United States Office of Air Quality EPA-450/3-84-014b
Environmental Protection Planning and Standards March 1987
Agency Research Triangle Park NC 27711
Air ' ——
v>EPA Review of National Final
Emission Standards EIS
for Mercury
-------
EPA-450/3-84-014b
Review of
National Emission Standards
for Mercury
Emission Standards and Engineering Division
Office of Air and Radiation
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park NC 27711 „ Environrr,ental Protection
March'1987 E-.^Ion 5, Library l51^1^
o -> s Dearborn Street, Room
Chicago, IL 60604
-------
This report has been reviewed'by the Emission Standards and Engineering Division ot the Office of Air Quality
Planning and Standards, EPA, and approved for publication. Mention of trade names or commercial products is not
intended to constitute endorsement or recommendation for use. Copies of this report are available through the
Library Services Office (MD-35), US Environmental Protection Agency, Research Triangle Park NC 27711, or from
National Technical Information Services, 5285 Port Royal Road, Springfield VA 22161.
-------
ENVIRONMENTAL PROTECTION AGENCY
Background Information
and Final
Environmental Impact Statement
for Revisions to the National Emission
tandards for Mercury
Prepared by:
Jack R. Farmer, Director (Date)
Emission Standards and Engineering Division
U.S. Environmental Protection Agency (MD-13)
Research Triangle Park, North Carolina 27711
1. The promulgated revisions to the national emission standards for
mercury would add monitoring, recordkeeping, and reporting require-
ments for mercury-cell chlor-alkali plants. The promulgated standards
implement Section 112 of the Clean Air Act and are based on the
Administrator's determination that mercury presents a 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.
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; State and Territorial
Air Pollution Program Administrators; EPA Regional Administrators; Local
Air Pollution Control Officials; Office of Management and Budget; and
other interested parties.
3. For additional information contact:
Mr. Gilbert H. Wood
Emission Standards and Engineering Division (MD-13)
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Telephone: (919) 541-5625
4. Copies of this document may be obtained from:
U.S. Environmental Protection Agency Library (MD-35)
Research Triangle Park, North Carolina 27711
Telephone: (919) 541-2777
National Technical Information Service
5285 Port Royal Road
Springfield, Virginia 22161
Telephone: (703)' 487-4650
-------
-------
TABLE OF CONTENTS
Section Ti 11 e
1.0 ' SUMMARY
1.1 Introduction 1-1
1.2 Summary of Major Changes Since Proposal 1-1
t.3 Summary of Impacts of Promulgated Action 1-2
1.3.1 Environmental Impacts 1-2
1.3.2 Energy and Economic Impacts 1-2
2.0 SUMMARY OF PUBLIC COMMENTS
2.1 Monitoring Requirements 2-3
2.2 Recordkeeping Requirements 2-15
2.3 Reporting Requirements 2-18
2.4 Indirect Exposures to Mercury 2-23
2.5 Unregulated Source Categories 2-27
2.5.1 Coal Combustion 2-27
2.5.2 Oil Shale Retorts 2-29
2.6 Miscellaneous Comments 2-30
-------
-------
1.0 SUMMARY
1.1 INTRODUCTION
On December 26, 1984, the Environmental Protection Agency (EPA)
proposed revisions to the national emissions standards for the hazardous
air pollutant mercury (49 FR 50146). The revisions were proposed under
authority of Section 112 of the Clean Air Act. Public comments on the
proposed revisions were requested, and 10 comment letters were received.
Eight of the letters were from representatives of the mercury-cell
chlor-alkali industry, one letter was from a State health department, and
one letter provided joint comments from two environmental organizations.
All comments and the Agency's responses to them are summarized in this
document. The comments and responses serve as the basis for the changes
that have been made to the proposed revisions.
1.2 SUMMARY OF MAJOR CHANGES SINCE PROPOSAL
One major change has been made to the recommended revisions since
proposal. This change is the addition of an alternative means of complying
with the monitoring requirements for mercury-cell chlor-alkali plants.
The standards now allow the source owner or operator to develop and
submit for the Administrator's approval a site-specific monitoring plan
for monitoring, recording, and reporting emissions from the hydrogen
and end-box ventilation streams. To be approved, an alternative monitor-
ing plan must adhere to the guidelines provided and must ensure not only
compliance with the emission limits but also proper operation and mainte-
nance of the emissions control systems.
-------
1.3 SUMMARY OF IMPACTS OF PROMULGATED ACTION
1.3.1 Environmental Impacts of the Promulgated Action
The promulgated action includes monitoring, recordkeeping, and
reporting requirements for mercury-cell chlor-alkali plants. The primary
purpose of these' requirements is'to encourage plant owners or operators *
to adopt the best practices for operating and maintaining process equipment
and control devices. While emission reductions are anticipated to occur
as a result of better control system practices, no emission reduction
impact has been quantified. Implementation of these requirements will
not result in any adverse solid waste or water impact.
1.3.2 Energy and Economic Impacts of the Promulgated Action
There are no adverse energy impacts associated with the promulgated
revisions. For plants not electing to file site-specific monitoring
plans, there will be an average yearly cost of approximately $9,000
per plant during the first three years that the revisions are in effect.
