EPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park, NC 27711
EPA-453/R-99-002b
May 1999
Air
NATIONAL EMISSION STANDARDS FOR
HAZARDOUS AIR POLLUTANTS (NESHAP)
FOR THE POLYETHER POLYOLS
MANUFACTURING INDUSTRY:
SUMMARY OF PUBLIC COMMENTS AND
RESPONSES
U S. Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson ppulevard, 12Wl
Chicago, \l 60604-3590
-------
EPA-453/R-99-0025
NATIONAL EMISSION STANDARDS FOR
HAZARDOUS AIR POLLUTANTS (NESHAP) FOR THE
POLYETHER POLYOLS MANUFACTURING INDUSTRY
Background Information for
Promulgated Standards - Summary of
Public Comments and Responses
Emission Standards Division
0 s Environmental Protection A8.nc,
60604-3590
U. S. Environmental Protection Agency
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
May 1999
-------
-------
Disclaimer
This report is issued by the Office of Air Quality Planning and Standards, U. S.
Environmental Protection Agency. Mention of trade names and/or commercial products is not
intended to constitute endorsement or recommendation for use. Copies of this report are
available free of charge to Federal employees, current contractors and grantees, and nonprofit
organizations-as supplies permit-from the Library Services Office (MD-35), U. S.
Environmental Protection Agency, Research Triangle Park, NC 27711, (919-541-2777) or, for a
nominal fee, from the National Technical Information Service, 5285 Port Royal Road,
Springfield, Virginia 22161, (703-487-4650).
in
-------
ENVIRONMENTAL PROTECTION AGENCY
National Emission Standards for Hazardous Air Pollutants for the
Polyether Polyols Manufacturing Industry-
Background Information for Promulgated Standards
Prepared by:
Sally L. Shaver (Date)
Director, Emission Standards Division
U. S. Environmental Protection Agency
Research Triangle Park, NC 27711
1 The final National Emission Standards for Hazardous Air Pollutants (NESHAP) will regulate
emissions of hazardous air pollutants from polyether polyols manufacturing operations.
Only those operations that are part of major sources under section 112(d) of the Clean Air
Act as amended in 1990 will be regulated.
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; and the Council on environmental Quality; members of the
State and Territorial Air Pollution program Administrators; the Association of Local Air
Pollution Control Officials; EPA Regional Administrators; and other interested parties.
3 For additional information contact:
Mr. David Svendsgaard
Organic Chemicals Group (MD-13)
U. S. Environmental Protection Agency
Research Triangle Park, NC 27711
Telephone: (919) 541-2380
4. Paper copies of this document may be obtained from:
National Technical Information Service (NTIS)
5285 Port Royal Road
Springfield, VA 22161
Telephone: (703) 487-4650
-------
U. S. EPA Library Services Office (MD-35)
U. S. Environmental Protection Agency
Research Triangle Park, NC 27711
5. Electronic copies of this document may be obtained from the EPA's OAR Technology
Transfer Network website (TTNWeb).
The TTNWeb is a collection of related Web sites containing information about many areas of
air pollution science, technology, regulation, measurement, and prevention. The TTNWeb is
directly accessible from the Internet via the World Wide Web at the following address:
http//www.epa.gov/ttn
VI
-------
TABLE OF CONTENTS
1.0 SUMMARY W
1.1 BACKGROUND M
1.2 SIGNIFICANT CHANGES SINCE PROPOSAL M
1.2.1 Primary Product Determination 1-3
1.2.2 Definition of "Polyether Polvol" M
1.2.3 The Definition of a "Process Vent" 1-5
1.2.4 Start-up and Shutdown Definitions 1-6
1.2.5 A Concentration Limit as an Alternative Epoxide Process Vent Emission
Limit for New Sources 1-6
1.2.6 Flares as a Reference Control Technology 1-8
1.2.7 Group Determination on an Individual Process Vent Basis for Nonepoxide
Emissions from Making or Altering the Product 1-8
1.2.8 Possibility of Dual Controls for Nonepoxide HAPs for Making or
Modifying the Product 1-10
1.2.9 Worst-Case Testing Requirements 1-11
1.2.10 Engineering Calculations as an Alternative to Performance Testing .. 1-13
1.2.11 Revisions to the Test Method Requirement for Control Efficiency
Determination 1-16
1.2.12 Flexibility of the Determination of a Site-specific Onset of Extended
Cookout 1-17
1.2.13 Parameter Monitoring Excursion Definitions 1-18
1.2.14 Start-up. Shutdown, and Malfunction Plan 1-20
2.0 SUMMARY OF PUBLIC COMMENTS AND RESPONSES 24
2.1 APPLICABILITY 24
2.1.1 PMPUs Without Organic HAP 24
2.1.2 Definition of Affected Source 2-3
2.1.3 Processes Exempted from Affected Sources 2-5
2.1.4 Necessity of Primary Product Redetermination 2-6
2.1.5 Primary Product Determination and Applicability 2-9
2.1.6 Primary Product Determination for Non-Flexible Operation Units ... 2-10
2.1.7 Applicability for Flexible Operation Units 2-10
2.1.8 Requirements for Flexible Operation Units when Producing Non-polyether
Polvols 2-11
2.1.9 NonPMPUs that Produce Polvether Polvols 2-13
2.1.10 Changes or Additions to Plant Sites 2-14
2.1.11 Applicability of this subpart except during periods of start-up, shutdown.
malfunction, or non-operation 2-17
2.1.12 Storage Tank Predominant Use 2-18
2.1.13 §63.1420(e¥5)(ii) 2-19
2.1.14 Section 63.1420(c)(l) and (c¥8) Appear to be Duplicative 2-20
vii
-------
TABLE OF CONTENTS (continued)
2.2 COMPLIANCE AND RELATIONSHIP TO OTHER RULES 2-20
2.2.1 Time Periods 2-20
2.2.2 Changes to the General Provisions 2-2.1
2.2.3 Cross-references with Other Subparts 2-22
2.2.4 Interfacing with Title V Operating Permit 2-24
2.2.5 Complexity of the Proposed Rule 2-25
2.3 DEFINITIONS 2-26
2.3.1 Annual Average Concentration 2-26
2.3.2 Batch Cycle 2-26
2.3.3 Cross-referencing of Definitions from Other Rules 2-27
2.3.4 Epoxide 2-31
2.3.5 Extended Cookout 2-32
2.3.6 Impurity 2-32
2.3.7 Make or Modify the Product 2-33
2.3.8 Non-epoxide HAP 2-33
2.3.9 Organic HAP 2-34
2.3.10 Override Definitions 2-36
2.3.11 PMPU 2-36
2.3.12 Polvether Polvol 2-37
2.3.13 Pressure Decay Curve 2-39
2.3.14 Process 2-40
2.3.15 Process Condenser 2-41
2.3.16 Process Vent 2-41
2.3.17 Product Class 2-43
2.3.18 Purification 2-45
2.3.19 Recovery Device 2-45
2.3.20 "Start-up" and "Shutdown" 2-46
2.3.21 Storage Vessel 2-48
2.3.22 Unit Operation 2-48
2.4 PROCESS VENT CONTROL REQUIREMENTS 2-49
2.4.1 3 Percent Oxygen Correction 2-49
2.4.2 A Concentration Limit as an Alternative Process Vent Emission Limit
2-60
2.4.3 Basis for Outlet Concentration Testing as an Alternative Process Vent
Limit 2-64
2.4.4 Group Determination for Nonepoxide HAP Process Vent Emissions on a
Vent-bv-Vent Basis 2-65
2.4.5 Relationship to Polymers and Resins I Changes 2-71
2.4.6 Group Redetermination 2-72
2.4.7 Non-epoxide versus Epoxide Process Vent Emission Limits 2-72
2.4.8 Alternative Emission Factor 2-73
vm
-------
TABLE OF CONTENTS (continued)
2.4.9 Batch/Continuous Process Units versus Batch/Continuous Unit Operations
2-74
2.4.10 Controls for New Source Process Vent Emission Limits 2-74
2.4.11 Calculation Methods for Emissions from Batch Operation Units .... 2-75
2.4.12 Continuous Unit Operations Group Determinations 2-76
2.4.13 Summing the Values 2-76
2.4.14 Flares as a Reference Control Technology for Existing and New Sources
2-77
2.4.15 Use of Multiple Compliance Methods 2-79
2.4.16 Uses and Emission Point Locations of Nonepoxides 2-80
2.4.17 Possibility of Dual Controls For Nonepoxide HAPs from Making or
Modifying the Product 2-81
2.4.18 Clarification of "Schedule for Compliance" 2-83
2.4.19 THE PMPU Exemption from Reporting 2-84
2-84
2.4.20 Clarification to Condenser Identity 2-85
2.4.21 Small Combustion Device Performance Testing 2-86
2.4.22 Concentration Compliance 2-87
2.4.23 "Organic HAP" versus "HAP of Concern" 2-87
2.4.24 Rename "Product" as "Reactor Liquid" 2-88
2.4.25 §63.1425(e)(l)(i) 2-89
2.5 EXTENDED COOKOUT AS A CONTROL OPTION 2-90
2.5.1 In Support of the Inclusion of ECO 2-90
2.5.2 ECO Compliance Demonstration 2-90
2.5.3 Flexibility of the Determination of a Site-specific Onset of ECO 2-91
2.5.4 ECO Requirements for Monitoring 2-94
2.5.5 Accuracy "Buffer" for ECO Emissions Calculations 2-96
2.5.6 Clarification of the Definition of "Emissions" in ECO Destruction
Efficiency Calculation 2-97
2.5.7 First Order for Epoxide Reactions 2-99
2.5.8 Update of Product Class List 2-99
2.5.9 Product Class Definition 2-100
2.5.10 Pressure Measurements 2-10.1
2.5.11 ECO Recordkeeping 2-101
2.6 MONITORING REQUIREMENTS FOR PROCESS VENTS 2-102
2.6.1 Monitoring Requirement Exemptions for the Process Vent Requirements
2-102
2.6.2 Scrubber Monitoring Requirements 2-103
2.6.3 Monitoring of Multiple Absorbers 2-104
2.6.4 Alternative to §63.1429(a¥7) 2-104
2.6.5 Flow Indicator Specifications 2-105
IX
-------
TABLE OF CONTENTS (continued)
2.7 GROUP DETERMINATION FOR NONEPOXIDE HAP EMISSIONS 2-106
2.7.1 Group Determination Records When the TRE Index is Between 1.0 and
fLQ 2-106
2.8 PROCESS VENT RECORDKEEPING AND REPORTING REQUIREMENTS
2-107
2.8.1 "Up-to-date" Process Vent Records 2-107
2.8.2 Flares Compliance Demonstration 2-107
2.8.3 Records for Start-ups, Shutdowns, Malfunctions, and Periods of Non-
operation 2-108
2.8.4 Records Related to Group Determination 2-109
2.8.5 Recordkeeping Exemption 2-110
2.8.6 Process Change Resulting in a Change in the TRE 2-111
2.9 EMISSION FACTOR 2-112
2.9.1 Emission Factor Plan 2-112
2.10 STORAGE VESSEL 2-113
2.10.1 Storage Vessel Provisions 2-113
2.10.2 Previous Performance Testing of Storage Vessel Control Device ... 2-113.
2.11 WASTEWATER PROVISIONS 2-114
2.11.1 Wastewater Cross-Referencing with the HON 2-114
2.11.2 Definition of Wastewater in the Basis and Purpose for Proposed Standards
2-115
2.11.3 Classification of Wastewater Receiving Facilities 2-116
2.11.4 Clarifying the Definition of Residuals 2-117
2.11.5 Wastewater Control Options 2-117
2.11.6 "Inprocess" Aqueous Streams Should not be Identified as Wastewater
Streams 2-118
2.11.7 Selection of Compounds Subject to the Wastewater Provisions .... 2-119
2.11.8 Wastewater Nationwide Baseline Emissions Estimate: Steam Jet System's
Condensate Estimate 2-120
2.11.9 Wastewater Nationwide Baseline Emissions Estimate: The Fraction
Emitted Factors Used 2-122
2.11.10 Inclusion of Biological Treatment as Wastewater Control Option
2-123
2.11.11 PEPO Chemical List Versus HON's Table 8 2-124
2.12 EQUIPMENT LEAK PROVISIONS 2-125
2.12.1 Method 21 for Equipment Leak Detection 2-125
2.12.2 "Delay of Repair" Example in the Preamble 2-125
2.12.3 Cost-Effective Alternatives to the HON LDAR 2-126
2.12.4 De minimis Equipment Count 2-126
2.12.5 Applicability of Subpart 1 2-127
2.12.6 Inclusion of Phase-in Option 2-127
-------
TABLE OF CONTENTS (continued)
2.12.7 Exemption for Heat Exchanger Units not Using HAP 2-128
2.13 TESTING 2-129
2.13.1 Consistency Between the Test Method Required in the Proposed Rule and
the Test Method Used to Establish the New Source Process Vent MACT
Floor 2-129
2.13.2 Use of Method 25A Without Method 301 Validation 2-129
2.13.3 Allow any Testing to Demonstrate Compliance 2-131
2.13.4 Notice for Rescheduling of a Test 2-132
2.13.5 Engineering Calculations for Worst-case Requirements 2-133
2.13.6 Daily Averages versus Compliance 2-134
2.13.7 Method 1 or 1A 2-135
2.13.8 Engineering Calculations as an Alternative Compliance Demonstration to
Performance Testing 2-136
2.13.9 Request for Exemption from Testing Multiple Similar Controls ... 2-139
2.13.10 Worst-Case Testing 2-140
2.13.11 Determination of Emission Profile for Worst-Case Testing . 2-144
2.14 PARAMETRIC MONITORING 2-145
2.14.1 Operating Permit Requirements 2-145
2.14.2 Average versus Maximum Value for Monitoring 2-145
2.14.3 Compliance Determination 2-146
2.14.4 Excursion Provisions for Storage Vessels Exempt from Continuous
Monitoring, Process Vents from Batch Unit Operations, and Extended
Cookout 2-150
2.14.5 Continuous Monitoring 2-151
2.14.6 Clarification to Text 2-153
2.15 GENERAL RECORDKEEPING AND REPORTING 2-154
2.15.1 Elimination of Initial Notification 2-154
2.15.2 Provide Examples of Emission Points 2-154
2.15.3 Periodic Report 2-155
2.15.4 Include all Records in One Section 2-155
2.15.5 Parametric Monitoring During Periods of Start-up, Shutdown or
Malfunction 2-156
2.15.6 Exclusion of Monitoring Data from Daily Averaging 2-161
2.15.7 Retention of Superseded Start-up. Shutdown, and Malfunction Plan
2-162
2.15.8 Exclusion to Continuous Recordkeeping Requirements 2-162
2.15.9 Delete Redundancy in Recordkeeping Requirements 2-163
2.15.10 Recordkeeping Waiver 2-163
2.15.11 "Document on Demand" 2-164
2.15.12 Notification of Compliance Status 2-165
2.15.13 Addition of a Provision for Reporting Updates 2-166
XI
-------
TABLE OF CONTENTS (continued)
2.15.14 Revise Exemptions from Recordkeeping 2-168
2.15.15 Recordkeeping Retention Revision 2-169
2.15.16 Exclusion of Equipment Leaks to the Periodic Reporting
Requirements 2-170
2.15.17 Record of Preparation of Standards 2-171
2.15.18 Periodic Reports 2-171
2.15.19 Predominant Use Reporting 2-173
2.15.20 Alternative Continuous Monitoring and Recordkeeping .... 2-174
2.15.21 Consolidation of Periodic Reporting 2-174
2.15.22 Group Status Change Reporting 2-175
2.15.23 Excursions 2-176
2.15.24 Alternative Continuous Monitoring for Storage Vessels .... 2-177
2.15.25 Cross-referencing in Reduced Recordkeeping 2-178
2.15.26 Operating Permit Application 2-178
2.15.27 Manual Reading 2-179
2.15.28 "Any Other Information" in Sections 63.1439(e)(5Ki)(B) and
63.1439(eX6Kiv)(B) 2-179
2.15.29 §63.1439(g)(3) 2-180
2.16 EDITORIAL 2-181
2.16.1 "The Owner or Operator" Versus "Each Owner or Operator"
2-181
2.16.2 Section 63.1420(e)(3) 2-181
2.16.3 Section 63.1420(e)(5)(i) 2-182
2.16.4 Section 63.1420(e)(5)(iii) 2-182
2.16.5 Sections 63.1427fh)f 1) and fh¥2) 2-183
2.16.6 Section 63.1427(l)(iv) 2-183
2.16.7 Section 63.1433(aW) 2-183
2.16.8 Section 63.1433(b) 2-183
2.16.9 Section 63.143l(f)(l) 2-184
2.16.10 62 FR 46812. col. 3 2-184
2.16.11 Section 63.1438(b¥3) 2-185
2.16.12 Heading in §63.1420(c) 2-185
2.16.13 Section 63.1420(c¥2) 2-185
2.16.14 Section 63.1420(d) 2-186
2.16.15 Section 63.1420(e)m 2-186
2.16.16 Section 63.1422(d)(2Xiv) and (dX5) 2-186
2.16.17 Section 63.1420(f)(7)fi) 2-187
2.16.18 Section 63.1424(a) 2-187
2.16.19 Section 63.1425(c)(4)(D 2-187
2.16.20 Section 63.1426(c) 2-188
2.16.21 Section 63:i426(c)m(D(q(iD. (c¥3)(iD. and (e¥D 2-188
XII
-------
TABLE OF CONTENTS (continued)
2.16.22 Section 63.1426(e¥l) 2-189
2.16.23 "Pressure Decline Curve" Versus "Pressure Decay Curve" .. 2-190
2.16.24 Section 63.1427(l)(3)(ii) 2-190
2.16.25 Section 63.1428(c) 2-190
2.16.26 Section 63.1435fd) 2-191
2.16.27 Section 63.1439(b) 2-191
2.16.28 Section 63.1439fe¥6¥iii¥A) 2-192
2.16.29 Section 63.1439(e¥6)(iii¥D¥l) and (f)(3) 2-192
2.16.30 Section 63.1426(c¥4¥iv). (c). and (c¥l¥i¥A) 2-193
2.16.31 Equation 9 2-193
2.16.32 Section 63.142Sfe¥2¥i) 2-194
2.16.33 Section 63.1425ffl 2-194
2.16.34 Section 63.1429fd¥3) 2-195
2.16.35 Section 63.1439(e¥6¥v) 2-195
2.16.36 Section 63.1439(h¥l) 2-195
2.16.37 Section 63.1427 2-196
2.16.38 Section 63.1439(e¥5)(ii)(D) 2-196
2.16.39 Section 63.14390?) 2-197
2.16.40 Actual Date Versus "Date of Promulgation" 2-198
2.16.41 Changes to Several Equations 2-198
2.16.42 Section 63.1420(h¥3) 2-199
2.16.43 Table 1 2-199
2.16.44 Table 1 2-200
2.16.45 Table 2 2-200
2.16.46 Table 2 2-201
2.16.47 Table 4 2-201
2.16.48 Table 5 2-201
2.17 LEGAL CONSIDERATIONS 2-202
2.17.1 Executive Order 12866 2-202
Xlll
-------
1.0 SUMMARY
1.1 BACKGROUND
On September 4, 1997 (62 FR 46804), the United States
Environmental Protection Agency proposed National Emission
Standards for Hazardous Air Pollutants (NESHAP) for Polyether
Polyols Production under Section 112(d) of the Act.
Public comments were requested on the proposed standard and
comment letters were received from industry representatives and
governmental entities. A total of 13 comment letters were
received. Table 1-1 presents a listing of all persons that
submitted written comments, their affiliation, and their docket
item number. A public hearing was not requested.
The written comments that were submitted on the proposed
rule have been summarized, and responses to the comments are
included in the following sections. This summary of comments and
responses serves as the basis for revisions made to the NESHAP
between proposal and promulgation.
1-1
-------
TABLE 1-1. LIST OF COMMENTERS ON PROPOSED NATIONAL EMISSION
STANDARDS FOR HAZARDOUS AIR POLLUTANTS
Air Docket A-96-38 Commenter and affiliation
Item Number
IV-D-01 T.A. Threet, Counsel, The Dow Chemical Company,
Midland, MI
IV-D-02 M. Wax, Deputy Director, The Institute of Clean
Air Companies, Washington, D.C.
IV-D-03 J.P. Keigher, Responsible Care Manager, Olin
Chemicals Group, Brandenburg, KY
IV-D-04 D.W. Gustafson, EH&S Regulatory Management; T.A.
Threet, Legal, The Dow Chemical Company, Midland,
MI
IV-D-05 M.A. Healey, Director, Federal Environment and
Transportation Issues, The Society of the Plastics
Industry, Inc., Washington, D.C.
IV-D-06 M. Manning, Corporate Ecology & Safety, BASF
Corporation, Enka, NC
IV-D-07 J.C. Hovious, Assistant Director, Environmental
Affairs, Union Carbide Corporation, Danbury, CT
IV-D-08 M.L. Mullins, Vice President, Regulatory Affairs,
Chemical Manufacturers Association, Arlington, VA
IV-D-09 J.A. Dege, Jr., Manager - Air Programs, DuPont SHE
Excellence Center, Wilmington, DE
IV-D-10 D.C. Boyle, Director, Environmental, Health and
Safety, ARCO Chemical Company
IV-D-11 D.C. Boyle, Director, Environmental, Health &
Safety, ARCO Chemical Company, Newtown Square, PA
IV-G-01 C.F. Johnston, Environmental Manager, ICI
Chemicals & Polymers, New Castle, DE
IV-G-02 J.C. Hovious, Assistant Director, Environmental
Affairs, Union Carbide Corp, Supplemental
Comments, Danbury, CT
1-2
-------
1.2 SIGNIFICANT CHANGES SINCE PROPOSAL
In response to comments received on the proposed standards,
several changes have been made to the final rule. A summary of
the substantive changes made since the proposal in response to
comments is provided in the following sections. Additional
information on the final rule is contained in the docket for this
rule (Docket A-96-38).
1.2.1 Primary Product Determination
One commenter expressed confusion over aspects of the
primary product determination in the proposed rule, particularly
the provision that specified how a non-PMPU could become a PMPU
after the initial determination based on actual production. The
EPA agreed that this portion of the proposed primary product
provisions needed clarification. In fact, the EPA conducted an
overall review of the proposed primary product provisions, and
concluded that several structural and clarifying changes were
needed. In addition, the EPA noted some potential situations
that could occur that were not addressed in the proposed
provisions.
