bay Area AQMD also require periodic
inspection of valves in the chemical and refining industries.
3-20
-------
3.4 REFERENCES
1. Erikson, D. G., and V. Kalcevic. Emissions Control Options for the
Synthetic Organic Chemicals Manufacturing Industry, Fugitive Emis-
sions Report, Draft Final. Hydroscience, Inc., 1979. p. II-2.
2. Ref. 1.
3. Ref. 1, p. II-3.
4. Ramsden, J. H. How to Choose and Install Mechanical Seals. Chem.
E., 85J22):97-102. 1978.
5. Ref. 1, p. II-3.
6. Ref. 4, p. 99.
7. Ref. 4, p. 100.
8. Ref. 4, p. 101.
9. Ref. 4, p. 99.
10. Perry, R. H., and C. H. Chilton. Chemical Engineers' Handbook, 5th
Ed. New York, McGraw-Hill Book Company, 1973. p. 6-8.
11. Ref. 10, p. 6-12.
12. Ref. 10, p. 6-13.
13. Ref. 10, p. 6-13.
14. Ref. 1, p. II-7.
15. Ref. 1 , p. II-8.
16. Lyons, J. L., and C. L. Ashland, Jr. Lyons' Encyclopedia of Valves,
New York, Van Nostrand Reinhold Co., 1975. 290 p.
17. Templeton, H. C. Valve Installation, Operation and Maintenance.
Chem. E., 78^(23)141-149, 1971.
18. Ref. 1, p. II-5.
19. Ref. 17, p. 147-148.
20. Ref. 17, p. 148.
21. Ref. 17, p. 148.
22. Ref. 17, p. 148.
3-21
-------
23. Pikulik, A. Manually Operated Valves. Chem. E., April 3, 1979.
p. 121.
24. Ref. 23, p. 121. -
25. Steigerwald, B. J. Emissions of Hydrocarbons to the Atmosphere from
Seals on Pumps and Compressors. Report No. 6, PB 216 582, Joint
District, Federal and State Project for the Evaluation of Refinery
Emissions. Air Pollution Control District, County of Los Angeles,
California. April 1958. 37 p.
26. Ref. 1, p. II-7.
27. Ref. 1, p. II-6.
28. Cooling Tower Fundamentals and Application Principles. Kansas City,
Missouri, The Marley Company, 1969. p. 4.
29. Ramsey, W. D., and G. C. Zoller. How the Design of Shafts, Seals
and Impellers Affects Agitator Performance. Chem. E., 83(18):
101-108. 1976.
30. Ref. 29, p. 105.
31. Ref. 29, p. 105.
32. Ref. 29, p. 105.
33. Ref. 29, p. 106.
34. Ref. 29, p. 106.
35. Ref. 29, p. 106.
36. McFarland, I. Preventing Flange Fires. Chem. E. Prog., 65(8):
59-61. 1969.
37. Wetherold, R. G., et al. Emission Factors and Frequency of Leak
Occurrence for Fittings in Refinery Process Units, interim report.
EPA Contract No. 68-02-2665. Austin, Texas, Radian Corporation,
February 1979. p. 22.
38. Radian Corporation. The Assessment of Environmental Emissions From
Oil Refining. Draft Report, Appendix B. EPA Contract No. 68-02-2147,
Exhibit B. Austin, Texas. August, 1979.
39. Letter with Attachments from J. M. Johnson, Exxon Company, U.S.A.,
to.Robert T. Walsh, U.S. EPA. July 28, 1977.
3-22
-------
4. EMISSION CONTROL TECHNIQUES
Sources of fugitive VOC emissions from SOCMI plants were identified
in Chapter 3 of this document. The potential emission control techniques
that can be applied to SOCMI fugitive emission sources are discussed in
this chapter. The applicability and estimated control effectiveness of
each technique are also presented. The quantitative control effective-
ness for many of the control techniques is not known. Qualitative
discussions of effectiveness and references to technology, transfer from
similar industries are presented wherever applicable.
4.1 LEAK DETECTION AND REPAIR METHODS
Leak detection and repair methods can be applied in order to reduce
fugitive emissions from any source. Leak detection methods are used to
identify equipment components that are emitting significant amounts of
VOC. Emissions from leaking sources may be reduced by three general
methods: repair, modification, or replacement of the source.
4.1.1 Leak Detection Methods
Leak detection methods include individual component surveys, area
(walk-through) surveys, and fixed point monitors. They are described in
this order since the first method is also included as part of the other
methods.
4.1.1.1 Individual Component Survey. Each fugitive emission source
(pump, valve, compressor, etc.) is checked for VOC leakage in an individ-
ual component survey. The source may be checked for leakage by visual,
audible, olfactory, soap bubble, or instrument techniques. Visual methods
are good for locating liquid leaks, especially pump seal failures.
Observation of a visible leak does not necessarily indicate VOC emissions,
since the leak may be composed of non-VOC compounds. High pressure leaks
4-1
-------
may be detected by hearing the escaping vapors, and leaks of odorous
materials may be detected by smelling the odor. Current industry
practices include leak detection by visual, audible, and olfactory
methods. However, in many instances, even very large VOC leaks are not
detected by these methods.
Spraying soap bubbles on equipment components is another individual
survey method. If the soap bubbles expand or are blown away, it is an
indication that something is leaking from the component. A disadvantage
of this method is that it does not distinguish leaks of non-VOC compounds
from VOC leaks. Consequently, air or steam leaks would produce the same
observed effect as VOC leaks. This method is only semtqiiantitattve since it
requires that the observer subjectively determine the rate of leakage
based on behavior of the soap bubbles. This method is limited to "cool"
sources, since temperatures above 100°C would cause the water in the soap
solution to boil away. This method is also not suited for moving shafts on
pumps or compressors, since the motion of the shaft may interfere with the
motion of the bubbles caused by a leak.
Portable hydrocarbon detection instruments are the best method for
identifying leaks of VOC from equipment components. The instrument is
used to sample and analyze the air in close proximity to the potential
leak surface by traversing the sampling probe tip over the entire area
where leaks may occur. This sampling traverse is called "monitoring" in
subsequent descriptions. The hydrocarbon concentration of the sampled air
is displayed on the instrument meter. The performance criteria for moni-
toring instruments and a description of instrument survey methods are
included in Appendix D. The hydrocarbon concentration observed during
monitoring of a component is proportional to the VOC emission rate from
the component. Data from petroleum refineries have been used to develop
relationships between monitoring concentration and mass emission rates.
The hydrocarbon concentration which defines a component needing mainte-
nance must be chosen. Components which have indicated concentrations
higher than this "action level" are marked for repair. Data from
petroleum refineries indicate that large variations in mass emission rate
4-2
-------
may occur over short time periods for an individual equipment component.
More frequent monitoring intervals tend to reduce the chance of missing
"large leaks" because of their variable leak rates.
4.1.1.2 Area Survey. An area survey (also known as a walk-through
survey) requires the use of a portable hydrocarbon detector and a strip
chart recorder. The procedure involves carrying the instrument within one
meter of the upwind and downwind sides of process equipment and associated
fugitive emission sources. An increase in observed concentration indi-
cates leaking fugitive emission sources. The instrument is then used for
an individual component survey in the suspected leak area. The efficiency
of this method for locating leaks is not well established. It has been
estimated that the walk-through survey combined with selected individual
surveys will detect about 50 percent of the number of leaks identified in a
complete individual survey. The time and labor requirements for the
walk-through are much lower. This method will not detect leaks from
sources such as elevated valves or relief valves. Leaks from adjacent
units and adverse meteorological conditions can also interfere with the
walk-through survey. Consequently, the walk-through survey is best for
locating only large leaks with a small resource expenditure.
4.1.1.3 Fixed Point Monitors. This method consists of placing
several automatic hydrocarbon sampling and analysis instruments at
various locations in the process unit. The instruments may sample the
ambient air intermittently or continuously. Elevated hydrocarbon concen-
trations indicate a leaking component. As in the walk-through method, an
individual component survey is required to identify the specific leaking
component in the area. For this method, the portable hydrocarbon detec-
tor is also required. Leaks from adjacent units and adverse meteorologi-
cal conditions may interfere with the method. The efficiency of this
method is not well established, but it has been estimated that 33 percent
of the number of leaks identified by a complete individual component
2
survey could be located by fixed-point monitors. Fixed-point monitors
are more expensive, multiple units may be required, and the portable
instrument is also required to locate the specific leaking component.
4-3
-------
Calibration and maintenance costs may be higher. Fixed-point monitors have
been used to detect emissions of hazardous or toxic substances (such as vinyl
chloride) as well as potentially explosive conditions. Fixed-point monitors
have an advantage for these cases, since a particular compound can be selected
as the sampling criterion.
4.1.2 Repair Methods
The following descriptions of repair methods include only those
features of each fugitive emission source (pump, valve, etc.) which need
to be considered in assessing the applicability and effectiveness of each
method. They are not intended to be complete repair procedures. The
effectiveness of repairs in reducing fugitive emissions has not been
well documented; however, data for valve repairs have been collected in
various petroleum refineries. In many cases, perfect repair will not be
achieved, but whenever repairs are performed, the portable hydrocarbon
detector should be used to identify the lowest achievable emission rate.
4.1.2.1 Pumps. Many pumps have spares which can be operated while
the leaking pump is being repaired. Leaks from packed seals may be reduced
by tightening the packing gland. At some point, the packing may deteriorate
to the point where further tightening would have no effect or possibly even
increase fugitive emissions from the seal. The packing can be replaced with
the pump out of service. When mechanical seals are utilized, the pump must
be dismantled so the leaking seal can be repaired or replaced. Dismantling
pumps will result in spillage of some process fluid and evaporate emissions of
VOC. These temporary emissions may be greater than the continued leak from the
seal, if the seal leak is small.
4.1.2.2 Compressors. Leaks from packed seals may be reduced by the
same repair procedure that was described for pumps. Other types of seals
require that the compressor be out of service for repair. Since most compressors
do not have spares, repair or replacement of the seal would require a shut-
down of the process. Temporary emissions resulting from a shutdown may be
greater than the emissions from the seal if it was allowed to leak until the
next scheduled shutdown.
4-4
-------
4.1.2.3 Relief Valves. In general, relief valves which leak must
be removed in order to repair the leak. In some cases of improper reseat-
ing, manual release of the valve may improve the seat seal. In order to
remove the relief valve without shutting down the process, a block valve
is required upstream of the relief valve. A spare relief valve should be
attached while the faulty valve is repaired and tested. After a relief
valve has been repaired and replaced, there is no guarantee that the
next over-pressure relief will not result in another leak.
4.1.2.4 Valves. Most valves have a packing gland which can be
tightened while in service. Although this procedure should decrease the
emissions from the valve, in some cases it may actually increase the
emission rate if the packing is old and brittle or has been overtightened.
Plug type valves can be lubricated with grease to reduce emissions around
the plug. Some types of valves have no means of in-service repair and
must be isolated from the process and removed for repair or replacement.
Other valves, such as control valves, may be excluded from in-service
repair by operating or safety procedures. In many cases, valves cannot
be isolated from the process for removal. Most control valves have a
manual bypass loop which allows them to be isolated and removed. Most
block valves cannot be isolated easily although temporary changes in
process operation may allow isolation in some cases. If a process unit
must be shut down in order to isolate a leaking valve, the emissions
resulting from the shutdown will probably be greater than the emissions
from the valve if allowed to leak until the next process change which
permits isolation for repair.
Depending on site specific factors, it may be possible to repair process
valves by injection of a sealing fluid into the source. This type of repair
may affect the operability of the valve such that replacement of the source
might be necessary within a short time after its repair, and the emissions
that could result due to the replacement of the source should be evaluated
when considering this type of repair. It should be noted that injection of
sealing fluid has been successfully used to repair leaks from valves in.
petroleum refineries in California.
4-5
-------
4.1.2.5 Flanges. In some cases, leaks from flanges can be reduced
by replacing the flange gaskets. Most flanges cannot be isolated to
permit replacement of the gasket. Data from petroleum refineries show
that flanges emit very small amounts of VOC.
4.1.3 Control Effectiveness of Leak Detection and Repair Methods
The instrument survey of individual components is the only type of leak
detection method for which control effectiveness has been quantified. Therefore,
the following estimations of control effectiveness do not pertain to the
soap bubble leak detection method, area surveys, or fixed-point monitoring
methods.
There are several factors which determine the control effectiveness of
individual component surveys; these include
Action level or leak definition,
Inspection interval or monitoring frequency,
Achievable emission reduction of maintenance, and
Interval between detection and repair of the leak.
Some of these factors can be estimated by using data collected from
*t
petroleum refineries.
4.1.3.1 Action Level. The action level is the minimum hydrocarbon
concentration observed during monitoring which defines a leaking component
which requires repair. The choice of the action level for defining a
leak is influenced by a number of important considerations. First, the
percent of total mass emissions which can potentially be controlled by
the monitoring and repair program can be affected by varying the leak
definition, or action level. Table 4-1 gives the percent of total mass
emissions affected by various action levels for a number of equipment
types. The data in this table, indicate that, in general, a low action
level results in larger potential emission reductions. However, the
choice of an appropriate leak definition is most importantly limited by
the ability to repair leaking components. Test data indicate that about
50 percent of valve leaks with initial screening values equal to or greater
than 10,000 ppmv can be successfully repaired. Similar data indicate
that attempted repair of valve leaks with initial screening values
4-6
-------
TABLE 4-1. FRACTION OF TOTAL MASS EMISSIONS FROM VARIOUS SOURCE TYPES THAT
WOULD BE.CONTROLLED BY DIFFERENT ACTION LEVELS
Action level9 (ppmv)
Fraction of mass emissions (as
100.000
50.000
10.000
1.000
Source type
Pump seals
Light liquid service
Heavy liquid service
56
0
68
0
87
21
97
66
In-line valves
Vapor service 85
Light liquid service 49
Heavy liquid service 0
92
62
0
98
84
0
99
96
23
Safety/relief valves
Compressor seals
Flanges
20
28
33
48
0
69
84
0
92
98
48
Level of emission at which repair of the source is required.
These data show the fraction of the total emissions from a given source
type that is attributable to sources with leaks above the various action
levels.
4-7
-------
of less than 10,000 ppmv can increase instead of decrease emissions
from these values. From these data it is concluded that repairing leaks
with screening values in the 1,000-10,000 ppmv range may not result in a
net reduction in mass emissions. The nature of repair techniques for
pipeline valves, for instance, are such that to repair leaks below
a certain level by tightening valve packing may actually result in an
increase in emissions. In practice, valve packing material becomes hard
and brittle after extended use. As the packing loses its resiliency, the
valve packing gland must be tightened to prevent loss of product to the
atmosphere. Excessive tightening, however, may cause cracks in the packing,
thus increasing the leak rate. Unbalanced tightening of the packing gland
may also cause the packing material to be positioned improperly in the valve
and allow leakage. Valves which are not often used can build up a "static"
seal of paint or hardened lubricant which could be broken by tightening
the packing gland. Therefore, it may be important not to cause small
leaks to become large leaks by requiring tightening of valves to meet a
very low leak repair action level.
4.1.3.2 Inspection Interval. A monitoring plan may include annual,
quarterly, monthly, or even weekly inspections. The length of time
between inspections should depend on the expected occurrence and
recurrence of leaks after a piece of equipment has been checked and/or
repaired. This interval can be related to the type of equipment and
service conditions, and different intervals can be specified for different
pieces of equipment after appropriate equipment histories have been
developed. In the refinery VOC leak Control Techniques Guideline (CTG)
document,7 the recommended monitoring intervals are: annualpump seals,
pipeline valves in liquid service, and process drains; quarterly--
compressor seals, pipeline valves in gas service, and pressure relief
valves in gas service; weeklyvisual inspection of pump seals; and no
individual monitoringpipeline flanges and other connections, and
pressure relief valves in liquid service. The choice of the interval
affects the emission reduction achievable since more frequent inspection
will result in leaking sources being found and fixed sooner. In order
to evaluate the effectiveness of different inspection intervals, it is
4-8
-------
necessary to estimate the rate at which new leaks will occur and repaired
leaks will recur. The estimates which have been used to evaluate yearly,
quarterly, and monthly inspections are shown in Table 4-2.
4.1.3.3 Allowable Interval Before Repair. If a leak is detected,
the equipment should be repaired within a certain time period. The
allowable repair time should reflect an interest in eliminating a source
of VOC emissions but should also allow the plant operator sufficient time
to obtain necessary repair parts and maintain some degree of flexibility
in overall plant maintenance scheduling. The determination of this
allowable repair time will affect emission reductions by influencing the
length of time that leaking sources are allowed to continue to emit
pollutants. Some of the components with concentrations in excess of the
leak definition action level may not be able to be repaired until the
next scheduled unit shutdown, e.g., a unit turnaround.
The effects of different allowable repair intervals are shown in
Table 4-3. The percentages shown in the table are the percent of emis-
sions from the component which will be affected by the repair. The
emissions which occur between the time the leak is detected and repair is
attempted are increased with increasing allowable repair intervals.
4.1.3.4 Achievable Emission Reduction. Repair of leaking components
will not always result in complete emission reduction. The repair of
components which have initial monitoring levels below 1,000 ppm has not
been adequately demonstrated. Repair of those components with low initial
leak rates, mav actually result in an emission rate increase. In order to
estimate repair effectiveness, it was assumed that emission would be
reduced to a level equal to components with screening values of 1,000
ppm. The average emission rates of components above 10,000 ppm and at
1,000 ppm are shown in Table 4-4.
4.1.3.5 Development of Controlled Emission Factors. The uncon-
trolled emission levels for the emission sources that are typically found
in the model plants were previously presented in Chapter 3 (Table 3-1).
Controlled VOC emission levels can be calculated by a "controlled emission"
factor. This factor can be developed for each type of emission source by
using the general expression:
4-9
-------
TABLE 4-2. ESTIMATED OCCURRENCE AND RECURRENCE RATE FOR VARIOUS MONITORING INTERVALS
I
o
Estimated perci
of sources leal
at above 10,000
Source type initially3
Pump seals
Light liquid service 23
Heavy liquid service 2
In-line valves
Vapor service 10
... Light liquid service 12
Heavy liquid service 0
Estimated percent of
*nt initial leaks which
^in9 are found leaking at
PP subsequent inspections'3
Annual
20
20
20
-.20
20
Quarterly
10
10
10
10
10
Monthly
5
5
5
.5 .
5
Estimated. percent of
sources which are
found leaking at
subsequent inspections c
Annual
4.6
0.4
2.0
2.4
0.0
Quarterly
2.3
0.2
1.0
' 1.2
0.0
Monthly
1.2
0.1
0.5
. 0.6
0.0
Safety/relief valves
8
20
10
1.6
0.8
0.4
Compressor seals
33
20
10
6.3
3.3
1.7
Flanges
20
10
0.0
0.0
0.0
^Approximate fraction of sources having leaks equal to or greater than 10,000 ppm prior to repair.
Approximate fraction of leaking sources that were repaired but found to leak during subsequent
inspections. These approximations are based on engineering judgment.
Approximate fraction of sources that were repaired but found to leak during a subsequent inspection.
These approximations are the product of the information presented in footnotes a and b.
-------
TABLE 4-3. PERCENT OF MASS EMISSIONS AFFECTED BY VARIOUS REPAIR INTERVALS
Allowable repair interval (days)
30
15
Percent of emissions affected
95.9 97.9 99.3 99.9
TABLE 4-4. AVERAGE EMISSION RATES FROM SOURCES
ABOVE 10,000 PPMV AND AT 1000 PPMV9
Source type
Pump seals
Light liquid service
Heavy liquid service
In-line valves
Vapor service
Light liquid service
Heavy liquid service
Safety/relief valves
Compressor seals
Flanges
(Y)
Emission rate
from sources above
10,000 ppmva
(kg/hr)
0.45
0.21
0.21
0.07
0.005
1.4
l.'l
0.003
Emission rate
from sources .at
1000 ppmvD
(kg/hr)
0.035
0.035
0.001
0.004
0.004
0.035
0.035
0.002
/Xx
Percentage
reduction
92.0
83.0
99.5
94.0
20.0
97.5
97.0
33.0
a. . . ,, .... , , .
screening values above 10,000 ppmv.
'Emission rate of all sources, within a source type,.having screening
values of 1000 ppmv.
4-11
-------
Controlled emission factor = Uncontrolled factor - uncontrolled
factor x emission reduction efficiency
The reduction efficiency can be developed by the following expression and
correction factors:
Reduction efficiency =AxBxCxD
Where:
A = Theoretical Maximum Control Efficiency = fraction of total mass
emissions for each source type with VOC concentrations greater
than the action level (Table 4-1, Figure 4-1).
B = Leak Occurrence and Recurrence Correction Factor = correction
factor to account for sources which start to leak between
inspections (occurrence) and for sources which are found to
be leaking, are repaired and start to leak again before the
next inspection (recurrence) (Table 4-2, 4-6).
C = Non-Instantaneous Repair Correction Factor = correction factor
to account for emissions which occur between detection of a leak
and subsequent repair; that is, repair is not instantaneous
(Table 4-3).
D = Imperfect Repair Correction Factor = correction factor to
account for the fact that some sources which are repaired are
not reduced to zero emission levels. For computational pur-
poses, all sources which are repaired are assumed to be reduced
to a 1000 ppm emission level (Table 4-4).
These correction factors can, in turn, be determined from the following
expressions:
rf
0) B = 1 - /
ty\ r - 365 - t
u; L " 365
(3) D = 1 - I
4-12
-------
Where:
n = Average number of leaks occurring and recurring over the
m .
monitoring interval.
N = Total number of sources at or above the action level (Figure
4-2).
t = Average time before repairs are made (with a 15-day repair limit,
7.5 is the average used).
f = Average emission factor for sources at the average screening
value achieved by repair.
F = Average emission factor for all sources at or above the action
level.
An example of a control effectiveness calculation is presentee! in Table 4-5.
Support data for this calculation are presented in Tables 4-1, 4-2, 4-3,
4-4, and 4-6, as well as in Figures 4-1 and 4-2.
4.2 EQUIPMENT SPECIFICATIONS
Equipment specifications for each emission source are described
below. Some of the specifications may be applicable to more than one
type of source. In these cases, references are made to the preceding
description with any differences in applicability or effectiveness notes.
4.2.1 Pumps
Fugitive emissions from pumps occur at the junction of a moving shaft
and a stationary casing. Equipment specifications that may be implemented
for pumps include elimination of this junction, improvement of the seal at
the junction, or collection and control of the emissions from the junction.
4.2.1.1 Seal!ess Pumps. Pumps such as diaphragm type pumps or
"canned" pumps do not have a shaft/casing junction and therefore do not
leak the pumped fluid as a normal course of operation. Failure of the
diaphragm may result in temporary emissions of VOC. Sealless pumps are
used primarily in SOCMI processes where the pumped fluid is hazardous or
toxic, and every effort must be made to prevent leaks of the fluid.
4-13
-------
TABLE 4-5. EXAMPLE OF CONTROL EFFICIENCY CALCULATION
Assume:
1) A leak detection and repair program to reduce emissions from
valves in gas/vapor source.
2) Action level = 10,000 ppm.
3) Average screening value after directed repair = 1,000 ppm.
4) Leak detection monitoring interval = 3 months.
5) Allowable repair interval =15 days.
6) Number of valves having new or recurring leaks between repair
intervals, n = 0.2N (see Table 4-6).
m
Calculations:
A = 0.98 (from Figure 4-1 for a screening value of 10,000 ppmv)
B = 0.9 (from Table 4-6)
C = 0.979 (from Table 4-3 for 15-day interval)
where:
c - A(Avg. uncontrolled emission factor)9
Fraction of sources screening > 10,000 ppmD
= (0.98)(0.021:kg/hr)/0.10 = 0.206 kg/hr
f = Emission factor at 1000 ppme
= 0.001 kg/hr
and D = (1 - jj^-) = 0.995
Overall percentage reduction =AxBxCxD
= (0.98) x (0.9) x (0.979) x (0.995)
= 86 Percent
Therefore:
Control effectiveness factor = 0.021 kg/hr - (0.86)(0.021 kg/hr)
= 0.003 kg/hr
? Reference 10. '
From Figure 4-2.
c Reference 11.
4-14
-------
TABLE 4-6. IMPACT OF MONITORING INTERVAL ON CORRECTION FACTOR ACCOUNTING
FOR LEAK OCCURRENCE/RECURRENCE (FOR EXAMPLE CALCULATION)
Monitoring -
interval m m
1 month 0.1N 0.05N 0.95
x
3 months 0.2N 0.1N 0.90
1 year 0.4N 0.2N 0.80
a n = Total number of leaks which occur, recur, and remain between
m monitoring intervals.
n" = Average number of leaks over the monitoring interval.
m
c B = Correction factor accounting for leak occurrence/recurrence.
4-15
-------
O
C/J
CO
O
EC
100
90
80
70
60
50
40
30
20
10
0
11
i 111 i i ni i i_i 11 jrrn | rm r
'""" W-^.
-i_>iU<
VS
I
I I I 1 I I J II I 1 t 11 I 111
UPPER LIMIT OF 90%
CONFIDENCE INTERVAL
ESTIMATED PERCENT OF
TOTAL MASS EMISSIONS
LOWER LIMIT OF 90%
CONFIDENCE INTERVAL
PERCENT OF TOTAL MASS
EMISSIONS -
INDICATES THE PERCENT OF
TOTAL EMISSIONS ATTRIBU-
TABLE TO SOURCES WITH
SCREENING VALUES GREATER
THAN THE SELECTED VALUE
10 100 1000 10,000 100.000 1.000.000
SCREENING VALUE (ppmv)
-------
4.2.1.2 Double Mechanical Seals. Double mechanical seals consist
of two mechanical sealing elements with a barrier fluid in the chamber
between the seals. This chamber is either flushed with circulating
barrier fluid or is flooded with static barrier fluid. The pressure of
the static barrier fluid can be monitored to de.tect failure of the inner
seal.14 Any leaks through the inner seal may be dissolved or suspended in
the barrier fluid, and subsequent degassing of the sealing fluid may
result in emission of VOC. Therefore, barrier fluid degassing vents must
be controlled in order to provide maximum control effectiveness of double
mechanical seals. After extended periods of use, double seals may also
develop leaks at the outer seal/shaft junction.
Emissions of VOC from degassing vents can be controlled by a closed
vent system which consists of piping and, if necessary, flow inducing
devices to transport the degassing emissions to a control device such as
a process heater, or vapor recovery system. Control effectiveness of a
double mechanical seal and closed vent system is dependent on the effective-
ness of the heater, or vapor recovery system, and the frequency of seal
failure. Failure of both the inner and outer seals can result in relatively
large VOC emissions at the seal area of the pump. As noted, the pressure
monitoring of the static barrier fluid may be used in order to detect failure
of the seals. In addition, visual inspection of the seal area also can be
effective.for detecting failure of the outer seals. Upon seal failure, the
leaking pump would have to be shut down for repair.
Double mechanical seals are used in many SOCMI process applications;
however, there are some conditions that preclude use of double mechanical
seals. Their maximum service temperature is usually limited to less than
260°C, and mechanical seals cannot be used on pumps with reciprocating
shaft motion. Process fluids containing catalyst fines or other abrasive
materials may not be suitable for use with mechanical seals.
4.2.1.3 Closed Vent Systems. The system described above for con-
trolling degassing vent emissions could also be applied to control
emissions from the seal area of pumps. This application would require
the use of some type of flow inducing device to transport the emissions
from the seal area to the control device. The seal area would be enclosed
4-17
-------
in order to collect the emissions and a vacuum eductor or a compressor
could be used to remove vapors from the seal area. However, normal pump
operating practices may require frequent visual inspection or mechanical
adjustments in the seal area. This would not be possible with a closed
vent system at the seal area. A potential problem with this approach is
that explosive mixtures may be created by enclosing the pump seal area,
and therefore safety and operating practices may limit the use of closed
vent systems for pump seal areas.
4.2.2 Compressors
Fugitive emissions from compressors occur at the junction of a moving
shaft and a stationary casing. Emission reductions from this source type
may be achieved by improving the seal at the junction, or collecting and
controlling the emissions from the junction.
4.2.2.1 Double Mechanical Seals. Double mechanical seals for
compressors are similar to those described above for pump applications,
and reciprocating shafts cannot be fitted with mechanical seals.
Labyrinth type seals may also have barrier fluid systems. Existing
compressors may have double mechanical seals and seal oil flush systems,
but seal oil reservoir degassing vents must be controlled with closed
vent systems as described above. Control efficiency is dependent on the
control device efficiency and the frequency of seal failures.
4.2.2.2 Closed Vent Systems. The seal area of a compressor may
be enclosed, and the VOC emissions routed to a control device through
a closed vent system. However, flow inducing*, devices may be required to
transport vapors to the control device. Although the formation of
explosive mixtures in the enclosed seal area may prohibit application
of this equipment modification, closed vent systems have been applied to
compressor seal areas in petroleum refineries.
4.2.3 Pressure Relief Devices
Pressure relief devices include rupture disks and safety/relief
valves. Fugitive emissions from these devices occur because of improper
seating or partial failure of the device. These fugitive emissions do
not include emissions which result from normal operation of the devices
4-18
-------
caused by overpressure of the process or vessel which the device protects.
Fugitive emissions from rupture disks may be caused by pinhole leaks in
the disk itself caused by corrosion or fatigue. Fugitive emission from
relief valves may be caused by failure of the valve seating surfaces,
improper reseating after overpressure relieving, or process operation
near the relief valve set pressure which may cause "simmering".
4.2.3.1 Rupture Disks. Although they are also pressure relief devices,
rupture disks can be installed upstream of a safety/relief valve in order
to prevent fugitive emissions through the relief valve seat. This procedure
may require use of a larger size relief valve because of operating codes. The
disk/valve combination may also require appropriate piping changes to prevent
disk fragments from lodging in and damaging the relief valve when relieving
overpressure. A block valve upstream of the rupture disk is also required
in order to permit in-service replacement of the disk after overpressuring.
If the disk could not be replaced, the first overpressure would result in
the relief valve being the same as an uncontrolled relief valve, and it may
actually be worse since disk fragments may prevent proper reseating of the
relief valve. In some chemical plants, installation of a block valve up-
stream of a pressure relief device may be a common practice. In others, it
may be forbidden by operating or safety procedures. Tandem pressure
relief devices with a three-way valve can be used to avoid operation without
overpressure protection. Rupture disk/relief valve combinations must have
some provision for testing the integrity of the disk. The area between the
rupture disk and relief valve must be connected to a pressure indicator,
recorder, or alarm. If the process fluid is not hazardous or toxic, a
simple bubbler apparatus could be used to test disk integrity by connecting
the bubbler to the disk/valve area. The control efficiency of the disk
valve combination is assumed to be 100 percent for fugitive emissions. If
the disk integrity is not maintained or if the disk is not replaced after
overpressure relief, the control efficiency would be lowered. The disk/valve
combination has no effect on emissions which result from overpressure
relieving.
4-19
-------
4.2.3.2 Resilient Seat Relief Valves. Manufacturers of relief
valves state that resilient seat or "0-ring" relief valves provide better
reseat qualities compared to standard relief valves. No test data are
available to verify these statements. These improvements would have no
effect on overpressure emissions or fugitive emissions due to seal
failure or "simmering'.'..
4.2.3.3 Closed Vent Systems. A closed vent system can be used to
transport the discharge or leakage of pressure relief devices to a
control device such as a flare. Since overpressure discharges as well as
fugitive emissions are routed to the control device, it must be sized
appropriately. A larger pressure relief device may be required for use
with a closed vent system. The control efficiency of a closed vent
system is dependent on the effectiveness of the control device. Typical
flare systems may be only 60 percent effective for fugitive emission
15
destruction. This efficiency reflects the fact that many flare systems
are not of optimum design. As a result flares that are designed to handle
large volumes of vapors associated with overpressure releases are used to
handle low volumes of fugitive emissions. With such designs, optimum mixing
is not achieved because the vent gas exit velocity is low and large
16
flares generally cannot properly inject steam into low volume streams. A
properly designed flare system typically exhibits a 99 percent hydrocarbon
17
destruction efficiency. Closed vent systems for pressure relief devices
are used in existing SOCMI processes especially where the emissions may be
hazardous or toxic.