For plants electing to submit site-specific plans, the cost is expected to
be considerably less. These costs would not result in significantly
adverse economic impacts.
1-2
-------
2.0 SUMMARY Of PUBLIC COMMENTS
This document includes responses to all public comments received on
the revisions to the standards that were proposed on December 26, 1984.
A list of the nine commenters, their affiliations, and the EPA docket
numbers assigned to their correspondence is provided on page 2-2.
For the purpose of orderly presentation,-the comments have been
categorized under the following topics:
1. Monitoring Requirements
2. Recordkeeping Requirements
3. Reporting Requirements
4. Indirect Exposures to Mercury
5. Unregulated Source Categories
6. Miscellaneous Comments
The comments and responses to each comment are presented in the following
sections of this document.
-------
LIST OF COMMENTERS ON PROPOSED REVISIONS TO THE MERCURY STANDARDS
Docket item number3 Commenter and Affiliation
IV-D-1 Mr. George P. Ferreri
Air Management Administration
State of Maryland
Dept. of Health and Mental Hygiene
201 W. Preston St.
Baltimore, MD 21201
IV-D-2 Mr. Michael R. Foresman
Monsanto Industrial Chemicals Co.
800 N. Lindbergh Blvd.
St. Louis, MO 63167
IV-D-3 Mr. Don P. DeNoon
LCP Chemicals & Plastics, Inc.
Raritan Plaza II
CN #3106
Edison, NO 08818
IV-D-4 Mr. P. N. Tobia
Vulcan Chemicals
P. 0. Box 161
Port Edwards, WI 54469
IV-D-5 Mr. David Vaughn
01 in Chemicals
P. 0. Box 248
Charleston, TN 37310
IV-D-6 Mr. James W. Call
Stauffer Chemical Co.
Westport, CT 06881
IV-D-7 Mr. Richard J. Samel son
PPG Industries, Inc.
One PPG Place
Pittsburgh, PA 15272
IV-D-8 Mr. Edmund J. Laubusch
The Chlorine Institute, Inc.
70 West 40th St.
New York, NY 10018
IV-D-9 Mr. L. K. Arehart
Diamond Shamrock Corporation
717 North Harwood St.
Dallas, TX 75201
2-2
-------
LIST OF COMMENTERS ON PROPOSED REVISIONS TO THE MERCURY STANDARDS (con't)
Docket Item number Commenter and Affiliation
IV-D-10 Mr. David D. Doniger
Natural Resources Defense Council, Inc.
1350 New York Avenue, N.W.
Washington, D.C. 20005
and
Environmental Defense Fund
1525"l8th Street, N.W.
Washington, D.C. 20036
aThe docket number for this project is A-82-41. Dockets are on file at
EPA's Central Docket Section (LE-131), West Tower Lobby; Gallery 1,
Waterside Mall, 401 M Street, S.W., Washington, D.C. 20460
2-3
-------
2.1 MONITORING REQUIREMENTS
Comment:
Several commenters believed the standards should allow submittal (to
the Administrator) of plant-specific compliance plans as an alternative to
the proposed monitoring requirements. In support of this alternative, one
commenter (IV-D-2) stated that the parameters ..listed in the proposed revisions
would not be appropriate indicators of mercury removal efforts at one of his
company's facilities. He provided examples to indicate that other parameters
such as the concentration of mercury in the treated hydrogen stream would
be better indicators of compliance at that facility.
Commenter IV-D-7 said that both mercury-cell chlor-alkali plants operated
by his company employ the hydrogen stream for production of ammonia rather
than venting or burning the stream. He stated that these plants maintain
mercury levels in the hydrogen stream far below the amounts required by the
standards. For this reason, the commenter believed that additional monitoring
of this stream was not necessary.
Commenter IV-D-8 stated that in many plants redundant, back-up control
equipment has been installed to ensure compliance and that, as a result,
several monitored parameters could be exceeded for long periods wihout
emissions exceeding the limits of the standards. This commenter believed
that in such cases the standards should not require that all parameters be
monitored. This commenter believed site-specific conditions should
determine the parameters to be monitored.
Commenter IV-D-5 also stated that only one significant control factor
should be used for monitoring and reporting purposes. He stated that the
outlet temperature of the final cooling unit on the primary cooling system
2-4
-------
is the appropriate parameter for plants using cooling systems as the
primary mercury removal system (with or without subsequent control systems).
Experience at his plant indicated that if the outlet temperature of the
gas stream from the final cooler is maintained at a proper temperature then
compliance with _the emission standard is assured. If, on the other hand.,
the outlet temperature increases significantly for an extended period of
time, elevated emissions may be experienced with or without downstream
control equipment.
Commenter IV-D-4 stated that hourly monitoring of parameters was not
i
necessary at all plants. His plant had designed and installed extra
abatement equipment with performance such that upsets, including a total
scrubber bypass, of several hours duration would not result in an
exceedance of the 1,000 grams per day emission limit for the hydrogen and
end-box ventilation streams. He recommended that the monitoring frequency
be established on a site-specific basis, related to the amount of time
that could elapse during a control system failure before the emission
limit would be exceeded.