The specific concern raised by the commenter was addressed
by clearly stipulating how owners or operators of non-PMPUs are
to determine whether they have become subject to the rule after
the initial primary product determination. The final rule
specifies that non-PMPUs that have produced polyether polyols in
the past five years are to annually re-determine the primary
product using actual production values. The rule also specifies
how a non-PMPU process unit is to determine the primary product
if it has not produced polyether polyols in the past five years,
but plans to produce polyether polyols in the future.
The proposed provisions required that initial primary
product determination be based on a five-year prediction of
anticipated production by the owner or operator. The EPA is
aware that, in some instances, the owner or operator may not be
able to make such a prediction. Clarifications and/or revisions
were made to the primary product provisions to address this
situation. First, in the initial determination, the time frame
1-3
-------
for which production must be anticipated for new process units
was changed to one year. Also, provisions were added for owners
or operators that cannot determine the primary product based on
anticipated five-year (or one-year) production. In such
situations, the process unit is designated as a PMPU and is
subject to the existing source provisions of subpart PPP, if
polyether polyols have been produced in an existing process unit
for 5 percent or greater of the time since September 4, 1997.
For new process units, if polyether polyols will be produced at
any time during the first year of production, then the unit is a
PMPU and subject to the new source provisions of subpart PPP.
In addition to the provisions discussed above that specify
how non-PMPUs are to determine if they become PMPUs and subject
to subpart PPP, the EPA has also clarified and expanded the
provisions that specify how the PMPU designation can be removed
from a process unit. The first case,, which is retained from the
proposed rule, is where production of polyether polyols ceases
and the owner or operator does not anticipate the production of
polyether polyols in the future. Also, the EPA has added
provisions that specify procedures for a primary product re-
evaluation based on actual production. If an owner or operator
of a PMPU finds that another product has been produced for a
greater amount of time than polyether polyols over a specified
time period (previous five years or since beginning the
production of polyether polyols), then the PMPU designation would
be removed provided that production of the "new" primary product
must make the process unit subject to another part 63 NESHAP. If
the new primary product is not subject to another part 63 NESHAP
and polyether polyols continue to be produced, the process unit
continues to be classified as a PMPU and continues to be subject
to subpart PPP.
The EPA has also added provisions addressing the
determination of the primary product in situations where two or
more products are produced simultaneously. Also, clarifications
were made in the reporting and recordkeeping requirements
associated with the primary product determination.
1-4
-------
1.2.2 Definition of "Polvether Polvol"
In the proposed rule a "Polyether Polyol" was defined as "a
compound formed through the polymerization of ethylene oxide (EO)
or propylene oxide (PO) or other cyclic ethers with compounds
having one or more reactive hydrogens (i.e., a hydrogen atom
bonded to nitrogen, oxygen, phosphorus, sulfur, etc.) to form
polyethers. This definition excludes materials regulated under
the HON, such as glycols and glycol ethers."
One commenter requested that EPA revise the definition of
"polyether polyol" to clarify that the production of typical
alkanolamines, which lack repeating ether units, is not regulated
under subpart PPP. Another commenter explained that hydroxy
ethyl cellulose is formed through the reaction of EO on cellulose
polymer molecules. This commenter requested that the EPA clarify
whether hydroxy ethyl cellulose manufacturing is included or
excluded from the definition of "polyether polyol."
The EPA has revised the definition of "polyether polyol" in
the final rule addressing both these issues by excluding the
production of hydroxy ethyl cellulose and by specifying that a
polyether must have more than two ether bonds.
1.2.3 The Definition of a "Process Vent"
The definition of "process vent" in the proposed rule did
not include any cutoffs based on the flow or HAP concentration of
the process vent. One commenter was concerned that the
definition of "process vent" did not have a de minimis cutoff, as
does the definition of "process vent" in the HON. The cutoff
suggested by the commenter (0.005 weight-percent total organic
HAP) has been incorporated into the final definition of a process
vent, for process vent from continuous unit operations. This
decision was based on the fact that the EPA considers it to be
impractical to impose requirements for process vent streams with
such low HAP concentrations (less than 0.005 weight percent
organic HAP). For similar reasons, a de minimis cutoff for
process vents from batch unit operations was also added in the
final rule. In the Polymers and Resins I and IV NESHAP, the
batch process vent definition contains a de minimis cutoff of 225
1-5
-------
kg/yr uncontrolled HAP emissions. The EPA believes that this
level is also an appropriate de minimis level for process vents
from batch unit operations in the polyether polyols industry.
1.2.4 Start-up and Shutdown Definitions
One commenter noted that the definitions of "start-up" and
"shutdown" in the proposed rule were not parallel. The
definitions were revised in the final rule.
1.2.5 A Concentration Limit as an Alternative Epoxide Process
Vent Emission Limit for New Sources
The proposed rule did not include a concentration limit as
an alternative epoxide process vent emission limit for new
sources. The preamble to the proposed rule solicited comments on
this subject, to which four commenters responded. All four
recommended a 20 ppmv alternative concentration limit. The
commenters indicated that the preambles for the New Source
Performance Standard for VOC Emissions from Synthetic Organic
Chemical Manufacturing Industry (SOCMI) Distillation Operations
(40 CFR 60, subpart NNN), and the HON (40 CFR 63, subpart G)
provided rationales for a 20 ppmv limitation that also are
applicable to the polyether polyols rule.
In subpart NNN's preamble (48 FR 48932, October 21, 1983),
the EPA stated that the outlet concentration of 20 ppmv was
established based on kinetic calculations of incinerators. It
was demonstrated that, at a given temperature and residence time,
a stream with a low inlet concentration could not demonstrate an
outlet concentration below 20 ppmv. In the preamble to the
proposed amendments to the HON (61 FR 43698, August 26, 1996),
the EPA expanded the application of this lower bound
concentration performance standard to control/recovery devices
other than incinerators. In the HON preamble, the EPA explained
that recovery devices are designed to typically reduce emissions
to the same outlet concentration level given a relatively wide
range of inlet concentrations. When the inlet concentration is
substantially below the design maximum leading conditions (and
begins to approach the residual level in the outlet stream) the
recovery device efficiency will decrease.
1-6
-------
Therefore, the final rule contains an alternative
concentration limit of 20 ppmv, which is measured at the outlet
of the combustion, recovery, or recapture device. This
alternative provides owners or operators of affected sources with
the ability to comply with the standard when the inlet
concentration to the combustion, recovery, or recapture device
drops below the point where optimum control efficiency can be
achieved, and when it is not feasible to require optimum percent
reduction levels to be met. This rationale for allowing the 20
ppmv alternative is applicable for both new and existing sources,
even though the new source percent reduction limitation is more
stringent than the existing source percent reduction limitation.
Therefore, the final rule allows this alternative for both new
and existing sources.
Another commenter advocated that the alternative 20 ppmv
concentration limit should apply more broadly to process vents
that do not utilize a combustion, recovery, or recapture device
to reduce epoxide emissions. The examples provided by the
commenter included vents from equipment practicing a very long
extended cookout or vents from equipment where the epoxide
content is very low and emissions are very small.
The EPA understands that the outlet concentration after
extended cookout (ECO) may be as low as that after a combustion,
recovery, or recapture device. However, this is not based on
technological limitations of ECO, as is the basis for the 20 ppmv
concentration limit for combustion, recovery, and recapture
devices. Therefore, the EPA believes that allowing the 20 ppmv
concentration limit for ECO is not appropriate.
Further, the EPA does not believe that it is appropriate to
use this alternative concentration requirement as a de minimis
cutoff for vents where the epoxide content is very low and
emissions are very small. The EPA believes that the HAP
concentration and emission de minimis cutoffs in definition of
the process vent (discussed above in Section 1.2.3) adequately
address these vents.
1-7
-------
Finally, the proposed existing source concentration limit
was 20 ppmv of total epoxides. Other rules, such as the HON,
allow the option of determining outlet concentration limits on a
TOG basis. In many instances in the polyether polyols industry,
all the TOC in the emission stream will be epoxides, making the
TOC and epoxide concentration equivalent. In fact, if there were
other TOC in the stream, compliance with a 20 ppmv TOC limit
would mean that the epoxide concentration would necessarily be
less than 20 ppmv. For these reasons, the EPA believes that
having the alternative concentration limits based on total
epoxides or TOC is appropriate for this rule. As discussed later
in Section 1.2.11, the EPA decided to allow Method 25A (which is
designed to measure TOC) to determine compliance with the
alternative concentration limits.
1.2.6 Flares as a Reference Control Technology
Two commenters requested that the EPA allow flares as a
reference control technology for process vents at existing and
new sources. The EPA agreed with the commenters that flares are
an acceptable reference control technology for situations where
the required organic HAP emission reduction is 98 percent or
less. The final rule was revised to allow flares as a reference
control technology for epoxide process vent emissions at existing
sources, for Group 1 nonepoxide HAP process vent emissions at new
and existing sources, and for nonepoxide HAP process vent
emissions from catalyst extraction at new and existing sources.
However, the data presented by the commenters do not support a
destruction efficiency of 99.9 percent for flares combusting EO
and PO, which is the equivalent percent reduction efficiency for
the epoxide process vent limitation for new sources. Therefore,
the EPA cannot allow flares as a reference control technology for
epoxide process vent emissions at new sources.
1.2.7 Group Determination on an Individual Process Vent Basis
for Nonepoxide Emissions from Making or Altering the
Product
At proposal, the rule required "group" determinations (to
determine whether control is required) for nonepoxide process
1-8
-------
vent emissions from making or modifying the product be made for
the combination of all the associated process vents in the PMPU.
Two commenters raised the point that the equations and other
criteria for deciding whether a vent is Group 1 or Group 2 were
based on cost-effectiveness decisions of controlling individual
process vents, and were borrowed from other rules that apply them
on an individual vent basis. The commenters requested that
owners or operators have the option of making the group
determinations for nonepoxide process vents on a vent-by-vent
basis, rather than being required to do the group determination
for the combination of all process vents.
The EPA agrees with the statement that the Group 1 criteria
is essentially a cost-effectiveness decision. The EPA also
agrees that the group determination criteria in subpart PPP were
borrowed from other MACT standards, specifically the HON (for
process vents from continuous unit operations) and Polymers and
Resins I and IV (for process vents from batch unit operations).
Finally, the EPA recognizes that in all three of the rules cited
above, the group determination has been applied to individual
process vents.
The EPA agrees that the total resource effectiveness (TRE)
index approach was developed for, and has been applied to,
individual vents. The EPA further agrees that applying the TRE
approach to the combination of process vents from continuous unit
operations in a PMPU is not appropriate without conducting an
analysis to validate the equations for the combination of vents,
or to develop new equations. Rather than take this approach, the
EPA decided to apply the Group 1 criteria for process vents from
continuous unit operations that use nonepoxide organic HAP to
make or modify the product to individual process vents.
For process vents from batch unit operations that use
nonepoxide organic HAP to make or modify the product, the Group 1
equations are the same equations employed in the Polymers and
Resins I and IV MACT standards (40 CFR 63, subparts U and JJJ,
respectively). The EPA agrees with the commenters that in the
polymers and resins standards, the Group criteria are applied to
1-9
-------
individual vents. However, unlike the TRE for process vents from
continuous unit operations, the group determination approach that
is used in subparts U, JJJ, and PPP, was originally developed to
be used for either individual vents or the combination of vents.
The original source of the batch vent group determination
approach is the EPA document "Control of Volatile Organic
Compound Emissions From Batch Processes - Alternative Control
Techniques Information Document" (EPA-453/R-94-020), i.e., the
Batch ACT. On page 7-5 of this document, the EPA states "The
control option requirements presented in Chapter 6 apply to (1)
individual batch VOC process vents to which the annual mass
emissions and average flowrate cutoffs are applied directly, and
(2) aggregated VOC process vents for which a singular annual mass
emission total and average flowrate cutoff value is calculated
and for which the option is applied across an aggregate of
sources." Therefore, for process vents from batch unit
operations, the EPA disagrees with the statements that the group
determination equations are being used "in a totally different
context" and that there is no supporting rationale for using
them. The final rule retains the requirement that the Group
criteria be applied to the nonepoxide organic HAP emissions from
the combination of process vents from batch unit operations
associated with the use of nonepoxide organic HAP to make or
modify the product.
1.2.8 Possibility of Dual Controls for Nonepoxide HAPs for
Making or Modifying the Product
The proposed rule required group determinations the
nonepoxide HAP process vent emissions from making or modifying
the product. One commenter pointed out that the proposed rule
was not clear about when and where to make this group
determination. The commenter also noted that a process vent that
uses a control technique for epoxides only (e.g., a scrubber or
ECO) would require a second control technique for the nonepoxide
HAP emissions.
The EPA considered the commenter's points and the options
suggested by the commenter. The final rule requires that the
1-10
-------
group determination for nonepoxide HAP emissions be made after
the stream has been controlled for the epoxide emissions. The
EPA believes that this approach addresses the situation regarding
the possibility of dual control. If the epoxide control device
also reduces nonepoxide emissions, then that control would impact
whether the vent (or group of batch vents) is Group 1.
Therefore, control of nonepoxide emissions along with the
epoxides will impact whether controls are required at all. If
the vent (or group of vents) still has sufficient nonepoxide
organic HAP emissions after the epoxide control device to satisfy
the Group 1 criteria, the EPA does not believe it is unreasonable
to require an additional control device to achieve the specified
percent reduction of the nonepoxide emissions.
1.2.9 Worst-Case Testing Requirements
The proposed rule required that performance tests for
process vents be conducted during worst-case operating conditions
for the process. Four commenters requested that this requirement
be deleted from the rule.
Worst-case testing requirements were not deleted from the
final rule, but were revised. The EPA's reason for requiring
compliance testing under worst case conditions is so that the
reduction efficiency of the control device is documented under
the most challenging conditions for that control device,
especially since commenters noted how difficult it is to
represent a typical venting episode. The phrase "worst-case" in
the proposed rule referred to the operating conditions of the
process (or PMPU). The worst-case testing requirement has been
revised to require testing during the worst-case conditions with
respect to the combustion, recovery, or recapture (i.e., control)
device.
Presumably, the control device should function as well or
better under conditions that are not as challenging. By revising
the rule to require testing during the worst-case conditions with
respect to the control device, continuous monitoring of operating
parameters established during the test provides a reasonable
1-11
-------
measure of continuous compliance with the efficiency requirement
under all conditions.
The commenters asserted that there is no obvious
technological difference that would require a different approach
to performance testing in this rule as from other regulations
have allowed performance tests during representative operating
conditions. The EPA disagrees with the commenters' rationale.
The EPA believes that there are obvious technological differences
from the polyether polyols industry to industries previously
regulated (particularly SOCMI type industries) since polyether
polyols are produced on a batch basis. There is much more
variance in the process vent parameters (i.e., flow and
concentration) for process vent streams from batch unit
operations, compared to process vents from continuous unit
operations. In fact, this point was stressed by commenters. The
EPA believes that it is more appropriate to compare the
requirements of this rule with other rules that also regulate
industries that operate on a batch basis. For this rule the EPA
not only compared the worst-case testing conditions with other
rules regulating batch processes, but adopted similar language to
that which is used in the Pharmaceutical Production NESHAP (40
CFR 63, subpart GGG).
The EPA would like to clarify a misconception related to
these worst-case testing provisions. It is not the intent that
production schedules be significantly altered, or that
impractical scenarios be created for testing that would never
occur in actual production. In other words, the EPA intends that
testing be conducted for the worst-case situation that can
reasonably be expected to occur during normal production. In
order to clarify this intent, the EPA has added language in
§63.1438, the general testing section of the rule. This new
language specifies that absolute worst case testing conditions
does not include situations that could cause damage to equipment,
situations that necessitate that the owner or operator make
product that does not meet an existing specification for sale to
1-12
-------
a customer, or situations that necessitate that the owner or
operator make product in excess of demand.
The added language in §63.1438 also specifies the time
period in which the worst-case conditions are to be determined.
This time period is either the 6-month period that ends 2 months
before the Notification of Compliance Status is due, or the 6-
month period that begins 3 months before the performance test and
ends 3 months after the performance test. By limiting the worst-
case conditions to one of these 6-month periods, the rule
eliminates the need for an owner or operator to consider endless
possible production scenarios, and allows them to focus on those
production scenarios in the 6-month period selected by the owner
or operator.
In conclusion, the EPA believes that requiring that
performance tests for process vents from batch unit operations
during absolute worst-case conditions is necessary to ensure that
the emission limitations in the rule are achieved. The EPA also
believes that, with the modifications to the rule made after
proposal, that the worst case provisions are reasonable and
workable for the polyether polyols industry.
1.2.10 Engineering Calculations as an Alternative to
Performance Testing
Three commenters voiced concern over the feasibility,
accuracy, expense, and safety of measuring emissions from process
vents from batch unit operations. The commenters stated that a
performance test on these short duration, variable vents is
likely to be very inaccurate and potentially dangerous. Two of
the commenters recommended that a material balance based on
common engineering calculations should be allowed in the final
rule as a compliance demonstration option. The commenters stated
that engineering calculations would provide a more accurate, less
costly, and significantly safer means to verify compliance.
The EPA recognizes that there are issues related to the
feasibility, accuracy, and expense of testing process vents from
batch unit operations. The EPA would refer readers to Section
7.3 of EPA's "Control of Volatile Organic Compound Emissions from
1-13
-------
Batch Processes - Alternative Control Techniques Information
Document" EPA-453/R-94-020 for a detailed discussion of these
issues. However, the EPA does believe that accurate emission
tests can be conducted for these process vents.
One reason that the EPA has historically required
performance testing for control devices that reduce emissions
from process vents, when engineering analyses is allowed for
other emission sources (such as storage vessels), is that
emissions from process vents are typically significantly larger
than those from other emission sources. When emissions are
larger, the EPA believes that it is important that the
effectiveness of the control device be accurately determined by a
performance test.
Given that the magnitude of the emissions was a part of the
basis for requiring performance tests, the EPA believes that it
is reasonable to allow an alternative to performance testing for
a process vent control device if emissions being routed to the
device are comparable to the emissions that would be vented to
control devices for other emission sources for which performance
tests are not required. Therefore, the EPA decided that
engineering assessments could be allowed in lieu of performance
testing for "small" control devices that reduce HAP emissions
from process vents. For the Pharmaceutical Production NESHAP,
the EPA also determined that it was appropriate to allow
engineering calculations as an alternative to performance testing
for small control devices, where a small control device is
defined as one with uncontrolled annual HAP emissions of less
than 10 tons per year. The EPA believes that this level of
uncontrolled emissions is also appropriate to define a small
control device for the polyether polyols industry. Therefore,
the final rule allows the use of a design evaluation instead of a
performance test if the control device receives less than 10 tons
per year uncontrolled emissions from one or more PMPUs.
The exemption from performance testing for small control
devices discussed above should help to alleviate some of the
concerns raised by the commenters. Many of the concerns related
1-14
-------
to the feasibility, accuracy, and expense of testing these batch
vents are due to the short duration, variable nature of batch
venting episodes. The EPA believes that if a control device
receives more than 10 tons per year of uncontrolled HAP
emissions, it is likely that the vent streams being routed to the
device are of longer duration and less variable, thus making it
easier to conduct the performance test.
However, the EPA also recognizes that the small control
device exemption will not totally eliminate the concerns raised
by the commenters. Therefore, the EPA made other changes to the
testing requirements to address potential problems related to the
testing of batch process vents, which are briefly discussed
below.
Since batch emission episodes can be less than one hour, the
rule was changed to specify that test runs be conducted for the
complete duration of the batch venting episode or one hour,
whichever is less. Other references to one-hour periods were
also removed.
The proposed rule required the use of Method 1 or 1A to
select sampling sites. Commenters claimed that, in many
instances, neither method would be appropriate for the batch vent
streams. The rule was restructured by separating the paragraph
addressing the use of Method 1 or 1A for sample or velocity
traverses from the paragraphs specifying the sampling site
location. In other words, if the owner or operator conducts a
sample or velocity traverse, the final rule requires that Method
1 or 1A be used. However, it does not require that these methods
be used to select sampling sites.
With regard to the safety issue, the final rule states that,
in cases where it is imperative to limit any leakage of emissions
into the work atmosphere, a sampling port with a double seal
should be installed so that the probe can be inserted and removed
without any leakage of exhaust gas into the work atmosphere.
Further, the final rule requires that permanent sampling ports be
installed at the inlet to the control device during a period when
it is most convenient (or least disruptive) to shut the process
1-15
-------
down (e.g., during a scheduled maintenance outage). In addition
to these specific requirements, a general requirement was added
that allows owners or operators to eliminate potential testing
scenarios if the test could create a situation which could cause
plant or testing personnel to be subject to unsafe conditions.
In conclusion, the EPA acknowledges that issues exist with
regard to the testing of emissions from batch process units.
Changes have been made to the final rule to address these issues.
However, the Agency maintains that numerous other industries that
utilize batch processes are regulated by MACT standards, and are
able to conduct performance tests. The EPA believes that the
commenters did not provide sufficient rationale why the polyether
polyols industry presents unique testing problems that are not
present in these other industries that utilize batch processes.
Therefore, the final rule requires that control devices that
receive more than 10 tons per year of uncontrolled organic HAP
emissions conduct tests to demonstrate control device
performance.
1.2.11 Revisions to the Test Method Recruirement for Control
Efficiency Determination
The proposed rule required Test Method 18, or any other
method or data that has been validated according to Method 301
for control efficiency determinations. Three commenters noted
that this requirement was inconsistent with the test methods used
by the facility whose data established the new source MACT floor
for epoxide process vent emissions (Method 25A was used). These
commenters discussed the expense of Method 301 validation, and
noted that the proposed rule relied on Method 25A in other parts
of the rule (for wastewater) , and that other rules (such as the
Polymers and Resins IV rule) allowed Method 25A without Method
301 validation.
The EPA agrees that allowing of the use of Method 25A would
provide more flexibility, and potentially provide the opportunity
for less costly testing. However, the EPA believes that Method
25A should be used only after an accurate response factor has
been determined. The importance of calibrating a flame
1-16
-------
ionization detector (FID) reading obtained using Method 25A with
respect to a certain compound (adjustment by response factor)
depends on how the Method will be used to demonstrate compliance
with the standard. In general, the EPA believes that an accurate
response factor is necessary in cases where Method 25A is used to
demonstrate control efficiency across a device where the
composition of the stream may change, or in situations where
multiple components, including non-HAP VOCs, are present.