4.2.4 Open-Ended Valves
Fugitive emissions from open-ended valves are caused by leakage through
the seat of the valve. Emissions may also occur through the stem and gland
of the valve, and these emissions may be controlled by methods described
for valves in Section 4.1.2. Approximately 28 percent of SOCMI valves
l!8
(excluding safety/relief and check valves) in VOC service are open-ended.
They include drain, purge, sample, and vent valves. Fugitive emissions from
open-ended valves can be controlled by installing a cap, plug, flange, or
second valve to the open end of the valve. In the case of a second valve, the
4-20
-------
upstream valve should always be closed first after use of the valves. Each
time the cap, plug, flange, or second valve is opened, any VOC which has
leaked through the first valve seat will be released. These emissions have not
been quantified. The control efficiency of these devices is assumed to be
100 percent. The actual efficiency will be dependent on the frequency of
removal of the cap or plug. Caps, plugs, etc. for open-ended valves do not
affect emissions which may occur during use of the valve. These emissions may
be caused by line purging for sampling, draining, or venting through the
open-ended valve. Caps, plugs, flanges, or second valves for open-ended
19
valves are required by California regulations..
4.2.5 Sampling Connections
Fugitive emissions from sampling connections occur as a result of
purging the sampling line in order to obtain a representative sample of the
process fluid. Approximately 25 percent of open-ended valves are used for
20
sampling connections. Fugitive emissions from sampling connections can be
reduced by using a closed loop sampling system. The closed loop system is
designed so that the purged fluid is returned to the process at a point of
lower pressure. A throttle valve or other device is required to induce the
pressure drop across the sample loop. The purged fluid could also be directed
to a control device such as a flare. In this case the control efficiency would
be dependent on the flare efficiency for hydrocarbon destruction. Closed
loop sampling is assumed to be 100 percent effective for controlling fugitive
emissions. Since some pressure drop is required to purge sample through the
loop, low pressure processes or tankage may not be amenable to closed loop
sampling. Safety requirements may prohibit closed loop sampling in some
instances.
4.2.6 In-Line Valves
Fugitive emissions from valves occur at the stem or gland area of the
valve body. Diaphragm and bellows seal valves do not have a stem or gland
and therefore are not prone to fugitive emissions. Diaphragm valves are
generally used where hazardous or toxic process fluids are present and
fugitive emissions must be eliminated. Their control effectiveness is
approximately 100 percent, although failure of the diaphragm may cause
4-21
-------
large temporary emissions. The applicability of diaphragm valves is
limited by the strength of the diaphragm. Diaphragm valves may not be
suitable for many applications because of process conditions or cost
considerations.
4.2.7 Effectiveness of Equipment Specifications
In order to quantify the environmental and economic impacts of apply-
ing controls, the control efficiency must be determined. In some cases,
there are many complicating factors which make it difficult to accurately
estimate control efficiency. For example, the efficiency of caps or plugs
for open-ended valves is dependent on 1) the frequency of removal of the
cap or plug, since this removal will result in emission of fluids trapped
by the cap or plug, and 2) the emission rate through the valve seat. The
estimated control efficiencies for various equipment modifications are
shown in Table 4-7. These estimates represent the maximum emission re-
duction possible for the equipment modifications. In some instances, the
actual emission reduction will depend on other factors such as the effi-
ciency of control devices attached to closed vent systems. Carbon absorp-
tion or vapor recovery systems would approach the 100 percent efficiency,
but flares may be only 60 percent effective for hydrocarbon destruction.
These estimates of effectiveness are used to calculate environmental and
economic impacts of regulatory alternatives in Chapters 7 and 8 of this
document.
4-22
-------
TABLE 4-7. EFFECTIVENESS OF EQUIPMENT MODIFICATIONS
Source type/ Control efficiency
equipment modification (%)
Pumps
Sealless pumps 100
Double mechanical seals/closed vent system ^100
Closed vent system on seal area ^100
Double mechanical seals/closed vent system %100a
Compressors
Double me
Closed vent system on seal area
Safety/relief valves
Closed vent system
Rupture disks 100
Closed vent system 60
Open-ended lines
Caps, plugs, blinds, second valves 100
Sampling connections
Closed loop sampling 100
In-line valves
Diaphragm valves 100
aAlthough a control efficiency is not attained in all cases, it is
achievable in some cases.
This control effectiveness reflects the fact that a closed vent system is
normally sized for emergency relief.21
This is the control efficiency reflects the.use of these devices downstream
of an irntial valve with VOC on one side and.atmosphere on the other
4-23
-------
4.3 REFERENCES
1. Hustvedt, K. C., and R. C. Weber. Detection of Volatile Organic
Compound Emissions from Equipment Leaks. Presented at 71st Annual
Air Pollution Control Association Meeting, Houston, Texas, June 25-30,
1978.
2. Ref. 1.
3. Wetherold, R. G., and L. P. Provost. Emission Factors and Frequency
of Leak Occurrence for Fittings in Refinery Process Units. Interim
Report. EPA/600/2-79-044. Radian Corporation. February 1979. p. 2.
4. Ref. 3.
5. Ref. 3.
6. Valve Repair Summary and Memo from F. R. Bottomley, Union Oil Company.
Rodeo, California. To Milton Feldstein, Bay Area Quality Management
District. April 10, 1979.
7. Environmental Protection Agency. Control of Volatile Organic Com-
pound Leaks from Petroleum Refinery Equipment. EPA-450/2-78-036.
OAQPS No. 1.2-111. Research Triangle Park, North Carolina. June 1978.
8. Ref. 3.
9. Ref. 3.
10. Ref. 3.
11. Ref. 3.
12. Ref. 3. '
13. Ref. 3.
14. Erikson, D. G., and V.: Kalcevic. Emission Control Options for the
Synthetic Organic Chemicals Manufacturing Industry, Fugitive Emissions
Report, Draft Final. Hydroscience, Inc. 1979. p. III-l.
15. Draft-of Background Information Document (BID), Chapter 4, for
a National Emission Standard for Hazardous Air Pollutants from the
Ethylbenzene/Styrene Industry. U.S. Environmental Protection
Agency, RTP, North Carolina, October 1979.
4-24
-------
16. Ref. 15.
17. Environmental Protection Agency. Control of Volatile Organic
Emissions from Existing Stationary Sources. Volume 1: Control
Methods for Surface Coating Operations. EPA-450/2-76-028.
Research Triangle Park, North Carolina. November 1976, p. 42.
18. Ref. 14, p. III-5.
19. Ref. 14, p. III-5.
20. Ref. 14.
21. Ref. 15.
4-25
-------
5. MODIFICATION AND RECONSTRUCTION
In accordance with the provisions of 40 Code of Federal Regulation (CFR),
Sections 60.14 and 60.15, an "existing facility" can become an affected
facility and, subsequently, subject to the standards of performance if it
is modified or reconstructed. An existing facility, as defined in 40 CFR
60.2 (aa), is a facility of the type for which standards of performance have
been promulgated and the construction or modification of which was begun
prior to the proposal date of the applicable standards.
The applicability of provisions 40 CFR 60.14 and 60.15 to the SOCMI,
and the conditions, as outlined in these provisions, under which existing
facilities could become subject to standards of performance are discussed
below.
5.1 GENERAL DISCUSSION OF MODIFICATION AND RECONSTRUCTION PROVISIONS
5.1.1 Modification
"Modification" is defined in 40 CFR 60.14.(a) as any physical or
operational change of an existing facility which increases the emission rate
of any pollutant to which a standard applies. Exceptions to this definition
are presented in paragraphs (d), (e), and (f) of Section 60.14. These
exceptions are as follows:
Paragraph (d) - In accordance with the paragraph, an
existing facility may undergo a physical or operational
change, which increases the emission rate of any pollurant
to which standards of performance apply, but not judged to
be a modification, if the owner or operator can demonstrate
to the Administrator's satisfaction (by any of the pro-
cedures prescribed in paragraph (b) of this section) that
5-1
-------
the total emission rate of that pollutant has not increased
from the facility.
Paragraph (e) - Physical or operational changes to an
existing facility which will not be considered modifica-
tions are specified in this portion of Section 60.14.
These changes include:
a. Routine maintenance, repair, and replacement.
b. An increase in the production rate not requiring
a capital expenditure as defined in Section
60.2(bb).
c. An increase in the hours of operation.
d. Use of an alternative fuel or raw material if
prior to the standard, the existing facility
was designed to accommodate that alternate fuel
or raw material.
e. The addition or use of any system or device
whose primary function is the reduction of
air pollutants, except when an emission control
system is removed or replaced by a system con-
sidered to be less efficient.
Paragraph (f) - This paragraph provides for superceding
any conflicting provisions of this section.
Upon modification, an existing facility becomes an affected facility
for each pollutant to which a standard applies and for which there is an
increase in the emission rate to the atmosphere. Paragraph (c) also declares
that the addition of an affected facility to a stantionary source through
any mechanismnew construction, modification, or reconstructiondoes not
make any other facility within the stationary source subject to the applicable
standards.
5.1.2 Reconstruction
Under the provisions of Section 60.15, an existing facility becomes
an affected facility upon reconstruction, irrespective of any change in
emission rate. Generally, reconstruction is considered to occur upon the
replacement of components if the fixed capital cost of the new components
exceeds 50 percent of the fixed capital cost that would be required to
5-2
-------
construct a comparable entirely new facility, and it is economically and
technically feasible for the facility to comply with the applicable
standards of performance. The final judgments on what replacement con-
stitutes reconstruction and when it is technologically and economically
feasible to comply with the applicable standards of performance is made
by the Administrator. The Administrator's final determinations are made
on the following bases:
(1) comparison of the fixed capital costs of the replacement
components and a newly constructed comparable facility,
(2) the estimated life of the facility after the replacements
compared to the life of a comparable entirely new facility,
(3) the extent to which the components being replaced cause
or contribute to the emissions from the facility, and
(4) any economic or technical limitations on compliance with
applicable standards of performance which are inherent in
the proposed replacements.
The purpose of this provision is to ensure that an owner or operator
does not perpetuate an existing facility by replacing all but vestigial
components, support structures, frames, housing, etc., rather than
totally replacing it in order to avoid subjugation to applicable standards
of performance. In accordance with Section 60.5, EPA will, upon request,
determine if the action taken constitutes construction (including reconstruction)
5.2 APPLICABILITY OF MODIFICATION AND RECONSTRUCTION PROVISIONS TO THE
SOCMI
5.2.1 Modification
Several operating conditions that could be encountered in an organic
chemical plant are presented below. These conditions may or may not result
in an increase in emissions.
The replacement of a potential fugitive emission source such as a pump
or valve commonly occurs in an organic chemical plant. If such a source
is replaced with an equivalent source (such as is done during routine
repair and replacement), the fugitive emissions from the facility should not
increase because the number of potential sources in the same vapor pressure
service (handling the same organic chemical) remains unchanged.
5-3
-------
Process equipment pieces such as heat exchangers, reactors, distilla-
tion columns, reboilers, filters and separators, or new control loops are
commonly added to existing facilities in the organic chemical industry to
increase the capacity of or to optimize a process. The addition of this
equipment would normally increase fugitive emissions from a facility due to
the increased number of potential emission sources (pumps, valves, sampling
connections, etc.) that are associated with the process equipment. However,
in those cases where some sources are physically removed from service, the
addition of new fugitive emission sources would not necessarily increase the
level of fugitive emissions from the stationary source.
In some cases a facility in the organic chemical industry can be
converted from the production of one chemical to the production of a second
chemical. This normally occurs when production of the second chemical results
in greater profits. In such a case, whenever either the number of fugitive
emission sources or the vapor pressure of the second chemical increases during
this conversion, the level of VOC emissions from the facility could be
expected to increase.
Changes may be made to a process, although the chemical being produced
remains the same. One such case would be a change in catalyst for producing
a given chemical. In such a case the level of fugitive emissions would not
be expected to change because the number of sources nor the vapor pressure of
the chemical would change.
In many cases, there may be a desire to increase the capacity of an
existing facility. This may be achieved by replacing certain process equip-
ment (pumps, heat exchangers, reactors, etc.) with similar equipment but of
larger capacity or addition of process equipment. If this replacement or
addition does not increase the number of fugitive emission sources handling
the given organic chemical, the level of fugitive emissions would not be
expected to increase. However, if the number of sources were to increase due
to this replacement or addition, then VOC emissions could be expected to
increase.
5-4
-------
5.2.2 Reconstruction
When an owner or operator replaces several components of an existing
facility, that facility may or may not become subject to applicable
standards of performance under the provisions of Section 60.15. For
example, if an owner or operator replaces several reactors in an existing
facility, reconstruction is considered to have occurred if the fixed
capital costs for these reactors exceeds 50 percent of the costs that
would be required to construct an entirely new facility. Replacement
of other major equipment components such as heat exchangers, and distillation
columns may also be considered as reconstruction if the fixed capital
costs for the replaced equipment exceeds 50 percent of the costs of construct-
ing an entirely new facility.
5-5
-------
6. MODEL PROCESS UNITS AND REGULATORY ALTERNATIVES
This chapter presents model process unit parameters and alternative
emission controls considered for reduction of fugitive emissions from SOCMI
sources. The model units were selected to represent the range of processing
complexity in the industry to provide a basis for comparing environmental
and economic impacts of the regulatory alternatives. The selected regulatory
alternatives provide varying levels of emission control.
6.1 MODEL UNITS
Available data show that fugitive emissions are proportional to the
number of potential sources, but are not related to capacity, throughput,
age, temperature, or pressure.1 SOCMI model units therefore represent
different levels of process complexity (number of sources) rather than
unit size.
6.1.1 Sources of Fugitive Emissions
The various potential fugitive emission sources in a SOCMI process
unit were described in Chapter 3. Data from petroleum refineries indicate
that cooling towers are very small sources of VOC emissions.2 Differences
in SOCMI operating procedures, such as recirculation of process water, might
result in cooling tower VOC emissions, but no data are available to verify
this. The number of agitator seals in SOCMI is not known. Furthermore, the
emission rate from SOCMI agitator seals has not been measured. Since there
are no data from similar sources in other industries, no estimates of emission
rate can be made. Because of these uncertainties, cooling towers and agitator
seals are not included in the Model Units.
6-1
-------
6.1.2 Model Unit Parameters
SOCMI process units vary considerably in size, complexity, age, and
types of products manufactured. In order to estimate emissions, control
costs, and environmental impacts on a unit specific basis, three model
units were developed. The technical parameters for the model units are
shown in Table 6-1. These three model units represent the range of
emission source populations that may exist in .SOCMI process units. The
technical parameters were developed from a data base compiled by Hydro-
3
science, Inc. The data base included equipment source counts from 62 SOCMI
plants which produce 35 different chemicals. These plant sites represent
approximately 5 percent of the total existing SOCMI plants and include large
and small capacities, batch and continuous production methods, and varying
levels of process complexity. Hydroscience estimates that 52 percent of
existing SOCMI plants are similar to Model Unit A, 33 percent are similar
to B, and 15 percent are similar to C. The source counts for the 35 chemicals
include pumps, valves, and compressors. These counts were used in com-
bination with the number of sites which produce each chemical in order to
4
determine the average number of sources per site.
Data from petroleum refineries indicate that emission rates of sources
decrease as the vapor pressure (volatility) of the process fluid decreases.
Three classes of volatility have been established based on the petroleum
refinery data. These include gas/vapor service, light liquid service, and
heavy liquid service.5 The split between light and heavy liquids for the
refinery data is between streams called naphtha and kerosene. Since simi-
lar stream names may have different vapor pressures, depending on site
specific factors, it is difficult to quantify the light-heavy split. The
break point is approximately at a vapor pressure of 0.3 kPa at 20°C.
The data collected by Hydroscience were used to estimate the split between
6
gas/vap.or and liquid service for each source type. In order to apply
emission factors for light and heavy liquid service, it is assumed that
one half of SOCMI liquid service sources are in light liquid service. There
are no data available on the actual distribution of sources in volatility
ranges. It is assumed that all SOCMI packed seal pumps are in heavy liquid
service. This assumption is reasonable, since more volatile liquids are
6-2
-------
TABLE 6-1. FUGITIVE EMISSION SOURCES FOR THREE MODEL UNITS
Number of components in model unitc
Equipment component3
Pump seals
Light liquid service
Single mechanical
Double mechanical
Seal less
Heavy liquid service
Single mechanical
Packed
In-line valves
Vapor service
Light liquid service
Heavy liquid service
Safety/ relief valves
Vapor service
Light liquid service
Heavy liquid service
Open-ended valves and lines
Vapor service
Light liquid service
Heavy liquid service
Compressor seals
Sampling connections0
Flanges
Cooling towers
Agitator seals
^Equipment components in VOC
Sample, drain, purge valves
^Based on 25% of open-ended
^-* C O "/ A f j-i**^*»+-T»-*j-t i i n ? 4- « rt v* rt
Model unit
A
5
3
0
5
2
90
84
84
11
1
1
9
47
48
1
26
600
c
600e
service only.
valves. From Ref.
*» T m 4 1 a w» -t- f\ Mrtrltt 1 1!
Model unit
B
19
10
1
24
6
365
335
335
42
4
4
37
189
189
2
1.04
2400
c
2400e
3, pg. IV-3.
U44. A
Model unit
C
60
31
1
73
20
1117
1037
1037 ;
;
130 1
13
14
115
581
581
8
320 ;
7400 <
e :'
7400e
3,3% of existing units
15% of existing units
Ref. 3, pg. IV-1.
Data not available.
are similar to Model Unit B.
are similar to Model Unit C.
6-3
-------
more suitable for mechanical seal applications, and newer process units tend
to use fewer packed seals. Sampling connections are a subset of the open-
ended valve category. Approximately 25 percent of open-ended valves are used
for sampling connections.7 Emissions which occur through the valve stem,
gland, and open-end are included in the open-ended valve category. The
emission factor for sampling connections applies only to emissions which
result from sample purging.
6.2 REGULATORY ALTERNATIVES
Regulatory alternatives represent comprehensive programs for reduction
of emissions by combining the individual control techniques described in
Chapter 4. The regulatory alternatives described in this section contain
feasible control techniques for reducing fugitive emissions of VOC from
SOCMI sources.
The purpose of developing different regulatory alternatives is to
provide a basis, along with model unit parameters, for determing the air-
quality and non air-quality environmental impacts, energy requirements, and
the costs associated with varying degrees of VOC fugitive emissions reduction.
The regulatory alternatives selected for analysis include a "status quo of
fugitive emission control" case and three increasingly restrictive levels of
emission control requirements. The "status quo" case allows for the analysis
of not implementing standards of performance. The three increasingly restric-
tive control requirements allow for analysis of the impacts of different
systems with varying degrees of emission reduction. The requirements for
each of these regulatory alternatives are summarized in Table 6-2 and are
described below.
6.2.1 Regulatory Alternative I
Alternative I represents the general level of control that would exist
in the absence of establishing any VOC fugitive emission control requirement.
For this case, SOCMI facilities located in oxidant National Ambient Air
Quality Standard (NAAQS) attainment areas, in general, would not be subject
to any requirements. However, some states may require leak detection and
repair programs to control fugitive emissions of VOC through prevention of
significant deterioration (PSD) statutes. SOCMI facilities located in
6-4
-------
TABLE 6-2. REGULATORY ALTERNATIVES FOR FUGITIVE EMISSION SOURCES IN SOCMI
Source typed
Pumos
Light liquids
with single mechanical seals
with double mechanical seals
with no seals
Heavy liquids
with packed seals
with single mechanical seals
Valves (in-line)
Gas
Light liquid
Heavy liquid
Safety/relief valves
Gas
Light liquid
Heavy liquid
Open-ended valves and lines
Gas
Light liquid
Heavy liquid
Flanges
Sampling connections
Compressor seals
^Sources In VOC service.
Monitoring
interval
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
I
Equipment
specification
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
Monitoring
interval
Annually
Annual ly°
None
None
Hone
Quarterly
Annually
None
Quarterly0
None
None
Quarterly
Annually
None
None
Noned
Quarterly
Regulatory <
I!
Equipment
specification
None
Hone
None
None
None
N.
None
None
None
None
None
None
Capsf
Capsf
Caps
None
None
None
ilternative
Monitoring
interval
Monthly6
Monthly0
None
None
None
Monthly
Monthly
None
Monthly0
None
None
Monthly
Monthly
None
None
Noned
Monthly
111
Equipment
specification
None
None
None
None
None
None
None
None
None
None
None
Capsf
Capsf
Capsf
None
None
None
Monitoring
interval
Noneb
None6
None
None
None
Monthly
Monthly
None
Nonec
None
None
Monthly
Monthly
None
None
Noned
None
IV
Equipment
specification
Double seals;
degassing vents
connected to
control device
Degassing vents
connected to
control device6
None :
None
None
None
None
None
Upstream
rupture disks
None
None
Capsf
Capsf
Capsf
None
Closed loop
sampling
Seal area or
degassing vents
connected to
control device
"Plus weekly visual inspection. If liquid leak is observed, instrument monitoring is required to determine if action level is being exceeded.
Monitoring is required after each over pressure release. If it Is found to be leaking, the valve will be repaired
Included 1n open-ended valves.
eSealless pumps nay also be used.
'Or blinds, plugs', second valves. "
-------
non-attainment areas would be subject to the applicable SIP regulations and
other permitting requirements. In some areas control of fugitive VOC emissions
may be used to achieve hydrocarbon emission offsets. However, no present or
anticipated SIP regulations would be generally applicable to SOCMI. This
baseline control alternative merely presents a generalized fugitive emission
control level that can be used to compare the impacts of the more stringent
alternatives. As such, this alternative does not consider the levels of
control for specific facilities.
6.2.2 ReguaTatory Alternative II
This alternative would require leak detection and repair methods as
in the petroleum refinery control techniques guideline (CTG), EPA-450/2-78-036.
Leak detection would be accomplished by checking equipment components for
emissions of VOC using a portable VOC detection instrument to sample and
analyze the air in close proximity to the potential leak area. A measured
VOC concentration greater than some predetermined level, known as an "action
level", would be defined as a leak that would require equipment repair. A
measured VOC concentration less than the action level would not require equip-
ment repair. The action level is defined as 10,000 ppmv VOC concentration
for all cases.
Quarterly monitoring of compressors, gas service relief valves, inline
valves, and open-ended valves would be required. Annual monitoring of light
liquid service pumps and valves would be required. Weekly visual inspections
of light liquid pump seals would also be required. Leaks detected visually
would require instrument monitoring to determine if the action level is
exceeded. Relief valve monitoring after over pressure relieving would be
required. Open-ended valves would be required to be sealed with a cap, blind,
plug, or another valve.
6.2.3 Regulatory Alternative III
Regulatory Alternative III would provide for more restrictive control
than Alternative II by increasing the inspections for all applicable equipment
to monthly. Increasing the inspections would result in a reduction of
emissions from residual leaking sources; i.e., those sources which are found
6-6
-------
leaking and are repaired and recur before the next inspection and those
sources that begin leaking between inspection. Thus, although this alterna-
tive is similar in approach to Alternative II, it provides for more emissions
reduction. The requirements for weekly visual pump seal inspections, relief
valve monitoring.after over pressure, and caps for open-ended valves are the
same as Alternative II.
6.2.4 Regulatory Alternative IV
Alternative IV would require equipment specifications instead of more
frequent equipment inspections. This alternative would provide a more
restrictive level of control than the other alternatives. Several equipment
specifications would be required, including caps for open-ended valves as in
Alternatives II and III. Closed loop sampling techniques would be required
and rupture disks would be required on gas service relief valves venting to
atmosphere. The integrity of the disk would be required and replacement of
the disk would be required whenever a failure is detected. No monitoring
would be required for relief valves which have rupture disks upstream or
which vent to a control device header. Compressor seal areas and degassing
vents from seal oil reservoirs, or both, would be required to be connected
to a control device with a closed vent system. Pumps in light liquid
service would be required to have double mechanical seals with a seal oil
flushing system. The degassing vent from the seal oil reservoir would be
required to be connected to a control device with a closed vent system.
6.3 REFERENCES
1. Wetherold, R.G., L.P. Provost, D.D. Rosebrook, and C.D. Smith.
Emissions Factors and Frequency of Leak Occurrence for Fittings in
Refinery Process Units. Interim Report. Radian Corporation.
Austin, Texas. EPA Contract Number 68-02-2665. p. 11-49.
2. The Assessment of Environmental Emissions from Oil Refining. Draft
Report, Appendix B. Radian Corporation. Austin, Texas. August 1979.
pp. 3-5 through 3-16.
3. Erikson, D.G., and V. Kalcevic. Emissions Control Options for the
Synthetic Organic Chemicals Manufacturing Industry. Draft Report.
Hydroscience, Inc. Knoxville, Tennessee, EPA Contract Number 68-02-2577.
p. IV-1,2.
6-7
-------
4. Ref. 3, p. II-9-13.
5. Ref. 1, p. 11-23.
6. Ref. 3, p. 11-10.
7. Ref. 3, p. IV-8.
6-8
-------
7. ENVIRONMENTAL IMPACT
The environmental impacts that would result from implementing the
regulatory alternatives being considered in this study are examined in
this chapter. Included in this chapter are estimations of the controlled
VOC fugitive emissions and the incremental reductions in uncontrolled VOC
emissions that could be achieved under each of the alternatives. Also,
the impacts of these regulatory alternatives on water quality, waste water
generation and treatment, solid waste generation and treatment or disposal,
and energy consumption or savings are discussed.
7.1 IMPACT ON ATMOSPHERIC EMISSIONS
Implementation of Regulatory Alternatives II, III, or IV, would
reduce VOC fugitive emissions from the SOCMI. To quantify these reduc-
tions, the controlled VOC emission levels from emission sources in the
model units (described in Chapter 6) were estimated for each alternative.
These emission levels are developed below for individual emission sources, for
model units in SOCMI, and then for SOCMI as a whole.
7.1.1 Emission Source Characterization
As indicated in Chapter 6, a SOCMI model unit typically consists of
several types of process equipment that contribute to fugitive VOC emis-
sions. Under Regulatory Alternative I (baseline case), all these sources
are "uncontrolled" emission sources. However, if Regulatory Alternative
II, III, or IV were implemented, the emissions from some uncontrolled
sources would be reduced; these sources would subsequently become "con-
trolled" sources. Both the controlled and uncontrolled sources are
important because the total fugitive VOC emissions from the model units
and ultimately the SOCMI are the sum of emissions from both types of
sources.
7-1
-------
7.1.2 Development of VOC Emission Levels .
The uncontrolled emission levels were previously presented in Chapter
3 (Table 3-1). Controlled emission levels were developed for those
sources that would be controlled by the implementation of a regulatory
alternative. These controlled fugitive emission levels were calculated by
multiplying the uncontrolled emissions from this equipment by a "control
efficiency" presented in Chapter 4, Tables 4-2 through 4-4. The resulting
controlled VOC emission factors for each source are presented in Tables
7-1, 7-2, and 7-3 for Regulatory Alternatives II, II, and IV, respectively.
The total VOC fugitive emissions from Model A, Model B, and Model C
units in the SOCMI were determined under each regulatory alternative.
Initially, emissions from each source type within a model unit were
estimated by using the model unit equipment inventories presented in
Table 6-1 and the source emission factors presented in Tables 7-1, 7-2,
and 7-3. These emissions were then used to estimate the VOC fugitive
emissions from each of the three model units. An example calculation is
presented in Table 7-4 to illustrate the procedure used to estimate the
total VOC fugitive emissions from a model unit under Regulatory Alterna-
tive II. The total VOC fugitive emissions calculated for the respective
model units under each regulatory alternative are presented in Table 7-5.
Also, presented in this table are the average percent reductions in the
baseline emission levels that result from implementing Regulatory Alterna-
tive II, III, or IV. The incremental reductions in fugitive emission
levels achieved by implementing the alternatives are also presented in .
Table 7-5.
7-1-3 Fujbure Impact on V_OC_FucjHive Emissions
In order to assess the future impacts of the various regulatory
alternatives on VOC fugitive emissions from the SOCMI, the levels of
these emissions were estimated for a period of five years after adoption
of a regulatory alternative. These emissions were estimated by using:
1) the emission factors presented in Tables 7-1, 7-2,
and 7-3;
2) the industry population for the assumed base year
of 1980;
7-2
-------
TABLE 7-1. EMISSION FACTORS FOR SOURCES CONTROLLED UNDER REGULATORY ALTERNATIVE II '
I
CO
Uncontrolled
emission source
Pumps
Light liquid service
Valves
Gas service
Light liquid service
Safety/relief valves
Gas service
Compressors
Inspection9
interval
Yearly
Quarterly
Yearly
Quarterly
Quarterly
Uncontrollf
emission
factor,
kg/hr
0.
0.
0.
0.
0.
120
021
010
160
440
;d Correction
factors
Ac
0.87
0.98
0.84
0.69
0.84
Bd
0.80
0.90
.0.80
0.90
0.90
. ce
0.98
0.98
0.98
0.98
0.98
Df
0.92
0.99
0.94
0.97
0.97
Control
efficiency
(AxBxCxD)
0.
0.
0.
0.
0.
63
86
62
59
72
Controlled9
emission
factor,
kg/hr
0.
0.
0.
0.
0.
044
003
004
067
126
aFrom Table 6-2.
bFrom Table 3-1.
theoretical maximum control efficiency.1
Leak occurrence and reoccurrence correction factor - assumed to be 0.80 for yearly inspection, 0.90
for quarterly inspection, and 0.95 for monthly inspection.2
eNon-instantaneous repair correction factor - for a 15-day maximum allowable repair time, the 7.5-day
average repair time yields a 0.98 yearly correction factor [365 - (15/2)] * 365.3
Imperfect repair correction factor - calculated as 1 - (f v F). Where f = average emission rate for
sources at 1000 ppm and F = average rate for emission sources greater than 10,000 ppm. *
^Controlled emission factor = uncontrolled emission factor x [1 - (A x B x C x D)].
-------
TABLE 7-2. EMISSION FACTORS FOR SOURCES CONTROLLED UNDER REGULATORY ALTERNATIVE III
Uncontrolled
emission source
Pumps
Light liquid service
Valves
Gas service
Light liquid service
Safety/relief valves
Gas service
Compressors
Inspection9
interval
Monthly
Monthly
Monthly
Monthly
Monthly
Uncontrolled
emission
factor,
kg/hr
0.
0.
0.
0.
0.
120
021
010
160
440
b
AC
0.87
0.98
0.84
0.69
0.84
Correction
factors
Bd
0.95
0.95
0.95
0.95
0.95
ce
0.98
0.98
0.98
0.98
0.98
Df
0.92
0.99
0.94
0.97
0.97
Control
efficiency
(AxBxCxD)
0.
0.
0.
0.
0.
75
90
74
62
76
Controlled9
emission
factor,
kg/hr
0.
0.
0.
0.
0.
030
002
003
061
108
From Table 6-2.
bFrom Table 3-1.
C 6
Theoretical maximum control efficiency-
Leak occurrence and reoccurrence correction factor - assumed to be 0.80 for yearly inspection, 0.90
for quarterly inspection, and 0.95 for monthly inspection.
Non-instantaneous repair correction factor - for a 15-day maximum allowable repair time, the 7.5-day
average repair time yields a 0.98 yearly correction factor [365 - (15/2)] * 365. 8
Imperfect repair correction factor - calculated as 1 - (f -r F). Hhere f = average emission rate for
sources at 1000 ppm and F = average rate for emission sources greater than 10,000 ppm. '10
^Controlled emission factor = uncontrolled emission factor x [1 - (A x B x C x D)].