A second commenter (IV-D-7) also believed hourly monitoring was
unnecessary since a complete failure of control equipment would rarely,
in his opinion, result in an emissions violation within a one-hour period.
He further stated that, in many cases, redundant control equipment is
installed in series so that failure of one item of equipment does not
necessarily indicate a violation of the standard. Because each control
system at each plant is different, this commenter believed that testing
and reporting schemes should be plant specific and that the preamble
should describe the frequency of monitoring that would be appropriate
2-5
-------
relative to the length of time which is critical for compliance. As an
example, the commenter suggested that recording of a parameter which
could cause an excursion after no less than 12 hours would need to occur
only once per four-hour period.
Response:
The monitoring requirements that were proposed were intended to serve
two purposes: to ensure compliance with the emission limits of the standards
and to ensure proper operation and maintenance of the emission reduction
system(s). The Agency recognizes that some of the mercury-cell chlor-alkali
plants operate at emission levels well below the limits of the'standards.
However, the Administrator believes that sound environmental policy dictates
that emission control systems, installed to reduce emissions to meet Federal
health standards, should be operated and maintained .properly on a continuous
basis even if such systems can be operated improperly and still comply with
the emission limits. As indicators of proper operation and maintenance, the
monitoring requirements are considered important elements of the standards
and are being maintained in the final standards.
However, the Agency recognizes, based on the comments received, that
additional monitoring options may be necessary to ensure that the parameter(s)
being monitored is the most appropriate one. To achieve this purpose,
the Agency has added a provision to the regulation that allows individual
plant owners or operators to develop site-specific monitoring plans and
submit them to the Administrator for approval. These plans must assure
proper operation and maintenance of the emission reduction system as well
as compliance with the emission limits. For this reason, the plans are not
to be considered as compliance plans (as stated in the comments) but as
2-6
-------
alternative monitoring plans. It should be noted that owners, or operators
are not required to submit site-specific plans. Submittal of such
plans, followed by adherence to the approved monitoring procedures, is
being provided as an-alternative to meeting the monitoring requirements
of the standards.
If alternative monitoring plans are submitted for approval, they must,
at a minimum, include the following:
1. Identification of the critical parameter or parameters for the
hydrogen stream and for the end-box ventilation stream that are to be
monitored and an explanation of why the critical parameter(s) is the best
indicator of proper control system performance and of mercury emission
rates.
2. Identification of the maximum or minimum value of each parameter
(e.g., degrees temperature, concentration of mercury) that is not to be
exceeded. The level(s) is to be directly correlated to the results of a
recent performance test when the facility was in compliance with the
emission limits of the standards.
3. Designation of the monitoring/recording frequency, with
justification if the frequency is less than hourly. The monitoring/
recording frequency must be justified on the basis of the amount of
time that could elapse during periods of process or control system
upsets before the emission limits would be exceeded, and consideration is
to be given to the time that would be necessary to repair the failure.
For example,'if emissions could exceed the limits of the standards within
8 hours of a control system failure and the repairs would typically
require 2 hours, then by-hourly monitoring may be appropriate on shifts
when repairs can be assured within two hours. If, however, maintenance
2-7
-------
and repair technicians are not available until the next shift, or the
repairs may require 6 hours to complete, then nothing short of hourly
monitoring would appear to be appropriate.
4. Designation of the immediate actions to be taken in the event
of an excursion beyond the level of the parameter identified in 3.
5. Provisions for reporting, semiannually, parameter excursions and
the corrective actions taken, and provisions for reporting, within 10
days, any significant excursion.
6. Identification of the accuracy of the monitoring device(s) or
of the readings obtained.
7. Recordkeeping requirements for certifications and calibrations.
The Agency believes that, once approved, alternative monitoring plans
that include all of the above information would be appropriate indicators
of compliance and of proper operation and maintenance and that the option to
submit alternative plans provides the flexiblity sought by the commenters.
Comment:
One commenter (IV-D-6) said that the average regeneration temperature
for a molecular sieve was not the appropriate temperature to monitor
during the performance test. He stated that a better indicator of the
effectiveness of bed regeneration would be the maximum temperature reached
during the heating cycle. He suggested that EPA establish a temperature
level that must be reached during the heating cycle.
Response:
The Administrator has accepted the commenter's statement that the average
regeneration temperature is not as good an indicator of regeneration effec-
tiveness as the maximum temperature reached during the heating cycle. Thus,
the monitoring requirement has been changed to require measuring/recording
2-8
-------
of the maximum temperature reached during desorption, i.e., the heating
cycle. After the reference temperature is recorded, the owner or operator
is required to record the maximum temperature reached during each subsequent
heating cycle. Failure to reach the reference temperature during any future
heating cycle will be considered improper operation and maintenance of the
molecular sieve.
