23 July 1982
EPA-450/3-81-001b
BENZENE EMISSIONS FROM
MALEIC ANHYDRIDE PLANTS-BACKGROUND
INFORMATION FOR PROMULGATED STANDARD
Emission Standards and Engineering Divisipn
U.S. Environmental Protection Agency
Office of A1r, Noise, and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
July 1982
-------
This report has been reviewed by the Emission Standards and
Engineering Division of the Office of Air Quality Planningand
Standards, U.S. Environmental Protection Agency, 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 Office
(MD-35), U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina 27711, or from National Technical Information
Services, 5285 Port Royal Road, Springfield, Virginia 22161.
PUBLICATION NO. EPA-450/3-80-001b
11
-------
ENVIRONMENTAL PROTECTION AGENCY
Background Information
and Final
Environmental Impact Statement
for Benzene Emissions from Maleic
Anhydride Manufacture
Prepared by:
Don R. Goodwin (Date)
Director, Emission Standards and Engineering Division
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
1. The promulgated national emission standards will limit benzene emissions
from existing and new maleic anhydride plants. The promulgated standards
implement Section 112 of the Clean Air Act and are based on the
Administrator's determination of June 8, 1977 (42 FR 29332) that benzene
presents significant risk to human health as a result of air emissions
from one or more stationary source categories and is therefore a
hazardous air pollutant. This standard will affect plants in Illinois,
Indiana, Missouri, New Jersey, Pennsylvania, Texas, and West Virginia.
2. Copies of this document have been sent to the following Federal
departments: Labor, Health and Human Services, Defense, Transportation,
Agriculture, Commerce, Interior, and Energy; the National Science
Foundation; the Council on Environmental Quality; members of the State
and Territorial Air Pollution Program Administrators; the Association of
Local Air Pollution Control Officials; EPA Regional Administrators; and
other interested parties.
3. For additional information, contact:
Susan Wyatt
Standards Development Branch (MD-13)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
telephone: (919) 541-5578.
4. Copies of this document may be obtained from:
U.S. EPA Library (MD-35)
Research Triangle Park, NC 27711
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
11 i
-------
TABLE OF CONTENTS
Page
1. SUMMARY 1-1
1.1 Summary of Changes Since Proposal 1-1
1.2 Summary of Impacts of Promulgated Action 1-2
1.2.1 Alternatives to Promulgated Action 1-2
1.2.2 Environmental and Health Impacts of
Promulgated Action. .. 1-2
1.2.3 Energy and Economic Impacts of
Promulgated Action 1-3
1.2.4 Other Considerations 1-3
1.2.4.1 Irreversible and Irretrievable
Commitment of Resources 1-3
1.2.4.2 Environmental and Energy Impacts of
Delayed Standard 1-4
2. SUMMARY OF PUBLIC COMMENTS 2-1
2.1 Selection of Maleic Anhydride for Regulation 2-1
2.1.1 Significant Source. ....... 2-1
2.1.2 Priority of Regulating Source Categories. ..... 2-12
2.1.3 Risk Estimate Consideration 2-13
2.1.4 Generic Benzene Standard. .... 2-14
2.1.5 Fugitive and Storage Emissions . 2-14
2.1.6 Existing Applicable Standards 2-14
2.2 Health and Environmental Impacts 2-15
2.2.1 Oispersion Modeling 2-15
2.2.2 Current Health Impact 2-15
2.2.3 Impact on Ambient Concentrations at USS Chemicals . 2-17
2.2.4 Risk and Expected Plant Life 2-17
2.2.5 Health Impact of Standard at Reichhold 2-17
2.2.6 Health Impact of Plant Closures - 2-17
2.2.7 Population "At Risk" 2-18
2.3 Emission Control Technology 2-18
2.3.1 Feasibility of 97 Percent Control ......... 2-18
2.3.2 Feasibility of Carbon Adsorption 2-19
2.3.3 Use of Catalytic Incineration 2-19
2.4 Cost and Economic Impact 2-20
2.4.1 Capital Costs . 2-20
2.4.2 Medical Costs 2-21
2.4.3 Costs Per Leukemia Case 2-23
2.5 Selection of Basis of the Standard for Existing Sources. . 2-23
2.5.1 Basis of Standard for Continuous Emissions 2-23
2.5.1.1 Generic Benzene Control 2-23
2.5.1.2 Economic Reasonableness of Standard. . . . 2-25
2.5.1.3 Selection of Regulatory Alternatives . . . 2-25
2.5.1.4 Selecton of Control System 2-30
2.5.2 Basis of Standard for Excess Emissions 2-31
2.5.2.1 Startup Emissions Allowance 2-31
2.5.2.2 Shutdown Emissions Allowance ....... 2-34
iv
-------
TABLE OF CONTENTS (Continued)
Page
2.5.2.3 Shutdown Time 2-34
2.5.2.4 Control Device Malfunctions 2-35
2.5.2.5 Malfunction Definition 2-42
2.5.2.6 Determination of Control Device
Malfunction ...... 2-42
2.5.2.7 Control Device Maintenance 2-43
2.5.2.8 Design, Equipment, Work Practice, or
Operational Standards 2-43
2.5.3 Selection of Promulgated Standard for Existing
Sources ..................... 2-44
2.6 Selection of Basis of the Standard for New Sources .... 2-45
2.6.1 Zero Emission Standards 2-45
2.6.2 n-Butane Technology 2-46
2.6.3 Banning of Benzene 2-47
2.6.4 Economic Considerations 2-48
2.7 Modification and Reconstruction 2-50
2.8 Test Methods and Monitoring 2-53
2.8.1 Gas Chromatography 2-53
2.8.2 Process Parameter Monitoring 2-54
2.8.3 Total Hydrocarbon Monitoring 2-55
2.8.4 Monitor Malfunction 2-56
2.8.5 Permeation Tubes. 2-56
2.8.6 Process Vent Benzene Concentrations 2-57
2.8.7 Calibration 2-57
2.8.8 Chromatograph Response 2-58
2.9 Reporting and Recordkeeping 2-59
2.10 Legal 2-59
2.10.1 Airborne Carcinogen Policy as Basis for Rulemaking. 2-59
2.10.2 BDT as Control Strategy 2-60
2.10.3 Alternatives to Uniform Standard 2-62
2.10.4 New Listing of Benzene . 2-64
2.11 Miscellaneous 2-64
Appendix A: Development of Revised Costs for Control of Benzene
Emissions from Maleic Anhydride Plants A- 1
Appendix B: Energy Requirement Calculations B- 1
Appendix C: Addendum to the Economic Analysis of NESHAP for the
Maleic Anhydride and Fumaric Acid Industries C- 1
Appendix D: Methodology for Estimating Leukemia Incidence and
Maximum Lifetime Risk from Exposure to Benzene
Emissions from Maleic Anhydride Process Vents ..... D- 1
v
-------
LIST OF TABLES
Number
Page
2-1 List of Commenters on the Proposed National Emission
Standard for Benzene Emissions from Maleic Anhydride
Plants 2-2
2-2 Comparison of Capital Cost Estimates for 97 Percent
Control Using Thermal Incineration 2-22
2-3 Impacts of Regulatory Alternatives for Existint Plants. . . 2-28
2-4 Troubleshooting Time for Thermal Incinerator (Total
Venting Time Required) 2-38
2-5 Shutdown Emissions at Full Capacity and at 50 Percent
Capacity. 2-40
2-6 Comparison of Transfer Prices for n-Butane and Benzene
Processes 2-49
VI
-------
1. SUMMARY
1.1 SUMMARY OF CHANGES SINCE PROPOSAL
Since proposal of these standards, a number of changes have been made,
the most important being that the promulgated standard allows 1.0 kilogram
of benzene to be emitted to the atmosphere per 100 kilograms (1 lb per 100
lb) of benzene fed to the maleic anhydride reactor. The proposed standard
would have allowed 0.3 kilogram of benzene to be emitted per 100 kilograms
(0.3 lb per 100 lb) fed to the reactor. After reevaluating costs due to
the standard, based in part on plant-specific information industry
provided, the Administrator concluded that, in comparing the proposed
emission limit with the promulgated emission limit, the incremental costs
of the proposed emission limit were unreasonably high in light of the
incremental emissions reduction achieved. The difference is due largely to
one plant that already has controls sufficient to meet the promulgated
standard but that would have had to scrap current systems and install new
ones to meet the proposed standard.
Another significant change is in the standard's excess emissions
requirements. The proposed standard would have allowed excess emissions
only when a control device malfunction occurred, and then only up to the
amount of emissions that would have occurred if the plant were otherwise
shut down. Two changes were made in the promulgated standard. The
promulgated standard allows a maximum of 8 hours of excess emissions for
total plant startups and a maximum of 1.5 hours of excess emissions for
individual reactor startups. In addition, the emissions limit during a
control device malfunction was revised to allow more time to diagnose the
problem and make standard repairs. The promulgated standard allows a
maximum of 6 hours of excess emissions after onset of control device
malfunction, based on 6 hours for diagnosis and repair for nonmajor
1-1
-------
malfunctions or shutdown. In addition, as at proposal, the plant owner or
operator must demonstrate that excess emissions were due to a malfunction.
The promulgated standard allows an existing benzene-based plant to
switch to an alternative feedstock and reconvert to benzene if it so chose
when the initial nonbenzene catalyst was spent. The standards'
reconstruction provisions have been deleted.
1.2 SUMMARY OF IMPACTS OF PROMULGATED ACTION
1.2.1 Alternatives to Promulgated Action
Regulatory options were discussed in Chapter 3 of the Background
Information Document (BID) for the proposed standard. After proposal, an
additional alternative of 90 percent control was considered. The
90-percent option was evaluated because revised capital and annualized
costs of equipping three plants to meet the proposed 97-percent standard
were estimated to have been $8.6 million and $3.5 million, respectively, a
significant increase since proposal. At proposal, the capital and
annualized costs were estimated to be $6.6 million and $2.5 million,
respectively, for equipping five plants to meet the 97-percent standard.
Because costs for this alternative have increased significantly since
proposal and because most of the costs would be borne by two plants, the
less stringent alternative was examined and subsequently selected as the
basis of the standard.
1.2.2 Environmental and Health Impacts of Promulgated Action
The promulgated standard will reduce nationwide benzene emissions from
maleic anhydride units from about 1000 Mg/yr (1,100 tons/yr) to about 420
Mg/yr (460 tons/yr). Estimated maximum lifetime risk will be reduced from
a range of 2.9 x 10"^ to 2.0 x 10"^ to a range of 6.9 x 10~6 to 4.7 x 10*5.
Maximum lifetime risk is the probability of someone within the assumed
exposed population, who is exposed to the highest maximum annual average
benzene concentration during an entire lifetime (70 years), of contracting
leukemia. Estimated leukemia incidence will be reduced from a range
of 0.013 to 0.092 cases per year to a range of 0.011 to 0.079 cases per year.
Due to the assumptions which were made in calculating the maximum lifetime
risk and leukemia incidence numbers, there is considerable uncertainty
associated with them beyond the ranges presented here. They may represent
1-2
-------
overestimates or underestimates. The uncertainties associated with the risk
numbers are explained in section 2.1.1.
Analyses of n-butane conversion impacts, water pollution impacts, and
solid waste disposal impacts have not changed since proposal.
1.2.3 Energy and Economic Impacts of Promulgated Action
Energy impacts were discussed in Subsection 4.4 of the proposed
standard's BIO. These impacts have changed because of plant-specific
information provided by the industry and changes that have occurred to the
industry. The following impacts were calculated based on plant-specific
waste gas temperatures, flow rates, and benzene concentrations.
The promulgated standard will increase energy use 7,300 GJ/yr (1,180
barrels of fuel oil equivalent/yr) if Pfizer installs a carbon adsorption
unit and 61,600 GJ/yr (9,970 barrels of fuel oil equivalent/yr) if Pfizer
installs a thermal incineration unit to meet the standard. Thermal
incineration impacts assume 50 percent heat recovery.
Because of industry changes and revised cost estimates, the economic
impacts discussed in Chapter 5 of the proposed standard's BIO have been
modified. Industrywide installed capital cost to meet the promulgated
standard by carbon adsorption will be about $1.4 million, and the same cost
to meet the promulgated standard by thermal incineration will be about $1.8
million. Total annualized cost due to the promulgated standard will be
about $0.27 million to use carbon adsorption and $0.75 million to use
thermal incineration.
Product price impacts also have been revised. No increase in the
price of maleic anhydride is expected to occur as a result of the
standards. The standard could increase fumaric acid prices 3 to 4
percent.
No plant closures are anticipated to result from the promulgated
standard.
1.2.4 Other Considerations
1.2.4.1 Irreversible and Irretrievable Commitment of Resources.
Subsection 4.6.1 of the BID for the proposed standard discussed resource
commitments. No changes have occurred since proposal.
1-3
-------
1.2.4.2 Environmental and Energy Impacts of Delayed Standard.
Environmental impacts of a delayed standard have decreased because of
industry changes. Current benzene emissions have decreased by about 4,800
Mg/yr (1.1 x 10^ lb/yr) since proposal. Energy impacts of a delayed
standard, as discussed in Appendix 8 of Volume 1 of the 810, have not
changed.
1-4
-------
2. SUMMARY OF PUBLIC COMMENTS
Commenters, affiliations, and EPA docket number assigned each comment
are shown in Table 2-1. Thirty-five letters and documents on the proposed
standard and its background information document (BID) were received.
Significant comments have been divided into the following 11 categories:
1.
Selection of Maleic Anhydride for Regulation
2.
Health and Environmental Impacts
3.
Emission Control Technology
4.
Cost and Economic Impact
5.
Selection of Basis of the Standard for Existing Sources
6.
Selection of Basis of the Standard for New Sources
7.
Modification and Reconstruction
8.
Test Methods and Monitoring
9.
Reporting and Recordkeeping
10.
Legal
11. Miscellaneous
Comments, issues, and responses are discussed in the following
sections of this chapter. Changes to the regulation are summarized in
Subsection 1.2 of Chapter 1.
Responses to comments on the health effects, listing, and regulation
of benzene are contained in Responses to Public Comments on EPA's Listing
of Benzene Under Section 112 and Relevant Procedures for the Regulation of
Hazardous Air Pollutants, EPA-450/5-82-003.
2.1 SELECTION OF MALEIC ANHYDRIDE FOR REGULATION
2-1.1 Significant Source
Comment: According to one commenter, the Administrator has recognized
that risk from pollutant sources may be insignificant even if the pollutant
is an airborne carcinogen listed under Section 112. According to the
Administrator, "This may occur, for example, because . . . sources have
installed adequate controls on their own initiative or in response to other
regulatory requirements." The commenter believes the maleic anhydride
2-1
-------
TABLE 2-1. LIST OF COMMENTERS ON THE PROPOSED NATIONAL
EMISSION STANDARD FOR BENZENE EMISSIONS FROM MALEIC
ANHYDRIDE PLANTS
Commenter Docket No. Affiliation
J. T. Barr
E. A. Treanor
D. B. Rathbun
C. H. Fishman
N. J. King
F. S. Lisella
J. J. Moon
P. F. Infante
Chemical
Manufacturer's
IV-D-1, Part I
IV-D-5, Part II
IV-D-2, Part I
IV-D-3, Part I
IV-D-4, Part I
IV-D-5, Part I
IV-D-6, Part I
IV-D-19, Part II
IV-D-7, Part I
IV-D-21, Part II
IV-D-8, Part I
IV-D-9, Part I
IV-D-22, Part II
Association (CMA) IV-F-8, Part II
J. M. DeMeester IV-D-10, Part I
Air Products and Chemicals, Inc.
Box 538
Allentown, Pennsylvania 18105
American Petroleum Institute (API)
2101 L Street, Northwest
Washington, D.C. 20037
API
Wilmer & Pickering
1666 K Street, N.W.
Washington, D.C. 20006
Wilmer & Pickering
Center for Disease Control
Department of Health and Human
Services
U.S. Public Health Service
Atlanta, Georgia 30333
Phillips Petroleum Company
Bartlesville, Oklahoma 74004
Occupational Safety and
Health Administration (OSHA)
U.S. Department of Labor
Washington, D.C. 20210
Chemical Manufacturers
Association (CMA)
1825 Connecticut Avenue, N.W.
Washington, D.C. 20009
Dow Chemical Company
Bennett Building
2030 Dow Center
Midland, Michigan 48640
(Continued)
-------
TABLE 2-1. LIST OF COMMENTERS ON THE PROPOSED NATIONAL
EMISSION STANDARD FOR BENZENE EMISSIONS FROM MALEIC
ANHYDRIDE PLANTS (Continued)
Commenter
Docket No.
Affiliation
R. K. Meyers
IV-D-11, Part I
R. C. Sterrett IV-D-12, Part I
API
IV-D-13, Part I
IV-F-9, Part II
G. C. Iannelli IV-D-14, Part I
F. M. Brower IV-D-2, Part II
L. Behr IV-D-4, Part II
A. F. Montgomery IV-D-6, Part II
M. L. Joseph
D. Rector
IV-D-7, Part II
IV-D-20, Part II
IV-D-8, Part II
H. H. Hovey, Jr. IV-D-9, Part II
J. Ruspi
D. J. Goodwin
IV-D-10, Part II
IV-D-11, Part II
Texaco, Inc.
P.O. Box 509
Beacon, New York
12508
Ashland Chemical Company
P.O. Box 2219
Columbus, Ohio 43216
API
General Council of the United
States Department of Commerce
U.S. Department of Commerce
Washington, D.C. 20230
Dow Chemical Company
Private Citizen
64 Maple Lane
Greens Farms, Connecticut
06346
National Science Foundation (NSF)
Washington, D.C. 20550
OS HA
State of Michigan
Department of Natural Resources
Box 30028
Lansing, Michigan 48909
New York State Department
of Environmental Conservation
50 Wolf Road
Albany, New York 12233
The Aerospace Corporation
20030 Century Boulevard
Germantown, Maryland 20767
Illinois Environmental
Protection Agency
2200 Churchill Road
Springfield, Illinois 62706
2-3
(Continued)
-------
TABLE 2-1. LIST OF COMMENTERS ON THE PROPOSED NATIONAL
EMISSION STANDARD FOR BENZENE EMISSIONS FROM MALEIC
ANHYDRIDE PLANTS (Continued)
Commenter
Docket No.
Affiliation
L. D. Johnson
IV—D—14, Part II
Rohm and Haas Company
Environmental Control Department
Box 584
Bristol, Pennsylvania 19007
R. M. Gifford
IV-D-15, Part II
Pfizer Chemicals Division
235 East 42nd Street
New York, New York 10017
M. Lennon
IV-D-16, Part II
API
N. B. Galluzzo
IV-D-17, Part II
IV-D-18, Part II
Monsanto Plastics and Resins Co.
800 N. Lindbergh Boulevard
St. Louis, Missouri 63166
R. W. Russell
IV-D-23, Part II
Council on Wage and Price Stability
Winder Building
600 17th Street, N.W.
Washington, D.C. 20506
M. R. Foresman
IV-D-25, Part II
Monsanto Chemical Intermediates Co.
800 N. Lindbergh Boulevard
St. Louis, Missouri 63166
D. E. Rickert
IV-F-2, Part II
Chemical Industry Institute
of Toxicology (CIIT)
P.O. Box 12445
Research Triangle Park, North
Carolina 27709
R. D. Irons
IV-F-3, Part II
IV-F-11, Part II
CIIT
D. Doniger
IV-F-4, Part II
Natural Resources Defense Council
1725 I Street, N.W., Suite 600
Washington, D.C. 20006
P. S. Hewett
IV-F-5, Part II
Relchhold Chemicals, Inc.
601-707 Woodward Heights Boulevard
Detroit, Michigan 48220
D. Glassman
IV-F-6, Part II
USS Chemicals
600 Grant Street
Pittsburgh, Pennsylvania 15230
(Continued)
2-4
-------
TABLE 2-1. LIST OF COMMENTERS ON THE PROPOSED NATIONAL
EMISSION STANDARD FOR BENZENE EMISSIONS FROM MALEIC
ANHYDRIDE PLANTS (Continued)
Commenter
Docket No.
Affi1iation
A. Meyer
IV-F-7, Part II
DENKA Chemical Corporation
8701 Park Place Boulevard
P.O. Box 87220
-
Houston, Texas 77017
H. A. Jewett
IV-F-10, Part II
Private Citizen
5451 42nd Street N.W.
Washington, D.C. 20015
2-5
-------
industry has installed such controls and cited the operating status of the
following plants as examples (Part II Docket Item IV-D-22).
The Reichhold (New Jersey) maleic anhydride plant has closed and will
be dismantled (p. 61-62).
Koppers does not intend to reopen its Bridgeville, Pennsylvania,
maleic anhydride plant unless very substantial changes in the economics of
the industry occurred. If the plant reopened, it would operate (as in the
past) with emission controls sufficient to meet the standard (Attachment
E). Monsanto has recently completed installation of a thermal oxidizer at
its St. Louis maleic anhydride plant (Attachment E).
Tenneco has registered its maleic anhydride plant with the New Jersey
Department of Environmental Protection. The State will require Tenneco to
install an incinerator or equivalent controls. Tenneco interprets this
requirement as equivalent to 97 percent control and intends to comply
(Attachment E).
Response: At proposal, the Administrator determined that maleic anhydride
process vents are a source category that should be regulated for two
reasons: they emit significant amounts of benzene, and based on estimated
maximum lifetime risk and number of deaths, they ranked as one of the
higher priority benzene source categories for regulation (Part II Docket
Item II-I-99). These conclusions were based on the maleic anhydride
industry's status during the proposed standard's development.
At proposal, the following maleic anhydride plants were
considered to be operating or operational:
USS Chemicals, Neville Island, Pennsylvania;
Reichhold Chemicals, Inc., Morris, Illinois;
Reichhold Chemicals, Elizabeth, New Jersey;
Ashland Chemical, Neal, West Virginia;
DENKA Chemical Company, Houston, Texas;
Koppers, Bridgeville, Pennsylvania;
Monsanto, St. Louis, Missouri (20 percent n-butane feedstock);
Tenneco, Fords, New Jersey; and
Amoco, Joliet, Illinois (n-butane feedstock).
In addition, several plants were considered to have process vent control
devices with the following emission reduction capabilities:
2-6
-------
Koppers, 99 percent control from a waste heat boiler;
DENKA, 97 percent control from a thermal incinerator;
Reichhold, Morris, 90 percent control from a carbon adsorption
unit;
Reichhold, Elizabeth, 97 percent control from a carbon adsorption
unit; and
USS Chemicals, 90 percent control from a catalytic incinerator.
Finally, the proposal noted that the standard would also apply to one
fumaric acid plant producing maleic anhydride as an intermediate product
using a benzene feedstock. This plant is owned by Pfizer, has a capacity
of 10,700 Mg/yr (fumaric acid), and is located in Terre Haute, Indiana.
Health, environmental, energy, and economic impacts for this plant were not
included in the proposal because EPA had only become aware of this plant
just prior to proposal. Unlike other fumaric acid producers, which make
both fumaric acid and maleic anhydride for sale (such as Monsanto and U.S.
Steel), Pfizer produces only fumaric acid for sale. All the maleic acid it
produces is used captively. Because of this, it was not originally
identified as a maleic anhydride producer. However, it produces maleic
acid with the same technology, air oxidation of benzene, as other maleic
anhydride producers covered by the standards. The only difference is that
Pfizer is riot a commercial seller of maleic anhydride or maleic acid.
Consequently, Pfizer's maleic anhydride production unit appropriately is
covered by the standards.
Based on information obtained prior to proposal, EPA made estimates of
the health impact due to emission from these plants. Commenters felt that
these quantitative estimates show the risks to be insignificant and that
the industry did not warrant regulation.
Quantitative risk estimates at ambient concentrations involve an
analysis of the effects of the substance in high-dose epideniological or
animal studies, and extrapolation of these high-dose results to relevant
human exposure routes at low doses. The mathematical models used for such
extrapolations are based on observed dose-response relationships for
carcinogens and assumptions about such relationships as the dose approaches
very low levels or zero.
The risk to public health from carcinogenic emissions may be estimated
by combining the dose-response relationship obtained from this
2-7
-------
carcinogenicity strength calculation with an analysis of the extent of
population exposure to the substance through the ambient air. Exposure in
this context is a function of both a substance's concentration and
duration.
The exposure analyses are based on air quality models, available
emission estimates from maleic anhydride plants, and approximate population
distributions near these sources.
The air quality models used estimated exposures of up to 20
kilometers, and population and growth statistics were examined. Along with
the existing carcinogenic strength determinations, the information
collected was used to provide estimates of the degree of risk to
individuals and the range of increased cancer incidence expected from
ambient air exposures associated with maleic anhydride plants at various
possible emissions levels.