-------
TABLE 7-3. EMISSION FACTORS FOR SOURCES CONTROLLED UNDER REGULATORY ALTERNATIVE IV
-vl
I
01
Uncontrolled
emission source
Pumps
Light liquid service
Valves
Gas service
Light liquid service
Safety/relief valves
Gas service
Compressors
Sampling connections
Inspection3
interval
None
Monthly
Monthly
None
None
None
Uncontrolled
-- -emission
factor,
kg/hr
0.120
0.021
0.010
0.160
0.440
0.015
b
Ac
u
NAh
0.98
0.84
NA
NA
NA
Correction
"factors -
j
Bd
NA
0.95
0.95
NA
NA
NA
Q
ce
NA
0.98
0.98
NA
NA
NA
---- ---Control " "
^ ell 1 L 1 cilLy
DT (AxBxCxD)
NA
0.99 0.90
0.94 0.74
NA
NA
NA
Controlled9
-'emission
factor,
kg/hr
0.0
0.002
0.003
0.0
0.0
0.0
11
From Table 6-2.
bFrom.Table 3-1.
°Theoretical maximum control efficiency.
Leak occurrence and recurrence correction factor - assumed to be 0.80 for yearly inspection, 0.90 for
quarterly inspection, and 0.95 for monthly inspection.12
eNon-instantaneous repair correction factor - for a 15-day maximum allowable repair time, the 7.5-day
average repair time yields a 0.98 yearly correction factor [365 - (15/2)] -r 365.13
Imperfect repair correction factor - calculated as 1 - (f 4- F). Where f = average emission rate for
sources at 1000 ppm and F = average rate for emission sources greater than 10,000 ppm. '
^Controlled emission factor = uncontrolled emission factor x [1 - (A x B x C x D)].
Since the equipment associated with this regulatory alternative essentially eliminates fugitive
emissions, these correction factors are not applicable.
-------
TABLE 7-4. EXAMPLE CALCULATION OF VOC FUGITIVE EMISSIONS FROM MODEL
UNIT A UNDER REGULATORY ALTERNATIVE II
Number of
sources in
model unit9
(N)
Emission Source:0
Pumps
Light liquidd single
mechanical seal
Light liquidd double
mechanical seal
Heavy liquid6 single
mechanical seal
Heavy liquid6 packed seal
In-line valves
Vapor service
Light liquidd service
Heavy liquid6 service
Safety/ relief valves
Vapor service
Light liquidd service
Heavy liquid6 service
Open-ended valves
Vapor service
Light liquidd service
Heavy liquid6 service
Compressors
Sampling connections
Flanges
5
3
5
2
90
84
84
11
1
1
9
47
48
1
26
600
Total
Emission
factor, b
kg/hr-source
(E)
0.044
0.044
0.020
0.020
0.003
0.004
0.0003
0.067
0.006
0.009
0.003
0.004
0.003
0.126
0.015
0.0003
emissions
Emissions
from sources
kg/hr
(N x E)
0.220
0.132
0.100
0.040
0.270
0.336
0.025
0.737
0.006
0.009
0.027
0.188
0.014
0.126
0.390
0.180
2.800
Model units are characterized in Table 6-1.
Emission factors from Tables 3-1 and 7-1.
cSources in VOC service.
Light liquid service means that the fugitive emission source contains a
liquid which has a vapor pressure equal to or greater than 0.3 kPa at
20°C.
"Heavy liquid service means that the fugitive emission source contains a
liquid which has a vapor pressure less than 0.3 kPa at 20°C.
Open-ended valve factor is equivalent to the in-line valve factor because
capping the open end is assumed to eliminate emissions from this source.
7-6
-------
TABLE 7-5. ESTIMATED EMISSIONS AND EMISSION REDUCTIONS ON A MODEL UNIT BASIS'
Regulatory
Alternative
I
II
III
l\l
Estimated emissions^ >c
(Mg/yr)
Model unit
ABC
67 260 800
24 94 290
21 80 250
8 34 106
Average percent
reduction from emissions
estimated under
Regulatory Alternative I
63
69
87
Average incremental
percent reduction
in emissions
63
6
18
The emissions and percentage reductions presented in this table were calculated using the following:
controlled and uncontrolled emission factors (see Tables 7-1, 7-2, and 7-3), and
emission sources given in Table 6-1.
3A year is assumed to be equivalent to 8,760 hours.
C1.0 Mg/yr = 2200 pounds/yr
-------
3) annual replacement of the industry population based on
a twenty-year equipment life ; and
4} annual growth rate of 5.9 percent for the industry.
Using these bases and the techniques presented in Appendix E, the total
number of affected facilities (model units) in operation in 1981 were
estimated to be 148. In 1985 the total number of affected facilities were
I Q
estimated to be 831.
Under Regulatory Alternative I, the total VOC fugitive emissions from
affected facilities were estimated to increase from 35 to 199 gigagrams
per year (Gg/yr) during the same five-year (1981-1985) period (see Table 7-6).
In the same time period, implementation of Regulatory Alternative II could
be expected to reduce the baseline case (Regulatory Alternative I) fugitive
emissions 65 percent. Implementation of Regulatory Alternative III would
reduce the baseline emissions by 69 percent. As Table 7-5 indicates,
Regulatory Alternative IV, the most stringent of all the proposed alterna-
tives, would reduce the baseline emissions by about 87 percent.
7.2 IMPACT ON WATER QUALITY
In the absence of standards to reduce fugitive emissions of VOC from
SOCMI and under normal equipment operation, liquid leaks from various
equipment components could increase the quantity of wastewater generated
by a "typical" SOCMI facility. Under Regulatory Alternative I, liquid leaks
could originate from pumps and process valves in light or heavy liquid
service as well as valves on open-ended lines in light or heavy liquid service
and enter the wastewater system as runoff. Although the uncontrolled emission
rates for these sources are given in Chapter 3, the gas-liquid split of
these emissions i.s not defined. Consequently, the increase in wastewater
from SOCMI due to liquid leaks from potential fugitive emission sources
cannot be quantified.
Implementation of Regulatory Alternative II could reduce the wastewater
from a "typical" SOCMI facility by reducing the fugitive liquid emissions
resulting under Alternative I. The reduced emissions would be due to the
use of caps, plugs or second valves on open-ended lines in gas and light or
heavy liquid service. For example, caps, plugs, or second valves required
7-8
-------
under Alternative II would reduce the VOC fugitive emission rate from open-
ended lines in light or heavy liquid service from 0.01 kg/hr under Alternative
I to 0.004 kg/hr. This reduction would reflect a reduction in gaseous
emissions and liquid leaks. Since the gas-liquid split of the emission from
a given source is site specific, the impact of Alternative II on waste-
water from SOCMI cannot be quantified. However, it is likely that this
impact would be minor.
Implementation of Alternative III would result in impacts on wastewater
from SOCMI similar to those resulting from Alternative II. However, the
impacts under Alternative III would be more pronounced due to the more
frequent inspection intervals required by this alternative. The more
frequent intervals would reduce the VOC fugitive emission rate from values
in light or heavy liquid service from 0.004 kg/hr under Alternative II to
0.003 kg/yr under Alternative III. Similarly, the fugitive emission rate
from pumps in light liquid service would be 0.044 kg/hr under Alternative II
and 0.03 kg/hr under Alternative III. Consequently, the potential for
wastewater production by possible fugitive emission sources in SOCMI would
be less under Alternative III than under Alternative II.
Of the alternatives being considered, Regulatory Alternative IV
could have the greatest impact on the quality of water that is discharged
from a "typical" SOCMI facility. Implementation of this alternative could
have positive (and possibly some negative) impacts on wastewater depending on
the specific control device requirements at each unit. Implementation of
Regulatory Alternative IV could reduce the amount of wastewater from a
SOCMI facility by reducing the fugitive liquid emissions resulting under
Alternative I. The reduction of these emission levels is primarily due to
the reduction of leaks from equipment in light liquid service, e.g., from
the use of double mechanical seals for pumps and closed loop sampling. Under
Regulatory Alternative IV, a double mechanical seal-degassing vent arrangement
reduces the emission rate of a pump seal in light liquid service under
Regulatory Alternative I from 0.12 to 0.0 kg/hr. A portion of this emission
reduction would be a reduction in liquids leaked to the ground or ditch.
However, the amount of liquids leaked to the ground or ditch that could enter
a plant wastewater system is not known.
7-9
-------
TABLE 7-6. TOTAL VOC FUGITIVE EMISSIONS FROM AFFECTED MODEL UNITS
FOR REGULATORY ALTERNATIVES
Year
1981
1982
I! 1983
o
1984
1985
Number of affected
model units9
A B C
77 49 22
158 100 46
244 155 71
335 213 97
432 274 125
Total fugitive emissions estimated
under Regulatory Alternative^5'0
I
(Gg/yr)
35.4
73.1
113.0
155
199
II
(Gg/yr)
12.9
26.7
41.2
56.5
72.8
III
(Gg/yr)
11.0
22.8
35.2
48.3
62.1
IV
(Gg/yr)
4.6
9.5
14.8
20.2
26.0
The bases for estimating the number of model units,-as detailed in Appendix E, are:
an industry growth rate of 5.9 percent per year,'
unit replacement based on a 20-year equipment life, and
a base year (1980) total of 872 Model A, 554 Model B, and 252 Model C Units.
Estimated total VOC fugitive emissions from Model Units A, B, and C.
Q
Does not include emissions from units in existence prior to 1981.
-------
Implementation of Regulatory Alternative IV could also result in a
negative impact on water quality due to the operation of an ultimate con-
trol device which "captures" the fugitive VOC's. If a carbon adsorption
device were used to capture any VOC released at the degassing vent, a waste-
water containing suspended solids and some dissolved organics could be pro-
duced during the carbon regeneration process if the carbon is regenerated
at the unit. The use of a refrigeration process as the ultimate control
device could possibly result in a condensate containing dissolved organics.
The wastewater flow rates would be quite small and would generally be suitable
for treatement in the existing unit wastewater treatment process. Overall
the impacts, both positive and negative, of Alternative IV on wastewaters
from SOCMI would be minor.
7.3 IMPACT ON SOLID WASTE
In the absence of standards to reduce fugitive emissions of VOC from
SOCMI and under normal operation, solid wastes that could result from SOCMI
include replaced seals, packing, rupture disks, equipment components such
as pumps and valves, spent catalysts, and polymerization products. Metal
solid wastes such as mechanical seals, rupture disks and disposed valves can
be sold as scrap metal to companies which can recycle the metal. This would
help to minimize the impact on solid waste. The quantity of used valve
packings and used batteries for monitoring instruments would not signifi-
cantly contribute to solid waste.
Implementation of Alternatives II and III would require the use of caps,
plugs, or second valves on open-ended lines in light or heavy liquid service,
and more frequent monitoring intervals. Implementing either of these
alternatives would have no greater impact on solid waste than Alternative I.
This is due to the relatively long life of caps, plugs, and second valves on
open-ended lines as well as the ability to sell discarded components such
as valves, mechanical seals, and rupture disks as scrap metal.
Implementation of Regulatory Alternative IV could res.ult in the
generation of solid waste if carbon adsorption were used as a control
device and the carbon disposed of instead of being regenerated. However,
the carbon could be sent back to the manufacturer for regeneration, and
thereby reducing the.solid waste problem at the facility. It is antici-
pated that.the manufacturer could incinerate or commercially dispose of
7-11
-------
any carbon that could not be regenerated (such as carbon fines) without
any serious environmental problems. Consequently, the negative impact of
implementing Alternative IV would be minor.
7.4 ENERGY IMPACT
Regulatory Alternatives II, III and IV call for passive controls on
euipment handling VOC streams (i.e., pump seals, process vent enclosures,
degassing vents, etc.); so implementing any of these alternatives will not
significantly increase the energy usage of a typical SOCMI plant. If a
control device such as carbon adsorption were used, steam (or another hot
regenerating medium) would be needed to regenerate the carbon at the unit;
however, the energy requirements would be quite small. The energy require-
ments of vapor recovery systems and of closed loop sampling would also be small
Because fugitive emissions of VOC have an energy value, implementation of
any of the alternatives being considered will result in a positive energy
impact.
The average energy value of the fugitive VOC emissions from SOCMI is
6 19
estimated to be approximately 31 x 10 joule/kg. The energy savings
resulting from the fugitive VOC emission reductions associated with
Alternatives II, III, and IV are presented in Table 7-7. Because Alterna-
tive IV is the most stringent, it will result in the greatest emission
reduction. As Table 7-7 indicates, implementation of this regulatory alter-
native would reduce the uncontrolled fugitive emissions by 173 Gg in the
fifth year and by a total of 520 Gg over a five-year period after implemen-
tation. These "recovered" VOC emissions have a total energy value of
13 6
1.55 x 10... joules based on an average heating value of 31 x 10 joule/kg.
6 20
Assuming an energy value of 5.8 x 10 Btu per barrel of crude oil, the
energy value of the total fugitive emissions recovered over the five-year
period is approximately equal to 2.5 million barrels of crude oil under
Regulatory Alternative IV. This corresponds to an average daily savings of
1390 bbl/day of crude oil over the five-year period.
7-12
-------
TABLE 7-7. ENERGY IMPACT OF EMISSION REDUCTIONS FOR REGULATORY ALTERNATIVES
Year
1981
1982
1983
1984
1985
5-year
total
Reduction from baseline Energy value of emission
emissions under reductions under Crude oil equivalent
Regulatory Alternatives, Regulatory Alternatives , of emission reductions,
Gga terajouleb thousand barrels
II III IV II III IV IIC IIIC IVC
22.4 24.4 30.8 694 756 955 113 124 . 156
46.4 50.3 63.6 1,440 1,560 1,970 235 255 322
71.8 77.8 98.2 2,230 2,410 3,040 364 394 497
98.3 106 135 3,050 3,290 4,180 498 538 683
127 137 173 3,940 4,250 5,360 644 695 876
366 396 500 11,350 12,270 15,500 1,855 2,005 2,530
Estimated total VOC fugitive emission reduction from Model Units A, B, and C.
Based on 1.55 x 10 joules/kg : This may be slightly over estimated if safety/
relief valves are controlled by a closed vent and flare system.
r 6
Based on 5.8 x 10 Btu/bbl crude oil.
-------
7.5 OTHER ENVIRONMENTAL CONCERNS
7.5.1 Irreversible and Irretrievable Commitment of Resources
Implementation of any of the various alternatives is not expected to
result in any irreversible or irretrievable commitment of resources. As
previously noted, the regulatory alternatives should help to save resources
due to the energy savings associated with the reductions in emissions.
7.5.2 Environmental Impact of Delayed Standards
As it was indicated above, implementation of the standards will only
have minor impacts on water and solid wastes. Consequently, delaying the
standards would have essentially no impact on these problems. However, a
delay in implementing the alternatives would have a greater impact on air
pollution and associated energy losses. The air and energy impacts of
delayed standards are shown in Table 7-7. The emission reductions and
associated energy savings shown would be irretrievably lost at the rates
shown for each of the five years.
7.6 REFERENCES
1. Wetherold, R. G., L. P. Provost, D. D. Rosebrook, and C. D. Smith.
Emission Factors and Frequency of Leak Occurrence for Fittings in'
Refinery Process Units. EPA Contract No. 68-02-2665, 600/2-79-44.
2. Tichenor, B. A., K. C. Hustvedt, and R. C. Weber. Controlling
Petroleum Refinery Fugitive Emissions Via Leak Detection and Repair.
Draft. Symposium on Atmospheric Emissions from Petroleum Refineries
Austin, Texas. '
3. Ref. 2.
4. Ref. 1.
5. Ref. 2.
6. Ref. 1.
7. Ref. 2.
8. Ref. 2.
9. Ref. 1.
7-14
-------
10. Ref. 2.
11. Ref. 1.
12. Ref. 2.
13. Ref. 2.
14. Ref. 1.
15. Ref. 2.
16. Letter from .Charles A. Muela, Radian Corporation, to K. C. Hustvedt,
EPA Office of Air Quality Planning and Standards. May 11, 1979.
Replacement Rate of Process Unit in the Organic Chemical Industry.
17. The American Economy, Prospects for Growth to 1991. New York:
McGraw-Hill, 1979."
18. Letter from Vincent Smith, Research Triangle Institute, to
Russell L. Honerkamp, Radian Corporation. November 30, 1979.
Projected Number of Affected Facilities and Average Product Value in
SOCMI.
19. Memo from J. R. Blacksmith, Radian Corporation, to K. C. Hustvedt,
EPA. December 14, 1979. Average Energy Value of SOCMI Chemicals.
20. American Petroleum Institute. Petroleum Facts and Figures.
Washington, D.C. 1971.
21. Ref. 19.
7-15
-------
8. COST ANALYSIS
8.1 COST ANALYSIS OF REGULATORY ALTERNATIVES
8.1.1 Introduction
c
The costs of implementing the regulatory alternatives for controlling
fugitive emissions of volatile organic compounds (VOC) from the synthetic
organic chemicals manufacturing industry (SOCMI) are presented in the
following sections. A detailed description of the model units and regula-
tory alternatives being considered in this analysis is presented in
Chapter 6. Three different model units (A, B, and C) have been chosen to
represent the range of emission source populations that exist in SOCMI
units. Regulatory Alternative I (no controls) is used as the baseline,
and Regulatory Alternatives II, III, and IV are increasingly restrictive
control alternatives.
8.1.2 New Facilities
8.1.2.1 Capital Costs. The bases for the capital costs for the
model units are presented in Table 8-1. The capital cost estimates for
each model unit under each regulatory alternative are given in Table 8-2.
Regulatory Alternative I requires no control of VOC emissions. Conse-
quently, there are no capital costs associated with this alternative.
The capital costs for the model units are the same under Regulatory
Alternative II as under Regulatory Alternative III, since the only change
is the monitoring frequency. These costs reflect the purchase of two VOC
monitoring instruments and caps for all open-ended lines. It is assumed
that one monitoring instrument is used as a standby spare.
Under Regulatory Alternative IV, like II and III, two monitoring
instruments and caps for all open-ended lines are purchased. In addition,
there are several other capital expenditures. All single seal pumps in
8-1
-------
TABLE 8-1. CAPITAL COST DATA
Item
Value used in analysis
(last quarter 1978 $)
1. Monitoring instrument
2. Caps for open-ended lines
3. . Double mechanical seals
4. Flush oil system for double mechanical seals
5. Closed vents for. degassing reservoirs of compressors
and double seal pumps
6. Rupture disks for relief valves
7. Closed loop sampling connections
2 x 4250 = 8500/model unit0
45/lineb
575/pump (new)0 H
850/pump (retrofit)
1500/pumpe>f
6530/compressor^
3265/pump3>h
1730/relief valve (new)1.
3110/relief valve (retrofit)J
k
460/sample connection
aOne instrument used as a spare (Ref. 1). Cost from Ref. 2.
bBased on installation of a 2.5 cm. screwed valve. Cost (1967)= $12 (Ref. 3, p. 450).
Cost index = 278.1 -f 113 (Ref. 4 and 5). Installation = 1 hour at $15/hour (Ref. 6,7, 8).
cFrom Ref. 6, p. IV-3. Seal cost = $560. Single seal credit = $225. Shop installation = $240.
dFrom Ref. 10, p. IV-3. Seal cost = $560. Field installation = $290.
eFrom Ref. 11, p. IV-3. Pressurized reservoir system = $700. Flush system cooler = $800.
Pumps that have double mechanical seals without regulatory requirement may not have the cost
of a flush system added. The flush system is assumed to be an integral part of the double .
seal system. .
9From Ref. 12, pp. IV-8,9.. Based on installation of a 122 m. length of 5.1 cm. diameter,
schedule 40 carbon steel pipe at a cost of $5200; plus three 5.1 cm. ca'st steel plug valves
and one metal gauze flame arrestor at a cost of $1330. These costs include connection of the
degassing reservoir to an existing enclosed combustion device or vapor recovery header. Cost
of a control device added specifically to control .the degassing vsnts is, therefore, not
included.
This cost is based on the assumption that two pumps (such as a pump and its spare) are
connected to a single degassing vent.
of rupture disk assembly from Ref. 13, p. IV-8. One 7.6 cm. rupture disk, stainless'
steel = $195. One 7.6 cm. rupture disk holder, carbon steel = $325. One 0.6 cm. pressure
gauge, dial face = $15. One 0..6 cm. bleed valve, carbon steel, gate = $25. Installation =
$240. .In order to allow in-service disk replacement, a block valve must De installed
upstream of the rupture disk. Cost (1967) from Ref. 14, p. 451, for one 7.6 cm. gate valve -
$240. Cost- index = 278.1 T 113 (Ref. 15 and 16). Installation = 10 hours at $15/hour
(Ref .17,18,19). In order to prevent damage to the relief valve by disk fragments, an
offset mounting is required. Cost (1967) from Ref. 20, p. 450 for one 10.2 cm. tee and
one 10.2 cm. elbow = $7.. 30. Cost index = 278.1 v 113 (Ref. 21 and 22). Installation = 8
hours at $15/hour (Ref -23,24,25).
JCosts for the rupture disk, holder, and block valve are the same as for the new applications.
An additional cost is added to replace the de-rated relief valve. No credit is assumed for
the used relief valve. Cost (1967) for one 7.6 cm. pressure reducing valve, stainless steel
body and trim from Ref. 3, p. 452 = $500. Cost index = 278.1 v 113 (Ref .26 and 27).
Installation = 10 hours at $15/hour (Ref. 28,29,30)'.
k
Based on installation of a 6 m. length of 2.5 cm. diameter, schedule 40, carbon steel pipe
and three 2.5 cm. carbon steel ball valves. Costs from Ref.31 , p. IV-8. Installation =
18 hours at $15/hour.
8-2
-------
TABLE 8-2. CAPITAL COST ESTIMATES FOR NEW MODEL UNITS
(thousands of last quarter 1978 dollars)
Model
1.
2.
3.
4.
5.
6.
7.
8.
Model
1.
2.
3.
4.
5.
6.
7.
8.
Model
1.
2.
3.
4.
5.
6.
7.
8.
Capital cost item3 I
Unit A
Monitoring instrument
Caps for open-ended lines
Double mechanical seals
Seals
Installation
Flush oil system for double mech. seals
Vents for compressor degassing reservoirs
Vents for pump degassing reservoirs
Rupture disks for relief valves
Disks
'Holders, block valves, installation
Closed loop sampling connections
Total 0.0
Unit B
Monitoring instrument
Caps for open-ended lines
Double mechanical seals
Seals
Installation
Flush oil system for double mech. seals
Vents for compressor degassing reservoirs
Vents for pump degassing reservoirs
Rupture disks for relief valves
Disks
Holders, block valves, installation
Closed loop sampling connections
Total 0.0
Unit C
Monitoring instrument
Caps for open-ended lines
Double mechanical seals
Seals
Installation
Flush oil system for double mech. seals
Vents for compressor degassing reservoirs
Vents for pump degassing reservoirs
Rupture disks for relief valves
Disks
Holders, block valves, installation
Closed loop sampling connections
Total 0.0
Regulatory alternative
II III IV
8.50 8.50 8.50
4.68 4.68 4.68
1.68
1.20
7.5
6.53
26.1
2.14
16.8
12.0
13.2 13.2 87.1
8.50 8.50 8.50
18.7 18.7 18.7
6.36
4.56
28.5
13.1
94.7
8.19
64.4
47.8
27.2 27.2 295
8.50 8.50 8.50
57.5 57.5 57.5
20.1
14.4
90.0
52.2
297
25.4
199
147
66.0 66.0 911
From Tables 6-1 and 8-1.
8-3
-------
light liquid service must have double mechanical seals installed. This
is at a cost of $575/pump. A flush oil system ($1500/pump) must also be
used in conjunction with the double mechanical seals. Existing pumps
with double mechanical seals are assumed to have a flush oil system
already incorporated. Hence, there is no additional, capital expenditure
for the double seals or flush system.
Also, under Regulatory Alternative IV, compressor seals and pump
seals must have the seal oil degassing vents that are connected to a
control device such as a vapor recovery system or an enclosed combustion
device. The cost is estimated to be $6530 per compressor and $3265 per
pump. This cost is based on the assumption that one closed vent system
is required for each compressor. Since main pumps and spares are
generally located in close proximity to each other, one closed vent
system is required for each pair of pumps. These costs are based on
connecting the closed vent system to an existing control device.
The costs of purchasing and installing rupture disks is $1590 per
relief valve. The rupture disks are to be installed upstream of relief
valves in gas service. The cost includes the purchase of a shutoff
valve to allow the disk to be replaced after overpressure relief.
The closed loop sampling connection costs are based on an estimate
of $460 per sampling connection for installation of 6m. of pipe and three
valves.
8.1.2.2 Annual Costs. With the implementation of Regulatory
Alternatives II, III, or IV, visual and/or instrument monitoring of
potential sources of fugitive VOC emissions will be required. A summary
of the requirements for the different alternatives is presented in
Chapter 6. Tables 8-3, 8-4, and 8-5 give the monitoring labor-hour
requirements for Regulatory Alternatives II, III and IV, respectively.
The labor-hour requirements are calculated by taking the product of the
number of workers needed to monitor a component (1 for visual, 2 for
instrument), the time required to monitor, the number of components in
the model unit, and the number of times the component is monitored per
year. Monitoring labor costs are then calculated based on $15 per hour.'32»33>3£+
Regulatory Alternative III hasthe highest annual monitoring costs.
8-4
-------
TABLE 8-3. ANNUAL MONITORING AND LEAK REPAIR LABOR REQUIREMENTS
FOR REGULATORY ALTERNATIVE II
Monitoring
Number of
components per
model unit
Source type ABC
Pumps (light liquid)
Single mechanical 5 19 60
seals
Double mechanical 3 10 31
seals
Valves (in-line)
Gas 90 365 1117
Light liquid 84 335 1037
CD . Safety/relief valves 11 42 130
cj, (gas service)
Valves on open-ended
1 ines^1
Gas 9 37 115
Light liquid 47 189 581
Compressor seals 1 2 8
Monitoring
Type ofa time.b
monitoring min
Instrument 5
Visual 0.5
Instrument 5
Visual 0.5
Instrument 1
Instrument 1
Instrument 8
Instrument 1
Instrument 1
Instrument 10
Times
monitored
per year
1
52
1
52
4
1
4
4
1
4
Leak repair
Estimated
Monitoring labor- number of . Repair Leak repair labor-
hours required0 leaks per year time, hours required6
A B
1.0 3.2
2.2 8.2
1.0 1.7
V-l3'"* ( 4-3
12.0 49.0
2.8 11.2
11.7 44.8
1.2 4.9
1.6 6.3
1.3 2.7
C ABC
10.0 1 1 3
26.0
5.2 112
13.4
149.0 4 15 45
34.6 3 9 25
139.0
15.3 125
19.4 2 6 14
10.7 1 1 2
hrs ABC
80 b 80 80 240
80b 80 80 160
1.13f 4.5 17.0 50.9
1.13f 3.4 10.2 28.3
O9 0 0 0
1.13e 1.1 2.3 5.7
1.13e 2.3 6.8 15.8
40b 40 40 80
'2 workers for instrument monitoring, 1 for visual. Ref. 35, p. 4-3.
uRef. 36.
^Monitoring labor-hours = number of workers a number of components a time to monitor (total is minimum of 1 hr).
From Table 4-2.
p
Leak repair labor-hours = number of leaks x repair time.
f
h°UP
°f the
.
is assumed that these leaks are corrected by routine maintenance at no additional labor requirements. Ref. 38.
-------
TABLE. 8-4.. ANNUAL MONITORING AND LEAK REPAIR LABOR REQUIREMENTS
FOR REGULATORY ALTERNATIVE . 111.
oo
I
en
Monitoring
Number of
. components per
model unit
Source type ABC
Pumps (light liquid)
Single mechanical 5 19 60
seals
Double mechanical 3 10 31
seals
Valves (in-line)
Gas 90 365 11'17
Light liquid 84 335 1037
Safety/ relief valves 11 42 130
(gas service)
Valves on open-ended
1 inesh
Gas 9 37 115
Light liquid 47 189 581
Compressor seals 1 2 8
Type ofa
monitoring
Instrument
Visual
Instrument
Visual
Instrument
Instrument
Instrument
Instrument
Instrument
Instrument
Monitoring
time, b
min
5
0.5
5
0.5
1
1
8
1
1
10
Times
monitored
per year
12
52
12
52
12
12
12
12
12
12
Leak repair
Estimated
Monitoring labor- number of Repair Leak repair labor-
hours required0 leaks per yeard time, hours required6
A
10.0
2.2
6.0
1.3
36.0
33.6
35.2
3.6
18.8
4.0
B
38.0
8.2
20.0
4.3
146.0
134.0
134.4
14.8
75.6
8.0
C ABC
120.0 1 3 9
26.0
62.0 1 2 5
13.4 .
446.8 6 2'2 68
414.8 7 25 75
416.0
46.0 1 3 7
232.4 4 14 42
32.0 1 1 2
nrs A B C
80 b 80 240 720
80 b 80 160 400
1.13f 6.8 24.9 76.8
1.13f 7.9 28.3 84.8
O9 0 0 0
1.13e 1.1 3.4 7.9
1.13e 4.5 15.8 47.5
40b 40 40 80
a2 workers for instrument monitoring, 1 for visual. Ref. 39.
bRef. 40. .
GMonitoring labor-hours = number of workers x number of components x time to monitor (total is minimum of 1 hr] .
dFrom Table 4-2.
eLeak repair labor-hours = number of leaks x repair time.
Weighted average based on 75 percent of the leaks repaired on-line, requiring 0.17 hour per repair, and on 25 percent of the leaks repaired
off-line, requiring 4 hours per repair. Ref. 41 .
^It is assumed that these leaks are corrected by routine maintenance at no additional labor 'requirements. Ref. 42 .
The estimated number of leaks per year for open-ended valves is based on the same percent of sources used for in-line valves. This represents
leaks occurring through the stem and gland of the open-ended valve. Leaks through the seat of. the valve are eliminated by addinq caps for
Regulatory alternatives II, III, IV. '
-------
TABLE 8-5. ANNUAL MONITORING AND LEAK REPAIR LABOR REQUIREMENTS FOR
REGULATORY ALTERNATIVE IV.
CO
I
Monitoring
Number of
components per
model unit
Source type ABC
Pumps (light liquid)
Single mechanical 5 19 60
seals converted to
double seals
Double mechanical 3 10 31
seals
Valves (in-line)
Gas 90 365 1117
Light liquid 84 335 1037
Safety/relief valves 11 42 130
(gas service)
Valves on open-ended
lines'*
Gas 9 37 115
Light' liquid 47 189 581
Compressor seals 1 2 8
Type ofa
monitoring
Instrument
Visual
Instrument
Visual
Instrument
Instrument
Instrument
Instrument
Instrument
Instrument
Monitoring
time, b
min
5
0.5
5 .
0.5
1
1
8
1
1
10
Times
monitored
per year
Of
52
Of
52
12
12
of
12
12
Of '
Monitoring labor-
hours requiredc
A
0
2.2
0
1.3
36.0
33.6
0
3.6
18.8
0
B
0
8.2
0
4.3
146.0
134.0
0
14.8
75.6
0
C
0
26.0
0
13.4
446.8
414.8
0
46.0
232.4
0
Estimated
number of
leaks per year''
A B
of of
of of
6 22
7 25
Of Of
1 3
4 14
Of O'f
C
of
of
68
75
Of
7
42
Of
Leak repair
Repair Leak repair labor-
time, hours required6
hrs .A B C
80b 0 0 0
80b 000
1.13g 6.8 24.9 76.8
1.139 7.9 28.3 84.8
Of)h 0 0 0
1.139 1.1 3.4 7.9
1.139 4.5 15.8 47.5
40b ' 0 0 0
2 worker's for instrument monitoring, 1 for visual. Ref. 43.
bRef. 44.
"Monitoring labor-hours = number of workers a number of components x time to monitor (total is a minimum of 1 hr)
From Table 4-2.
6
Leak repair labor-hours = number of leaks x repair time.
^No monitoring or leak repair required because equipment specifications eliminate leak potential
Jfffine^reTu^^ 'Kf!^! rePai>ed °n-line' reqU1>in9 °'17 h°Ur P6r repai>' and on 25 Patent of the leaks repaired .