Because the relative proportion of organic compounds may change
across the control device, appropriate response factors are
needed to accurately quantify TOC at the inlet and outlet of a
control device. In addition, the EPA believes that owners and
operators should have the opportunity to demonstrate compliance
at the outlet of a control device by measuring 20 ppmv TOC or
less. Therefore, the final rule does allow the use of Method 25A
under certain conditions. The following describes the choices of
test methods allowed in the final rule: (1) Method 18 to
determine HAP concentration in any control device efficiency
determination; (2) Method 25 to determine total gaseous
nonmethane organic concentration for control efficiency
determinations in combustion devices; (3) Method 25A to determine
the HAP or TOC concentration for control device efficiency
determinations under the conditions specified in Method 25 of
appendix A of part 60 for direct measurement of an effluent with
a flame ionization detector, or in demonstrating compliance with
the 20 ppmv TOC outlet standard.
1.2.12 Flexibility of the Determination of a Site-specific
Onset of Extended Cookout
The proposed rule defined the onset of the extended cookout
(ECO) as the point when the epoxide concentration in the liquid
is equal to 25 percent of the concentration of epoxide in the
liquid at the end of the epoxide feed. In addition to using this
"default" definition of the ECO onset, the proposed rule allowed
owners and operators the option of defining the onset of the ECO
for their specific process, at a point other than when the
reactor epoxide partial pressure equals 25 percent of the reactor
1-17
-------
partial pressure at the end of the epoxide feed. The factors in
setting a site-specific ECO onset were the profit variable margin
(the difference between variable costs (raw materials and energy)
of the product) and the cost of the raw material. One commenter
objected to allowing the establishment of a site-specific ECO
onset based on economics, stating that economics can be
subjective, making it easy to demonstrate a 98-percent emission
reduction.
A late submittal from one commenter refuted the first
commenter's argument that the onset of ECO is subjective, stating
that one of the pieces of economic information, the price of the
raw material, comes from the Chemical Market Reporter. However,
the other variable in defining the onset of ECO, the product
variable margin and the selling price, was the variable that
provoked the original commenter's concern. In fact, the
commenter providing the late comment stated that the product
variable margin has "a much stronger correlation between product
profitability and the economic onset of ECO."
Due to the subjectivity of the product variable margin, and
the strong correlation between the product variable margin and
the ECO onset, the EPA agreed with the first commenter.
Therefore, the EPA revised the final rule, and the determination
of a site-specific ECO onset is not allowed.
1.2.13 Parameter Monitoring Excursion Definitions
In reviewing the sections associated with parameter
monitoring excursions as a result of public comments, the EPA
decided to restructure and expand these provisions in order to
simplify and clarify these provisions in subpart PPP.
Basically, there are two ways an excursion can occur. The
first is if the average parameter value measured is above a
maximum, or below a minimum, established value. The second is if
insufficient monitoring data are collected. The proposed rule
had a definition of an excursion for each of these situations.
However, the EPA realized that the proposed definitions of
excursions were not well suited to emission points where
1-18
-------
intermittent venting episodes occur, such as storage vessels that
vent only when being filled and process vents from batch unit
operations. Therefore, paragraphs were added to the final rule
to define excursions for these situations.
Also, the EPA realized that while excursions for owners or
operators using extended cookout to comply with the epoxide
emission limitations were defined in the extended cookout section
of the rule, a reference to those definitions was needed in the
parameter monitoring levels and excursions section (§63.1438).
Therefore, another paragraph was added referring to the excursion
definitions for ECO.
With regard to calculating averages, §63.1439(d)(7) of the
final rule specifies that monitoring data collected during
periods of monitoring system breakdowns, repairs, calibration
checks, and zero (low-level) and high-level adjustments; start-
ups; shutdowns; malfunctions; and periods of non-operation of the
affected source that result in the cessation of emissions to
which the monitoring applies are not to be included when
calculating any average.
Language has also been added to §63.1438(f) to clarify when
monitoring data are insufficient. An excursion due to
insufficient monitoring data occurs if measured values are
unavailable for a specified percentage of time the control device
is in operation. First, the rule now clairifies the situations
that cause measured values to be unavailable: monitoring system
breakdowns, repairs, calibration checks, or zero (low-level) and
high-level adjustments. Second, the final rule clarifies that
periods of start-ups; shutdowns; malfunctions; and periods of
non-operation of the affected source that result in the cessation
of emissions to which the monitoring applies are not to be
included in defining the period of control device operation.
Finally, commenters requested that the EPA clarify the
appropriate use of parameter monitoring data for enforcement
purposes. Paragraph §63.1438(e) of the final rule has been
rewritten to add specificity regarding what the owner or operator
is out of compliance with when an excursion occurs (that is not
1-19
-------
an excused excursion). If an organic monitoring device is used
to monitor HAP or TOC concentration at the outlet of a recovery
or recapture device, the final rule clarifies that each excursion
where the daily average value of monitored parameters is above
the maximum, or below the minimum established parameter level,
represents a violation of the emission limit. Similarly, an
excursion where the daily average temperature is above the
maximum established temperature for a condenser represents a
violation of the emission limit. Other excursions where average
values are above the maximum, or below the minimum established
parameters represent violations of the operating limit, rather
than violations of the emission limit. Also, excursions due to
insufficient monitoring data are violations of the operating
limit.
1.2.14 Start-up, Shutdown, and Malfunction Plan
The proposed rule required that monitoring data be collected
during periods of SSM. Commenters requested that the EPA allow a
provision for ceasing to collect monitoring data at a particular
control device if operating that monitoring device during periods
of SSM would damage the monitoring device. The EPA revised the
final rule to allow the owner or operator to cease collecting
monitoring data if the owner or operator has illustrated that the
monitoring device would be damaged or destroyed if it were not
shut down during the SSM period. Such a provision must be
included in the Start-up, Shutdown, and Malfunction Plan.
Getting such a provision in the Start-up, Shutdown, and
Malfunction Plan is accomplished by submitting a request, and
rationale defending the request, in the Precompliance Report or
in a supplement to the Precompliance Report.
1-20
-------
2.0 SUMMARY OF PUBLIC COMMENTS AND RESPONSES
A total of twelve letters commenting on the proposed
standards and the supplementary information document (SID) for
the proposed standards were received. A public hearing was not
requested and, therefore, none was held. A list of the
commenters, their affiliations, and the EPA docket number
assigned to their correspondence is given in Table 2-1. .
For the purpose of orderly presentation, the comments have
been categorized under the following topics:
Applicability
Compliance and Relationship to Other Rules
Definitions
Process Vents Control Requirements
Extended Cookout as a Control Option
Process Vent Monitoring Requirements
Group Determination for Nonepoxide HAP Emissions
Emission Factor
Storage Vessel
Wastewater
Equipment Leak Provisions
Testing Requirements
Parametric Monitoring
General Recordkeeping and Reporting
Editorial Comments
Legal Considerations
In the comment summaries and responses contained in the
following sections, when a change to specific rule language is
discussed, the new language is represented by underlining, while
the text that the commenter recommended removing or the EPA has
decided to remove is represented in strikeout font.
2-1
-------
2.1 APPLICABILITY
2.1.1 PMPUs Without Organic HAP
Comment: Three commenters (IV-D-04, IV-D-05, and IV-D-07)
stated that PMPUs that do not use or manufacture any organic HAP
should comply with 63.1420(b)(1) or (b)(2), not (b)(1) and
(b)(2). The equivalent of (b)(2) originated in the HON as a
settlement provision. According to the commenters, its intent is
to provide an alternative to (b) (1) , so that if an owner or
operator (some years after the compliance date) cannot find the
original documentation that a process is exempt, this does not
result in civil or criminal penalties for noncompliance with a
recordkeeping requirement even though the process really does
qualify for the exemption. According to the commenters, this
approach also allows facilities to reduce the recordkeeping
burdens by choosing not to keep records perpetually, so long as
they accept the duty to demonstrate exemption on demand.
Response: The EPA agrees that §63.1420(b) should allow
owners or operators at PMPU's without organic HAP to comply with
either §63 .1420 (b) (1) or §63 .1420 (b) (2) , and the EPA has made the
suggested changes.
Comment: One commenter {IV-D-05) supported the exclusion of
PMPUs without organic HAP from applicability to the rule. To
clarify when a facility may qualify as not manufacturing or using
an organic HAP, the commenter recommended adding a new sentence
after the first sentence in §63.1420(b), as follows:
"A facility does not use or manufacture organic HAP if (a) a
HAP is used for reasons other than making the product, such as,
but not limited to, occasional cleaning of parts, or in paint
that is applied to components of the facility, or (b) a HAP is
produced as a trace impurity."
Commenter (IV-D-09) suggested that the phrase "or contains
organic HAP as impurities only" be added for clarity so that
§63.1420 (b) would read: "The owner or operator of a PMPU that is
part of an affected source, as defined in paragraph (a) of this
2-2
-------
section, that does not use or manufacture any organic HAP, or
contains organic HAP as impurities only, shall comply with
paragraphs (b)(l) and (b)(2) of this section...." The commenter
stated that this change would also be consistent with
§63.1420(c)(8), which exempts "Vessels and equipment storing
and/or handling material that contains no organic HAP or organic
HAP as impurities only."
Response: The EPA does not feel that the new sentence that
the commenter requested adding to §63.1420(b) is necessary. The
EPA believes that the requirement to retain information to
document the basis for the determination that the PMPU does not
use any organic HAP, in §63.1420(b)(1), will verify to
enforcement personnel that even when a PMPU uses HAP in cleaning
supplies, or produces a product that contains a HAP as a trace
impurity, the PMPU still fulfills the category of "PMPUs without
organic HAP." Further, the EPA agrees with Commenter IV-D-09's
suggestion to clarify §63.1420(b) by addressing the fact that
HAP may be present as impurities only; however, the wording
suggested by the commenter is not consistent with the section.
Therefore, the EPA revised §63.1420(b) in the final rule to read
as follows:
"(b) PMPUs without organic HAP. The owner or operator of a
PMPU that is part of an affected source, as defined in
paragraph (a) of this section, but that does not use or
manufacture any organic HAP shall comply with the
requirements of either paragraph (b)(1) andor (b)(2) of this
section. Such a PMPU is not subject to any other provisions
of this subpart and the owner or operator is not required to
comply with the provisions of subpart A of this part.
Products or raw material(s) containing organic HAP as
impurities only are not considered organic HAP for the
purposes of this paragraph."
2.1.2 Definition of Affected Source
Comment: One commenter (IV-D-05) stated that §63.1420
(a)(2), (a)(3), and (a)(4)(i) refer to emission points and
equipment "associated with each group" of PMPUs in an affected
source. This implies there could be more than one group.
2-3
-------
Instead, the commenter suggested this section should read
"associated with the group" of PMPUs in an affected source.
There is only one group, and only one affected source, at a site.
Therefore, the commenter recommended changing the text in
§63.1420(a)(2) as follows:
"Emission points and equipment. The affected source also
includes the emission points and equipment specified in
paragraphs (a)(2)(i) through (a)(2)(iv)(vi) of this section that
are associated with the each group of polyether polyol
manufacturing process units (PMPU) making up an affected source,
as defined in §63.1423."
The commenter requested that similar changes also be made in
§63.1420(a)(3) and (a)(4)(i).
Response: The EPA agreed with the commenter, and has
replaced the term "each group" with the term "the group,"
throughout §63.1420(a). In addition, to clarify that the
equipment that was listed in §63.1420(a)(2) at proposal is part
of the affected source, but separate from the PMPU, the list
proposed under §63.1420(a)(2) has been moved to §63.1420(a)(4),
the other subparagraphs in §63.1420(a) have been re-lettered
accordingly, and, in the final rule, these paragraphs refer to
"the group of one or more PMPU and associated equipment, as
listed in paragraph (a) (4) of this section," instead of just
referring to the "group of one or more PMPU," as was done at
proposal.
As a point of clarification, the commenter is incorrect in
their assertion that there can only be one affected source at a
plant site. Under most circumstances, this would be the case.
However, consider a situation where a new PMPU with potential HAP
emissions greater than 10/25 tons per year is added to a plant
site that already had an existing source consisting of a group of
PMPUs. In accordance with §63.1420(g), the newly added PMPU would
be a new affected source. Therefore, there would be two affected
sources at the plant site, a new affected source and an existing
affected source.
2-4
-------
2.1.3 Processes Exempted from Affected Sources
Comment: One commenter (IV-D-05) requested that EPA delete
the words "separate entities and" in §63.1420(d)(2). According
to the commenter, they are not necessary, and they are
potentially inaccurate. The term "entity" is typically
interpreted as meaning a separate company, which the commenter
believes was not the EPA's intent.
Response: The EPA agrees with this comment and has deleted
the words as suggested. In the final rule, §63.1420(d)(2) reads
"...(TSDF) requiring a permit under 40 CFR part 270 that are
separate entities and not part of a PMPU...."
Comment: One commenter (IV-D-05) stated that §63.1420(d)(3)
exempts reactions and processing that occur "after the
manufacture of polyether polyol products." The commenter agrees
that once the basic chemical reaction has occurred that produces
the polyol, further reaction to form derivatives should be
exempt. However, one possible interpretation of this section is
that a polyether polyol "product" does not exist until after some
of these additional steps have occurred. Therefore, for clarity,
the commenter recommended changing these provisions as follows:
"reactions or processing that occur after the manufacture of
polvcthcr polvol products epoxide polymerization is complete and
after any catalyst removal step is complete."
Response: At proposal, §63.1420(d)(3) exempted "reactions or
processing that occur after the manufacture of polyether polyol
products." The commenter's suggested language is preferable,
since it omits the use of the word "product." This eliminates
the possible interpretation that a polyether polyol "product"
does not exist until after some of the additional steps have
occurred. However, there may be more than one catalyst removal
step; therefore, the final rule reads "reactions or processing
that occur after the manufacture of polycthcr polyol products
2-5
-------
epoxide polymerization is complete and after all catalyst removal
steps are complete."
2.1.4 Necessity of Primary Product Redetermination
Comment: One commenter (IV-D-04) requested that EPA clarify
whether redetermination of the primary product is mandatory or
optional in §63.1420(e)(4)(i) of the proposed rule. The
commenter currently interprets this paragraph as allowing, but
not requiring, a redetermination of the primary product.
Response: The EPA agreed that the portion of the primary
product provisions cited by the commenter needed clarification.
In fact, the EPA conducted an overall review of the proposed
primary product provisions, and concluded that several structural
and clarifying changes were needed. In addition, the EPA noted
some potential situations that could occur that were not
addressed in the proposed provisions. The portions of the final
rule that address the commenter's specific concern are discussed
below, followed by a general summary of other changes to the
primary product determination provisions.
With regard to the specific concern cited by the commenter,
specifies procedures for a required annual applicability
determination (beginning five years after the promulgation of the
final rule) for non-PMPU's that have produced a polyether polyol
product at any time in the preceding 5-year period or since the
date that the unit began production of any product, whichever is
shorter. The method for performing this annual applicability
determination requires the owner or operator to calculate the
percentage of total operating time each product was produced
during the applicable time period. If a polyether polyol product
was the product with the highest percentage of total operating
time over that period, then the flexible operation unit is
designated as a PMPU.
If a process unit has not produced any polyether polyol in
the previous five-year period, but the owner or operator
2-6
-------
anticipates that their non-PMPU will begin manufacturing a
polyether polyol product in the near future, the provisions in
paragraph 63.1420(e)(4) in the final rule apply. This provision
basically requires the owner or operator to redetermine the
primary product for the process unit based on their prediction on
the anticipated production for the five years (or one year, for
new process units) following the date that production of a
polyether polyol will be initiated.
The proposed provisions required that the initial primary
product determination be based on a five-year prediction of
anticipated production by the owner or operator. The EPA was
made aware that, in some instances, the owner or operator may not
be able to make such a prediction. Clarifications and/or
revisions were made to the primary product provisions to address
this situation. First, in the initial determination, the time
frame for which production must be*anticipated for new process
units was changed to one year. Also, provisions were added for
owners or operators that cannot determine the primary product
based on anticipated 5-year (or 1-year) production. First, the
process unit is not a PMPU if the owner or operator cannot
determine the primary product, but can determine that the primary
product is not polyether polyol. If the owner or operator cannot
determine a primary product, and cannot determine that polyether
polyol is not the primary product, then the process unit is
designated as a PMPU and subject to the existing source
provisions of subpart PPP if polyether polyols have been produced
in an existing process unit for 5 percent or greater of the time
since September 4, 1997. The EPA believes that if production is
so uncertain that an owner or operator cannot determine the
primary product based on future production, the fact that
polyether polyols have been produced in the unit since proposal
of the rule for even a small amount of time is sufficient basis
for having the process unit be subject to subpart PPP. For new
process units, if polyether polyols will be produced at any time
2-7
-------
during the first year of production, then the unit is a PMPU and
subject to the new source provisions of subpart PPP.
In addition to the provisions discussed above that specify
how non-PMPUs are to determine if they become PMPUs and subject
to subpart PPP, the EPA has also clarified and expanded the
provisions that specify how the PMPU designation can be removed
from a process unit. The first case, which is retained from the
proposed rule, is where production of polyether polyols ceases
and the owner or operator does not anticipate the production of
polyether polyols in the future. Also, the EPA has added
provisions that specify procedures for a primary product re-
evaluation based on actual production. If an owner or operator
of a PMPU finds that another product has been produced for a
greater amount of time than polyether polyols over a specified
time period (previous five years or since beginning the
production of polyether polyols), then the PMPU designation would
possibly be removed. The stipulation is that production of the
"new" primary product must make the process unit subject to
another part 63 NESHAP. If the new primary product is not
subject to another part 63 NESHAP and polyether polyols continue
to be produced, the process unit continues to be classified as a
PMPU and continues to be subject to subpart PPP.
The EPA has also added provisions addressing the
determination of the primary product in situations where two or
more products are produced simultaneously. Also, clarifications
were made in the reporting and recordkeeping requirements
associated with the primary product determination.
Comment: Two commenters (IV-D-04, IV-D-05) stated that
§63.1420(e)(4)(ii) of the proposed rule applies "if a process
unit meets the criteria of paragraph (e) (4) (i) . . . . " However,
Commenter IV-D-04 maintained that it is literally impossible for
any individual process unit to meet all of the criteria of
paragraph (e)(4)(i), because that paragraph has two opposite
kinds of criteria. It has criteria for process units that are
subject to the rule, and criteria for process units that are not
2-8
-------
subject to the rule. So, in order to meet "the criteria" of
paragraph (e)(4)(i), a process unit would have to be both subject
to the rule, and not subject to the rule. To clarify this
situation, the coiranenters recommended changing the text of
paragraph (e)(4)(ii) as follows:
"If a process unit mcota the criteria of paragraph—(c) (4) (i)
of this acctionis subject to this subpart,..."
Response: As mentioned in the response to the previous
comment, the EPA has largely re-written and re-structured
§63.1420(e), to resolve problems such as the one pointed out by
the coiranenters (IV-D-04 and IV-D-05). The paragraph referred to
by the coiranenters [§63.1420(e) (4) (ii)] no longer contains the
language that the commenters were concerned about, and conflicts
such as the one mentioned by the commenters should not occur in
the final version of §63.1420(e).
2.1.5 Primary Product Determination and Applicability
Comment: One commenter (IV-D-05) maintained that the
sentence in §63.1420(e) that reads "Paragraphs (e)(3) through
(e)(4) of this section describe whether or not a process unit is
subject to this subpart" is inaccurate. The commenter stated
that the referenced paragraphs merely describe the relevance of
the primary product determination to whether or not a process
unit is subject to this subpart. Therefore, for clarity, the
commenter recommended changing this sentence, as follows:
"A process unit may be subject to the requirements of this
subpart if it meets the criteria of paragraphsParaqrapha—(c) (3)
through (e)(4) or (e)(5) of this section dcacribc whether or not
a proccaa unit ia oubjcct to thia aubpart."
Response: As discussed earlier, the EPA has re-structured
§63.1420(e), and the language that the commenter objected to no
longer exists in the final rule. The language in the final
version of §63.1420(e) is much more explicit about the roles of
2-9
-------
the different sub-paragraphs under §63.1420(e) than the proposed
version was, and should cause no similar confusion.
2.1.6 Primary Product Determination for Non-Flexible
Operation Units
Comment: One commenter (IV-D-05) requested that a paragraph
be added between §§63.1420(e)(1) and (e)(2), telling how to
determine the primary product if (a) the unit is not a flexible
operation unit, and (b) the unit produces two or more products.
According to the commenter, the HON and similar rules have such a
provision. There is simply no way to tell what the primary
product is if a non-flexible operation unit has two or more
"product" streams simultaneously. Therefore, the commenter
recommended making this provision consistent with the HON rule
(§63.100(d)(1) and (2)) by inserting the following new subsection
and renumbering the subsequent clauses accordingly:
"(e)(2) If a process unit is not designed and operated as a
flexible operation unit and the unit produces more than one
intended product, the product with the greatest annual design
capacity on a mass basis represents the primary product of the
process unit."
Response: The EPA agrees that §63.1420(e) needed to include
a provision for process units that are not flexible operation
units but that produce two or more products simultaneously.
Under the newly re-structured §63.1420(e), §63.1420(e) (1) (ii)
provides provisions addressing such a situation. Although the
commenter's recommended change was not incorporated exactly as
suggested, the EPA believes that the changes to §63.1420(e) will
alleviate the commenter's concern.
2.1.7 Applicability for Flexible Operation Units
Comment; One commenter (IV-D-07) referred to the provisions
in §63.1420 and §63.1423 of the proposed rule which flexible
operation units must use to determine whether they are covered by
the rule. The commenter requested that EPA provide the option in
2-10
-------
the final rule for flexible operation units that do not
technically qualify as PMPUs to be able to "opt in" to the rule
if it makes administrative sense to the owner/operator to do so.
For example, the commenter might prefer to "opt in" to subpart
PPP rather than be subject to a future MACT standard such as the
Miscellaneous Organic NESHAP (MON), so as to have only one
standard to meet in the administrative unit. The commenter added
that the flexible operation unit concept is workable and should
be retained.
Response: The EPA appreciates the commenter's desire to be
subject to only one MACT standard; however, the change requested
by the commenter (where a flexible operation unit that is not a
PMPU, could opt into the Polyether Polyols NESHAP because the
non-PMPU flexible operation unit is in the same "administrative
unit" as the facility's other PMPUs) is not being granted. While
the EPA recognizes that the owner/operator would benefit from
greater ease of compliance, it is the EPA's position that an
affected source should be defined by the products that it makes,
and not solely by the products made near it. Defining the
affected source solely based on the owner or operator's choice of
standards would allow a source that might otherwise be subject to
a more stringent rule to "opt in" to subpart PPP. However, in
the final rule, provisions have been added [e.g.,
§63.1420(e) (10) (iii)] that ensure that a source that is no longer
making a polyether polyol as its primary product will not be
subject to subpart PPP as well as to another MACT standard at the
same time.