Comment:
Commenters IV-D-5 and IV-D-7 said that the need for additional
monitoring and reporting requirements had not been demonstrated. Commenter
IV-D-5 cited a statement made in the preamble to the proposed revisions
that all mercury-cell chlor-alkali plants are presently in compliance
with the NESHAP for mercury. He also stated that the Agency did not
indicate any specific problems in maintaining compliance. This commenter
believed additional monitoring and reporting requirements are unnecessary.
Commenter IV-D-7 said that no evidence was presented to show that a reduc-
tion in emissions would result from implementing the proposed changes.
Response:
While compliance with the NESHAP has been demonstrated (by a
performance test) at least once in the past at each of the mercury-cell
chlor-alkali plants, information that the Agency received prior to proposal
of the revisions indicated that (1) some plants had experienced problems
with controlling mercury emissions from the hydrogen stream, and (2) some
facilities emit at levels just below the emission limits so that exceed-
ances of the emission limits are likely to occur during process or control
system upsets at these plants. Monitoring requirements were proposed,
in part, to prevent such exceedances. It is the Agency's belief that
careful monitoring of the process and control system(s) is necessary to
2-9
-------
alert owners/operators of problems in time to correct them before the emis-
sion limits are exceeded. In addition to assuring compliance, monitoring
requirements also serve to ensure that control systems are properly
operated and maintained on a continuous basis. (See response to earlier
comment in this section.)
The emission reduction to be gained from the addition of monitoring
requirements could not be quantified since the Agency did not have suffi-
cient information on the number of preventable exceedances that have
occurred in the past,or on the emission levels that could have been reduced
by better operation of the control system. Without monitoring and report-
ing requirements, such information is rarely available. However, since
problems with controlling mercury emissions from the hydrogen stream have
been encountered and because the Agency believes that monitoring require-
ments will result in earlier knowledge and repair of problems, the Agency
believes that there will be further emission reductions as a result of the
monitoring and reporting requirements.
Comment:
Commenter IV-D-6 asked if new values for operating parameters could
be established when changes in plant operations take place that affect
the values of the monitored parameters (provided that the new values
are established on the basis of a new performance test and that proper
notification of the test and of the intent to revise parameter values
is given).
Response:
New values for monitored parameters (either process or emission
control system parameters) may be established when changes in plant opera-
tions take place that affect the values established previously. As the
2-10
-------
commenter assumed, such new values would have to be!based on values measured
during a performance test conducted under the new operating conditions. As
with all performance tests, at least 30 days' prior notice of the test is
required.
Comment: ~
Commenter IV-D-7 questioned the usefulness to enforcement personnel
of the monitoring program being required. He commented that the proposed
control program requires deliberate detuning of emissions equipment
during testing to establish a 'worst-case baseline followed by frequent
monitoring and reporting of parameters which exceed those values
established during testing. He questioned how enforcement personnel
would use the information to determine if additional emission tests
are necessary.
Response: '
The reason for allowing plant owners or operators to conduct
performance tests under the upper range of operating conditions that
can reasonably be expected to occur under normal operating conditions
is to minimize the reporting burden while maintaining the purpose of
the monitoring requirements (which is to ensure proper operation and
maintenance of emission control systems). The standards do not require
detuning of emissions control equipment during testing; in fact, detuning
is not allowed. The test may be conducted under conditions expected to
provide the highest values of the monitored parameters (or, in the case
of chemical absorption systems, lowest values) that could be expected to
occur under conditions of proper operation and maintenance. Thus, while
a test may be conducted during the summer when ambient cooling waters
would approximate the highest temperature readings that are expected to
2-11
-------
occur, the test must still be conducted under normal operating conditions
with all control systems functioning properly.
Because the acceptable levels of monitored parameters would be
established near the highest (or, where appropriate, the lowest) value
expected to occur while the facility is in compliance, any exceedance
beyond these levels would indicate to enforcement personnel that there
are problems with the emission control systems. Based on the extent of
the excursion and its duration, enforcement personnel woul-d have an
indication of whether the emission limits are be;ing exceeded. If there
is reason to believe that the limits are being exceeded, a performance
test could be required to establish compliance status.
Comment:
Commenter IV-D-7 stated that testing to establish the proper "baseline
condition" could be costly. He believed that an owner or operator would
attempt to establish test conditions at a "barely pass" level to avoid
future reporting. In his opinion as many as 10 tests might be conducted
(at $12,000 per test) to establish the "barely pass" conditions. Further,
he suggested that the phrase "other than conditions of malfunction" be
deleted from the preamble discussion of "testing under the upper range of
operating conditions (other than conditions of malfunctions) that can
reasonably be expected to occur on a routine basis." (49 FR 50150) , He
believed that malfunction conditions are undefined, and, consequently,
insignificant deviations from normal operations may be interpreted as
malfunctions.
Response:
As stated in the response to the previous comment, the Agency allows
testing under the upper range of operating conditions that can reasonably
2-12
-------
be expected 'to occur given conditions of proper operation and maintenance.
