National Emission Standards for Hazardous Air Pollutants From Petroleum Refineries, 55670-55692 [E9-25454]

Agencies

• ENVIRONMENTAL PROTECTION AGENCY

[Federal Register Volume 74, Number 207 (Wednesday, October 28, 2009)]
[Rules and Regulations]
[Pages 55670-55692]
From the Federal Register Online via the Government Printing Office [www.gpo.gov]
[FR Doc No: E9-25454]



[[Page 55669]]

-----------------------------------------------------------------------

Part III





Environmental Protection Agency





-----------------------------------------------------------------------



40 CFR Parts 9 and 63



National Emission Standards for Hazardous Air Pollutants From Petroleum 
Refineries; Final Rule

Federal Register / Vol. 74, No. 207 / Wednesday, October 28, 2009 / 
Rules and Regulations

[[Page 55670]]


-----------------------------------------------------------------------

ENVIRONMENTAL PROTECTION AGENCY

40 CFR Parts 9 and 63

[EPA-HQ-OAR-2003-0146; FRL-8972-4]
RIN 2060-AO55


National Emission Standards for Hazardous Air Pollutants From 
Petroleum Refineries

AGENCY: Environmental Protection Agency (EPA).

ACTION: Final rule.

-----------------------------------------------------------------------

SUMMARY: This action amends the national emission standards for 
petroleum refineries to add maximum achievable control technology 
standards for heat exchange systems. This action also amends the 
general provisions cross-reference table and corrects section 
references.

DATES: The final amendments are effective on October 28, 2009. The 
incorporation by reference of certain publications listed in the final 
rule amendments is approved by the Director of the Federal Register as 
of October 28, 2009.

ADDRESSES: The EPA has established a docket for this action under 
Docket ID No. EPA-HQ-OAR-2003-0146. All documents in the docket are 
listed in the https://www.regulations.gov index. Although listed in the 
index, some information is not publicly available, e.g., confidential 
business information or other information whose disclosure is 
restricted by statute. Certain other material, such as copyrighted 
material, is not placed on the Internet and will be publicly available 
only in hard copy. Publicly available docket materials are available 
either electronically in https://www.regulations.gov or in hard copy at 
the EPA Docket Center, Environmental Protection Agency, EPA West 
Building, Room 3334, 1301 Constitution Ave., NW., Washington, DC. The 
Public Reading Room is open from 8:30 a.m. to 4:30 p.m., Monday through 
Friday, excluding legal holidays. The telephone number for the Public 
Reading Room is (202) 566-1744, and the telephone number for the Air 
and Radiation Docket is (202) 566-1742.

FOR FURTHER INFORMATION CONTACT: Mr. Robert Lucas, Office of Air 
Quality Planning and Standards, Sector Policies and Programs Division, 
Coatings and Chemicals Group (E143-01), Environmental Protection 
Agency, Research Triangle Park, North Carolina 27711, telephone number 
(919) 541-0884; fax number (919) 541-0246; e-mail address: 
lucas.bob@epa.gov.

SUPPLEMENTARY INFORMATION: The information in this preamble is 
organized as follows:

I. General Information
    A. Does this action apply to me?
    B. Where can I get a copy of this document?
    C. Judicial Review
II. Background Information
III. Summary of the Final Amendments to NESHAP for Petroleum 
Refineries and Changes Since Proposal
    A. What requirements for heat exchange systems are we 
promulgating pursuant to CAA section 112(d)(2)?
    B. What other revisions and clarifications are we making?
    C. What is the compliance schedule for the final amendments?
IV. Summary of Comments and Responses
    A. Heat Exchange Systems
    B. General Provisions Applicability
V. Summary of Impacts
VI. Statutory and Executive Order Reviews
    A. Executive Order 12866: Regulatory Planning and Review
    B. Paperwork Reduction Act
    C. Regulatory Flexibility Act
    D. Unfunded Mandates Reform Act
    E. Executive Order 13132: Federalism
    F. Executive Order 13175: Consultation and Coordination With 
Indian Tribal Governments
    G. Executive Order 13045: Protection of Children From 
Environmental Health Risks and Safety Risks
    H. Executive Order 13211: Actions Concerning Regulations That 
Significantly Affect Energy Supply, Distribution, or Use
    I. National Technology Transfer and Advancement Act
    J. Executive Order 12898: Federal Actions To Address 
Environmental Justice in Minority Populations and Low-Income 
Populations
    K. Congressional Review Act

I. General Information

A. Does this action apply to me?

    The regulated category and entities potentially affected by this 
final action include:

------------------------------------------------------------------------
                                                   Examples of regulated
           Category               NAICS \1\ code          entities
------------------------------------------------------------------------
Industry......................  324110...........  Petroleum refineries
                                                    located at a major
                                                    source that are
                                                    subject to 40 CFR
                                                    part 63, subpart CC.
------------------------------------------------------------------------
\1\ North American Industry Classification System.

    This table is not intended to be exhaustive, but rather provides a 
guide for readers regarding entities likely to be regulated by this 
final rule. To determine whether your facility is regulated by this 
action, you should carefully examine the applicability criteria in 40 
CFR 63.640 of subpart CC (National Emission Standards for Hazardous Air 
Pollutants From Petroleum Refineries). If you have any questions 
regarding the applicability of this action to a particular entity, 
contact either the air permit authority for the entity or your EPA 
regional representative as listed in 40 CFR 63.13 of subpart A (General 
Provisions).

B. Where can I get a copy of this document?

    In addition to being available in the docket, an electronic copy of 
this final action will also be available on the Worldwide Web through 
the Technology Transfer Network (TTN). Following signature, a copy of 
this final action will be posted on the TTN's policy and guidance page 
for newly proposed or promulgated rules at the following address: 
https://www.epa.gov/ttn/oarpg/. The TTN provides information and 
technology exchange in various areas of air pollution control.

C. Judicial Review

    Under section 307(b)(1) of the Clean Air Act (CAA), judicial review 
of this final rule is available only by filing a petition for review in 
the United States Court of Appeals for the District of Columbia Circuit 
by December 28, 2009. Under section 307(d)(7)(B) of the CAA, only an 
objection to these final rules that was raised with reasonable 
specificity during the period for public comment can be raised during 
judicial review. Moreover, under section 307(b)(2) of the CAA, the 
requirements established by these final rules may not be challenged 
separately in any civil or criminal proceedings brought by EPA to 
enforce these requirements.
    Section 307(d)(7)(B) of the CAA also provides a mechanism for us to 
convene a proceeding for reconsideration, ``[i]f the person raising an 
objection can demonstrate to the EPA that it was impracticable to raise 
such objection within [the period for public comment] or if the grounds 
for such objection

[[Page 55671]]

arose after the period for public comment (but within the time 
specified for judicial review) and if such objection is of central 
relevance to the outcome of the rule.'' Any person seeking to make such 
a demonstration to us should submit a Petition for Reconsideration to 
the Office of the Administrator, Environmental Protection Agency, Room 
3000, Ariel Rios Building, 1200 Pennsylvania Ave., NW., Washington, DC 
20460, with a copy to the person listed in the preceding FOR FURTHER 
INFORMATION CONTACT section, and the Associate General Counsel for the 
Air and Radiation Law Office, Office of General Counsel (Mail Code 
2344A), Environmental Protection Agency, 1200 Pennsylvania Ave., NW., 
Washington, DC 20460.