2.1.8 Requirements for Flexible Operation Units when
Producing Non-polvether Polvols
Comment: One commenter (IV-D-07) expressed concern that the
proposed rule could be interpreted as requiring installation of
additional controls for periods when the flexible operation unit
is producing a product other than the primary product and
believes that clarification is needed to confirm that this is not
2-11
-------
a valid interpretation. In order to clarify the compliance
concern, the commenter suggested that the following sentence be
added as a new §63.1420(e)(7)(iv):
"So long as the owner or operator of a flexible operating
unit is able to demonstrate compliance during the production of
the primary product, then no additional control device, recovery
device, and/or recapture device is required to be installed (to
otherwise demonstrate compliance) for periods when the flexible
operating unit is producing a product other than the primary
product."
The commenter stressed that similar wording was added in the
HON (§63.103(b)(6)).
Response: The EPA clearly did not consider the option that
flexible unit operations could be required to install additional
controls for periods when the flexible operation unit is
producing a product other than the primary product. The addition
suggested by the commenter has no effect other than to warrant
against this possibility. Therefore, the sentence was added as a
new paragraph under §63.1420(e)(5), as §63.1420(e)(5)(iii),
rather than at §63.1420(e)(7)(iv) as the commenter suggested, due
to the general restructuring of §63.1420(e). The language has
been edited slightly from that suggested by the commenter. The
new language in §63.1420(e)(5)(iii) reads:
(iii) So long as the owner or operator of a flexible
operation unit is able to demonstrate compliance with this
subpart during the production of polvether polyols, then no
additional combustion device, recovery device, and/or
recapture device is required to be installed (to otherwise
demonstrate compliance) for periods when the flexible
operation unit is producing a product other than a polyether
polvol. However, while a product other than polyether
polvol is being produced, the owner or operator shall
continue to operate any existing combustion, recovery,
and/or recapture devices that are recruired for compliance
during production of the primary product. If extended
cookout (ECO) is the control technique chosen for epoxide
emission reduction, then ECO or a control techniorue
providing an equivalent reduction in epoxide emissions
should continue to be used for epoxide emission reduction,
if the non-polvether polvol being produced uses epoxide
2-12
-------
monomers. If ECO is used, the parameter monitored for ECO
shall be averaged for all product classes produced. The ECO
for non-polvether polvol production shall be performed so
that the averaged parameter is maintained when the ECO is
used as a control technique during the production of
nonpolvether polvols.
2.1.9 NonPMPUs that Produce Polvether Polvols
Comment: One commenter (IV-D-05) expressed concern that the
last sentence in §63.1420(e)(6) imposes a recordkeeping
requirement on all processes that are determined not to be PMPUs.
The commenter suggested two changes to ease the burden on sources
that are not intended to be covered by the rule. First, this
sentence should apply only to processes that produce some
polyether polyol product. For example, a source should not have
to keep records proving that a dedicated chlorine plant,
magnesium plant, or methyl cellulose plant is not a PMPU.
Second, there should be two options: keep records, or document,
on demand, that the process is not a PMPU. That way, if a
facility does not have the records available many years later,
but the process really is not a PMPU, there is no violation. The
commenter believed these changes would be consistent with EPA's
intent. Therefore, the commenter recommended changing the text
as follows:
"For a process unit that produces some polvether polvol but
If the primary product of that process unit is determined to be
something other than a polyether polyol product, the owner or
operator shall retain information, data, and analysis used to
document the basis for the determination that the primary product
is not a polyether polyol product or must be able to document
that the process unit is not a PMPU."
Response: The suggestion that the owner or operator of a
process unit dedicated to the production of a product other than
a polyether polyol would have to maintain records is a
misinterpretation of proposed §63.1420(e)(6), since that
paragraph addressed flexible operation units, not dedicated
2-13
-------
operation units. In the final rule, a flexible operation unit is
defined as a process unit that manufactures different chemical
products in addition to polyether polyols, by periodically
alternating raw materials fed to the process unit or operating
conditions at the process unit.
The commenter's second request was that owners or operators
of flexible operation units with a primary product that is not a
polyether polyol have the option of keeping records documenting
the fact that the process unit is not a PMPU, or of providing
such documentation "on demand," to the Administrator, if
requested to do so. The EPA agrees that this is a reasonable
request, and has incorporated this type of provision into the
restructured §63.1420(e), which appears in the final rule. The
proposed paragraph §63.1420(e)(6) no longer exists, and
§63.1420(e)(8) in the final rule provides the "document on
demand" option that the commenter requested.
2.1.10 Changes or Additions to Plant Sites
Comment: One commenter (IV-D-07) stated that
§63.1420(g)(1)(i)(D) indicates that if a new PMPU is added to a
site where a polyether polyol is not currently produced as the
primary product of an affected source (and the plant site is a
major source either before or after the addition), the unit will
be subject to new source MACT regardless of the emissions from
the new unit. The commenter claimed that this was an arbitrary
criterion for first time addition of a PMPU to an integrated
complex and it removes substantial pollution prevention incentive
to keep new units under the 10/25 criteria and makes no sense
from a policy standpoint. The commenter suggested that this
paragraph be deleted in the final rule.
Response: The EPA disagrees with the statement that the
proposed provisions related to the first-time addition of a PMPU
to a plant site are arbitrary. Rather, these provisions are
consistent with the Clean Air Act and the General Provisions of
part 63 (40 CFR part 63, subpart A). In the Clean Air Act, a new
2-14
-------
source is defined as "a stationary source the construction or
reconstruction of which is commenced after the Administrator
first proposes regulations under this section establishing an
emission standard applicable to such source." The definition in
the General Provisions mirrors the Clean Air Act definition,
except that it uses the term "affected source" rather than
"stationary source."
If, after September 4, 1997 (proposal date of subpart PPP),
a PMPU is constructed at a major source plant site where no PMPUs
previously existed, then the new PMPU cannot be considered to be
"an addition" to an "existing" affected source, since there was no
"existing affected source" at the facility. The criteria for
whether or not one or more newly added PMPUs constitute a "new
affected source" relies on whether or not the plant site is a
"major source" (before or after the addition of the PMPU).
However, if the plant was not a major source (i.e., was an area
source) before the addition of the one or more PMPUs, and is
still not a major source after the addition of the PMPU(s), it
remains exempt from the requirements of this rule, as an area
source.
These provisions have not changed since proposal, although
the intent of the requirements has been clarified in this final
rule. The EPA believes that these provisions are both reasonable
and consistent with the intent of the Clean Air Act and the
General Provisions to part 63.
Comment: One commenter (IV-D-07) maintained that the entire
affected source should not be subject to new source MACT if
reconstructed. The impact of the proposed wording in
§63.1420(g)(2)(i) is that if a PMPU in a location makes a
substantial investment to modernize such that reconstruction is
triggered, other PMPUs at the location would be required to
install new source MACT, even though they are not undergoing
changes. The commenter suggested that the HON approach be used
under subpart PPP to avoid this impact. The wording used in the
HON (§63.100(1)(2)) states that "If any change is made to a
2-15
-------
chemical manufacturing process unit subject to this subpart, the
change shall be subject to the requirements of a new source in
subparts F, G, and H of this part...."
Response: The impact that the commenter objected to (i.e.,
that all PMPUs in the affected source become subject to the new
source MACT) is the intended impact of the provisions of
§63.1420(g)(2)(i). If activity at an existing affected source
constitutes reconstruction, the existing source becomes a new
source by definition. Reconstruction is triggered when the cost
of a modernization project exceeds 50 percent of the fixed
capital cost that would be required to construct a comparable new
affected source. In subpart PPP, the EPA defined the "affected
source" as all process units that produce polyether polyols as
their primary product at a plant site, along with all waste
management units, maintenance wastewater, heat exchange systems,
and equipment used to comply with subpart PPP that are associated
with the PMPUs. Therefore, the EPA expects that it would have to
be a substantial modernization for the cost to exceed the cost of
50 percent of the entire affected source.
To address the commenter's example, assume that there are
four PMPUs at a plant site. By definition, the group of four
PMPUs, along with the associated equipment, make up an existing
affected source. If an investment was made to modernize one
PMPU, that investment would need to exceed the cost of
constructing four new PMPUs comparable to those at the site to
trigger reconstruction. The EPA believes that it is unlikely
that the modernization of one PMPU would ever cost more than 50
percent of the construction of four new comparable PMPUs.
Therefore, all four PMPUs would remain an existing affected
source. If, however, the cost of that modernization did exceed
50 percent of the cost of four new PMPUs, the EPA believes that
it is reasonable, and in fact consistent with the intent of the
definition of reconstruction, for the entire affected source to
be subject to the new source requirements in subpart PPP.
2-16
-------
Comment: One commenter (IV-D-05) maintained that if
"reconstruction" causes "the entire affected source" to become a
new affected source, then the rule should at least clarify that
"the entire affected source" is the capital base for deciding
whether a reconstruction has occurred. To clarify when a
modification is considered reconstruction such that the process
unit becomes subject to this subpart, the commenter recommended
changing the text in §63.1420(g)(2)(i) as follows:
" (i) If any process change is made or emission
point is added to an existing affected source, the
entire affected source shall be a new affected source
and shall be subject to the requirements for a new
affected source in this subpart upon initial start-up
or by [insert date of promulgation], whichever is
later, if the process change or addition meets the
criteria specified in paragraphs (g)(2)(i)(A) through
(g) (2) (i) (B) of this section:
(A) It is a process change or addition that meets
the definition of reconstruction in §63.2 of subpart A_.
For purposes of determining whether the fixed capital
cost of the new components exceeds 50 percent of the
fixed capital cost that would be required to construct
an entire affected source, the equivalent capital cost
shall be the entire potentially affected source; and
Response: The commenter's requested change to paragraph
63.1420(g)(2)(i) clarifies the EPA's intent at proposal, which
was to be consistent with the fact that the definition of
reconstruction is based on the entire potentially affected new
source. The EPA has changed added the sentence requested by the
commenter (with minor edits) to §63.1420(g)(2)(i)(A) in the final
rule.
2.1.11 Applicability of this subpart except during periods of
start-up, shutdown, malfunction, or non-operation
Comment: One commenter (IV-D-04) requested that EPA add the
words "to the extent practical" in §63.1420(h)(3) of the final
rule. These words were added in the HON as part of a litigation
settlement amendment. Their intended purpose is to clarify the
extent to which emissions must be minimized in certain unusual
2-17
-------
situations where the standard's "normal" requirements do not
apply. If the words are not added, there will be uncertainty as
to what degree of emission reduction must be achieved.
Response: The EPA agrees with this comment, and the phrase
"to the extent practical" has been added to §63.1420(h)(4) in the
final rule [which replaces proposed paragraph §63.1420(h)(3)].
The additional language is consistent with the HON
[§63.102(a)(4)] as well as with the proposed amendments to the
Polymers and Resins I and IV NESHAP [§63.480 (j) (4) and
§63.1310 (j) (4), respectively.]
2.1.12 Storage Tank Predominant Use
Comment: One commenter (IV-D-04) suggested that EPA limit
the applicability of §63.1420(f) (8) of the proposed rule, which
requires industry to redetermine storage vessel ownership if
something happens that could reasonably change the predominant
use of the storage vessel. The commenter stated that this
assumes that the original ownership was based on "predominant
use," rather than being based on "dedicated to a single process
unit" or some other basis. It also assumes that possible future
changes in predominant use were not already taken into account in
the original ownership determination. The commenter stated that
either of these assumptions could be false and cited two examples
to support the claim. The commenter requested that EPA revise
§63.1420(f)(8) as follows:
"(8) If a storage vessel has been assigned to a process
unit that is not a PMPU on the basis of predominant use under
(f)(3) of this section, and there is a change in the utilization
of the storage vessel that could reasonably be expected to change
the predominant use, the owner or operator shall redetermine to
which process unit the storage vessel belongs by reperforming the
procedures specified in paragraphs (f)(2) through (f)(7) of this
section, as appropriate."
Response: The EPA agrees with the commenter that the
proposed criteria for when a storage vessel ownership
redetermination is required [in §63.1420(f)(8)] should be more
2-18
-------
specific. However, the EPA does not believe that the language
suggested by the commenter is the most appropriate solution. In
consideration of this issue, the EPA reexamined the circumstances
under which it is necessary to reevaluate storage vessel
ownership. The EPA concluded that it is not necessary to require
a storage vessel ownership redetermination unless the storage
vessel has begun receiving material from (or sending material to)
a process unit that was not included in the initial
determination, or has ceased to receive material from (or send
material to) a process unit that was included in the initial
determination. Therefore, the proposed §63.1420(f)(8) has been
replaced with the following paragraph:
(8) If there ia a change in the utilization of the storage
vessel that could reasonably bo expected to change the
predominant uoc,—begins receiving material from (or sending
material to) a process unit that was not included in the
initial determination, or ceases to receive material from
(or send material to) a process unit that was included in
the initial determination, the owner or operator shall
rc-dctcrminc to which procoao unit the storage voaaol bolongo
by performing the procedures cpocificd in paragraphs—(f) (2)
through—(f) (7)—of this section,—ao appropriate reevaluate
the applicability of this subpart to that storage vessel.
2.1.13 S63.1420(e)(5)(ii)
Comment: For clarity, one commenter (IV-D-05) recommended
changing paragraph 63.1420(e)(5)(ii) to read:
"Alternatively, each the owner or operator shall determine
the applicability of the provisions of this subpart (e.g., Group
status) to each emission point ...Based on these findings, the
owner or operator shall comply with the applicable requirements—
that apply at any time based on emission point characteristics at
that time, as appropriate, regardless of what product is being
produced...."
Response: The first two suggested revisions are ones the
EPA has agreed to in response to other comments, and will
certainly generalize to paragraph 63.1420(e)(5)(ii) as well. The
last revision that the commenter suggested is somewhat awkward,
and could possibly be misinterpreted. The wording "based on
2-19
-------
emission point characteristics at that time" could be construed
as meaning that the applicability status could change depending
on that day's emissions. Further, the first part of the
suggested revision, "requirements that apply at any time," is
unnecessary since "any time" is such a nebulous term, and since
the requirements have monthly compliance demonstration
requirements to prove that they are being met (at any time).
Therefore, the first two suggested revisions were incorporated
into the final rule; however, the third suggested revision was
not made.
2.1.14 Section 63 .1420(c) (1) and (c) (8) Appear to be
Duplicative
Comment: One commenter (IV-D-05) stated that paragraphs
(c)(1) and (c)(8) in §63.1420 appear to duplicate one another.
If this is the case, the commenter recommended replacing
paragraph (c)(1) with (c)(8), as follows, and renumbering the
subsequent paragraphs (c)(9) through (12):
" (1) Vessels and ecruipment storing and/or handling material
that contains no organic HAP or organic HAP as impurities
onlyEcruipmcnt that dooo not contain organic HAP and io—located at
a PMPU that ia part of an affected source."
Response: The EPA agrees that there is confusing overlap
between paragraphs §63.1420(c)(1) and (c)(8). However, the EPA
does not believe that the change suggested by the commenter
accurately communicates the intent of the two paragraphs.
Paragraph §63.1420(c)(8) of the final rule has been revised to
the following:
" (8) Vessels and equipment atorina that store and/or handle
handling material that contains no organic HAP or organic HAP as
impurities only."
2.2 COMPLIANCE AND RELATIONSHIP TO OTHER RULES
2.2.1 Time Periods
Comment: Two commenters (IV-D-04, IV-D-05) requested that a
paragraph similar to §63.100(k)(9) in subpart F of the HON be
2-20
-------
added to §63.1422 as paragraph(k). Commenter IV-D-04 provided
the following wording:
"All terms in this subpart that define a period of time for
completion of recruired tasks (e.g., daily, weekly, monthly,
quarterly, annual), unless specified otherwise in the section or
subsection that imposes the requirement, refer to the standard
calendar periods." Commenter IV-D-05 suggested adding: "...,
unless altered by mutual agreement between the owner or operator
and the Administrator in accordance with §63.1422(k)."
Commenter IV-D-04 maintained that this paragraph is
important for the following reasons: (I) it ends the questions
about whether a month means 30 rolling days, and what to do if a
month has more or less days than that; (2) it eliminates the need
for a large number of definitions; and (3) it avoids having
irreconcilable conflicts with the HON, which would arise if this
rule defined time periods differently than the HON.
Response: The EPA agrees that a paragraph based on
§63.100(k)(9) in subpart F of the HON was needed in the Polyether
Polyols NESHAP. The EPA has added language to what is
§63.1422(1) 'in the final rule, mirroring the language found in
§63.100(k)(9) in subpart F of the HON. In addition, the EPA has
added the condition ". . . , unless altered by mutual agreement
between the owner or operator and the Administrator in accordance
with paragraph (1)(1) of this section," to the end of §63.1422(1)
in the final rule.
2.2.2 Changes to the General Provisions
Comment: One commenter (IV-D-04) suggested that additional
notice and comment rulemaking would be necessary if EPA makes
future changes in the General Provisions that apply to this rule.
The commenter referred to the following statement in the proposal
preamble (62 FR 46812, col. 2): "If this subpart is promulgated
subsequent to the promulgation of the amendments to the General
Provisions, the amended General Provisions will be incorporated
into this subpart." The commenter stated that, according to the
2-21
-------
CAA's administrative procedures requirements, any amendment of
the polyether polyols rule (such as changing General Provisions
that are incorporated into the rule by reference) would require
rulemaking.
Response: It is important that owners and operators realize
that provisions in subpart PPP that cross-reference the HON, the
General Provisions, or any other regulation, refer to the most
recent, promulgated versions of those rules. When such rules are
amended, the EPA will provide an opportunity for comment on the
effect that such changes will have on standards that cross-
reference these rules. The EPA believes that this practice meets
the requirements of the Clean Air Act.
2.2.3 Cross-references with Other Subparts
Comment: One commenter (IV-D-04) requested that EPA address
the overlap between the proposed rule and other MACT standards
for wastewater and/or heat exchange systems. In many instances,
the waste management units and heat exchangers that serve PMPUs
may already be subject to another MACT standard (generally the
HON) by the compliance date for this rule. There should be no
need to comply with each rule separately (duplicating the
reporting and recordkeeping requirements) when the substantive
requirements are essentially the same. Therefore, the commenter
suggested specific regulatory language that would address overlap
between subpart PPP and other MACT standards for wastewater
and/or heat exchange systems.
Response: The EPA agrees that it is appropriate to add a
paragraph at §63.1422(k) to address instances in which
requirements from other regulations overlap for the same heat
exchange system(s) or waste management unit(s) that are subject
to subpart PPP. The language in §63.1422(k) in the final rule
states that owners and operators of affected sources that share
heat exchange systems with sources that are subject to subpart F
of part 63 or any other subpart of part 63 that references the
2-22
-------
heat exchange system requirements in §63.104 in subpart F (e.g.,
subpart U) will be considered to be in compliance with subpart
PPP for heat exchange systems, if the shared heat exchange
systems are in compliance with the heat exchange system
requirements in that other subpart. Similarly, owners and
operators of affected sources that share waste management units
with sources that are subject to subpart G of part 63 or any
other subpart of part 63 that references the waste management
unit requirements in §63.132 through §63.147 of subpart G will be
considered to be in compliance with subpart PPP for the shared
waste management unit, if the shared waste management unit is in
compliance with the waste management unit requirements in that
other subpart. This change was made to ensure that owners and
operators are not subject to multiple sets of monitoring,
recordkeeping, or reporting requirements for the same equipment
due to the type of regulatory overlap that the commenter
described.
Comment: One commenter (IV-D-07) referred to §63.1422(j) of
the proposed rule, which discusses overlap with other regulatory
requirements, including RCRA. The commenter requested that the
provision be explicitly extended to "hazardous waste tanks"
regulated under 264 Subpart CC and 265 Subpart CC. The subject
hazardous waste tank emission controls include options other than
combustion and recovery devices; for example, pressure tanks and
certain types of pressure relief devices for low-vapor pressure
tanks are allowed. The commenter asserted that duplicative
monitoring, recordkeeping and reporting requirements should not
be imposed for these systems.
Response: Section 63.1422(j) addresses overlap with other
regulations for monitoring, recordkeeping or reporting with
respect to combustion devices, recovery devices, or recapture
devices. The subjects of the monitoring, recordkeeping, and
reporting requirements in this paragraph are the combustion,
recovery or recapture devices, and not the source of the emission
2-23
-------
control. Therefore, a specific provision for the hazardous waste
tanks regulated under 40 CFR part 264 subpart AA or CC and 40 CFR
part 265 subpart AA or CC is not appropriate. Further, the
requirements in 40 CFR part 264 subpart AA or CC and 40 CFR part
265 subpart AA or CC are already specifically cited in
§63.1422(j) .
2.2.4 Interfacing with Title V Operating Permit
Comment: One commenter (IV-D-07) expressed concern that the
interface between this proposed rule and the Title V Operating
Permits program is not well defined and will result in
inadvertent compliance issues. For example, in a number of
sections of the proposal, reference is made to the need to
incorporate certain items into the operating permit or permit
application. Then in §63.1439(e)(8), the information to be
included in the application is limited to the information listed
in §63.1439(e)(4). The commenter suggested that EPA conduct a
search of the final rule to identify all interfaces with
Operating Permits and include an inclusive listing in the final
rule.
Response: It is not the EPA's intent to incorporate all of
the Title V Operating Permit requirements into subpart PPP. The
owner or operator will always need to consult Title V in order to
determine all of the Operating Permit requirements that pertain
to a particular affected source, because Title V requirements are
site specific. Subpart PPP simply allows some information to be
submitted in either the Operating Permit application or the
Notification of Compliance Status, and allows the owner or
operator to choose between submitting special requests (e.g., for
permission to use alternative monitoring parameters or controls)
in the Precompliance Report or in the Operating Permit
application. Information that is submitted in the Operating
Permit application will, once approved, become part of the
affected source's Title V Operating Permit. Therefore, the EPA
has not included an "inclusive listing" in the final rule of all
2-24
-------
interfaces with Operating Permits or operating permit
applications, particularly because the requirements in subpart
PPP related to Operating Permits or Operating Permit applications
are provided as alternatives to submitting information in other
reports. However, the EPA has made every effort to clarify this
distinction in the final rule.