The phrase "other than conditions of malfunctions" is intended to clarify
that the test conditions must be representative of proper operation and
maintenance.- The phrase is necessary to prevent detuning of control
systems (see previous comment) or any kind of change to the system that
is not representative of proper operation. For this reason, the phrase
is necessary and has not been deleted.
The standards do not require repetitions of the performance test to
achieve "barely pass" conditions, nor was the provision that is the subject
of the comment intended to encourage such repetitions. Only one performance
test is required under the proposed revisions. Conducting performance tests
would be costly only if the plant owner or operator elected to repeat them.
Comment:
Commenter IV-D-6 stated that it was not feasible to conduct continuous
stack testing during the entire six-hour (minimum) period of the performance
test.
Response:
The standards require the completion of one performance test to use
as a basis for establishing the limits for parameters to be monitored.
The performance test result is the average of the results of 3 individual
repetitions of the test methods. Each of these 3 sample runs should last
a minimum of 2 hours (See Section 7.1.2 of Method 101), thus yielding
a minimum total sampling time of 6 hours. However, continuous stack
testing during any consecutive 6-hour clock period is not required by
the reference methods. The three sample runs should be conducted over
a period or periods that accurately reflect the maximum emissions that
occur in a 24-hour period, but these periods need not be consecutive.
2-13
-------
The commenter was contacted to ascertain if there would be problems
conducting stack tests according to the procedure outlined above. He
indicated that the procedures, as described, would not present a problem.
2-H
-------
2.2 RECORDKEEPING REQUIREMENTS
Comment:
Several of the commenters believed that the requirement to record
all incidences of mercury leaks or spills should be changed to require
recording only incidences of unpredictable leaks or spills. Commenters
IV-D-3, IV-D-5, and IV-0-8 stated that predictable events such as those
that occur during routine maintenance should be excluded since clean-up
procedures are designated for these types of leaks or spills. These
commenters believed that only unpredictable leaks/spills that require
immediate corrective actions should be reported.
Commenter IV-D-6 suggested that criteria for determining "significant"
leaks/spills be established to define those leaks/spills that should be
recorded. Shift personnel at his company's plants do not formally document
their observations unless the leaks/spills are significant and require
immediate corrective action. No definition of "significant" was offered
by the commenter.
Response:
The EPA agrees that "significant" leaks are the type of leak/spill
that is of interest to the Agency. However, the Agency is unable to
define levels of parameters (e.g., quantity, volume, duration) that would
constitute a "significant" leak or spill. In a meeting with representatives
from several companies that own or operate mercury-cell chlor-alkali
plants (February 28, 1984; see docket entry II-D-61), EPA asked these
industry representatives for their suggestions of ways to define significant
leaks or spills. No definition of "significant" could be suggested.
2-15
-------
Instead, these industry representatives indicated that predictable leaks/spills
should be excluded from the recordkeeping requirements. However, in the
Agency's opinion, there could be predictable leaks/spills that, due to
their number or quantity, would be "significant" in that they could cause
an exceedance of the emission limit if not corrected immediately. The
EPA is unaware of any definite relationship between a leak or spill's
predictability and its significance. Also, subjective opinions about
what is predictable are likely to vary from plant to plant so that some
plants; would record certain leaks/spills while another plant would not.
i i
For these reasons, the commenters' suggestion could not be adopted.
In the February 28, 1984, meeting referred to earlier, mercury-cell
chlor-alkali plant representatives gave EPA the impression that leaks and
spills of any magnitude were infrequent occurrences if the cell room were
operated properly. However, based on the written comments received, the
Agency is concerned that leaks, labeled as predictable, may occur on a
somewhat frequent basis and that sufficient steps may not be taken to
avoid such occurrences in the future. Because of the hazardous health
effects of mercury, it seems to EPA that a plant owner/operator would
want to be informed of any mercury leaks or spills that occur and of the
corrective actions taken not only to clean up the spill but to ensure
that future leaks or spills are avoided as much as possible.
The Administrator believes it is appropriate to record all leaks or
spills on a daily basis and to record the corrective actions that were
taken. With such records available, company and enforcement personnel
can review the history of mercury leaks/spills at a specific plant and
determine if corrective actions have been timely and appropriate to avoid
2-16
-------
recurrent leaks/spills of the same type. Thus, the requirement for daily
recording of all mercury leaks/spills has not been changed.
2-17
-------
2.3 REPORTING REQUIREMENTS
Comment:
One commenter (IV-D-5) believed that reporting incidences when a
monitored parameter is not within the range established by the performance
test should be required on an annual basis rather than on a semiannual basis.
Response:
Failure to maintain process or control device- parameters within
acceptable ranges established by a site-specific performance test consti-
tutes a failure to operate and maintain equipment properly so as to prevent
exceedances of the emission limits. Enforcement personnel need to know,
as early as possible, about such failures so that they can work with
plant personnel to correct the situation as early as possible. Considering
that failure to maintain the equipment within acceptable ranges that were
established during a performance test conducted under worst-case conditions
could lead to exceedances of the emission limit, the Administrator believes
it is prudent to require, on a frequent basis, reports of periods when
the systems are not operating properly. Semiannual reporting is judged to
be reasonable for these purposes.