II. Background Information

    Section 112 of the CAA establishes a regulatory process to address 
emissions of hazardous air pollutants (HAP) from stationary sources. 
After EPA has identified categories of sources emitting one or more of 
the HAP listed in section 112(b) of the CAA, section 112(d) calls for 
us to promulgate national emission standards for hazardous air 
pollutants (NESHAP) for those sources. For ``major sources'' that emit 
or have the potential to emit any single HAP at a rate of 10 tons or 
more per year or any combination of HAP at a rate of 25 tons or more 
per year, these technology-based standards must reflect the maximum 
reductions of HAP achievable (after considering cost, energy 
requirements, and non-air quality health and environmental impacts) and 
are commonly referred to as maximum achievable control technology 
(MACT) standards.
    For MACT standards, the statute specifies certain minimum 
stringency requirements, which are referred to as floor requirements. 
See CAA section 112(d)(3). Specifically, for new sources, the MACT 
floor cannot be less stringent than the emission control that is 
achieved in practice by the best-controlled similar source. The MACT 
standards for existing sources can be less stringent than standards for 
new sources, but they cannot be less stringent than the average 
emission limitation achieved by the best-performing 12 percent of 
existing sources in the category or subcategory (or the best-performing 
five sources for categories or subcategories with fewer than 30 
sources). In developing MACT, we must also consider control options 
that are more stringent than the floor. We may establish standards more 
stringent than the floor based on the consideration of the cost of 
achieving the emissions reductions, any non-air quality health and 
environmental impacts, and energy requirements.
    We published the final MACT standards for petroleum refineries (40 
CFR part 63, subpart CC) on August 18, 1995 (60 FR 43620). These 
standards are commonly referred to as the ``Refinery MACT 1'' standards 
because certain process vents were excluded from this source category 
and subsequently regulated under a second MACT standard specific to 
these petroleum refinery process vents (40 CFR part 63, subpart UUU, 
referred to as ``Refinery MACT 2'').
    In developing this rule, we first issued an advanced notice of 
proposed rulemaking (ANPR) on March 29, 2007. The purpose of the ANPR, 
which covered the sources subject to the Refinery MACT 1 rule and other 
source categories, was to solicit additional emissions data and any 
corrections to the data we already had. We issued an initial proposed 
rule for the petroleum refineries subject to the Refinery MACT 1 on 
September 4, 2007, and held a public hearing in Houston, Texas, on 
November 27, 2007. In response to public comments on the initial 
proposal, we collected additional information and revised our analysis 
of the MACT floor. Based on the results of these additional analyses, 
we issued a supplemental proposal on November 10, 2008, that 
established a new MACT floor for heat exchange systems. A public 
hearing for the supplemental proposal was held in Research Triangle 
Park, North Carolina, on November 25, 2008. We are now taking final 
action to establish standards for heat exchange systems in the Refinery 
MACT 1 standards (40 CFR part 63, subpart CC) and to update and amend 
Table 6 to 40 CFR part 63, subpart CC.\1\
---------------------------------------------------------------------------

    \1\ We were also required by a Consent Decree to consider and 
address the application of the NESHAP General Provisions in 40 CFR 
part 63, subpart A to the existing Refinery MACT 1 rule (subpart 
CC).
---------------------------------------------------------------------------

III. Summary of Final Amendments to NESHAP for Petroleum Refineries and 
Changes Since Proposal

A. What requirements for heat exchange systems are we promulgating 
pursuant to CAA section 112(d)(2)?

    On September 4, 2007, we proposed, under CAA section 112(d)(2), two 
options for work practice standards for cooling towers: Option 1 was 
proposed based on our initial assessment of the MACT floor and Option 2 
was a beyond-the-floor option. These options would require the owner or 
operator of a new or existing source to monitor for leaks in the 
cooling tower return lines from heat exchangers in organic HAP service 
(i.e., lines that contain or contact fluids with 5 percent by weight or 
greater of total organic HAP listed in Table 1 of the rule) and, where 
leaks are detected, to repair such leaks within a specified period of 
time.
    On November 10, 2008, we issued a supplemental proposal that 
significantly modified the proposed monitoring methods, leak 
definitions, and corrective action timeframe based on a revised MACT 
floor and beyond-the-floor analysis. In the supplemental proposal, we 
also redefined the requirements in terms of heat exchange systems to 
include the heat exchangers, for which corrective actions are targeted, 
as part of the source and to specifically address once-through cooling 
systems.
    After considering public comments, for purposes of establishing 
MACT under CAA section 112(d)(2), we have selected the MACT floor 
requirements specified in the supplemental proposal for heat exchange 
systems in organic HAP service at petroleum refineries. We rejected the 
beyond-the-floor option because it is not cost-effective.
    Under these selected requirements, owners and operators of heat 
exchange systems that are in organic HAP service at new and existing 
sources are required to conduct monthly sampling and analyses using the 
Texas Commission on Environmental Quality's (TCEQ) Modified El Paso 
Method, Revision Number One, dated January 2003.\2\ For existing 
sources, a leak is defined as 6.2 parts per million by volume (ppmv) 
total strippable volatile organic compounds (VOC) in the stripping gas 
collected via the Modified El Paso Method. For new sources, a leak is 
defined as 3.1 ppmv total strippable VOC collected via the Modified El 
Paso Method. The amendments require the repair of leaks in heat 
exchangers in organic HAP service within 45 days of the sampling event 
in which the leak is detected, unless a delay in repair is allowed. 
Delay in repair of the leak is allowed until the next shutdown if the 
repair of the leak requires the process unit served by the leaking heat 
exchanger to be shut down and the total strippable VOC concentration is 
less than 62 ppmv. Delay in repair of the leak is also allowed for up 
to 120 days

[[Page 55672]]

if the total strippable VOC concentration is less than 62 ppmv and if 
critical parts or personnel are not available. The owner or operator is 
required to continue monthly monitoring and to repair the heat 
exchanger within 30 days if sampling results show that the leak exceeds 
62 ppmv total strippable VOC.
---------------------------------------------------------------------------

    \2\ ``Air Stripping Method (Modified El Paso Method) for 
Determination of Volatile Organic Compound Emissions from Water 
Sources,'' Revision Number One, dated January 2003, Sampling 
Procedures Manual, Appendix P: Cooling Tower Monitoring, prepared by 
Texas Commission on Environmental Quality, January 31, 2003 
(incorporated by reference--see Sec.  63.14).
---------------------------------------------------------------------------