2.2.5 Complexity of the Proposed Rule
Comment: Two commenters (IV-D-02, IV-D-03) expressed
concern regarding the complexity of the proposed rule and stated
that it would be costly for polyether polyol manufacturers to
interpret and comply with the rule, and for permitting
authorities to administer it. Commenter IV-D-03 stated that the
rule was unnecessarily complex due to frequent references to the
HON and the Polymer and Resins I and IV NESHAP. The commenter
recommended that the cited references to the HON and to the
Polymer and Resins NESHAP be replaced by the appropriate
regulatory language in the Polyether Polyols final rule.
Response: The EPA realizes that the Polyether Polyols
Production NESHAP is relatively complex. The development of this
rule began with a preliminary maximum achievable control
technology (PMACT) partnership between the EPA and industry
representatives. During the roundtable PMACT discussions,
industry representatives requested unique emission limits for
three of the emission sources: epoxide emissions from
polymerization; non-epoxide emissions from polymerization; and
non-epoxide emissions from catalyst extraction. This NESHAP
introduces extended cookout (ECO) as a new control option for
process vents. The explanation of how to measure and monitor the
effectiveness of this new control option contributes further to
the complexity of this rule. The EPA has attempted to keep the
rule brief by cross-referencing sections of the HON or subpart U
that apply to this rule. However, the EPA realizes that making
the rule shorter through cross-referencing other subparts of part
63 makes the rule more complicated. The EPA believes that cross-
2-25
-------
referencing other part 63 subparts is necessary in subpart PPP to
ensure that the source is subject to consistent requirements
across all the rules that cite other part 63 subparts.
2.3 DEFINITIONS
2.3.1 Annual Average Concentration
Comment: One commenter (IV-D-05) stated that the definition
of "annual average concentration" should not reference subpart G
if the chemical lists (Table 8 and/or Table 9 compounds) differ
in this rule. To clarify, the commenter recommended revising the
text, as follows:
"Annual average concentration, as used in conjunction with
the wastewater provisions, means the flow-weighted annual average
concentration and is determined by the procedures in §63.144(b)
of subpart G, except as provided in §63.1433(a)(2)."
Response: The addition requested by the commenter cites
§63.1433(a)(2), which in turn lists all the exceptions in
applicability to §63.132 through §63.149. The EPA agrees that
this revision clarifies the meaning of this definition, and has
added the phrase that the commenter suggested to this definition
in the final rule.
2.3.2 Batch Cycle
Comment: One commenter (IV-D-05) claimed that the term
"batch cycle" does not match industry usage. To avoid confusion,
the commenter recommended changing the defined term to "batch
unit operation cycle."
Response: The definition of the term "batch cycle" clearly
states that the batch cycle means the steps that occur in a batch
unit operation. The rule then defines a batch unit operation.
The EPA does not think that the phrase "batch cycle" differs
significantly from "batch unit operation cycle," and so has
decided not to revise the phrase.
2-26
-------
2.3.3 Cross-referencing of Definitions from Other Rules
Comment: One commenter (IV-D-04) stated that the EPA should
not just refer to the HON or the Group I Polymers and Resins rule
for definitions that need different chemical lists, equations, or
calculations in this rule. The commenter requested that
additional clarifications be made in this regard in the following
areas:
(1) The definition of "in organic hazardous air pollutant
service" is borrowed from the HON, subpart H, which is
appropriate. However, EPA should clarify that the PEPO
definition of organic HAP applies.
(2) The definitions of "process wastewater" and "process
wastewater stream" are borrowed from the HON, subpart G, which is
appropriate. However, these defined terms ultimately depend on
whether something is "wastewater." Since the HON and this rule
each have a different definition of "wastewater," EPA should
clarify which one applies.
(3) The definition of "total resource effectiveness (TRE)
index value" is borrowed from the HON. However, the TRE
definition relies on a TRE equation, and the equation was written
to derive an appropriate cost-effectiveness cutoff for control of
individual process vents. In contrast, this rule requires
determination of the TRE index value for combinations of process
vents. Ideally, EPA should allow an option to conduct TRE
determinations on each individual process vent. However, if the
concept of combinations of process vents is retained, EPA should
revise the TRE equation (and hence, the definition).
(4) The definitions of "combination of process vents that
are Group 1" and "Combination of process vents that are Group 2"
require use of §63.115 of the HON for TRE calculations, and
section 63.499(b) of subpart U for annual average flow
calculations. Both the HON and subpart U were written for
individual process vents, not for combinations of process vents.
It seems unlikely that the same methods of calculation would work
appropriately in both contexts. Another commenter (IV-D-05)
expressed the same concern.
2-27
-------
(5) The definition of "maintenance wastewater" relies on
subpart F of the HON, with some exceptions. The exceptions
apparently do not include chemical lists. The commenter believed
that the polyether polyols chemical lists would apply instead of
the HON chemical lists and requested that the definition of
Maintenance wastewater in the final rule be clarified
accordingly.
(6) The definition of "maximum true vapor pressure" relies
on subpart G of the HON, with some exceptions. The exceptions do
not include chemical lists. The commenter asked if that was
intentional and requested that the definition in the final rule
be revised to clarify that the polyether polyols chemical lists
apply.
(7) The definition of "residual" in the proposed rule is
confusing. It says that, instead of using the HON terminology
"Table 9 compounds," it uses the phrase "organic HAP listed in
Table 9 of subpart G." The commenter questioned this and stated
that if EPA really does intend to keep referring to the HON Table
9 but with different words than the HON uses, then EPA should
explain what the change is intended to accomplish. Contrarily,
if EPA intended to refer to a polyether polyols table, then the
definition should be revised accordingly.
Response: (1) and (2): The EPA agrees that clarification is
needed. In the case of clarifying the definition of "in organic
hazardous air pollutant service", "process wastewater," and
"process wastewater stream," the EPA believes that it would be
clearer and simpler to copy the definitions from the HON into
subpart PPP. Therefore, the final rule contains definitions for
these three terms.
(3): As discussed in section 2.4.4 of this document, the EPA
has revised the rule such that the TRE equation is not applied to
aggregated streams. Therefore, the TRE equation applies only to
individual process vent streams (as in the HON), and the final
rule continues to reference the HON definition for "Total
resource effectiveness (TRE) index value".
2-28
-------
(4): The concerns raised by these commenters regarding the
combination of process vents for the group determination is
addressed in section 2.4.4 of this document. To summarize, the
EPA agreed with the commenters regarding the group determination
for process vents from continuous unit operations that are
associated with the use of a nonepoxide organic HAP to make or
modify the product, and the final rule requires that these group
determinations be conducted on an individual vent basis.
However, the final rule continues to require that the group
status of process vents from batch unit operations be determined
for the combination of process vents associated with the use of a
nonepoxide organic HAP to make or modify the product.
The change noted above necessitated a change in the proposed
definitions of "combination of process vents that are Group 1"
and "combination of process vents that are Group 2." These
proposed definitions, which addressed both process vents from
continuous unit operations and process vents from batch unit
operations, were no longer appropriate for the final rule.
Therefore, the EPA defined separate terms for Group 1 and Group 2
process vents from continuous and batch unit operations.
The new terms used for process vents from batch unit
operations, which are provided below (as they appear in the final
rule), are "Group 1 combination of batch process vents" and "Group
2 combination of batch process vents."
Group 1 combination of batch process vents means a
collection of process vents in a PMPU from batch unit
operations that are associated with the use of a
nonepoxide organic HAP to make or modify the product
that meet all of the following conditions:
(1) Has annual nonepoxide organic HAP emissions,
determined in accordance with §63.1428(b), of 11,800
kg/yr or greater, and
(2) Has a cutoff flow rate, determined in
accordance with §63.1428(e), that is greater than or
equal to the annual average flow rate, determined in
accordance with §63.1428(d).
Group 2 combination of batch process vents means a
collection of process vents in a PMPU from batch unit
operations that are associated with the use of a
nonepoxide organic HAP to make or modify the product
2-29
-------
that is not classified as a Group 1 combination of
batch process vents.
In the proposed rule, the relationship of the batch vent portion
of the definition of the "Combination of process vents that are
Group 1" to the provisions in §63.1428(f) was confusing. The
11,800 kg/yr annual nonepoxide organic HAP emissions cutoff was
not included in the definition, and the comparison of the cutoff
flow rate to the annual average flow rate was included in both
places. In order to clarify this situation, the definition of
"Group 1 combination of batch process vents" shown above includes
all criteria. This change in the final rule made §63.1428(f)
redundant with this definition. Therefore, the proposed
provisions in §63.1428(f) have been removed, and, in the final
rule, paragraph §63.1428(f) is reserved.
For process vents from continuous unit operations that are
associated with the use of a nonepoxide organic HAP to make or
modify the product, the terms in the final rule are "Group 1
continuous process vent" and "Group 2 continuous process vent."
In addition to changing the basis for this group determination to
an individual vent basis, the final rule also incorporates other
criteria not included in the proposed definition. The HON
definition of a Group 1 process vent includes three criteria:
flow rate, organic HAP concentration, and TRE index value. The
EPA intended for the proposed definition to mirror the HON
definition, but failed to include the flow rate and organic HAP
concentration criteria. Therefore, the final rule defines a
Group 1 continuous process vent as follows.
Group 1 continuous process vent means a process
vent from a continuous unit operation that is
associated with the use of a nonepoxide organic HAP to
make or modify the product that meets all of the
following conditions:
(1) Has a flow rate greater than or equal to 0.005
standard cubic meter per minute,
(2) Has a total organic HAP concentration greater
than or equal to 50 parts per million by volume, and
(3) Has a total resource effectiveness value,
calculated in accordance with §63.1428(h)(1), less than
or equal to 1.0.
2-30
-------
(5) and (6): The HON definitions of "maintenance
wastewater" and "maximum true vapor pressure" do not include
references to chemical lists, so the EPA does not understand why
there would be confusion regarding which chemical lists to use.
However, the EPA has made a small edit to the definition of
"maintenance wastewater," to clarify that the generation of
wastewater from the routine rinsing or washing of equipment in
batch operation between batches is not considered to be
"maintenance wastewater," for the purposes of subpart PPP.
(7): The phrase "organic HAP listed in Table 9 of subpart
G" means the same thing as the HON's language "Table 9
compounds." The EPA made this distinction so that the language
in the Polyether Polyol's phrase would be consistent with the
language in the rest of the rule. The reason that the EPA is
referring to Table 9 in subpart G is that Table 9 in subpart G
plays an instrumental part in the wastewater provisions in the
HON, which subpart PPP cross-references.
2.3.4 Epoxide
Comment: One commenter (IV-D-04) maintained that the
definition of "epoxide" is needlessly broad and complex. The
commenter declared there are only two epoxides, for purposes of
this rule, and each has a name. The commenter recommended the
following revised definition: "Epoxide means ethylene oxide
and/or propylene oxide."
Another commenter (IV-D-05) also recommended shortening the
definition of "epoxide," similarly as follows:
"Epoxide means ethylene oxide and propylene oxide for
purposes of this subpart."
Response: The EPA has reviewed the current definition of
epoxide and the revisions suggested by the commenters. The EPA
does not agree that the definition of epoxide would benefit from
a revision. Due to the fact that other epoxides are used to make
polyether polyols, the definition of "epoxide" cannot be limited
to EO and PO.
2-31
-------
2.3.5 Extended Cookout
Comment: One cominenter (IV-D-05) claimed that to conform
with industry usage, the definition of "extended cookout" should
be revised as follows:
Extended Cookout means a control technique that
reduces the amount of unreacted EO and/or PO (epoxides)
in the reactor. This is accomplished by allowing the
product to react for a longer time period, thereby
having less unreacted epoxides and reducing epoxides
emissions that may have otherwise been emitted
occurred.
Response: The EPA appreciates the comment, and has modified
the definition of "extended cookout" in the final rule,
accordingly.
2.3.6 Impurity
Comment: Two commenters (IV-D-05, IV-D-07) claimed that
since the provisions for process vents apply to all process vents
in the process, the definition of "impurity" should be changed to
reflect that low epoxide levels which remain in the product are
indeed impurities. Commenter (IV-D-05) stated that the existing
HON definition is inadequate in this regard and the incorporation
of the subpart F definition should be deleted in this section.
Both commenters suggested that the following definition be pulled
into subpart PPP from the HON, with the following revisions:
"Impurity means a substance that is produced coincidentally
with the primary product, or is present in a raw material, or is
a residual raw material that remains with the product after
production. An impurity does not serve a useful purpose in the
use of the primary product and is not isolated." Commenter (IV-
D-07) noted that this revised definition is consistent with the
concept of extended cookout in §63.1427 and informal guidance
provided to sources subject to the HON.
Response: There is a fundamental difference between an
impurity and unreacted HAP-reactant. The unreacted HAP-reactant
is a primary source of emissions from polyether polyols
2-32
-------
production. The EPA does not consider it appropriate to classify
unreacted reactants as impurities. Therefore, the final rule
continues to cross-reference the HON definition in §63.101.
2.3.7 Make or Modify the Product
Comment: One commenter (IV-D-05) claimed that, for clarity,
the definition of the phrase "make or modify the product" should
be revised as follows:
"Make or modify the product means to produce the polyether
polyol by polymerization of epoxides or other cyclic ethers with
compounds having one or more reactive hydrogens, and to add any
preservativeSj -/-antioxidants or diluents in order to maintain the
quality of the finished products, before shipping. Making and
modifying the product for this rule does not include grafting,
polymerizing the polyol, or modifying reacting it with compounds
components other than EO or PO.
Response: The EPA appreciates the comment and has made the
revisions requested. However, the revised definition lists
additives parenthetically. The definition in the final rule reads
as follows:
"Make or modify the product means to produce the polyether
polyol by polymerization of with epoxides or other cyclic
ethers with compounds having one or more reactive hydrogens,
and to add any prcacrvativoa/antioxidanto incorporate
additives (e.g., preservatives, antioxidants or diluents)
in order to maintain the quality of the finished products.
before shipping. Making and modifying the product for this
rule regulation does not include grafting, polymerizing the
polyol, or modifying reacting it with compounds other than
EO or PO."
2.3.8 Non-epoxide HAP
Comment: One commenter (IV-D-05) stated that the phrase
"non-epoxide HAP," which appeared in the May 1997 draft rule that
was distributed for review, has been deleted from the definitions
section of subpart PPP (§63.1423). The May 1997 draft of the
rule based the Group 1/Group 2 determination solely on non-
epoxide HAP emissions. According to the commenter, omitting the
2-33
-------
term seems to include the epoxide emissions in the Group 1/Group
2 evaluation, which is a significant change. Section 63.1428,
which defines the Group I/Group 2 evaluation, does not use the
epoxide emissions to determine the classification of the vent.
Response: Section 63.1428 is entitled "Process vent
requirements for group determination of PMPUs using a nonepoxide
organic HAP to make or modify the product." The EPA did not
intend to create any confusion over the fact that the group
determination described in §63.1428(f) was based on nonepoxide
organic HAP emissions. The commenters are mistaken about the
phrase "non-epoxide HAP" appearing in the definition section of
the May 1997 version of the proposed rule. In the May 1997
version of the proposed rule the term "nonepoxide organic HAP"
was used throughout the rule, but never defined. The EPA does
not find it necessary to define this term since "epoxide" is
defined and "organic HAP" is defined. Furthermore, as described
earlier in this document, §63.1428(f) has been reserved in the
final rule, and the definitions of "Group 1 combination of batch
process vents" and "Group 2 combination of batch process vents"
(in §63.1423) now contain all of the criteria for the group
determination, including the fact that only nonepoxide organic
HAP are used in the group determination.
2.3.9 Organic HAP
Comment: Three commenters (IV-D-04, IV-D-05, IV-D-07)
expressed serious concern about the clause "or has been or will
be reported under any Federal or State program, such as EPCRA
section 311, 312, or 313 or Title V" in the definition of
"organic hazardous air pollutant" and requested that it be
deleted from the definition. One commenter (IV-D-04) posed
several questions regarding future reports of chemicals: (1) how
can they know what chemicals will be reported in the future; (2)
how far into the future must they predict; and (3) if they make a
mistake in future predictions, does that mean they retroactively
2-34
-------
have 25 years of violations for not considering that chemical an
organic HAP?
The commenter added that they could probably tell what
chemicals have been reported under the specific' reporting laws
mentioned. However, they could not figure out everything that
has been reported under the vast number of other laws that might
be included in this definition. The commenter concluded that
they would probably have some difficulty excluding any law and
gave examples, which included OSHA, TSCA, and State requirements.
In addition, commenter IV-D-07 requested that the definition
be restricted such that glycol ethers that have low volatility
will not be included in the definition of Organic Hazardous Air
Pollutant. The commenter produces a number of products and
intermediates which technically meet the wide CAAA definition of
"glycol ethers." The commenter suggested that the limited
listing of glycol ethers used in Table 4 or 9 of the HON would be
an appropriate sub-list for inclusion as HAPs under subpart PPP.
Commenter IV-D-05 suggested the following language:
"Organic hazardous air pollutant(s) (organic HAP) means one
or more of the chemicals listed in Table 4 of this subpart or any
other chemical whictH-
(-t) •£•& is knowingly introduced into the manufacturing
process other than as an impurity, or haa boon or will bo
reported under any Federal or State program,—such aa EPCRA
section 311, 312, or 313 or Title V; and
&J 3t& is listed in Table 2 of subpart F of thio part. "
Response: The EPA has amended the definition of "organic
hazardous air pollutant." The definition that appears in the
final rule states that only chemicals listed in Table 4 of
subpart PPP, or chemicals listed in Table 2 of subpart F, that
are "knowingly produced or introduced" into the manufacturing
process constitute organic HAP for the purposes of subpart PPP.
However, with regard to the comment that requested that low
volatility glycol ethers be exempted from the definition of
organic HAP (or that a limited listing of glycol ethers, such as
2-35
-------
that used in table 4 and table 9 of subpart G, be used), table 4
and table 9 in subpart G both apply to the wastewater provisions
in subpart G, which subpart PPP directly references. Therefore,
the EPA has decided that, with regard to wastewater, the "limited
listing" that the commenter mentions (in tables 4 and 9 of
subpart G) is applicable, while a limited listing of glycol
ethers for the other provisions in this subpart would be
inappropriate for subpart PPP.
2.3.10 Override Definitions
Comment: One commenter (IV-D-04) expressed concern about
paragraph (c) in §63.1423 of the Definitions section. This
paragraph addresses what to do if a referenced subpart of the HON
uses a term that is defined in the HON, in the proposed rule, or
both. The commenter believed that the introductory sentences of
§63.1423 (b) addressed this and thus there is no need for proposed
§63.1423(c). Also, paragraph (c) leaves a number of other
scenarios unexplained. The following are two of four examples
cited by the commenter where doubt is created by the use of
paragraph (c) : the HON uses a term that is defined in the General
Provisions, and (1) the term is not otherwise defined in the HON
or in the proposed rule, and neither the HON nor the proposed
rule expressly borrows the General Provisions definition; or (2)
the HON does not expressly borrow that definition from the
General Provisions, but the proposed rule does. The commenter
requested that EPA eliminate the doubt and confusion by deleting
paragraph (c) in §63.1423.
Response: The EPA agrees with the commenter, and has removed
§63.1423(c) from the final rule. The EPA decided that
§63.1423(b) was sufficiently clear regarding which definition
should be used with regard to the subpart PPP requirements.
2.3.11 PMPU
Comment: One commenter (IV-D-05) stated that the definition
of polyether polyol manufacturing processes (PMPU) uses the term
2-36
-------
"polyether polyol product," which is not defined. For clarity,
the commenter recommended changing the text to read:
"... means a collection of equipment aoaomblod and conncctod
by process pipes or ducto,—used to process raw materials and to
systems that are associated with the manufacture of—a polyether
polyol product QD ita primary product."
Response: The definition of a PMPU in the proposed rule
read: "polyether polyol manufacturing process unit (PMPU) means
a collection of equipment assembled and connected by process
pipes or ducts, used to process raw materials and to manufacture
a polyether polyol product as its primary product." The
commenter correctly pointed out that the term "polyether polyol
product" was not defined. Therefore, the definition of
"polyether polyol manufacturing process unit" (PMPU) has been
modified by deleting the word "product" from the end of the
phrase "polyether polyol product", instead of by using the
commenter's suggested language. The language suggested by the
commenter was not used because the word "system" is not defined
in the rule.
2.3.12 Polvether Polvol
Comment: One commenter (IV-D-04) requested that EPA revise
the definition of "polyether polyol" to clarify that the
production of typical alkanolamines is not regulated under
subpart PPP. The commenter stated that the proposed definition
was worded broadly enough that it might be misinterpreted to
include alkanolamines. One alkanolamine (diethanolamine) is a
HAP; however, its production is regulated under the HON. The
commenter presented reasons why it does not believe that the EPA
intended to regulate alkanolamines under subpart PPP. However,
in the case where a manufacturer further reacts an alkanolamine
until it possesses repeating ether units, the end result is a
polyether polyol derivative of an amine. The commenter believed
that, due to the batch nature of these processes and the chemical
structure of the derivative, these products should be considered
2-37
-------
polyether polyols. However, they are a small exception and are
not typical of alkanolamines. Typical alkanolamines lack
repeating ether units, which are the hallmark of polyether
polyols. Therefore, the commenter urged EPA to add the following
sentence to the end of the definition of "polyether polyol" to
clarify that the production of typical alkanolamines is not
regulated under this rule:
"Polvether polyols do not include alkanolamines, in which
nitrogen is intentionally attached directly to the carbon of an
alkvl alcohol, unless the alkanolamine is further reacted to form
a molecule with more than three repeating ether units."
Response: The EPA agrees that the proposed definition of
"polyether polyol" needed some clarification. In the final rule,
"polyethers" are described parenthetically (in the definition of
"polyether polyol" as "compounds with two or more ether bonds".
The EPA believes that explaining what is meant by the term
"polyether" eliminates the possibility of owners or operators
interpreting alkanolamines as being part of the polyether polyol
source category. Therefore, the EPA did not add the sentence
suggested by the commenter to the final rule, but did revise the
final rule in a manner that should resolve the commenter's
concern.
Comment: One commenter (IV-D-07) owns and operates a hydroxy
ethyl cellulose manufacturing facility. Hydroxy ethyl cellulose
is formed through the reaction of ethylene oxide with multiple
reactive hydrogen sites (actually hydroxyl sites) on cellulose
polymer molecules. According to the commenter, some standard
references classify hydroxy ethyl as a "polyether ether."
Previous indications were that this type of manufacturing would
not be covered under this rule. For purposes of applicability,
the commenter requested that EPA clarify whether hydroxy ethyl
cellulose manufacturing is included or excluded from the
definition of a polyether polyol.