.It is assumed that these leaks are corrected by routine maintenance at no additional labor requirements Ref 46
-------
Leak repair labor is the cost of repairing those components in which
leaks develop after initial repair. The leaks are discovered during the
periodic monitoring required by the regulatory alternatives. The number
of estimated leaks and the labor hours required for repair are given in
Tables 8-3, 8-4, and 8-5. Leak repair labor is calculated based on $15 per
hour1/7,1*8,49 Maintenance labor costs are greatest under Regulatory Alter-
native III and least under Alternative IV. The costs are reduced under
Alternative IV because the required installation of double mechanical
seals with seal oil degassing vents eliminates the most time-consuming
repair items. .
Administrative and support costs are estimated at 40 percent of the
sum of monitoring and leak repair labor costs. Monitoring labor, leak
repair labor, and administrative/support costs are the recurring annual
costs for each Regulatory Alternative.
8.1.2.3 Annualized Costs. The bases for the annual ized control
costs are presented in Table 8-6. The annualized capital, maintenance,
and miscellaneous costs are calculated by taking the 'appropriate factor
from Table 8-6 and applying it to the corresponding capital cost from
Table 8-2. The capital recovery factors were calculated using the
equation:
CRF - "
(1 + i)n - 1
where i = interest rate, expressed as a decimal,
n = economic life of the component, years.
The interest rate used was 10 percent (last quarter 1978). The expected 1
life of the monitoring instrument is 6 years compared to 10 years for other
control equipment components. Double seals and rupture disks are assumed
to have a 2 year life.
The implementation of any of the Regulatory Alternatives (except I)
will result in the initial discovery of leaking components. It is
estimated that fewer leaks will be found at subsequent inspections. The
cost of repairing initial leaks is amortized over a 10-year period, since
this is a one-time cost. Repair of leaks found at subsequent inspections
is included as a recurring annual cost in 8.1.2.2. The estimated
8-8
-------
TABLE 8-6. DERIVATION OF ANNUALIZED LABOR, ADMINISTRATIVE,
MAINTENANCE AND CAPITAL CHARGES
1. Capital recovery factor for capital
charges
* Double seals and rupture disks
' Other control equipment
* Monitoring instruments
2. Annual maintenance charges
* Control equipment
* Monitoring instruments
3. Annual miscellaneous charges
(taxes, insurance, administration)
Control equipment
Monitoring instruments
4. Labor charges
5. Administrative and support costs to
implement regulatory alternative
6. Annualized charge for initial leak
repairs
0.58 x capital .
0.163 x capital'
0.23 x capital0
0.05 x capital
$2700e
0.04 x capitaU
0.04 x capital1"
$15/hour9
0.4 x (monitoring labor +
maintenance labor)n
^(estimated number of leaking
components per model unit1 x
repair time1) x $15/hr9 x 1.4"
x 0.163J
aApplies to cost of seals ($335) and disk ($195) only. Two year life,
ten percent interest.
Ten year life, ten percent interest. From Ref. 50, pp. IV-3,4.
cSix year life, ten percent interest. From Ref. 51, pp. IV-9,10.
dFrom Ref. 52, pp. IV-3,4.
elncludes materials and labor for maintenance and calibration. Cost (last
quarter 1977) from Ref. 53, p. 4-2. Cost index = 221.7 ^ 209.1 (Ref. 54
and 55).
fFrom Ref. 56, pp. IV-3,4,9,10.
Includes wages plus 40 percent for labor-related administrative and
overhead costs. Cost (last quarter 1977) from Ref. 6, pp. 4-4,5. Cost
index = 190.3 ^ 180.9 (Ref. 58 and 59).
hFrom Ref. 60, pp. IV-9,10.
""Shown in Tables 8-18, 8-19, 8-20.
JInitial leak repair amortized for ten years at ten percent interest.
8-9
-------
percentage of initial leaks per component is shown in Table 4-2. This
percentage is applied to the number of components in the model unit
under consideration. Fractions are rounded up to the next integer, since
in practice it is the whole .valve, or seal , that is replaced and not just
part of one. The time required to repair each component type is given
in Table 8-7. The initial repair cost is determined by taking the product
of the number of initial leaks, the repair time, and the labor rate, $15
per hour.' ' ' Forty percent is added for administrative and support
costs. Finally, the total is multiplied by 0.163, the capital recovery
factor. As shown in Table 8-7, the cost of initial leak repair under
Regulatory Alternative IV is substantially less for each of the model
units than under Alternatives II and III. The main reason for this
reduction is the required installation of double mechanical seals and
seal oil degassing vents that reduce the leak potential of pumps and
compressors. Although the total number of pumps and compressors is not
great, the repair time for a single pump or compressor seal is very much
greater than the repair time for a valve.
8.1.2.4 Recovery Credits. The annual VOC emissions, total emission
reductions, and annual recovered product credits for each model unit
under each Regulatory Alternative are shown in Table 8-8. Regulatory
Alternative I represents the uncontrolled emissions from each model unit.
The annual emission reduction is calculated by subtracting the.controlled
emission factor from the uncontrolled emission factor for each source.
To obtain an annual rate, the result is multiplied by 8760 hours per year.
The recovery credit is figured at $360 per Mg of recovered product.
8.1.2.5 Net Annualized Costs. The net annualized costs, shown in
Tables 8-9, 8-10, and 8-11, are determined by subtracting the annual
recovered product credit from the total cost before credit. For example,
Model Unit A, under Regulatory Alternative II has a net annualized credit
of $3300, as a result of $12,100 in costs and $15,400 for recovery
credits.
8.1.2.6 Cost Effectiveness. The cost effectiveness of each regula-
tory alternative for each model unit is shown in Table 8-12. Regulatory
Alternatives II and III have a net annualized credit for all model units,
and cost effectiveness numbers are negative. Since Regulatory Alternative
8-10
-------
TABLE 8-7. LABOR-HOUR REQUIREMENTS FOR INITIAL LEAK REPAIR
oo
i
Regulatory alternative II
Number of
components
per model
unit
Source type
Pumps (light liquid)
Single mechanical seal
Double mechanical seal
Valves (in-line)
Gas
Light liquid
Safety/relief valves3
(gas service)
Valves on open-ended lines
Gas
Light liquid
Compressor seals
A
5
3
90
84
11
9
47
1
B C
19 60
10 31
365 1117
335 1037
42 130
37 115
189 581
2 8
Estimated
number of
initial
leaks b
A
2
1
9
11
0
1
6
1
B C
5 14
3 8
37 112
41 125
0 0
4 12
23 70
1 3
Repair
time,
hrs
80C
80C
1.13d
1.13d
0
1.13d
1.13d
40C
Regulatory alternative III Regulatory alternative IV
Estimated
Labor-hours "^^f
required '»^b
ABC A
160 400 1120 2
80 240 640 1
10 42 127 9
12 46 141 11
0000
1 5 14 1
7 26 79 6
40 40 120 1
B C
5 14
3 8
37 112
41 125
0 0
4 12
23 70
1 3
Repair
time,
hrs
80C
80C
1.13d
1.13d
0
1.13d
1.13d
40C
Estimated
number of
Labor-hours initial.
required leaks
ABC ABC
160 400 1120 Oe Oe Oe
80 240 640 Oe Oe O6
10 42 127 9 37 112
12 46 141 11 41 125
00 0 Oe Oe Oe
1 5 14 1 4 12
7 26 79 6 23 70
40 40 120 Oe Oe Oe
Repair
time,
hrs
80C
80C
1.13d
1.13d
0
1.13d
1.13d
40C
Labor-hours
required
A
0
0
10
12
0
1
7
0
B
0
0
42
46
0
5
26
0
C
0
0
127
141
0
14
79
0
alt is assumed that these leaks are corrected by routine maintenance at no additional labor requirements. Ref.65
Based on the percent of sources leaking at > 10,000 ppm. From Table 4-2.
cRef.66
Weighted average based on 75 percent of the leaks repaired on-line, requiring 0.17 hours per repair, and on 25 percent of the leaks repaired off-line,
requiring 4 hours per repair. Ref. 67.
eNo maintenance required because equipment specification eliminates leak potential.
The estimated number of initial leaks for open-ended valves is based on the same percentage of sources used for in-line valves. This represents leaks
occurring through the stem and gland of the open-ended valve. Leaks through the valve seat are eliminated by adding caps for Regulatory Alternatives
II, III, IV.
-------
TABLE 8-8. RECOVERY CREDITS.
Regulatory
alternative
I
II
III
co
-L iv
VOC
emissions,
Mq/yr
67.2
24.5
20.8
8.46
Model unit A
Emission reduction
from uncontrolled,
Mq/yr
42.7
46.4
58.7
Recovered3
product
value,
$/yr
15,400
16,700
21,100
VOC
emissions,
Mg/yr
257
93.7
79.8
34.3
Model unit B
Emission reduction
from uncontrolled,
Mg/yr
163
177
223
Recovered"1
product
value,
$/yr
--
58,800
63,800
80,200
VOC
emissions,
Mg/yr
800
293
249
106
Model unit C
Emission reduction
from uncontrol led,
Mq/yr
.
507
551
694
Recovered3
product
value,
S/yr
--
182,500
198,400
249,800
Last quarter 1978 dollars. Based on an average price of $36o/Mg. Ref. 68.
-------
TABLE 8-9. ANNUALIZED CONTROL COST ESTIMATES FOR MODEL UNIT A
(thousands of last quarter 1978 dollars).
Cost item . I
Annuali zed capital charges
1. Control equipment
a. Instrument
b. Caps
c. Double seals
Seals
Installation
d. Flush oil system
e. Vents - pumps and compressors
f. Rupture disks
Disks
Holders, etc.
g. Closed loop sampling
2. Initial leak repair
Operating costs
1. Maintenance charges
a. Instrument
b. Caps
c. Double seals
d. Flush oil system
e. Vents - pumps and compressors
f. Rupture disks
g. Closed loop sampling
2. Miscellaneous (taxes, insurance,
administration)
a. Instrument
b. Caps
c. Double seals
d. Flush oil system
e. Vents - pumps and compressors
f. Rupture disks
g. Closed loop sampling
3. Labor
a. Monitoring labor
b. Leak repair labor
c. Administrative and support
Total before credit 0.0
Recovery credits 0.0
Net annual i zed cost 0.0 .
Regulatory
II
1.96
.763
1.06
2.70
.234
.340
.187
0.54
3.17
1.51
12.1
15.4
(-3.3)
alternative
III
1.96
.763
1.06
2.70
.234
.340
.187
2.26
3.30
2.22
15.0
16.7
(-1.7)
IV
1.96
.763
0.974
.196
1.22
5.32
. 1.24
2.74
1.96
0.10
2.70
.234
0.144
.375
1.63
.950
0.60
.340
.187
.115
0.30
1.31
.758
.48
1.43
0.304
0.692
29.0
21.1
7.9
3Based on 40 percent of monitoring plus leak repair labor. Ref.69.
Based on an average price of $360/Mg. Ref.70
c(-xx) =*> net credit
8-13
-------
TABLE 8-10. ANNUALIZED CONTROL COST ESTIMATES FOR MODEL UNIT B
(thousands of last quarter 1978 dollars)
Cost item
Annual ized capital charges
1. Control equipment
a. Instrument
b. Caps
c. Double seals
Seals
Installation
d. Flush oil system
e. Vents - pumps and compressors
f. Rupture disks
Disks
Holders, etc.
g. Closed loop sampling
2. Initial leak repair
Operating costs
1. Maintenance charges
a. Instrument
b. Caps
c. Double seals
d. Flush oil system
e. Vents - pumps and compressors
f. Rupture disks
g. Closed loop sampling
2. Miscellaneous (taxes, insurance,
administration)
a. Instrument
b. Caps
c. Double seals
d. Flush oil system
e. Vents - pumps and compressors
f. Rupture disks
g. Closed loop sampling
3. Labor
a. Monitoring labor
b. Leak repair labor
c. Administrative and support
Total before credit
Recovery credits
Net annual ized cost0
Regulatory
I II
1.96
3.05
2.73
2.7
.935
0.34
.748
2.04
3.54 .
2.23
0.0 20.3
0.0 58.8
0.0 (-38.5)
alternative
III
1.96
3.05
2.73
2.7
.935
0.34
.748
8.75
7.69
6.58
35.5
63.8
(-28.3)
IV
1.96
3.05
3.69
.743
4.65
17.6
4.75
10.5
7.79
0.41
2.7
.935
.546
1.42
5.39
3.63
2.39
0.34
.748
.437
1.14
4.31
2.90
1.91
5.74
1.09
2.73
93.5
80.2
13.3
aSee footnote from preceeding Table 8-9, Ref. 71.
Based on an average
c(-xx)=> net credit
Based on an average price of $360/Mg. Ref.72.
8-14
-------
TABLE 8-11. ANNUALIZED CONTROL COST ESTIMATES FOR MODEL UNIT C
(thousands of last quarter 1978 dollars)
Cost item
Annuali zed capital charges
1. Control equipment
a. Instrument
b. Caps
c. Double seals
Seals
Installation
d. Flush oil system
e. Vents - pumps and compressors
f. Rupture disks
Disks
Holders, etc.
g. Closed loop sampling
2. Initial leak repair
Operating Costs
1. Maintenance charges
a. Instrument
b. Caps
c. Double seals
d. Flush oil system
e. Vents - pumps and compressors
f. Rupture disks
g. Closed loop sampling
2. Miscellaneous (taxes, insurance,
administration)
a. Instrument
b. Caps
c. Double seals
d. Flush oil system
e. Vents - pumps and compressors
f. Rupture disks
g. Closed loop sampling
3. Labor
a. Monitoring labor
b. Leak repair labor
c. Administrative and support
Total before credit
Recovery credits
Net annual uzed cost0
Regulatory
I II
1.96
9.37
7.67
2.70
2.88
0.340
2.30
6.33
8.71
6.02
0.0 48.3
0.0 182.
0.0 (-134.)
alternative
III
1.96
9.37
7.67
2.70
2.88
0.340
2.30
27.14
21.3
19.4
95.1
198.
(-103.)
IV
1.96
9.37
11.7
2.35
14.7
56.9
14.7
32.4
24.0
1.23
2.70
2.88
1.72
4.50
17.5
11.2
7.35
0.340
2.30
1.38
3.60
14.0
8.98
5.88
17.7
3.25
8.38
283.
250.
33.0
aBased on 40 percent of monitoring plus leak repair labor. Ref.73
Based on an average price of $360/Mg. Ref. 74.
c(-xx) =i»-net credit.
8-15
-------
TABLE 8-12. COST EFFECTIVENESS FOR MODEL UNITS
(last quarter 1978 dollars)
Model unit Aa
Regulatory alternative
Total capital cost ($1000)
Total annualized cost ($1000)
Total annual recovery credit ($1000)
oo . Net annualized cost ($1000)d
cr>
Total VOC reduction (Mg/yr)
Cost effectiveness ,
(annual $/Mg VOCr
I II
0.0 13.2
.0.0 12.1
0.0 15.4
0.0 (-3.3)
0.0 42.7
- (-77.3)
III
13.2
15.0
16.7
(-1.7)
46.4
(-36.6)
IV
87.1
.29.0
21.1
7.9
58.7
135.
Model unit Bb
I II
0.0 27.2
0.0 20.3
0.0 58.8
0.0 (-38.5)
0.0 163
- (-236.)
Ill
27.2
35.5
63.8
(-38.3)
177
(-160.)
IV
295
93.5
80.2
13.3
223
59.6
Model
I II
0.0 66.0
0.0 48.3
0.0 182.
0.0 (-134.)
0.0 507
- (-264.)
unit Cc
III
66.0
95.1
198.
(-103.)
551
(-187.)
IV
911
283
250.
33.0
694 .
47.6
52 percent of the units in the SOCMI are similar to Model Unit A. Ref. 75.
b33 percent of the units in the SOCMI are similar to Model Unit B. Ref. 76.
C15 percent of the units in the SOCMI are similar to Model Unit C. Ref. 78.
. (-xx) = Control method net credit
-------
IV is the only one with a positive net cost, comparisons of cost
effectiveness in the normal sense are meaningless. The highest cost of
VOC control under Regulatory Alternative IV is for model unit A. Although
this cost ($128/Mg) is much larger than the cost for model unit C ($40/Mg),
the total annualized cost for model unit A is only $7500. This amount is
insignificant compared to the annual operating cost of the process unit
itself.
8.1.3 Modified/Reconstructed Facilities
8.1.3.1 Capital Costs. The bases for determining the capital costs
for modified/reconstructed facilities are presented in Table 8-1. The
capital costs for these units are the same under Regulatory Alternatives
II and III as are those for new units. There are no costs associated
with Alternative I. The capital costs for the monitoring instruments,
the caps for open-ended lines, the flush-oil systems, the vents for
degassing reservoirs, and the closed loop sampling connections are also
the same as for new units.
The estimated cost of retrofitting double mechanical seals for
single seal pumps is estimated at $850 per pump. This figure includes
$560 for a new double mechanical seal plus $290 labor for field installa-
tion.
Rupture disks for relief valves, required under Regulatory Alterna-
tive IV, are estimated to cost $2970 per relief valve. The original
relief valve must be replaced with a larger relief valve. Credit for
the removed valve is not included.
The total capital cost estimates for modified/reconstructed facilities
are presented in Table 8-13. As noted above, the costs associated with
Regulatory Alternatives I, II, and III are the same as for new units.
8.1.3.2 Annualized Costs. The annualized control costs for
modified/reconstructed units, presented in Table 8-14, are derived from
the same basis as new units (see Table 8-2). The only changes from new
unit costs occur under Regulatory Alternative IV because of the increased
capital costs for double mechanical seals and rupture disks. The
recovered product credits for the modified/reconstructed units are the
same as for the new model units.
8-17
-------
TABLE 8-13. CAPITAL COST ESTIMATES FOR MODIFIED/
RECONSTRUCTED FACILITIES
(thousands of last quarter 1978 dollars)
1.
2.
3.
4.
5.
6.
Capital cost item3
Monitoring instrument
Caps for open-ended lines
Double mechanical seals
. Seals
Installation
Flush oil systems for double
mechanical seals
Vents for compressor degassing
reservoirs
Vents for pump degassing
Regul
A
8.5
4.68
2.8
1.45
7.50
6.53
26.1
atory alternative
Model unit
B
8.5
18.7
10.6
5.51
28.5
13.1
94.7
IVb
C
8.5
57.5
33.6
17.4
90.0
52.2
297
reservoirs
7. Rupture disks for relief
valves
. Disks 2.14 8.19 25.4
Holders, block valves, 16.8 64.4 199
installation : .
Replacement relief valve and 15.2 58.0 179
installation
8. Closed loop sampling connections 12.0 . 47.8 147
Total 104 358 1107
aFrom Tables 6-1 and 8-1.
bFor Regulatory Alternatives I, II, III the capital costs for modified/
reconstructed facilities are the same as for new units (Table 8-2).
8-18
-------
TABLE 8-14. ANNUALIZED CONTROL COST ESTIMATES FOR MODIFIED/
RECONSTRUCTED MODFL UNITS UNDER REGULATORY ALTERNATIVE IVa
(thousands of last quarter 1978 dollars)
Model h Model Model .
Cost item unit A unit B unit C
Annualized capital charges
1. Control equipment
a. Instrument 1.96 1.96 1.96
b. Caps .763 3.05 9.37
c. Double seals
Seals 1.62 6.15 19.5
Installation .236 0.898 2.84
d. Flush oil system 1.22 4.65 14.7
e. Vents for pumps and compressors 5.32 17.6 56.9
f. Rupture disks
Disks 1.24 4.75 14.7
Holders, etc. 2.74 10.5 32.4
Relief valves 2.48 9.45 29.2
g. Closed loop sampling 1.96 7.79 24.0
2. In.itial leak repair 0.10 0.41 1.23
Operating costs
1. Maintenance charges
a. Instrument
b. Caps
c. Double seals
d. Flush oil system
e. Vents for pumps and compressors
f. Rupture disks
g. Closed loop sampling
2. Miscellaneous (taxes, insurance,
administration)
a. Instrument
b. Caps
c. Double seals
d. Flush oil system
e. Vents for pumps and compressors
f. Rupture disks
g. Closed loop sampling
3. Labor
a. Monitoring labor
b. Leak repair labor
c. Administrative and support6
Total before credit
Recovery credits
Net annual i zed cost
Total VOC reduction (Mg/yr)
Cost effectiveness ($/Mg VOC)
2.70
.234
0.213
.375
1.63
1.71
.60
.340
.187
0.170
0.30
1.31
1.37
.480
1.43
0.304
0.692
33.7
21.1
12.6
58.7
215.
2.70
.935
0.806
1.42
5.39
6.53
2.39
.340
.748
.644
1.14
4.31
5.22
1.91
5.74
1.09
2.73
111.
80.2
30.8
223.
138.
2.70
2.88
2.55
4.50
17.5
20.2
7.35
.340
2.30
2.04
3.60
14.0
16.2
5.88
17.7
3.25
8.38
338.
250.
88.
694
127.
aFor Regulatory Alternatives I, II, III, the annualized control costs and cost
effectiveness for modified/reconstructed facilities are the same as for new units
(Tables 8-7, 8-8, 8-9).
52 percent of existing units are similar to Model Unit A. Ref. 78.
C33 percent of existing units are similar to Model Unit B. Ref. 79.
15 percent of existing units are similar to Model Unit C. Ref. 80.
eBased on 40 percent of monitoring plus leak repair labor. Ref. 81.
Based on an average price of $360/Mg. Ref. 82.
8-19
-------
8.1.3.3 Cost Effectiveness. The cost effectiveness figures for
modified/reconstructed facilities are also shown in Table 8-14. The cost
effectiveness under Regulatory Alternatives I, II, and III is the same as
for the new model units. The cost effectiveness under Regulatory Alter-
native IV is a net cost of $134 per Mg for Model Units B and C, and $208
per Mg for Model Unit A.
8.1.4 Projected Cost Impacts
The regulatory alternatives are assumed to go into effect by 1981,
using 1980 as the base year.. The industry is estimated to grow at. a rate
of 5.9 percent.83 SOCMI facilities are estimated to be replaced at a rate
based on a 20-year equipment life (see Appendix E). The estimated numbers
of projected new units are presented in Tables 7-5, 7-6, and 7-7. The
estimated costs to the industry for the years 1981 through 1985 are
presented in Tables 8-15 through 8-17. Capital costs shown are only for
units which begin operation in the indicated year. All other costs shown
are for all units subject to NSPS in the indicated year.
8-20
-------
PO
TABLE 8-15. NATIONWIDE COSTS FOR THE INDUSTRY UNDER REGULATORY ALTERNATIVE II
(last quarter 1978 dollars)
Cost item9
Total capital cost ($1000)b
Total annualized cost ($1000)
Total annual recovery credit
Net annualized cost ($1000)
1981
3,800
c 2,990
($1000) 8,070
(-5,080)
1982
4,040
6,160
16,700
(-10,500)
1983
4,280
9,530
25,800
'(-16,300)
1984
4,490
13,100
35,300
(-22,200)
1985
4,790
16,800
45,500
(-28,700)
^aFrom Tables 8-2, 8-9, 8-10, 8-11.
Capital costs for model units which begin operation in the years shown.
GAnnualized costs for all model units subject to NSPS in the years shown.
(-xx) => net credit
-------
ro
ro
TABLE 8-16. NATIONWIDE COSTS FOR THE INDUSTRY UNDER REGULATORY ALTERNATIVE III
(last quarter 1978 dollars)
Cost item3
Total capital cost ($1000)b
Total annual ized cost ($1000)
Total annual recovery credit
Net annualized cost ($1000)
1981
3,800
C 4,990
($1000) 8,770
(-3,780)
1982
4,040
10,300
18,100
(-7,800)
1983
4,280
15,900
28,000
(-12,100) .
. 1984
4,490
21,800
38,400
(-16,600)
1985
4,790
23,100
49,400
(-21 ,300)
aFrom Tables 7-6, 8-7, 8-8, 8-9.
Capital costs for model units which begin operation in the years shown.
cAnnualized costs for all model units subject to NSPS in the years shown.
(-xx) => net credit
-------
TABLE 8-17. NATIONWIDE COSTS FOR THE INDUSTRY UNDER REGULATORY ALTERNATIVE IV
(last quarter 1978 dollars)
ro
oo
'Cost item3
Total capital cost ($1000)b
Total annual ized cost ($1000)°
Total annual recovery credit
($1000)
Net annual ized cost ($1000)d
1981
41,200
13,000
11,100
1,900
1982
44,000
27,000
22,900 .
4,100
1983
46,500
41,700
35,400
6,300
1984
48,700
57,100
48,500
8,600
1985
51,900
73,500
62,400
11,100
Trom'Tables 7-6, 8-7, 8-8, 8-9.
Capital costs for model units which begin operation in the years shown.
'Annualized costs for all model units subject to NSPS in the years shown.
(-xx)=*> net credit
-------
8.2 OTHER COST.CONSIDERATIONS
Environmental Safety and Health Statues which are applicable to
SOCMI plants are listed in Table 8-18. The provisions, requirements, and
regulations, listed are those which may cause an outlay of funds by an
organic chemical manufacturer.
Specific costs of each of these provisions or requirements to the
industry .defined as SOCMI were unavailable. Total costs to SOCMI for
complying with environmental, safety and health standards were also
unavai Table.
The entire chemical industry is planning to spend an estimated $639
million on pollution control in 1979 according to a McGraw-Hill Survey.8tf
Although this is a sizeable sum of money, the industry has enjoyed three .
decades of rapid growth and high profits. The economic health of the
industry is better than that of many other industries.85 The substantial
pollution problems encountered in the industry and the large expenditures
necessary for this solution are expected to affect the smaller firms more
adversely than the larger firms. However, few plant closings are expected
due solely to costs of compliance with standards and regulations.86
The costs incurred by SOCMI in complying with all health, safety,
and environmental requirements are not expected to prevent compliance
with the proposed NSPS for fugitive emissions.
8-24
-------
TABLE 8-18. STATUTES THAT MAY BE APPLICABLE TO SOCMI
Statute
Applicable provision, regulation or
requirement of statute
Approximate cost incurred due to
enactment of statute
Model unit
Industry
Clean Air Act and Amendments
State implementation plans
National emission standards for hazardous
air pollutants
Benzene fugitive emissions
New source performance standards
Air oxidation
Volatile organic liquid storage
PSD construction permits
Non-attainment construction permits
Total
$249 million3
Clean Water Act (Federal
Water Pollution Act)
oo
ro
en
Discharge permits
Effluent limitations guidelines
New source performance standards
Control of oil spills and discharges
Pretreatment requirements
Monitoring and reporting
Permitting of industrial projects that
impinge on wetlands or public waters
Environmental impact statements
Total
$414 million0
Resource Conservation and
Recovery Act
Permits for treatment, storage, and
disposal of hazardous wastes
Establishes system to track hazardous
wastes
Establishes recordkeeping, reporting,
labelling and monitoring system for
hazardous wastes
Superfund
Total
$200 million0
Superfund-less than 2* of profits
or $200 million maximum annual
rate on petrochemical
feedstock^
Production costs for the industry
are expected to increase by an
average of 0.6%'and a maximum
of 5%.e
Toxic Substances Control
Act
Premanufacture notification
Labelling, recordkeeping
Reporting requirements
Toxicity testing
Total
$100-200 million per year
Preinventory notification cost:
$1200-1500 per chemical
(Continued)
-------
TABLE 8-18. (Cont.
Statute
Applicable provision, regulation or
requirement of statute
Approximate cost incurred due to
enactment of statute
Model unit
Industry
Occupational Safety and Health
Act
Walking-working surface standards
Means of egress standards
Occupational health and environmental
control standards
Hazardous material standards
Personal protective equipment standards.
General environmental control standards
Medical and first aid standards
Fire protection standards
Compressed gas and compressed air
equipment
Welding, brazing, and cutting standards
Total
S220/year per worker
CO
i
rv>
cr>
Coastal Zone Management Act
Power Plant and Industrial
Fuel Use Act
States may veto federal permits for
plants to be sited in coastal zone
Prohibits new, major, industrial power-
plants which utilize fuel oil or
natural gas
National Environmental Policy
Act
Requires environmental impact statements
Safe Drinking Water Act
Requires underground injection control
permits
Marine Sanctuary Act
Ocean pumping permits
Recordkeeping and reporting
Expenditure, by entire chemical industry, on air pollution control;. SOCMI's portion of expenditure not delineated. (Ref.87 )
Expenditure, by entire chemical industry, on water pollution control; SOCMI's portion of expenditure not delineated. (Ref.83
cCost reflects entire organic industry; SOCMI's cost not delineated. (Ref.89 ) . '
Cost reflects entire organic industry; SOCMI's cost not delineated. (Ref.90,91)
eCost reflects entire organic industry; SOCHI's cost not delineated. (Ref. 92) . .
Cost incurred by entire chemical industry; SOCMI's portion, of expenditure not delineated.. (Ref. 93)
Cost incurred by entire chemical industry; SOCMI's portion of expenditure not delineated. (Ref. 94)
nCost incurred by entire chemical industry; SOCMI's portion of expendItuve not delineated. (Ref. ?3)
-------
8.3 REFERENCES
1. Erikson, D. G., and V. Kalcevic. Emissions Control Options for
the Synthetic Organic Chemical Manufacturing Industry, Fugitive
Emissions Report. Draft Report. EPA Contract No. 68-02-2577.
Knoxville, Tennessee, Hydroscience, Inc., March 1979. p. IV-9.
2. Letter from Guy C. Amey, Century Systems Corporation, to James C.
Seme, PES, Inc. October 17, 1979. Cost data for VOC monitoring
instrument.
3. Peters, Max S., and K. D. Timmerhaus. Plant design and Economics for
Chemical Engineers. Second Edition. New York, McGraw-Hill. 1968.
4. Kohn, P. M. CE Cost Indexes Maintain 13-Year Ascent. Chem. Eng.
18(11):189-190. May 1978.
5. Economic Indicators. Chem. Eng. Vol. 87 #1. January 14, 1980.
6. Ref. 5.
7. Environmental Protection Agency, Chemical and Petroleum Branch.
OAQPS Guideline Series. Control of Volatile Organic Compound Leaks
from Petroleum Refinery Equipment. EPA-450/2-78-036, OAQPS
No. 1.2-111. June 1978. p. 4-5.
8. Economic Indicators. Chem. Eng. Vol 86 #2. January 15, 1979.
9. Ref. 1, p. IV-3.
10. Ref. 1, p. IV-3.
11. Ref. 1, p. IV-3.
12. Ref. 1, pp. IV-8, 9.
13. Ref. 1, p. IV-8.
14. Ref. 3, p. 451.
15. Ref. 4.
16. Ref. 5.
17. Ref. 5.
18. Ref. 7.
8-27
-------
19. Ref. 8.
20. Ref. 3, p. 450.
21. Ref. 4.
22. Ref. 5.
23. Ref. 5.
24. Ref. 7.
25. Ref. 8.
26. Ref. 4.
27. Ref. 5.
28. Ref. 5.
29. Ref. 7.
30. Ref. 8.
31. Ref. 1, p. IV-8.
32. Ref. 5.
33. Ref. 7.
34. Ref. 8.
35. Ref. 7, p. 4-3.
36. Letter with Attachments from J. M. Johnson, Exxon Company, U.S.A.,
to Robert T. Walsh, U. S. EPA. July 28, 1977.
37. Ref. 1, p. B-12.
38. Ref. 36.
39. Ref. 7, p. 4-3.
40. Ref. 36.
41. Ref. 1, p. B-12.
42. Ref. 36.
8-28
-------
43. Ref. 7, p. 4-3.
44. Ref. 36.
45. Ref. 1, p. B-12.
46. Ref. 36.
47'. Ref. 5.