    Sampling for leaks can be done for individual or combined heat 
exchangers. For heat exchange systems including a cooling tower, 
sampling can be conducted at the combined cooling tower inlet water 
location. Similarly, for once-through heat exchange systems, the 
sampling can be conducted after the heat exchanger water is combined 
and prior to discharge where it will be open to atmosphere. For both 
cooling tower and once-through heat exchange systems, sampling can be 
conducted at individual heat exchangers in the return or ``exit'' lines 
(i.e., water lines returning the water from the heat exchangers to the 
cooling tower or to the discharge point). That is, if the cooling tower 
or once-through system services multiple heat exchangers, the owner or 
operator may elect to monitor only the heat exchangers ``in organic HAP 
service'' or monitor at branch points that combine several heat 
exchanger exit lines, or monitor at the combined stream for the entire 
system. If a leak is detected (the measured VOC concentration exceeds 
the applicable leak definition) at the combined cooling tower inlet or 
once-through system, the owner or operator may either fix the leak 
(reduce the VOC concentration to less than the applicable leak 
definition) or sample heat exchanger exit lines for combinations of 
heat exchanger exit lines or sample each heat exchanger ``in organic 
HAP service'' as necessary to document that the leak is not originating 
from a heat exchanger ``in organic HAP service.'' If a leak is detected 
in an individual heat exchanger ``in organic HAP service,'' that leak 
must be repaired.
    All new or existing refineries with a heat exchange system ``in 
organic HAP service'' are required to maintain records of all heat 
exchangers and which of those heat exchangers are in organic HAP 
service, the cooling towers and once-through systems associated with 
heat exchangers in organic HAP service, monthly monitoring results, and 
information for any delays in repair of a leak.
    These requirements will apply to sources on a continuous basis, 
including periods of startup, shutdown, and malfunction (SSM). As 
provided in the response to comments below, properly operating heat 
exchangers will not leak HAP into the cooling water, so HAP will not be 
emitted from the cooling tower or once-through discharges. It is only 
when they malfunction (i.e., there are leaks) that there may be HAP 
emissions. The MACT standard for heat exchange units addresses these 
emissions. Furthermore, there are no HAP emissions associated with 
start-up and shutdown.
    The requirements outlined above are based on the MACT floor 
determination. We evaluated the following beyond-the-floor options: 
having a leak definition of 3.1 ppmv for existing sources (beyond-the-
floor option for existing sources) and requiring continuous monitoring 
(beyond-the-floor options for both new and existing sources). As 
described in our supplemental proposal, we determined that these 
beyond-the-floor options were not cost-effective and concluded that 
MACT was the floor level of control.
    The final MACT requirements for heat exchange systems will reduce 
HAP emissions by 630 tons per year (ton/yr). The final requirements for 
heat exchange systems will also reduce VOC emissions by 4,100 ton/yr. 
Reducing VOC emissions may provide the added benefit of reducing 
ambient concentrations of ozone and may reduce fine particulate matter. 
The annualized nationwide cost impacts of these final standards for 
heat exchange systems are estimated to be $3.0 million. Our economic 
analysis indicates that this cost will have little impact on the price 
and output of petroleum products.

B. What other revisions and clarifications are we making?

    As proposed, we are amending 40 CFR 63.650(a) of subpart CC to 
replace ``gasoline loading racks'' with ``Group 1 gasoline loading 
racks'' to clarify the applicability of the requirements. Furthermore, 
as we proposed on November 10, 2008, we are also finalizing proposed 
amendments to the cross-references to subparts R and Y of 40 CFR part 
63 in the rule text and in Tables 4 and 5 of subpart CC because 
subparts R and Y were amended and the revised cross-references clarify 
the requirements of subpart CC.
    We are finalizing amendments to Table 6 to 40 CFR part 63, subpart 
CC (General Provisions Applicability to Subpart CC) to bring the table 
up-to-date with requirements of the General Provisions that have been 
amended since this table was created, to correct cross-references, and 
to incorporate additional sections of the General Provisions that are 
necessary to implement other subparts that are cross-referenced by this 
rule. With respect to the exemption from emission standards during 
periods of SSM in the General Provisions (see, e.g., 40 CFR 63.6(f) and 
(h)), we note that on December 19, 2008, in a decision addressing a 
challenge to the 2002, 2004, and 2006 amendments to those provisions, 
the Court of Appeals for the District of Columbia Circuit vacated the 
SSM exemption. Sierra Club v. EPA (D.C. Cir. No. 02-1135).
    The CAA section 112(d)(2) and (3) MACT standard we are promulgating 
today for heat exchange systems is not implicated by that decision 
because it does not rely on or reference the provisions of the vacated 
rule and because the MACT standard applies at all times. We are 
amending Table 6 to clarify that the MACT standard for heat exchange 
systems applies at all times.
    We are still evaluating the recent court decision. At this time, we 
are not making any additional changes to Table 6 with respect to the 
SSM provisions in 40 CFR 63.6(f)(1) and (h)(1). We have completed our 
initial assessment of the General Provisions and their application to 
subpart CC of part 63. The recent court decision requires further 
analysis, and we are currently evaluating how to address SSM events for 
Refinery MACT 1 sources in light of the court decision.
    We are also finalizing amendments to Table 1 and Table 7 to delete 
methyl ethyl ketone (also known as 2-butanone) from the HAP listed in 
those tables because methyl ethyl ketone has been delisted as a HAP. We 
are finalizing amendments to clarify the applicability sections by 
changing general references to ``the promulgation date'' to specify the 
actual promulgation date of the original subpart CC of part 63. 
Finally, we are also finalizing amendments to clarify how owners and 
operators should comply with overlapping standards for equipment leaks.

C. What is the compliance schedule for the final amendments?

    The final amendments to the Refinery MACT 1 rule will be effective 
on October 28, 2009. Under section 112(i)(1) of the CAA, any new 
facility must comply upon startup or on the effective date of the rule, 
whichever is later. For purposes of determining compliance with these 
amendments, a new source is a source that commenced construction or 
reconstruction after September 4, 2007 (the initial date of proposal 
for these regulations). Consistent with the requirements of CAA section 
112(1)(3), the owner or operator of an existing source (including an 
existing source for these amendments that is currently subject to 1995 
Refinery MACT 1 standards for new sources) must comply with the heat 
exchange system requirements no later than

[[Page 55673]]

October 29, 2012. The basis for the 3-year compliance period is set 
forth below in our responses to comment.

IV. Summary of Comments and Responses

    This preamble and the document ``National Emission Standards for 
Hazardous Air Pollutants from Petroleum Refineries: Background 
Information for Final Standards for Heat Exchange Systems--Summary of 
Public Comments and Responses'' (``Response to Comments'') located in 
the docket (Docket ID No. EPA-HQ-OAR-2003-0146) include only comment 
summaries and responses to issues related to heat exchange systems and 
other clarifying amendments. The major comments on those issues and our 
responses are summarized in the following sections. A summary of the 
remainder of the comments and responses related to those issues can be 
found in the Response to Comments document.
    Comments regarding other issues raised as a result of the proposed 
and supplemental proposed rules are not included in this preamble or 
the Response to Comments document; they will be addressed, as 
appropriate, in future rulemakings addressing the residual risk and 
technology reviews for Refinery MACT 1.