2-38
-------
Response: The EPA did not intend to make hydroxy ethyl
cellulose manufacturing facilities subject to the polyether
polyols rule, due to the cellulose component of the product.
Therefore, to rule out the likelihood that others might consider
the production of hydroxy ethyls to be subject to subpart PPP,
the EPA has revised the definition of "polyether polyol" in the
final rule, clarifying that the production of hydroxy ethyls is
not subject to subpart PPP. The revised definition of "polyether
polyol in the final rule reads:
Polvether polvol means a compound formed through the
polymerization of ethylene oxide (EO) or propylene oxide
(PO) or other cyclic ethers with compounds having one or
more reactive hydrogens (i.e., a hydrogen atom bonded to
nitrogen, oxygen, phosphorus, sulfur, etc.) to form
polyethers (i.e., compounds with two or more ether bonds).
This definition of "polyether polyol" excludes hydroxy ethyl
cellulose and materials regulated under the HON, such as
glycols and glycol ethers.
2.3.13 Pressure Decay Curve
Comment: One commenter (IV-D-05) asserted that the term
"pressure decay curve" is used in the rule without definition.
According to the commenter, this is not a common technical term.
Therefore, the commenter suggested adding a definition, as
follows:
"Pressure decay curve is the graph of the reactor pressure
versus time from the point when epoxide feed is stopped until the
reactor pressure is constant, indicating that most of the epoxide
has reacted out of the vapor and licruid phases. This curve must
be determined with no leaks or vents from the reactor. The
pressure decay curves for products that may have different
starting and finishing pressures may be compared by graphically
determining the time when the pressure has fallen to half the
total pressure drop;
Phalf=(Pinital-Pfinal)72 Equation x"
Response: The commenter was correct that the term "pressure
decay curve" was not defined in the definition section of the
rule. The EPA agrees with the definition presented by the
2-39
-------
commenter, except for the last sentence. The last sentence,
which gives a reference point for comparison with other pressure
decay curves, is not germane to the definition of a pressure
decay curve for this rule. Therefore, the last sentence was not
included in the definition in the final rule. However, the last
sentence is important in determining a point of comparison
between two different pressure decay curves, and this concept was
incorporated under §63.1427(h) in the final rule.
2.3.14 Process
Comment: One commenter (IV-D-05) recommended deleting the
definition of "process" for two reasons. First, the word
"process" by itself does not appear to be a significant
regulatory term in this proposed rule. Second, the definition
could cause confusion. For example, the definition says a
"process" makes a polyether polyol. Yet, companies have a
"process" to make HON products, or to make epoxy products, etc.
Similarly, the definition says a "process" may consist of one or
more unit operations. In contrast, a "process unit" (such as a
HON "chemical manufacturing process unit" or a PMPU for this
rule) consists of two or more unit operations. If a "process" is
not coextensive with a "process unit, " what is it? Moreover, the
definition says a "process" includes "all or a combination of"
various processing steps, "or other activity, operation,
manufacture, or treatment" which are used to produce a polyether
polyol. Apparently, then, there could be multiple "processes" in
a "process unit." As a result of this, there is no way to tell
what a "process" is or how to tell where one process ends and
another begins. Therefore, for clarity, the commenter
recommended deleting the definition of "process" because it is
adequate to use this term as part of other defined terms.
Response: The EPA agrees with the commenter that the word
"process" by itself is not a significant regulatory term.
Further the EPA agrees that the definition of "process" in the
2-40
-------
proposed rule is confusing. For these two reasons the EPA
deleted the definition of "process" from the final rule.
2.3.15 Process Condenser
Comment: One commenter (IV-D-05) claimed that the definition
of "process condenser" is not necessary because the term is not
used in the rule. Therefore, the commenter recommended deleting
the definition.
Response: The commenter is correct; the term "process
condenser" does not appear in the rule. Therefore, the term
"process condenser" has been deleted from the definition section
of the final rule.
2.3.16 Process Vent
Comment: One commenter (IV-D-07) requested that the
provisions for process vents from operations that handle
materials with HAP "as impurities only" be clarified. For
example, in one of their company's units which practices ECO as a
control technology, subsequent unit operations which provide
product treatment manage product containing small quantities of
residual epoxide, which the commenter claimed to be an impurity
at this point in the PMPU. The operation has about 15 emission
points and none of them are tied together into a manifold; to do
so would be very difficult and quite expensive. The commenter
noted that the proposal exempts these emission points from the
process vent control requirements under §63.1420(c)(8). However,
the commenter was concerned about the requirements in §63.1425 to
manage emissions from the combination of all process vents.
Therefore, the commenter requested that the final rule include a
clear provision to exempt process vents from equipment handling
HAP only as impurities, to address these situations. According
to the commenter, similar provisions are included in the HON,
Polymers and Resins I and IV, and other standards. The commenter
suggested that the -last sentence of the definition of process
vent be modified as follows: "Process vents exclude pressure
2-41
-------
relief valve discharges, gaseous streams routed to a fuel gas
system(s), vents from equipment that contain no organic HAP or
organic HAP only as impurities, and leaks from equipment
regulated under §63.1434."
The commenter also noted that the definition of impurity
will also have to be changed to fully accomplish this objective.
Response: Although the EPA has not taken the commenter's
advice, insofar as editing the definitions of "impurity" and
"process vent," the EPA has added the following sentence to the
end of the definition of "process vent," which should alleviate
the commenter's concern about the status of post-control unit
operations:
"A gaseous emission stream is no longer considered to be a
process vent after the stream has been controlled and
monitored in accordance with the applicable provisions of
this subpart."
Comment: One commenter (IV-D-05) was concerned that the
definition of "process vent" does not have a de minimis cutoff,
as does the definition of "process vent" in the HON. Also, the
commenter pointed out that, under this rule, a process vent may
originate from any unit operation, rather than from only
specified unit operations as in the HON, and that, unlike the
HON, this rule seems to say that the point at which a process
vent exists (and, thus, presumably where one would expect to have
to determine its characteristics) is where the stream leaves the
unit operation. To be consistent with the HON, the commenter
recommended deleting the definition and replacing it with the
following definition:
"Process vent means a gas stream containing greater than
0.005 weight-percent total organic HAP that is discharged during
operation of the PMPU. Process vents are gas streams that are
discharged to the atmosphere (with or without passing through a
control device) either directly or after passing through one or
more recovery devices. Process vents exclude relief valve
2-42
-------
discharges, gaseous streams routed to a full gas system(s), and
leaks from equipment regulated under §63.1434."
Response: The cutoff suggested by the commenter (0.005
weight-percent total organic HAP) has been incorporated into the
final definition of a process vent, for process vent from
continuous unit operations. This decision was based on the fact
that the EPA considers it to be impractical to impose
requirements for process vent streams with such low HAP
concentrations (less than 0.005 weight percent organic HAP). For
similar reasons, a de minimis cutoff for process vents from batch
unit operations was also added in the final rule. In the
Polymers and Resins I and IV NESHAP, the batch process vent
definition contains a de minimis cutoff of 225 kg/yr uncontrolled
HAP emissions. The EPA believes that this level is also an
appropriate de minimis level for process vents from batch unit
operations in the polyether polyols industry. The revised
definition of process vents in the final rule reads as follows:
Process vent means a point of emission from a unit operation
having a gaseous omiDDion stream that is discharged to the
atmosphere either directly or after passing through one or
more combustion, recovery, or recapture devices. A process
vent from a continuous unit operation is a gaseous emission
stream containing more than 0.005 weight-percent total
organic HAP. A process vent from a batch unit operation is
a gaseous emission stream containing more than 225 kilograms
per year of organic HAP emissions. Unit operations that may
have process vents are condensers, distillation units,
reactors, or other unit operations within the PMPU.—Process
vcnto arc pointa of omiooion from a unit operation having a
gaaoouo stream that io discharged to the atmosphere either
directly or after passing through one or more combustion,
recovery,—or recapture devices. Process vents exclude
pressure relief valve discharges, gaseous streams routed to
a fuel gas system(s), and leaks from equipment regulated
under §63.1434. A gaseous emission stream is no longer
considered to be a process vent after the stream has been
controlled and monitored in accordance with the applicable
provisions of this subpart.
2.3.17 Product Class
2-43
-------
Comment: One commenter (IV-D-07) suggested that the
definition of "product class" be modified to allow products with
similar or faster pressure decay curves to be included in the
same product class. The net result of this change would be to
allow sources to include more products in a product class with no
increase in emissions. As a result, the commenter stated that
fewer alternate scenarios and compliance tests would be required.
Another commenter (IV-D-05) stated that the definition of
"product class" is incomplete and makes little sense unless the
reader already knows what is implied. The commenter recommended
revising the definition, as follows:
"Product class means a group of polyether polyols with a
similar pressure decay curve representing the decline in proapurc
versus time. that are manufactured within a given set of
operating conditions..."
Response: The EPA modified the definition of "product
class" to include products with similar or faster pressure decay
curves, as requested by commenter IV-D-07. This change will
allow sources to include more products in a product class with no
resulting increase in emissions. The EPA has decided that
Commenter IV-D-05's suggested revisions to the definition of
"product class" add clarity to the definition, and has
incorporated those revisions into the final rule. The definition
of "product class" in the final rule reads:
Product class means a group of polyether polyols with a
similar pressure decay curve (or faster pressure decay
curves) that are manufactured within a given set of
operating conditions representing the decline in pressure
versus time. All products within a product class shall will
have an essentially similar pressure decay decline curve,
and operate within a given set of operating conditions.
These operating conditions are: a minimum reaction
temperature; the number of -OH groups in the polyol; a
minimum catalyst concentration; the type of catalyst (e.g.,
self-catalyzed, base catalyst, or acid catalyst); the
epoxide ratio, or a range for that ratio-7-j_ and^ the reaction
conditions of the system (e.g., the size of the reactor, or
the size of the batch).
2-44
-------
2.3.18 Purification
Comment: One commenter (IV-D-07) requested that a definition
be added for the term "purification." The commenter stated that
the proposed rule implied, but did not explicitly state, that
product purification is part of the PMPU process. As an
alternative, the commenter suggested that the definition of
"PMPU" could be modified by adding the words "and purify" to the
first sentence after the word "manufacture" in the final ,rule.
Response: The EPA agrees that purification of the product
was implied, but not directly stated, as being part of the PMPU
in the proposed rule. In the final rule, the EPA has revised the
third sentence in the definition of "PMPU," so that it states
that the collection of equipment "includes purification systems,
reactors and their associated product separators and recovery
devices...."
2.3.19 Recovery Device
Comment: Two commenters (IV-D-04, IV-D-05) asserted that the
definition of "recovery device" in the proposed rule should be
revised to be the same as the definition in the HON. Instead of
saying that recapture devices are considered to be recovery
devices "for the purpose of" monitoring, recordkeeping and
reporting requirements (as the HON does), the proposed definition
stated that recapture devices were considered to be recovery
devices "when" the rule required compliance with monitoring,
recordkeeping and recording (should be "reporting") requirements.
The word "when" referred to time, and not purpose. The
commenters requested that the last sentence be changed to match
the HON. Once revised, the definition would read, "For purposes
of the monitoring, recordkeeping, or reporting requirements of
this subpart, recapture devices are considered recovery devices,"
rather than "When...."
Also, the proposed definition states that reflux condensers
are part of the reactor unit operation. Commenter IV-D-04
believed that a reflux condenser on a distillation unit should be
2-45
-------
considered part of the distillation unit operation, instead of
the reactor unit operation. The commenter recommended that the
EPA consider revising this part of the definition to simply say
that reflux condensers are part of "a unit operation in the
process unit."
Response: The definition of "recovery device" is different
for the Polyether Polyols NESHAP than for the HON because the HON
covers continuous processes, while the Polyether Polyols NESHAP
includes both batch and continuous processes. However, the EPA
agrees that the definition of "recovery device" in the Polyether
Polyols NESHAP should use the wording from the HON, so that
"when" is replaced with "for the purpose of." This change has
been made in the final rule.
The EPA agrees with Commenter IV-D-04's comment that reflux
condensers are not necessarily part of a reactor unit operation,
and has revised the definition of "recovery device" accordingly.
2.3.20 "Start-up" and "Shutdown"
Comment: One commenter (IV-D-05) was concerned about the
fact that the definitions of "start-up" and "shutdown" in the
proposed rule were not parallel. The commenter stated that, in
the HON, considerable care was taken to make the definitions of
those two terms parallel, and that the same care is needed in
subpart PPP. The definition of "start-up" draws several
distinctions between batch and continuous processes, or unit
operations. The definition of "shutdown" does not draw those
distinctions. Thus, some equipment could be started up, but not
shut down, or vice-versa. The commenter stated that if these
distinctions are appropriate, they should be in both definitions,
worded identically.
In addition, Commenter IV-D-05 stated that the proposed
definition of "start-up" correctly mentioned an affected source,
a PMPU, a unit operation, "or" equipment required or used for
compliance, while the definition of "shutdown" mentions an
affected source, a PMPU, a unit operation, "including" equipment
2-46
-------
required or used for compliance. The commenter saw no reason why
these two definitions should use the words "or" and "including"
differently, and stated that the word "including" in the proposed
definition of "shutdown" raises a problem. This wording implies
that the only type of "compliance" equipment that counts, for the
purposes of a "shutdown," would be compliance equipment included
in a unit operation. However, the commenter pointed out that
compliance equipment is seldom included in a unit operation.
Instead, it is typically "add-on" equipment, rather than part of
the process. Thus, according to the commenter, under the
proposed definition, almost no compliance equipment could ever
have a "shutdown," although it might have occasional "start-ups."
Response: The EPA agrees with the commenter, and, in the
final rule the definitions of "start-up" and shutdown" read as
follows:
Shutdown means for purposes including, but not limited
to, periodic maintenance, replacement of equipment, or
repair, the cessation of operation of an affected source, a
PMPU within an affected source, a waste management unit or—a
unit operation within an affected source, including
equipment required or used to comply with this subpart, or
the emptying or degassing of a storage vessel. Shutdown
does not include the normal periods between batch cycles.
For continuous unit operations, shutdown includes
transitional conditions due to changes in product for
flexible operation units. For batch unit operations,
shutdown does not include transitional conditions due to
changes in product for flexible operation units. For
purposes of the wastewater provisions, shutdown does not
include the routine rinsing or washing of equipment between
batch cycles.
Start-up means the setting into operation of an
affected source, a PMPU within the affected source, a waste
management unit or a—unit operation within an affected
source, or equipment required or used to comply with this
subpart, or a storage vessel after emptying and degassing.
For all processes, start-up includes initial start-up and
operation solely for testing equipment. Start-up does not
include the recharging of batch unit operations. For
continuous unit operations, start-up includes transitional
conditions due to changes in product for flexible operation
units. For batch unit operations, start-up does not include
transitional conditions due to changes in product for
flexible operation units.
2-47
-------
2.3.21 Storage Vessel
Comment: Coimnenters IV-D-04 and IV-D-05 stated that the
following clause (highlighted) in the definition of "storage
vessel" is unnecessary, and that it causes problems and should be
deleted from the definition: "Storage vessel means a tank or
other vessel that is used to store liquids that contain one or
more organic HAP and that has been assigned, according to the
procedures in §63.1420(f), to a PMPU that is subject to this
subpart." According to the definition, nothing can be a storage
vessel until it is assigned to a PMPU. In addition, because it
is not a "storage vessel," it cannot be assigned to a process
unit under §63.1420(f). The commenter presented examples as to
why this clause is problematic.
Commenter IV-D-05 also recommended adding subparagraph (7)
as follows:
(7) Storage vessels assigned to another process unit
regulated under another subpart of Part 63.
Response: The EPA agrees with commenters IV-D-04 and IV-D-
05 on this point. The definition of "storage vessel" in subpart
PPP has been revised in the final rule as suggested (except that
the "p" in "part" is not capitalized) .
2.3.22 Unit Operation
Comment; One commenter (IV-D-04) recommended that the
definition of "unit operation" in the proposed rule be revised to
say "distillation units" instead of "distillation columns." This
change was made in the HON because the unit operation may include
more than just a "column." Additionally, there may be other
equipment (such as a reflux condenser) that is part of the same
unit operation, even though it is not a "column."
Response: The EPA agrees that subpart F of the HON and
subpart PPP should be consistent in how they define a "unit
operation." Subpart PPP has been changed to reflect the change
made via the HON amendments, as the commenter requested.
2-48
-------
2.4 PROCESS VENT CONTROL REQUIREMENTS
2.4.1 3 Percent Oxygen Correction
Comment: One commenter (IV-D-02) disagreed with the
proposed rule's requirements for sources to demonstrate
compliance with outlet concentration limits in §63.1425(b)(1)(i)
and (b)(2)(ii) and §63.1426(c)(3) at a 3 percent reference oxygen
level. While 3 percent oxygen is an appropriate reference level
for boilers, the commenter claimed it is not a reasonable
requirement for thermal and catalytic oxidizers, which typically
run at around 20 percent oxygen. The effect of using a 3 percent
oxygen level is to make the standard excessively stringent for
those sources using thermal and catalytic oxidizers. The
commenter suggested that the rule allow the use of a higher
reference oxygen level for these and similar technologies.
Response: The EPA is aware of situations where the 3 percent
oxygen correction is not appropriate. However, the commenter did
not provide sufficient rationale or information to support the
claim that this cutoff was not appropriate for the polyether
polyols industry. The EPA discussed this issue with polyether
polyols producers, and found that they did not share the concern
raised by the commenter. Therefore, no change was made in the
final rule in response to this comment.
Comment: Five commenters (IV-D-03, IV-D-04, IV-D-05, IV-D-
07, IV-D-08) maintained that the new source MACT floor of 99.9
percent for epoxide emissions from process vents is not
appropriate. The commenters recommended that the EPA establish a
separate category for facilities that are like the facility that
was used to set the new source MACT standard (i.e., Facility M).
The commenters elaborated that Facility M is not similar to
other sources in the source category because:
(a) its method of operation is substantially different from
the typical facility, resulting in significantly different
uncontrolled emissions;
2-49
-------
(b) it has a polyether polyols production capacity many
times higher than that of other sources, and;
(c) it has two incinerators.
In regard to their claim that Facility M's method of
operation is substantially different from a typical polyether
polyols production facility, the commenters stated that the
Agency already has adequate data to evaluate how differences in
venting may affect the emissions profile of a facility in .this
source category. According to the commenters, great differences
in emissions are associated with whether a facility operates with
a closed reactor that is vented only at the end of the epoxide
feed (i.e., closed-vent), or with one that is continually or
periodically vented during the epoxide feed (i.e., open-vent).
The commenters stated that existing data available to the EPA
demonstrate that Facility M is unlike other facilities in the
source category in that it emits significantly more epoxide to
the control devices on an essentially continuous basis.
In addition, one of the commenters (IV-D-07) pointed out
that the high levels of uncontrolled emissions occurring during
the vented mode of operation creates a significant difference
between the ability to demonstrate very high levels of reduction
in the control device at vented and non-vented sources.
Commenter IV-D-05 presented both a hypothetical and actual
comparison of emissions profiles from facilities that operate in
a closed-vent mode versus a similar facility venting in an open-
vented mode. The hypothetical case compared PO emissions, both
controlled and uncontrolled (after a water-cooled condenser) for
two reactor systems with the same physical parameters. Results
indicated that uncontrolled emissions from the vented facility
were 27 times greater than those from the nonvented facility;
however, controlled emissions (assuming 99.9 percent control for
the vented facility and 98 percent for the non-vented facility)
were 94.5 and 84.7 Ib/yr for the vented and nonvented facilities,
respectively. For the actual facility comparison, the commenter
compared the uncontrolled emission estimates of Facility M with
those from Facility I, which were reported to be similar sources.
2-50
-------
The commenter calculated uncontrolled emissions for Facility M
from Facility M's test report, by using the emission rates
reported during the test, and scaling these rates to 100-percent
capacity, with the assumption that both incinerators present at
Facility M were operating at 100-percent capacity at the same
time. The emission rate from Facility I were provided by a
representative of the corporation that owns Facility I, with
updated control efficiency and emissions estimates from those
originally submitted to the EPA for the MACT floor analysis for
1993. For the actual facility comparison, the uncontrolled
emissions from the vented facility were 17 times greater than
those from the nonvented facility, whereas controlled emissions
for the vented facility were 989 Ib/yr and those for the
nonvented facility were 1,160 Ib/yr. The commenter stated that
these analyses "confirmed the hypothesis that less effective
control gives equivalent emissions for nonvented reactors."
Further, commenter IV-D-07 provided a comparison of two
facilities owned by the corporation that the commenter
represents. The commenter explained that both facilities have
"similarly sized units." The commenter explained that the vented
reactor produces high molecular weight polyethylene glycols and
is equipped with a refrigerated condenser, and the nonvented
reactor produces a lower molecular weight polyethylene glycol.
The annual uncontrolled emissions were 860 Ib/yr for the vented
reactor and 390 Ib/yr for the nonvented reactor. The commenter
concluded that the "design considerations and emissions differ
significantly for vented and non-vented systems."
Commenter IV-D-05 stated that explanations of the need to
subcategorize Facility M based on size and the presence of two
incinerators was explained in detail before proposal. Commenter
IV-D-03 noted that the production capacity for Facility M is five
times larger than the average source cited in Table 2 of the
Supplementary Information Document for Proposed Standards (EPA-
453/R-97-010c, May 1997). Further, commenter IV-D-05 noted that
Facility M is unlike other facilities in the source category
2-51
-------
because Facility M had two incinerators, compared to other
facilities that did not have any incinerators.
Response: The EPA disagrees with the statement that, at
proposal, information was available to the Agency to demonstrate
that Facility M is unlike other facilities in the source category
with regard to the method of operation. Prior to proposal, an
extensive amount of information was provided to the EPA related
to the mode of operation at Facility M. However, while this
information made the EPA quite knowledgeable regarding Facility
M's mode of operation, only two of the other facilities in the
database provided any information regarding their mode of
operation. Without information about the majority of the other
facilities, it was impossible for the EPA to evaluate the
uniqueness of the mode of operation at Facility M prior to
proposal. At proposal (62 FR 46815), the EPA requested specific
information (including information on the mode of operation) from
polyether polyol facilities to allow the evaluation of whether a
subcategory was appropriate.
In response to this request, commenters presented three
comparisons of uncontrolled and controlled epoxide emissions for
vented and nonvented facilities. The EPA appreciated these
comparisons. However, several inconsistencies and assumptions
were identified that caused the Agency to conclude that these
comparisons do not, independently, provide a sufficient basis for
subcategorizing the polyether polyols source category into vented
and nonvented subcategbries. Some of EPA's concerns with these
comparisons are discussed below.