It should be emphasized that no reports are required at all if the
equipment is operated and maintained so as to remain within the specified
acceptable ranges.
Comment:
Two commenters (IV-0-7 and IV-D-8) stated that reporting failures to
maintain process or control system parameters within site-specific acceptable
2-18
-------
ranges for periods exceeding 24 hours in duration need not be required
within 10 days of the failure. These commenters believed that the requirement
is redundant, since parameter excursions would be reported again as part
of the semiannual reports, and creates an unnecessary paperwork burden on
the regulated facilities.
Response:
The semiannual reporting period is believed to be appropriate for
reporting infrequent or unprolonged failures to operate equipment within
those site-specific ranges established during the performance test.
However, excursions beyond these ranges that are not corrected within 24
hours could represent major equipment failures with increased risk of
exceeding the emission limit of the standard. For this reason, enforcement
personnel need to know about prolonged excursions and the corrective
actions taken to correct them as quickly as possible. Reporting such
information within 10 days is considered the maximum reasonable amount of
time to delay the report of prolonged excursions but sufficiently long
for the operator to be able to report steps that were taken to fully
correct the problem.
In this way, enforcement personnel are made aware of potentially
significant equipment failures and attempts to correct them in a timely
manner so that they can evaluate the efficacy of the corrective actions
and determine whether the potential exists for a significant exposure risk.
For these reasons the Administrator believes that prolonged excursions
need to be reported in as timely a manner as possible. The relevant
semiannual report would include a copy of the earlier report as well as
a description of any additional excursions. The paperwork burden associated
with 10-day reports is estimated to be less than 5 manhours per report.
2-19
-------
This burden is not viewed as unnecessary or unreasonable in light of the
potential emission exceedances that may be prevented as a result of
the reports. As noted previously, it should be emphasized that no 10-day
reports would be required at all if the equipment is operated and maintained
so as to remain'within the specified acceptable ranges.
Comment:
Commenter IV-D-5 said that the sentence in the preamble stating that
24 hours should be sufficient time to repair most conditions causing
parameter excursions is true for most conditions but is not true for
equipment malfunctions requiring major repair or replacement. In these
instances, repairs could require several days.
Response:
The sentence referred to by the commenter related to the consecutive
24-hour period being allowed for hourly parameter excursions before the
owner/operator is required to report the excursions within a 10-day period.
Available information indicates that the majority of control system upsets
can be repaired and the parameters returned to acceptable limits within
24-hours. By requiring reports within 10 days of only those hourly parameter
excursions that continue for extended period (24 hours or more) owners or
operators do not have to report periods of excursions that are repaired
within a few hours, bringing the parameters back within acceptable levels.
However, periods of parameter excursions that continue for prolonged periods
(24 hours or more) indicate significant control system failures that are
of interest to enforcement personnel. Therefore, the standards require
reports of such extended excursions to be filed early enough for enforce-
ment personnel to assess the corrective actions being taken. The Agency
2-20
-------
recognizes that some equipment failures may require more than 24 hours to
repair, ,and it is these failures that are of primary concern.
Comment j
Three commenters (IV-D-4, IV-D-5, and IV-D-8) asked that the Agency
describe in the standards the purpose of reports of incidences of excursions
I
of process or control system parameters and the way that enforcement personnel
would use such reports. The commenter believed such a description would help
ensure that parameter excursions would not be misconstrued as exceedances
i ' i
of the emission limits. Another commenter (IV-D-3) asked if there will be
penalties for reporting excursions of control device parameters.
Response:
Failure to operate process or control system parameters within acceptable
ranges (as determined by site-specific performance tests) constitutes a
failure to operate and maintain control systems properly. Such failures
do not constitute a violation of the emission limits; noncompliance with
the emission limits of this standard can be determined only on the basis
of a performance test. Thus, plant owners or operators cannot be cited
or otherwise penalized for noncompliance with the emission limits based
on reports of parameter excursions. However, enforcement personnel could
issue a violation based on improper operation and maintenance and could
require a performance test to establish compliance with the emission limits.
Section 61.55(b) of the regulation has been revised to clarify that
parameter excursions will be considered unacceptable operation and mainte-
nance of the facility but that such excursions cannot, by themselves, be
interpreted as exceedances of the emission limits.
2-21
-------
Comment:
Two commenters (IV-D-6 and IV-D-7) believed that the period for filing
reports of performance tests should be extended from 15 days following
receipt of the test results to 30 days (IV-D-6) or 60 days (IV-D-7). This
additional time would allow such owners or operators to verify the validity
of the test data and, if errors are suspected, to resample and retest.
Commenter IV-D-7 stated that an operator may conduct up to 10 tests in an
attempt to establish test conditions at a "barely pass" level to avoid
future reporting of parameter excursions. He believed the resulting
testing and analytical load is sufficient to delay the report of the test
results by 60 days.
Response:
Prior to proposal of revisions to the standards, the mercury NESHAP
allowed only one day, following receipt of test results, to report the
results to the Administrator. Because one day was considered insufficient
time for the source owner/operator to review the results and verify their
validity, an additional two week period was proposed. This period is believed
sufficient to review the results before forwarding them to the Administrator.