48. Ref. 7.
49. Ref. 8.
50. Ref. 1, pp. IV-3, 4.
51. Ref. 1, pp. IV-9, 10.
52. Ref. 1, pp. IV-3, 4.
53. Ref. 7, p. 4-2.
54. Ref. 5.
55. Ref. 8.
56. Ref. 1, pp. IV-3, 4, 9, 10.
57. Ref. 7, pp. 4-4. 5.
58. Ref. 5.
59. Ref. 8.
60. Ref. 1, pp. IV-9, 10.
61. Ref. 5.
62. Ref. 7.
63. Ref. 8.
64. Letter from Vincent Smith, Research Triangle Institute to Russell
L. Honerkamp, Radian Corporation. November 30, 1979. Projected
Number of Affected Facilities and Average Product Value in SOCMI.
65. Ref. 36.
66. Ref. 36.
8-29
-------
67. Ref. 1, p. B-12.
68. .Ref. 64.
69. Ref. 1, pp. IV-9, 10.
70. Ref. 64.
71. Ref. 1, pp. IV-9, 10.
72. Ref. 64.
73. Ref. 1, pp. IV-9, 10.
74. Ref. 64.
75. Ref. 1, p. IV-1.
76. Ref. 1, p. IV-1.
77. Ref. 1, p. IV-1.
78. Ref. 1, p. IV-1.
79. Ref. 1, p. IV-1. .
80. Ref. 1, p. IV-1.
81. Ref. 1, pp. IV-9, 10.
82. Ref. 64.
83. Letter from Vincent Smith, Research Triangle Institute, to Russell
L. Honerkamp, Radian Corporation. August 13, 1979. Growth Rate of
SOCMI.
84. News Flashes. Chemical Engineering, Vol. 86,. No. 12. 1979. p. 77.
85. Environmental Quality, The Ninth Annual^Report of the Council on
Environmental Quality. U.S. Government Printing Office, Washington,
D.C. December 1978.
86. Ref. 85.
87. Ref. 85.
88. Ref. 85.
89. Solid Waste Facts, A Statistical Handbook. U. S. Environmental
Protection Agency, Office of Public Awareness. U. S. Government
Printing Office, Washington, D. C. August 1978.
8-30
-------
90. EPA Charges Chemical Trade Seeks Lowest Denominator.as its Position
on Superfund. Chemical Marketing Reporter. N.Y. (216) 10. McGraw-
Hill, Sept. 3, 1979. p.3.
91. Tough Version of Superfund Would Cook Industry, $1.6 Billion for
Cleanup. Chemical Marketing Reporter. N.Y. (215) (25).
McGraw-Hill, Jan. 18, 1979.
92. Ref. 85.
93. Preproposal of Premanufacture Notification Notice Form and Provision
of Rules 40 CFR Part 720. 44(201) Oct. 16, 1979.
94. Ref. 93.
95. Cost of Government Regulation Study. Arthur Anderson and Co.,
Washington, D.C. March 1979.
8-31
-------
9. ECONOMIC ANALYSIS
9.1 Industry Profile
9.1.1 Introduction
The synthetic organic chemicals manufacturing industry (SOCMI) has
been defined as the producers of 378 synthetic organic chemicals, listed in
Appendix F. This profile gives a general qualitative description of the in-
dustry, supported by quantitative information wherever possible. Because
SOCMI does not directly correspond to industrial classifications used for re-
porting information by secondary data sources, a weighting technique was used
to develop industry statistics (see Appendix El).
Synthetic organic chemicals (SOCs) are substances containing at least
carbon and hydrogen. They exhibit three basic molecular structures: ali-
phatic or acyclic, cyclic, and combinations of aliphatic and cyclic. Acyclic
compounds are composed groups of atoms arranged in a straight chain. Ex-
amples are alcohols, ethers, ketones, and carbohydrates. Cyclic compounds
have the atoms of their component elements arranged in the form a closed
ring. Examples include aromatic hydrocarbons, napthenes, and thiazoles.
Certain amino acids and terpene hydrocarbons represent combinations of cyclic
and aliphatic molecular structures.1
SOCMI chemicals may be used as primary feedstocks, chemical intermedi-
ates, or end use chemicals. Primary feedstocks are produced from crude raw
materials and used in the manufacture of other chemicals. Chemical interme-
diates are the product of primary feedstocks and are also used to produce
other chemicals. End use chemicals are products of chemical intermediates
and/or primary feedstocks and are used either as final goods or as inputs to
production processes outside the chemical industry. Many synthetic organic
chemicals are used in more than one of these categories. Figure 3.1 illu-
strates the general relationships among the various organic chemicals.
9-1
-------
Detailed flow charts identifying inputs and product uses for many of the SOCMI
chemicals have been presented elsewhere.2
9.1.2 Production Processes and Capacities
Most of the SOCMI chemicals produced in the United States are derived
from crude petroleum and natural gas. Oil, shale, coal, and biomass (non-
prehistoric plant tissue) are also sources of primary feedstocks.3 A wide
variety of processes are used to manufacture the 378 synthetic organic chem-
icals included in the definition of SOCMI. Frequently individual chemicals
can be manufactured in several different ways. Consequently, as relative
prices change, chemical producers may alter the mix of primary feedstocks
used to produce SOCs.
After chemical feedstocks are manufactured from petroleum, natural gas,
and other raw materials, they are processed into chemical intermediates and
end use chemicals. Some of the chemicals included in SOCMI are the product
of a simple distillation process, while others are produced from a series of
cracking processes.
The 1976 Organic Chemical Producers Data Base4 reports 1,270 units pro-
ducing SOCMI chemicals in the United States.* Table 9-1 presents a distribu-
tion of those units and estimated capacity by state. New Jersey, Texas, and
California have the largest number of units producing SOCMI chemicals. Texas
and Louisiana have the largest total production capacities. These states are
major producers because of their petroleum deposits and good sea port facili-
ties. Table 9-2 presents a geographical distribution of units by reported
capacity. Approximately 12 percent of these units produce fewer than 5,000
Mg. Another 12 percent of the units have production capacities in excess of
*The 1976 version of the Organic Chemical Producers Data Base is used because
it was the most recent version available.
9-2
-------
TABLE 9-1. ESTIMATED ANNUAL PRODUCTION CAPACITY
BY STATE, 19764
State
Total
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Hawai i
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New York
North Carolina
Ohio
Oklahoma
Oregon
Pennsylvania
Puerto Rico
Rhode Island
South Carolina
Tennessee
Texas
Utah
Vermont
Virgin Islands
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Number
of units
1,270
25
2
1
12
120
5
18
14
14
20
1
2
85
31
11
1
27
54
1
17
27
28
6
15
16
1
4
2
5
131
52
50
88
9
17
75
13
8
27
24
126
4
1
2
22;
13
24
18
1
Percentage of
units reporting
capacity
40
52
50
00
33
' 23
20
26
50
43
30
- 00
50
31
26
64
00
56
74
00
35
48
25
17
47
37
100
100
50
40
24
19
38
26
78
53
27
85
00
48
46
76
25
100
100
68
46
63
22
100
Estimated total .
capacity,
(103 Mg)
319,835
5,174
399
91
1,982
19,650
644
2,765
2,031
3,257
3,459
91
97
16,517
3,551
1,698
390
6,062
31,810
390
2,160
4,835
9,735
574
1,999
4,072
222
103
122
483
28,070
10,586
7,283
14,576
702
3,838
14,634
7,259
815
3,875
6,809
77,189
628
2
643
3,581
2,502
9,242
3,514
24
Capacities were estimated by calculating the mean of reported unit capacity
for each chemical. This was substituted for any missing values of unit
capacity for each chemical. If no units reported capacity for a chemical,
then the mean of all chemicals was substituted for the missing value. Esti-
mated capacity represents the sum of reported-capacities, means of reported
capacity for some specific chemicals, and industry mean reported capacity
for other chemicals.
9-3
-------
TABLE 9-2. DISTRIBUTION OF UNITS BY UNIT CAPACITY AND REGION, 19764
Number of
units
Unit capacity ranges (103 Mg)
Region
North east
New England
Mid-Atlantic
North central
East
West
South
East south
central
West south
central
South
Atlantic
West
Mountain
Pacific
0-5
17
7
10
16
13
3
23
2
4
17
3
0
3
5-10
5
0
5
5
3
2
7
3
3
1
2
1
1
10-25
12
4
8
13
10
3
34
12
10
12
17
3
14
25-50
16
5
11
13
9
4
38
8
18
12
12
2
10
50-100
17
4
13
13
9
4
44
5
20
19
5
0
5
100-250
10
0
10
17
16
1
50
9
24
17
8
0
8
250-500
5
1
4
6
6
0
34
3
24
7
3
0
3
500+
1
0
1
3
2
1
50
4
44
2
0
0
0
Units
reporting
capacity
83
21
62
86
68
18
280
46
147
87
50
6
44
Units not
reporting
capacity
235
39
196
202
182
20
200
45
54
101
119
10
109
Total
units
318
60
258
288
250
38
480
91
201
188
169
16
153
Total
59
19
76
79
79
85
48
54
499
756
1,255
-------
500,000 Mg. Seventy-five percent of these large facilities are located in
the southwest central region of the United States, which includes Texas and
Louisiana. Table 9-3 presents the total reported capacity for each region
by unit size.
9.1.3 Production and Sales
Production and sales data for the SOCMI are presented in Table 9-4. The
production of SOCMI chemicals increased from 58,050 Gg in 1968 to 84,530 Gg
in 1978, at an average annual growth rate of approximately 3.5 percent. How-
ever, output levels have fluctuated widely since 1974. The effects of the
oil embargo, the increase in energy and feedstock prices, and the sharply re-
duced demand resulting from a major economic recession caused the industry to
cut back production by 13.2 percent in 1975. In 1976 output rose only
slightly, but, in 1977, as real prices for energy and feedstocks fell, the
economy recovered, and the need to increase inventories became urgent, pro-
duction increased by 50.4 percent. In 1978 energy and feedstock prices began
to increase again and the need to replenish inventories disappeared. Output
declined that year by 28.6 percent. Nevertheless, production in 1978 was
greater than in 1974, suggesting that the industry may have substantially
adjusted to the shocks experienced in 1974 and 1975.
Sales and production trends were virtually identical over the period
1967-1978.* The two variables are likely to remain highly correlated in the
future, because the industry's feedstock requirements are closely tied to its
production levels. The absolute level of sales was much lower than the level
of production (45.6 percent of production) over the period 1967 to 1978. The
*The estimated correlation coefficient for the two variables over this period
is 0.97.
9-5
-------
TABLE 9-3. DISTRIBUTION OF INDUSTRY CAPACITY BY UNIT CAPACITY AND REGION, 19764
Industry capacity
Unit capacity
Region
North east
New England
Mid-Atlantic
North central
East
West
South
DEast south
^ central
West south
central
South Atlantic
West
Mountain
Pacific
Caribbean
Total
0-5
37.7
19.1
18.6
42.2
38.1
4.1
62.9
1.3
12.2
49.4
5.0
5.0
--
147.8
5-10
43.1
43.1
38.8
22.5
16.3
58.8
23.6
28.8
6.4
13.2
6.4
6.8
--
153.9
10-25
199.2
70.8
128.4
205.9
142.4
63.5
605.6
204.6
194.6
206.4
298.4
48.5
249.9
1,309.1
25-50
548.4
176.9
371.5
478.1
350.2
127.9
1,442.0
299.4
710.8
431.8
477.2
72.6
404.6
109.8
3,055.5
50-100
1,177.6
263.1
914.5
953.0
664.1
288.9
3,236.1
316.2
1,532.3
1,387.6
342.5
342.5
72.6
5,781.8
(103 Mg)
100-250
1,392.6
1,392.6
2,544.7
2,406.8
137.9
8,369.
1,494.6
4,411.7
2,463.5
1,055.1
1,055.1
--
13,362.2
150-500
1,811.2
299.4
1,511.8
2,024.9
2,024.9
11,910.9
984.5
8,596.6
2,329.8
1,024.2
1,024.2
1,324.1
18,095.3
^500
742.5
742.5
2,430.8
1,906.9
523.9
67,600.5
3,132.6
63,197.8
1,270.1
--
5,659.1
76,432.9
Total
5,952
829
5,123
8,718
7,556
1,163
93,287
6,457
78,685
8,145
3,216
127
3,088
7,166
118,339
-------
TABLE 9-4. ANNUAL PRODUCTION AND SALES OF SYNTHETIC ORGANIC CHEMICALS'
Year
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
Production3
(Gg)
51,380
58,050
65,210
68,140
69,020
76,740
81,220
83,720
72,660
76,030
114,320
84,530
Sales volume3
(Gg)
23,440
26,960
30,360
32,090
33,020
36,930
39,420
38,450
32,920
32,520
49,470
35,310
Sales value3
($106)
3,085.55
3,411.91
3,590.07
3,702.20
3,724.03
4,173.97
4,991.53
9,357.99
8,411.34
10,187.76
15,317.72
12,951.16
Average
unit value3
($Ag)
0.13
0.13
0.11
0.12
0.11
0.11
0.13
0.24
0.26
0.29
0.31
0.36
3See Appendix El for a discussion of the methodology used to compute these
data.
9-7
-------
difference between output and sales represents captive consumption, indicat-
ing that the industry has a relatively high degree of vertical integration.
9.1.4 Resource Use
Estimates of employment, assets, cost of materials and energy used in
SOCMI from 1972 to 1976 are presented in Table 9-5. In general, resource use
increased with production. Total industry employment, including administra-
tive, clerical, marketing and service employees as well as production work-
ers, increased 5 percent from 1972 to 1976. Employment of production workers
increased 4.1 percent during this period, although the number of production
workers declined during the adjustment period following the 1973-74 oil em-
bargo. Cost of materials increased substantially during the 1972-1976 peri-
od; however, much of this increase can be attributed to rapidly increasing
prices of crude petroleum. Consumption of energy for heat and power has
fluctuated, resulting in an overall decrease of about 1 percent from 1972 to
1976. Value of assets increased each year from 1972 to 1976. The total in-
crease during that period was approximately 53 percent, much of which can be
accounted for by changes in the value of buildings and equipment. The stock
of physical assets increased at a much slower rate.
9.1.5 Consumption
The chemicals in SOCMI have a wide variety of end uses as fuels, sol-
vents, pesticides, and pigments, and as feedstocks for the production of
plastics, synthetic fibers and textiles, soaps and detergents, rubber pro-
ducts, medicines and fertilizers. It is not possible to estimate consistent-
ly apparent consumption, because import and export data presented in Table
9-6 for SOCMI are not compatible with the production and sales data presented
in Table 9-4. However, it is probable that historical consumption trends have
9-8
-------
TABLE 9-5. SOCMI RESOURCE USE6
Year
1972
1973
1974
1975
1976
Total
employment
. do3)
130.6
132.3
130.1
132.7
137.1
Production
workers
(103)
83.2
85.1
84.0
82.7
86.6
Total
assets
($106)
12,287.8
13,048.3
13,919.5
16,198.2
18,788.3
Cost of
materials
($106)
5,338.5
6,311.8
10,388
11,569
14,503.1
Energy
purchased
for heat
and power
(109 joules)
l,220.1a
l,286.6a
l,322.7a
l,154.4a
1,202.4
These data were estimated by multiplying the 1976 estimate of energy use by
the ratio of production in each of the previous years to 1976 production
levels. Thus, for example, energy use in 1972 was estimated by multiplying
energy use in 1976 by the ratio of production in 1972 to production in 1976.
9-9
-------
TABLE 9-6. INDUSTRIAL ORGANIC CHEMICALS:
U.S. IMPORTS AND EXPORTS, 1966-779
Year
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
Imports
($106)
48
48
67
84
91
129
150
169
259
205
294
326
Exports3
($106)
211
231
292
290
336
304
320
484
930
779
1,008
995
Includes exports of some finished products. Figures include estimates
and are not strictly comparable with imports or production.
9-10
-------
been similar to historical production and sales trends. Certainly, over the
period 1967-1978, consumption increased, although since 1974, if the sales
data presented in Table 9-4 can be regarded as an indicator of consumption,
consumption exhibited wide year-to-year variations for the reasons discussed
in Section 9.1.3.
9.1.6 Prices
The general level of prices for SOCMI chemicals more than tripled be-
tween 1967 and 1978. Most of the increase occurred after 1973. From 1967 to
1973, the average unit price of SOCMI chemicals remained close to $0.12/kg.
Following the 1973-1976 adjustments in oil prices, average prices in SOCMI
doubled, rising to $0.24/kg. After that time average unit prices increased
at a rate of approximately 11 percent annually, to a price of $0.36/kg in
1978. It is important to realize that these are average prices per unit of
all SOCMI chemicals. In 1976, prices for individual chemicals ranged from
$0.11/kg for formaldehyde to $4.30/kg for benzophenone. Changes in the unit
price for individual chemicals may vary substantially from the changes in
average prices. Table 9-4 presents annual statistics of production, sales
volume, sales value and average unit value for the industry. The data are
weighted using the procedures described in Appendix El to reflect the behav-
ior of the industry as accurately as possible.
9.1.7 International Trade
Chemical imports were first made subject to tariffs at the beginning
of the 20th century. The tariffs were initiated to protect the infant chemi-
cal industry from foreign competition. Since 1936, tariffs have been pro-
gressively lowered on chemical products.7 The U.S. International Trade Com-
mission reports 824 benzenoid intermediates on which tariffs are collected.
Of these, 179 are assessed duties competitively using import prices as the
9-11
-------
basis for tariffs. Another 430 of these products are classified noncompeti-
tive, with tariffs based on U.S. domestic prices. The competitive status of
15 products is not available.8 The remainder are not tariffed.
Accurate data concerning imports and exports of SOCMI chemicals are not
available. The most reasonable approximation of trade statistics for SOCMI
are provided by the U.S. International Trade Commission.5 Annual value of
imports and exports for the period 1966-1977 is presented in Table 9-6. In
each of these years, U.S. exports exceeded U.S. imports of industrial organic
chemicals. Table 9-7 presents imports, exports and trade balance of indus-
trial organic chemicals in 1976 and 1977 between the United States and its
principal trading partners. These countries include West Germany, Italy, the
United Kingdom, Switzerland, France, Belgium, the Netherlands, Canada, Japan,
Mexico, Brazil and Argentina. In 1977 the U.S. experienced a deficit in its
balance of trade in chemicals with West Germany, Japan, Italy, the United
Kingdom, Switzerland and France. It experienced a surplus in its balance of
trade in chemicals with Belgium, Canada, the Netherlands, Mexico, Argentina
and Brazil. Table 9-8 presents the value of imports for consumption from
principle sources from 1972 to 1977. These imports amounted to a total of
about $326 million in 1977.
9.1.8 Industry Growth
A number of forecasts of economic growth in the organic chemical in-
dustry are available. The annual growth rate used here, 5.9 percent, was
estimated by McGraw Hill10 for the basic organic chemicals industry. The
McGraw Hill estimate was selected for the following reasons. First, the growth
rate was calculated for a group of chemicals which closely corresponds to the
378 SOCMI chemicals. Second, the method used by McGraw Hill to develop the
growth rate is internally consistent and takes account of forecasted develop-
9-12
-------
TABLE 9-7. INDUSTRIAL ORGANIC CHEMICALS:
U.S. TRADE, BY PRINCIPAL TRADING PARTNERS, 1976 AND 19779
($103)
Source
Imports1
Exports
Trade balance
1976:
West Germany
Japan
Italy
United Kingdom
Switzerland
France
Belgium
Canada
Netherlands
Mexico
Argentina
Brazil
All other
Total
1977:
West Germany
Japan
Italy
United Kingdom
Switzerland
France
Belgium
Canada
Netherlands
Mexico
Argentina
Brazil
All Other
Total
94,768
61,228
30,678
24,709
17,280
12,371
2,154
8,081
8,987
3,452
1,927
98
28,103
293,836
105,172
65,770
32,711
31,132
21,956
15,763
9,839
7,270
4,858
4,673
3,353
538
22,865
325,900
10,487
27,380
N.A.
15,497
2,681
11,401
46,779
93,471
178,111
63,964
N.A.C
59,444
498,985
1,008,200
5,038
30,736
N.A.C
27,458
6,541
N.A.C
61,126
82,676
156,581
62,965
6,283
78,512
477,469
995,385
84,281
33,848
30,000
9,212
14,599
970
44,625
85,390
169,124
60,512
- 1,500
59,346
470,882
714,364
100,134
- 35,034
32,500
- 3,674
15,415
15,500
51,287
75,406
151,723
58,292
2,930
77,974
454,604
669,485
Data represent customs import value, the value appraised by the U.S. Customs
Service in accordance with the legal requirements of sec. 402 and 402a of
the Tariff Act of 1930, as amended.
Includes exports of some finished products.
are not strictly comparable with imports.
CN.A. = Not available.
Figures include estimates and
9-13
-------
TABLE 9-8. INDUSTRIAL ORGANIC CHEMICALS:
U.S. IMPORTS FOR CONSUMPTION, BY PRINCIPAL SOURCES, 1972-779
($103)3
Source 1972 1973 1974 1975 1976 1977
West Germany
Japan
Italy
United Kingdom
Switzerland
France
Belgium
Canada
Netherlands
Mexico
Argentina
All other
Total
66,085
36,181
11,305
7,605
11,593
1,611
1,220
4,301
5,067
35
3
5,031
150,037
72,715
29,793
10,705
10,433
16,063
4,233
7,919
5,515
4,724
486
--
6,892
169,478
84,059
65,027
17,323
21,119
15,846
8,585
10,494
4,826
10,291
1,812
--
19,190
258,572
62,145
49,243
19,073
18,820
14,773
9,797
1,871
4,352
6,738
388
657
17,625
205,482
94,768
61,228
30,678
24,709
17,280
12,371
2,154
8,081
8,987
3,452
1,927
28,201
293,836
105,172
65,770
32,711
31,132
21,956
15,763
9,839
7,270
4,858
4,673
3,353
23,403
325,900
aCustoms import value, the value appraised by the U.S. Customs Service in
accordance with the legal requirements of sec. 402 and 402a of the Tariff Act
of 1934, as amended.
9-14
-------
ments in the U.S. economy. Third, the projections are developed for the the
period 1979-1991, entirely covering the forecast period of interest in this
study (1981-1985).
In order to estimate the number of new model units covered by the regu-
latory alternatives, it is assumed that the number of operating facilities
will grow at the same rate as the industry's output. It is further assumed
that any regulatory alternative will take effect on January 1, 1981, and
therefore that the fifth year of the impact analysis is 1985. In 1976 (the
most recent year for which data are available), 1,334 facilities manufactured
SOCMI chemicals in the U.S.11 If the industry grows at an annual rate of 5.9
percent, by the beginning of 1981 this number will have risen to 1,678 facil-
ities and by the end of 1985 to 2,235 facilities. Thus, an estimated 557
units built to provide additional capacity for the industry will be covered
by the regulatory alternatives.
The regulatory alternatives will also cover units constructed to replace
existing capacity which "wears out" during the period. The number of replace-
ment units is estimated on the basis of the following assumptions. First,
units have a working life of 20 years. Second, the historical growth rate for
SOCMI prior to 1977 was 6 percent per year. Using these assumptions, it is
estimated that 274 new units will be required to replace the part of the
existing capacity that will "wear out" over the period 1981 to 1985. The
methodology used to compute this estimate is described in detail in Appendix
E2.
The estimates for entirely new units, combined with estimates for
replacement facilities, indicate a total of 831 units that will be affected
by the regulation. To estimate the number of A, B and C model units (identi-
fied in Section 6.1) that will be constructed between 1981 and 1985, it is
9-15
-------
assumed that the mix of model units will not change over time, and that the
percentages of A, B and C model units are as follows:
Model unit Percent of existing units
A 52
B 33
C 15 !
If these percentages are applied to the estimate of the total number of
units presented above, they imply that 432 A units, 274 B units and 125 C
units will be affected by the regulatory alternatives.
9.2 Economic Impact Analysis
9.2.1 Market Structure and Financial Profile
SOCMI producers manufacture 378 chemicals, each of which has its own
national and regional markets. Consequently, SOCMI firms encounter a wide
range of market situations for the different chemicals they produce. Many
SOCMI chemicals, for example, formaldehyde, urea and benzene, are manufactured
by a relatively large number cf firms using an array of different processes.
The products have a wide range of end uses in which substitute materials can
often be used. Thus industry-wide elasticities of demand for the chemicals
are relatively high. In this type of market situation, producers have little
or no ability to pass on cost increases to consumers in the form of higher
market prices. Other SOCMI chemicals, for example, succinonitrile, isoamy-
lene, and methyl butynol, are manufactured by a small number of producers and
in some cases only one producer, and have no close substitutes in their end
uses. In these oligopolistic and monopolistic markets, producers may be able
to exercise considerable influence on market prices and to pass on a large
part or all of any production cost increases in the form of higher prices.
9-16
-------
The ability of firms to pass on cost increases in the form of price in-
creases is influenced by the extent to which the industry is vertically and
horizontally integrated. There is extensive vertical integration within the
SOCMI. Captive consumption in the industry averaged 53.7 percent* of total
output during the period 1967-1978, and this ratio varied only slightly from
year to year. The precise degree of horizontal integration within SOCMI is
I
difficult to evaluate because it varies iconsiderably among products. However,
a general assessment of the industry-wide situation may be made using the ca-
pacity share data presented in Table 9-9i These data suggest that no one com-
pany or group of companies has a dominant position within the industry. In
1976, the top four companies owned only 18.3 percent and the top twenty firms
1
45.4 percent of total SOCMI capacity. There is no reason to believe that
1
the extent of industry-wide market concentration has altered significantly
since that time. I
Data on the returns on equity, returns on debt, returns on preferred
i
stock, debt-asset ratios, equity-asset ratiios and preferred stock-asset ra-
tios were collected for a sample of 100 chemical manufacturing firms for the
most recent available years.t These data are presented in Table E3-2. The
data have been used to estimate the cost of capital to firms in the SOCMI,
using the assumption that the sample of firms in Table E3-2 is unbiased and
normally distributed. A detailed discussion of the methodology used to esti-
mate the cost of capital is presented in Appendix E3.
The estimated cost of capital, presented in Table 9-10, is used in Sec-
tion 9.2.3 to estimate the economic impacts of SOCMI fugitive emissions regu-
This figure is estimated from data presented in Table 9-4.
tData on the ratio variables and rates of return were available for 1977 and
1978, respectively.
9-17
-------
TABLE 9-9. INDUSTRY CONCENTRATION, 19764
Number Percent Estimated Percent of
of firms of firms capacity (gg) industry capacity
Top 4 0.72 58.75 18.3
Top 8 1.43 91.82 28.6
Top 20 3.58 145.75 45.4
Top 40 7.17 186.68 58.1
9-18
-------
TABLE 9-10. ESTIMATED COST OF CAPITAL FOR FIRMS IN SOCMI3
Mean
Standard
deviation
Minimum
Maximum
Aftertax cost
of capital
10.807%
0.930
8.015%
12.798%
Pretax cost
of capital
20.783%
1.789
15.414%
24.612%
aSee Appendix E3 for details of the data and methodology used to estimate
the cost of capital for firms in SOCMI.
9-19
-------
latory alternatives. Note that the average aftertax cost of capital for
chemical firms is 10.8 percent. On a pretax basis, this figure increases to
20.8 percent. If, as was assumed, capital costs are normally distributed,
then 95 percent of the firms in the industry face aftertax costs of capital
in the range of 9.0 percent to 12.7 percent and pretax costs of capital in
the range of 17.2 percent to 24.4 percent.
9.2.2 Regulatory Alternatives
The four regulatory alternatives being considered are described in de-
tail in Section 6.2. The baseline regulatory alternative (alternative I)
does not require producers to implement additional control techniques. Con-
sequently, model units complying with this alternative would not incur any
incremental costs* and no economic impacts would result from its implementa-
tion. Regulatory alternatives II, III, and IV require successively more
stringent equipment inspections and equipment specifications. Firms comply-
ing with regulatory alternatives II, III, and IV would therefore incur incre-
mental costs, and consequently economic impacts would result from their imple-
mentation.
9.2.3 Economic Methodolgy
9.2.3.1 Regulatory Scenarios
Economic impacts are estimated for regulatory alternatives II, III
and IV, but not for regulatory alternative I, since firms will not incur in-
cremental costs in complying with that alternative. The economic impacts
associated with alternatives II, III and IV are estimated under two alterna-
*Incremental costs of a regulatory alternative are those additional costs a
firm incurs in meeting the regulatory alternative that it would not incur in
meeting the baseline alternative.
9-20
-------
tive assumptions about firm pricing behavior: (1) full cost absorption and
i
(2) full cost pricing. Combining the three regulatory alternatives with the
two alternative pricing models yields six regulatory scenarios:
Regulatory Alternative Pricing Policy
Scenario 1 Alternative II Full Cost Absorption
Scenario 2 Alternative II Full Cost Pricing
Scenario 3 Alternative III Full Cost Absorption
Scenario 4 Alternative III Full Cost Pricing
Scenario 5 Alternative IV Full Cost Absorption
Scenario 6 Alternative IV Full Cost Pricing
Under full cost absorption, the affected firm bears the full incremental
costs of environmental controls, accepting a lower rate of return on its cap-
ital investment. Under full cost pricing, the firm adjusts product prices so
as to maintain its current aftertax rate of return on capital investment.
The alternative assumptions about firm pricing behavior are associated
with different market conditions in the affected industry. In both cases,
firms are assumed to have no monopsony power in resource markets. Thus, they
cannot pass back cost increases to resource suppliers. In the cost absorp-
tion case, the domestic industry as a whole is assumed to be a price taker,
unable to affect the market price of its product either because of the exist-
9-21
-------
ence of close product substitutes, or because of strong international compe-
tition in domestic and foreign markets. However, full cost pricing will take
place if the industry produces a commodity for which no domestic or imported
substitutes exist, or if the industry has constant costs. A constant-cost in-
dustry is one in which unit costs remain constant as industry output in-
creases. Firms in such industries experience constant returns to scale.
In fact, firms in SOCMI face a wide variety of product market situations
(see Section 9.2.1). Some firms will be able to fully pass through cost in-
creases to consumers in the form of higher prices. Some will be able to pass
on only a part of the cost increases. Others will be forced to fully absorb
all regulatory control costs, leaving product prices unchanged. Consequent-
ly, the full cost pass through and full cost absorption scenarios evaluated
below provide estimates of the maximum range of possible price and rate of
return impacts for the different products and firms in SOCMI.
9.2.3.2 Estimation of Regulatory Price Impacts Under Full Cost Pricing
Under full cost pricing, the firm is assumed to respond to cost in-
creases by adjusting product price to maintain a target rate of return on in-
vestment. The required price change (dP) may be calculated using the follow-
ing equation:*
HD _ dTOC + r dK/(l-t) (1)
dK - Q
where
dP = required change in product price
dTOC = total annual operating costs of compliance
dK = total initial costs of compliance
Q = total annual unit output
*The derivations of Equations (1) and (2) are presented in Appendix E4.
9-22
-------
r = target rate of return
t = tax rate
9.2.3.3 Estimation of Rate of Return Impacts Under Full Cost Absorption
Under full cost absorption, an increase in facility production costs re-
sults in a lower rate of return on investment for the firm, because market
conditions prohibit it from passing on cost increases to the consumer. The
impact on the facility's rate of return on investment is given by the follow-
ing equation:
-dr = r * dK + P-V dTOC (2)
K
where
dr = change in rate of return, and
K = preregulation level of capital investment.
Note that pretax rate of return impacts may be calculated by setting the tax
rate variable, t, equal to zero in Equations (1) and (2). Also note that
price and rate of return impacts are estimated on the assumption that capa-
city utilization rates remain constant (that is, Q remains unchanged). To
the extent that the regulatory alternatives result in decreases (increases)
in capacity utilization rates, price and rate of return impacts will be larg-
er (smaller) than those estimated using Equations (1) and (2) because of eco-
nomies of scale in the use of control techniques.
9.2.3.4 Other Economic Impacts
The price and rate of return impacts estimated by the above techniques
are used to make a quantitative assessment of the probable impacts of the reg
ulatory alternatives II, III, and IV on industry growth, new facility open-
ings, the replacement of existing facilities, and investment levels. These
data are then used to assess the extent of interindustry and macroeconomic
impacts associated with the various regulatory alternatives.