First, the hypothetical analysis assumed that a water-cooled
condenser was used at the reactor vent. The EPA believes that
the use of more efficient refrigerated condensers, which would
result in considerably lower uncontrolled emissions, is more
representative of practice in the industry.
With regard to the comparison of the actual facilities,
Facility I and Facility M, the EPA found that the epoxide
emission estimates used for Facility M in the commenter's
2-52
-------
comparison were drastically different from the emission data that
were directly submitted to the EPA by representatives of
Facility M. Also, the emission data from Facility I had been
updated from the data originally submitted during an EPA plant
site visit to that facility. The estimates provided in the
comments were lower than the original estimates due to process
improvements at the facility (that were not related to the method
of operation). The EPA conducted a similar comparison of the
uncontrolled epoxide emissions at these same two facilities using
the data originally submitted to the EPA by the two companies.
The results were not in accordance with those presented by the
commenter. In fact, the uncontrolled emission factor for
Facility I was higher than Facility M's factor. Clearly, the
analysis of the data available to the Agency does not support
this commenter's analysis.
The actual facility analysis conducted by commenter IV-D-07
stated that their analysis consisted of two facilities owned by
the commenter that were "similarly sized units." However, the
EPA found that the production capacity for the nonvented reactor
was larger than that for the vented reactor, and the emissions
were not adjusted accordingly.
Given these and other inconsistencies in the facility
comparisons provided by commenters, the EPA could not conclude
that subcategorization was necessary, based solely on these
comparisons. No commenters submitted the facility-specific data
that were requested in the proposal preamble. Therefore, even if
the examples provided by the commenters had led to the conclusion
that subcategorization was warranted, the EPA did not have
sufficient facility information to allow a complete
subcategorization evaluation.
However, the Agency still wanted to attempt to address the
commenters' concerns on this issue. Given the lack of data
provided by the industry prior to proposal and during the public
comment period, the EPA conducted a brief telephone survey to
inquire specifically about the mode of operation at polyether
polyol production facilities. Representatives from all the
2-53
-------
facilities in the process vent data base were called and asked to
describe their method of venting during epoxide feed. Of the
facilities for which the EPA was able to collect mode-of-venting
data, 24 percent (including Facility M) reported venting during
the epoxide feed step, and 76 percent reported that their
facilities did not vent during the epoxide feed step. Therefore,
the EPA concluded that the manner of operation of facility M was
not unique to the source category, as claimed by the commenters.
The EPA sought to determine whether the different venting
modes during epoxide feed resulted in "differences in the amount
and pattern of emissions and the achievable degree of emission
reduction," (Memorandum, from Seaman, J.C., EC/R Incorporated to
Svendsgaard, D, EPA/OCG. January 15, 1999. Documentation of the
Calculation of Uncontrolled Emission Factors. Docket Item: IV-B-
01). The EPA determined that a facility's uncontrolled emission
factor (mass emissions per mass of polyol product produced) was
the best method of comparison, and calculated such a factor for
each facility for which sufficient information was available.
For the "vented" facilities, the median uncontrolled emission
factor was 0.17 (Ib HAP emissions per 1000 Ib of product). The
data points were considered to have too varied a distribution,
with two orders of magnitude making up the difference between the
highest and lowest emission factor, for the mean value to be an
adequate representation of central tendency. For the "nonvented"
facilities the median uncontrolled emission factor was 1.09. The
commenters asserted that uncontrolled epoxide emissions at vented
facilities are considerably higher than those at nonvented
facilities. However, the results of the EPA's analysis, based on
the best information available, clearly do not support this
assertion, since the median uncontrolled emission factor
calculated for nonvented facilities is over six times higher than
the median uncontrolled emission factor for vented facilities.
In conclusion, based on all of the information available to
the Agency, the EPA was unable to determine a different emission
trend between the vented and nonvented groups from the data made
available to the Agency between proposal and promulgation.
2-54
-------
Therefore, the EPA did not subcategorize the industry based on
the method of operation.
The commenters' second rationale to support their claim that
Facility M is not a similar source was that the production
capacity at Facility M is many times that of other sources in the
source category. Subcategories, or subsets of similar emission
sources within a source category, may be defined if technical
differences in emissions characteristics, processes, control
device applicability, or opportunities for pollution prevention
exist within the source category (Federal Register, Vol. 57, No.
137, Initial List of Categories of Sources Under Section
112(c)(1) of the Clean Air Act Amendments of 1990). The EPA does
not believe that the fact that Facility M has a larger production
capacity satisfies any of these criteria. Further, since one
facility in the process vent database has a capacity that is 83
percent of Facility M's capacity, the EPA also disagrees that the
production capacity is unique.
The third argument given by the commenters to support the
claim that Facility M is not similar to the other affected
sources, was that Facility M has two incinerators, and that no
other sources have incinerators. The EPA disagrees with the
commenters' claim that Facility M is the only source with an
incinerator, since there is another facility in the database that
also uses incineration. Further, the fact that a source has a
better control than all other facilities in the source category
through the use of one or more incinerators is not a sufficient
basis for asserting that the source should be subcategorized.
The purpose of MACT is to ensure that regulated sources meet the
control standards achieved by the best performing sources in the
category. Subcategorization on the basis of the control
technology utilized would undermine the very concept of MACT.
In addition to the evaluation of the individual points
raised by commenters, the EPA also considered whether these
characteristics of Facility M collectively form a basis for
Subcategorization. The EPA concluded that, based on the
facility-specific process, emissions, and emissions control
2-55
-------
information provided to the Agency by the polyether polyol
industry, a separate subcategory should not be created solely for
Facility M.
Comment: Three commenters (IV-D-04, IV-D-05, IV-D-08)
requested that the combustion efficiency be set at 98-percent,
with a 20 ppmv concentration cutoff, for new sources for this
source category, to be consistent with the policy established for
other MACT standards as well as with data furnished to the EPA.
The commenters referred to other MACT standards, including the
HON, where the EPA has exercised such discretion. They stated
that the EPA had selected the 98-percent or less efficiency level
in some of the other source categories despite individual test
results indicating that greater than 98-percent reduction could
be achieved under specific test conditions. One commenter (IV-D-
08) noted that the EPA (in the HON) has required new source
controls for chemical industry process vents to meet a 98-percent
emission reduction, recognizing that a 99.9-percent control
efficiency was not achievable for these industries (see HON BID,
Section 12, page 2). One commenter (IV-D-03) asserted that the
overall expected emission reduction from a new source MACT of
99.9-percent, as opposed to 98-percent, would be trivial, even if
uniformly applied to all sources nationwide.
Response: The EPA disagrees with the commenters1 statement
that the EPA has an "established policy" that the combustion
efficiency be set at 98-percent. The EPA has no such policy,
even though previous rules may have established 98-percent
destruction efficiency as the standard, along with a 20 ppmv
alternative. However, more important than a precedent set by
previous rules, was the test data provided by the facility used
to set the MACT floor level of control for epoxide emissions from
new sources, Facility M, and the permit conditions with which the
facility must comply. The EPA has a responsibility to scrutinize
the test and permit data, and use it in setting a standard,
whenever possible. Therefore, the EPA could not simply go by the
2-56
-------
precedent set on other rules since the EPA had test data and
permit conditions from Facility M that could not be ignored.
Further, the test data provided by Facility M were calibrated to
the predominant epoxide in the vent stream, and EO and PO were
the overwhelmingly predominant HAP in the process vent stream.
This situation is unlike the HON. Since the HON regulated such a
large number of HAP, even if an individual facility had a tested
and reported destruction efficiency greater than 98 percent, this
destruction efficiency could not be generalized to all the HAP
regulated by the HON, due to the large variety of flammability
characteristics of the HAP at HON facilities. The EPA could not
address the commenters' statement regarding information the EPA
had available to set the MACT floors in "other MACT standards,"
since the commenter did not make specific references.
Comment: Two commenters (IV-D-03, IV-D-07) explained that
the combustion technology utilized by Facility M results in an
increase in criteria pollutants (C02 and NOX) , which were not
included in EPA's MACT floor analysis, while alternative control
technologies, such as scrubber or extended cookout, would be
expected to cause significantly lower NOX emissions.
Additionally, the commenters claimed that the EPA has failed to
account for potential process safety considerations associated
with the combustion of EO, noting that explosions at a number of
facilities that use or produce EO have already prompted the EPA
to delay enforcement of the December 6, 1994 air toxics rule for
EO sterilization facilities. Commenter IV-D-07 added that the
EPA should encourage standards that can be met using non-
combustion control strategies (achieving 98-percent reduction).
Response: The EPA is aware that incineration has secondary
criteria pollutant emissions. However, MACT floor decisions are
based on the reduction of HAP emissions, and cannot be based
primarily on their secondary impacts. The EPA is aware that the
use of incineration, resulting in an increase in sulfur dioxide
emissions, which may trigger Prevention of Significant
2-57
-------
Deterioration(PSD) and/or New Source Review (NSR). The EPA has
addressed this issue in previous NESHAP, by referring to a July
1, 1994 guidance memorandum issued by the EPA (available on the
Technology Transfer Network; see "Pollution Control Projects
(PCP) and New Source Review (NSR) Applicability" from John S.
Seitz, Director, OAQPS to EPA Regional Air Division Directors).
In this memorandum the EPA provided guidance for permitting
authorities on their ability to approve PCP exemptions (from PSD
review and major NSR) for source categories other than electric
utilities that use add-on controls and fuel switches to less
polluting fuels. In the July 1, 1994 guidance memorandum, the
EPA specifically identified the combustion of organic toxic
pollutants as an example of an add-on control that could be
considered a PCP and an appropriate candidate for a case-by-case
exclusion from major NSR. The EPA is alert to potential NSR
conflicts, and feels that this memorandum will alleviate most
NSR/PSD review concerns. In the event that it will not, the EPA
will attempt to create implementation flexibility on a case-by-
case basis.
The EPA does consider secondary impacts such as water
pollution, energy costs, costs to control, and emission of air
pollutants other than the 188 HAP in developing a MACT standard.
The estimated secondary impacts are presented in the proposal
SID. Further, the safety issues of incineration of epoxides were
adequately addressed at Facility M and the other facility in the
database that has incineration. Therefore, the EPA did not find
these reasons to be sufficient to justify eliminating Facility
M's data from the determination of the MACT floor for new sources
based on the fact that Facility M uses incineration.
Comment: Two commenters (IV-D-05, IV-D-08) maintained that
data from Facility M do not support the new source standard
because the Agency used State permit information and
corresponding performance test reports for Facility M. They
claimed that these data were submitted to the State agency to
demonstrate compliance with permit emission limitations for VOCs,
2-58
-------
not HAPs, and to document that the incinerators were meeting the
required VOC destruction efficiency. They noted that there are
several significant inconsistences between the test reports and
the proposed standards (these inconsistencies were discussed in
more detail under section 2.1.6, Test methods and procedures).
The commenters concluded that Facility M itself has not
demonstrated that it is able to meet the proposed rule's epoxide
emission limits, noting that the rule requires Method 301
validation of Method 25A and Facility M did not perform
validation by Method 301.
Response: The EPA disagrees with the commenters' statement
that the data from Facility M do not support the new source
standard because the performance test was conducted to determine
VOC destruction efficiency instead of epoxide, and the permit
conditions are for VOC. The primary pollutant in the stream was
PO, and this is the pollutant for which Method 18 at the inlet of
the incinerator, and Method 25A at the outlet of the incinerator,
were calibrated. Therefore, even though the test and permit cite
VOC destruction efficiency, it is clear that it is the
destruction of PO that was tested and regulated at Facility M.
The commenters' concerns about inconsistencies between the
test reports and the proposed standards are discussed in section
2.1.6 of this document. In summary, the performance test
performed by Facility M is consistent with the performance test
requirements in the final rule.
Comment: Two commenters (IV-D-04, IV-D-05) requested that
the EPA clarify that, in all instances, two or more devices in
combination may be used to meet an emission limitation. For
example, commenter IV-D-04 stated that Facility M has two
incinerators, and there may be other facilities that use a
combination of control devices or .recovery devices to achieve
emission limitations. Both commenters requested that the EPA
clarify, both in §63.1431(e) and in the preamble for the final
rule, that combinations of devices are permissible. The
2-59
-------
commenters also recommended the following revision in §63.1431(e)
and the first sentence in §63.1431(e)(1):
"(e) Compliance with the epoxide emission factor limitation
through the use of extended cookout in conjunction with a—one or
more combustion, recovery, and/or recapture devices.. (1) The
owner or operator shall notify the Agency of the intent to use
extended cookout in conjunction with a—one or more combustion.
recovery, and/or recapture devices, to comply . . . . "
Response: The Agency intended to allow for multiple control
techniques in series, and has amended §63.1431(e) as recommended
by the commenters.
2.4.2 A Concentration Limit as an Alternative Process Vent
Emission Limit
Comment: In response to EPA's request for comments on the
determination of an alternative concentration limit for new
source process vents (§63.1425(b)(1)(ii)), four commenters (IV-D-
03, IV-D-05, IV-D-07, IV-D-08) recommended a 20-ppm cutoff
concentration limit. Two commenters (IV-D-03 and IV-D-05)
agreed that ther'e was ample data available to EPA to support this
limit. One commenter (IV-D-03) referred to HON stack emission
test data submitted to EPA Region IV and an attached emissions
summary table taken from the test report, which they believed
supported their claim. The commenter noted, however, that these
data were generated to demonstrate HON compliance for a
continuous process, which may not be equivalent to the expected
performance of a batch process to be regulated under the
Polyether Polyols Production NESHAP.
Commenter IV-D-05 pointed to the HON compliance trials as
proof that the 20-ppmv concentration limit used in the HON is a
conservative value as an alternative to the new source MACT
standard and should be adopted in this standard, regardless of
whether a 98 or 99.9-percent efficiency limit is established.
This commenter noted that a 20-ppmv cutoff is recognized as
appropriate in other MACT standards, and in most NSPS standards
2-60
-------
that apply to process vents from the chemical industry. Because
most facilities in this source category use batch processes, the
commenter asserted that a lower concentration cutoff will be
difficult to develop, demonstrate compliance with, and enforce in
a practical manner.
Commenter IV-D-07 supported the comments of commenter IV-D-
05, and added that from a source owner/operator perspective,
compliance with the 20-ppmv criteria is significantly simpler to
demonstrate than compliance with the 98-percent reduction, since
only the outlet from the control device needs to be tested to
demonstrate compliance with the 20-ppmv criteria. In addition,
the compliance difficulties that might result from owners and
operators comparing two measured numbers would be eliminated, and
problems in sampling and analyzing highly variable streams from
batch process vents containing significant concentrations of
epoxides would be eliminated, by allowing the 20 ppmv cutoff
concentration limit.
The commenter (IV-D-07) also provided another reason to
support an alternative concentration limit of 20-ppmv for new
sources. In the Boiler and Industrial Furnace Interim rules, EPA
has established a precedent where a 20-ppmv hydrocarbon limit is
used as a compliance limit for certain systems required to comply
with a destruction efficiency greater than 99.9 percent. The
commenter concluded that a 20-ppmv concentration limit is an
acceptable control level for both combustion facilities and
uncontrolled and non-combustion control devices, and that it is
justified for both new and existing sources.
One commenter (IV-D-08) supported the reasons above to
maintain a 20-ppmv concentration cutoff, and added that the EPA
has previously stated that 20 ppmv is the lowest outlet
concentration of total organic compounds achievable by combustion
of low organic concentrations (reference was made by the
commenter to the preamble to the NSPS for subpart NNN).
In addition, this commenter cited the following reasons for
not establishing a concentration limit of 1 ppmv for new source
process vents:
2-61
-------
(1) A concentration limit of 1 ppmv may not be achievable by
combustion over the long compliance duration required by the
rule;
(2) A concentration limit of 1 ppmv would be closer to the
analytical detection limit, and have greater uncertainty would be
associated with the analytical results than there would be with a
concentration limit of 20 ppmv.
Response: The Agency agrees with Commenter IV-D-08's
statement that the EPA previously stated that 20 ppmv is the
lowest outlet concentration of total organic compounds achievable
by combustion of low organic concentrations (an inlet
concentration of 2000 ppmv), referencing the preamble to the
proposed NSPS for Air Oxidation Unit Process (48 FR 48932,
October 21, 1983). As stated in subpart NNN's preamble, the
outlet concentration of 20 ppmv was established based on kinetic
calculations of incinerators. It was demonstrated that, at a
given temperature and residence time, a stream with a low inlet
concentration (approximately 2000 ppmv) could not demonstrate an
outlet concentration below 20 ppmv. Further, in the preamble to
the proposed amendments to the HON (61 FR 43698, August 26,
1996), the EPA expanded the application of this lower bound
concentration performance standard to control/recovery devices
other than incinerators. The HON's preamble explained that
recovery devices are designed to typically reduce emissions to
the same outlet concentration level given a relatively wide range
of inlet concentrations. When the inlet concentration is
substantially below the design maximum leading conditions (and
begins to approach the residual level in the outlet stream) the
recovery device efficiency will decrease. Therefore, the final
rule contains an alternative concentration limit of 20 ppmv for
both new and existing sources.
At proposal, the existing source concentration limit was 20
ppmv of total epoxides. In evaluating the new source limitation,
the EPA considered whether this limitation should be "total
epoxides or TOC." Other rules, such as the HON, allowed the
2-62
-------
option of determining outlet concentration limits on a TOC basis.
Since the EPA desired to allow Method 25A (which is designed to
measure TOC) to determine compliance with this concentration
limit, and since other standards allowed the option of compliance
on a TOC basis, the concentration limits in the final rule for
new and existing sources are 20 ppmv total epoxides or TOC.
Comment: One commenter (IV-D-07) advocated that the
alternative 20-ppmv concentration limit should apply more broadly
to process vents without controls. For example, there might be
vents from equipment practicing a very long extended cookout or
vents from equipment where the epoxide content is very low and
emissions are very small. The commenter noted a variety of
precedents in MACT standards (particularly the HON) applying to
chemical industry sources, to support the concept of making the
limit broadly applicable.
Response: First, the commenter is incorrect in stating that
the HON allows a 20-ppmv concentration limit for process vents
that do not control. Paragraph §63.113(a) in the HON specifies
the control devices and recovery devices that are permissible for
achieving the 20-ppmv concentration limit.
The 20 ppmv outlet concentration limit recognizes that there
is a lower outlet concentration boundary, below which combustion,
recapture and control devices cannot achieve when the inlet to
the device is below approximately 2000 ppmv. The EPA understands
that the outlet concentration after extended cookout may be as
low as that after a combustion, recovery, or recapture device.
However, this is not based on technological limitations of ECO,
as is the basis for the 20 ppmv concentration limit for
combustion, recovery, and recapture devices. Therefore, the EPA
believes that allowing the 20 ppmv concentration limit for ECO is
not appropriate.
Further, the EPA does not believe that it is appropriate to
use this alternative concentration requirement as a de minimis
cutoff for vents where the epoxide content is very low and
2-63
-------
emissions are very small. The EPA believes that the HAP
concentration and emission de minimis cutoffs in definition of
the process vent (discussed above in Section 1.2.3) adequately
address these vents.
2.4.3 Basis for Outlet Concentration Testing as an
Alternative Process Vent Limit
Comment: One commenter (IV-D-05) strongly supported the
level for existing sources and the use of alternative cutoff
levels in §63.1425(b)(2)(ii). However, the commenter noted that
the word "average" in §63.1425(b)(2)(ii), which was between
"outlet" and "concentration" in a draft version of the proposed
rule that was shared with industry, has been deleted. This is a
significant change for batch processes. According to the
commenter, a 20-ppmv average outlet concentration is a much
different limit than a 20-ppmv maximum outlet concentration.
When a reactor is vented down, the initial concentration will be
high, decreasing as the venting continues. The overall vent in a
given situation may well meet the 20-ppmv average, but be
significantly above this limit for a short portion of the venting
period. Also, the commenter noted that §63.1425(b)(2)(ii) unlike
§63.1425(b) (2) (i), does not include "process vents." Since this
paragraph is in a section dealing with process vents, the
commenter believed that it was the EPA's intent to include them
in this paragraph. Therefore, the commenter recommended revising
the text, as follows:
"Maintain an average outlet concentration for process vents
of total epoxides...."
Response; The word "average" between "outlet" and
"concentration" was deleted in the proposed rule from a draft
shared with the public because the term was inappropriate at that
location. Initial compliance is determined by the procedures
specified in §63.1426(c)(3), which in turn cites Method 18. For
process vents from continuous unit operations Method 18 is
conducted for 3, 1-hour runs; for process vents from batch unit
2-64
-------
operations Method 18 is conducted during worst-case conditions.
The term "average" comes into play in relation to continuous
compliance. For continuous compliance, a daily average must be
maintained. This daily average can be determined using either
GEMS data or parametric monitoring data. Further, the word
"average" was not used in any of the other subsections of
§63.1425(b)(1) or (2), for the same reason.
2.4.4 Group Determination for Nonepoxide HAP Process Vent
Emissions on a Vent-by-Vent Basis
Comment: Three commenters (IV-D-04, IV-D-05, and IV-D-08)
requested that owners or operators have the option of making the
group determinations for nonepoxide process vents on a vent-by-
vent basis, rather than being required to do the group
determination for the combination of all process vents. The
commenters maintained that the distinction between Group 1 vents
(requiring control) and Group 2 vents (not requiring control) is
essentially a cost-effectiveness decision borrowed, in this rule,
from previous MACT standards such as the HON. However, all
previous MACT standards that have required Group determinations
for process vents have specified that the determinations be
conducted on individual vents. According to the commenters, the
EPA appears to be borrowing those same equations and criteria,
and employing them in a totally different context, without making
the adjustments that would be necessary for that context. One of
the commenters also noted that there was no supporting rationale,
and lacks legal justification for setting the MACT floor level of
control more stringent than any other MACT standard that
previously used these Group Determination equations. Commenter
IV-D-08 maintained that the proposed rule sets a dangerous
precedent for future MACT standards that might impact the
chemical industry.
These commenters also expressed the following specific
objections to this approach.
(1) Commenter IV-D-08 stated that the use of an aggregated
vent approach implies that all process vents in a unit are
2-65
-------
manifolded to a common control device. The commenter reported
safety and construction concerns if this were the case.
(2) Commenter IV-D-04 asserted that Group determinations for
combinations of process vents would be excessively biased toward
finding vents to be Group I. For example, a threshold flow rate
of 0.005 standard cubic meters per minute may be realistic for
deciding the cost-effectiveness of controlling an individual
process vent, but it would be virtually impossible to find any
process unit having such a low flow rate for the combination of
all its process vents.