If repeated tests are subsequently conducted and the owner/operator wants
these results to supercede the first test results, then the results of
these tests would also be submitted to the Administrator within 15 days
following the date that the results are known. The extension is not
intended to provide additional time for repeating tests. All test results
should be reported, and the owner or operator is being provided only the
time believed necessary to review the results before reporting them, to
identify any errors. For, this reason, the 15-day period is not being
revised.
2-22
-------
2.4 INDIRECT EXPOSURES TO MERCURY
Comment: Commenter IV-D-10 stated that the 1.0 yg/m3 ambient air guideline
is based exclusively on the toxicity of inhaled mercury and ignores the
health effects resulting from indirect exposures to mercury. He stated
that atmospheric mercury emissions are deposited on land, water, and other
surfaces, and that the deposited mercury may be transformed into methylated
mercury, the most toxic form of mercury. The.-commenter believed that the
Agency failed to assess all means through which mercury can endanger public
health and thus failed to meet its Clean Air Act mandate to consider all
sources of exposure to the pollutant in determining health risk. The Agency
. also did not draw any conclusion as to how methylation processes increase
the health risk from emissions of metallic and inorganic mercury.
This commenter also stated that little attention was paid to background
mercury levels in humans and in the ecosystem,in evaluating the hazards
of incremental increases resulting from uncontrolled emissions sources.
The commenter believed that EPA must evaluate the risks associated with
ongoing industrial activities over and above background levels resulting
from previous activities. He believed that the marginal contribution of
some current sources may be relatively small, but for certain populations,
such as fish eaters in the Great Lakes region, the incremental exposure
could be significant.
The commenter believed that EPA must fully evaluate the risk from
indirect exposures to mercury and then re-evaluate the adequacy of the
standard for each currently regulated category as well as the Agency's
decisions not to set standards for other categories. Only after such
evaluation will EPA meet its statutory duty to review and, if necessary,
revise the mercury standard.
2-23
-------
Response:
While the ambient air guideline of 1.0 yg/m^ is based primarily on
the toxicity of inhaled mercury (Hg), the Agency did not ignore the effects
resulting from indirect exposure to mecury in developing the ambient air
guideline. To determine the ambient air level of mercury that did not ~
impair human health, the impact of the airborne burden (inhalation of
mercury vapor) was considered in conjunction with the water arid food
burdens (ingestion of methyl mercury). This analysis assumes that expo-
sures to methyl mercury (diet/ingestion) and mercury vapor (inhalation)
are equivalent and additive. The prolonged intake level of the most
hazardous form of mercury was used to develop the inhalation guideline.
This level (approximately 300 yg Hg as methyl mercury/day/70 Kg body
weight) was based on the occurrence of the first symptoms of mercury
intoxication, i.e., non-specific symptoms such as paresthesia. A safety
factor of ten (30 yg Hg/day) was added to provide protection against
genetic lesions and effects in children and the fetus. The mercury
intake level for inhalation (20 yg Hg/day/70 Kg adult) used in developing
the standard considered the impact of the average intake level from the
ingestion of mercury (10 yg Hg/day).
An average dietary intake level of 10 yg Hg/day/adult was used to
assess the impact of indirect mercury exposure from the ingestion of food.
This intake level is well above the 2.9 to 3.9 yg Hg/day average adult
intake from the ingestion of food as reported in the current Food and Drug
Administration's (FDA) Revised Total Diet Study. The current intake levels
are based on data obtained from the eight revised market baskets collected
during the period April 1982 through March 1984 (Docket item IV-D-JZ). The
FDA tests foods throughout the United States to estimate levels of mercury
2-24
-------
and other trace materials. A review of mercury dietary intake levels for
the foods tested by FDA for the period 1973 to 1982 indicate average
mercury levels in foods are not significantly increasing [ranging from 2
ug Hg/day for Fiscal Year (FY) 1974 to 7 ug Hg/day for FY 1976 and back
to 3 ug Hg/day for FY 1981/82] and are below the 10 ug Hg/day intake
level used in the NESHAP analysis to represent indirect exposures. While
the Agency does not continuously assess the background levels of mercury
in humans, the FDA does continuously monitor the mercury levels in food
through its Revised Total Diet Study to provide an indication of changes
in the dietary intake levels of the general public.
The effects of atmospheric mercury on other environments such as
drinking water and the accumulation of methyl mercury in food (primarily
fish) were not fully addressed in the NESHAP review. As a result of the
acid rain issue, the Agency is presently reviewing available information
concerning these effects, and studies are currently underway to gather
necessary data. These include studies of biochemical mechanisms (for
example", the biochemical cycling of mercury) and health and environmental
effects (for example, the bioaccumulation of methylmercury in fish) from
the deposition of mercury. A preliminary report of the results of studies
addressing the bioaccumulation of mercury in fish (the primary source of
ingested mercury) is scheduled for 1989 with an integrated report on
mercury bioaccumulation scheduled for 1992. As the results of these
studies become available, the Agency will take action as appropriate.