The decision to employ estimates of carcinogenic risks despite their
lack of precision rests on the belief that, although they are subject to
considerable uncertainties, current analytical models and techniques can,
with due consideration of the uncertainties, provide useful estimates of
relative carcinogenic strength and of the probable general ranges of excess
cancer incidence and individual risks. This view has been supported by the
National Academy of Sciences, the National Cancer Advisory Board, and
others.
The health impacts estimated at proposal were bounded by a range. The
ranges presented represent uncertainty in estimates of benzene
concentrations to which workers were exposed in occupational studies of
Infante, Aksoy, and Ott that served as the basis for developing the benzene
unit risk factor (Part I Docket Item II-A-31). Ranges are based on a
95-percent confidence interval that assumes estimated benzene
concentrations to which workers were exposed are within a factor of two of
actual concentrations.
Other uncertainties associated with estimating health impacts are not
quantified here. Maximum lifetime risk and leukemia incidence were
calculated based on a no-threshold linear extrapolation of leukemia risk
associated with a presumably healthy white male cohort of workers exposed
to benzene concentrations in the parts per million to the general
2-8
-------
population, which includes men, women, children, nonwhites, the aged, and
the unhealthy, who are exposed to concentrations in the parts per billion.
These widely diverse population segments may or may not have differing
susceptibility to leukemia than do workers in the studies. In addition,
the exposed population is assumed to be immobile, remaining at the same
location 24 hours per day, 365 days per year, for a lifetime (70 years).
This assumption is counterbalanced to some extent (at least in the
calculation of incidence) by the assumption that no one moves into the
exposure area either as a permanent resident or as a transient.
Assumptions that must be made in order to estimate ambient concentrations
by dispersion modeling and exposed populations by census tract also
introduce uncertainties into the risk estimates. Modeled ambient benzene
concentrations depend upon (1) plant configuration, which is difficult to
determine for more than a few plants; (2) emission point characteristics,
which can be different from plant to plant and are difficult to obtain for
more than a few plants; (3) emission rates which may vary over time; and
from plant to plant; and (4) meteorology, which is seldom available for a
specific plant. The particular dispersion modeling used can also influence
the numbers. The best model to use (ISC) is usually too resource intensive
for modeling a large number of sources. Less complex models introduce
further uncertainty through a greater number of generalizing assumptions.
Dispersion models also assume that the terrain in the vicinity of the
source is flat. For sources located in complex terrain, the maximum annual
concentration could be underestimated by several fold due to this
assumption. Furthermore, leukemia incidence is the only benzene health
effect considered. Other health effects, such as aplastic anemia and
chromosomal aberrations, are not as easily quantifiable and are not
reflected in the risk estimates. Although these other health effects have
been observed at occupational levels, it is not clear if they occur at
ambient exposure levels. Additionally, benefits to the general population
of controlling other organic emissions from maleic anhydride production are
not quantified and are not reflected in the risk estimates. Overlapping
benzene exposures from other source categories are also not included in the
estimates. An individual living near a maleic anhydride plant, for
example, is also exposed to benzene emissions from automobiles and service
stations. Finally, these estimates do not include cumulative or
2-9
-------
synergistic effects of concurrent exposure to benzene and other substances.
As a result of these uncertainties, the leukemia incidence and the maximum
lifetime risk calculated around maleic anhydride plants could be
overestimated. More importantly, however, they could just as likely be
underestimated for the same reasons.
The status of the maleic anhydride industry has changed since
proposal. Several commenters felt that these changes have reduced benzene
emissions and associated health risks effectively, and thus the impacts of
a promulgated standard have been reduced. First, two plants have ceased
operation permanently. Koppers shut down its Bridgeville, Pennsylvania,
plant (15,400 Mg/yr capacity) in March 1979 and has no plans to reactivate
it for maleic anhydride production (Part II Docket Items IV-D-22,
Attachment E, and II-I-38). Reichhold shut down its Elizabeth, New Jersey,
plant (13,600 Mg/yr capacity) in August 1979, which is being dismantled
(Part II Docket Item IV-D-22, Attachment E).
The largest source of uncontrolled benzene emissions, the Monsanto
plant, has recently installed a thermal incinerator with a waste heat
boiler that can achieve 97 percent control of benzene emissions (Part II
2-9-A
-------
Docket Item IV-E-12). Its installation is part of a Monsanto environmental
control and energy conservation program. Although not yet completely
operational on a full-time basis, the device will control volatile organic
compounds (VOC's) from either the n-butane-based or the benzene-based
process (Part II Docket Item IV-D-22, Attachment E). When the oxidizer is
fully operational, 30 percent of the plant's steam requirements will be
supplied by the maleic anhydrice process. The Monsanto plant currently
uses n-butane for 20 percent of its feedstock needs and plans to convert
renaining capacity to n-butane by 1985 (Part II Docket Item II-I-42).
The Tenneco plant has become subject to New Jersey regulations on the
"Control and Prohibition of Air Pollution by Toxic Substances" (Subchapter
17 of the New Jersey Administrative Code) (Part II Docket Item IV-D-22,
Attachment E). Under this regulation, benzene is defined as a toxic
volatile organic substance (TVOS). Tenneco is required by the new Jerseyu
Bureau of Air Pollution Control either to reduce benzene emissions by 99
percent or to control 97 percent of n-butane emissions if n-butane instead
of benzene is used as a feedstock (Part II Docket Item IV-D-30).
The DENKA plant currently is using n-butane for all of its capacity
(Part II Docket Item IV-E-7), and the Ashland plant has converted all of
its capacity to n-butane (Part II Docket Item IV-E-8). Capable of
switching back and forth as economic conditions change, the Pfizer plant
currently is purchasing only about 10 percent of its maleic anhydride, with
the remainder produced from benzene for its fumaric acid process (Part II
Docket Item IV-E-11). Finally, the Reichhold, Morris, Illinois, plant and
USS Chemicals plants have not changed status since proposal.
Industry commenters attribute reduced benzene emissions and associated
health risks to these changes and consider the impacts of any promulgated
standard reduced to the extent that no regulation is required. They
considered the estimated risk posed by benzene emissions from maleic
anhydride process vents insignificant.
In view of the changes in the maleic anhydride industry following
proposal of the standard; EPA examined the current emission levels, risks,
and other factors to determine whether maleic anhydride process vents pose
a significant risk of leukemia and whether a standard is needed.
2-10
-------
While maleic anhydride plants are currently neither the sole cause of
benzene emissions to the atmosphere nor the largest source category
emitter, they contribute significantly to nationwide benzene emissions. In
the absence of a standard, the industry would still be emitting about 1,000
Mg/yr from process vents (Part II Docket Item IV-F-8). Pfizer, the
uncontrolled plant, alone will contribute over 650 Mg/yr to this total.
Benzene is causally related to leukemia induction in humans. The
estimated 4 million people living within 20 kilometers of benzene-emitting
maleic anhydride plants are exposed to higher levels of benzene than if
they lived at greater distances. Because no known threshold exists for
benzene's carcinogenic effects, these people not only incur a higher
benzene exposure but also run greater risk of contracting leukemia due to
that exposure.
2-10-A
-------
In the absence of a standard, estimated maximum lifetime risk would
range from 2.9 x 10"^ to 2.0 x 10~4 to the most exposed individuals.
Maximum lifetime risk is the probability of someone within the assumed
exposed population contracting leukemia who is exposed to the highest
maximum annual average benzene concentration during an entire lifetime (70
years). Additionally, a range of 0.013 to 0.092 leukemia cases per year
due to benzene emissions from maleic anhydride process vents is estimated
(see Appendix D).
Thus, based on the human carcinogenicity of benzene, the amount of
benzene currently emitted from maleic anhydride process vents, the number
of people exposed to benzene emissions from maleic anhydride process vents,
and the uncertainties associated with the estimates of maximum risks and
leukemia incidence, the Administrator has concluded that benzene emissions
from maleic anhydride process vents pose a significant risk of leukemia to
the general public.
Several other factors were considered in the Administrator's
determination that a standard for maleic anhydride process vents is needed.
First, if no standard were promulgated, new benzene-based sources could be
built and could remain uncontrolled. Such construction could increase
maximum lifetime leukemia risk and would increase estimated leukemia cases
per year. Second, if no standard were promulgated, one existing source
(Pfizer) would remain uncontrolled, with emissions of over 650 Mg/yr.
Third, although some existing sources have installed controls to meet State
regulations, another is doing so voluntarily as part of a company
environmental and energy conservation program, and not to comply with a
specific pollution standard. A nationwide benzene standard for maleic
anhydride plants would provide additional assurance that existing sources
control emissions on a continuing basis. Fourth, existing sources that
currently use a butane feedstock and uncontrolled sources that plan to
convert to butane could reconvert to benzene if they so chose, thereby
remaining uncontrolled and increasing benzene emissions over what they
would have been under the standard. Fifth, although plants have already
taken steps necessary to meet the proposed standard, thereby reducing the
health impact of a promulgated standard from levels projected at proposal,
cost and economic impacts of a promulgated standard likewise would be
2-11
-------
reduced. Plants that have taken steps to meet the proposed standards would
bear no control equipment costs as a result of the standards.
Congress writes legislation such as the Clean Air Act and implicitly
sets priorities for government regulation of risks. In setting these
priorities, Congress has found that it is not feasible to regulate every
risky activity, technology, or event, but that it is desirable to regulate
those risks that can be reasonably reduced. Political processes direct
priorities towards those risks that are most repugnant to the public.
Ultimately, government agencies, through Congressional actions, are given
regulatory authority; but these agencies are likely to interpret their
authority differently, thus addressing risk differently.
Accordingly, EPA is required by Section 112 of the Clean Air Act to
identify and regulate hazardous air pollutants. As described previously in
this response, EPA has determined, based on risk and other factors, that
maleic anhydride process vents represent a significant source of benzene.
2.1.2 Priority of Regulating Source Categories
Comment: One commenter felt that EPA had not ordered its priorities
properly and was attempting to regulate first one of the smaller sources of
risk due to benzene emissions (Part II Docket IV-D-5). Conversely,
another felt that control of all benzene emissions should be given high
priority (Part II Docket Item IV-D-9).
Response: There is no one principal stationary source of benzene
emissions. Reducing the population's exposure to benzene requires
controlling several individual source categories, each of which represents
a relatively small proportion of the total emissions. All major stationary
benzene sources were identified (Part I Docket Item II-A-21), and presently
standards are being developed for several sources under Section 112 of the
Clean Air Act. In addition to the maleic anhydride standard, standards
have been proposed for ethyl benzene/styrene plants (45 FR 8344 8), benzene
storage tanks (45 FR 83952), and benzene fugitive emissions (46 FR 1165).
A fifth, for benzene emissions from coke by-product plants, is scheduled to
be proposed in 1982. Mobile sources are not covered by Section 112 of the
Act, but their emissions are being reduced by catalytic converters. A
study by General Motors Research Laboratories (Part I Docket Item 11-1-98)
indicated that catalytic converters reduced benzene emissions up to 90
percent.
2-12
-------
Initially, after benzene was listed in 1977 as a hazardous air
pollutant (42 FR 29332), standards development for all stationary benzene
source categories was given high priority. As more information was
gathered, it became apparent that some source categories emit much smaller
amounts of benzene than others, indicating that some posed less relative
risk than others. Consequently, to use resources better, the Agency
arranged source categories by priority in terms of estimated maximum risk
and leukemia incidence. This ordering identified source categories
presenting greater risk for which regulatory efforts should proceed
immediately. This estimation ranked maleic anhydride process emissions as
one of the higher priority source categories for regulation (Part I Docket
Item II-I-99).
Despite the fact that industry changes (see Subsection 2.1.1) have
decreased estimated maximum lifetime risk and leukemia incidence, maleic
anhydride process vents still emit significant quantities of benzene (1,000
Mg/yr) to which a significant number of people are exposed (4.0 million).
In addition, several other nonrisk factors support continued standards
development (see Subsection 2.1.1).
2.1.3 Risk Estimate Consideration
Comment: After recalculating the number of leukemia cases per year due to
benzene emissions from maleic anhydride process vents as 0.079, one
commenter (Part II Docket Item IV-D-22) considered this number close enough
to the number of leukemia cases that EPA deemed reasonable at proposal to
say that no regulation was necessary.
Response: Best demonstrated technology (BDT), the first step in choosing
the basis of the standard, is selected based on environmental, economic,
and energy impacts, which can be measured relatively accurately. However,
quantitative health impacts, as discussed previously, are uncertain and in
isolation are poor absolute measures. Because of the considerable
uncertainty in attempting to quantify cancer risks, the Agency believes
controls applied to significant sources of airborne carcinogens should, at
a minimum, represent BDT. The Administrator then examines the residual
risks remaining after application of BDT to determine whether or not they
are unreasonable in view of the risk reduction that would be gained by
requiring a control level beyond BDT and the associated cost increase.
Therefore, the decision that the residual risk is "not unreasonable" is a
2-13
-------
function not of its magnitude but of the reduction in risk that could be
gained with additional resources. In the absence of consideration of the
economic consequences of further action, the label "reasonable" or
"unreasonable" cannot be applied to risk estimates solely.
2.1.4 Generic Benzene Standard
Comment: One commenter supported postponement of a standard for benzene
emissions from maleic anhydride plants until it can be incorporated into a
larger, general benzene emissions standard (Part II Docket Item IV-D-15).
Response: In light of benzene's carcinogenicity and its adverse health
effects (Response to Public Comments on EPA's Listing of Benzene Under
Section 112 and Relevant Procedures for the Regulation of Hazardous Air
Pol 1utants, EPA-450/5-82-003), postponement and subsequent incorporation of
this standard into a larger, general benzene standard would not be prudent.
In addition, the numerous source categories, their varying characteristics,
and the resulting different methods to control each source's emissions make
a general benzene emission standard infeasible.
2.1.5 Fugitive and Storage Emissions
Comment: According to one commenter, division of process, fugitive, and
storage emissions into separate documents is confusing and makes comparison
of appropriate information difficult (Part II Docket IV-D-19).
Response: Because various sources emit benzene, categories were
established according to similar emission characteristics and applicable
control technology. Although standards for benzene fugitive and storage
emissions could have been incorporated into the maleic anhydride
regulation, these standards would have been specific only to sources at
maleic anhydride plants. To cover all benzene fugitive and storage
sources, EPA would need to develop identical fugitive and storage standards
for each kind of chemical plant and refinery using or making benzene. To
avoid this redundancy, to develop standards more efficiently, and to cover
all like benzene fugitive and storage sources with uniform requirements,
EPA developed standards for these sources separately from those for process
vents and applied than to several source categories (46 FR 1165 and 45 FR
83 952).
2.1.6 Existing Applicable Standards
Comment: One commenter said the proposed standard will have little health
impact because a standard for organic anissions from existing chemical
2-14
-------
plants will control VOC emissions to nearly the level of the proposed
standard (Part II Docket Item IV-D-14).
Response: Any standard for VOC emissions from chemical plants would be
designated on a State level as part of a State Implementation Plan (SIP)
and would not necessarily apply statewide. Even if such State regulations
were equivalent to the proposed regulation, total reduction in benzene
emissions would not be as great because plants in attainment areas would
not be required to comply. The only uncontrolled maleic anhydride plant
is in an attainment area. Also, benzene is only negligibly photoreactive
and some States may not choose to regulate specifically emissions from
maleic anhydride plants to achieve the ambient air quality standard for
ozone.
2.2 HEALTH AND ENVIRONMENTAL IMPACTS
2.2.1 Dispersion Modeling
Comment: The dispersion model used by Cramer, unlike the model developed
by CMA, fails to consider site-specific factors, according to one commenter
(Part II Docket Item IV-D-22, Attachment E). Specific factors that cause
the models to differ include:
Variations in product recovery absorber (PRA) emission rates,
Cramer used 2.34 x 10"3 g/s per MT/yr; CMA used an estimated
industry average of 1.86 x 10~3.
Variations in stack height. Cramer used 27.4 meters; four plants
have higher stacks.
Variations in meteorological data. Cramer used worst-case data
for Pittsburgh; CMA data reflected local conditions over an
entire year.
Variations in storage emissions. Cramer assuned storage sources
to be uncontrolled; CMA took account of existing controls on
storage sources.
Response: Based on much more detailed emissions data than it had at
proposal, EPA has substantially refined the dispersion modeling. Since
proposal, EPA has modeled each plant individually, based on the emissions
data supplied by the CMA during the public comment period. The revised
2-15
-------
modeling also takes into consideration variations in stack parameters and
uses meteorological data from STAR* stations near the plants. Since only
process emissions were considered in the promulgated standards, variation
in storage emissions was not a factor in the revised modeling. A more
detailed description can be found in Appendix D.
2.2.2 Current Health Impact
Comment: Because of changes in the maleic anhydride industry (see
Subsection 2.1.1), one commenter (Part II Docket Item IV-F-8, p. 9-12)
contended that EPA's estimate of 0.496 leukemia cases per year under
current control conditions is too high and the correct estimate is 0.0037
leukemia cases per year based on:
Installation of 97 percent control at the Monsanto and Tenneco
pi ants,
Closure of the Koppers (Pennsylvania) and Reichhold (New Jersey)
pi ants,
Use of a 97-percent conversion rate for the Ashland plant,
Use of plant-specific data in dispersion modeling, and
Use of the Lamm risk factor (0.031 death/106 ppb-person-years).
Response: As discussed in the previous response, the dispersion modeling
has been revised to account for the current operating and control status of
the industry and for the plant-specific data provided by CMA. The results
are contained in Appendix D.
As discussed in Response to Public Comments on EPA's Listing of
Benzene Under Section 112 and Relevant Procedures for the Regulation of
Hazardous Air Pollutants (EPA-450/5-82-003), EPA has reviewed the comments
regarding its unit risk factor and the Lamm unit risk factor. In light of
the comments, the Agency has revised its unit risk factor accordingly,
although not to the extent desired by the commenter. The revised unit risk
factor is lower by about 7 percent.
Based on the revised dispersion modeling and the Carcinogen Assessment
Group's revised unit risk factor, the Agency has recalculated the estimated
*STAR (stability array) data are standard climatological frequence of
occurrence summaries formulated for use in EPA models and are available
for major U.S. sites from the National Climatic Center, Asheville, North
Carolina. The data consist of frequencies tabulated as functions of wind
speed stability and wind direction classes.
2-16
-------
leukemia incidence and maximum lifetime risk under assumed current control
levels (see Appendix D). The leukemia incidence per year is estimated to
range from 0.013 to 0.092. The maximum lifetime risk is estimated to range
from 2.9 x 10" 5 to 2.0 x 10*4.
2.2.3 Impact on Ambient Concentrations at USS Chemicals
Comment: One commenter maintained that the proposed standards' impact on
USS Chemicals' maleic anhydride plant would be to reduce maximum annual
mean groundlevel concentration of benzene from 0.00013 ppmv to 0.000087
ppmv.
Response: Revised atmospheric dispersion modeling, which uses plant-
specific parameters provided by industry, estimates the proposed standard
would have reduced maximum annual average groundlevel concentration of
benzene from 0.57 ppm to 0.27 ppm (Part II Docket Item IV-J-16 and
Appendix D).
2.2.4 Risk and Expected Plant Life
Comment: One commenter stated that calculation of estimated lifetime risks
and estimated leukemia cases per year should include consideration of
existing sources' expected operating lifetimes (Part II Docket Item
IV-D-8).
Response: No expected standard operating lifetime for maleic anhydride
plants has been determined (Part II Docket Item IV-D-22). The approximate
risk and leukemia incidence for any time period is obtained by prorating
the lifetime (70 years) risk and annual leukemia incidence values.
2.2.5 Health Impact of Standard at Reichhold
Comment: Reichhold believes the health effects of replacing the current
control system at its Morris, Illinois, plant with a thermal incinerator
that meets the proposed standard (a reduction in leukemia cases due to
process emissions, from 0.000540 to 0.000358 per year) could not be
detected (Part II Docket Items IV-F-5; IV-F-1).
Response: As discussed in Subsection 2.5.1.3, the standard has been
revised and Reichhold is not expected to have to replace its current
control system.
2.2.6 Health Impact of Plant Closures
Comment: According to one commenter, analysis should include human health
benefits from the proposed standard's associated plant closures and
subsequent ending of benzene exposures (Part II Docket item IV-D-8).
2-17
-------
Response: The Agency examines possible health, environmental, and economic
impacts in selecting a regulatory alternative as the basis for a standard.
At proposal, the economic analysis predicted one possible closure because
the plant may not have the capital budget to finance additional benzene
pollution controls (Part II Docket Item III-B-1). Because of this
uncertainty, however, emissions from all existing plants were used to
determine impacts. The promulgation analysis accounts for plant closures
that have been confirmed, such as those of Koppers and Reichhold. Health
benefits due to closure are elimination of illnesses, including leukemia,
related to benzene exposure from those plants. No closures are expected as
a result of the standard, however.
2.2.7 Population "At Risk"
Comment: EPA gave no explanation for choosing a 20-kilometer radius for
assessing human health impacts (Part II Docket Item D-8).
Response: The reason for estimating exposure to within 20-ki1ometers of
stationary sources is based on modeling considerations, not on health
effects criteria nor on EPA policy. Twenty kilometers was chosen as a
practical modeling stop-point. The results of dispersion models are felt
to be reasonably accurate within that distance. The dispersion
coefficients used in modeling are based on empirical measurements made
within 10 kilometers of sources. These coefficients become less applicable
at long distances from the source, and the modeling results become more
uncertain.
2.3 EMISSION CONTROL TECHNOLOGY
2.3.1 Feasibility of 97 Percent Control
Comment: One commenter considered 99 percent control of benzene emissions
possible based on technology transfer and 97 percent control immediately
feasible (Part II Docket Item IV-D-19).
Response: Ninety-nine percent control of benzene emissions is possible and
ninety-seven percent is technologically feasible. However, factors other
than technological feasibility; e.g., environmental, energy, and economic
impacts of retrofitting existing sources are considered in selection of
best demonstrated technology (BDT). When considering these impacts, BOT
for existing maleic anhydride plants was determined to be 90 percent
control of benzene emissions. Selection of BDT is discussed more fully in
Subsection 2.5.1.3.
2-18
-------
2.3.2 Feasibility of Carbon Adsorption
Comment: Reichhold Chemicals' two maleic anhydride plants have each been
equipped with carbon adsorption systems to control benzene emissions. The
commenter stated that neither system has been able to sustain a removal
efficiency of even 96 percent. Reichhold believes that carbon adsorption
cannot be used to meet the proposed standard (Part II Docket Items IV-F-5;
IV-F-l).
Response: Reichhold1s carbon adsorption system at its Morris, Illinois
facility had only three beds, which, as described in Subsection 2.2.1 of
Volume I of the BID and in a plant visit trip report (Part II Docket Item
II-B-01), did not allow sufficient time to complete the regeneration cycle
(including the cooling and drying cycle) before breakthrough can occur.
The technical parameters of a system designed to meet the proposed standard
are also described in Subsection 2.2.1 of Volume I of the BID.
However, since proposal, selection of BDT for existing sources has
been reevaluated to account for additional information received during the
public comment period. As a result of this analysis, the standard for
existing sources has been revised from a 97-percent benzene emission
control level to 90 percent control (see Subsection 2.5.1). Because the
existing carbon adsorption system at Reichhold's Morris, Illinois, plant
can achieve benzene removal efficiencies of 92 to 94 percent (Part II
Docket Item IV-B-1), it could be used to meet the promulgated standard.
2.3.3 Use of Catalytic Incineration
Comment: USS Chemicals uses a catalytic incineration system that removes
up to 97 percent of the waste gas stream's carbon monoxide but could not
meet a 97-percent benzene standard. A thermal incineration system which
would have to replace the catalytic incinerator system to achieve 97
percent benzene renoval may not be able to accomplish as high a carbon
monoxide removal efficiency (Part II Docket Item IV-F-6).
Response: While a thermal incineration system may not achieve as great a
carbon monoxide (CO) reduction, a thermal incinerator could achieve a
greater benzene reduction, which is the standard's purpose. The commenter
did not indicate how much less CO reduction he believed a thermal
incinerator would incur but only said it was questionable. However, since
2-19
-------
proposal, the standard has been revised downward, permitting USS Chemicals
to continue using its catalytic incinerator.