9-23
-------
9.2.3.5 Estimation Data
Estimation of price and rate of return impacts for different model units
requires data on the following variables: (1) total acquisition and instal-
lation costs of the control equipment (dK), (2) total annual operating costs
of the control equipment and monitoring procedures (dTOC), (3) the preregula-
tion capital stock (K), (4) the target rate of return (r), (5) the tax rate
(t), and (6) model unit production levels (Q). Data on dK and dTOC for each
of the three model units identified as representative of the industry* were
obtained from Section 8.1. The tax rate is assumed to be 48 percent. Data
on model unit production levels were obtained from the 1976 Organic Chemical
Producers Data Base for each of the units covered by the regulatory alterna-
tives. The 831 model units are assumed to be distributed by capacity in an
identical manner to the 1,105 units for which both value of product and quan-
tity data are available in the 1976 Organic Chemical Producers Data Base.
Thus, the number of new units with a given capacity, say 100 Gg, is assumed
to be equal to the number of units in the Organic Chemical Producers Data
Base with the same capacity, multiplied by the ratio of the number of new
units to the number of units in the data base (831/1,105). Actual unit output
levels are obtained by applying a capacity utilization rate to the estimated
unit capacities.
To evaluate industry-wide impacts, the cost data from Section 8.1 were
adjusted to allow for higher or lower product recovery credits for chemicals
with a value greater or less than $0.36/kg. For such chemicals, product re-
covery credits were estimated by multiplying estimated product savings by the
price of the chemical in question.
*See Chaper 6 for a detailed discussion of the model units.
9-24
-------
Data on the value of the preregulation capital stock for plants of dif-
ferent capacities were calculated as follows. A capital-capacity coefficient
for firms in SOCMI was obtained by dividing the estimated total value of in-
dustry assets in 1976 by the volume of output produced in that year.* The
estimate of the capital-output coefficient was converted into a capital-
capacity coefficient by multiplying the capital-output coefficient by an
assumed industry-wide capacity utilization rate. The assumed capacity utili-
zation rate for 1976 was 50 percent. This capacity utilization estimate was
based on the assumption that the typical capacity utilization for the indus-
try is 75 percent. In 1976, output was 9.2 percent below the industry-wide
high level of output achieved in 1974. Between 1974 and 1976 it is probable
that some additions to industry capacity were made. Hence, the assumption of
a 50 percent capacity utilization rate for 1976, though somewhat arbitrary,
is not unreasonable. The 1976 capital-capacity coefficient was updated to
last quarter 1978 dollars using the machinery and equipment price index com-
puted by the United States Department of Commerce.12 The capital-capacity
coefficient estimated by the above procedure was $125/Mg of product. This
coefficient was multiplied by model unit capacity to obtain an estimate of K
for each model unit considered in the analysis.
Estimates of pretax and post-tax rates of return used in the analysis
are presented in Table 9-10. These data were obtained from an analysis of a
sample of 100 firms in the SOCMI industry. Details of the analysis are
contained in Appendix E3.
''See Tables 9-4 and 9-5 for data on production and total industry assets.
9-25
-------
9.2.4 Economic Impacts
9.2.4.1 Rate of Return Impacts
Data on unit capacity, product value, capital investment and tax rates
are available for 1,105 units in the 1976 Organic Chemical Producers Data
Base.4 Price data were updated using the Chemical Marketing Reporter.13 Cap-
ital stock estimates were also expressed in 1978 prices. These data were
used in conjunction with the cost information presented in Section 8.1 to cal-
culate full cost absorption rate of return impacts of regulatory alternatives
II, III, and IV for the 831 model units projected to be built. It is assumed
in estimating the rate of return impacts presented here that the 831 new model
units will have the same capacity and product value distributions as the units
in the Organic Chemical Producers Data Base, and that capacity utilization
for each unit is 50 percent. This relatively low capacity utilization rate
is used to estimate unit output levels because it represents a feasible
worst-case economic scenerio for the industry (that is, economic conditions
similar to those experienced in 1976). As a-result, actual impacts are like-
ly to be less adverse than those presented below.
Rate of return impacts are estimated on the basis of these assumptions
for each of the 831 new model units covered by the regulatory alternatives.
It is probable that the assumption of a constant product price distribution
also results in an overestimate of adverse rate of return and price impacts,
since the prices of the products manufactured by SOCMI are expected to in-
crease between 1979 and 1985 as energy and feedstock costs rise. Any real
increase in product prices will raise the value of product recovery credits,
lower the net costs of compliance associated with any given regulatory alter-
native, and thereby reduce adverse rate of return and price impacts.
Rate of return impacts for A, B and C average model units under each reg-
9-26
-------
ulatory alternative are presented in Table 9-11. Each of these average model
units is assumed to manufacture products valued at approximately $0.36/kg, to
have an annual capacity of 84,678 Mg and to have an existing cost of capital
of 10.81 percent. These average model units differ only in terms of the com-
plexity of the processes they use to manufacture the chemicals. The product
value and rate of return data represent the means for each variable in the
samples used in the analysis. Under regulatory alternatives II and III, each
average model unit experiences a very small increase, not a decrease, in its
aftertax rate of return on investment, regardless of the process it uses.
This result is obtained because at a price of $0.36/kg for recovered product,
product recovery credits exceed total annualized costs of control. Under reg-
ulatory alternative IV, average model unit of types A and B experience small
decreases in aftertax rates of return on investment. Model C units experience
rate of return decreases amounting to 1.12 percentage points, still quite
small adverse impacts.
The data presented in Table 9-11 suggest that some firms may benefit from
the implementation of any regulatory alternative. These results are subject
to the following qualifications. In the above analysis, it is assumed that
firms will not independently implement the emissions controls proposed in the
regulatory alternatives. In fact, if there are significant net cost reduc-
tions to be achieved from additional emissions controls, firms will voluntar-
ily adopt them. Under such circumstances, the cost reductions associated
with any regulatory alternative will be considerably smaller. Note that in-
centives for voluntary emissions controls increase as the value of the manu-
factured product increases. As some SOCMI producers manufacture highly valued
products with prices in excess of $0.50/kg, they are likely to be willing to
use extensive emissions control techniques in the absence of any NSPS.
9-27
-------
TABLE 9-11. AVERAGE RATE OF RETURN IMPACTS3
Change in rate of return (percentage points)
Model
Units
Unit A
Unit B
Unit C
Alternative II
+0.000
+0.003
+0.006
Alternative III
+0.000
+0.001
+0.005
Alternative IV
-0.16
-0.37
-1.12
almpacts are estimated on the assumption that the initial aftertax rate of reti;vn
on investment is 10.807 percent, the mean cost of capital presented in Table 9-10;
the initial price of the product is $0.36/kg; plant capacity is 84,678 Mg; and
the capacity utilization rate is 50 percent.
9-28
-------
Although in general firms will not be affected by the implementation of
regulatory alternatives II, III, and IV, because of wide variation in produc-
tive capacity and value of product among model units, some will experience
adverse rate of return impacts. Estimates of the numbers of model units exper-
iencing rate of return decreases in excess of one and two percentage points
as a result of the implementation of each regulatory alternative are present-
ed in Table 9-12. These estimates were obtained by calculating rate of return
impacts for each of the 831 new model units under the assumption that 52 per-
cent, 33 percent, and 15 percent of all units of all sizes are A, B and C
model units, respectively. Under regulatory alternatives II and III, the
estimated number of adversely affected units is very small; only 6 and 12
units, out of a total of 831 model units, experience rate of return decreases
of more than one percentage point. Under regulatory alternative IV, a much
larger number of units, 93 in all, are estimated to experience rate of return
decreases in excess of 1 percent under a full cost absorption scenario. It
should be noted that most of these adversely affected units are B and C model
units rather than A model units. In fact, all B and C model units with capa-
cities in excess of 26,464 and 89,121 Mg, respectively, producing chemicals
with prices exceeding $0.15/kg, will experience rate of return impacts small-
er than one percentage point even under alternative IV. The EPA estimates
that virtually all B and C model units do in fact have capacities in excess
of this figure,14 and furthermore, industry sources indicate that most pro-
duce chemicals that have prices in excess of $0.15/kg.13 If the estimated
impacts on B and C model units are ignored, only 25 units are likely to be
adversely affected by regulatory alternative IV.
9-29
-------
TABLE 9-12. MODEL UNITS EXPERIENCING SIGNIFICANT RATE OF RETURN IMPACTS
UNDER FULL COST ABSORPTION3
Model
units
Unit A
Unit B
Unit C
Total
Alternative
dr < -1% dr
6
0
0
6
II
< -2%
4
0
0
4
Alternative
dr < -1% dr
7
4
1
12
III
< -2%
5
2
0
7
Alternative
dr < -1% dr
25
34
34
93
IV
< -2%
12
16
20
48
adr denotes the percentage point change in firms' rates of return on investment.
9-30
-------
9.2.4.2 Price Impacts
The potential price impacts of regulatory alternatives II, III, and IV
are also estimated under the assumption that capacity and value of product
distributions will remain constant over the forecast period, 1979-1985. The
price impact estimates are therefore subject to the same limitations as the
rate of return impact estimates discussed above. Potential price impacts for
A, B, and C model units with average capacities of 84,678 Mg and product val-
ues of $0.36/kg are presented in Table 9-13. Under regulatory alternatives
II and III, price impacts are negative for each type of model unit because
annual product recovery credits exceed the total annualized cost of the moni-
toring procedures and capital equipment required under these alternatives.
Under regulatory alternative IV, extremely small positive price impacts
occur. In general most units will not increase product prices as a result of
the implementation of regulatory alternatives II, III, or IV. However, be-
cause of the variations in capacity and product value within the industry,
some firms may have to raise product prices in order to maintain existing
rates of return on investment. In some cases, the price increases required
by individual facilities are in excess of 5 percent and even 6 percent of the
current product price. Data on the estimated numbers of such units are pre-
sented in Table 9-14. These estimates are also obtained by calculating price
impacts for each of the 831 new model units under the assumption that units
are distributed among A, B, and C model units in the manner described above
and operate at 50 percent of unit capacity. Under alternatives II and III,
only five A and eight B model units would have to increase product prices by
more than 5 percent. Under alternative IV, 30 units must increase prices by
more than 5 percent to maintain preregulation rates of return on investment.
However, it should be noted that these estimates may overstate the extent of
9-31
-------
TABLE 9-13. AVERAGE PERCENTAGE PRICE IMPACTS OF REGULATORY ALTERNATIVES3
Model
units
Unit A
Unit B
Unit C
Alternative II
-0.000
-0.002
-0.009
Price changes (percent)
Alternative III
-0.000
-0.002
-0.007
Alternative IV
+0.000
+0.000
+0.733
Impacts are estimated on the assumption that the target rate of return is
10.807 percent, the average cost of capital presented in Table 9-10; the
initial price of the product is $0.36/kg; plant capacity is 84,678 Mg; and
the capacity utilization rate is 50 percent.
9-32
-------
TABLE 9-14. MODEL UNITS REQUIRING SIGNIFICANT PRICE INCREASES
TO MAINTAIN TARGET RATES OF RETURN
Model
Units
Unit A
Unit B
Unit C
Total
Alternative II
Price increase
£ 5% ^6%
4
0
0
4
4
0
0
4
Alternative III
Price increase
^5% £ 6%
4
2
0
6
4
2
0
6
Alternative IV
Price increase
^5% ^6%
6
10
14
30
5
9
12
26
9-33
-------
significant price impacts under regulatory alternative IV. Most of the units
estimated to require price increases in excess of 5 percent are C model
units. In fact, C model units that manufacture chemicals with prices in ex-
cess of $0.15/kg and have capacities greater than 71,550 Mg do not have to
increase product prices by more than 5 percent to maintain their target rates
of return on investment. The EPA estimates that virtually all plants using
processes with the same degree of complexity as that assumed for C model units
have larger capacities and produce products with higher values than these.13
In addition, model units are assumed to operate at the relatively low capacity
utilization rate of 50 percent.
A final caveat concerning price and rate of return impacts should be
noted. The impact estimates presented in Tables 9-12 and 9-14 were developed
on the assumption that feedstock prices are unaffected by the implementation
of any regulatory alternative. However, as the industry extensively uses its
own products as feedstocks, this assumption is not strictly valid and intro-
duces a systematic upward bias in the estimated size of adverse rate of re-
turn and potential price impacts. The upward bias occurs because, in general,
firms adopting alternative II, III, and IV control technologies will achieve
net cost reductions and, at least in competitive markets, will tend to reduce
rather than increase the prices of products used as feedstocks by the indus-
try.
9.2.4.3 Investment Impacts
It is difficult to assess the impact of any of the standards on the num-
ber of units to be constructed between January 1, 1981, and December 31, 1985,
because of the variations in these impacts across units. Some smaller facili-
ties may not be erected as a result of the standard because of adverse impacts
on rates of return and price competiveness. Other larger facilities may be
9-34
-------
built because production costs fall as a result of emissions reductions and
product recovery credits. Therefore, in this analysis it is assumed that im-
plementation of regulatory alternatives II, III, IV will have no measurable
impact on the number of new facilities constructed between 1981 and 1985, the
5-year period following proposal of any regulatory alternative. Industry-
wide investment impacts are therefore simply the incremental capital costs
associated with the acquisition of the capital and monitoring equipment re-
quired under each regulatory alternative by the 831 new units expected to be
constructed between 1981 and 1985.
Data on these investment impacts are presented in Table 9-15. The esti-
mates are obtained by assuming that 432 A model plants, 274 B model units,
and 125 C model units will be constructed and that, as a result of each regu-
latory alternative, these units incur incremental capital costs equal to those
presented in Section 8.2. Under regulatory alternatives II and III,
industry-wide investment impacts are quite small, less than $22 million.
Under regulatory alternative IV, they increase substantially to almost $233
million. Nevertheless, even under regulatory alternative IV, the total 5-year
investment impacts of any of the regulatory alternatives would be less than
1.14 percent of total industry assets in 1976 and less than 1.64 percent of
the value of industry sales in 1978.* It appears, therefore, that the indus-
try as a whole will not have much difficulty in obtaining the investment funds
to acquire required control equipment under any of the regulatory alterna-
tives.
*Data on total industry value of assets and industry sales are presented in
Table 9-4 and 9-5.
9-35
-------
TABLE 9-15. INVESTMENT IMPACTS
Model
units
Number of
model units
Incremental
model unit
costs of control
($106)
Incremental indus-.:
costs of control
($106)
Regulatory
alternative
Unit A
Unit B
Unit C
II
Total
432
274
125
831
Total
831
0.0132
0.0272
0.0660
Total
831
5.7
7.5
8.3
21.5
Regulatory
alternative III
Unit A
Unit B
Unit C
432
274
125
0.0132
0.0272
0.0660
5.7
7.5
8.3
21.5
Regulatory
alternative IV
Unit A
Unit B
Unit C
432
274
125
0.0871
0.2950
0.9110
37.6
80.8
113.9
232.3
9-36
-------
9.2.4.4 Employment Impacts
Regulatory alternatives II, III, and IV will each have small but measur-
able impacts on employment in SOCMI because they require firms to intensify
monitoring and maintenance schedules to control fugitive emissions. Estimates
of the number of additional workers required as a result of each regulatory
alternative are presented in Table 9-16. The estimates were obtained by mul-
tiplying the projected numbers of each type of affected facility by the unit-
by-unit, person-year monitoring and maintance requirements for each standard
presented in section 8.1.* The largest employment impacts (400 workers)
are associated with regulatory alternative III, which requires more stringent
monitoring programs than alternative II. Under alternative IV, some alterna-
tive III monitoring requirements are replaced by equipment controls, reducing
incremental employment requirements to approximately 225 workers. The employ-
ment impacts of each of the standards are small relative to total employment
in the industry, representing no more than 0.6 percent of the 1976 SOCMI work
force in each case.
9.2.4.5 Total Annualized Costs of Control
Total incremental annualized costs of control for the fifth year follow-
ing promulgation of alternatives II, III, or IV are presented in Table 9-17.
Product recovery credits are calculated using the fourth quarter 1978
industry-wide average product price of $0.36/kg. Under regulatory alterna-
tives II and III, the industry as a whole is estimated to reduce annual ized
production costs by $28.73 million and $21.35 million, respectively. Under
regulatory alternative IV, annualized production costs are estimated to
*A person-year is assumed to consist of 2,000 person-hours.
9-37
-------
TABLE 9-16. EMPLOYMENT IMPACTS
(Person-years)
Model
unit
Alternative II
Alternative III
Unit Industry
Unit Industry
Alternative IV
Unit Industry
Unit A
Unit B
Unit C
0.1237 53.43
0.1863 51.05
0.5017 62.71
0.1855 80.14
0.5079 139.16
1.4532 181.65
0.0579 25.01
0.2277 62.39
1.0982 137.27
All units
167.19
400.95
224.67
9-38
-------
TABLE 9-17. MODEL UNIT AND INDUSTRY ANNUALIZED CONTROL COSTS
Regulatory
alternative
No. of
model
units
Incremental
unit
annualized costs
without product
recovery credit
($103)
Incremental
unit
annualized costs
with product .
recovery credit0
($103)
Incremental
industry
annualized costs
with product
recovery credit
($103)
Alternative II
Unit A
Unit B
Unit C
Total
Alternative III
Unit A
Unit B
Unit C
Total
Alternative
Unit A
Unit B
Unit C
IV
432
274
125
432
274
125
432
274
125
12.1
20.3
48.3
15.0
35.5
95.1
29.0
93.5
283.0
3.3
38.5
-134.0
- 1.7
28.3
-103.0
7.9
13.3
33.0
Total
- 1,430
-10,550
-16,750
-28,730
730
- 7,750
-12,870
-21,350
3,410
3,640
4,120
11,170
Product recovery credits estimated on the basis of an assumed product value of
$0.36/kg.
9-39
-------
increase by $11.17 million. If the above estimates are accurate in the mini-
mal sense that they indicate the direction in which production costs will
move and their approximate order of magnitude, then it may be concluded that
none of the regulatory alternatives will result in any measurable industry-
wide increase in prices.
9.2.4.6 Interindustry Impacts
Interindustry impacts will be negligible, because net annualized costs
of control are extremely small relative to the value of total industry output,
representing less than 0.03 percent of the value of 1978 output in even the
most adverse case (regulatory alternative IV).
9.3 Socio-Economic and Inflationary Impacts
The socio-economic and inflationary impacts of alternatives II, III and
IV will be very small.
(1) Annualized Costs: In the fifth year following promulgation, the regula-
tory alternatives, if implemented, are estimated to result in either
annualized cost reductions or very small annualized cost increases.
Consequently, none of the alternatives violates the regulatory criterion
of $100 million.
(2) Price Impacts: Because industry-wide annualized costs of compliance for
alternatives II, III and IV are estimated to be negative or extremely
small relative to the value of industry output, none of the standards is
likely to cause any industry-wide price increases.
9-40
-------
REFERENCES
1. Condensed Chemical Dictionary. Rev. by Gessner Hawley, 8th ed. New
York: Van Nostrand Reinhold Co, 1971.
2. Comer, James F. Synthetic Organic Chemicals Manufacturing Industry:
Inputs and Product Uses. Prepared for the U.S. Environmental Protection
Agency, 1979.
3. Proceedings of the Conference on Chemical Feedstock Alternatives.
American Institute of Chemical Engineers, Houston, TX, 1977.
4. Radian Corp. "Organic Chemical Producers Data Base, 1976." Prepared
for the U.S. Environmental Protection Agency under EPA Contract No.
68-03-2623. 1978.
5. United States International Trade Commission. Synthetic Organic Chemi-
cals, U.S. Production and Trade. Washington, D.C., 1967-1978.
6. United States Department of Commerce. Annual Survey of Manufactures,
Industry Profiles, 1976. Washington, D.C.
7. Russell, T.W. F. , M.W. Swartzlander, and J. Wei. The Structure of the
Chemical Processing Industries. New York: McGraw Hill, 1979. pp.
321-334.
8. United States International Trade Commission. Imports of Benzenoid
Chemicals and Products, 1978. Washington, D.C., July 1979.
9. United States International Trade Commission. "Import Penetration of
U.S. Markets for Cyclic Intermediates." Synthetic Organic Chemicals:
U.S. Production and Sales -1977. Washington, D.C., 1978.
10. The American Economy, Prospects for Growth to 1991. New York: McGraw
Hill, 1979.
11. Radian Corp. "Organic Chemical Producers Data Base, 1978." Prepared
for the U.S. Environmental Protection Agency under EPA Contract No.
68-03-2623. 1979.
12. United States Department of Commerce. Survey of Current Business.
Washington, D.C., 1976-1979.
13. Chemical Marketing Reporter. September 7, 1979.
14. Memo from Hustvedt, K.C. December 20, 1979. Model Unit Capacities.
9-41
-------
APPENDIX C. EMISSION SOURCE TEST DATA
-------
APPENDIX C
EMISSION SOURCE TEST DATA
The purpose of Appendix C is to describe testing results used in the
development of the Background Information Document (BID) for fugitive
emissions from the Synthetic Organic Chemicals Manufacturing Industry
(SOCMI). The information in this appendix consists of a description of
the tested facilities, and the sampling procedures and test results of
fugitive emissions studies in SOCMI and the petroleum refining industry.
Considerable data exist concerning both the incidence and magnitude
of fugitive emissions from petroleum refineries. The purpose of the SOCMI
study was, in part, to support the use of emission factors generated
during studies of emissions from petroleum refineries for similar sources
in the Synthetic Organic Chemicals Manufacturing Industry. The results
of the SOCMI study and data from a study of fugitive emissions from
petroleum refineries are discussed in Section C.I.
Section C.2 consists of the results of three studies on the effects
of maintenance on reducing fugitive VOC emissions from valves in petroleum
refineries and one study on maintenance of valves in a SOCMI process unit.
These results are included as an indication of the reduction in emissions
which could be expected as a function of the designated action level, and
by applying routine on-line maintenance procedures.
C.I FUGITIVE EMISSIONS TEST PROGRAMS
The SOCMI test program conducted by EPA personnel and EPA contractors
consisted of emissions testing in six chemical process units. Data were
collected pertaining to the percentage of fugitive emission sources found
to be leaking, as indicated by the VOC concentration measured at the source.
The results of a study on fugitive emissions from petroleum refineries
are also discussed in this section. Data on fugitive emissions were obtained
from thirteen refineries located in major refining areas throughout the
C-l
-------
country. Data on the effects of maintenance were obtained at the
last four of these refineries. These results are discussed later in
Section C.2 of this Appendix.
The test procedures and the results obtained for each of these
studies are described in detail in the following sections.
C.I.I Description and Results of SOCMI Study
The objective of this test program was to gather data on the percen-
tage of sources which leak (as defined by a VOC concentration at the leak
interface of >10,000 ppmv). To achieve this objective, an attempt was
made to screen all potential leak sources (generally excluding flanges)
on an individual component basis with a portable organic vapor analyzer.
The test crews relied on plant personnel to identify equipment handling
organics. Normally all pumps and compressor seals were examined, and the
percentage of valves carrying VOC which were screened ranged from 33 to
85 percent. All tests were performed with a Century Systems Corporation
Organic Vapor Analyzer, Model 108, with the probe placed as close to the
source as possible. The results of this study are shown in Table C-l.
Six chemical process units were screened. Unit A is a chlorinated
methanes production facility in the Gulf Coast area which uses methanol
as feedstock material. The individual component testing was conducted
during September 1978. Unit B is a relatively small ethylene production
facility on the West Coast which uses an ethane/propane feedstock.
Testing was conducted during October 1978. Unit C is a chlorinated
methanes production facility in the Midwest. This plant also uses
methanol as the basic organic feedstock. Over the last few years, several
pieces of equipment have been replaced with equipment the company feels is
more reliable. In particular, the company has installed certain types of
valves which they have found do not leak "as much" as other valves..
C-2
-------
The individual component testing was conducted during January 1979. Unit
D is an ethylene production facility on the Gulf Coast, using an ethane/
propane feed. The facility is associated with a major refinery, and
testing was conducted during March 1979. Units E and F are part of an
intermediate size integrated petroleum refinery located in the North
Central United States. Testing was conducted during November 1978. Unit E
is an aromatics extraction unit that produces benzene, toluene, and xylene
by extraction from refined petroleum feedstocks. Unit E is a new unit and
special attention was paid during the design and startup to minimize
equipment leaks. All valves were repacked before startup (adding 2 to 3
times the original packing) and all pumps in benzene service had double
mechanical seals with a barrier fluid. Unit F produces benzene by
hydrodealkylation of toluene. Unit F was originally designed to produce a
different chemical and was redesigned to produce benzene.
In general, chloromethane plants had fewer leaks than the ethylene
production facilities.
C.I.2 Description and Results of Refinery Fugitive Emissions Study
Data concerning the leak frequencies and emission factors for various
fugitive sources were obtained primarily at nine refineries. More
complete information for compressors and relief valves emissions was
obtained by sampling at four additional refineries. Refineries were
selected to provide a range of sizes and ages and all of the major
petroleum refinery processing units were studied. The type of process
units and the number of each studied in the first nine refineries are
listed in Table C-2.
C-3
-------
TABLE C-l FREQUENCY OF LEAKS FROM FUGITIVE EMISSION SOURCES IN
SYNTHETIC ORGANIC CHEMICAL UNITS
. _ Unit Ac. .... _ . . Unit Bc Unit_C° Unit De Unit Ef
Chloromethanes Ethyl ene Chloromethanes Ethyl ene BTX Recovery
Number Percent with Number Percent with Number Percent with Number Percent with Number Percent with
of screening of screening of screening of screening of screening
sources values sources values sources values sources values sources values
Equipment type tested >10,000 ppmv tested >10,000 ppmv tested >10,000 ppmv tested ^O.OOO ppmv tested >10,000 ppmv
Valves 600 1 2301 19 658 0.1 862 14 715 1.1
Open-ended lines 52. 2 386 11 -a 90 13 33 0.0
Pump seals : 47 15 51 21 39 3 63 33 33b 3.0
o
i
"^ Compressor seals -a 42 59 3 33 17 6 -3
Control valves 52 6 128 20 25 0 25 44 53 4.0
Pressure relief valves 7 0 -a -3 -a -a
Flanges 30 3 -a -a -a -a
Drains -a -a -a 39 10 -a
Unit Ff
Toluene HDA
Number Percent witn
of screening
sources values
tested >_! 0,000 ppmv
427 7.0
28 11.0
30 10.0
_a
44 11.0
_a
_a
_a
a No data
Pump seals in benzene service have double mechanical seals
c Source: Reference- 1
Source: Reference 2
e Source: Reference 3
-------
TABLE C-2. SAMPLED PROCESS UNITS FROM NINE REFINERIES
DURING REFINERY STUDY
Number of
Refinery process unit sampled units
Atmospheric distillation 7
Vacuum distillation 4
Thermal operations (coking) 2
Catalytic cracking 5
Catalytic reforming 6
Catalytic hydrocracking 2
Catalytic hydrorefining 2
Catalytic hydrotreating 7
Alkylation 6
Aromatics/isomerization 3
Lube oil manufacture 2
Asphalt manufacture 1
Fuel gas/light-ends processing - 11
LPG 2
Sulfur recovery 1
Other 3
Source: Ref. 5
C-5
-------
In each refinery, sources in six to nine process units were selected
for study. The approximate number of sources selected for study and
testing in each refinery is listed below:
Valves 250-300
Flanges 100-750
Pump seals 100-125
Compressor seals 10-20
Drains 20-40
Relief Valves 20-40
There were normally 500-600 sources selected in each refinery.
The distribution of sources among the process units was determined
before the selection and testing of individual sources was begun.
Individual sources were selected from piping and instrumentation diagrams
or process flow diagrams before a refinery processing area was entered.
Only those preselected sources were screened. In this way, bias based on
observation of individual sources was theoretically eliminated.
The screening of sources was accomplished with portable organic
vapor detectors. The principal device used in this study was the
J. W. Bacharach Instrument Co. "TLV Sniffer". The components were tested
on an individual basis, and only those components with VOC concentrations
in excess of 200 ppmv were considered for further study.
A substantial portion of these leaking sources were enclosed and
sampled to determine both the methane and nonmethane emission rates. An
important result of this program was the development of a correlation between
the maximum observed screening value (VOC concentration) and the measured
nonmethane leak rate.
Emission factors and leak frequency information generated during
this study are given in Table C-3.
C.I.3 Comparison of Fugitive Emissions Test Data
The results of the SOCMI study and those of the refinery emissions
study are compared in Table C-4. Fugitive emission leak frequencies for
similar source types appear to correlate, particularly for valves and
pump seals.
C-6
-------
TABLE C-3. LEAK FREQUENCIES AND EMISSION FACTORS FROM FUGITIVE
SOURCES IN PETROLEUM REFINERIES
Percent of Estimated emission
sources having factor for
Equipment screening values refinery sources,
type 2.10,000 ppmv kg/hr-source
Valves NA. NA
Gas service 10 0.021
Light liquid service 12 0.010
Heavy liquid service 0 0.0003
Pump seals NA NA
Light liquid service 23 0.12
Heavy liquid service 2 0.02
Compressor seals (hydrocarbon 33 0.44
service)
Pressure relief valves 8 0.086
Gas service 0.16
Light liquid service 0.006
Heavy liquid service 0.009
Flanges 0 0.0003
Open-ended lines NA NA
Gas service 0.025
Light liquid service 0.014
Heavy liquid service 0.003
Source: Ref. 5
C-7
-------
TABLE C-4. COMPARISON OF LEAK FREQUENCIES FOR FUGITIVE EMISSION SOURCES IN SOCMI UNITS AND PETROLEUM REFINERIES
Percent of SOCMI Percent of petroleum refinery sources
sources having screening having screening values ^10,000 ppmv
Equipment type values ^.10,000 ppmva
Valves (all) HC NA
Gas service ]^
Light liquid service ^
Heavy liquid service "
Open-ended lines (all) 10 NA
Gas service
Light liquid service
Heavy liquid service
« Pump seals (all) 17 NA
00 Light liquid service 23
Heavy liquid service 2
Compressor seals (hydrocarbon 43 33
service)
Pressure relief valves (all) 0 8
Gas service
Light liquid service
Heavy liquid service
Flanges (all) 3 0
jjsource: Table C-l.
DSource: Table C-2.
clncludes block and control valves.
-------
C.2 MAINTENANCE TEST PROGRAMS
The results of four studies on the effects of maintenance on fugitive
emissions from valves are discussed in this section. The first two studies
were conducted by refinery personnel at the Union Oil Co. refinery in
Rodeo, California, and the Shell Oil Co. refinery in Martinez, California.
These programs consisted of maintenance on leaking valves containing
fluids with vapor pressures greater than 1.5 Reid Vapor Pressure. The
third study was conducted by Radian Corporation, under contract to EPA.
Valves were selected and maintained at four refineries. The fourth study
was conducted by EPA and EPA contractors at Unit D (ethylene unit). The
study results and a description of each test program are given in the
following sections.
C.2.1 Description and Results of the Union Maintenance Study
The Union valve maintenance study consisted of performing undirected
maintenance on valves selected from 12 different process units. Undirected
maintenance consists of performing valve repairs without simultaneous
measurement of the effect of repair on the VOC concentration detected.
This is in contrast to directed maintenance where emissions are monitored
during the repair procedure. With directed maintenance, repair procedures
are continued until the VOC concentration detected drops to a specified level
or further reduction in the emission level is not possible. Also, maintenance
may be curtailed if increasing VOC concentrations result.
The Union data was obtained with a Century Systems Corporation
Organic Vapor Analyzer, OVA-108. All measurements were taken at a
distance of 1 cm from the seal.
Correlations developed by EPA have been used to convert this data
from readings taken at one centimeter to equivalent readings at the leak
interface. This facilitates comparison of data from different studies
and allows the estimation of emission rates based on screening value-
leak rate correlations.