However, at this time, the Agency does not have a sufficient basis for
revising the 1.0 ug/m^ ambient guideline level.
2-25
-------
Comment:
Commenter IV-D-5 responded to the Agency's request (49 FR 50147) for
comments on the effects of indirect exposures to mercury by stating that
he believed that no information has been developed which indicated that
indirect exposures to mercury have a significant impact on health or on
the environment. He believed that the 1.0 ug/m^ ambient guideline is
amply protective and should not be changed.
Response:
Please see response to previous comment.
2-26
-------
2.5 UNREGULATED SOURCE CATEGORIES
2.5.1 COAL COMBUSTION
Comme nt:
Commenter IV-D-10 believed that further analysis of mercury emissions
from coal-fired power plants was necessary. He stated that the Agency must
analyze the health risk associated with deposited mercury emisions from coal
combustion on a national basis and on a regional, or even plant-specific,
basis for coals with high mercury content. The commenter referred to reports
of mercury concentrations in some American coals as high as 1.6 parts per
million (ppm), a level four times higher than the concentation that was used
in the Agency's analysis. He stated that the Agency cannot conclude that
the ambient guideline will not be exceeded until an analysis of the ambient
concentrations expected from plants burning high-mercury coals is completed.
The commenter also objected to EPA's "piecemeal" approach to regulating
toxic emissions from coal-fired boilers. By analyzing toxic components of
boiler emissions one-by-one, there is a strong bias against control since
only a fraction of the total health risk is compared with the total control
cost. The commenter believed that EPA should abandon this approach and
should require the use of particulate control techniques to capture all
toxic emissions, including mercury.
Response:
To examine the potential for mercury emissions from coal-fired power
plants to exceed the ambient air guideline, the Agency reviewed the data
on the mercury content of coals available in the United States (Docket
item IV-B-1). The highest mercury level reported for the 48 contiguous
states is 8 parts per million (ppm) for subbituminous coal (640 samples)
and 3.3 ppm for bituminous coal (3527 samples) with an average of 0.1 ppm
2-27
-------
for subbituminous coal and 0.21 ppm for bituminous coal. The worst case
estimates for a large 4000 megawatt (MW) coal-fired power plant firing
8 ppm subbituminous coal is 870 pounds of mercury per day. According to
dispersion estimates, a 4000 MW plant emitting 790 pounds of mercury per
day would cause "a maximum ground"level concentration of 1.0 ug/m^.
This indicates that in the extreme case a large coal-fired power plant
could emit mercury at levels high enough to exceed the ambient guideline.
However, typically, mercury emissions from coal-fired power plants are
expected to be well below the ambient guideline level.
The Agency is currently studying the combined effect of identified
trace element (including mercury) emissions from fossil fuel combustion.
For mercury, estimates are being made of nationwide emissions and of
maximum concentrations associated with four sectors of coal burning:
utility, industrial, commercial, and residential combustors. The results
of this study will be used to determine the need and appropriate mechanism
for regulating mercury emissions from fossil-fuel combustion.
2-28
-------
2.5.2 OIL SHALE RETORTS
Comment:
Commenter IV-D-10 stated that there are data indicating that mercury
emissions from oil shale retort operations can equal or exceed emissions from
the currently regulated source categories. He believed these data demonstrate
the need to set a national emission standard for mercury emissions from
oil shale retorting and to examine the potential for mercury emissions
from other synthetic fuel processes that are under active consideration.
Response:
At the present time there is only one retort plant in operation in
the United States that is capable of processing more than 100 tons per day
of raw shale to produce crude oil. Estimates of mercury emissions from
this operation indicate that ambient mercury levels would be less than
0.04 ug/m3, a level well below the ambient guideline level of 1.0 yg/m3
(Docket item IV-B-2).
Construction of new retort operations or startup of existing plants
that have been shut down is not anticipated in the near future. Further-
more, projections of mercury emissions from hypothetical commercial-scale
operations indicate that emissions from a large size facility would still
be below the ambient guideline level (Docket item IV-B-2).
In view of the low level of emissions from the oil shale retort that
is currently in operation and the lack of anticipated growth in this
industry in the near future, oil shale retorting operations are not being
added as a source category to be regulated by the current mercury NESHAP.
If oil shale retort operations become economically feasible, the Agency
will review its decision not to regulate mercury emissions from these
operations under these standards.
2-29
-------
2.6 MISCELLANEOUS COMMENTS
Comment:
Commenter IV-D-6 stated that his company's initial estimates of
the annual cost to comply with the proposed revisions indicate that
the Agency's estimate of $9,000 may be low.
Response:
The commenter was contacted to obtain a copy o*,r his company's estimates
for comparison purposes. The commenter indica+--ed that nis comPiny's initial
estimates indicated costs may be slightly l n1gher than EPA's.stimate (around
$10-12,000 instead of $9,000) but tt t difference was in his opinion,
not large enough to pursue compa- >•' estimates. Tte, the Agency's
number has not been revised.
2-30
------- |