2.4 COST AND ECONOMIC IMPACT
2.4.1 Capital Costs
Comment: CMA stated that EPA's estimates of capital and annualized costs
are low because they are not site specific and because they are out of date
(Part II Docket Item IV-D-22, Attachment E). CMA capital cost estimates,
in millions of dollars, follow:
Company
Incinerator-
battery
1imits
instal1ed
Ductwork,
services
tie-in
Engineering
Si te
preparation,
other
Total
Ashland
—
—
—
7.0
Monsanto
5.1
0.95
0.94
—
7.0
Pfi zer
—
—
—
—
2.0
Reichhold
1.3
0.33
0.23
1.1
2.95
Tenneco
1.1 combined
0.2
0.6
2.0
U.S. Steel
4.4
1.3
0.6
0.95
7.25
Response: Because of data obtained since proposal, the Agency has
reinvestigated original cost estimates. Original EPA cost estimates and
industry estimates differ for several reasons. First, CMA costs (Part II
Docket Items IV-F-8 and IV-D-22) were in either 1980 or 1981 dollars,
depending on the company, while BID (Part II Docket Item III-B-1) costs had
an August 1979 basis. Second, CMA costs are based on plant-specific
parameters supplied to CMA by the individual plants for the public hearing
(Part II Docket Item IV-D-22, Attachment E), while BID costs are based on
costs of equipping a model pi ant with control equipment and scaling up or
down to accommodate capacity differences (Part II Docket Item III-B-1).
Third, CMA costs reflected specific costs for ductwork not included in
EPA's original analysis. For example, USS Chemicals has a ductwork cost of
$1.3 million, as estimated for CMA. EPA representatives who visited the
USS Chemicals, Neville Island, Pennsylvania, plant confirmed the need for
extensive ductwork should a 97-percent control level standard be
promulgated, because of the plant's cramped spatial configuration (Part II
Docket Item IV-B-3).
2-20
-------
Because of the wide range between original Agency cost estimates and
CMA cost estimates for individual plants, the Agency recosted thermal
incineration with 50 percent heat recovery and carbon adsorption control
systems for the following plants: Pfizer, Reichhold (Illinois), and USS
Chemicals. The costing methodology is discussed in Appendix A. Table 2-2
compares original EPA estimates with revised EPA estimates and CMA
estimates for a thermal incineration control device.
Revised EPA costs—lower than CMA estimates—better reflect some of
the plant-specific parameters provided during the public comment period.
These revised EPA costs were prepared using a methodology for costing
chemical process industry control devices (Part II Docket Item IV-A-3).
All revised EPA costs are within 20 percent of CMA costs except for those
of USS Chemical s.
USS Chemicals' costs are significantly higher than EPA's revised
costs, because their costs include an incinerator design capable of
withstanding high pressures of 10 to 15 psi. This design is based on their
experience with a phthalic anhydride incineration system (Part II Docket
Item IV-D-29). Cost estimates submitted by other maleic anhydride
producers did not include costs for such an incinerator design, and there
is nothing unusual about this plant that necessitates its use there. This
design, although safer, is not ordinarily required to meet the 90- or
97-percent air pollution control levels and was not costed. For detailed
information on the standards' economic impacts, see Appendix C.
The industry estimate for Monsanto is based on a system to provide 30
percent of the plant's steam requirements (Part II Docket Item IV-B-2) by
using excess air and fuel and by using boilers. Such a system is much more
expensive than the standard shell and tube heat exchanger used in original
Agency estimates. However, both this system and Tenneco's system are not
being installed to meet these standards and the costs are not attributable
to the standards. Therefore, no capital costs are currently estimated to
be required to meet the standard for this plant.
2.4.2 Medical Costs
Comment: One commenter said the 1.2-percent price increase in maleic
anhydride is negligible when compared to medical costs incurred by local
citizens due to benzene exposure (Part II Docket Item IV-D-4).
2-21
-------
TABLE 2-2. COMPARISON OF CAPITAL COST ESTIMATES
FOR 97 PERCENT CONTROL USING THERMAL INCINERATION
Company
EPA estimate in
BID (Part II
Docket Item
III-B-1)
Revised EPA
estimate
(Attachment A)
Industry estimate
(Part II Docket Item IV-D-22,
Attachment E)
Monsanto
1.73
Not costed
7.0
Pfizer
Not costed
1.68
2.0
Reichhold
(111inois)
1.07
2.59
2.95
Tenneco
0.73
0
0
USS Chemicals
1.75
4.24
7.25
aThis cost is for an Incinerator designed to handle the waste gas from a
plant with twice the capacity of the existing plant.
2-22
-------
Response: The projected price increase at proposal was one of several
factors considered in selection of "best demonstrated technology" (BDT) for
the industry and is considered reasonable. The Agency has not performed
the quantitative cost/benefit analysis alluded to by the commenter because
the uncertainty in risk numbers and medical costs is too great for such an
analysis to be performed accurately. Also, medical cost savings would be
only one of several benefits of a regulation. As a result of revisions to
the standards, however, the price of maleic anhydride is not expected to
increase.
2.4.3 Costs Per Leukemia Case
Comment: According to a Council on Wage and Price Stability study, the
cost per leukemia case avoided would be $3 million at 90 percent control
and $35 million at 97 percent control (Part II Docket Item IV-D-5).
Another commenter believed the cost would be $4.5 billion per leukemia case
(Part II Docket Item 1V-D-22).
Response: These standards are based on BDT. The selection of BDT is based
on technological and economic factors and not a cost/benefit analysis of
specific health effects. Even for the residual risk analysis, however,
because of the shortcomings of quantitative risk assessment and the
inability of cost-per-leukemia case calculations to deal with unacceptably
high risks to smaller groups, EPA does not believe that cost-per-leukemia
case calculations are appropriate as the principal means of assessing
residual risks. A more detailed discussion of the general methodology for
selecting the basis of the standard is found in Response to Public
Comments on EPA's Listing of Benzene Under Section 112 and Relevant
Procedures for the Regulation of Hazardous Air Pollutants
(EPA-450/5-82-003).
2.5 SELECTION OF BASIS OF THE STANDARD FOR EXISTING SOURCES
2.5.1 Basis of Standard for Continuous Emissions
2.5.1.1 Generic Benzene Control.
Comment: One commenter contended that EPA should evaluate and report the
costs and risk reduction benefits from generic controls; i.e., good
housekeeping rules, before regulating individual source categories
(Part II Docket Item IV-D-23, Attachment, p. 3).
2-23
-------
According to another commenter, EPA should permit consideration of
cost and health benefit comparisons before making final decisions on
control requirements more stringent than "generic" controls and should
investigate a full range of control levels in analyzing regulatory options
(Part II Docket Item IV-D-23).
Response: The generic standards mentioned above are those described in the
proposed Airborne Carcinogen Policy (Proposed Rulemaking 44 FR 58654):
"Upon the listing of a substance, previously developed
generic standards will be proposed for source categories of that
substance to which they could apply. Generic standards,
developed based upon the similarities among industrial processes,
will be 'tailored' as necessary to fit the source categories for
which they are proposed."
Benzene was listed as a hazardous air pollutant and standards
development for individual source categories began before the concept of
generic standards was proposed. The purpose of the generic standards would
be to apply easy and inexpensive emission-reducing steps until standards
for specific source categories can be developed. Therefore, they would
constitute a temporary, interim step and are not implemented in lieu of
source-specific standards. Sufficient information was (is) available to
develop permanent (as opposed to interim) standards for benzene sources, so
that any interim measure is unnecessary. The generic standards proposed
with the Airborne Carcinogen Policy were for fugitive emission sources, not
stack sources. A standard for benzene fugitive emission sources has been
proposed and would apply to maleic anhydride plants as well as to other
plants. For both fugitive emissions and stack emissions, cost and emission
reductions were considered in the selection of BDT. As explained in
Subsection 2.10.2, EPA believes it is appropriate to apply at least the
same level of control for carcinogens under Section 112 as it would apply
to new sources of criteria pollutants under Section 111(b) or recommend for
existing sources under Section 111(d). This would involve applying BDT to
each emission source in a plant type before considering a control level
beyond BDT. The commenters suggest requiring much less than this. They
would apply less than BDT to a particular kind of source within a plant and
then balance risks against costs to decide whether additional controls
2-24
-------
should be applied to that source, and if any controls should be applied to
the other sources within a plant type.
2.5.1.2 Economic Reasonableness of Standard.
Comment: According to one coinmenter, a regulation cannot be judged
reasonable simply because most or all existing manufacturers happen to
possess adequate resources and incentives to withstand the regulation's
initial economic impact and recoup the costs later. An industry's ability
to survive initial economic impact of regulation may depend more on the
industry's economic importance and its products than on the standard's
reasonableness as a public health measure (Part II Docket Item IV-D-22).
Response: The Agency considers inflationary impacts, intraindustry
competition impacts, installed capital cost, and total annualized cost in
evaluating economic impacts of a standard. These impacts were discussed at
proposal and based on these impacts, a 97-percent control level was
selected as the proposed standard. However, in light of comments regarding
costs received during the public comment period, this decision was
reassessed and a 90-percent control level was selected as the promulgated
standard (see Subsection 2.5.1.3). In any case, the coinmenter provided no
supporting data or information and did not elaborate on additional factors
that should be considered.
2.5.1.3 Selection of Regulatory Alternatives.
Comments: According to several commenters, EPA should investigate a full
range of control levels for all technologically feasible options in
regulatory options analysis. They felt EPA should have considered a
standard based upon 90 percent (or lower) control because, at that time,
five of the eight existing rnaleic anhydride plants already had 90 percent
or better control. They stated that the cost of changing from 90 to 97
percent control would not be justified by the reduction in benzene
emissions. Requiring 90 percent control would achieve 94 percent of the
health benefits (as measured by leukemia cases averted) at roughly 60
percent of the cost of a 97-percent standard (Part II Docket Items IV-D-23
and IV-D-21).
Response: The proposal (45 FR 26660, April 18, 1980) examined three
alternative benzene emission control techniques for maleic anhydride
plants: conversion to an n-butane feedstock, carbon adsorption, and
thermal incineration. Information on actual and potential use of these
2-25
-------
techniques served as the basis for the following regulatory alternatives:
1. 97 percent benzene control, based on best danonstrated
control level now being achieved at an existing maleic anhydride
plant and universally applicable to existing plants;
2. 99 percent benzene control, based on technology transfer; and
3. No detectable benzene emissions, based on conversion to an
n-butane feedstock.
Regulatory alternatives 1 and 2 were considered a viable basis for a
standard for existing sources. For new sources, an additional regulatory
alternative of no detectable benzene emissions (100 percent control) was
considered a viable basis for a standard.
In selecting BDT (called BAT at that time; see Subsection 2.10.2) for
existing sources at proposal, the .Administrator examined impacts of the two
regul atory alternatives for existing sources, A standard based on 99
percent control was estimated to reduce benzene emissions by 5,860 Mg/yr,
resulting in total capital costs of about $9.1 million, increased total
annualized costs of about $4.5 million, a .potential price increase in
maleic anhydride of 1.7 percent, increased energy usage of 85,000 barrels
(bbl) (525,000 GJ/yr or 4.97 x 1011 Btu/yr) of oil per year, and as many as
two projected plant closures. A standard based on 97 percent control to
reduce benzene emissions by 5,370 Mg/yr would have resulted in total
capital costs of about $6.6 million, increased total annualized costs of
about $2.5 million, a potential price increase in maleic anhydride of 1.2
percent, increased energy usage of about 50,000 bbl (310,000 GJ/yr or 2.94
x 10^ Btu/yr) of oil per year, and as many as one projected plant closure.
Imposing a less stringent control level would not have prevented this plant
closure.
The Administrator arrived at the following conclusions. First, 97
percent control was considered the best danonstrated technology thus far to
be achieved at an existing maleic anhydride plant with a control system
applicable to all other existing plants. Higher levels, such as 99
percent, were believed to be technically feasible, but only with technology
transfer and at higher cost and energy use. In addition, one plant not
projected to close with a 97-percent control standard was projected to
close with a 99-percent control standard if it could not successfully
2-26
-------
convert to n-butane. For these reasons, the Administrator selected the
97-percent control alternative as BDT.
After the 97-percent alternative was identified as the basis for BDT
for existing sources, estimated risks remaining after applying this
alternative were examined in view of health benefits and costs resulting
from a more stringent alternative.
Relatively small health benefits would have been gained for the cost
of an additional projected plant closure associated with requiring 99
percent rather than 97 percent control for process vents. Technical and
economic impacts associated with a substitute feedstock's requirements for
existing maleic anhydride sources were uncertain. For these reasons, the
Administrator concluded at proposal that risks remaining after application
of BDT to existing sources were not unreasonable and decided not to require
more stringent control than BDT for process vents.
EPA cost estimates at proposal reflected the higher costs for both
Reichhold (Illinois) and USS Chemicals plants because they currently have
at least 90 percent but not 97 percent control. Thus, these two plants
would have additional costs for removing the control device and installing
a new one. However, based on industry comments, original cost estimates
were revised, and costs for these two plants increased significantly.
Consequently, EPA examined a 90-percent control regul atory alternative
in addition to the options considered at proposal. The 90-percent control
level was examined because both the Reichhold (Illinois) and USS Chemicals
plants were achieving this level (Part II Docket Items II-A-7, II-A-11,
IV-B-1, IV-B-3, and IV-F-8). Consequently, these plants could meet a
90-percent standard without the high retrofit and other costs a 97-percent
standard would require. Table 2-3 summarizes impacts of the 90- and
97-percent alternatives and changes in the maleic anhydride industry that
have occurred since proposal.
To estimate impacts, EPA assumed the average conversion rates of the
plants were those provided by the industry (Part II Docket Item IV-D-22).
For Monsanto, an average conversion rate of 95 percent was assumed. Based
on these assumptions regarding control levels and conversion rates,
industrywide process vent benzene emissions are now estimated to be 960
Mg/yr.
2-27
-------
•
TABLE 2-3. IMPACTS
OF REGULATORY ALTERNATIVES FOR
EXISTING PLANTS
Cost effec-
All
owable benzene
Total
tiveness ($/Mg
Energy impactc
process
emissions (Mg/yr)
Total installed cost3
annualized cost*
removed per year)
(GJ /yr)
(J 1,000s)
¦(Jl.OOfc/yr)
90S 97%
Benzene
Current,
90S 974
Control 901 97i
90% 97X
control control
Plant and
Conversion
unregul ated
control control
efficiency control control
control control
(7) (81
90% 97%
location
Rateb
(1)
(2) (3)
(4) (5) (6)
(7) (8)
(1H2) {i)-t3j
control control
Koppers, PAd
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Reichhold, NJd
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Pfizer, IN6
95
650
65
20
0
1,260/
1,680
1,260/
1,680
235/
719
462/
719
401/
1,227
732/
1,140
7,300/
61,600
y 26,000/
61,600
Ashland, WVf
HA
0
0
0
NA
NA
NA
NA
NA "
NA
NA
0
0
Tenneco, NJ9
96.3
4.5
4.5
4.5
99
30*
301
8h
81
NA
NA
0
0
Relchhold, IL
94
150
150
45
90
3(P
1,540/
2,590
8h
738/
830
NA
7,235/
8,137
0
30,400/
70,000
OSS Chemicals, PA
97
130
130
39
90
3(P
3,354/
4,240
8h
1,168/
1,970
NA
12,559/
21,183
0
87,000/
123,000
DENKA, TXf
NA
0
0
0
NA
NA
NA
NA
NA
NA
NA
0
0
Monsanto, MO
95
69
69
69
97
30h
301
8h
8'
NA
NA
0
0
Amoco, II
NA
0
0
0
NA
NA
NA
NA
NA
NA
NA
NA
NA
Total
1.000J
42$
ie(P
1,380/
1,800
6,210/
8,570
267)
751
2,380/
3,540
460/
1,300
2,900/
4,300
7,300/
61,600
143,000/
255,000
NA: Not applicable.
a£ach pair of costs is based on carbon adsorption/thermal incineration with 50 percent heat recovery for 8,000 hr/yr.
bProvided by industry.
cBased on calculations in Attachment B.
dPlant closed permanently since proposal.
®fumaric acid plant not included in the original proposal.
'Boths plants have converted to n-butane production.
^New Jersey will require at least 99 percent control of benzene emissions.
iplant will meet 90 percent control level; only additional costs are for continuous monitoring.
'plant will meet 97 percent control level; only additional costs are for continuous monitoring.
^Emission totals rounded to two significant figures.
-------
In reexamining selection of the regulatory alternative for process
vents, EPA first examined 97 percent control, the basis of the proposed
standard. At proposal, EPA economic analysis projected this control level
might cause one plant closure due to substantially decreased profitability.
However, analysis indicated that if this plant closed as a result of the
standard, it would close regardless of level selected. Because of the
closure of a nearby maleic anhydride plant, this plant's competitive
position has improved and the likelihood of its closure has decreased.
This plant is required by a State Implementation Plan to reduce its benzene
emissions by 99 percent. Therefore, if it closed, it would close
regardless of whether or not this standard were promulgated.
A second plant has stated that it would probably close if the
proposed standard were promulgated (Part II Docket Item IV-D-22). If a
97-percent standard were promulgated, this plant would have to scrap its
present control system, which achieves 92 to 94 percent control, and
replace it with a 97-percent control device. Therefore, this plant more
likely would consider closing if a 97-percent standard were selected than
if a 90-percent standard were selected.
The 97-percent control regulatory alternative would reduce benzene
emissions by about 820 Mg/yr (900 tons/yr). Total installed capital costs
would be $8.6 million, and total annualized costs would be $3.5 million/yr.
The 90-percent control regulatory alternative would reduce emissions by
about 580 Mg/yr (640 tons/yr). Total installed capital costs would be $1.8
million, and total annualized costs would be $0.8 million/yr. Derivations
of the cost estimates are contained in Appendix A. The cost difference
between the two regulatory alternatives is due entirely to the two plants
that could meet the 90-percent control alternative with no additional
controls but that would have to replace current with new equipment to meet
a 97-percent control alternative. Reichhold would have to replace an
existing control device at a capital cost of $2.6 million to remove an
additional 100 Mg/yr (110 tons/yr) of benzene for an annualized cost of
$0.3 million. USS Chemicals also would have to replace an existing control
device at a capital cost of $4.2 million to remove an additional 93 Mg/yr
(100 tons/yr) for an annualized cost of $2.0 million.
2-29
-------
A 90-percent control level would require an additional 40,800 GJ/yr
{3.8 x lO^O Btu/yr) of energy, while a 97-percent standard would require an
additional 163,800 GJ/yr (1.5 x 1011 Btu/yr). These nunbers represent
energy impact for thermal incineration with 50 percent heat recovery. (See
Appendix B for details.)
After considering these costs and emission reductions, EPA concluded
that the additional emission reduction associated with the regulatory
alternative of 97 percent control compared to 90 percent control and cost
impacts associated with the two alternatives are unreasonably high.
2-29-A
-------
Reducing emissions further, from 90 percent reduction to 97 percent
reduction, would cost an additional $3,000 per rnegagram. In addition, 90
percent control would require one uncontrolled plant to install controls
while not imposing burdensome retrofit costs on two plants that currently
have controls and that would realize a relatively small emission reduction
if this control system were replaced with more efficient systems. Also,
selection of a 97-percent regulatory alternative would increase the
probability of one of these two plants closing. After considering these
impacts, the Administrator selected the 90-percent regulatory alternative
as the BDT for controlling continuous process vent emissions.
Accordingly, the numerical emission limit would also have to be
revised. In setting the emission limit, one must consider the conversion
rate as well as the percent emission reduction. Conversion rates typically
decrease over the lifetime of the catalyst. Therefore, selecting an
average conversion rate upon which to base the limit would mean that much
of the time--especially toward the end of the catalyst life--a plant would
be unable to meet the standard even if its control device were 90 percent
efficient. Consequently, a conversion rate below that expected toward the
end of the catalyst's useful life must be assuned to allow a plant to use a
catalyst through its normal life cycle. CMA has stated that the lowest
conversion rate within the industry is about 91 percent (Part II Docket
Item IV-D-22). Therefore, for the purpose of establishing a numerical
emission limit, a 90 percent conversion rate was assuned. Thus, the
resulting anission limit, based on 90 percent conversion and 90 percent
emission reduction, would be 1 kg benzene per 100 kg benzene feed.
2.5.1.4 Selection of Control System.
Comment: One commenter said the proposed standard should show preference
for or require control systems that allow the fewest total emissions
(normal plus excess) and limit either total annual emissions or total
excess emissions (Part II Docket Item IV-D-8).
Response: Section 112 of the Clean Air Act requires the Agency to set
numerical emission standards. The Agency can require specific systems or
equipment only if a numerical standard cannot be set. The level of the
standard is based on analyses of different control systems, discussed in
Chapter 2 of Volume I of the BID.
2-30
-------
The standard-setting procedure used bases the standards, at a minimum,
on BDT, considering the environmental, economic, and energy impacts.
Consequently, the standards do in fact show preference for control systems
that minimize emissions, but if the impacts of basing the standard on such
a system are adverse, a less stringent system would be selected. In this
case, because the cost of the proposed standards was considered
unreasonable, the Administrator selected a less stringent control level as
the basis of the standard, although by definition it is still BDT
considering environmental, economic, and energy impacts.
2.5.2 Basis of Standard for Excess Emissions
2.5.2.1 Startup Emissions Allowance.
Comment: In post-hearing comments (Part II Docket Item IV-D-22), CMA
stated that during reactor startup the benzene oxidation reaction may have
delayed or only partial startup. Also, during reactor startup there is an
unquantifiable risk that either the air or benzene flow control instrument
will not control the flow accurately and that too high a ratio of benzene
to air will be fed to the reactor. If benzene feed were inaccurately
controlled and benzene concentrations leaving the reactor were at or near
the lower explosive limit (LEL) due to incomplete oxidation, an explosive
mixture could be fed to the incinerator, possibly causing a flashback and
explosion in the PRA off-gas ducting, the heat exchanger, and the PRA.
Several commenters stated that the standard should include both an
allowance to bypass control equipment for up to 8 hours during startups and
to bypass control equipment for up to 1 1/2 hours during individual reactor
startups.
USS Chemicals, which uses a catalytic oxidation system to control
benzene emissions, noted (Part II Docket Item IV-B-3) the tendency for a
higher than normal benzene level in the reactor tail gas during reactor
startup (due to low benzene conversion rates and minimal dilution air) that
could overheat the system and damage the catalyst and equipment.
For thermal incinerators, the fuel content of the incinerator's
product recovery absorber (PRA) off-gas may be sufficient to increase the
incinerator's operating temperature suddenly by up to 200° C, causing
temperature and other safety controls to signal high-temperature shutdown
(Part II Docket Item IV-F-8).
2-31
-------
Both DENKA and Monsanto stated (Part II Docket Items IV-F-1, IV-F-8,
and IV-D-25) that they need the flexibility to start up the inaleic
anhydride process independently of control device startup to ensure safe
and smooth startups.
Response: The request for 8 hours of allowable excess emissions during
total plant startup is based on DENKA's experience with its thermal
incinerator (Part II Docket Item II-B-22). They stated in their testimony,
that if during startup the incinerator is operating at or near the normal
operating temperature of 760° C (to achieve required benzene control), the
incinerator's temperature when the waste stream initially is introduced may
increase enough due to energy content of the off-gas, thus causing
temperature and other safety controls to signal high-temperature shutdown.
The request for allowing 1.5 hours of excess emission for individual
reactor startups is based on USS Chemicals' experience (Part II Docket Item
IV-B-3). USS Chemicals requires 30 minutes to start up and stabilize each
individual reactor and an additional 60 minutes to stabilize the
incinerator. The time to stabilize the incinerator may be shortened by
venting the reactor fail gas to the atmosphere for about 1 hour and then
introducing it to the control system when operating conditions have
stabilized. These excess emissions during startup occur at USS Chemicals'
plant about 35 times per year for the control unit serving three reactors
and about 20 times per year for the control unit serving one reactor. (The
plant has two separate catalytic oxidizers.)
The Agency has noted that starting up the maleic anhydride process or
the incinerator requires the full attention of plant personnel to limit the
possibility of explosion. Based on comments received during the public
comment period, startup of a maleic anhydride plant depends on a series of
critical procedures, each requiring complete concentration of available
personnel. Incinerator startup also requires complete concentration of
available personnel. Because benzene is explosive, particular care must be
paid during each procedure to benzene feed rates and evidence of incomplete
or partial benzene oxidation. Attempts to accomplish these procedures
concurrently would burden available plant personnel. The normal LEL for
benzene is 1.5 mole percent, depending on temperature. If the initial
2-32
-------
conversion rate is less than 50 percent, the off-gas concentration is
greater than 33 percent of the LEL. Feeding a benzene stream of this
concentration to an incinerator would create an explosion risk, which is
exacerbated because the LEL drops as temperature rises. Flame arrestors
are not practical in these situations because they would not prevent
benzene explosions in the ductwork associated with the high temperature
generated by the control device. Probability of such explosions is small
but high enough for concern because such explosions could destroy a large
part of a plant.