A. Heat Exchange Systems

    On November 10, 2008, we issued a supplemental proposal with our 
revised MACT floor and beyond-the-floor analysis. In general, the 
comments received on the cooling tower requirements initially proposed 
on September 4, 2007, either have been addressed through the 
supplemental proposal or are not applicable to the final standards 
(e.g., clarifications to monitoring methods no longer required). Any 
general comments regarding cooling tower requirements received on the 
initial proposal that are still applicable are summarized in the 
Response to Comments document located in the docket (Docket ID No. EPA-
HQ-OAR-2003-0146). Significant comments received on the supplemental 
proposal are addressed in this section.
1. MACT Floor for Heat Exchange Systems
    Comment: A few commenters noted that the leak definition proposed 
for new heat exchange systems of 3.1 ppmv has not been ``demonstrated 
in practice.'' One commenter stated that the leak definition of 3.1 
ppmv was developed by the State of Texas from the AP-42 emission 
factor. The commenter stated that only one cooling tower is operating 
under a permit with that limit (the other cooling towers are under 
construction), and this cooling tower has only recently begun 
operating, so there is no significant experience operating with the 
identified new source limit or applying it to the range of operations 
and ages of exchangers in a typical refinery. The commenter asserted 
that some heat exchangers and heat exchange systems are difficult to 
control, and different leak definitions are appropriate for different 
situations within an individual refinery, so a set of requirements must 
be demonstrated to be workable on multiple heat exchange systems of 
varying services and ages before that set of requirements can be 
considered ``demonstrated in practice.'' Another commenter stated that 
there is no demonstration that there is technology that can be applied 
to new sources that improves the emission performance of these systems 
when considered across the operating life of the facilities. Both 
commenters recommended setting the new source and existing source 
requirements equivalent at 6.2 ppmv. (One of the commenters noted that 
EPA's analysis shows that the next best controlled source has a limit 
of 5 ppmv, but the commenter noted that there is not much difference 
between the reductions achieved by a leak definition of 5 ppmv and a 
leak definition of 6.2, and 5 ppmv is not cost-effective. The commenter 
urged EPA to review cooling towers and heat exchange systems under CAA 
sections 112(d)(6) and 112(f)(2) and consider factors such as cost 
rather than developing a standard under CAA section 112(d)(2).)
    One commenter noted that in the State of Texas, if a particular 
cooling tower cannot meet its normal leak definition of 80 parts per 
billion by weight (ppbw) VOC in the water, the State allows that source 
to set a leak definition of up to 150 ppbw VOC in the water. For 
flexibility when dealing with continuous small seepage or situations 
where the particular HAP or VOC present are not completely stripped by 
the cooling tower, the commenter suggested that in any 1-year period, 
if monitoring shows three leaks above 6.2 ppmv, but below 12 ppmv, EPA 
should allow that source to set a new leak definition of 12 ppmv.
    Commenters stated that the leak definition of 6.2 ppmv VOC in the 
stripping gas is not stringent enough. One commenter noted that during 
cooling tower leak investigations conducted by the City of Houston and 
TCEQ, a potential leak measured at 2 ppm required sampling by summa 
canister to confirm the leak, and EPA's regulation should be at least 
that stringent. The commenter stated that a stringent leak threshold of 
2 ppm will ensure that small leaks are found and repaired quickly, 
especially since the TCEQ leak threshold is 50 parts per billion by 
volume (ppbv).
    Several commenters supported using the Modified El Paso Method to 
detect leaks but suggested that cooling towers that have higher 
recirculation flow rates should have lower leak definitions than 
cooling towers with lower flows because the large cooling towers will 
have higher mass emissions at the same leak concentration.
    Commenters stated that EPA failed to consider the TCEQ Highly 
Reactive VOC (HRVOC) rule in establishing the MACT floor. The 
commenters believe the HRVOC rule is applicable to several refinery 
cooling towers, requires continuous monitoring, and it has a more 
stringent leak definition and leak repair schedule. One commenter also 
cited a California refinery that is required to install and operate a 
continuous hydrocarbon analyzer and repair leaks above an agreed 
threshold.
    Response: The TCEQ El Paso Method has been demonstrated at numerous 
refineries and other similar sources as an effective means of 
identifying leaks in heat exchange systems. The method has been used 
extensively for over 20 years. As suggested by some commenters, the 
detection limit of the El Paso Method is generally less than 2 ppmv, so 
leaks of 3.1 ppmv are quantifiable. Ongoing monitoring at refineries 
indicates that, when no leaks are present or after repairs are made, El 
Paso monitoring is able to detect leaks well below this leak threshold. 
As such, the monitoring method and the corrective action measures have 
been adequately demonstrated.
    In criticizing our new source leak definition of 3.1 ppmv, the 
commenter recognizes that heat exchangers connected to one refinery 
cooling tower are subject to a monitoring program with a leak 
definition of 3.1 ppmv. Section 112(d)(3) of the CAA provides that new 
source MACT cannot be less stringent than ``the emission control that 
is achieved in practice by the best controlled similar source.'' The 
commenter's concern that the facility has only recently begun operation 
and that there is not ``significant'' experience with the leak 
definition of 3.1 ppmv does not change the fact that this level is 
being achieved in practice and thus is the appropriate new source MACT 
floor. To the extent that the commenter suggests that the cooling 
towers meeting this limit are different and thus is presumably arguing 
that they must be subcategorized, the

[[Page 55674]]