The results of the Union study are given in Table C-5. Two sets of
results are provided; the first includes all repaired valves with initial
screening values greater than or equal to 5300 ppmv, and the second
includes valves with initial screening values below 5300 ppmv. A screening
value of 5300 ppmv, obtained, with the OVA at 1 cm from the leak interface,
is equivalent to a screening value of 10,000 ppmv measured by a Bacharach
C-9
-------
TABLE C-5. SUMMARY OF MAINTENANCE STUDY RESULTS FROM THE UNION OIL CO.
REFINERY IN RODEO, CALIFORNIA3
o
I
All valves
with initial
screening values
>5300 ppmvb
Number of repairs attempted
Estimated emissions before maintenance, kg/hr
Estimated emissions after maintenance, kg/hr
Number of successful repairs (<5300 ppmv after maintenance)
Number of valves with decreased emissions
Number of valves with increased emissions
Percent reduction in emissions
Percent successful repairs
Percent of valves with decreased emissions
Percent of valves with increased emissions
133
9.72
4.69
67
124
9
51.8
50.4
93.2
6.8
All valves
with initial
screening values
<5300 ppmv
21
0.323
0.422
--
13
8
-30.5
--
61.9
38.1
a
Source: Ref. 6.
DThe value 5300 ppmv, taken with the OVA-108 at 1 cm., generally corresponds to a value of 10,000 ppmv taken
with a "TLV Sniffer" at 0 cm.
-------
Instrument Co. "TLV Sniffer" directly at the leak interface. The OVA-
1 cm readings have been converted to equivalent TLV-0 cm readings because:
1) EPA correlations which estimate leak rates from screening
values were developed from TLV-0 cm data.
2) Additional maintenance study data exists in the TLV-0 cm format.
3) Method 21 specifies 0 cm screening procedures.
The results of this study indicate that maintenance on valves with
initial screening values above 5300 ppm is much more effective than
maintenance on valves leaking at lower rates. In fact, this study indi-
cates that emissions from valves are reduced 51.8 percent for valves
initially over 5300 ppmv while valves with lower initial screening values
experienced an increase in emissions of 30.5 percent.
w
C.2.2 Description and Results of the Shell Maintenance Study
The Shell maintenance program consisted of two parts. First, valve
repairs.were performed on 172 leaking valves. In the second part of the
program, 163 of these valves were rechecked and additional maintenance
was performed. The second part of the program was conducted approximately
one month after the initial maintenance period. It was not determined
whether the maintenance procedures were directed or undirected, based on
the information reported by Shell.
VOC emissions were measured using the OVA-108 and readings were
obtained one centimeter from the source. This data has been transformed
to TLV-0 cm values as was the Union data. And, the same methods of data
analysis described in Section C.2.1 have been applied to the Shell data.
The results of the Shell maintenance study are given in Table C-6.
g
C.2.3 Description and Results of the EPA/Kadian Maintenance Study
Repair data were collected on valves located in four refineries.
The effects of both directed and undirected maintenance were
C-ll
-------
TABLE C-6. SUMMARY OF MAINTENANCE STUDY RESULTS FROM THE SHELL OIL COMPANY
REFINERY IN MARTINEZ, CALIFORNIA
ro
March maintenance
Number of repairs attempted
Estimated emissions before maintenance, kg/hrc
Estimated emissions after maintenance, kg/hrc
Number of successful repairs (<5300 ppmv after
maintenance)
Number of valves with decreased emissions
Number of valves with increased emissions
Percent reduction in emissions
Percent successful repairs
Percent of valves with decreased emissions
Percent of valves with increased emissions
All repaired valves
with initial screening
values >5300' ppmvb
161
11.08
2.66
105
161
0
76.0
65.2
100.0
0.0
aSource: Ref. 8.
The value 5300 ppmv, taken with the OVA-108 at 1 cm., generally corresponds
GShell reported the screening value of all valves which measured <3000 ppmv
All repaired valves
with initial screening
values <5300 ppmv
11
0.159
0.0
11
0
100.0
--
100.0
0.0
to a value of 10,000 ppmv
(<1500 ppmv-TLV at 0 cm.)
April maintenance
All repaired valves with All repaired valves with
initial (March) screening initial (March) screening
values >5300 ppmv
-------
evaluated and other data, including valve size and type and the processes'
fluid characteristics, was obtained. Screening data were obtained with
the Pacharach Instrument Co. "TLV Sniffer" and readings were taken as
close to the source as possible.
Unlike the Shell and Union studies, emission rates were not based on
the screening value correlations. Rather each valve was sampled to determine
emission rates and after maintenance using techniques developed by EPA during
the refinery emission factor study. These values were used to evaluate
emissions reduction.
The results of this study are given in Table C-7. Of interest here
is a comparison of the emissions reduction for directed and undirected
maintenance. The results indicate that directed maintenance is more
effective in reducing emissions than is undirected maintenance, particu-
larly for valves with lower initial leak rates. The results shown an
increase in total emissions of 32.6% for valves with initial screening
values less than 10,000 ppmv which were subjected to undirected maintenance.
However, this increase is due to a large increase in the emission rate
of only one valve.
C.2.4 Description and Results of Unit D (Ethylene Unit) Maintenance Study
Maintenance was performed by Unit D personnel. VOC concentration
measurements were made using the OVA-108, and readings were obtained at
the closest distance possible to the source. The results of this study
are shown in Table C-8. Directed and undirected maintenance procedures
were used. The results show that directed maintenance results in more
repairs being successfully completed than when undirected maintenance is used.
C.2.5 Comparison of Maintenance Study Results
Generally speaking, the results of these maintenance programs would
tend to support the following conclusions:
A reduction in emissions may be obtained by
performing maintenance on valves with screening values
above 10,000 ppmv (measured at the source).
The reduction in emissions due to maintenance of valves
with screening values below 10,000 ppmv is not as dramatic
and may result in increased emissions.
C-13
-------
TABLE C-7. SUMMARY OF EPA REFINERY MAINTENANCE STUDY RESULTS
Repaired values with initial Repaired values with initial
screening values >1 0,000 ppmv screening values <1 0,000 ppmv
Number of valves repaired
Measured emissions before maintenance
kg/hr
Measured emissions after maintenance
kg/hr
Number of successful repairs
(<1 0,000 ppmv after maintenance)
Number of valves with decreased
emissions
Number of valves with increased
emissions
Percent reduction in emissions
Percent successful repairs
Percent of valves with decreased
emissions
Percent of valves with increased
emissions
Directed
Maintenance
. 9
0.107
0. 01 39
8
9
0
87.0
88.9
100.0
0.0
Undirected Directed Undirected
Maintenance Maintenance Maintenance
23 10 16
1.809 0.0332 0.120
0.318 0.0049 0.159
13
21 6 15
2 4 1
82.4 85.2 -32.6
56.5
91.3 60.0 93.8
8.7 40.0 6.3
Source: Ref.9
-------
TABLE C-8. UNIT D ETHYLENE UNIT BLOCK VALVE REPAIRS
Action Level: >10,000 ppm Instrument: "OVA-108" VOC detector
Distance from Source: Maximum concentration at seal interface
o
en
Tag Initial
Number Reading
32
28
16
10
7
4
367
366
364
362
>1 0,000
>10,000
>1 0,000
>1 0,000
>10,000
>10,000
>1 0,000
>10,000
>1 0,000
>10,000
>10,000
>1 0,000
>1 0,000
Date Maintenance
Screened Attempted
03/06/79
03/06/79
03/06/79
03/06/79
03/06/79
03/06/79
03/06/79
03/06/79
03/06/79
03/05/79
03/05/79
03/05/79
03/05/79
No
'No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes
Undirected Directed Maintenance
rid i ii icMaiiLc
Reading 1 2
>10,000 1,100
>10,000 >10,000 210,000
>1 0,000 >1 0,000 >10,000
2,000 100
2,000
>10,000 >10,000 >10,000
100
>10,000 >10,000
200
500
NCb
>10,000 >10,000
Readings
3 Comments
Only checked one valve
with tag meter lines
Only checked one valve
with tag
Only checked one valve
with tag
700 Repaired when valve was
backseated
Bolts all the way down
Bolts need replacing
Leak at gland, not stem
corrosion preventing good
seating of gland
-------
TABLE C-8. UNIT D ETHYLENE UNIT BLOCK VALVE REPAIRS (Continued)
o
01
Tag
Number
360
359
None
358
361
None
356
354
352
65
64
Initial
Reading
>1 0,000
>1 0,000
>10,000
>1 0,000
>1 0,000
>10,000
>1 0,000
>1 0,000
>10,000
>1 0,000
>1 0,000
Date
Screened
03/05/79
03/05/79
03/05/79
03/05/79
03/05/79
03/05/79
03/05/79
03/05/79
03/05/79
03/06/79
03/06/79
Maintenance
Attempted
Yes
Yes
No
Yes
Yes
No
Yes
Yes
Yes
Yes
Yes
(gland)
Undirected Directed Maintenance Readings
Reading 123
2,000
4,000
»1 0,000 »1 0,000
NCb
>10,000 >10,000
>10,000 >10,000
NCb
900
NCb
3,000
1,000
>10,000 >10,000 7,000
Comments
Mistagged originally so no
initial repair attempted
tightened bolts needs
new packing
Leak reduced but needs new
packing
Near No. 361 needs new
packing
Was not leaking before
maintenance (mistagged)
Leak detected by soap
Qnln-Hnn miscpH hv-
instrument operator
-------
TABLE C-8. UNIT D ETHYLENE UNIT BLOCK VALVE REPAIRS (Concluded)
Tag
Number
315
311
316
313
312
314
0
Initial
Reading
>10,000
NCb
>10,000
>1 0,000
>1 0,000
>1 0,000
Date
Screened
03/06/79
03/06/79
03/06/79
03/06/79
03/06/79
03/06/79
Maintenance
Attempted
Yes
Yes
Yes
Yes
Yes
No
Undirected Directed Maintenance Readings
Reading 1 2 3
3,000
NCb
>10,000 2,000
>10,000 >10,000
>10,000 >10,000 5,000
>1 0,000
Comments
Drain still >10,000
All the
packing
All the
packing
Bad bol
replaci
way down on
way down on
ts need
ng
JA11 readings are in parts per million by volume calibrated to hexane using OVA-108 detector.
3NC = No change detected in reading above ambient level.
-------
Directed maintenance is preferable to undirected maintenance
for valve repair.
The information presented in Tables C-5, C-6, C-7, and C-8 has been
compiled with the objective of placing the data on as consistent a basis
as possible. However, some differences were unavoidable and others may
have gone unrecognized, due to the limited amount of information concerning
the details of methods used in each study. Therefore, care should be
exercised before attempting to draw specific quantitative conclusions based
on direct comparison of the results of these studies.
C.3. REFERENCES
1. Muller, Christopher, memo to files, U.S. Environmental Protection
Agency, Emission Standards and Engineering Division, Research Triangle
Park, N.C., January 18, 1979. (Plants A & B).
2. Muller, Christopher, memo to files, U.S. Environmental Protection
Agency, Emission Standards and Engineering Division, Research Triangle
Park, N.C., March 19, 1979. (Plant C).
3. Muller, Christopher, memo to files, U.S. Environmental Protection
Agency, Emission Standards and Engineering Division, Research Triangle
Park, N.C. (Plant D).
4. Hustvedt, K.C., trip report to James F. Durham, Chief, Petroleum
Section, U.S. Environmental Protection Agency, January 5, 1979 (Plants
E & F).
5. Wetherold, R.G., and L.P. Provost, Emission Factors and Frequency of
Leak Occurrence for Fittings in Refinery Process Units. EPA-600/2-79-044.
Radian Corporation, Austin, Texas, February 1979.
6. Valve Repair Summary and Memo from F.R. Bottomley, Union Oil Company,
Rodeo, Calfironia, to Milton Feldstein, Bay Area Quality Management District,
April 10, 1979.
7. Honerkamp, R.L., L.P. Provost, J.W. Sawyer, and R.G. Wetherold,
Valve Screening Study at Six San Francisco Bay Area Petroleum Refineries,
Final Report. Radian Corporation, Austin, Texas, February 6, 1979.
8. Valve Repair Summary and Memo from R.M. Thompson, Shell Oil Company,
Martinez Manufacturing Complex, Martinez, California. To Milton Feldstein,
Bay Area Quality Management District, April 26, 1979.
9. Radian Corporation, The Assessment of Environmental Emissions From Oil
Refining, Draft report, Appendix B, detailed results, EPA Contract No.
68-02-2147, Exhibit B., Austin, Texas, August 1979.
10. Equipment Summary from Phillips Petroleum Company, Sweeney, Texas,
March 14, 1979.
C-18
-------
APPENDIX D. EMISSION MEASUREMENT AND CONTINUOUS MONITORING
-------
APPENDIX D - EMISSION MEASUREMENT AND CONTINUOUS MONITORING
D.I EMISSION MEASUREMENT METHODS
To develop data in support of standards for the control of fugitive
emissions, EPA conducted leak surveys at six petroleum refineries and
three synthetic organic chemical manufacturing plants. The resulting
leak determination procedures contained in Reference Method 21 were
developed during the course of this test program.
Prior to the first test, available methods for measurement of
fugitive leaks were reviewed, with emphasis on methods that would provide
data on emission rates from each source. To measure emission rates,
each individual piece of equipment must be enclosed in a temporary cover
for emission containment. After containment, the leak rate can be
determined using concentration change and flow measurements. This
(1 2 T
procedure has been used in several studies, ' ' and has been demonstrated
to be a feasible method for research purposes. It was not selected for
this study because direct measurement of emission rates from leaks is a
time-consuming and expensive procedure, and is not feasible or practical
for routine testing.
Procedures that yield qualitative or semi-quantitative indications
of leak rates were then reviewed. There are essentially two alternatives:
leak detection by spraying each component leak source with a soap solution
and observing whether or not bubbles were formed; and, the use of a
portable analyzer to survey for the presence of increased organic compound
concentration in the vicinity of a leak source. Visual, audible, or
olefactory inspections are too subjective to be used as indicators of
leakage in these applications. The use of a portable analyzer was selected
as a basis for the method because it would have been difficult to establish
a leak definition based on bubble formation rates. Also, the temperature
of the component, physical configuration, and relative movement of parts
often interfere with bubble formation.
D-l
-------
Once the basic detection principle was selected, it was then necessary
to define the procedures for use of the portable analyzer. Prior to
performance of the first field test, a procedure was reported that
(3]
conducted surveys at a distance of 5 cm from the components. ' This
information was used to formulate the test plan for initial testing. '
In addition, measurements were made at distances of 25 cm and 40 cm on
three perpendicular lines around individual sources. Of the three
distances, the most repeatable indicator of the presence of a leak was a
measurement at 5 cm, with a leak definition concentration of 100 or 1000
ppmv. The localized meteorological conditions affected dispersion
significantly at greater distances. Also, it was more difficult to
define a leak at greater distances because of the small changes from
ambient concentrations observed. Surveys were conducted at 5 cm from
the source during the next three facility tests.
The procedure was distributed for comment in a draft control techniques
guideline documents. Many commentors felt that a measurement distance
of 5 cm could not be accurately repeated during screening tests. Since
the concentration profile is rapfdly changing Between 0 and about 10 cm
from the source, a small variance from 5 cm could significantly effect
the concentration measurement. In response to these comments, the
procedures were changed so that measurements were made at the surface of
the interface, or essentially 0 cm. Additional testing at two refineries
and three chemical plants was performed by measuring volatile organic
concentrations at the interface surface.
A complication that this change introduces is that a very small
mass emission rate leak C'pin-hole leak") can Be totally captured by the
instrument and a high concentration result will be obtained. This has
occurred occasionally in EPA tests and a solution to this problem has
not been found.
The calibration basis for the analyzer was evaluated. It was
recognized that there are a number of potential vapor stream compositions
that can be expected. Since all analyzer types do not respond equally
to different compounds, i~t was necessary to establish a reference
calibration material. Based on the expected compounds and the limited
D-2
-------
information available on instrument response factors, hexane was chosen
as the reference calibration gas for EPA test programs. At the 5 cm
measurement distance, calibrations were conducted at approximately 100
or 1000 ppmv levels. After the measurement distance was changed,
calibrations at 10,000 ppmv levels were required. Commentors pointed
out that hexane standards at this concentration were not readily available
commercially. Consequently, modifications were incorporated in the
method to allow alternate standard preparation procedures or alternate
calibration gases.
The alternative of specifying a different calibration material for
each type stream and normalization factors for each instrument type was
not intensively investigated. There are at least four instrument types
available that might Be used in this procedure, and there are a large
number of potential stream compositions possible. The amount of prior
knowledge necessary to develop and subsequently use such factors would
make the method prohibitively complicated. Based on EPA test results,
the number of concentration measurements in the range where a variability
of 2 or 3 would change the decision as to whether or not a leak exists
is small in comparison to the total number of potential leak sources.
An alternative approach to leak detection was evaluated by EPA
during field testing. The approach used was an area survey, or walk-
through, using a portable analyzer. The unit area was surveyed by
walking through the unft positioning the instrument probe within 1 meter
of all valves and pumps. The concentration readings were recorded on a
portable strip chart recorder. After completion of the walkthrough, the
local wind conditions were used'with the chart data to locate the
approximate source of any increased ambient concentrations. This
procedure was found to yield mixed results. In some cases, the majority
of leaks located by individual component testing could be located by
walkthrough surveys. In other tests, prevailing dispersion conditions
and local elevated ambient concentrations complicated or prevented the
interpretation of the results. Additionally, it was not pqs.s.tb.le to
D-3
-------
develop a general criteria specifying how much of an ambient increase at
a distance of 1 meter is indicative of a 10000 ppm concentration at the
leak source. Because of the potential variability in results from site
to site, routine walkthrough surveys were not selected as a reference or
alternate test procedure.
D.2 CONTINUOUS MONITORING SYSTEMS AND DEVICES
Since the leak determination procedure is not a typical emission
measurement technique, there are no continuous monitoring approaches
that are directly applicable. Continual surveillance is achieved by
repeated monitoring or screening of all affected potential leak sources.
A continuous monitoring system or device could serve as an indicator
that a leak has developed between inspection intervals. EPA performed a
limited evaluation of fixed-point monitoring systems for their effective-
ness in leak detection. The systems consisted of both remote sensing
devices with a central readout and a central analyzer system (.gas chromatograch)
with remotely collected samples. The results of these tests indicated
that fixed point systems were not capable of sensing all leaks that were
found by individual component testing. This is to be expected since
these systems are significantly affected by local dispersion conditions
and would require either many individual point locations, or very low
detection sensitivities in order to achieve similar results to those
obtained using an individual component survey.
It is recommended that fixed-point monitoring systems not be required
since general specifications cannot be formulated to assure equivalent
results, and each installation would have to be evaluated individually.
D.3 PERFORMANCE TEST METHOD
The recommended fugitive VOC emission detection procedure is Method 21
This method incorporates the use of a portable analyzer to detect the
presence of volatile organic vapors at the surface of the interface
where direct leakage to the atmosphere could occur. The general approach
of this technique assumes that if an organic leak exists, there will be
an increased vapor concentration in the vicinity of the leak, and that
the measured concentration is generally proportional to the mass emission
rate of the organic compound.
D-4
-------
Method 21 is designed for use in many different source categories
and does not include the specification of a specific compound in instrument
calibration or a leak definition in terms of VOC concentration. These
criteria are given in the applicable standard.
There are at least four types of detection principles currently
available in commercial portable instruments. These are flame ioniza-
tion, catalytic oxidation, infrared absorption (NDIR) and photoionization.
Two types (flame ionization and catalytic oxidation) are known to be
available in factory mutual certified versions for use in hazardous
atmospheres.
The recommended test procedure includes a set of design and operating
specifications and evaluation procedures by which an analyzer's performance
can be evaluated. These parameters were selected based on the allowable
tolerances for data collection, and not on the performance of individual
instruments. Based on manufacturers' literature specifications, many
commercially available analyzers can meet these requirements.
The estimated purchase cost for an analyzer ranges from about $1000
to $5000 depending on the type and optional equipment. The cost of an
annual monitoring program per unit, including semiannual instrument
tests and reporting is estimated to be from $3,000 to $4,500. This
estimate is based on EPA contractor costs experienced during previous
test programs. Performance of monitoring by plant personnel may result
in lower costs. The above estimates do not include any costs associated
with leak repair after detection.
D.4 REFERENCES
1. "JOINT DISTRICT, FEDERAL, and STATE PROJECT for the Evaluation
of Refinery Emissions", Los Angeles County Air Pollution Control District,
Nine Reports, 1957-1958.
D-5
-------
2. "Emission Factors and Frequency of Leak Occurrence Fittings in
Refinery Process Units" Radian Corporation Contract No. 68-02-2147 and
No. 68-02-2665, EPA Report No. 600/2-79-044, February 1979.
3. Telephone Communication: Paul Harrison, Meteorology Research,
Inc. to K. C. Hustvedt, EPA, December 22, 1977.
4. EMB Report No. 77-CAT-6, "Miscellaneous Refinery Equipment VOC
Sources at Arco, Watson Refinery and Newhall Refining Co." ESED, EPA,
December, 1979.
5. "Control of Volatile Organic Compound Leaks from Petroleum
Refinery Equipment," OAQPS Guideline Series, EPA-45Q/2-78-036, June, 1978.
D-6
-------
APPENDIX £. METHODOLOGY FOR ECONOMIC ANALYSIS
-------
APPENDIX El: WEIGHTING PROCEDURES FOR THE ESTIMATION OF SOCMI
TIME-SERIES DATA
The chemicals produced by the synthetic organic chemicals manufacturing
industry (SOCMI) do not directly correspond to the industrial classifications
of organic chemicals used by sources reporting production and sales statis-
tics. Consequently, the weighting procedure described below was used to gen-
erate data that reflect the SOCMI as accurately as possible.
Production and sales data for synthetic organic chemicals are reported
annually in the United States International Trade Commission (ITC) report,
Synthetic Organic Chemicals: U.S. Production and Sales. The report presents
production, quantity of sales and value of sales data for 14 categories of
chemicals. Four of these categories, Tar and Tar Crudes, Primary Products
from Petroleum and Natural Gas for Chemical Conversion, Cyclic Intermediates,
and Miscellaneous Cyclic and Acyclic Chemicals, contain SOCMI chemicals, as
well as other chemicals. To derive appropriate estimates of data for the
SOCMI chemicals, production, quantity of sales and value of sales for 1977
were estimated for all SOCMI chemicals included in each of the four categor-
ies. The estimates for SOCMI chemicals for each variable were then divided
by the aggregate estimates for all chemicals within the category. Resulting
ratios were used as weights to calculate estimates of production, quantity of
sales and value of sales of SOCMI chemicals each in category over the period
1974-1978.
Prior to 1975, the chemicals included in the category, Miscellaneous
Cyclic and Acyclic Chemicals, were reported as Miscellaneous Synthetic Organ-
ic Chemicals. A weighting scheme based on 1974 data for this category was
developed using the procedure described above and was used to estimate pro-
E-l
-------
duction and sales of SOCMI chemicals in this category for the period 1968-
1974. Data on production and sales of SOCMI for the remaining three categor-
ies for the period 1968-1974 were estimated using the 1977 weights.
Table El-1 presents the SOCMI chemicals included in each of the ITC
categories. Table El-2 presents the estimated ratios used to weight the ITC
data in order to calculate production, quantity of sales and value of sales
of SOCMI chemicals.
E-2
-------
TABLE £1-1. SOCMI CHEMICALS INCLUDED IN ITC CATEGORY RATIOS
Tar and Tar Crudes*
Benzene
Toluene
Xylene
Solvent Naptha
Primary Products from Petroleum and Natural Gas for.Chemical Conversion
Benzene
Cumene
Cyclohexane
Ethyl benzene
Napthalene
Styrene
Toluene
Xylenes
o-Xylene
p-Xylene
All other aromatics and napthenes
Acetylene
Cyclic Intermediates
Aniline
Benzoic acid
Biphenyl
Cresols
Cresylic acid, refined
Cyclohexanone
Cylohexylamine
Miscellaneous Cyclic and Acyclic Chemicals
Benzyl alcohol
Caprolactam
Dioxane
p-Hydroxybenzoic acid
Maleic anhydride
Ethanolamines
Acenitrile
Acrylonitrile
Acetic acid
Acetic anhydride
Acrylic acid
Adipic acid
Fumaric acid
Propionic acid
Formaldehyde
Isobutyraldehyde
Acetone
Ethylene
Propylene
Butadiene and butylene fractions
1, 3-Butadiene, grade for rubber
1-Butene
Isobutylene
Isoprene
Dodecene
Petenes
Nonene
Polybutene
o-Dichlorobenzene
p-Dichlorobenzene
Hydroquinone
a-Methylstyrene
Nitrobenzene
Nonylphenol
Phenol
Butyl amines
Ethyl amines
Isopropylamirie, mono-
Methyl amines
All other amines
Pentaerythritol
Propylene glycol
All other polyhydric alcohols
2-(2-Butoxyethoxy) ethanol (Diethy-
lene glycol monobutyl ether)
2-[2-(2-Butoxyethoxy) ethoxy]
ethanol (Triethylene glycol,
monobutyl ether)
Diethylene glycol
Dipropylene glycol
2-(2-Ethoxyethoxy) ethanol (Diethy-
lene glycol monoethyl ether)
E-3
(continued)
-------
TABLE El-1. (continued)
Miscellaneous Cylic and Acyclic Chemicals (continued)
2-Butanone (methyl ethyl ketone)
4-Hydroxy-4-methyl-2-pentanone
(Diacetone alcohol)
4-Methyl-2-pentanone (methyl
isobutyl ketone)
4-methyl-3-penten-2-one (Mesityl
oxide)
All other ketones
n-Butyl alcohol (n-propylcarbinol)
Methanol
n-Butyl acetate
Ethyl acetate
Ethyl aerylate
Isobutyl acetate
Methyl acetate
Methyl methacrylate
Vinyl 'acetate
Ethylene glycol
Glycerol, synthetic
1,1,1-Trichloroethane (methyl
chloroform)
Trichloroethylene
Vinyl chloride, monomer (chloro-
ethylene)
All other chlorinated hydrocar-
bons
Chiorodi f1uoromethane
Dichlorodi.fi uoromethane
2-Methoxyethanol (ethylene glycol
monomethyl ether)
2-(2-Methoxyethoxy) ethanol (Diethy-
lene glycol monomethyl ether)
Polyethylene glycol
Polypropylene glycol
Propylene glycol
Triethylene glycol
Carbon tetrachloride
Chloroethane (ethyl chloride)
Chloroform
Chloromethane (methyl chloride)
1-2-Dichloroethane (ethylene
dichloride)
Dichloromethane (methylene chloride)
1-2-Dichloropropane (propylene di-
chloride)
Tetrachloroethylene (perchloro-
ethylene)
Trichlorofluoromethane
Carbon disulfide
Ethylene oxide
Ethyl ether
Propylene oxide
All other epoxides, ethers, and
acetals
Phosgene (carbonyl chloride)
^Derived from coal. Does not duplicate data contained in other categories.
E-4
-------
TABLE El-2. RATIOS USED TO WEIGHT ITC DATA3
Production
(1000 Ibs)
Quantity
(1000 Ibs)
Value
(1000 $)
Tar and Crudes
Total SOCMI Chemicals 3,265,976 1,543,188 976,660
ITC Grand Total 4,145,815 2,009,737 1,104,285
Ratio . 78.78% 76.79% 88.44%
Primary Products from
Petroleum and Natural Gas
Total SOCMI Chemicals 93,517,108 38,873,142 4,120,327
ITC Grand Total 126,133,316 61,008,376 5,820,390
Ratio ' . . 74.14% 63.72% 70.79%
Cyclic Intermediates
Total SOCMI Chemicals 15,699,616 6,139,015 1,878,235
ITC Grand Total 18,725,626 7,985,790 2,596,627
Ratio 83.84% 76.84% 72.33%
Miscellaneous Cyclic
and Acyclic Chemicals
Total SOCMI Chemicals 65,876,154 27,695,411 4,734,676
ITC Grand Total 86,968,069 38,753,311 7,919,082
Ratio 75.75% 71.47% 59.79%
Miscellaneous Chemicals
1974 Figures
Total SOCMI Chemicals 73,670,360 34,817,621 3,607,825
ITC Grand Total 100,604,375 47,430,967 7,815,487
Ratio 73.23% 73.41^ 46.16%
1977 figures except where indicated.
E-5
-------
APPENDIX E2: REPLACEMENT INVESTMENT PROJECTIONS
The methodology used to project SOCMI replacement investment is de-
scribed in this appendix. The projections are based on two key theoretical
assumptions: (I) the historical growth rate of capacity, p, has been con-
stant over time; and (II) model units have a fixed life of L years. These
assumptions are summarized in the following equations:
KT = (l+p)L KT_L (1)
IT = PKT + RT (2)
RT = IT_L (3)
where
K = industry capacity,
I = gross investment,
R = replacement investment,
T = time subscript,
and K, I and R are measured in terms of model units.
Equation (1) is an algebraic restatement of assumption I. Equation (2)
is simply a mathematical definition of gross investment, that is, gross
investment, I-,-, is equal to additions to new capacity, pKT, plus replacement
investment, R-.-. Equation (3) is an algebraic restatement of assumption II.
Appropriately lagging Equation (2) and back substituting from (2) into (3),
it can be shown that
RT = p ^ KT_iL (4)
Further, by substituting for the various KT .. in Equation (4) using Equation
(1), and rearranging terms, the following result is obtained:
E-6
-------
RT = KT-I Z P (5)
i I
1=0 (1+p)1
oo
The expression I p is a constant and, if p is assumed to be 0.06 and
i=0
L to be 20, approximately equal to 0.087. Equation (5) can be used to pro-
ject replacement investment in any year, T, if an estimate of the capital
stock in the (T-L)th year is available. For SOCMI, capital stock data are
available for 1976. This information, together with an assumed historical
growth rate of 6 percent, was used to estimate the capital stock for the
years 1961 to 1965 by means of Equation (1). The resulting capital stock
estimates are then used in Equation (5) to project replacement investment in
SOCMI for each of the five years following proposal of any regulatory alter-
natives (1981-1985), on the basis of the empirical assumption that each model
unit has a life of 20 years. The annual projections of replacement invest-
ment are then summed to obtain a projection of the number of replacement
facilities subject to the provisions of any regulatory alternative in the
fifth year following its proposal. The projections of replacement investment
obtained by applying this methodology are presented in Table E2-1.
E-7
-------
TABLE E2-1 PROJECTIONS OF REPLACEMENT INVESTMENT
Number of replacement capacity units
Year Annual Cumulative
1981 49 49
1982 . 51 100
1983 55 155
1984 58 213
1985 61 274
E-8
-------
APPENDIX E3: METHODOLOGY FOR COMPUTING COST OF CAPITAL
TO SYNTHETIC ORGANIC CHEMICAL MANUFACTURERS
This appendix describes the process used to estimate the cost of capital
for the chemical industry. The cost of capital for any new project is the
cost of equity, debt, and preferred stock, weighted by the percentage of
funds generated by each type of financing, that is,
k = k - + k - + k - (I)
kc el Ki I kp I U;
where
k =. cost of capital
k = cost of equity capital
k. = cost of debt capital
k s cost of preferred stock capital
E = the amount of equity used to finance a given investment
D = the amount of debt used to finance a given investment
P = the amount of preferred stock used to finance a given
investment
I '= the total funds needed for the investment
The first step in estimating Equation (1) is to determine the relevant
weights for the three types of financing. It is assumed that the proportion
of debt, equity, and preferred stock, to be used on any new project will be
the same as currently exists in the firm's capital structure. This implies
that the firm is currently using the optimal mix of financing. Figures for
the three types of funds came from the COMPUSTAT tapes, supplied by Standard
& Poor's Corporation, for each firm's fiscal year ending in 1977. Common
equity included the par value of common stock, retained earnings, capital
E-9
-------
surplus, self-insurance reserves, and capital premium, while debt included
all obligations due more than a year from the company's balance sheet date.
Preferred stock represented the net number of preferred shares outstanding at
year-end multiplied by the involuntary liquidating value per share.