According to CMA, during initial reactor startup partial oxidation
(causing incomplete benzene conversion) may occur due to uneven heating of
the tubes inside the reactor that contain the catalyst, malfunctioning
catalyst, inadequate benzene circulation, or other similar situations.
In addition, a catalyst often responds "sluggishly" to initial benzene
introduction to the reactor (Part II Docket Item IV-E-4), allowing a slug
of unreacted benzene to move through the system. Standard instrunentation
used at a plant would not detect the slug in time to respond to it. This
slow catalytic response would produce higher than normal benzene levels in
the gas leaving the reactor and in the PRA during startup. These slugs
would again present the risk of approaching the LEL in the incinerator.
Other causes of incomplete benzene conversion during startup include (Part
II Docket Item IV-B-3) impurities, such as toluene, in the benzene feed,
incorrect benzene feed control system operation, incomplete benzene
reaction in portions of the reactor, and uneven heating of the reactor.
Another difficulty during maleic anhydride plant startup is due to the
continuous nature of maleic anhydride plants (Part II Docket Item IV-E-4).
During startup, a system designed for continuous operation may require
cleaning of plugged ducts or valves, release of stuck valves, repair of
leaks, restarting of stubborn motors or fans, or similar measures.
Although any one of these problems would be solved easily during normal
operation, any combination may occur simultaneously during startup,
straining plant personnel resources. When a plant encounters problems
associated with incinerator startup at the same time it encounters problems
associated with reactor startup, its difficulties double (Part II Docket
Item IV-E-4)-
2-33
-------
After considering these comments, the Administrator, recognizing the
difficulties of achieving a relatively consistent continuous reaction,
determined that BDT for maleic anhydride plants does not include control of
benzene emissions during reactor startup.
2.5.2.2 Shutdown Emissions Allowance.
Comment: One commenter stated that routine shutdown of a maleic anhydride
reactor could result in a short period of excess emissions from control
systems that treat the tail gas from two or more reactors. The relatively
sudden change in load on the control unit, which would occur when the
benzene feed to the reactor is shut off, may upset operation of the
incinerator1s temperature controls and cause excess emissions.
Consequently, allowance for periods of excess emissions during routine
shutdown is a practical necessity and would help prevent the excess
emissions that would result from unnecessary reactor shutdowns or startups
required by the proposed standard (Part II Docket Items IV-F-6; IV-F-1, p.
113).
Response: Benzene feed to the reactor can be reduced gradually rather than
suddenly. Gradual reduction would allow time for a catalytic or thermal
incinerator to adjust to the changed benzene loading and continue to
achieve required benzene ranoval until all benzene feed had completely
stopped. Consequently, BDT for controlling excess emissions during routine
reactor shutdown remains as proposed; i.e., no excess emissions would be
allowed.
2.5.2.3 Shutdown Time.
Comment: One commenter stated that EPA should use more than 1 1/2 hours as
the standard time for shutting down three reactors used in calculating the
controlled shutdown mass emission limit (Part II Docket Items IV-F-8;
V-F-l, p. 199-200).
Response: The proposed standard did not specify a required shutdown time.
The time period the commenter referred to is the estimated time for
shutting down three reactors and was used only to calculate the proposal's
impacts. Under the proposal, each pi ant would have specified its own
shutdown mass emission limit based on its shutdown procedures. However, as
described in the next response, the Administrator does not require in the
promulgated standard calculation of emissions from controlled shutdown.
2-34
-------
2.5.2.4 Control Device Malfunctions.
Comment: Several commenters stated that if all control system malfunctions
necessitate shutoff of benzene feed, disruptive benzene feed interruptions
may occur, followed by startup problems due to items that could be
corrected in a short time. Such malfunctions may cause few or no excess
emissions but would still require immediate benzene shutoff under the
proposed regulations, according to the commenters.
Therefore, to permit diagnosis and correction of most problems without
feed interruption and associated restarting problems, the commenters
recommended that provisions be incorporated allowing continued operation
for 6 hours following malfunction of the emission system. (Part II Docket
Items IV-F-6; IV-F-8; IV-D-25; IV-F-1, p. 123).
Response: The proposed standard would not have required immediate shutdown
in the event of a control system malfunction but would have required that
emissions during malfunction be no greater than if the plant were shut
down. The plant thus could have continued to operate if the malfunction's
cause could be identified and repaired quickly. In such events emissions
would be fewer if the plant remained operative than if it shut down.
Emissions allowed would have been calculated by each plant, assuming full
capacity. By decreasing emissions during malfunctions, an operator could
extend repair time for a minor problem either by decreasing the production
rate or by increasing the conversion rate of the reaction. Additionally,
since plants seldom operate at full capacity, emissions would be fewer than
under the full-capacity emission estimate.
However, based on several industry comments, EPA reevaluated the
proposed excess emission requirement for control system malfunction. In
doing so, the Agency examined possible malfunctions and expected
durations.
The type of malfunction depends on the type of control device used.
Thermal incineration, catalytic oxidation, and carbon adsorption control
systems currently are used at maleic anhydride plants and are the only
denonstrated control techniques for existing benzene-based plants.
2-35
-------
1. Malfunctions of Catalytic Incineration Systems
USS Chemicals identified the following problans with its two
catalytic incinerators (Part II Docket Item IV-B-3). Controls may
malfunction or misread a parameter, automatically diverting waste gases to
the atmosphere. During normal operating hours, this malfunction can be
checked and repaired within 1 hour. However, if the problem occurs at
night or on weekends, 3 to 4 hours may be required to contact appropriate
personnel and arrange overtime work.
A problem related to catalyst life and use is masking the catalyst and
inhibiting or preventing chemical reaction due to deposition on the
catalyst of noncombustible material entrained in the PRA off-gas. Masking
the catalyst reduces its control efficiency, a problsn usually prevented
with a well-designed filter/mixer element located between the burner and
the catalyst. This filter is cleaned during normal shutdown.
2. Malfunctions of Carbon Adsorption Systems
Reichhold Chemicals has experience with carbon adsorption control
systems on two maleic anhydride plants, neither of which is still operating
(Part II Docket Items II-D-47 and II-D-48). Based on Reichhold1s
experience, most control device failures usually affect only one of three
carbon beds and may result from cycle timer malfunction, control valve
malfunction, or partial collapse of a carbon support bed. These
malfunctions can be corrected with reduced benzene feed rates and normal
operation of the renaining two units while the affected unit is repaired.
These repairs usually can be made in 1 to 4 hours.
Complete failure may be caused by ruptured disc opening, condenser
freezeup, and condenser temperature control malfunction.
3. Malfunctions of Thermal Incineration Systems
DENKA has noted that fire eyes or other flame-sensing devices may
become dirty, plugged, or otherwise inoperable (Part II Docket Item
II-B-42), which may cause an incorrect reading that signals the temperature
controller to shut off natural gas (or oil) feed to the incinerator. Until
the problem is diagnosed and corrected—usually within 2 hours—excess
benzene emissions will occur. Natural gas or other fuel impurities may
result in total flameouts, which industry predicts may occur up to 3 to 4
2-36
-------
times per year. Correction can be completed within 4 to 6 hours. Other
potential problems with thermal incineration systems are reviewed in the
following paragraphs.
Corrosion of exposed metal within the incinerator and associated
ductwork may affect gas flow and mixing (Part II Docket Item IV-B-4).
Problems can be diagnosed in a few hours unless the firebox must be
entered. In that case, several days are required to cool the firebox to
ambient temperatures, and repairs may require 1 to 4 weeks, depending on
location, nature, and extent of corrosion.
Plugged burners due to ash and carbon buildups may affect firebox flow
patterns, reducing destruction efficiencies (Part II Docket Item IV-B-4).
This problem can be diagnosed in a few hours and usually can be fixed
quickly; e.g., by blowing compressed air through the burners. Obviously,
major blockage or deterioration of the burner will require complete
incinerator shutdown for days or weeks.
Heat exchanger problems range from simple defects to breakdown of
recuperative heat exchanger seals (Part II Docket Item IV-B-4). These
problems may be caused by general corrosion, warping due to hot spots, or
improper heating and cooling during startup and shutdown. Such problems
may cause inlet waste gas to leak into the outlet flue gas without first
passing through the firebox. Diagnosis may take 1 to 4 hours, but repair
may require a day or weeks, depending on location and extent of damage.
For repairs that can be diagnosed and repaired in hours, no more than 4 to
6 hours of excess emissions would occur for each malfunction. Procedures
required to repair a minor malfunction and to restart a thermal
incinerator, based on Monsanto's experience (Part II Docket Item IV-B-2),
are shown in Table 2-4.
Based on experience with other oxidizers, Monsanto predicts most
malfunctions will be minor, requiring perhaps 1 hour to find and correct.
Total flameouts will be infrequent—approximately three to four per
year--and could take up to 6 hours to correct. Because 30 to 40 percent of
total steam supply will be generated by benzene oxidation at Monsanto's
J. F. Queeny plant, frequent shutdown of the maleic anhydride process would
burden not only that process but others conducted at the site. According
2-37
-------
TABLE 2-4. TROUBLESHOOTING TIME FOR THERMAL INCINERATOR
(Total Venting Time Required)
Process Time
1.
Open scrubber vent valves and close process damper
5 minutes
la.
Correct shutdown problem
30 to 90 minutes
2.-
Start up ambient air fans for oxidizer
10 minutes
3.
Purge oxidizer
10 minutes
4.
Ignite pi lot—burner A
5 minutes
5.
Establish main flame—burner A (low fire)
5 minutes
6.
Ignite pilot—burner B
5 minutes
7.
Establish main flame—burner B (low fire)
5 minutes
8.
Reheat oxidizer to 1,500° F
120 minutes
9.
Direct process from east scrubber to oxidizer
5 minutes
10.
Establish stable oxidizer operation
30 minutes
11.
Direct process from west scrubber to oxidizer
5 minutes
12.
Establish stable oxidizer operation
30 minutes
265 to 325 minutes
4 to 6 hours required
NOTE: Nuisance trip would require at minimum "1 hour" of reduced destruction
efficiency. Nuisance trip would be single-burner shutdown (based on
Texas City experience).
Source: Part II Docket Item IV-B-2.
2-38
-------
to industry, the regulation should allow continued operation for a short
period after the emission control system malfunctions. Continued operation
would permit diagnosis and correction of most problems without feed
interruption and associated restarting problems.
Although industry believed benzene feed would have to be decreased or
discontinued, such was not the case under the proposed standards. Some
flexibility would have existed because allowed shutdown emissions would
have been based on rated plant capacity and not on current production.
Minor problems cited by commenters could be corrected within the short time
periods they described without adjusting feed to the reactor. However, the
problem would almost have to be diagnosed and repairs started immediately,
requiring constant and immediate availability of appropriate diagnostic and
maintenance personnel, as noted by the commenters. Such a requirement is
unreasonable. Consequently, while not necessary, feed probably would be
adjusted during a control system upset, as a practical procedure under such
conditions.
Even if feed adjustments were made, plant personnel could still find
themselves in a difficult position under the proposed excess emission
requirements.
For example, under the proposed excess emission requirements, if a
typical plant could cut its benzene feed rate to 50 percent of capacity
immediately, it could continue operating for 2.5 hours without exceeding
its shutdown limit. However, in little over an hour plant personnel would
have to know whether or not the breakdown could be repaired in the
remaining time. Table 2-5 shows shutdown emissions for the model plant.
The estimated one hour diagnostic time for the model plan was calculated
using the mass emission shutdown limit of 253.2 kg benzene. If the benzene
feed rate is reduced by 50 percent, 126.5 kg benzene emissions would be
allowed during shutdown while the remaining 126.6 kg benzene emissions
would be allowed during a diagnostic period. If the plant emits 94.8 kg/hr
of benzene at 50 percent feed, then 1.3 hours of diagnosis time would be
allowed.
Because benzene feed reduction cannot be accomplished instantaneously,
plant personnel also would be involved in cutting the feed rate during part
of the first hour. Clearly, requiring maintenance personnel to locate and
2-39
-------
TABLE 2-5. SHUTDOWN EMISSIONS AT FULL CAPACITY AND AT 50 PERCENT CAPACITY
Shutdown
(When operating at full capacity)
Operating at
Shutdown
(When operating at 50% capacity)
Hours
Reactor
Period of
uncontrolled
emissions
(min)
Uncontrolled
emissions3
(kg/hr)
50% capacity
Uncontrolled
emissions3
(kg/hr)
Period of
uncontrolled
emissions
(min)
Uncontrolled
emissions3
(kg/hr)
1
1
20
21.1
31.6
20
10.5
2
60
63.3
31.6
60
31.6
3
60
63.3
31.6
60
31.6
2
1
0
0
31.6
0
0
2
20
21.1
31.6
20
10.5
3
60
63.3
31.6
60
31.6
3
1
0
0
31.6
0
0
2
0
0
31.6
0
0
3
20
21.1
31.6
20
10.5
TOTAL
Cumulative = 240
Elapsed = 140
253.2
284.9 Cumulative = 240
Elapsed = 140
126.4
a Normal uncontrolled emission rate for a typical plant is 63.3 kg/hr at full operating capacity.
-------
characterize the failure in 1 hour while much of the plant's personnel is
involved in cutting the feed rate would be unreasonable.
The proposed excess emission limits would not have provided adequate
time to diagnose control system malfunctions. Because the Administrator
has determined the proposed limit would unduly burden the plant operator,
it does not appear as viable a regulatory approach formaleic anhydride
plants as perceived at proposal. Consequently, the Administrator concluded
that the proposed excess emissions requirements were unreasonable and
considered the commenters1 suggestions.
Based on the above discussion of malfunctions, the Agency agrees that
6 hours is a reasonable length of time to find and repair most
malfunctions. Malfunctions not repairable within 6 hours are likely to
last much longer. Since the Agency believes it is reasonable to allow
plants to continue operation for a short time during a nonmajor malfunction
rather than to shut down and because the latter would result in more
emissions, BDT has been revised to allow a maximum of 6 hours of excess
emissions due to malfunction, based on a 6-hour period for a nonmajor
malfunction repair, or shutdown if a major malfunction. In other words, at
the end of 6 hours after onset of an upset, the plant operator must either
have corrected the upset or have shutdown the reactor(s), or he would be in
violation of the standard. It should be noted that if a plant had to
shutdown due to malfunction, the subsequent startup time would not be
included in the 6 hours allowed for malfunctions. However, it should be
emphasized that the plant owner or operator would have to demonstrate to
the Administrator that an upset was a malfunction; i.e., a sudden and
unavoidable upset.
This provision in the final standards would have some advantages over
the proposed provision. Such a provision would be less complicated and
would apply uniformly to all the plants, instead of each plant having its
own limit, as at proposal. Also, an owner or operator would not be
required to submit a detailed description of the plant's shutdown
procedures and estimated emissions. Therefore, the objective of the
standard, limiting excess emissions to a reasonable extent, could be met
with less burden.
2-41
-------
2.5.2.5 Malfunction Definition.
Comment: One commenter noted that control system design deficiencies that
result in violation of the standard are considered by definition to be
preventable. He thought that this definition would subject operators to
penalties even though it is often impossible to determine design flaws in
the absence of operating experience and that the phrase "deficiencies in
design" should be deleted from the definition of control system malfunction
in Section 61.91(b) (Part II Docket Item IV-D-25). Conversely, another
commenter supported the proposed standard's definition of "control system
malfunction" (Part II Docket Item IV-D-8).
Response: Design deficiencies are those that typically occur in the
construction or installation of the control system and includes such things
as improperly sized pollution control devices or improper installation of
piping. Because such deficiencies are avoidable during construction and
installation through careful planning and proper engineering calculations,
which normally accompany capital expenditures of this magnitude, they are
correctly excluded from being considered malfunctions, which by definition
are unavoidable.
There is no evidence to indicate the existence of any design flaws in
any existing control device that would prevent a plant from achieving the
standard on a consistent basis. Additionally, the only plant projected to
have to install a control device to meet the standard would likely install
an incinerator, a control device with which the chemical industry has much
experience and familiarity. Even in the unlikely event that a design
deficiency caused periods of excess emissions, it would be inappropriate
for the Agency to allow such occurrences, which can be corrected.
Furthermore, the commenter did not elaborate on what deficiencies might
occur, nor did he provide examples of past known deficiencies.
2.5.2.6 Determination of Control Device Malfunction.
Comment: One commenter maintained that EPA's determination of whether or
not a control device malfunction was unavoidable could not be accurate.
Plant inspections would be as effective as the proposed procedure in
limiting excess emissions. Exceeding a certain number of occurrences of
excess emissions could be used to determine the need for an inspection by
State or Federal agencies (Part II Docket Item IV-D-21).
Response: Although determining whether or not control device malfunctions
2-42
-------
are unavoidable may be difficult, the Administrator has concluded that such
an approach is better than allowing excess emissions regardless of cause.
Such an approach provides incentive to operate and maintain the air
pollution control system properly and would conform to BDT for reducing
emissions.
2.5.2.7 Control Device Maintenance.
Comment: One commenter stated that excess emissions should not be limited
as a result of control device malfunction. Instead, adequate maintenance
of the control device should be required (Part II Docket Item IV-D-15).
Response: Although required under Section 61.92(b) of the standard and
helping to limit emissions possible during preventable upsets, proper
operation and maintenance cannot prevent malfunctions, which by definition
are unavoidable. Excess emission limitations are necessary to minimize
emissions during malfunctions by not allowing a plant to continue operation
indefinitely during such periods. Consequently, requiring adequate
maintenance in addition to excess emission limits reduces the likelihood of
occurrences of excess emissions.
2.5.2.8 Design, Equipment, Work Practice, or Operational Standards.
Comment: One commenter (Part II Docket Item IV-D-22) thought the proposed
standards impermissibly incorporated design, equipment, work practice, or
operational requirements, citing Sections 61.92(c), 61.93(a), and 61.91(b)
as examples. He considered these sections vague and giving nearly
unlimited discretionary powers to the Administrator. The commenter also
referred to Section 112(e)(1) as allowing design, equipment, work practice,
or operational standards only if the Administrator determines that a
numerical emission standard is not feasible.
Response: The provisions referred to by the commenter are not "design,
equipment, work practice, or operational standards" under Section 112(e).
They are requirements relating to the operation and maintenance of maleic
anhydride plants intended to ensure that emissions are, to the extent
practicable, continuously controlled. The Clean Air Act specifically
authorizes such requirements as supplements to numerical emission limits.
Section 302(k) of the Act defines an emission standard as a requirement
"which limits the quantity, rate, or concentration of emissions of air
pollutants on a continuous basis, including any requiranent relating to the
operation or maintenance of a source to assure continuous emission
2-43
-------
reduction." (Emphasis added.) Necessarily, requirements for operation and
maintenance cannot be as precise as are numerical emission limits. As
discussed elsewhere, the promulgated operation and maintenance requirements
are as precise as possible, given the impossibility of enumerating every
possible factual circumstance that might arise.
2.5.3 Selection of Promulgated Standard for Existing Sources
After identifying the regulatory alternatives to apply to process
vents and excess emissions, the Administrator determined this combination
to be BDT for controlling benzene emissions from existing sources. The
estimated health impacts remaining after application of BDT were examined
in view of health benefits and costs that would result with application of
a more stringent option. Due to the assumptions used in calculating the
health numbers, there is considerable uncertainty associated with them
beyond that represented by the ranges presented here. They may represent
overestimates or underestimates. The uncertainties associated with these
numbers are explained in Section 2.1.1. Remaining leukemia cases are
estimated to range from 0.011 to 0.079 per year and the remaining maximum
lifetime risk of acquiring leukemia is estimated to range from 6.9 x 10
to 4.7 x 10~5 for the most exposed persons living in the vicinity of a
maleic anhydride plant.
The Administrator identified the next most stringent combination
beyond BDT as 97 percent control for process vents and the same excess
emissions control as selected for BDT. Health impacts of the 97 percent
control level were then examined to determine whether a more stringent
control level should be required.
In going from BDT to beyond BDT, estimated leukemia cases within 20
kilometers of maleic anhydride plants would be reduced from a range of
0.011 to 0.079 per year to a range of 0.0076 to 0.052 per year. Estimated
maximum lifetime risk at the point of maximum exposure caused by process
vent emissions would be reduced from a range of 6.9 x 10~6 to 4.7 x 10"^ to
a range of 2.3 x 10~6 to 1.6 x 10"5.
Requiring beyond BDT rather than BDT would increase capital cost from
$1.8 million to $8.6 million, total annualized cost from $0.75 million to
$3.5 million, percentage increase in maleic anhydride prices from 1 to as
much as 3 percent, and percentage increase in fumaric acid prices from 4
to 5 percent. Beyond BDT also could cause one plant closure.
2-44
-------
Because of the relatively small health benefits to be gained with the
additional costs (including a possible additional plant closure) of
requiring beyond BDT instead of BDT, the Administrator considers the risks
remaining after application of BDT to existing sources reasonable.
Consequently, no more stringent control than BDT for process vents is
required and BDT is judged to provide an ample margin of safety to protect
public health.
2-44-A
-------
To summarize, the final standard requires benzene emissions to be
reduced from existing maleic anhydrice production units to 1.0 kg of
benzene per 100 kg benzene feed to the reactor. Excess emissions during
total plant startup are restricted to 8 hours, and individual reactor
startup to 1 1/2 hours. Excess emissions during control device malfunction
are restricted to no more than 6 hours.
In addition, to ease the reporting burden, the requirement of
reporting excess emissions within 10 days after each occurrence has been
revised to submitting reports on a quarterly basis. The quarterly reports
would contain the same information that would have been required in the
10-day reports, except for the amount of emissions, and would essentially
be a listing of the occurrences and duration of excess emissions for each
3-month period.
2.6 SELECTION OF BASIS OF THE STANDARD FOR NEW SOURCES
2.6.1 Zero Emissions Standards
Comment: According to one commenter (Part II Docket Item IV-D-5), under
Section 112, EPA lacks the authority to differentiate between old and new
sources or to require zero emissions from a source.
Response: EPA has authority to differentiate between existing and new
sources. As noted above, in order to best meet its responsibilities under
Section 112, EPA regulates hazardous pollutants by source categories and
considers technical and economic factors in establishing standards for
pollutants for which no threshold has been identified. A distinction
between new and existing sources is a reasonable classification in this
regard, because of the important technical and economic differences between
installing control equipment as part of construction, as opposed to
retrofitting that technology. Congress has, in fact, recognized the
importance of this distinction; under Section 112(c), new sources must
install the technology required by the standard immediately, while existing
sources have a certain amount of time before the standard becomes
applicable.
This very distinction lies at the heart of the standard for new and
existing maleic anhydride plants. As noted elsewhere, BDT for new
2-45
-------
production units is the use of n-butane processes, since these can be
installed at reasonable cost during the design and construction of a plant.
But such costs may not be reasonable where such processes are used on a
retrofit basis. Accordingly, 8DT for existing production units is add-on
controls.
EPA also has authority to require zero emissions from a source
category. The only applicable statutory requirement is that a standard be
set "at the level which in [the Administrator's] judgment provides an ample
margin of safety to protect the public health from such hazardous
pollutant." Section 112(b)(1)(B). For nonthreshold pollutants, EPA has
interpreted this provision to mean that, at a minimum, a source category
must apply BDT. For new maleic anhydride plants, EPA has concluded that
BDT is the use of n-butane processes that will result in zero emissions of
benzene. Moreover, although Congress deleted a zero emissions presumption
in adopting Section 112, it by no means intended that zero emissions could
never be required. As Senator Muskie stated in presenting Section 112 to
the Senate, Section 112 "could include emission standards which allowed for
no measurable emissions." 1 Legislative of the Clean Air Act Amendments of
1970 at 133; see also jM. at 116 (remarks of Rep. Hechler). If EPA
concluded that emissions greater than zero resulted in unreasonable
residual risks, it would need to prescribe a zero emissions standard to
assure an ample margin of safety as described above. This would, of
course, be rare, and would not reflect a policy of eliminating all risks.
2.6.2 n-Butane Technology
Comment: One commenter stated that EPA had not determined whether
available n-butane processes are commercially viable or even workable (Part
II Docket Item IV-D-14).
Response: Developments within the last several years show that n-butane
technology for producing maleic anhydride is commercially viable. For
example, Amoco has been operating an n-butane-based plant since 1976.