commenter failed to submit any data supporting such a claim. As one 
commenter suggested, we cannot set the new source limit at 6.2 ppmv 
because we are establishing these requirements under CAA section 
112(d)(2), and we cannot consider cost in setting the MACT floor. The 
requirements for heat exchange systems are appropriately developed 
under CAA section 112(d)(2) because a MACT standard had not been 
previously developed for this emissions source.
    One commenter noted that the TCEQ allows some discretion in setting 
the total strippable VOC concentration limit or altering the limit 
based on the performance history of the cooling tower. We do recognize 
that the cooling tower leak definitions for total strippable VOC 
required in Texas refinery permits varied from 40 ppbw (or 3.1 ppmv) to 
280 ppbw (22 ppmv), including within this range leak definitions at 60 
ppbw, 80 ppbw, 150 ppbw, and 180 ppbw, but the 6th percentile facility 
had a leak definition of 80 ppbw, or 6.2 ppmv total strippable organics 
as methane. While some permits issued by TCEQ contain language that 
allows an alteration request or a permit amendment application, as the 
commenter noted, the permit issued for the 6th percentile cooling tower 
did not include this type of permit condition. As we cannot establish a 
requirement less stringent than the MACT floor, we do not provide a 12 
ppmv leak definition under any circumstances.
    Most of the commenters requesting lower leak definitions appear to 
misunderstand the stringency of the requirements for heat exchange 
systems included in the supplemental proposal. Based on the liquid and 
air flow rates specified in the TCEQ El Paso Method, and with the VOC 
measurements made as methane as required in the State permits and the 
supplemental proposal, a 3.1 ppmv VOC concentration in the gas stream 
from the El Paso stripping column is equivalent to 40 ppbw of 
strippable VOC (as methane) in the cooling water. The 6.2 ppmv leak 
threshold translates to a strippable VOC (as methane) in the cooling 
water of 80 ppbw.
    The TCEQ HRVOC rule sets an action level that is 50 ppbw in the 
cooling water, not 50 ppbv in the stripping air as the commenter 
suggested. As such, the TCEQ HRVOC rule action level is actually 
slightly less stringent than the leak definition in the new source MACT 
requirements. Furthermore, the 50 ppbw threshold only triggers 
calculations of emissions, and not necessarily corrective action. 
Therefore, we disagree with commenters that suggest the HRVOC rule 
requirements are more stringent than the new or existing MACT floor 
requirements we established.
    In our supplemental proposal, we specifically looked at lowering 
the leak definition for existing sources from 6.2 ppmv to 3.1 ppmv as 
part of our beyond-the-floor analysis, and determined that this was not 
cost-effective. Incrementally reducing the leak definition to 2 ppmv 
would be even less cost-effective than the option we evaluated. 
Furthermore, it would result in negligible additional emissions 
reductions, and it is very near the limit of detection of the El Paso 
Method. Therefore, we reject the option of setting the leak definition 
at 2 ppmv for new or existing sources because it is not cost-effective.
    The commenter requesting different leak definitions for different-
sized cooling towers is essentially asking for less control for small 
cooling towers (i.e., an effective leak definition greater than 6.2 
ppmv) and more control for larger cooling towers (i.e., an effective 
leak definition less than 6.2 ppmv, and in some cases less than 3.1 
ppmv). In our review of permits, we found no basis for subcategorizing 
the cooling towers by different recirculation rates. In addition, the 
suggested approach is inconsistent with the MACT floor requirements we 
identified for heat exchange systems.
    We also disagree with the comments that claim we did not consider 
the HRVOC rule in our decision-making process. We found that most 
cooling towers that are subject to the HRVOC rule are associated with 
ethylene production units, and not refinery process units. As we 
specifically collected recent permit requirements for Texas refineries, 
to the extent there might be refinery cooling towers subject to the 
HRVOC rule, those requirements were considered in the development of 
the MACT floor. As explained above, we also disagree with the 
commenter's characterization of the stringency of the HRVOC rule in 
comparison with the new and existing MACT floors.
    Our analysis indicated that repair provisions were more important 
in reducing heat exchange system emissions than using continuous 
monitoring. Contrary to the commenter's supposition, there are no 
repair schedules within the HRVOC cooling tower requirements. The 
commenter actually referenced the repair provisions for fugitive 
process equipment leaks (valves and pumps), which are not applicable to 
cooling towers. In the HRVOC rule, the action level is not a leak 
definition; rather, the leak definition is used to trigger more 
frequent monitoring for emission estimation and not specific repair 
requirements. In the HRVOC rules, facilities with cooling towers must 
meet an annual and an hourly site-wide HRVOC emissions cap. The hourly 
cap is quite high, and would not require any heat exchanger leaks to be 
repaired; the annual cap would tend to drive heat exchanger repairs. A 
medium-sized 30,000 gallon per minute cooling tower with a leak of 
1,000 ppbw total VOC containing 20 percent HRVOC (as defined in the 
Texas rule) would have to repair within 45 days under the MACT floor 
requirements of this rule, but would not necessarily have to repair in 
45 days to comply with the HRVOC rule, which sets a site-wide cap of 10 
ton/yr (45 days of emissions would release 1.6 tons of HRVOC, under 
this scenario).
    While different scenarios can be devised, the stringency of the 
Texas HRVOC rule is not as easy to categorize as the commenters 
suggest, and it could result in less emission reductions than the 
proposed new or existing source MACT floors.
    Contrary to the commenter's assertion, we also reviewed and 
evaluated the permit requirements for the cited California refinery, 
and the permit was included in the docket. The permit, dated April 17, 
2008, included a provision for a continuous monitor to be installed at 
a future date, to be determined, and the planned monitor was not being 
used at the time of our review. Additionally, based on the cooling 
tower's recirculation rate and the permitted VOC daily emission rate, 
the apparent action level (also not yet determined) is likely to be 
much higher than the leak definition for existing source MACT floors. 
In the cooling tower memorandum, we only summarized the information 
from the top-ranked cooling towers; the cooling tower at this 
California refinery was not included in the memorandum because, based 
on actual permit conditions, this cooling tower is not among the top-
performing 12 percent of cooling towers.
    While continuous monitoring was not used by the top-performing 
cooling towers, and, therefore, is not part of the floor requirements, 
we did evaluate requiring continuous monitoring in our beyond-the-floor 
analysis. However, the cost-effectiveness of this option exceeded half 
a million dollars per ton of HAP reduced, and, therefore, we did not 
require continuous monitoring as the standard. Rather, we adopted the 
floor as the MACT standard.

[[Page 55675]]