The next step in calculating Equation (1) is to estimate the cost of
equity financing. Two approaches are commonly used: the results derived from
the capital-asset pricing model (CAPM) and the results derived from the
dividend capitalization model (DCM). The CAPM examines the necessary returns
on a firm's stock in relation to a portfolio comprised of all existing
stocks, while the DCM evaluates the stream of dividends and the discount rate
needed to arrive at the firm's existing share price. The required return on
equity using the CAPM is:
ke = i + p (km-i) (2)
where
i = the expected risk free interest rate
k -i = the expected excess return on the market, and
P = the firm's beta coefficient.
The required return on equity using the DCM is:
Di
ke = r, *« (3)
where
D, = the dividend expected in period 1
P = the share price at the beginning of period 1
g = the expected rate of dividend growth, assumed to be constant.
Figures for Equation (2) were developed in the following manner. The
expected risk-free rate was assumed equal to the yield on a 3-month Treasury
Bill, as reported in the October 1, 1979, Wall Street Journal. The current
yield was 10.46 percent. This corresponds to the yield from a bond with no
E-10
-------
possibility of default and offering no chance of a capital loss and is
therefore riskless. The firm's beta coefficients came from the September 24,
1979, Value Line.Investment Survey. The expected excess return equalled
2.9646 percent, the 5-year average (July 1974-June 1979) of the monthly
excess returns on the Standard & Poor's .500 Stock Index multiplied by twelve.
Figures for Equation (3) came from two sources. Both share price and
expected yearly dividends came from figures reported in the October 1, 1979,
Wall Street Journal. The growth rate was calculated from data contained on
the COMPUSTAT tapes. Three different growth rates were tried: the 5-year
average growth of total assets, the 5-year average growth of per share
earnings, and the 5-year average growth of dividends.
A number of theoretical reasons exist for preferring the CAPM approach
to the DCM for estimating the required return on equity, but the figures
calculated revealed a more practical justification. Using growth estimated
from per share earnings or dividends' resulted in a number of firms having
negative required returns with the DCM method. Although using the growth in
assets resulted in only one firm with a negative required return, several
firms had extremely low returns (less than 10 percent). It is unreasonable
to expect that stockholders would demand a return on their stock that is less
than the existing yield on Treasury Bills, yet all three variants of the DCM
method led to this conclusion for a number of firms. From these
considerations it was decided to use the CAPM calculations as the required
return on equity.
The third step in estimating Equation (1) is calculating the cost of
debt financing. This would be a relatively easy estimation if interest rates
did not change over time. Past yields on old issues of bonds would suffice.
Since interest rates have been increasing, it was felt that a more forward-
E-ll
-------
looking rate was required. The method selected was to take the average yield
as given in the October 1 - September 3, 1979, Moody's Bond Survey for the
firm's bond ratings class as the necessary yield the firm must offer on
long-term debt. The firm's ratings class came from the September 1979
Moody's Bond Record or the 1979 Moody's Industrial Manual. A small number of
firms were not rated by Moody's. One firm was ranked in Standard and Poor's
Bond Guide and this was used to approximate a Moody's bond class.
Information on other firms was contained in the 1979 Moody's Industrial Manual
or the Standard & Poor's Corporation Records, concerning bank notes,
revolving credit, or term-loan agreements that tied the interest rate on
these types of debt to the current prime rate. This was used as the
necessary yield on long-term debt. Table E3-1 presents the yields by ratings
class and the prime rate (as of October 1, 1979) used for the cost of debt
funds.
TABLE E3-1. YIELDS BY RATING CLASS FOR COST OF DEBT FUNDS, 1979
(prime rate = 13.50 %)
Ratings Class Yield (percent)
AAA 9.25
AA 9.59
A 9.72
BAA 10.38
BA 11.97
B 12.395
The yield on long-term debt does not represent the aftertax cost of debt
financing since interest charges are tax deductable. To arrive at the after-
E-12
-------
tax cost, the yield must be multiplied by one minus the marginal tax rate.
k1 = k(l - t)
where
k = the yield on bonds
t = the marginal tax rate
It is assumed that the firms in the sample are profitable, so that taxes must
be paid, and that their marginal tax rate is 48 percent.
The last step in estimating Equation (1) is to arrive at the cost of
preferred stock financing. Unlike debt, preferred stock does not have a
maturity date, so that the current yield should approximate the yield on new
issues. The yield is:
V;
where
D = stated annual dividend
P = the price of a share of preferred stock
The figures for dividends and share price came from the October 1, 1979, Wall
Street Journal or, if not included in this source, from the January 1, 1979,
listing in the Daily Stock Price Record. A number of firms did not have
their preferred stock listed in either source, yet had preferred stock in
their capital structures. All used less than 15 percent preferred stock,
with the majority less than 5 percent. For these firms the yield on
preferred stock was set equal to the yield on long-term debt.
Table E3-2 lists the cost of capital for all 100 firms in the. sample,
along with some of the components of Equation (1). These firms represent
the best available sample of the approximately 600 firms in the industry.
However, it is likely that on the average they are larger than the firms
E-13
-------
TABLE E3-2. FINANCIAL DATA FOR 100 FIRMS IN SOCMI
1-11
Name
Abbott Labs
Akzona
Alco Standard Corp.
Allied Chem Corp.
American Cyanamid
Armco Steel Corp.
Atlantic Richfield
Beatrice Foods
Bendix Corp.
Bethlehem Steel Corp.
Borden Inc.
Borg-Warner Chem.
Brown Co.
CPC International
Inc.
Celanese Corp.
Charter International
Oil
Cities Service Co.
Combustion
Engineering
Continental Oil
Crompton & Knowles
Dart Indust.
Dayco Corp.
De Soto, Inc.
Diamond Shamrock
Corp.
Dow Chemical
Du Pont De Nemours
Eastern Gas & Fuel
Associates
Essex Chem. Corp.
Cost of
Capital
12.014
10.276
12.151
10.091
11.083
10.588
9.749
11.232
11.118
10.913
10.484
11.863
9.813
11.638
10.181
9.175
10.395
11.494
10.881
11.298
10.689
8.270
11.499
9.790
10.060
11.328
11.605
12.502
Return
On
Equity
14.018
13.276
13.425
13.721
13.425
13.276
13.128
12.832
13.425
14.018
12.683
13.128
12.387
13.128
13.128
14.166
12.980
14.314
13.721
13.425
14.166
12.980
13.128
13.721
14.018
13.573
14.018
14.166
Return
On
Debt
9.590
10.380
15.120
9.720
9.590
9.720
9.590
9.250
9.720
9.720
9.590
9.720
12.395
9.590
11.970
12.395
9.720
9.720
9.590
14.450
9.720
11.970
13.750
9.720
9.590
9.250
14. 180
12.395
Return
On'
Preferred
Stock
__a
--
--
--
--
6.461
--
7.429
3.333
--
--
--
.
--
10.084
--
--
--
2.564
--
4.211
6.071
--
--
--
8.654
--
--
Proportion
Of
Equity
.77262
. 61914
. 64134
.58118
.72252
. 66880
.51602
.79803
.72911
.65360
.71317
.82756
.56680
.81691
.53511
.27557
.67388
.68700
.67568
.53329
.63113
.30351
. 72746
.54639
.56176
.72512
.63681
. 78453
Proportion
Of
Debt
.216575
. 380859
.259343
. 418825
.277480
. 306858
.362174
.194329
. 248140
. 346402
.285155
.145263
.433202
. 183087
.396896 '
.623167
.326120
.296229
.321308
.375634
.231645
.666445
.272535
.453615
.438236
.232172
.363188
.215465
Proportion
Of
Preferred
Stock
. 010804
.000
.099317
.000
.000
.024337
.121802
.007644
.022754
.000
.001677
.027181
.000
.000
.067997
.101265
.000
.016774
.003009
.091078
.137221
. 030044
.000
.000
.000
.042712
.000
.000
-------
TABLE E3-2 (Continued)
Name
Exxon Corp.
FMC Corp.
Ferro Corp.
Firestone Tire &
Rubber
Ford Motor Co.
GAF Corp.
General Electric Co.
General Motors Corp.
General Tire & Rubber
Georgia-Pacific Corp.
Goodrich (B.F.) Co.
Goodyear Tire &
Rubber Co.
Gulf Oil Corp.
Hercules Inc.
Inland Steel
Insilco Corp.
Interlake, Inc.
International
Harvester
Kaiser Steel Corp.
Kraft Inc.
Marathon Oil Co.
Martin Marietta Chem.
Mead Corp.
Merck & Co.
Minnesota Mining &
Manuf .
Mobil Oil Corp.
Monsanto Co.
Morton-Norwich
Products
Cost of
Capital
11.875
10.183
12.369
10.610
12.069
9.398
12.130
12.798
11.440
10.793
10.430
10.101
11.745
11.177
10.092
9.339
11.331
10. 534
11.688
10.774
9.582
11.238
10.000
12.309
12.572
10.868
10.970
10.726
Return
On
Equity
13.276
13.573
13.276
12.980
13.276
13.573
13.721
13.425-
13.276
13.573
13.276
12.980
12.980
13.869
12.980
13.276
13.128
13.573
14.018
12.683
13.128
13.276
13.869
13.573
13.869
13.128
13.573
13.721
Return
On
Debt
9.250
9.720
9.720
9.720
9.250
10.380
9.250
9.250
11.970
9.590
10.380
9.720
9.250
9.720
9.590
11.970
9.720
9.720
14.000
9.250
9.720
9.720
9.720
9.250
9.250
9.250
9.590
9.720
Return
On
Preferred
Stock
_ _
6.250
--
--
--
7.559
--
8.715
--
8.864
--
--
--
7.752
--
--
--
--
--
--
4.308
--
--
--
5.000
--
Proportion
Of
Equity
.83450
.59257
.88968 .
.70096
.85743
. 44490
. 82148
.91962
.73287
.67625
.62957
.63679
. 84880
.69461
.62702
. 41885
.77736
.63297
.63274
.75752
.56074
.75212
.56423
.85481
.85677
.72833
.69690
.65441
Proportion.
Of
Debt
.165504
.339730
.110317
.299038
.142565
.387035
.178521
.063516
.258968
.323751
.349707
.363210
.151203
.305394
.352735
.475634
.222640
.348230
.345717
. 242479
.439257
. 247882
. 398718
.143358
.143235
.271665
.300335
.345589
Proportion
Of
Preferred
Stock
.000
.067701
.000
.000 .
.000
.168061
,000
.016862
.008163
.000
.020723
.000
.000
.000
.020249
.105511
.000
.018796
.021539
.000
.000
.000
.037048
.001827
.000
.000
..002767
.000
-------
TABLE E3-2 (Continued)
Name
National Distillers
& Chem.
National Steel Corp.
Northwest Indust.
Owens-Corning
Fiberglass
PPG Industries
Penwalt Corp.
Pfizer
Phillips Petroleum Co.
Procter & Gamble Co.
Quaker Oats Co.
Reeves Bros. Inc.
Reichold Chems.
Republic Steel Corp.
Riegel Textile Corp.
Rockwell International
Rohn and Haas Co.
SCM Corp.
Scott Paper Co.
Shakespeare Co.
Sherwin-Williams Co.
Squibb Corp.
A. E. Staley Mfg. Co.
Stauffer Chemical Co.
Sterling Drug
Sun Chem. Corp.
Sybron Corp.
Tenneco Inc.
Texaco
Texfi Indust.
Textron Inc.
Union Camp Corp.
Union Carbide Corp.
Cost of
Capital
11.037
9.909
8.015
11.653
10.596
9.013
11.244
11.670
11.824
10.946
10.629
10.647
11.305
11.201
9.589
10.739
10.835
10.784
11.229
9.617
11.266
10.428
10.188
12.595
10.427
10.786
9.155
11-230
10.090
10.085
11.359
10.775
Return
On
Equity
13.128
12.683
13.869
13.425
13.276
13.276
14.018
13.721
13.276
13.573
12.535
13.425
13.425
12.980
12.535
13.721
14.018
13.721
13.276
12.980
14.018
13.573
13.425
13.276
13.573
13.869
12.980
12.980
13.275
13.425
13.276
13.573
Return
On
Debt
9.720
9.590
10.380
9.720
9.590
9.720
9.590
9.250
9.250
9.720
10.380
10.380
9.720
11.970
9.720
9.720
10.380
9.590
14.000
10.380
9.590
9.720
9.720
9.590
12.395
9.720
10.380
9.250
16.000
9.720
9.590
9.590
Return
On
Preferred
Stock
9.193
--
2.9412
--
.
7.529
--
--
--
9.008
--
--
--
--
5.398
'
--
--
--
10.00
--
--
--
--
--
--
3.887
--
--
6.222
--
--
Proportion
Of
Equity
.73310
.63946
.32561
. 78828
.67661
.41712
.69289
. 76982
.82842
.651578
. 732870
.571986
. 746819
.736598
.602132
.655939
.630766
.660791
.658505
. 523981
.695345
. 629947
.613351
. 917816
. 558689
.616191
. 505890
.785863
.356904
.577353
. 768639
.674170
Proportion
Of
Debt
.251565
.360538
.617085
.211721
.323394
.369200
.307113
.230179
. 171428
.262094
.267130
.295871
.253181
.263402
.309032
. 344061
.369234
.333680
. 341495
.422439
.304655
. 368508
. 386649
. 082184
.441311
.319517
.442129
.214137
.643096
.252757
.231361
.325830
Proportion
Of
Preferred
Stock
.015334
.000
.057301
.000
.000
.213675
.000
.000
.000153
.086328
.000
. 132143
.000
.000
. 088836
.000
.000
.005529
.000
.053579
.000
. 001544
.000
.000
.000
.064292
.051981
.000
.000
.169890
.000
.000
-------
TABLE E3-2 (Continued)
m
Name
Union Oil , Calif.
Uni royal
U.S. Gypsum
U.S. Steel Corp.
Upjohn Co.
Vulcan Materials Co.
Walter (Jim) Corp.
Westinghouse Electric
Corp.
.Weyerhaeuser Co.
Wheeling-Pittsburgh
Steel
Whittaker Corp.
Wit Chem. Corp.
Cost of
Capital
10.577
10.514
10.726
10.919
11.052
10.675
9.019
12.596
10.402
11.238
10.070
10.736
Return
On
Equity
13.128
13.425
13.276
13.573
13.573
12.980
13.721
14.018
14.166
13.869
14.314
13.573
Return
On
Debt
9.590
11.970
9.590
9.590
9.590
9.720
11.970
9.720
. 9.590
14.000
11.970
9.720
Return
On
Preferred
Stock
16.000
5.539
--
--
--
4.444
8.837
.5.957.
12.739
--
3.313
Proportion
Of
Equity
.663994
.521603
.686341
. .690912
.706383
.709218
. 398726
.838775
.583685
.512893
.457808
.673790
Proportion
Of
Debt
.295934
.423786
.223477
.309088
.293617
.290782
.491966
.155115
.357341
.381136
.517470
.292825
Proportion
Of
Preferred
Stock
. 040072.
.054611
.090182
.000
.000
.000
.109308
.006110
..058973
.105972
.024722
.033385
Dashes indicated that data are unavailable.
-------
excluded, as many small firms do not have to publish detailed financial
records. This potential bias in the sample of firms used may have resulted
in a slight underestimation of the industry's cost of capital.12
E-18
-------
APPENDIX E4: METHODOLOGICAL CONSIDERATIONS
Price and Rate of Return Impacts
Let P denote product price, Q denote unit output, TOC denote total
operating costs, K denote the amount of capital invested in the unit, r
denote the rate of return on capital and t denote the tax rate in a given
year. The aftertax rate of return on capital invested in the unit may then
be defined as:
= (1-t) (PQ - TOC)
K
where [PQ - TOC] is the unit's pretax net revenues from its operations in
that year. Now, assume that the unit is required to change its operating
costs and level of capital investment in order to comply with the implemen-
tation of some regulatory alternative. Under the full cost absorption sce-
narios the unit will be unable to adjust the the price of its product or unit
output. Consequently, the rate of return on investment, r, will change. The
formula used to estimate this impact is obtained by totally differentiating
Equation (1) with respect to TOC and K; that is,
. _ (1-t) dTOC (1-t) (PO-TOC) dK
or - +
Substituting in (2) from (1) and rearranging terms, it follows that:
-dr = (1-t) dTOC + rdK (3)
K
Equation (3), identical to Equation (2) in section 9.2, is the formula used
to calculate the full cost absorption rate of return impacts presented in
Chapter 9.
Price impacts are estimated on the basis of the assumption that firms
will be able to maintain the preregulation rate of return (r) by increasing
E-19
-------
product prices. Thus, r is now a constant and P a variable. Rearranging
terms in Equation (1), it may be shown that:
p = TOC + r K/(l-t) (4)
In full cost pass through scenarios, changes in TOC and K leave r and Q
unaffected but result in a change in P. The formula for estimating this
change in P may be obtained by total differentiating Equation (4) with
respect to TOC and K; that is,
dp = dTOC + rdK /(1-t) (5)
Equation (5), identical to Equation (1) in Section 9.2, is the formula used
to estimate the full cost pass through price impacts presented in Chapter 9.
E-20
-------
APPENDIX E REFERENCES
1. COMPUSTAT. New York:. Standard & Poor1 s Corporation, 1978.
2. Daily Stock Price Record. New York: Standard & Poor's Corporation,
1979. .
3. Moody's Bond Record. New York: Mo.ody's. Investors Service, Inc., Sep-
tember 1979.
4. Moody's Bond Survey. New York: Moody's Investors Service, Inc., Octo-
ber 1 - September 3, 1979.
b. Moody's Industrial Manual. New York: Moody's Investors Service, Inc.,
1979.
6. Scherer, F. M. , et al. The Economics of Multi-Plant Operation. Cam-
bridge, Mass.: Harvard University Press, 1975.
7. Standard & Poor's Bond Guide. . New York: Standard & Poor's Corporation,
September 1979.
8. Standard & Poor's Corporation Records. New York: Standard & Poor's
Corporation, September 1979.
9. Standard & Poor's Statistical Service. New York: .Standard & Poor's
Corporation, October 1979.
10. Value Line Investment Survey. New York: Arnold Bernhard & Co., Inc.,
September 24, 1979. . ,
11. The Wall- Street Journal. New York: Dow Jones & Company, October 1,
1979.
E-21
-------
APPENDIX F - SYNTHETIC ORGANIC CHEMICALS MANUFACTURING INDUSTRY
F-l
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
OCPDB No.*
20
30
40
50
65
70
80
90
100
110
120
125
130
140
150
160
170
180
185
190
200
210
220
Chemical
Acetal
Acetaldehyde
Acetaldol
Acctamide
Acetanilide
Acetic acid
Acetic anhydride
Acetone
Acetone cyanohydrin
Acetonitrile
Acetophenone
Acetyl chloride
Acetylene
Acrolein
Acryl amide
Acrylic acid and esters
Acryl oni trile
Adipic acid
Adi poni trile
Alkyl naphthalenes
Ally! alcohol
Allyl chloride
Aminobenzoic acid
*The OCPDB Numbers are reference indices assigned to the various chemicals
in the Organic Chemical Producers Data Base developed by EPA.
F-2
-------
OCPDB No._ Chronical
24 230 Aiiiiiioethylethanolamine
25 235 p-nminophcnol
26 240 Amyl acetates
27 250 Amyl alcohols
28 260 .Amyl amine
29 270 Amyl chloride
30 280 Amyl mercaptans
31 290 Amyl phenol
32 - 300 Aniline
33 310 Aniline hydrochloride
34 320 Anisidine
35 330 Anisole
36 340 Anthranilic acid
37 350 Anthraqin'none
38 360 Benzaldehyde
39 370 Benzamide
40 . 380 Benzene
41 390 Benzenedisulfonic acid
42 400 Benzenesulfonic acid
43 410 Benzil
44 420 Benin lie acid
45 430 Benzoic acid
46 440 Benzoin
47 450 Benzonitrile
48 460 Benzophenone
49 4,30 Benzotrichloride
F-3
-------
Clif ;;n'cal
!>0 490 Bc-nzoyl chloride
!>1 1300 ' Benzyl alcohol
:?2 510 Benzyl ann'ne
53 520 Benzyl bcnzoate
54. 530 Benzyl chloride
55 540 Benzyl dichloride
56 550 Biphenyl
57 560 .Bisphenol A
58 570 Bromobenzene
59 580 Bromonaphthalene"
60 590 Butadiene
61 592 1-butene
62 600 n-butyl acetate
63 630 n-butyl acrylate
64 640 n-butyl alcohol
65 650 s-butyl alcohol
66 660 t-butyl alcohol
67 670 n-butylamine
68 680 s-butylamine
69 690 t-butylamine
70 700 p-tert-butyl benzoic acid
71 710 1,3-butylene glycol
72 750 n-butyraldehyde
73 760 Butyric acid
74 770 Butyric anhydride
75 780 Butyronitrile
.F-4
-------
' OCI'DIi No. . Cin-iiiical
76 785 Cnprolactam
II 790 Carbon disulfide
78 800 Carbon tetrabromide
79 810 Carbon tetrachloride
80 820 Cellulose acetate
81 840 Chloroacetic acid
82 850 m-chloroaniline
83 860 o-chloroaniline
84 870 p-chloroaniline
85 880 Chlorobenzaldehyde
86 890 Chlorobenzene
87 900 Chlorobenzoic acid
88 905 Chlorobenzotrichloride
89 910 Chlorobenzoyl chloride
90 920 Chlorodifluoroethane
91 921 Chlorodifluoroniethane
92 930 Chloroform
93 940 Chloronapthalene
94 950 .o-chloronitrobenzene
95 951 p-chloronitrobenzene
96 960 Chlorophenols
97 964 Chloroprene
98 965 Chlorosulfonic acid
99 970 in-chlorotoluene
100 980 o-chlorotoluene
101 990 p-chlorotoluene
F-5
-------
OU'UB No. Clir.-ini_cal_s
102 992 Chlorotrifluoroiiiethane
103 1000 m-crcsol
104 1010 o-cresol
105 1020 p-cresol
106 1021 . Mixed cresols
107 1030 Crosylic acid
108 1040 . Crotonaldehyde
109 1050 Crotonic acid
110 .1060 Cuinene
111 1070 Cuinene hydroperoxide
112 1080 Cyanoacetic acid
113 1090 Cyanogen chloride
114 1100 Cyanuric acid
115 1110 Cyanuric chloride
116 1120 Cyclohexane
117 1130 Cyclohexanol
118 1140 Cyclohexanone
119 1150 Cyclohexene
120 1160 Cyclohexylamine
121 1170 Cyclooctadiene
122 1180 .Decanol
123 1190 Diacetone alcohol
124 1200 Diaminobenzoic acid
125 1210 Dichloroanillne
126 1215 m-dichlorobenzene
127 1216 o-dichlorobenzene
F-6
-------
128
129
130
131
132.
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
(H.I'hB N'o.
1220
1221
1244
1240
1250
1270
1280
1290
1300
1 304
1305
1310
1320
1330
1340
1360
1420
1430
1440
1442
1444
1450
1460
1470
1480
1490
Chemical
p-dichlorobonzene
Dichlorodi fl uoromothano
1 ,2-dichloroothane (HOC)
Dichloroethyl ether
Dichlorohydrin
Dichloropropene
Dicyclohexylamine
Di ethyl ami ne
Diethylene glycol
Diethylene glycol diethyl ether
Diethylene glycol dimethyl ether
Diethylene glycol monobutyl ether
DiethyTene glycol monobutyl ether acetate
Diethylene glycol monoethyl ether .
Diethylene glycol monoethyl ether acetate
Diethylene glycol rnonomethyl ether
Diethyl sulfate
Difluoroethane
Diisobutylene
Diisodecyl phthalate
Diisooctyl phthalate
Diketene
Dimethyl ami ne
N,N -dimethyl aniline
N,N-dimethyl ether.
N ,N-dirnethyl formamide
F-7
-------
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
1/3
174
175
176
177
178
OCPUI3 N_o_._
1495
1500
1510
1520
1530
1540
1545
.1550
1560
1570
1580
1590
1600
1610
1620
1630
1640
1650
1660
1661
1670
1680
1690
1700
1710
Chc-inical
Dimethyl hydrazine
Dimethyl sulfate
Dimethyl sulfide
Dimethyl sulfoxide
Dimethyl terephthalate
3, 5-di nitrobenzene acid
Dinitrophenol
Dinitrotoluene
Dioxane
Dioxolane
Diphenylamine
Diphenyl oxidq
Diphenyl thiourea
Dipropylene glycol
Dodecene
Dodecyl aniline
Dodecyl phenol
Epichlorohydrin
Ethanol
Ethanol amines
Ethyl acetate
Ethyl acetoacetate
Ethyl acrylate
Ethylamine
Ethylbenzene
F-8
-------
UU'i)B No. Chf.-iiiicajs
179 1720 Ethyl bromide
180 1730 Ethylcellulose
181 1740 Ethyl chloride
182 1750 Ethyl chloroacetate
183 1760 Ethyl cyanoaceta'te
184 1770 Ethylene
185 1780 Ethylene carbonate
186 1790' " Ethylcne chlorohydrin
187 1800 Ethylenediamine
188 1810 Ethylene dibromide
189 1830 Ethylene glycol
190 1840 Ethylene glycol diacetate
191 1870 Ethylene glycol dimethyl ether
192 1890 Ethylene glycol monobutyl ether
193 1900 Ethylene glycol monobutyl ether acetate
194 1910 Ethylene glycol monoethyl ether
195 . 1920 Ethylene glycol monoethyl ether acetate
196 1930 Ethylene glycol monoinethyl ether
197 1940 Ethylene glycol inonomethyl ether acetate
198 1960 Ethylcne glycol monophenyl e.ther
199 1970 Ethylene glycol monopropyl ether
200 1980 Ethylene oxide
201 1990 Ethyl ether
202 2000 2-ethylhexanol
203 2010 Ethyl orthoformate
204 2020 Ethyl oxalate
F-9
-------
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
Oi.i'ltf I!o.
20:10
2040
20'JO
2060
2070
2073
2090
2091
2100
2110
2120
2145
2150
2160
2165
2170
2180
2190
2200
2210
2240
2250
2260
2261
2270
2280
Ch.-.-inical
Ethyl sodium oxnlacetate
Formaldehyde
Form a mi da
Formic acid
Fumaric acid
Furfural
Glycerol (Synthetic)
Glycerol dichlorohydrin
Glycerol tri ether
Glycine
Glyoxal
Hexachlorobenzene
Hexachloroethane
Hexadecyl alcohol
Hexamethylenediami ne
Hcxame thy lone glycol
Hexamethylenetetramine
Hydrogen cyanide
Hydroquinane
p-hydroxybenzoic acid
Isoamylene
Isobutanol
Isobutyl acetate
Isobutylene
Isobutyraldehyde
. Isobutyric acid
F-10
-------
231
232
7.33
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
GU'ljB No.
2300
2320
2321
2330
2340
2350
2360
2370
2380
2390
2400
2410
2414
2417
2420
2430
2440
2450
2455
2460
2490
2500
2510
2520
2530.
2540
Chemical
.Isodecanol
Isooctyl alcohol
Isopentane
Isophorone
Isophthalic acid
Isoprcne
Isopropanol
Isopropyl acetate
Isopropylamine
Isopropyl chloride
Isopropyl phenol
Ketene
Linear alkyl sulfonate
Linear alkylbenzene
Maleic acid
Maleic anhydride
Malic acid
Mesityl oxide
Metanilic acid
I'ethacrylic acid
Methallyl chloride
Methanol
. Methyl acetate
Methyl acetoacetate
Methyl ami ne
n-methylani 1 ine
F-ll
-------
257 2545 Methyl bromide
258 2550 Methyl butynol
2159 2560 Methyl chloride
260 2570 Methyl cyclohexane
261 2590 Methyl cyclohoxanone
262 2620 Methylene chloride
263 2530 Methylene dianiline
264 2635 Methylene diphonyl diisocyanate
265 2640 Methyl ethyl ketone
266 2645 Methyl formate
267 2650 Methyl isobutyl carbinol
268 2660 Methyl isobutyl ketone
269 2665 Methyl methacrylate
270 2670 Methyl pentynol
271 2690 a-methylstyrene
272 2700 Morpholine
273 2710 a-naphthalene sulfonic acid
274 2720 g-naohthalene sulfonic acid
275 2730 a-naphthol
276 2740 g-naphthol
277 2750 Neopentanoic acid
278 . 2756 o-nitroani1ine
279 2757 p-nitroani1ine
280 2760 o-nitroanisole
281 2762 p-nUroanisole
282 2770 Nitrobenzene
F-12
-------
. fKi'DB Np_._ Chemical. __
283 2780 Nitrobenzpic acid (o, r, and p).
284 2790 ' Nitroethane
285 2791 Nitromethane
286 2792 Nitrophenol
287 2795 Nitropropane
288 2800 Nitrotoluene
289 2810 Nonene
290 2820 Nonyl phenol
291 2830 Octyl phenol
292 2840 Paraldehyde
293 2850 Pentaerythritol
294 2851 n-penta.ne
295 . 2855 1-pentene
296 2860 '. Perchloroethylene
297 . 2882 Perchloromethyl mercaptan
298 2890 o-phenetidine
299 2900 . p-phenetidine
300 2910 Phenol
301 2920 Phenolsulfonic acids
302 2930 . Phenyl anthranilic acid
303 2940 . Phenylenediamine
304 2950 Phosgene
305 2960 Phthalic anhydride
306 2970 Phthalimide
307 2973 B-picoline
308 2976 Piperazine
F-13
-------
nr.i'Mij ::o. ci.-.ncai
309 3000 Polybutenes
310 3010 Polyethylene glycol
311 3025 Polypropylene glycol
312 3063 Propionaldehyde
313 3066 Propionic acid
314 3070 n-propyl alcohol
315 3075 Propylamine
316 3080 . Propyl chloride
317 3090 Propylene
318 3100 Propylene chlorohydrin
319 3110 Propylene dichloride
320 3111 Propylene glycol
321 3120 Propylene oxide
322. 3130 Pyridine
323 3140 Quinone
324 3150 Resorcinol
325 3160 Resorcylic acid
326 3170 'Salicylic acid
327 3180 Sodium acetate
328 3181 Sodium benzoate
329 3190 Sodium carboxymethyl cellulose
330 3191 Sodium chloroacetate
331 32.00 Sodium formate
332 3210 Sodium phenate
333 3220 Sorbic acid
334 3230 Styrene
335 3240 Succinic acid
F-14
-------
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
OU'DB Mo.
3250
3251
3260
3270
3280
3290 & 3291
3300
3310
3320
3330
3335
3340
3341
3349
. 3350
3354
3355
3360
3370
3380
3381
3390,3391
& 3393
3395
3400
Chemical
Succinitrile
Sulfanilic acid
Sulfolane
Tannic .acid
Terephthalic acid
Tetrachloroe thanes
Tetrachlorophthalic anhydride
Tetraethyllead
Tetrahydronapthalene
Tetrahydrophthalic anhydride
Tetramethyllead
Tetramethyl enedi ami ne
Tetramethy 1 e thy! enedi ami ne
Toluene
Toluene-2,4-diamine
To! uene-2 ,4-di i socyanate
Toluene diisocyanates (mixture)
Toluene sulfonamide
Toluene sulfonic acids
Toluene sulfonyl chloride
To! ui dines
Trichlorobenzenes
1 ,1 ,1-trichloroethane
1 ,1 ,2-trichloroethane
F-15
-------
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
3/7
3/8
CO PUB No.
3410
3411
3420
3430
3450
3460
3470
3480
3490
3500
3510
3520
3530
3540
3541
3560
3570
3580
3590
Chemical
Trichloroethylcne
Trichlorof luoroincthiine
1 ,2,3-trichloropropane
1 ,1 ,2-trichloro-l ,2,2-trifluoroethane
Tri ethyl ami ne
Triethylene glycol
Tri ethyl ene glycol dimethyl ether
Triisobutylene
Tr i methyl ami ne
Urea
Vinyl acetate
Vinyl chloride
Vinylidene chloride
Vinyl toluene
Xylenes (mixed)
o-xylene
p-xylene
Xylenol
Xylidine
F-16
------- |