Monsanto converted 20 percent of its existing plant to n-butane in the
midseventies, is building a new n-butane-based plant, and is planning to
convert all existing benzene-based capacity to n-butane by 1985 (Part II
Docket Item II-I-42). DENKA currently is using a conventional n-butane
process and, in conjunction with the Badger Corporation, has announced
plans to build a pilot plant using a fluidized-bed catalyst for oxidizing
2-46
-------
n-butane to maleic anhydride. Ashland is in the process of converting to
n-butane, while USS Chemicals, Pfizer, and Tenneco are considering the
n-butane option. Finally, Scientific Design Company, Inc. (SD), and Halcon
Industries market process designs and catalysts, respectively, for
converting n-butane to maleic anhydride. These developments clearly
indicate that the n-butane process is both commercially viable and
workable.
2.6.3 Banning of Benzene
Comment: Three commenters stated that the proposed ban, which causes
Government interference with future resource allocation, is unwise.
Chemical producers should be free to choose the most economically favored
feedstock, which under certain circumstances could again be benzene. The
commenters believe banning benzene from new plants would restrict
competition (Part II Docket Items IV-D-14; IV-D-21). Another commenter
(Part II Docket Item IV-D-25) thought such a requirement would establish
a precedent for similar action in the future.
Response: The Agency does not believe requiring 100 percent control
of benzene emissions from new plants would restrict competition. Indeed,
as discussed in the previous response, much of the industry is switching or
has switched to n-butane in order to gain a competitive edge.
The capital cost of an n-butane-based plant is greater than that of a
benzene-based plant of equal capacity because the n-butane catalyst
operates at peak efficiency at a lower productivity than does the benzene
catalyst. For any given capacity, an n-butane plant will be larger than a
comparable benzene-based plant. Conversely, for a given size plant, a
butane-based plant's capacity will be less than that of a benzene-based
plant. The capacity ratio, n-butane to benzene, is about 0.7 for most
process equipment (Part II Docket Item IV-J-30). Although, for any given
size plant, an n-butane-based process will make less maleic anhydride than
will a benzene-based process, the significant cost difference between
n-butane and benzene favors the n-butane-based process.
From January 19, 1981 to May 17, 1982, according to the Chemical
Marketing Reporter, benzene cost varied from $1.50 to $1.80 per gallon, or
about 2\i to 25£ per pound, while recent n-butane prices have been about
2-47
-------
10£ per pound. At a price difference of 3£ per pound between benzene and
n-butane, the price of maleic anhydride produced would be about the same
(Part II Docket Item IV-J-3). The lltf to 15£ difference between benzene
and n-butane prices clearly favors a new plant based on n-butane feed.
Capital and feedstock costs for the two processes are compared in Table
2-6.
Currently, benzene supplies are expected to remain tight for at least
the next several years (Part II Docket Item IV-J-6). Its current price--
$1.55 per gallon (May, 1982)--reflects this tight supply, which is caused
by high demand for octane boosters in gasoline, due in part to increased
unleaded gasoline use. More octane boosters are required to achieve a
particular octane number for unleaded than for leaded gasoline. One of the
major octane boosters used today is toluene, which is also a major source
of benzene via hydrodealkylation (HDA). As toluene's value as an octane
booster climbs, benzene's supply decreases and its cost increases.
A factor influencing n-butane supply is demand for light hydrocarbons,
including n-butane, which have three sources:, natural gas-processing
plants, refineries, and imports. A recent article on light hydrocarbons
projects that n-butane supplies and availability will be adequate through
1990 (Part II Docket Item IV-J-7).
The commenter did not supply any information suggesting that the
standard would have significant anticompetitive effects. Moreover,
Congress contemplated that compliance with Section 112 standards might
require the use of technologies under patent. Section 308 of the Act
provides for mandatory licensing of such technologies to prevent a
substantial lessening of competition.
These requirements do not establish a precedent for subsequent
NESHAP's and, if any similar situations arise in the future, they will be
judged on their own merits.
2.6.4 Economic Considerations
Comment: One commenter was concerned about whether or not provisions
existed that would permit revision of the standard in the event of changed
economic conditions such as benzene becoming less costly than butane (Part
II Docket Item IV-D-23, Attachment, p. 40).
2-48
-------
Response: If major economic changes limited n-butane availability for
maleic anhydride, several options would be available. First, EPA plans to
review this standard every 5 years, allowing analysis of and reaction to
significant changes in the standard's economic impact. Second, if
interested persons believe changes in n-butane availability justify
standard revision, they may petition the Agency for revision (01jato
Chapter of the Navajo Tribe, v. EPA 515 F.2d 654, 666 [D.C. Ci r. 1975]).
2-48-A
-------
TABLE 2-6. COMPARISON OF TRANSFER PRICES FOR N-BUTANE AND BENZENE PROCESSES
Basis: butane at 10?!/lb
benzene at 21^/1b
Production: 60 million Ib/yr of maleic anhydride
N-butane Benzene
Battery-limits capital cost (1979 costs) $15,700,000 $13,000,000
Maleic anhydride transfer price 0/lb g/lb
Raw materials
1. Feedstock 11.65 23.87
2. Catalyst and chemicals 2.00 1.12
13.65 24.99
Utilities (benzene oxidation is an exothermic 0.20 (0.33)
reaction producing net energy)
Labor related 0.59 0.59
Capital related (19% of capital costs) 5.12 4.22
19.56 29.47
20% return before taxes 5.24 4.32
Total transfer price 24.80 33.79
Source: Part II Docket Item IV-0-1.
2-49
-------
2.7 MODIFICATION AND RECONSTRUCTION
Comment: One cornmenter (Part II Docket Item IV-D-6) stated that new
facilities will not use benzene as a feed stock because n-butane is
cheaper, and that current users of benzene should be encouraged to convert
to n-butane. However, several commenters stated that the proposed standard
could delay conversion of existing plants to n-butane (Part II Docket Items
IV-F-7, IV-F-8, IV-D-25, IV-F-1, p. 101, 102 ). If a plant has converted to
n-butane technology, it cannot be easily reconverted to benzene feedstock.
Conversion of an existing benzene-based maleic anhydride plant to n-butane
technology or subsequent reconversion to benzene should not constitute
modification or reconstruction. In addition, the cominenters believed
process improvements that do not meet the definition of reconstruction and
that increase benzene emissions should not be considered modifications, if
the numerical emission limit is not exceeded.
According to the commenters, the modification and reconstruction
portion of the standard should be altered to allow:
Existing benzene-based plants that convert to n-butane to
reconvert to a benzene feedstock, and
Existing benzene-based plants to implement small process
improvements and alterations that increase emissions if the
numerical benzene emission limit is not exceeded.
Response: If an existing benzene-based plant converted to n-butane, the
change would not be considered a modification under this standard because
the benzene emission rate would not increase. (However, should the
n-butane process increase VOC emissions, the source could be considered
modified under a forthcoming New Source Performance Standard [NSPS] for air
oxidation processes.) This plant, though now converted to n-butane, would
still be considered an existing source for the purpose of this standard
because no modification, as defined in 40 CFR 61.02, occurred when the
plant was switched to an n-butane feedstock.
If, after having switched to n-butane, a plant returned to a benzene
feedstock, that change would have been considered a modification under the
proposed standard because the benzene emission rate would increase.
2-50
-------
Consequently, a plant that converted to n-butane would not have the option
of using benzene, while a plant that had not converted to n-butane would.
Industry has argued that a firm would be penalized for reducing benzene
emissions by converting a benzene-based plant to n-butane.
A difficulty associated with benzene to n-butane conversion is fine
tuning of downstream separation processes, necessary because the benzene
and n-butane processes, although similar, are not identical. A different
by-product mix would certainly result. While existing separation equipment
could probably do the job, a slightly different set of operating conditions
would probably be required to achieve optimum performance.
Because conversion of an existing benzene-based plant to n-butane may
entail technical experimentation and adjustments, as discussed above, the
Administrator has revised the standard to allow a plant to convert
experimentally to an alternative feedstock and to use that feedstock for up
to one catalyst charge after initial conversion. The duration of one
catalyst charge means the time from first introduction of feed to the
reactor until the catalyst in the reactor is either regenerated or
replaced. If the firm decided to reconvert to a benzene feedstock during
that period, it is allowed to do so under the promulgated rule without
violating the standard. One life cycle would be allowed for the n-butane
catalyst so the plant could determine whether the switch to n-butane was
technically viable and could become permanent. This situation applies only
to existing benzene-based sources. Considering economic incentives, this
allowance should sufficiently encourage existing plants to convert from
benzene to n-butane.
It should be noted that industry trends since proposal indicate the
proposed standards have had little or no deterring effect on conversion to
butane from benzene. In fact, two plants have completely converted to
butane and other plants are still considering such a switch.
The second part of the comment deals with minor plant improvements,
most of which would not constitute modification. The standard is not
intended to cause an existing source that increases production rate and
consequently emission rate to become a new source through minor process
changes. Therefore, the standard contains a provision to allow an existing
source to increase production rate without that change being considered a
modification, if that increase can be accomplished without capital
2-51
-------
expenditure. As defined in the final rule, "capital expenditure" is:
An expenditure for a physical or operational change to a
stationary source which exceeds the product of the applicable
"annual asset guideline repair allowance percentage" specified in
the latest edition of Internal Revenue Service (IRS) Publication
534 and the stationary source's basis, as defined by Section 1012
of the Internal Revenue Code.
However, the total expenditure for a physical or operational change to
a stationary source must not be reduced by any "excluded additions"
as defined in IRS Publication 534, as would be done for tax purposes.
IRS Asset Guideline Class 28.0, "Manufacture of Chemicals of Allied
Products," which includes maleic anhydride plants, has an "Annual Asset
Guideline Repair Allowance Percentage" of 12.5 percent. (Internal Revenue
Service Publication 534, "Depreciation," December 1981, U.S. Government
Printing Office). Thus, a firm could spend up to 12.5 percent of its
basis, or cost of plant and associated improvements, for changes to
increase production without the changes being considered a modification.
Expendable funds would be a function of the plant's cost basis, which
varies from plant to plant. An older plant may have a basis of only $8
million, while a newer plant may have a basis of $20 million. Generally,
older plants would not be used as the basis for capital investment unless
the plant's operating life could be extended sufficiently to allow
amortization of that investment.
Increased production output due to process changes, such as using a
new catalyst, can be accomplished at costs that would be less than a
"capital expenditure" and would therefore not qualify as a modification.
Thus, a plant could make such changes and continue to use a benzene
feedstock. Obviously, major changes in process equipment that increase
benzene emissions, such as a new reactor or increased capacity downstream
of the reactor, would probably require a "capital expenditure" and be
considered a "modification."
The standards' reconstruction requirements have been deleted.
2-52
-------
2.8 TEST METHODS AND MONITORING
2.8.1 Gas Chromatography
Comment: According to several commenters, gas chromatography is well
established for laboratory use but has riot been proven reliable in field
applications for long-term continuous compliance monitoring or process
control. If gas chromatography is to be a workable monitoring system for
industry, it must be designed on a plant-specific basis, requiring
substantial time, expense, and effort. Moreover, the commenters doubted
that reliable on-stream time greater than 80 percent can be attained with
current technology (Part II Docket Item IV-F-8).
Response: This comment by CMA is based on experiences of Pfizer,
Reichhold, and Monsanto, all of whom have gas chromatography/f1ame
ionization detector (GC/FID) monitors on similar process streams.
Specifically, Pfizer says it has found from experience that GC data are
unreliable because of sample contamination (Part II Docket Items IV-F-8 and
IV-D-22). Reichhold says it has six GC units that have never operated for
more than 4 consecutive days and that have been difficult to calibrate and
standardize because of moisture and problems with the sample line. These
six Reichhold GC units were established to operate continuously on air
oxidation units at plants other than Reichhold's maleic anhydride plant,
where process measurements were made through laboratory analysis (Part II
Docket Item IV-B-1). Monsanto currently uses a GC unit to monitor the
off-gas from its reactors for process control. This unit produces reliable
readings only 80 to 90 percent of the time, primarily because of sampling
problems; i.e., plugging and moisture condensation (Part II Docket Item
IV-B-2).
A GC/FID system could experience several types of problems, including
plugged sample lines, moisture condensation in sample lines, span drift,
and standardization and calibration difficulties. Although none would be
serious or time-consuming to solve, these problems could occur frequently
enough to cause losses in on-line time ranging from 10 to 20 percent (Part
II Docket Items IV-F-8, IV-D-22, IV-B-1, and IV-B-2). EPA recognizes that
these types of problems can occur with the GC/FID, but with proper care and
maintenance, 80 percent reliability should be considered a worst case. EPA
does not consider such a reliability for a GC/FID unacceptable. Some
2-53
-------
commenters had the incorrect impression that if their monitor
malfunctioned, they would have to shut down the plant. This is not the
case, and the regulation has been revised to clarify this point.
2.8.2 Process Parameter Monitoring
Comment: Several commenters said operating parameter monitoring and
emission testing would be sufficient to determine whether or not the
standard were exceeded. Benzene fed to the reactor is now measured
continuously, and the gas stream leaving the reactor is sampled and
analyzed routinely and periodically for benzene. Additional monitoring of
temperature, oxygen, and flow rates in the incinerator would be sufficient
to determine compliance with the standard if an emission test showed that
the standard were met under these conditions (Part II Docket Item IV-F-8).
Response: Operating parameter monitoring data would have to be correlated
with emission test data, including specific information on relationships
among incinerator destruction efficiency, temperature, and flow rate, to
provide meaningful information on benzene destruction efficiency.
Even if emission test and operating parameter data could be correlated
effectively, determining incinerator destruction efficiency would not be
sufficient to determine compliance because the standard is based on overall
benzene removal in both the reactor and the incinerator. Thus, information
on the reactor's benzene conversion rate would also be required. The
benzene conversion rate is expected to decline slowly but not to fluctuate
over the catalyst's life. Therefore, daily sampling and testing of reactor
off-gas would be sufficient to determine the benzene conversion rate.
However, the plant may wish to consider its conversion rate confidential.
Although operating parameter monitoring does not directly determine
benzene emissions, it is more reliable than GC/FID. Specifically,
operating parameter monitoring would provide reliable results essentially
100 percent of the time, as compared to only 80 percent of the time for
GC/FID, according to industry (Part II Docket Item IV-F-8). However, as
stated in the previous response, EPA does not consider 80 percent
reliability unacceptable, but feels if the monitoring requirements are
followed, 80 percent will be a worst case. In addition, as the commenter
stated, the owner or operator has an incentive to maintain the temperature
and flow monitors to ensure safe operations but not to maintain a GC/FID.
However, GC/FID remains the method specified by the standards, for several
2-54
-------
reasons. First, GC/FID would measure the effluent's benzene concentration,
and subsequent calculations would show a comparison between the emissions
and emission limit. Second, GC/FID can be used with all control options,
while operating parameter monitoring would have a different set of
parameters for each control device type and could not be used with carbon
adsorption. Third, operating parameter limits would vary from plant to
plant and could not be known before a control system was built. Therefore,
it would not be suitable for the Agency to try to set specific limits.
However, after receipt and consideration of written application, EPA is
willing to approve operating parameter monitoring as an acceptable
alternative to GC/FID on a case-by-case basis.
2.8.3 Total Hydrocarbon Monitoring
Comment: Several commenters said that the standard should permit
continuous monitoring by total hydrocarbon analysis employing an FID. This
method has proven effective over several years at USS Chemicals' maleic
anhydride plant, which controls benzene emissions by use of a catalytic
oxidizer (Part II Docket Items IV-F-7; IV-F-8; IV-F-1).
Response: For monitoring, the Administrator has considered using an FID
without GC as a total VOC monitor and assuming all VOCs are benzene.
However, because all VOCs from maleic anhydride plants are not benzene,
this alternative would indicate higher levels of benzene than are actually
present. With waste gas concentrations similar to those shown in Table 1-4
of the BID for proposal (EPA-450/3-80-001a) an FID might report a VOC
concentration 30 percent higher than actual benzene concentration.
VOC concentration detected by the FID would be assumed to be 100
percent benzene; when multiplied by the flow rate of the incinerator
off-gas, it would give the mass of "benzene" in the off-gas. Compliance
would be determined just as it is for a GC/FID: 1.0 kilograms "benzene"
emitted per 100 kilograms benzene fed to the reactor.
An assumed constant ratio of benzene to VOC in the off-gas would not be
suitable because this ratio can fluctuate 30 percent or more. Total VOC
monitoring with an FID may be advantageous in certain instances,
particularly in plants that use carbon adsorption (where high humidity may
cause problems for a GC/FID—although the problems should be correctable),
and would be cheaper than a GC/FID. However, many GC/FID sampling
problems, such as with standardization and calibration (see Subsection
2-55
-------
2.10.1), also would affect FID. Therefore, FID does not seem to be a
better approach than does GC/FID, but if a plant were willing to accept its
limitations, the Administrator would approve it as an acceptable
alternative, upon written application.
2.8.4 Monitor Malfunction
Comment: One commenter stated that production should be halted whenever
the continuous monitoring system is not fully operational (Part II Docket
Item IV-D-8).
Response: The Agency recognizes that monitoring systems do not operate 100
percent of the time. If operated and maintained according to monitoring
provisions, systems should be inoperable a minimal amount of time. If a
plant were required to shut down whenever a problem occurred in the
continuous monitoring system, total plant benzene emissions would increase
due to the excess emissions that would occur during shutdown and subsequent
startup. Such a requirement would be impractical and would increase total
benzene emissions (see Subsection 2.5.2.4).
The final rule contains a provision that monitoring data recorded
during monitoring system breakdowns shall not be included in average
emission calculations. Additionally, the rule specifies that such
breakdowns shall be corrected as soon as practicable. Each source's owner
or operator shall also maintain a record of frequency and duration of such
breakdowns to be included for the previous 3 months in each quarterly
report.
2.8.5 Permeation Tubes
Comment: In Method 110, EPA explicitly should allow use of permeation
tubes to generate benzene standards for the chromatograph, according to one
commenter (Part II Docket Items IV-D-8; IV-F-1, p. 211, 212).
Response: Benzene standards can be generated with permeation tubes, but
EPA considers it impractical in this instance due to the low permeation
rate for benzene. However, as provided for in Section 61.94 (f), upon
approval by the Administrator, EPA will accept a technically sound
deviation in a test method provided the technique is adequately documented
and the method's accuracy and precision are not impaired. Therefore, the
fact that EPA does not cite permeation tubes in Test Method 110 does not
mean they have been excluded as a possible means of generating standards.
2-55
-------
2.8.6 Process Vent Benzene Concentrations
Comment: The Federal Register does not discuss expected benzene
concentration in the process vent emissions although the test method
sensitivity is stated. Therefore, according to the commenter, it is not
clear that the proposed monitoring requirement will ensure that excess
emissions are detected or that the standard is met (Part II Docket Item
IV-D-8).
Response: The standard is expressed in terms of mass of benzene emissions
per amount of benzene feedstock. Therefore, the outlet mass emissions
depends on the benzene concentration and flow rate of the outlet stream,
which are subject to considerable variability depending upon the type of
control device used. Consequently, the required sensitivity of the
monitoring system also depends on the type of control device used and the
expected outlet benzene concentrations. Test data on a thermal incinerator
(Part II Docket Item II-A-4) showed outlet benzene concentrations varied
between 11.1 and 14.4 ppm. Test data on a carbon adsorber (Part II Docket
Item II-A-11) showed outlet benzene concentrations varied between 42.2 ppm
and 81.4 ppm. For each GC/FID, as required by §61.95 (g), a daily span
check is to be conducted with reference gas containing a concentration of
benzene determined to be equivalent to the emission limit for that source
based on emission tests required by §61.93. In addition, calibration of
the GC/FID monitor is to be done in accordance with Section 7 of Test
Method 110. The specified range of Test Method 110 is 0.1 to 70 ppm, the
upper limit of which can be extended by diluting the sample or extending
the calibration range. Because the concentrations of benzene expected to
be encountered in the outlet gases of maleic anhydride process vents are
greater than the lower end of the range specified in Test Method 110, the
sensitivity of the GC/FID monitor would be more than adequate to detect
excess emissions.
2.8.7 Calibration
Comment: The National Bureau of Standards (NBS) of the Department of
Commerce comments that it normally will not perform tests of the type
specified by EPA. Rather, NBS prepares and sells Standard Reference
Materials (SRM's) to enable results of such tests to be standardized.
Accordingly, the phrase "by direct analysis by the National Bureau of
Standards" appearing in the third paragraph under the preamble discussion
2-57
-------
entitled "Selection of Emission Monitoring Requirements" (45 FR 26673)
should be changed to read, "by calibration against benzene cylinder
Standard Reference Materials (SRM's) prepared by the National Bureau of
Standards, if such SRM's are available." Similarly, the phrase "having it
analyzed by the National Bureau of Standards" appearing in Paragraph
5.2.3.2 of proposed Test Method 110 (45 FR 26679) should be changed to read
"calibrating it against benzene cylinder Standard Reference Materials
(SRM's) prepared by the National Bureau of Standards, if such SRMs are
available." (Part I Docket Item IV-D-14).
Response: In instances cited by the commenter (Part I Docket Item
IV-D-14), NBS, EPA implied that NBS would analyze cylinders other than its
own. However, NBS has emphasized that it would not routinely analyze
cylinders other than its own. It should be noted that the two cited
requirements for NBS cylinder analysis were optional specifications of
Method 110. The change NBS suggested (allowing calibration against benzene
cylinder SRMs prepared by NBS if such SRMs are available), provides a
possible option satisfactory to EPA and is included in the final
regulation.
2.8.8 Chromatograph Response
Comment: One commenter stated that the statement in Paragraph 4.3.1 of
Test Method 110, "chromatographic system shall be capable of producing a
response to 0.1 ppm benzene that is at least as great as the average noise
level", appears to recommend a minimum response in terms of a signal-to-
noise ratio (S/N) of two to one. A higher response (an S/N of up to 10 to
1) is feasible for 0.1 ppm benzene. Furthermore, data reported at an S/N
of two to one can lead to interpretation error. Therefore, the phrase
"that is at least as great as the average noise level" in the second
sentence of Paragraph 4.3.1 should be changed to "that is at least five
times the average noise level (Part I Docket Item IV-D-14).
Response: EPA considered it sufficient to specify that the chromatograph
system must be capable of producing a response to 0.1 ppm benzene at least
as great as the average noise level in view of expected benzene stack gas
concentrations of several parts per million. However, as the commenter
noted, modern chromatographic systems are capable of far more sensitivity,
and it would be reasonable to specify that the system must be capable of
producing a response to 0.1 ppm benzene at least five times the average
2-58
-------
noise level. Also, because EPA specified that the range of Method 110 must
be 0.1 to 70 ppm, the commenter1s suggested specification will prevent
interpretation errors at benzene concentrations near the range's lower
limit. Therefore, the suggested change has been made.
2.9 REPORTING AND RECORDKEEPING
Comment: Several cornmenters stated that the proposed standard requires a
company to submit benzene conversion rates, benzene feed rate to the
reactor, and production rates at the beginning of and during the control
system malfunction if excess emissions are claimed to have resulted from a
control system malfunction. They said such information is proprietary and
should be made available only confidentially for EPA inspection on the
plant operator's premises (Part II Docket Items IV-F-7; IV-F-8; IV-D-25;
IV-F-1, p. 105).
Response: The proposed standard's excess emissions requirements have been
changed in the final rule and are based on duration and not quantities of
emissions. Therefore, information in the above comment is no longer
required. Instead, reported time and duration of each occurrence of excess
emissions is required. Also, the promulgated standard requires excess
emissions reports only quarterly, instead of within 10 days of occurrence.
2.10 LEGAL
2.10.1 Airborne Carcinogen Policy as Basis for Rulemaking
Comment: The proposed standard is based on the proposed Airborne
Carcinogen Policy. It is improper for EPA to implement the policy before
it has been promulgated (Part II Docket Item IV-D-5).
Response: While the proposed standards were developed "consistent with"
the proposed policy, standards development was not based on the proposed
policy, and the standards development methodology was presented to be
judged on its own. The standards are based on the authority of Section
112. Support for the standards is contained in the rulemaking docket
(0AQPS 79-3) for these standards and summarized in today's notice and the
notice for the proposed rulemaking.
EPA noted in the proposed policy that the procedures outlined would
"generally" be followed for actions taken in the interim between proposal
and promulgation. As a proposal, the policy does not bind the Agency to
the procedures described, nor did EPA intend that changes were responsive
2-59
-------
to public comment on the proposed policy or relevant rulemakings. Further,
EPA recognizes that defense of a regulatory decision cannot be based on a
proposed policy. While the Agency has been guided by the proposed policy,
therefore, the final rule and the basis for the rule are presented as a
separate, independent rulemaking.