    Comment: One commenter noted that the proposed recordkeeping and 
reporting requirements for heat exchange systems are unnecessarily 
burdensome, go far beyond the requirements for the MACT floor, and 
should be revised. For the Notice of Compliance Status, the commenter 
noted that ``heat exchange systems'' are an artifact of the regulation, 
do not normally have specific names, and will change from time to time, 
so the requirement to identify the heat exchange systems that are 
subject to the requirements of this subpart should be changed to a list 
of cooling towers that serve any heat exchange system or systems in 
organic HAP service. For periodic reports, the commenter stated that: 
(1) The number of heat exchange systems in HAP service will change over 
time, so the requirement to report that number should be deleted; (2) 
the requirement to report the number of heat exchange systems in HAP 
service found to be leaking should be changed to a request to identify 
exchangers found to be leaking; (3) the requirement to report the 
number of leaks in Sec.  63.655(g)(9)(iii) duplicates the requirement 
in Sec.  63.655(g)(9)(ii); (4) Sec.  63.655(g)(9)(iii) should not 
require the reporting of measurements below the leak definition and 
should only ask for a summary of the leaks identified during the 
reporting period; (5) each 6-month period will include a lot of leaks, 
so there is no need to report the date of every leak (a record should 
be sufficient); (6) Sec.  63.655(g)(9)(v) should be revised to reflect 
all delays and to address situations when a leak is detected in one 
reporting period and repaired in the next; and (7) reporting the 
estimate of VOC emissions for delay of repair should only be required 
when the delay of repair option was invoked. For recordkeeping, the 
commenter stated that: (1) Calculating the requested information for 
each heat exchanger in a refinery will take an estimated 40 hours per 
refinery and must be repeated every year; these burdens were not 
included in the information collection request (ICR) burden estimate 
and do not add value for exchangers that will not be monitored due to 
low HAP content, that do not contact HAP, or would not leak into the 
cooling water; (2) although sources will need a record of which heat 
exchange systems include exchangers in organic HAP service to comply 
with the monitoring requirements, identification of all heat exchangers 
is not necessary; and (3) the information requested in Sec.  
63.655(i)(4)(iii)(E) is sometimes available for whole cooling towers 
but not readily available for heat exchange exit lines or cooling tower 
return lines. The commenter stated that temporary heat exchangers and 
sample coolers should be excluded from these recordkeeping and 
reporting requirements.
    Response: We reviewed the recordkeeping and reporting requirements 
identified by the commenter. We do not see how the heat exchange system 
will be as variable as the commenter suggested. We have revised the 
definition of heat exchange system to clarify our intent. We also: (1) 
Amended Sec.  63.655(g)(9)(v) to more clearly indicate that all delayed 
repairs must be included and that delays may occur across reporting 
periods; (2) amended the reporting requirements in Sec.  
63.655(g)(9)(vi) to clarify that leak emission estimates are only 
required for an actual delay of repair; and (3) clarified in Sec.  
63.655(g)(9)(vi) that the flow rate is for the location where the 
monitoring occurs. It is anticipated that facilities will monitor at 
locations where the flow rate is known based on pump curves, heat 
balance calculations, or other engineering methods. A continuous flow 
monitor is not required, but a flow rate at the monitoring location is 
needed to assess the potential mass emissions associated with a leak. 
For the other comments, we find that the recordkeeping and reporting 
requirements are needed to document compliance with the rule. 
Specifically, identifying heat exchangers and heat exchange systems 
that are in organic HAP service, maintaining monitoring results, and 
reporting the date a leak is identified and repaired is essential for 
demonstrating compliance with the monitoring requirements.
2. Applicability Issues
    Comment: One commenter supported changing the affected source from 
``cooling towers'' to ``heat exchange systems,'' noting that it allows 
the facilities flexibility in monthly monitoring, leak tracking, and 
determining best sampling locations. Other commenters stated that 
Refinery MACT 1 should only apply to heat exchange systems that are 
part of cooling tower systems and should not apply to once-through 
cooling water systems. The commenters suggested that the supporting 
documentation indicates that only cooling tower heat exchange systems 
were evaluated, and, if EPA wants to finalize requirements for once-
through cooling water systems, the requirements must be properly 
evaluated and the analyses provided for comment. One commenter stated 
that the emissions from once-through cooling systems are fundamentally 
different than systems with cooling towers since once-through systems 
do not have the air contact and stripping properties of cooling towers, 
and, as a result, a cost analysis of the two systems would show 
considerably different costs. The commenter also noted that the 
monitoring and repair techniques employed for the once-through systems 
are different than the monitoring for cooling tower systems, and these 
techniques should be evaluated for best demonstrated control technology 
(BDT) if once-through cooling systems are included in the rule. One 
commenter noted that, as proposed, the heat exchange system 
requirements apply to systems where the pressure gradient would not 
allow leakage into the cooling water. The commenter noted that these 
systems do not need monitoring, and a pressure gradient threshold of 35 
kilopascals (kPa) should be included in the definition of ``heat 
exchange system'' to exempt these types of systems from Refinery MACT 
1. Finally, the commenter stated that including the term ``cooling 
tower'' in the definition of ``heat exchange system'' could lead to 
confusion over the monitoring location requirements.
    Response: EPA has developed MACT standards, such as the Hazardous 
Organic NESHAP (HON) and Ethylene MACT, for heat exchange systems, and 
these standards include once-through cooling water systems. Generally, 
the HON and Ethylene MACT standards allow alternative surrogate means 
of compliance that are equivalent to those standards. We considered and 
rejected these alternatives in the development of the requirements that 
we proposed for heat exchange systems and that we are now finalizing 
because the HON and Ethylene MACT standards are less stringent than our 
floor. We are not aware of any means of surrogate monitoring that would 
achieve identification of leaks equivalent to the floor level of 
monitoring required for refinery heat exchange systems.
    We believe that control of once-through heat exchanger cooling 
systems is appropriate for several reasons, as outlined below. First, 
emissions of volatile HAP such as benzene occur readily from open water 
sources, which is why the Benzene Waste Operations NESHAP and the 
Refinery MACT 1 wastewater provisions require wastewater streams with 
benzene (as a surrogate for volatile HAP) to be covered and controlled 
until an appropriate treatment process is used to recover or destroy 
the benzene. While the stripping process may not be as fast as in a 
cooling tower, the once-through cooling

[[Page 55676]]

water will have a much longer exposure to the atmosphere than a system 
with a cooling tower. Thus, while the emissions may occur over a longer 
time period (over a larger area), all available scientific evidence and 
fate modeling studies of open water systems leads us to conclude that 
essentially all volatile HAP will be released into the atmosphere. As 
such, we see no reason why HAP leaks from heat exchange systems into 
once-through cooling water should be treated any differently than HAP 
leaks from heat exchange systems that have cooling towers.
    Second, in conducting the MACT floor analysis for heat exchange 
systems presented in the supplemental proposal, we assumed that once-
through cooling waters were included and that emissions from the once-
through systems would be similar to those with recirculation of cooling 
waters. In reviewing the permits that formed the basis of the MACT 
floor analysis, we found that the majority did not indicate whether the 
system was once-through or recirculating. However, we note that some 
permits included text for monitoring of ``cooling towers'' and 
``cooling tower water'' and some specified monitoring for ``heat 
exchanger system cooling water.'' The latter permits would appear to 
include once-through systems. Based on review of multiple references, 
the use of once-through cooling water in the petroleum refinery 
industry has been declining over the last 40 years, and is now a very 
small subset of the heat exchanger water systems. One reference 
indicated that a sample of facilities surveyed back in 1967 showed that 
only 5 percent of petroleum refineries were still using once-through 
cooling.\3\ No more recent data could be found on how many refineries 
use once-through systems. A more recent study on once-through cooling 
systems for cogeneration facilities indicated that approximately 11 
percent of non-utility plants that cogenerated power use once-through 
cooling; the 123 non-utility facilities included pulp and paper, 
chemical, iron and steel, aluminum, and petroleum refining 
industries.\4\ Of the 123 facilities in the survey, four were confirmed 
petroleum refineries and three of these four sources provided a 
response to the survey. None of the three reported that once-through 
cooling systems were used.
---------------------------------------------------------------------------