In any event, the basis for the rulemaking is Section 112 of the Clean
Air Act, under which the Administrator is required to publish proposed
regulations establishing emission standards for hazardous air pollutants
listed under Section 112. Although no approach for developing such
standards is specifically addressed in Section 112, the Administrator
believes the approach used for developing benzene standards for maleic
anhydride plants meets the requirements of Section 112, is sensible, and
results in significant benzene reduction for reasonable cost.
2.10.2 BDT as Control Strategy
Comment: One commenter (Part II Docket Item IV-D-22) thought that some
control level lower than BAT (BDT) may be appropriate when risk is low from
the source to be regulated. Similarly, another commenter (Part II Docket
Item IV-D-5) thought there was no justification for BAT (BDT) as the only
admissible control strategy and no statutory support for such a position.
Response: These comments refer to the Administrator's selection at
proposal of 97 percent control (then defined as BAT), as a minimum, without
consideration of less stringent control levels. Section 112 requires that
a standard protect the public health with an "ample margin of safety."
Since the Agency believes there is no "safe" (risk-free) level for exposure
to a carcinogen, a strict reading of the statute implies that the Agency
could require only 100 percent control of carcinogens. As previously
discussed, the Agency believes that such an interpretation is unreasonable
and not the intent of Congress.
In setting standards, therefore, EPA must determine what level of
residual risk the standard will permit. Given that this decision cannot be
made on the sole basis of the absence of possible health risks, the
technological and economic implications of various levels of control, as
well as the associated risks, are logical factors to consider, as long as
the emphasis remains on risks.
2-60
-------
Such an approach is roughly equivalent to that required for new
sources under Section 111, which requires best demonstrated technology
(BDT). Standards under Section 111 (d) are promulgated for source
categories whose emissions "cause, or contribute significantly to, air
pollution that may reasonably be anticipated to endanger public health and
welfare." Section 111(b)(1)(A). Similarily, BDT is required for existing
sources listed under Section 111(d), where the source category is listed
under the test articulated in Section 111(b)(1)(A). The criteria for
listing under Section 111 address pollutants that are less hazardous than
those listed under Section 112. Accordingly, since BDT is the required
standard for sources emitting pollutants that may "endanger public health,"
it must also be at least a minimum standard for sources emitting hazardous
(here, carcinogenic) pollutants "with an ample margin of safety." Although
EPA has viewed the concept of BAT under Section 112 as essentially
equivalent to the concept of BDT as defined in Section 111, commenters have
inferred from the difference in nomenclature that BAT represents a more
stringent approach. To clarify EPA's position, the term BAT has been
dropped and BDT substituted in the final rulemaking of these documents.
The term BDT as used in this document is equivalent to the term BAT used
in the proposed documents.
In addition, use of BDT as a minimum makes good sense with regard to
regulation of hazardous substances. BDT represents a level of technology
that reduces emissions as much as possible, given cost, energy,
environmental, and technological factors. As such, it is a reasonable
control that at a minimum best protects public health without incurring
unreasonable costs. In view of the carcinogenic nature of the pollutants
regulated under Section 112, application of BDT to source categories
emitting significant quantities of those pollutants therefore imposes the
least stringent possible level of control that ordinarily can be said to
protect public health with an ample margin of safety (assuming, of course,
that there is no remaining "unreasonable residual risk" after application
of BDT).
EPA notes that it will not regulate a source category under Section
112, where the risk from that source category is not significant.
2-61
-------
Finally, EPA has revised its designation of BDT in this standard to 90
percent control, based on emissions and cost impacts (see Subsection
2.5.1.3).
2.10.3 Alternatives to Uniform Standard
Comment: Several commenters suggested alternatives to a uniform nationwide
standard. One commenter (Part II Docket Item IV-D-21) thought such an
alternative should be considered, since he estimated that control of one of
the plants would derive most of the health benefits. Another commenter
(Part II Docket Item IV-D-23) thought an appropriate alternative would be
to rely on regulatory efforts such as existing State regulatory programs.
Another commenter (Part II Docket Item IV-D-15) thought emission sources
located beyond a certain distance from populated areas should be exempt
from the standards. He suggested a distance of 0.5 km as adequate
to safeguard the population. Still another commenter (Part II Docket Item
IV-D-23) thought that EPA should consider defining plants in different
population density categories to be different source categories.
Response: As previously noted, EPA has examined a variety of existing
regulatory programs (including relying solely on State programs) and has
concluded that despite existing controls, benzene emission from maleic
anhydride plants pose a significant risk to public health. Having reached
this decision, EPA lacks authority to promulgate other than a national
standard for this source category. Section 112 clearly contemplates that
NESHAP standards be uniform regardless of facility location and proximity
to populated areas, for it speaks of promulgating a "standard" that is
meant to protect public health. Moreover, Section 112(e)(2)(B) speaks of a
"particular class of sources," which suggests that Congress intended
standards to be applicable to such a class as a whole. Finally, nothing in
Section 112 or its legislative history offers any indication that EPA could
promulgate anything other than a uniform national standard for a source
class, based on technical and economic differences, and not on location and
proximity to populated areas. However, it should be noted that Congress
has distinguished between new and existing sources in Section 112, and
because of this distinction, there is nothing to prohibit applying
different standards to new and existing sources based on the technical and
economic differences between such sources.
2-62
-------
EPA has carefully considered defining source categories according to
the population density around a source and has decided that this approach
is not feasible. First, this would mean that different sources of the same
type would have different standards applicable to them, which, as noted
above, is contrary to Section 112. Second, such an approach would be very
difficult to implement because it would entail examining sources of quite
different types (e.g., maleic anhydride plants and benzene storage vessels)
and determining a standard applicable to all of them. Finally, the comment
implies that the distinguishing factor should be "cost per leukemia case,"
which entails the use of quantitative risk assessments for individual
plants. But as discussed elsewhere in this notice, such assessments are
too unreliable to be used as a factor in determining 8DT. Moreover, this
approach would ignore the constant changes in population and the importance
of risks to the most exposed groups.
Finally, EPA sees no legal or factual basis for exempting certain
sources from a standard merely because they are a certain distance from
populated areas. EPA does not believe that an individual source exemption
based on the distance to the nearest private residence would be workable or
appropriate. Even where this distance is prescribed by company fencelines,
other factors that enter into the exposure calculation such as emission
rate and meterological conditions are subject to change. In addition, the
exposure estimate is a rough calculation subject to significant
uncertainties. Further, a simple limit on distance, while considering the
potential hazard to the most exposed individuals, ignores the situation in
2-63
-------
which a large population may be exposed to lower concentrations. Finally,
source categories are defined for regulatory purposes on the basis of
technological similarities. Section 112 does not provide for distinctions
on the basis of population parameters. In addition, the commenter has
provided no basis for the suggested 0.5-km cutoff.
2.10.4 New Listing of Benzene
Comment: One commenter (Part I Docket Item IV-D-10) felt that the maleic
anhydride proposal, in addition to being a proposed standard, constituted a
new "listing" of benzene as a hazardous air pollutant under Section 112,
and as such, lacked sufficient justification for the "listing."
Response: The standard for maleic anhydride plants is not a new listing of
benzene, for benzene has already been listed as a hazardous pollutant. In
any event, this comment is academic since EPA clearly has an adequate basis
both to list benzene as a hazardous pollutant and to set a standard for
maleic anhydride plants under Section 112.
2.11 MISCELLANEOUS
Comment: One commenter (Part II Docket Item IV-D-11) stated that the
specific proposal date should be included in Section 61.90, "Applicability
and Designation of Source."
Response: Although the Clean Air Act specifies that a source that
commences construction or modification after the proposal date is a new
source, the commenter's suggestion has been noted and the date is also
included in the promulgated standard.
2-64
-------
APPENDIX A: DEVELOPMENT OF REVISED COSTS
FOR CONTROL OF BENZENE EMISSIONS FROM MALEIC ANHYDRIDE PLANTS
-------
APPENDIX A
DEVELOPMENT OF REVISED COSTS FOR CONTROL OF BENZENE EMISSIONS
FROM MALEIC ANHYDRIDE PLANTS
A. 1 INTRODUCTION
This appendix describes procedures used to estimate costs of
installing retrofit control equipment in the maleic anhydride industry.
Costs are derived from the Enviroscience Reports, Emission Control Options
for the Synthetic Organic Chemicals Manufacturing Industry (Part II Docket
Item II-A-7), Control Device Evaluation: Thermal Incineration (Part II
Docket Item IV-A-3), and Control Device Evaluation: Carbon Adsorption
(Part II Docket Item IV-A-2). The first section describes how
industry-supplied ductwork costs were evaluated, the second describes how
retrofit costs were obtained for thermal incineration, and the final
describes how retrofit costs were obtained for carbon adsorption.
A.2 EVALUATING THE REASONABLENESS OF INDUSTRY DUCTWORK COST ESTIMATES
USS Chemicals has estimated it would need 600 linear feet of
42-inch-diameter stainless steel duct to connect its process vents to a new
control device (Trip Report to USS Chemicals, Neville Island, December 19,
1980). Stainless steel ducting is required due to the corrosive nature of
the waste gas from the product recovery absorber (PRA). The ductwork would
need to be steam traced and insulated to prevent the benzene from freezing
in the duct. (Benzene freezes at 42° F and would enter the ductwork at
only 100° F.) USS Chemicals estimated the ductwork's installed cost to be
$1.3 mil 1 ion.
Independent cost estimates of steam-traced and insulated ductwork
were not located. However, Capital and Operation Costs of Selected Air
Pollution Control System, referred to as the "GARD Report" (Part II Docket
A-2
-------
Item II-A-2), provides information on water-cooled ductwork fabrication
prices. Costs based on water-cooled ductwork provide a low cost estimate
for steam-traced and insulated ductwork because water cooling does
not require insulation. Another factor tempering this estimate is the lack
of allowance for elbows in the ductwork costs.
The GARD water-cooled costs had to be adjusted because they are based
on carbon steel ductwork. The report provides plain duct costs for both
carbon steel and stainless steel ductwork. EPA adjusted the GARD
water-cooled costs by assuming the cost difference between water-cooled
stainless steel duct and water-cooled carbon steel duct would be the same as
the difference between that of stainless steel duct and carbon steel duct.
The following costs are for ductwork 42 inches in diameter with
3/8-inch-thick walls. Figure 4-8 in the GARD report cites $414 per linear
foot for stainless steel ductwork, while Figure 4-7 cites $97 per linear
foot for carbon steel ductwork. Therefore, use of stainless steel results
in a $317 increase per linear feet over use of carbon steel. Figure 4-9 of
the GARD report cites $239 per linear foot for water-cooled carbon steel
ductwork. Adding to that cost the increase for stainless steel found above
gives a cost of $556 per linear foot for water-cooled stainless steel
ductwork. This price is based on December 1977 data. The fabricated
equipment index, one of the Chemical Engineering plant cost indices, was
used to update the price to December 1980. Multiplying $556 by the
appropriate ratio, 304/210, results in the December 1980 price of $805 per
linear foot. For 600 linear feet, the cost would be $483,000. A retrofit
installation component factor of 2.3 was found through methods described in
the Mascone memo (see Subsection A.3). A total installed cost of $1.1
million was found by multiplying the retrofit component factor by the
purchase cost. The following calculation was used:
[239 + (414 - 97)] x (304/210) x 600 x 2.3 = $1.1 million.
cost of stain-
less steel water-
cooled ductwork
in December 1977
jecember
iltfSl
ratio of
fabricated
equi pment
indices
(December
1980/Decem-
ber 1977)
(Chemi cal
Eny Inuring)
length
of duct-
work
li red
Chemical s)
retrofit
component
instal-
lation
factor
(Mascone
memo)
cost of
water-cooled
ductwork
installed in
a preexisting
Slant in
ecember
1980 ($)
A-3
-------
USS Chemicals' estimated $1.3 million compares favorably with this estimate,
and Reichhold Chemical's estimated $162,500 for delivered ductwork also
appears reasonable (Part II Docket Item IV-B-l). An estimate based on the
Enviroscience thermal incineration report with no allowance for special
types of duct or retrofit is $150,000 installed, as shown in Figure V-16 of
that report.
Because ductwork costs provided by these two plants appear specific and
reasonable, they were incorporated into EPA's revised costs.
A.3 ESTIMATION OF RETROFIT INSTALLATION COMPONENT FACTORS FOR THERMAL
INCINERATION
Retrofit installation factors used in this report are based on those
developed by David Mascone of the Chemicals and Petroleum Branch, Emissions
Standards and Engineering Division (ESED), U.S. Environmental Protection
Agency (EPA), as described in a memorandum from Jim Galloway to the
Synthetic Organic Chemicals Manufacturing Industry (SOCMI) Air Oxidation
file (Part II Docket Item IV-E-O).
Attachment A of this memorandum (referred to as the "Mascone memo") is
a table showing, as a percentage of the budget price, the cost of 10
individual installation cost components (foundation, erection, and piping).
Installation of three items is considered: combustion chamber, including
ductwork and stacks; recuperative heat exchanger; and waste heat boiler.
New source installation factors, retrofit labor factors, and total retrofit
factors are presented for each item. The basis for the estimates is
described in the memo and in the original Enviroscience thermal incineration
document. These factors are then combined with various contingency and
correction factors to produce overall installation factors, as shown at the
bottom of Attachment A of the Mascone memo. (Note that the number entered
for Contingencies, Fees, and Site Development should be 1.35 instead of
0.35. This is a correction of a typographical error.)
A consequence of using this method is that the graph for Total
Installed Costs for Thermal Incineration Systems, Figure V-2 of the
Enviroscience report, cannot be used because the overall installation factor
varies for the different items. Instead, individual Enviroscience graphs
A-4
-------
for combustion chamber, heat exchanger, and ductwork must be used: Figures
V-8, V-12, and V-16, respectively. The Enviroscience cost is multiplied by
the ratio of the overall retrofit factor to the overall new source factor
to obtain a retrofitted cost from the new source-installed costs cited in
the Enviroscience report. Finally, these costs were scaled to December
1980 from December 1979 by multiplying by 304/274. A sample calculation
for Ashland's heat exchanger is shown below:
720,000 >
cost of heat
exchanger
installed in
a new plant
in December
1979, from
Figu^ V-12
(Enviroscience
Report)
3.5
overal1
retrofit
instal-
lation
factor
(Mascone
Memo)
2.5 x (304/274)
overal1
new
source
instal-
1ation
factor
(Mascone
Memo)
ratio of
fabricated
equipment
indices
(December
1980/Decem-
ber 1979)
(Chemical.
1,118,000.
cost of heat
exchanger
installed in
a preexisting
Slant in
ecemjjj^r 1980
(Note that even though Ashland has converted 50 percent of its capacity to
n-butane, control device costs remain the same because the plant only has
one PRA through which all reactor waste gases must pass.)
For all plants except Reichhold and USS Chemicals, cost development
procedures were exactly as outlined above. The procedure for Reichhold and
USS Chemicals is slightly more complicated because, as discussed in
Subsection A.2, these plants provided specific and appropriate costs
for their ductwork. Only costs for the combustion chamber and heat
exchanger were obtained from the Enviroscience report for these two plants.
The heat exchanger could be scaled in the usual way because all duct costs
were included in Mascone's combustion chamber installation factor. For the
combustion chamber, a different new source and a different retrofit factor
were derived by eliminating piping from the list of 10 factors considered.
Overall factors were calculated by the original procedure, to provide
installation factors of 3.7 and 5.0, respectively. The ratio of the two
factors is then used as before to obtain the retrofitted cost. Because
USS Chemicals' ductwork cost was a retrofit-installed cost, it was merely
added to costs of the combustion chamber and the heat exchanger for a total
cost. Calculation for the new retrofit factor from the Mascone memo and
A-5
-------
for the retrofit-installed cost of the combustion chamber and ductwork is
shown below.
= 5.0.
overal1
retrofit
i nstal-
lation
factor,
less
piping
2,815,000.
cost of
combustion
chamber
and duct-
work
installed
in an
existing
plant in
December
1980 ($)
For the Reichhold plant, the new source-installed combustion chamber
cost ($655,000) was divided by the ductless overall new source installation
factor, less piping (3.7), to obtain a purchase price ($177,000) for the
combustion chamber. This price was added to the ductwork purchase price
($162,000) provided by Reichhold. The result was multiplied by the overall
retrofit installation factor, less piping (5.0). Table A-l illustrates the
method described here, and Tables A-2 and A-3 show total costs for thermal
incineration.
The water-cooled ductwork cost used to evaluate USS Chemicals'
ductwork in Subsection A.2 cost did not include installation, so a retrofit
installation figure for ductwork was needed. Because the ductwork cost
included an estimate of additional piping needed, the retrofit labor
estimate (0.30) was used with the new source estimate for everything else
(0.10). The result was an estimate that 40 percent of the total budget
price would be needed to install ductwork into an existing plant. No
attempt was made to change this estimate to reflect the fact that the
percentage would apply only to the ductwork and not to the combustion
chamber and ductwork. The error resulted in a conservative bias. The
ductwork component figure was multiplied by the correction factors just as
A-6
{1
budget
price
t- 0.2
unspeci-
fied
equipment
+ [(2.09 - 0.60) x 1.7]}
total retro-
fit instal-
lation
component,
less piping
total
i nstal-
lation
component
correction
factor
( 1.35
overal1
correc-
tion
factor
[1,010,000 x (304/274) x (5.0 * 3.7)] + [1,300,000] =
cost of
combustion
chamber
installed
in a new
plant in
December
979, from
i^ure V-8
Envi ro-
science)
ratio of
fabri cated
equipment
indices
(December
l9807Decem-
ber 1974]
[Chemical
Engineering)
overal1
retrofit
instal-
lation
factor,
less
iping
Mascone
lemo)
overal1
new
source
instal-
lation
factor,
less
Piping
(Mascone
Memo)
cost of
ductwork
installed
in USS
Chemicals
plant ($)
fuss
Chemicals)
-------
TABLE A-l. CALCULATION OF TOTAL RETROFITTED INSTALLED COST FOR THERMAL INCINERATION
Heat exchanger cost
Combustion chamber
and ductwork cost
Total retrofit
installed cost
New
Retrofit''
Update0
New source
Retrofit*3
Retrofit"1
Combustion
Retrofit"1
x 5.0
(E)
source
heat
exchanger3
x 3-5
TZ
x 304
27T
(A)
Combustion Combustion
chamber chamber3
and duct*
Update0
x 304
TH
x 6-5
-------
TABLE A-2. EXISTING PLANT COSTS FOR ACHIEVING 97 PERCENT BENZENE EMISSION REDUCTION
(Control Method: Thermal Incineration with Primary Heat Recovery )
Plant name
and location
Capacity
(Mg/yr)
Total
i nstal led
cost
($l,000s)
ACC
($1,000s/yr)
Direct operating
cost for 8,000
hr/yr ($1,000s/yr)
Energy recovery
credit for 8,000
hr/yr ($/yr)
Total annualized cost
for 8,000 hr/yr
($l,000s/yr) ($/Hg product)
Pfi zer
10,700
1,680
341
1,007
(629)
719
67.2
Ashland
27,200
2,790
566
2,134
(1,442)
1,258
46.2
Reichhold
20,000
2,590
525
1,437
(1,132)
830
41.5
Tennecob
9,980
30
6
2
—
8
0.8
U.S. Steel
36,400
4,240
860
4,160
(3,050)
1,970
54.1
Monsantoc
22,700
30
6
2
—
8
0.4
DENKAC
38,100
30
6
2
—
8
0.2
aA11 costs include continuous monitoring cost.
''Plant is assumed to have installed a control device with at least 99 percent removal efficiency.
cPlant is assumed to have installed a control device with at least 97 percent removal efficiency.
Source: Enviroscience Control Device Evaluation of Thermal Incineration (Part II Docket Item IV-A-3).
Chemical Manufacturers Association Hearing and Post-Hearing comments (Part II Docket Item IV-D-22).
-------
TABLE A-3. EXISTING PLANT COSTS FOR ACHIEVING 90 PERCENT BENZENE EMISSION REDUCTION
(Control Method: Thermal Incineration with Primary Heat Recovery3)
Plant name
and location
Capacity
(Mg/yr)
Total
installed
cost
($1,000s)
Direct operating
ACC cost for 8,000
($l,000s/yr) hr/yr ($l,000s/yr)
Energy recovery
credit for 6,000
hr/yr ($/yr)
Total annualized cost
for 8,000 hr/yr
($l,000s/yr) ($/Mg product)
Pfizer
10,700
1,680
341
1,007
(629)
719
67.2
Ashland
27,200
2,790
566
2,134
(1,442)
1,258
46.2
Reichholdb
20,000
30
6
2
—
8
41.5
Tennecoc
9,980
30
6
2
—
8
0.8
U.S. Steelb
36,400
30
6
2
—
8
54.1
Monsantod
22,700
30
6
2
—
8
0.4
DENKA<>
38,100
30
6
2
—
8
0.2
aAll costs include continuous monitoring.
bPlant already has 90 percent control.
cPlant is assumed to have installed a 99-percent efficient control device.
dPlant is assumed to have installed a control device with at least 97 percent removal efficiency.
Source: Enviroscience Control Device Evaluation of Thermal Incineration (Part II Docket Item IV-A-3).
Chemical Manufacturers Association Hearing and Post-Hearing comments (Part II Docket Item IV-D-22).
-------
TABLE A-4. EXISTING PLANT COSTS FOR ACHIEVING 97 PERCENT BENZENE EMISSION REDUCTION
(Control Method: Carbon Adsorption3)
Plant name
and location
Capacity
(Mg/yr)
Total
installed
cost
($1,000s)
ACC
($1,000s/yr)
Direct operating
cost for 8,000
hr/yr ($l,000s/yr)
Energy recovery
credit for 8,000
hr/yr ($l,000s/yr)
Total annualized cost
for 8,000 hr/yr
($l,0D0s/yr) (J/Hg product)
Pfi zer
Terre Haute, IN
10,700
3.Z60
256
502
(296)
462
43.1
Ashland
Neal, UV
27,200
1,960
397
521
(233)
685
25.2
Reichhold
Morris, IL
20,000
1,540
312
1,222
(796)
738
36.9
Tenneco*5
Ford, NJ
9,980
30
6
2
—
8
0.8
U.S. Steel
Neville Island, PA
36,400
3,354
680
1,480
(992)
1,168
32.1
Honsantoc
22,700
30
6
2
—
8
0.4
DENKAC
38,100
30
6
2
—
8
0.2
aAll costs Include continuous monitoring cost.
bpiant is assumed to have installed a 99-percent efficient control device.
cplant is assumed to have installed a control device with at least 97 percent efficiency.
Source: Enviroscience Control Device Evaluation of Carbon Adsorption (Part II Docket Item IV-A-2).
Chemical Manufacturers Association Hearing and Post-Hearing comments (Part II Docket Item IV-D-22).
-------
TABLE A-5. EXISTING PLANT COSTS FOR ACHIEVING 90 PERCENT BENZENE EMISSION REDUCTION
(Control Method: Carbon Adsorption3)
Plant name
and location
Capacity
(Mg/yr)
Total
installed
cost
(Jl.OOOs)
ACC
($l,000s/yr)
Direct operating
cost for 8,000
hr/yr ($l,000s/yr)
Energy recovery
credit for 8,000
hr/yr ($l,000s/yr)
Total annualized cost
for 8,000 hr/yr
($l,000s/yr) ($/Mg product)
Pfizer
Terre Haute, IN
10,700
1,260
256
254
(275)
235
22.0
Ashland
Neal, UV
27,200
1,960
397
333
(216)
514
18.9
Reichhold^
Morris, IL
20,000
30
6
2
—
8
0.4
Tennecoc
Ford, NJ
9,980
30
6
2
—
8
0.8
U.S. Steelb
Neville Island, PA
36,400
30
6
2
—
8
0.2
Monsantod
22,700
30
6
2
—
8
0.4
DENKAd
38,100
30
6
2
—
8
0.2
aAll costs include continuous monitoring cost.
''Plant already has 90 percent control.
cPlant is assumed to have installed a 99-percent efficient control device.
dPlant is assumed to have installed a control device with at least 97 percent removal efficiency.
Source: Enviroscience Control Device Evaluation of Carbon Adsorption (Part II Docket Item IV-A-2).
Chemical Manufacturers Association Hearing and Post-Hearing comments (Part II Docket Item (IV-D-22).
-------
the total component factors had been for an overall piping installation
factor of 2.3, as stated in Subsection A-2.
A-4. ESTIMATION OF RETROFIT INSTALLATION FACTORS FOR CARBON ADSORPTION
No detailed guides, such as the Mascone memo, are available for
retrofit costs for carbon adsorption. However, Enviroscience used methods
similar to those used for thermal incineration to develop new source-
installed costs for carbon adsorption. Table IV-2 in the carbon adsorption
evaluation, Factors Used for Estimating Total Installed Costs, is identical
to Table V-l in the thermal incineration evaluation. Retrofit factors for
carbon adsorption were assumed to be similar to those for thermal
incineration.
The ratio of overall retrofit factor to overall new source factor for
thermal incineration varies from 1.625 to 1.4. Because carbon adsorption
units are generally easier to install than are thermal incineration units,
the low end of the scale--l.4--was used. No allowance for additional
ductwork was made for Reichhold because ducts used in its current carbon
adsorption system could be used in an upgraded system. Ductwork was added
to USS Chemicals as before. Therefore, a retrofit of 1.3 will be used for
USS Chemicals to avoid counting ductwork retrofit twice. The costs will be
updated as for thermal incineration. Table A-4 shows the costs for 97
percent control, while Table A-5 shows the costs for 90 percent control.
A-12
-------
APPENDIX B: ENERGY REQUIREMENT CALCULATIONS
-------
APPENDIX B
ENERGY REQUIREMENT CALCULATIONS
Calculations were made to determine the energy impacts on each facility for
each of the regulatory alternatives. A sample calculation for the Pfizer plant is
given below. Personnel at each of the other plants calculated energy impact using
the same equations but substituting individual plant parameters. Table B-l shows
energy impacts for all of the plants.
SAMPLE CALCULATION OF ENERGY REQUIREMENTS*—PFIZER PLANT
(90 AND 97 PERCENT BENZENE EMISSION REDUCTION BY CARBON ADSORPTION)
Ninety percent reduction
191 lb benzene
hr
0.90
100 lb carbon
6 lb benzene
from CMA testimony
(Part II Docket
Item IV-F-8)
carbon adsorption
control efficiency
(Part II Docket
Item IV-A-2, p. I1-4)
0.3 lb steam
lb carbon
8,000 hr 1,000 Btu 1.054 x 10"6 GJ GJ
x ' x — = 7,200 — .
yr lb steam Btu yr
(Part II Docket
Item IV-A-2, p.
11-17)
Energy required
for steam per
year
B-2
-------
Ninety-seven percent reduction
191 lb benzene 100 lb carbon
x 0.97 x x
hr 6 lb benzene
from CMA testimony carbon adsorption (Part II Docket Item
(Part II Docket Item control efficiency IV-A-2, p. I1-4)
IV-F-8)
1.0 lb steam 8,000 hr 1,000 Btu 1.054 x 10"® GJ 26,000 GJ
x x x = .
lb carbon yr lb steam Btu yr
(Part II Docket Energy required
Item IV-A-2, for steam per
p. 11-17) year
SAMPLE CALCULATION OF ENERGY REQUIREMENT—PFIZER PLANT
(97 PERCENT BENZENE EMISSION REDUCTION BY THERMAL INCINERATION)
Heat value of benzene in waste gas
191 lb benzene/hr x 17,270 Btu/lb benzene x 0.97 = 3.2 10® Btu/hr.
from CMA AH comb fraction from benzene .
testimony (net, 1,400° F) of benzene
(Part II Docket Item, consumed
IV-F-8)
Heat content of nonbenzene VOCs in waste gas for model plant
# of moles 41 ooo g/hr 8,000 g/hr 8,000 g/hr
from Table 1-4 —- — + - - + - - - 1,610 g-mol/hr.
BID (model plant) 30 g/g-mol 46 g/g-mol 116 g/g-mol
(Part II Docket
Item, III-B-l) CH2O CH202 C4H202(0H)2
B-3
-------
Temperature of waste gas after heat exchanger
h is the enthalpy _ hH0 hj . o.92
(all enthalpies 0.5 = - .
for air) ^1,400 " ^110 26.13 - 0.92
hT = 13.52 Btu/SCF; therefore, T = 778° F.
Gross incinerator heat requirement
^1,400 ~ ^110 = 26.13 - 0.92 = 25.21 Btu/SCF.
Heat recovery credit
h778 - huo = 13.52 - 0.92 .= 12.60 Btu/SCF.
Conversion of volumetric flow rate to standard cubic feet
530° R
12.03 m3/s x x 35.3 ft3/m3 x 3,600 s/hr = 1,420,000 SCF/hr.
569° R
CMA Testimony CMA
(Part II Docket Testimony
Item, IV-F-8)
Incinerator fuel heat requirement
25.21 Btu/SCF x 1,420,000 SCF/hr = 35.8 x 106 Btu/hr.
gross incinerator
heat requirement
12.03 Btu/SCF x 1,420,000 SCF/hr = 17.1 x 106 Btu/hr.
heat recovery
credit
B-5
-------
1,610 mol/hr x 123.5 kcal/mol x
1 Btu
0.252 kcal
= 0.79 x 106 Btu/hr.
AH combustion
formaldehyde
(Perry*s Chemical
Engineering Handbook,
5th 3., p. 3-137)
Heat value of VOCs in waste gas for Pfizer
0.79 x 106 Btu/hr x
10.7
22.7
0.97
Pfizer capacity fraction
model pi ant of VOCs
capacity consumed
0.36 x 10® Btu/hr.
from benzene.
Heat value of CO in waste gas (Part II Docket Item II-A-4)
10-7 c
4,244 lb CO/hr x 4,344 Btu/lb x x 0.90 = 7.8 x 105 Btu/hr (from CO).
22.7
Total heat value of waste gas
(3.2 + 0.36 + 7.8) x 106 Btu/hr = 11.4 x 106 Btu/hr.
Incinerator System
(Incinerator temperature from Part II Docket Items II-B-1 and IV-J-10)
exhaust to atmosphere
50% heat recovery
adsorber waste gas
110° F
400° F
6
fuel
-------
(35.8 - 17.1 - 11.5) x 106 Btu/hr = 7.2 x 106 Btu/hr.
incinerator heat total heat total heat
heat recovery value of added as
requirement credit waste gas fuel
Yearly fuel requirement
7.2 x 106 Btu/hr x 8,000 hr/yr - 57,600 x 106 Btu/yr.
B-6
-------
APPENDIX C: ADDENDUM TO THE ECONOMIC ANALYSIS OF NESHAP
FOR THE MALEIC ANHYDRIDE AND FUMARIC ACID INDUSTRIES
-------
APPENDIX C
ADDENDUM TO THE ECONOMIC ANALYSIS OF NESHAP FOR THE MALEIC
ANHYDRIDE AND FUMARIC ACID INDUSTRIES
C.l INTRODUCTION
C.l.l Scope of Work
This appendix updates the economic analysis portion of the February
1980 draft Environmental Impact Statement (EIS), Benzene Emissions from
Maleic Anhydride Industry—Background Information for Proposed Standards.
It highlights changes in the control alternatives, industry structure, and
economic effects since 1978, the latest year for data in that document.
The economic analysis presented below summarizes impacts resulting
from control of benzene emissions at maleic anhydride and fumaric acid
plants. Costs of an incinerator installed to meet the promulgated
standards were evaluated to determine impacts on the industry (e.g., total
annualized costs, increased prices, changes in balance of trade) and
impacts on firms in the industry (e.g., shifts in competitive advantage due
to price differentials and effect of transportation on firms' penetration
of other firms' market areas).
Costs for each firm were adjusted in the economic analysis to reflect
the lowest and highest industry capacity use rates over the last 5 years,
thus providing a range of costs on the industry level. In the case of U.S.
Steel and Tenneco, which produce both maleic anhydride and fumaric acid,
costs were allocated to each product based on relative capacities and
expected capacity use rates in each industry. Before-tax annualized costs
for each firm were aggregated for total-cost yield for the maleic anhydride
and fumaric acid industries to determine industry impacts of the National
Emission Standards for Hazardous Air Pollutants (NESHAP). In examination
of intraindustry impacts, costs were evaluated on an after-tax basis.
C-2
-------
C.1.2 Baseline Conditions
Not all of the firms in the maleic anhydride and fumaric acid
industries will incur costs under the promulgated standards. In the maleic
anhydride industry, Monsanto has installed equipment to meet a control
level equal to 97 percent. Amoco, Denka, Ashland, and Koppers are not
affected by the benzene standards because they produce maleic anhydride by
a butane process or as a byproduct in phthalic anhydride manufacture.
Tenneco is expected to install pollution control equipment to meet a
99-percent control level under New Jersey's State law and, therefore,
Tenneco will only incur the cost of a monitor under the NESHAP.
It is assumed that no plant in the maleic anhydride industry will have
to install any control equipment to meet the promulgated standards, since
all benzene-based plants achieve at least 90 percent reduction or better.
However, Reichhold, U.S. Steel, Monsanto, and Tenneco might incur small
costs to comply with monitoring requirements.
Regarding the fumaric acid industry, Monsanto is not affected by the
standards since its fumaric acid production is not benzene based. Under
the promulgated standards, only Pfizer would have to install an
incinerator; U.S. Steel and Tenneco are required to install monitors only.
C.2 STRUCTURE
C.2.1 Maleic Anhydride
Several noteworthy changes in the maleic anhydride industry occurred
between 1977 and 1980. First, two plants closed (Reichhold1s Elizabeth,
New Jersey, plant and Kopper's Bridgeville, Pennsylvania, plant) and
another reduced its capacity (Reichhold's Morris, Illinois, plant),
reducing industry capacity from 238 Gg/year in 1978 to 202 Gg/year in
December 1980. At the same time these smaller plants closed, Monsanto
expanded its St. Louis plant capacity to 52 Gg/year and currently plans to
build a 59-Gg/year plant at Pensacola, Florida. The trend, therefore, is
toward greater concentration within the industry. Following construction
of its new plant, Monsanto's market share will increase from its 1978 rate
of 20 percent to 43 percent, and the four-firm concentration ratio will
C-3
-------
jump from its 1978 rate of 66 percent to 77 percent. As shown in Table
C-l, market shares for other firms in the industry following construction
of Monsanto's new plant will be as follows: U.S. Steel--14 percent,
Amoco--10 percent, Ashland--10 percent, DENKA—9 percent, Reichhold--8
percent, Tenneco--4 percent, and Koppers--2 percent.
Production of maleic anhydride has increased from 133 gigagrams in
1977 to 160 gigagrams in 1979.1 Maleic anhydride demand is expected to
grow at an average annual rate of 8 percent per year over the next several
years,2 which would boost production to 187 gigagrams in 1981.
Over the past 5 years, capacity use in the maleic anhydride industry
has ranged from a low of 56 percent in 1977 to a high of about 86 percent
in 1980.3* with Monsanto's new plant coming online, capacity use will drop
to 72 percent, assuming no other changes in industry capacity.
A second noteworthy trend in the maleic anhydride industry has been
conversion from benzene to n-butane feedstocks. Because n-butane costs
less than benzene, new plants are usually butane-based and an increasing
number are converting to the n-butane process. Monsanto will complete
conversion of its St. Louis plant to n-butane by 1985, and Denka and
Ashland have switched to n-butane. Amoco is butane-based, as will be
Monsanto's new Pensacola, Florida, plant. Except for Koppers, which
produces maleic anhydride as a phthalic anhydride byproduct, all other
plants are benzene based.
The import/export situation for U.S. producers in the maleic anhydride
industry has reversed from unfavorable to favorable over the past few
years. In 1977, maleic anhydride imports reached a high of 2.7 gigagrams,4
purportedly because subsidized production and excess capacity abroad drove
prices of foreign produced maleic anhydride below prices in the United
States. The industry's use is calculated by dividing production (or
estimated 1980 production) by industry capacity. However, partly because
of the decline of the U.S. dollar vs. other currencies, maleic anhydride
imports have dropped over the past several years to 0.03 gigagram in 1980.5
*The industry's capacity use is calculated by dividing production (or
estimated 1980 production) by industry capacity.
C-4
-------
TABLE C-l. FIRM CAPACITIES FOR MALEIC ANHYDRIDE INDUSTRY9
Capacity
Market share
Firm
(Gg/year)
(Percent)
Monsanto
111.1
42.6
U.S. Steel
36.3
13.9
Amoco
27.2
10.4
Ashland
27.2
10.4
DENKA
22.7
8.7
Reichhold
20.0
7.7
Tenneco
11.8
4.5
Koppers
4.5
1.7
TOTAL
260.8
100
Capacities reflect each firm's 1980 end-year capacity, except for Monsanto's.
Monsanto's capacity includes a 52-Gg/year facility in Pensacola, Florida,
with startup slated for the early 1980s.
SOURCE: Chemical Marketing Reporter. December 15, 1980.
C-5
-------
Conversely, exports have increased from 0.6 gigagram in 1977 to about 2.1
gigagrams in 1980.6
Prices in the maleic anhydride industry have continued to fluctuate in
recent years, reflecting volatile demand for the product. Thus, the market
price of maleic anhydride fell from 71^/kg in 1976 to 53tf/kg in 1978 and
increased to 79tf/kg in 1979. The 1980 market price is estimated to be
93^/kg, derived by assuming the market price is 27 percent below the
average August 1980 list price of $1.17/kg.® The fact that market prices
have averaged close to 27 percent below list prices over the past 5 years®
reflects producers' willingness to discount prices when excess capacity
exists.
C.2.2 Fumaric Acid
Four firms comprise the fumaric acid industry, including Monsanto,
Pfizer, Tenneco, and U.S. Steel. Based on 1980 data, total industry
capacity is approximately 33 Gg/yr.l^ Monsanto and Pfizer share top
industry positions, with market shares of 40.8 and 32.0 percent,
respectively; Tenneco and U.S Steel each have a market share of 13.6
percent.H All but Pfizer have the capability of producing maleic
anhydride in addition to fumaric acid.
Production of fumaric acid declined to 13 gigagrams in 1978 from a
1973 level 12 0f 24 gigagrams because 1t lost its primary market for food
acidulants.1^ Although 1979 production figures show a surge from the 1978
level of 13 gigagrams to 23 gigagrams,14 the projected industry growth rate
over the next few years has been estimated to be a low or negative
percent.*5 Capacity use over the past few years has ranged from a low of
33 percent in 1978 to 59 percent in 1979,1® indicating industry probably
suffers from excess capacity.
The balance of trade situation for the fumaric acid industry is
similar to that of the maleic anhydride industry. Low-priced fumaric acid
from foreign producers resulted in 1977 imports of 1.4 gigagrams,^ which
fell to 0.05 gigragram in 1980.18 Although exports have not increased to
the same degree for fumaric acid as for maleic anhydride, they increased
from 1.6 to 2.2 gigagrams between 1978 and 1980.
C-6
-------
Fumaric acid market prices follow a pattern similar to that for maleic
anhydride market prices. The latter dropped from 93g/kg in 1976 to 90^/kg
in 1978 and increased to $1.04/kg in 1979,20 which is 4 percent below the
list price of $1-21/kg in August 1980.The market and list prices of
fumaric acid have remained close, deviating no more than 5 percent from one
another over the past 5 years.
C.3 RESULTS
The following tables summarize the economic impacts resulting from the
promulgated standards. Table C-2 shows industry impacts, including total
annualized costs, percent price increase, number of firms affected, and
concentration effects. Table C-3 summarizes intraindustry impacts,
specifying which firms suffer cost disadvantages compared to competitors as
a results of the standard.
C.3.1 Intra-Industry Impacts
Maleic anhydride prices are not expected to increase as a result of
the standard, since no plant will have to install a control system.
In the funamir acid industry, Pfizer could be at a 4- to 7-percent
cost disadvantage in relation to both Monsanto and U.S. Steel. U.S.
Steel's and Tenneco's cost disadvantages would be very small and therefore
should not lead to loss in market share to competitors. Any gain in
Monsanto's market share would lead to greater concentrations in the fumaric
acid industry.
Although the analysis shows potentially significant impacts on Pfizer,
it is unlikely that the standards will cause closure for two reasons.
First, Pfizer can buy maleic anhydride and occasionally does when
production costs exceed prices. In fact, they are currently purchasing
about 10 percent of their needs. They would probably opt to purchase
maleic anhydride instead of installing a control device. Second, even if
they elected to install a control device instead of putting them at a
disadvantage, this would only erase an advantage they have had over other
fumaric acid producers, which have previously had to install controls for
other regulations.
C-7
-------
TABLE C-2. ECONOMIC IMPACTS OF THE PROMULGATED STANDARDS ON MALEIC
ANHYDRIDE AND FUMARIC ACID PRODUCERS1
Total Annualized Costs
Product Price Increase^
Firms Affected^
Concentration Effects
$0.03MM
S0.71MM
Negligible
None
3% - 4%
Pfizer
None
Likely increase in industry
concentration as Monsanto
increases market share at
expense of Pfizer
1 Ranges indicate variation of impacts according to whether high or low
capacity utilization was assumed. All costs are based on the assumption
that no plants would close.
2 The product price increase represents the average increase in prices
expected in the industry following promulgation of the NESHAP. It is
determined by dividing the total annualized costs resulting under the
standard by industry-wide sales, calculated by multiplying expected price
and production levels associated with both low and high capacity utilization
3 Firms affects only by monitoring costs ranging between $500 to $8,000 are
excluded.
rates.
C-8
-------
TABLE C-3. SUMMARY OF INTRA-INDUSTRY IMPACTS OF THE PROMULGATED STANDARDS
ON MALEIC ANHYDRIDE AND FUMARIC ACID PRODUCERS
Maleic Anhydride
• Reichhold, U.S. Steel, Monsanto,
and Tenneco would have a negligible
cost disadvantage (less than 0.1
percent) due to monitoring
requi rements.
Fumaric Acid
Pfizer would suffer a 4-5% cost
disadvantage vs. both Monsanto
and U.S. Steel
U.S. Steel, Monsanto, and Tenneco
would have a negligible cost
disadvantage (less than 0.1
percent) due to monitoring require-
ments.
C-9
-------
C.3.2 Industry Impacts
Total annualized costs as a result of the standards for the combined
maleic anhydride and fumaric acid industries would be about $0.75 million.
Maleic anhydride prices are not expected to increase, and fumaric acid
prices may increase by 4 percent.
There are no clearly identifiable adverse employment effects, although
some regional shifts may occur as the markets shift slightly among firms.
Balance of trade effects are expected to be small because the firm with the
highest control costs--Pfizer--has a small export market.
C.4 REFERENCES
1. U.S. International Trade Commission. Synthetic Organic Chemicals--
United States Production and Sales. 1977-79.
2. SRI International. CEH Marketing Research Report on Maleic Anhydride.
In: Chemical Economics Handbook. Menlo Park, California. November
1979.
3. U.S. International Trade Commission. Synthetic Organic Chemicals--
U.S. Production and Sales, 1977-1979. Chemical Marketing Reporter.
December 15, 1980, and February 13, 1978. Maleic Makers Build on
Hopes for Polyester. Chemical Week. February 2, 1977.
4. Bureau of Census, U.S. Department of Commerce. U.S. Imports for
Consumption. I.M. 146.
5. Bureau of Census, U.S. Department of Commerce. U.S. Imports for
Consumption. I.M. 146.
6. Bureau of Census, U.S. Department of Commerce. U.S. Exports. E.M.
546.
7. Chemical Marketing Reporter. May 7, 1979, September 4, 1978;
September 5, 1977; and September 13, 1976.
8. Chemical Marketing Reporter. August 18, 1980.
C-10
-------
9. Chemical Marketing Reporter. August 18, 1980. U.S. International
Trade Commission. Synthetic Organic Chemicals--U.S. Production and
Sales. 1977-1979.
10. Chemical Profile--Fumaric Acid. Chemical Marketing Reporter.
September 11, 1978. Chemical Briefs--Fumaric Acid. Chemical
Purchasing. May 1980.
11. Chemical Profile--Fumaric Acid. Chemical Marketing Reporter.
September 11, 1978. Chemical Briefs--Fumaric Acid. Chemical
Purchasing. May 1980.
12. U.S. International Trade Commission. Synthetic Organic Chemicals--
U.S. Production and Sales. 1977-1979.
13. Chemical Profile--Fumaric Acid. Chemical Purchasing. September 11,
1978.
14. U.S. International Trade Commission. Synthetic Organic Chemicals--
U.S. Production and Sales. 1977-1979.
15. Chemical Profile--Fumaric Acid. Chemical Purchasing. September 11,
1978.
16. U.S. International Trade Commission. Synthetic Organic Chemicals--
U.S. Production and Sales. 1977-1979.
17. Bureau of Census, U.S. Department of Commerce. U.S. Imports for
Consumption. I.M. 146.
18. Bureau of Census, U.S. Department of Commerce. U.S. Imports for
Consumption. I.M. 146.
19. Bureau of Census, U.S. Department of Commerce. U.S. Exports. E.M.
546.
20. U.S. International Trade Commission. Synthetic Organic Chemicals—
U.S. Production and Sales. 1977-1979.
21. Chemical Marketing Reporter. August 18, 1980.
C-ll
-------
APPENDIX D
METHODOLOGY FOR ESTIMATING LEUKEMIA INCIDENCE AND MAXIMUM
LIFETIME RISK FROM EXPOSURE TO BENZENE EMISSIONS FROM
MALEIC ANHYDRIDE PROCESS VENTS
-------
APPENDIX D
METHODOLOGY FOR ESTIMATING LEUKEMIA INCIDENCE AND MAXIMUM
LIFETIME RISK FROM EXPOSURE TO BENZENE EMISSIONS FROM
MALEIC ANHYDRIDE PROCESS VENTS
D.l INTRODUCTION
The purpose of this appendix is to describe the methodology and to provide
the information used to estimate leukemia incidence and maximum lifetime risk
from population exposure to benzene emissions from maleic anhydride process
vents. The methodology consists of four major components: estimating annual
average concentration patterns of benzene in the region surrounding each plant,
estimating the population associated with each computed concentration, computing
dosage by summing the products of the concentrations and associated populations,
and estimating annual incidence and maximum lifetime risk from dosage and
concentration estimates. Due to the assumptions made in each of these four
steps of the methodology, there is considerable uncertainty associated with the
lifetime individual risk and leukemia incidence numbers calculated in this
appendix. These uncertainties are explained in Section D.6 of this appendix.
A description of the health effects of benzene are not included in this appendix;
however, information on health effects is contained in EPA docket number OAQPS 79-3
and Response to Public Comments on EPA's Listing of Benzene Under Section 112 and
Relevant Procedures for the Regulation of Hazardous Air Pollutants, EPA 450/5-82-003.
D.2 ATMOSPHERIC DISPERSION MODELING
The long-term version of the Industrial Source Complex (ISCLT) dispersion
i ....
model was used to estimate annual average benzene concentrations in the vicinity
of five maleic anhydride plants. (Denka and Ashland have switched to n-butane
since proposal.)
Seasonal or annual stability array (STAR) summaries are principal meteoro-
logical input to the ISCLT dispersion model. STAR data are standard climatological
frequence of occurrence summaries formulated for use in EPA models and available
for major U.S. sites from the National Climatic Center, Asheville, N.C. A STAR
summary is a joint frequency of occurrence of wind speed stability and wind
direction categories, classified according to the Pasquill stability categories.
For this modeling analysis, seasonal STAR summaries were used. Urban mixing
heights and rural mixing heights were used for plants in urban and rural areas,
D-l
-------
respectively. The ISCLT dispersion model also required user input of ambient
temperatures by stability category and mixing heights by stability and wind
speed categories, for each season. Seasonal temperature and mixing height input
data were computed by averaging hourly CRSTER meteorological preprocessor data
for each category.
The model receptor grid consists of 10 downwind distances located along
76 radials. The radials are separated by 22.5° intervals beginning with 0.0°
and proceeding clockwise to 337.5°. The 10 downwind distances for each radial
are 0.2, 0.3, 0.5, 0.7, 1.0, 2.0, 5.0, 10.0, 15.0, and 20.0 kilometers. For
plants with only one stack, the stack was assumed to be at the center of the
receptor grid. For the U.S. Steel plant with two sources, stacks 502 and 1602
(see Section D.5.1) were assumed to be at the center of the receptor grid and
stacks 501 and 1601 assumed to be 61 meters southeast of the grid center.
The ISCLT output for all plants modeled, consisting of annual concentration
estimates at all 160 receptors, is contained in the docket (Part II, Docket
item IV-J-16). ISCLT dispersion model concentration estimates have been found
2
to be within a factor of two of measured concentrations in most tests.