    \3\ Gibbons, DC. The Economic Value of Water. Published by 
Resources for the Future. 1986.
    \4\ Veil, J., M. Pruder, D. Littleton, and D. Moses. ``Cooling 
Water Use Patterns at U.S. Nonutility Electric Generating 
Facilities.'' Environmental Science and Policy. 2000.
---------------------------------------------------------------------------

    Hypothetically, if we assumed that there were additional once-
through cooling systems that were not included in our MACT floor 
analysis, we could assume that approximately 5 to 11 percent of the 
total cooling systems were once-through. The original number of cooling 
tower systems included in the MACT floor analysis was 520. If we assume 
that 5 to 11 percent of the cooling systems are once-through systems, 
then the total hypothetical number of cooling systems could range from 
547 to 584 cooling systems. The MACT floor for these cooling systems 
would be based on the average emissions limitations achieved by the top 
12 percent of cooling systems; the 6th percentile would be represented 
by the 33rd and the 35th cooling systems, respectively, for the 
hypothetical total number of cooling systems estimated to be 547 and 
584. There would be no change in the MACT floor for existing sources 
for this hypothetical case. The MACT floor would be identical to the 
requirements in the supplemental proposal, i.e., the 33rd and 35th 
ranked cooling systems have requirements to implement corrective action 
and heat exchange leak repairs when the strippable total VOC 
concentration in stripped air exceeds 6.2 ppmv. The owner or operator 
must identify the leaking heat exchanger, and repair at the earliest 
opportunity and no later than the next scheduled shutdown.
    To the extent the commenters are suggesting that once-through 
systems should be treated as a separate subcategory, they have provided 
no information to support that subcategorization is appropriate.
    We agree with the commenter and have clarified in Sec.  
63.654(b)(1) that the requirements do not apply to heat exchange 
systems where the minimum water-side pressure is 35 kPa greater than 
the maximum process-side pressure. We have also revised the definition 
of ``heat exchange system'' to identify the equipment that is included 
for closed-loop recirculation systems (systems with cooling towers), to 
identify the equipment that is included in the once-through systems, 
and to clarify that once-through systems are also regulated. 
Furthermore, definitions are provided for ``cooling tower return line'' 
and ``heat exchanger exit line'' to clarify the appropriate sampling 
locations. Sampling at either location is allowed; for once-through 
cooling systems, sampling is allowed at an aggregated location as long 
as it is before exposure to the atmosphere. To clarify this 
requirement, we have modified the definition of ``heat exchange exit 
line'' to be ``the cooling water line from the exit of one or more heat 
exchangers (where cooling water leaves the heat exchangers) to either 
the entrance of the cooling tower return line or prior to exposure to 
the atmosphere, whichever occurs first.''
3. Compliance Schedule for Heat Exchange Systems
    Comment: Several commenters supported the originally proposed 
compliance date of 3 years and 90 days. One commenter noted that the 
reference to 90 days in CAA section 112(f)(4) has been misread by some 
to limit compliance time, but since it is expected that installation of 
controls necessitates a longer time to comply, the waiver provisions 
should only be considered if EPA set a compliance deadline less than 3 
years. Some commenters noted that 18 months should be sufficient for 
all new requirements, as industry is already familiar with many of the 
processes to be controlled and are already regulating these emissions.
    Several commenters addressed the compliance dates relative to the 
supplemental proposal. For new sources, commenters noted that these 
requirements will be promulgated only 2 months after they were proposed 
in the supplemental proposal, which is inadequate time in which to have 
monitors purchased and operating. The commenters asserted that EPA 
should provide 1 year for new sources to comply with the standards.
    Commenters specifically noted that although many Texas refiners are 
currently familiar with the monitoring methods required for heat 
exchange systems, it took years for them to gain that familiarity, and 
it will take time for other refiners to learn to perform the methods 
efficiently. One commenter noted that when monitoring begins, there 
will be an initial period in which multiple repairs are necessary, some 
of which may require shutdowns. The commenters recommended that EPA 
provide the full 3 years provided by the CAA for compliance with heat 
exchange system requirements; this additional time would allow refiners 
to become familiar with the monitoring method and to complete initial 
repairs during already scheduled shutdowns and turnarounds. Conversely, 
several commenters stated that the cooling tower standards should be 
implemented in 1 year rather than progressively over 3 years as 
proposed in the supplemental proposal. Another commenter stated that 
the 18-month compliance schedule for heat exchange systems in the 
supplemental proposal is preferable to

[[Page 55677]]