D.3 POPULATION AROUND MALEIC ANHYDRIDE PLANTS
The human exposure model (HEM)3 was used to estimate the population that
resides in the vicinity of each receptor coordinate surrounding each maleic
anhydride plant. A slightly modified version of the "Master Enumeration District
List-- Extended" (MED-X) data base is contained in the HEM and used for population
pattern estimation. This data base is broken down into enumeration district/
block group (ED/BG) values. MED-X contains the population centroid coordinates
(latitude and longitude) and the 1970 population of each ED/BG in the United
States (50 States plus the District of Columbia). For human exposure estimations,
MED-X has been reduced from its complete form (including descriptive and summary
data) to produce a randomly accessible computer file of only the data necessary
for the estimation. A separate file of county-level growth factors, based on
the 1970 to 1980 growth factor at the county level, has also been created for
use in estimating 1980 population figures for each ED/BG. The population "at
risk" to benzene exposure was considered to be persons residing within 20 km of
maleic anhydride plants. The population around each plant was identified by
specifying the geographical coordinates of that plant.
D-2
-------
D.4 POPULATION DOSAGE METHODOLOGY
D.4.1 Dosage Methodology
The plant's geographical coordinates and the concentration patterns computed
by the ISCLT were used as input to the HEM. (The HEM also has its own atmospheric
dispersion model. However, the HEM dispersion model, still under development,
is not as detailed as the ISCLT and less well suited for large, point emission
sources.)
For each receptor coordinate, the concentration of benzene estimated by the
ISCLT and the population estimated by the HEM to be exposed to that particular
concentration are identified. The HEM multiplies these two numbers to produce
population dosage estimates and sums these products for each plant. A two-level
scheme has been adopted in order to pair concentrations and populations prior to
the computation of dosage. The two-level approach is used because the concen-
trations are defined on a radius-azimuth (polar) grid pattern with nonuniform
spacing. At small radii, the grid cells are much smaller than ED/BG's; at large
radii, the grid cells are much larger than ED/BG's. The area surrounding the
source is divided into two regions, and each ED/BG is classified by the region
in which its centroid lies. Population dosage is calculated differently for the
ED/BG's located within each region.
For ED/BG centroids located between 0.1 km and 2.8 km from the emission
source, populations are divided between neighboring concentration grid points.
There are 96 (6 x 16) polar grid points within this range. Each grid point has
a polar sector defined by two concentric arcs and two wind direction radials.
Each of these grid points is assigned to the nearest ED/BG centroid identified
from MED-X. The population associated with the ED/BG centroid is then divided
among all concentration grid points assigned to it. The exact land area within
each polar sector is considered in the apportionment.
For population centroids between 2.8 km and 20 km from the source, a
concentration grid cell, the area approximating a rectangular shape bounded by
four receptors, is much larger than the area of a typical ED/BG (usually 1 km in
diameter). Since there is a linear relationship between the logarithm of
concentration and the logarithm of distance for receptors more than 2 km from
the source, the entire population of the ED/BG is assumed to be exposed to the
concentration that is geometrically interpolated radially and azimuthally from
the four receptors bounding the grid cell. Concentration estimates for 80
D-3
-------
(5 x 16) grid cell receptors at 2.0, 5.0, 10.0, 15.0, and 20.0 km from the
source along each of 16 wind directions are used as reference points for this
interpolation.
In summary, two approaches are used to arrive at coincident concentra-
tion/population data points. For the 96 concentration points within 2.8 km of
the source, the pairing occurs at the polar grid points using an apportionment
of ED/BG population by land area. For the remaining portions of the grid,
pairing occurs at the ED/BG centroids themselves, through the use of log-log
linear interpolation. (For a more detailed discussion of the methodology used
to estimate dosage, see Reference 3.)
D.4.2 Total Dosage
3
Total dosage (persons-pg/m ) is the sum of all multiplied pairs of
concentration-population computed by the previously discussed methodology:
N
Total dosage = I (P.C.) (1)
i^L1 1
where
Pj = population associated with point i,
C. = annual average benzene concentration at point i, and
N = total number of polar grid points between 0 and 2.8 km and ED/BG
centroids between 2.8 and 20 km.
The computed total dosage is then used with the unit risk factor to estimate
leukemia incidence and maximum lifetime individual risk. This methodology is described
in the following sections.
D.4.3 Unit Risk Factor
_8
The unit risk factor (URF) for benzene is 9.9 x 10 (leukemia cases per year)/
3
((jg/m -person years), as calculated by EPA's Carcinogen Assessment Group (CAG).
The derivation of the URF can be found in the CAG report on population risk to
4
ambient benzene exposure and updated in Response to Public Comments on EPA s
Listing of Benzene Under Section 112 and Relevant Procedures for the Regulation
of Hazardous Air Pollutants, EPA 450/5-82-003.
D.4.4 Calculation of Estimated Annual Leukemia Incidence
The annual leukemia incidence associated with a given plant under
a given regulatory alternative is the product of the total dosage around that
3 -8
plant in |jg/m -persons and the unit risk factor, 9.9 x 10 . Thus,
D-4
-------
Annual Leukemia Incidence = (total dosage) x (unit risk factor) (2)
where total dosage is calculated according to Equation 1.
D.4.5 Calculation of Maximum Lifetime Risk
The populations in areas surrounding maleic anhydride plants have various
risk levels of leukemia incidence from exposure to benzene emissions. Using the
maximum annual average concentration of benzene to which any person is exposed,
it is possible to calculate the maximum lifetime risk of leukemia (lifetime
probability of leukemia to persons exposed to the highest concentration of
benzene) attributable to benzene emissions using the following equation:
Maximum lifetime risk = C. x (URF) x 70, (3)
1 y fllaX
where
Ci max = the maximum annual average concentration at any receptor
' location where exposed persons reside,
-8
URF = the unit risk factor, 9.9 x 10 , and
70 years = average individual's life span.
D.5 MORTALITY AND MAXIMUM LIFETIME RISK
0.5.1 Input Data, Assumptions, and Methodology
Population exposures were computed for several different control scenarios
for each plant in order to estimate the leukemia incidence and maximum lifetime
risk for the current level of control and for evaluating the health impacts
between best demonstrated technology (BDT) and beyond BDT.
These emission scenarios consisted of: (1) no control device (also the
assumed startup mode), (2) properly operating (meeting the specified control
level) control device, and (3) malfunction. The emission rates and other
dispersion model inputs are shown in Table D-l. The original emission rates
supplied by industry (Part II Docket Item IV-F-8) were adjusted for the modeling
to conform with assumptions made at proposal concerning conversion rates. The
modeling results reflect these adjustments. However, it was determined after
the modeling was performed that such adjustments were unnecessary. Rather
than repeat the modeling effort, the results were prorated using a ratio of the
emission rates supplied by industry (Part II Docket Item IV D-22) to those that
were actually modeled. These results were subsequently used in calculating the
health impacts. The corrected emission rates are shown in Table D-2. It should
D-5
-------
TABLE D-l. PLANT SPECIFIC CHARACTERISTICS
ISC model
Plant Control Scenario source no.
Emission Stack Exit Exit
rate height velocity temperature Diameter
(g/s) (m) (m/s) (° K) (m)
U.S. Steel TI
malfunction
101
49.14
33.5
9.7
317
2.17
U.S. Steel CI
malfunction, NCD
501
502
30.24
18.9
33.5
33.5
10.0
9.1
317
317
1.68
1.37
U.S. Steel
90% (CI)
1.601
1.602
10.08
6.30
33.5
33.5
18.6
17.0
589
589
1.68
1.37
U.S. Steel
97% (TI)
2,001
4.914
33.5
18.0
589
2.17
Reichhold TI
malfunction
2
59.2
30.5
11.7
324
1.52
Reichhold CA
malfunction,
NCD
6
59.2
29.3
21.1
324
1.067
Reichhold
90% (CA)
17
10.07
29.3
21.1
324
1.067
Reichhold
97% (TI)
21
3.02
30.5
15.2
422
1. 52
Monsanto TI
malfunction
3
82.9
45.7
16.4
317
2.134
Monsanto
NCD
11
82.9
24.5
44.2
317
0.915
Monsanto
97% (TI)
9
2.64
45.7
16.4
404
2.134
Pfizer TI
malfunction
7
22.05
30.5
41.2
316
1.067
Pfizer
NCD
13
22.05
23.2
41.2
316
0.610
Pfizer
90% (TI)
18
4.41
30.5
17.2
405
1.067
Tenneco TI
malfunction
9
11.46
18.0
10.3
308
1.11
Tenneco
NCD
14
11.46
18.3
30.7
308
0.610
Tenneco
99% (TI)
22
0.40
18.0
18.6
447
1.11
TI = thermal incineration.
CI = catalytic incineration.
CA = carbon adsorption.
NCD = no control device.
D-6
-------
TABLE D-2. ADJUSTED BENZENE EMISSION RATES
PERCENT ,
PLANT CONTROL SCENARIO CONVERSION RATE1 EMISSION RATE, g/s
U.S. Steel
90% (CI)
97
7.371
97% (TI)
97
2.21
NCD, CI Malfunction
97
73.71
TI
Mai function
97
73.71
REICHHOLD
90% (CA)2
94
5.92
97% (TI)
94
1.78
NCD, CA Malfunction
94
59.2
TI
Malfunction
94
59.2
MONSANTO
97% (TI)2
3
953
2.64
NCD
953
88
TI
Malfunction
95
88
PFIZER
2
NCD
95
22.05
90% (TI)
95
2.205
97% (TI)
95
0.662
TI
Malfunction
95
22.05
TENNECO
99% (TI)2
96.3
0.1146
NCD
96.3
11.46
TI
Malfunction
96.3
11.46
TI = Thermal Incineration
CI = Catalytic Incineration
CA = Carbon Adsorption
NCD = No Control Device
1
Conversion rates supplied by industry in Part II Docket Item IV-D-22.
2
Assumed current level of control.
3
Average of operating range of 93 to 97 percent.
D-7
-------
_ 8
also be noted that the CAG URF was revised after proposal to 9.9 x 10 from
_ 7
1.06 x 10 . Plant location and meteorological inputs are provided in Table D-3.
Total dosage for each plant under each scenario is shown in Table D-4. The
3
dosage (person~ng/m ) calculated for each scenario was then prorated according
to the estimated hours per year a given plant would operate under that scenario.
The adjusted dosages for each plant were used in estimating the dosage associated
with each regulatory alternative.
Based on information from two currently controlled maleic anhydride
5 '6
plants, the hours per year a plant would be under each scenario were estimated.
Three total plant startups with 8 hours of uncontrolled emissions each and
15 single-reactor startups with 1.5 hours of uncontrolled emissions each were
assumed for a total estimate of 46.5 hours of startup emissions per year.
Twenty malfunctions with 8 hours (maximum allowable time for excess emissions
from malfunctions) of uncontrolled emissions each were assumed for a total
estimate of 160 hours of malfunction emissions per year. Each plant was assumed
to operate at full capacity with no control for both startup and malfunction
emissions. The plant was assumed to operate properly for the remainder of an
8,000-hour operating year, or 7,793.5 hours. Population dosage for each plant
under each regulatory alternative was then calculated as follows:
Population dosage = 8jq00 S + 8,000 M + 8,000 P' ^
where
S = dosage from startup (Table D-4),
M = dosage from malfunction (Table D-4), and
P = dosage from proper control device operation under each
regulatory alternative (Table D-4).
After population dosage under each regulatory alternative was computed,
leukemia incidence estimates could be made. Population dosage was multiplied by
-8
the CAG leukemia risk factor of 9.9 x 10 -to estimate the number of leukemia
cases per year for each plant.
A similar computation is made for determining maximum lifetime risk. The
maximum annual average benzene concentration under each scenario was prorated
the same way as described above for the dosage calculation. Estimated maximum
annual average concentrations for each plant under each scenario are shown in
Table D-4. These, in turn, were used in estimating the maximum annual average
concentration for each plant for each alternative as follows:
D-8
-------
TABLE D-3. PLANT LOCATIONS AND METEOROLOGICAL INPUTS
Plant
Location
Latitude Longitude
Urban or
rural
mixing
heights
Seasonal
STAR summary
CRSTER model
Meteorologically prepared data
used for ambient temperature
mixing height determinations
Surface data Upper air data
U.S. Steel Neville Island, PA 40.5000
80.0833 Urban
Pittsburgh, PA
1 hr data 1973-77
Pittsburgh, PA
1974
Pittsburgh, PA
1974
Reichhold Morris, IL
42.3839
88.2989 Rural
Rockford, IL
3 hr data 1973-77
Rockford, IL
1974
Peoria, IL
1974
Monsanto St. Louis, MO
38.5833
90.2000 Urban
St. Louis, MO
3 hr data 1973-77
St. Louis, MO
1977
Salem, IL
1977
Pfi zer
Terre Haute, IN 39.3650
87.4150 Rural
Indianapolis, IN
1 hr data 1973-77
Indianapolis, IN Patterson, OH
1977
1977
Tenneco
Fords, NJ
40.5117
74.3189 Urban
Newark, NJ
Newark, NJ
1 hr data 1970-74 1974
JFK, NY
1974
-------
TABLE D-4. DOSAGE AND MAXIMUM ANNUAL AVERAGE CONCENTRATIONS
Total Dgsage Maximum Annual
(|jg/m - Average Concegtrations
Plant Case person) (pg/m )
U.S. Steel
NCD, CI Malfunction
TI Malfunction
CI (90 percent)*
TI (97 percent)
1,830,000
1,660,000
120,000
28,500
35.0
25.2
0.932
0.166
Reichhold
NCD, CA Malfunction
TI Malfunction
CA (90 percent)*
TI (97 percent)
227,000
225,000
22,800
5,070
20.8
21.2
2.09
0.294
Pfizer
NCD*
TI Malfunction
TI (90 percent)
TI (97 percent)
50,800
34,900
4,210
1,260
11.0
4.38
0.481
0.144
Monsanto
NCD
TI Malfunction
TI (97 percent)*
2,730,000
1,930,000
47,300
36.4
11.7
0.231
Tenneco
NCD
TI Malfunction
TI (99 percent)*
334,000
343,000
2,730
13.7
18.9
0.0562
^Assumed current control level (includes known future control levels).
NCD = No control device; equivalent to startup emissions for situations in
which a plant has a control device.
TI = Thermal incinerator.
CI = Catalytic incinerator.
CA = Carbon adsorber.
D-10
-------
c. 46.5 C. + 160 C. + 7,793.5 C. (5)
i ,max = i^max iM,max —1 ip, max v
8,000 5 8,000 " 8,000 K
where
C-t = maximum annual average concentration for a plant under a given
1 9 iTloX
regulatory alternative,
C• = maximum annual average concentration during startup (Table D-4),
S'
C,- m=w = maximum annual average concentration during malfunction
¦ jui i "laX
(Table D-4), and
ip,max
= maximum annual average concentration during proper control
device operation (Table D-4).
Maximum lifetime risk for a given plant under each regulatory alternative
can be found by multiplying the maximum annual average concentration by the
¦¦ 8
unit risk factor of 9.9 x 10 times 70 years (to obtain a lifetime estimate).
The maximum lifetime risk for the industry under the different alternatives
presented in the BID is that due to the plant with the highest maximum lifetime
risk in the industry for the specified alternative.
D.5.2 Example Calculations
D.5.2.1 Leukemia Incidence. As an example for calculating leukemia incidence,
the U.S. Steel plant is used. In the case of BDT (90 percent control), the pop-
ulation dosage is computed according to Equation 4 as follows:
Population dosage = 46,5 (1,830,000) + 160 (1,830,000) + 7>793-5 (120,000)
8,000 8,000 8,000
3
Population dosage = 164,000 person-pg/m .
Therefore, leukemia cases per year (from Equation 2) are:
Leukemia cases per year = 164,000 x 9.9 x 10-8
Leukemia cases per year = 0.16.
D-ll
-------
In the case of beyond BDT (97 percent control), population dosage is:
Population dosage = (1,830,000)+ g^gg-0 ,660,000)+ (28,500)
Population dosage = 71,600
Further, annual leukemia incidence is:
Leukemia cases per year = 71,600 x 9.9 x 10~8
Leukemia cases per year = 0.0071.
0.5.2.2 Maximum Lifetime Risk. Again, U.S. Steel is used to illustrate the
calculation. In the case of BDT, the maximum annual average concentration as
defined by Equation 5 is as follows:
Ci,max = 57555" <35'°> + 57555" (35'0) * 24765r <°'932)
Ci,max = 1'81
Maximum lifetime risk according to Equation 3 is as follows:
.8
Maximum lifetime risk = 1.81 x 9.9 x 10 x 70
_5
Maximum lifetime risk = 1.3 x 10
Likewise, for beyond BDT:
ci,max = 57555" <35'°> * 5^ <25'2> + Zt755T <0166>
Cl,max = °'869 and
_ 8
Maximum lifetime risk = 0.869 x 9.9 x 10 x 70
_6
Maximum lifetime risk = 6.0 x 10
0-12
-------
D.5.3 Summary of Impacts
The methodology for calculating annual incidence and maximum lifetime risk
(described in Section D.5.1) was extended to each plant for the baseline, BDT,
and beyond BDT. The baseline scenario is the current level of control or the
level of control to be implemented in the near future that would have occurred
regardless of this standard. The estimated annual leukemia incidence is shown in
Table D-5. The estimated nationwide annual leukemia incidence under the assumed
baseline level of control ranges from 0.013 to 0.092.* The estimated maximum
lifetime risk is shown in Table D-6. The estimated maximum lifetime risk under
the assumed baseline level of control ranges from 2.9 x 10"^ to 2.0 x 10"^.*
D.6 UNCERTAINTIES
Estimates of both leukemia incidence and maximum lifetime risk are primarily
functions of estimated benzene concentrations, populations, the unit risk factor,
and the exposure model. The calculations of these variables are subject to a
number of uncertainties of various degrees. Some of the major uncertainties are
identified below.
D.6.1 Benzene Concentrations
Modeled ambient benzene concentrations depend upon: (1) plant configuration,
which is difficult to determine for more than a few plants; (2) emission point
characteristics, which can be different from plant to plant and are difficult to
obtain for more than a few plants; (3) emission rates which may vary over time;
and from plant to plant; and (4) meteorology, which is seldom available for a
specific plant. The particular dispersion modeling used can also influence the
numbers. The best model to use (ISC) is usually too resource intensive for modeling
a large number of sources. Less complex models introduce further uncertainty
through a greater number of generalizing assumptions. Dispersion models also
assume that the terrain in the vicinity of the source is flat. For sources located
in complex terrain, the maximum annual concentration could be underestimated by
several fold due to this assumption. Assuming the inputs to the dispersion model
are accurate, the predicted benzene concentrations are considered to be accurate
to within a factor of 2.
*These ranges represent the uncertainty of estimates concerning benzene
concentrations to which workers were exposed in the occupational studies of
Infante, Aksoy, and Ott that served as the basis for developing the unit risk
factor, and are based on a 95 percent confidence interval that assumes the
estimated benzene concentrations are within a factor of 2.
D-13
-------
TABLE D-5. ESTIMATED ANNUAL LEUKEMIA INCIDENCE (X10~2)*
Plant Baseline BDT Beyond BDT
U.S. Steel
0.61-4.2
0.61-4.2
0.27-1.9
Reichhold
0.11-0.73
0.11-0.73
0.042-0.29
Pfizer
0.19-1.3
0.019-0.13
0.0084-0.058
Monsanto
0.38-2.6
0.38-2.6
0.38-2.6
Tenneco
0.042-0.29
0.042-0.29
0.042-0.29
Total
1.3-9.2
1.1-7.9
0.76-5.2
These ranges represent the uncertainty of estimates concerning benzene
concentrations to which workers were exposed in the occupational studies of
Infante, Aksoy, and Ott that served as the basis for developing the unit risk
factor, and are based on a 95 percent confidence interval that assumes the
estimated benzene concentrations are within a factor of 2.
D-14
-------
_ 6
TABLE D-6. ESTIMATED MAXIMUM LIFETIME RISK (xlO )*
Plant
Baseline
BDT
Beyond BDT
U.S. Steel
5.0-34
5.0-34
2.3-16
Reichhold
6.9-47
6.9-47
2.2-15
Pfizer
29-200
1.6-11
0.76-5.2
Monsanto
1.8-12
1.8-12
1.8-12
Tenneco
1.3-9.2
1.3-9.2
1.3-9.2
These ranges represent the uncertainty of estimates concerning benzene
concentrations to which workers were exposed in the occupational studies of
Infante, Aksoy, and Ott that served as the basis for developing the unit risk
factor, and are based on a 95 percent confidence interval that assumes the
estimated benzene concentrations are within a factor of 2.
0-15
-------
D.6.2 Exposed Populations
Several simplifying assumptions were made with respect to the assumed
exposed population. In addition, the exposed population is assumed to be
unmobile, remaining at the same location 24 hours per day, 365 days per year,
for a lifetime (70 years). This assumption is counterbalanced to some extent
(at least in the calculation of incidence) by the assumption that no one moves
into the exposure area whetheras a resident or as a transient. The
population "at risk" was assumed to reside within 20 km of each plant, regardless
of the estimated concentration at that point. The selection of 20 km is considered
to be a practical modeling stop-point. The results of dispersion modeling are
felt to be reasonably accurate within that distance. The dispersion coefficients
used in modeling are based on empirical measurements made within 10 kilometers
of sources. These coefficients become less applicable at long distances from
the source, and the modeling results become more uncertain. A numerical estimate
of the accuracy of these assumptions regarding the exposed population is not
available.
D.6.3 Unit Risk Factor
The unit risk factor contains the uncertainty of estimates concerning
benzene concentrations to which workers were exposed in the occupational studies
4
of Infante, Aksoy, and Ott, which serve as the basis for the unit risk factor.
The ranges represent a 95 percent confidence interval that assumes the estimated
benzene concentrations to which the workers were exposed are within a factor of
2 of the actual concentrations. Other uncertainties regarding the occupational
studies and the workers exposed that may affect the unit risk factor were raised
during the public comment period and focus on assumptions and inconclusive data
contained in the studies. However, those uncertainties have not been quantified.
D.6.4 Other Uncertainties
There are several other uncertainties associated with estimating health impacts.
Maximum lifetime risk and annual incidence were calculated based on a no-threshold
linear extrapolation of leukemia risk associated with a presumably healthy white
male cohort of workers exposed to benzene concentrations in the parts per million
range compared to the risk associated with the general population, which includes
men, women, children, nonwhites, the aged, and the unhealthy, who are exposed to
concentrations in the parts per billion range. It is uncertain whether these
widely diverse segments of the population have susceptabi1ities to leukemia that
differ from that of workers in the studies. Furthermore, while death from
leukemia is the only benzene health effect considered in these calculations, it
D-16
-------
is not the only possible health effect. Other health effects, such as aplastic
anemia and chromosomal aberrations, are not as easily quantifiable and are not
reflected in the risk estimates. Although these other health effects have been
observed at occupational levels, it is not clear if they can result from ambient
benzene exposure levels. Additionally, benefits that would affect the general
population as the result of indirect control of other organic emissions in the
process of controlling benzene emissions from maleic anhydride plants are not
quantified. Possible benzene exposures from other sources also are not included
in the estimate. For example, an individual living near a maleic anhydride
plant is also exposed to benzene emissions from automobiles. Finally, these
estimates do not include cumulative or synergistic effects of concurrent expo-
sure to benzene and other substances.
D. 7 REFERENCES
1. U.S. Environmental Protection Agency. Industrial Source Complex
(ISC) Dispersion Model User's Guide, Volume I. Research Triangle
Park, North Carolina. EPA-450/4-79-031. 1979
2. U.S. Environmental Protection Agency. An Evaluation Study for the
Industrial Source Complex (ISC) Dispersion Model. Research Triangle
Park, North Carolina. EPA-450/4-81-002. 1981.
3. Systems Applications, Inc. Human Exposure to Atmospheric Concentrations
of Selected Chemicals. (Prepared for the U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina). Volume I (NTIS
No. PB81 193252) and Volume II (NTIS No. PB81 193260). May 1980.
4. Albert , R. E. Carcinogen Assessment Group's Final Report on Popula-
tion Risk to Ambient Benzene Exposures. U.S. Environmental Protection
Agency. Publication No. EPA-450/5-80-004. January 1979.
5. Letter from Basil, J. A., Reichhold Chemicals, Inc., to J. L. Warren,
Research Triangle Institute, May 2, 1979, (Part II Docket Item II-D-48).
6. Letter from Meyer, A. J., DENKA Chemical Corporation, to J. L. Warren,
Research Triangle Institute, May 30, 1979, (Part II Docket Item II-D-51).
D-17
------- |