the 3-year (and 90 days) compliance schedule in the original proposal.
    Response: As an initial matter, we note that the originally 
proposed compliance schedule (i.e., 3 years and 90 days) should not 
have included the additional 90 days. Section 112(i)(3) of the CAA 
provides that existing sources must comply within ``3 years after the 
effective date'' of the standard. With respect to the 18-month 
compliance timeframe specified in our supplemental proposal, we agree 
that the commenters have made valid points supporting adoption of a 3-
year compliance period instead. The comments that many refineries do 
not have experience with the TCEQ El Paso Method is supported by our 
review of cooling tower requirements for different States. We believe 
that some sources will need up to the full 3 years allowed under CAA 
section 112(i)(3) based on the estimated length of time required for 
refiners to survey the heat exchangers, identify those in organic HAP 
service, install the necessary sampling ports, purchase the Modified El 
Paso sampling system, familiarize themselves with the test method, and 
provide training to their employees. In addition, refiners will need to 
take steps to be prepared to repair leaking heat exchange systems. This 
includes performing initial sampling to identify heat exchangers that 
are prone to leakage or are in critical service, identify means to 
isolate or repair heat exchangers online, and to order and stock 
necessary equipment and spare parts.
    With respect to new source requirements, the CAA specifies that 
such sources must comply upon start-up or the date of publication of 
the final rule, whichever is later. We note that, based on the 
definition of an affected source in the Refinery MACT 1 rule, a 
construction project significant enough to trigger the new source 
provisions is likely to take years to complete, and that any source 
undertaking such project has been on notice since our initial proposal 
that cooling tower monitoring (or heat exchange system monitoring) 
would be required.
4. Delay of Repair Provisions
    Comment: Commenters noted that the new source delay of repair 
standards are based on cooling towers that are not yet operational, so 
those permit conditions are not ``achieved in practice.'' The 
commenters argued that it takes time after startup of new facilities to 
determine if new, previously untested requirements are achievable or 
whether permit modifications are needed; it is also unknown if Texas 
will allow deviations from permit conditions and under what conditions 
for heat exchange system repairs. The commenters stated that the new 
source delay of repair standards must instead be based on ``Repair and 
Delay 2'' as described in Table 1 of EPA's supporting memorandum (which 
the commenter thought were the requirements for the existing source 
floor).
    One commenter supported the 45-day repair allowance and delay of 
repair allowances. Another commenter stated that the maximum delay of 
repair should be 60 days because refineries already have 18 months to 
comply. Some commenters expressed concern that EPA proposed to disallow 
delay of repair for leaks above 62 ppmv after 3 years and noted that 
EPA has not demonstrated the rationale for removing that allowance. One 
commenter stated that EPA needs to address the situation in which 
multiple small leaks occur at multiple heat exchangers and the 
cumulative effect at the cooling tower return line is a leak above 62 
ppmv. The commenters stated that unplanned shutdowns are expensive and 
disruptive, but would be necessary when repair is infeasible without a 
shutdown. One commenter requested that EPA allow owners and operators 
to request delay of repair on a case-by-case basis when justified.
    Response: The supplemental proposed MACT floor for both new and 
existing sources is repair within 45 days for leaks of 62 ppmv or 
greater. In establishing the floor, we found that the no delay of 
repairs requirement for large leaks has been implemented and required 
for 35 cooling towers at numerous facilities. Also, both the top-ranked 
and 6th percentile cooling tower had identical requirements excluding 
large leaks from delay of repair. As such, this requirement has been 
implemented and has been adequately demonstrated and it establishes the 
minimum floor requirement. In the supplemental proposal, we proposed to 
allow delay of repair for large leaks for the 18 month phase-in of the 
repair requirements, which correspond to the ``Repair and Delay 2'' 
provisions cited by the commenter. However, we have concluded that 
these temporary delay of repair provisions were not equivalent to the 
requirements for the MACT floor for existing heat exchange systems, 
which is why they were only temporary provisions in the supplemental 
proposal. Additionally, the 3-year compliance timeframe in the final 
rule will allow facilities sufficient time to resolve these initial 
problems. As discussed previously, we are now implementing all heat 
exchange system requirements for existing sources on the same 3-year 
schedule. Upon implementation of the required monitoring provisions, it 
is anticipated that leaks will be identified well before they become 
large. Thus, while delay of repairs are allowed for small leaks, it is 
the refinery owner or operator's responsibility to order necessary 
parts and schedule a repair before the leak exceeds the 62 ppmv 
threshold. Negligence on the part of the owner or operator regarding 
this responsibility is not a reasonable justification for providing 
delay of repair provisions for large leaks. Consistent with the 
requirements that apply to the units which provided the basis for the 
MACT floor, any leak greater than 62 ppmv that is not repaired in the 
timelines provided in the rule is a deviation of the standard and 
subject to enforcement actions at the discretion of the Agency or 
permitting authority.
5. Monitoring Alternatives
    Comment: Commenters noted that the concentration of heavy organic 
HAP and water soluble HAP can build up in recirculating cooling tower 
systems, and since the El Paso Method involves more vigorous stripping 
than occurs in a cooling tower, monitoring might falsely indicate a 
leak. The commenters suggested that, as an alternative, sources should 
be allowed to use methods they are presently using, including testing 
the inlet water to a heat exchange system and using the difference 
between the outlet and the inlet to determine if the leak definition is 
exceeded. One commenter noted that if once-through cooling systems 
continue to be considered affected facilities by EPA, it is important 
for the requirements to consider the baseline of HAP (or surrogate VOC) 
emissions in the inlet to the system so that facilities are only 
responsible for assessing any ``increase'' in the pollutant attributed 
to the operating facility, not pollutants in the water basin upstream 
of the facility. Another commenter requested that EPA allow owners or 
operators to demonstrate that another monitoring method such as a 
continuous emission monitoring system or parameter monitoring is 
equivalent to the monitoring methods specified for heat exchange 
systems. One commenter requested that EPA continue to allow the method 
originally proposed as well as a relatively new analytical method for 
early detection developed by Baker Petrolite. Another commenter stated 
that the El Paso Method measures VOC in the air, and EPA should allow 
any monitoring method that has adequate sensitivity to measure 80 ppbw 
of

[[Page 55678]]

strippable VOC in the water or for a surrogate that can be correlated 
to strippable VOC and can be measured at a level that would indicate a 
leak of 80 ppbw of strippable VOC in the water for a particular heat 
exchange system. This monitoring flexibility would be helpful to 
confirm El Paso results as well as more efficient for sources that are 
required to conduct other types of monitoring by their State or local 
agency or for compliance with another Federal regulation (such as the 
HON).
    Response: We acknowledge that some refineries have specific 
monitoring systems in-place and that the use of these monitoring 
systems would ease the burden on the refinery owner or operator. 
However, we are not aware of any practical alternatives that we can 
specify that provide an equivalent measure of strippable organics. Nor 
have any of the commenters provided evidence that a specific 
alternative method would result in an equivalent measure. For example, 
we have reviewed the ``method for early detection developed by Baker 
Petrolite'' and found that the detection level for most individual 
compounds is much higher than the total strippable VOC concentrations 
that define a leak for the MACT floor facility. That is, this method 
would not be able to identify small to medium-sized leaks that would be 
identified and would be required to be fixed by the MACT requirements 
for heat exchange systems.
    Although we expect the El Paso column to mimic the stripping that 
occurs in the cooling tower, the amount of stripping that occurs in the 
cooling tower is dependent on the design and operation of the cooling 
tower. Moreover, the purpose for the use of the El Paso Method is to 
detect leaks in heat exchange systems, not to estimate emissions. 
Consequently, we do not believe that analytical methods based on the 
measurement of single constituents or that employ inlet/outlet cooling 
tower water sampling are equivalent to the El Paso Method for 
determining strippable VOC. That is, these alternative methods would 
not result in the same corrective action thresholds as the prescribed 
monitoring technique.
    The commenters have provided no evidence that a build-up of heavy 
organics would cause a heat exchange system to exceed a leak definition 
of 6.2 ppmv total strippable VOC, nor have they provided compelling 
evidence that such a leak would not result in any air emissions. While 
we agree that the relative stripping efficiency of a given cooling 
tower will not necessarily match the stripping efficiency of the El 
Paso stripping column, it is unreasonable to conclude that the cooling 
tower will have no HAP emissions. Furthermore, the majority of HAP 
included in Table 1 are volatile. Thus, for a heat exchange system that 
is ``in HAP service,'' we believe it is appropriate to initiate 
corrective action if the leak threshold is exceeded because that 
corrective action will result in reduced HAP emissions.
    As stated previously, the goal of the heat exchange system 
provisions is to identify and fix leaks at the heat exchanger to reduce 
subsequent emissions of HAP. For once-through cooling systems, we 
believe it is unlikely that the strippable organics concentration in 
the inlet water would exceed the leak threshold. Further, the 
commenters have provided no evidence that the fresh water feed for a 
once-through heat exchange system could contain enough strippable 
organics to cause a heat exchange system to exceed a leak definition of 
6.2 ppmv total strippable VOC. Therefore, we have not provided any 
alternative leak detection procedure for once-through heat exchangers.
    Comment: Commenters supported allowing the facility to demonstrate 
that a leak is not in a heat exchanger that is in HAP service. One 
commenter stated that if VOC testing indicates a leak in a heat 
exchange system, the facility should be allowed to speciate the 
compounds in the leak to determine if the leak is a HAP le