(3) Commenter IV-D-8 expressed concern that the proposed
rule does not provide for appropriate batch process applicability
cutoffs, such as annual emission limits and cutoff flow.
(4) Commenter IV-D-04 claimed that by using the combination
of process vent Group determination approach, the EPA would
provide a disincentive to the very type of emission reduction
efforts which, in previous rules such as the HON, were a desired
outcome. For example, under the HON an owner or operator is
allowed to make process changes that increase the TRE index
value. However, the proposed rule provides no incentive for
making beneficial changes to a single process vent unless there
is a realistic chance to get the TRE index value into Group 2 for
the combination of all process vents. The commenter stated that
there is virtually no chance that a process change could make the
entire combination of vents Group 2.
In order to address these concerns, the commenters made
several suggestions. Commenter IV-D-04 suggested that the EPA
either validate the "borrowed" equations and criteria in the
context of combined process vents, or develop and validate
entirely new equations and criteria, in order to allow Group
determinations to be established based on "combinations" of
process vents. Commenter IV-D-05 suggested that the EPA simply
allow owners and operators to conduct group determinations on a
vent-by-vent basis.
2-66
-------
Response: The EPA agrees with the statement that the Group 1
criteria is essentially a cost-effectiveness decision. The EPA
also agrees that the criteria in subpart PPP were borrowed from
other MACT standards, specifically the HON (for process vents
from continuous unit operations) and Polymers and Resins I and IV
(for process vents from batch unit operations).
The EPA agrees that the TRE index approach was developed
for, and has been applied on, individual vents. Therefore, the
EPA further agrees that in order to apply the TRE approach to the
combination of process vents from continuous unit operations in a
PMPU is not appropriate without conducting an analysis to
validate the equations for the combination of vents, or to
develop new equations. Rather than take this approach, the EPA
decided to apply the Group 1 criteria for process vents from
continuous unit operations that use nonepoxide organic HAP to
make or modify the product to individual process vents.
For process vents from batch unit operations that use
nonepoxide organic HAP to make or modify the product, the Group 1
equations are the same equations employed in the Polymers and
Resins I and IV MACT standards (40 CFR 63, subparts U and JJJ,
respectively). The EPA agrees with the commenters that in the
polymers and resins standards, the Group criteria are applied to
individual vents. However, unlike the TRE for process vents from
continuous unit operations, the group determination approach that
is used in subparts U, JJJ, and PPP, was originally developed to
be used for either individual vents or the combination of vents.
The original source of the batch vent group determination
approach is the EPA document "Control of Volatile Organic
Compound Emissions From Batch Processes - Alternative Control
Techniques Information Document" (EPA-453/R-94-020), i.e., the
Batch ACT. On page 7-5 of this document, the EPA states "The
control option requirements presented in Chapter 6 apply to (1)
individual batch VOC process vents to which the annual mass
emissions and average flowrate cutoffs are applied directly, and
(2) aggregated VOC process vents for which a singular annual mass
emission total and average flowrate cutoff value is calculated
2-67
-------
and for which the option is applied across an aggregate of
sources." Therefore, for process vents from batch unit
operations, the EPA disagrees with the statements that the group
determination equations are being used "in a totally different
context" and that there is no supporting rationale for using
them. The final rule retains the requirement that the Group
criteria be applied to the nonepoxide organic HAP emissions from
the combination of process vents from batch unit operations
associated with the use of nonepoxide organic HAP to make or
modify the product.
With regard to the specific concerns raised by the
commenters, the EPA does not agree that applying the group
criteria to the combination of process vents in a PMPU implies
that all process vents are manifolded together. The EPA clearly
recognizes that not all process vents are manifolded together,
and that there could be safety and construction concerns with
doing so. Applying the group criteria to the combination of
vents means that the decision whether to control process vents in
the PMPU is based on the characteristics of all process vents.
If the combination of process vents is determined to be Group 1,
the EPA believes the rule provides considerable flexibility to
the owner or operator in how to achieve the specified emission
reduction for emissions from all process vents. There is no
requirement that any process vents be combined.
The examples provided in second and fourth concerns are
specific to the Group 1 criteria for process vents from
continuous unit operations. As noted earlier, the EPA has
changed the final rule so that the group criteria for these vents
is applied on an individual vent basis. Therefore, the examples
should no longer be of concern.
With regard to the third concern, the commenter indicated
that the proposed rule did not include "appropriate batch
process applicability cutoffs, such as annual emission limits and
cutoff flow." At §63.1428 (c), the proposed rule did have an
annual emission limit cutoff of 11,800 kilograms per year.
Therefore, if total nonepoxide organic HAP emissions from all
2-68
-------
process vents from batch unit operations that use nonepoxide
organic HAP to make or modify the product were less than 11,800
kilograms per year, the combination of process vents would be
Group 2. The proposed rule also contained the concept of a
"cutoff flow rate" at §63.1428(e). The cutoff flow rate is
calculated from the annual nonepoxide organic HAP emissions, and
compared to the actual flow rate. In addition to these
"cutoffs," which are retained in the final rule, the EPA has
clarified, in the definition of process vent, that process vents
from batch unit operations must have annual organic HAP emissions
of 250 kilograms per year or greater.
In conclusion, the EPA agrees with the commenters that the
group determination for process vents from continuous unit
operations that use nonepoxide organic HAP to make or modify the
product should be made on an individual vent basis, and has
modified the final rule accordingly. However, the EPA disagrees
that the group determination for process vents from batch unit
operations that use nonepoxide organic HAP to make or modify the
product should be on an individual vent basis. The final rule
requires that this group determination be made on the combination
of all batch vents in the PMPU.
Comment: One commenter (IV-D-04) stated that the proposed
rule's requirement for separate Group determinations for
continuous process vents and batch process vents is inconsistent
with the real-life scenario in which the two types of vents are
ducted together. Criteria such as HAP concentrations, flow
rates, etc. can be determined on a vent-by-vent basis, but the
commenter is unsure how to determine them for a combination of
batch and continuous vents, and the proposed rule does not
explain how to do this. The solution suggested by Commenter IV-
D-05 involved revising §63.1424(b) as follows:
"(b) When emissions of different kinds (i.e., emissions
from process vents subject to §63.1425 through §63.1430,
storage vessels subject to §63.1432, process wastewater,
and/or in-process equipment subject to §63.149 of subpart G)
are combined, and at least one of the emission streams would
2-69
-------
require control according to the applicable provision (e.g.,
is Group 1. or where applicable, belongs to a combination of
process vents that is Group 1) in the absence of combination
with other emission streams, the owner or operator shall
comply with the requirements of either paragraph (b)(1) or
(b)(2) of this section."
Response: The EPA agrees that the final rule needed more
specific requirements for streams that have been ducted together
or otherwise combined. However, the commenter's suggested rule
language and their actual comment referred to two different
situations. The comment cited concerns with how to conduct the
group determination for combined streams, but the paragraph that
the commenter suggested language for, §63.1424(b), specified how
to control combined streams. The EPA made changes to address
these concerns in both of these instances.
As requested by the commenter, the final provisions in
§63.1424(b) state how to comply for combined streams from
different types of emission points; however, the EPA selected a
more straightforward approach. Paragraph §63.1424(b) of the
final rule states that when emission streams are combined, the
owner or operator has the option to comply with the individual
requirements for each type of emission stream in the combined
stream, or to comply with the most stringent requirement for any
stream in the combined stream.
In order to provide guidance regarding Group determinations
in combined streams, the EPA found it necessary to add new
requirements, as §63.1428(i). Paragraph §63.1428(1) specifies
that the Group determination for a stream containing a
combination of process vents from batch unit operations and
process vents from continuous unit operation, both associated
with the use of a nonepoxide organic HAP to make 'or modify the
product, shall be determined as for any other process vent from a
continuous unit operation, except that the TRE must be calculated
when nonepoxide organic HAP emissions are being generated by the
batch unit operation that feeds into the combined stream.
In making this change to the final rule, the EPA also
realized that clarification was needed with regard to when the
2-70
-------
owner or operator should collect the information needed to make a
Group determination for uncombined process vents from batch unit
operations, or from continuous unit operations, as well. As a
result, §63.1428(a) has been revised to state that if the owner
or operator is using a combustion, recovery, or recapture device
to reduce epoxide emissions from process vents from batch unit
operations, then the location at which the annual uncontrolled
nonepoxide organic HAP emissions and annual average flow rate are
determined must be at the exit of the combustion, recovery, or
recapture device.
In addition, §63.1428(h)(1) has been revised to provide
specifications regarding where the owner or operator must conduct
the TRE index value determination. This location is after the
last nonepoxide recovery device, if the owner or operator uses
one or more nonepoxide recovery devices after all control
techniques to reduce epoxide emissions; at the exit of at the
exit of the combustion, recovery, or recapture device, if the
owner or operator does not use a nonepoxide recovery device after
a combustion, recovery, or recapture device to reduce epoxide
emissions; or at the exit from the continuous unit operation, if
the owner or operator does not use a nonepoxide recovery device
after extended cookout to reduce epoxide emissions. The TRE
index value is one of the factors that determines the Group
status of a process vent from a continuous unit operation,
according to the final definition of Group 1 continuous process
vent in §63.1423.
2.4.5 Relationship to Polymers and Resins I Changes
Comment: One commenter (IV-D-04) noted that the proposed
requirements for batch process vents were borrowed from the Group
1 Polymers and Resins standard, which is in litigation.
Therefore, the commenter requested that, when the result of the
Polymers and Resins Group 1 litigation is final, the EPA take
additional public comments on the concept of incorporating any
changes to that rule into this rule.
2-71
-------
Response: The EPA agrees with the commenter, and the
preamble for the proposed amendments to the Group 1 Polymers and
Resins NESHAP (64 FR 11560, March 9, 1999) requested comments on
the concept of applying the amended batch requirements
automatically to other subparts that reference the Group 1
Polymers and Resins NESHAP.
2.4.6 Group Redetermination
Comment: One commenter (IV-D-04) maintained that reducing
production capacity or production rate should not trigger
mandatory redetermination of process vent Group status, as
proposed. The commenter stated that decreasing the production
rate or production capacity would not be expected to move process
vents from Group 2 to Group 1, since typically emissions decrease
with decreases in production. Therefore, the commenter suggested
that §63.1428(g)(1) and (h)(2) be revised as follows:
"Examples of process changes include, but are not limited
to, increases changed in production capacity-r or production
rate, changes in feedstock type-r or catalyst type^-t- or . . . . "
Response: The EPA agrees with the commenter, and has
incorporated the suggested changes into the final rule.
2.4.7 Non-epoxide versus Epoxide Process Vent Emission Limits
Comment: One commenter (IV-D-05) stated that, in the
process vent control requirements of §63.1425(a), it is not clear
whether emissions of epoxides would be subject to one emission
limit or two. The first emission limit is specifically for
epoxide emissions. The second emission limit is for "organic HAP
emissions resulting from the use of nonepoxide organic HAP (in
addition to epoxides) to make or modify the polyether polyol
product." Since epoxides are organic HAP, it appears that they
may be included in the term "organic HAP emissions" in the second
emission limit. The commenter thought that the EPA intended for
the first emission limit to cover epoxides, and for the second
emission limit to apply only to non-epoxide organic HAP. To make
2-72
-------
this clear, the commenter recommends revising the text, as
follows:
"...paragraph (c) of this section contains limitations for
nonepoxide organic HAP emissions resulting from the use of
nonepoxide organic HAP (in addition to epoxides) to make or
modify the polyether polyol product;..."
Response: The Agency agrees that the proposed language was
confusing, and has made the suggested change to the final rule.
2.4.8 Alternative Emission Factor
Comment: One commenter (IV-D-05) supported the inclusion of
an alternative emissions factor as proposed by the EPA for
process vents from new and existing sources. However, the
commenter urged the EPA to clarify [in §63.1425(b)] that owners
and operators may choose to demonstrate compliance by meeting the
appropriate control efficiency, by maintaining outlet
concentrations on individual process vents, or by maintaining the
PMPU-wide emission factor. The commenter pointed out that the
emission factor in §63.1425 (b) (1) (iii) and (b)(2)(iii) is not
expressly limited to process vents, and instead is "PMPU-wide,"
which could be mistaken to include equipment leaks, such as minor
emissions from flanges or valves, plus any emissions from
wastewater, storage vessels, etc. Since these PMPU-wide emission
factor limits are meant to apply to process vent emissions, the
commenter suggested minor changes to §63.1425(b)(1)(iii) and
(b)(2)(iii) to clarify their intent, as follows:
(1) ***
(iii) Maintain an a PMPU wide emission factor of no greater
than 4.43 X 10-3 kilogram epoxide emissions per megagram of
product for all process vents in the PMPU.
*****
(2)***
(iii) Maintain an a PMPU wide emission factor of no greater
than 1.69 X 10-2 kilogram epoxide emissions per megagram of
product for all process vents in the PMPU."
Another commenter (IV-D-07) also requested that these
paragraphs be clarified, for the same reasons presented above.
2-73
-------
This commenter's suggested revision to these paragraphs was
slightly different:
"Maintain a PMPU-wide emission factor for process vents of
no greater than...."
Response: The Agency agrees with the commenters, and has
changed §63.1425(b) in accordance with Commenter IV-D-05's
suggestions. However, please note that proposed
§63.1425(b)(2)(iii) is §63.1425(b)(2)(iv) in the final rule.
2.4.9 Batch/Continuous Process Units versus Batch/Continuous
Unit Operations
Comment: To be consistent with the definitions provided in
§63.1423(b), one commenter (IV-D-05) recommended changing the
terms in §63.1425(b) as follows: "... batch proacaa unit
operation" and "continuous process unit operation."
Response: The EPA agrees that "batch process unit operation"
and "continuous process unit operation" were redundant terms, and
that the word "process" was not necessary in those phrases. The
Agency appreciates the comment and has revised these phrases in
the final rule, as suggested by the commenter.
2.4.10 Controls for New Source Process Vent Emission Limits
Comment: One commenter (IV-D-05) stated that it is not clear
in the proposed rule that new sources may use extended cookout as
a mechanism to meet the epoxide emission limits, and, therefore,
the commenter recommended that a sentence be added at the end of
§63.1425(b)(1) as follows:
"Extended cookout may be used to meet any of these
standards."
Response: The extended cookout control options are
contained in §63.1427. The Agency does not think that it is
appropriate to single out extended cookout in §63.1425(b)(1), as
suggested by the commenter.
2-74
-------
2.4.11 Calculation Methods for Emissions from Batch Operation
Units
Comment: One cornmenter (IV-D-07) noted that, in order to
verify compliance with the emission factor limits for process
vents, the owner or operator is required to calculate annual
epoxide emissions and divide the emissions by the annual
polyether polyol production rate. Emissions from batch
operations are calculated based on the procedures in
§63.488(b)(1) through (b)(7) of subpart U. The commenter
requested that the EPA, in order to avoid duplicative work and
duplicate data sets for the same emission points, make it easy
for facility owner/operators to use the alternative engineering
assessments contained in §63.488(b)(6) of subpart U to estimate
emissions. That is, §63.488(b)(6)(ii)(B) of subpart U should
specifically state that other methods of estimating emissions,
such as those used for past permit applications, emission
inventories, or SARA 313 reports, may be used as part of the "any
other means" mentioned in §63.488 (b) (6) (ii) (B) .
Response: The EPA does not consider it necessary to clarify
the meaning of the phrase "any other means." However, under the
proposed amendments to subpart U, the promulgated requirements
found in §63.488(b)(6)(ii)(B) have been removed, and greater
latitude is offered to owners and operators by allowing them an
open-ended ability to "request approval" to use engineering
assessment (including the use of previous test results, as long
as the previous test was conducted under conditions that are
representative of current operating conditions), via proposed
§63.488(b)(6)(i)(C). This proposed amendment, if promulgated,
should reduce the amount of "duplicative" work for owners and
operators, which should appease the commenter's concern (as
stated above). Further, the EPA is seeking comments on the
impact of the proposed subpart U amendments on polyether polyol
facilities in the notice announcing these proposed amendments.
2-75
-------
2.4.12 Continuous Unit Operations Group Determinations
Comment: One coirimenter (IV-D-05) stated that the regulatory
text in §63.1428(h) does not provide a flow rate or concentration
cutoff (these are included in the Group 1 process vent definition
of subpart G) and refers only to §63.115(d) of the HON, which
addresses the determination of the TRE index value. Further, the
proposed rule is not consistent with the proposed preamble (62 FR
46810) because it does not completely state how to make the Group
1 determination for continuous processes. The commenter
recommended revising this section to include appropriate
language.
Response: The Agency agrees that, as proposed, subpart PPP
was unclear about flow rate and concentration cutoffs, for the
purpose of determining whether or not a "combination of process
vents" was Group 1. Since proposal, the EPA revised the Group
determination for process vents from continuous unit operations,
so that the group determination is conducted on an individual
vent basis (see Section 2.4.4 of this document). Therefore, the
definition of "combination of process vents that are Group 1" was
revised to pertain only to process vents from batch unit
operations. In the proposed rule, the EPA intended to include
flow rate and concentration cutoffs in this definition and in the
definitions of "process vents," and "combination of process vents
that are Group 2." Therefore, the final rule includes these
cutoffs. However, the EPA did not feel that a revision was
necessary in §63.1428(h), in order to address the commenter's
concern.
2.4.13 Summing the Values
Comment; One commenter (IV-D-04) requested that the EPA
clarify what "summing the values" means in §63.1428(h)(1). This
section in the proposed rule describes how to determine the Group
status of "combinations" of continuous process vents by using
standard HON "TRE" calculation procedures, except for the
following difference: "summing the values in the individual
2-76
-------
process vent streams." This could be interpreted either of two
ways: summing the stream data (such as concentrations and flow
rates), or summing the TRE index values. The commenter believed
the former was intended.
Response: This comment is no longer relevant since the EPA
decided to conduct Group determinations on an individual stream
basis for process vents from continuous unit operations (see
Section 2.4.4 of this document).
2.4.14 Flares as a Reference Control Technology for Existing
and New Sources
Comment: Two commenters (IV-D-05 and IV-D-07) requested that
flares be allowed as a reference control technology for existing
and new source process vents. Commenter IV-D-05 stated that the
proposed rule (§63.1425(b)(2)(i)) does not specifically allow the
use of a flare as a control technology, as has been done in all
other standards impacting SOCMI type sources. However, according
to the commenter, the provisions of §63.1426(a) that require
owners or operators who use a flare to comply with the provisions
of §63.11(b), in conjunction with the exception from the
requirement to demonstrate the control efficiency, provide what
is essentially a reference control technology approach for
existing sources. The commenter also noted that in §63.1426(a),
in order to use only a flare to comply with the new source
epoxide standard for process vents in §63.1425(b) (1) (i) , the
owner/operator must submit a request in accordance with §63.6(g).
The commenter concluded that the issue seems to be whether the
destruction of epoxides in flares is significantly different from
that in other control devices, particularly in setting the new
source MACT floor level of control.
Commenter IV-D-07 provided three attachments (test data and
general descriptions of flare use and destruction efficiency)
generated by the EPA and other key regulatory agencies, which is
at least equal in quality to the information available for new
sources in the MACT data base. Based on these data, the
2-77
-------
coinmenter requested that the final rule adopt flares as a
reference control technology for both existing and new sources,
and the commenter suggested revisions to §63.1425(b)(1) and
(b)(2), accordingly. The commenter added that allowing flares to
be used as a reference control technology would result in a much
simpler compliance demonstration, with greatly simplified
monitoring, recordkeeping, and reporting burdens. In the event
that the EPA declines to adopt the flare as an alternative
control technology for new sources, the commenter noted that the
language of §63.1426(e)(2)(i) would need to be changed to refer
to §63.1426(a), since §63.1426(a) allows the owner or operator to
attribute more than 98-percent control efficiency to a flare
(potentially), if a §63.6(g) request is made (where alternative
control devices are approved). Another commenter (IV-D-05)
claimed that new sources should not have to make complicated
demonstrations that compliance is achieved, and thus the
commenter recommended revising §63.1426(a) by deleting the second
sentence.
Response: The EPA agrees that flares should be listed as a
reference control technology in §63.1425(b)(2)(i), for existing
sources, for three reasons. The primary reason for adopting
flares as a reference control technology for existing sources is
because the EPA believes that flares, when operated properly,
effectively meet the emission limit. Additionally, a precedent
has been set in other rules to allow flares as a reference
control technology. Thirdly, this revision to the rule would
simplify the compliance demonstration and reduce the monitoring,
recordkeeping, and reporting requirements. Therefore, the EPA
has added this option to §63.1425(b) (2) (i) . The EPA applied this
same rationale to Group 1 vents for making or modifying the
product and for process vents from catalyst extraction, and added
flares as a reference control technology for existing and new
sources with those emission points.
However, the EPA does not agree with the commenters
suggestion that flares should be listed as a reference control
2-78
-------
technology for all new sources. Commenter IV-D-07 provided data
from three reports stating that a flare can achieve a destruction
efficiency of 99.5 percent for epoxides. These data do not
support equivalence with the 99.9 percent destruction efficiency
new source requirement for EO and PO.
However, the EPA does agree with Commenter IV-D-07's request
that the language of §63.1426(e)(2)(i) be revised to allow for
the fact that flares may have been assigned a control efficiency
greater than 98 percent, if approval was previously granted by
the EPA in accordance with §63.6(g) of the General Provisions.
2.4.15 Use of Multiple Compliance Methods
Comment: One commenter (IV-D-07) requested that the EPA
establish that source owners/operators can mix and match
appropriate compliance methods for epoxide emissions. For
example, under §63.1425(b)(2), if a PMPU has more than one vent,
the commenter wanted to know if it is possible to control some
vents to 98-percent reduction efficiency by extended cookout and
to control others to <20 ppmv using a recovery, recapture, or
combustion device. The commenter stated that this allowance in
the final rule could be accomplished by rewording §63.1425(b)(2)
to require compliance with (i) and/or (ii), or (iii).
Response: First, the EPA would like to clarify that the
emission limit for an aggregated control efficiency allows for
some process vents within a PMPU to be controlled to different
levels, or some process vents to go uncontrolled, as long as the
overall control efficiency for emissions from all the process
vents within the PMPU equals 98 percent (for existing sources).
So, for these instances, there is flexibility for controls within
a PMPU.
Specifically though, the commenter is asking about complying
with different emission limitation formats for different process
vents within a PMPU. The EPA has considered the commenter's
request and has decided that the owner or operator can use either
the emission reduction format or the concentration cutoff within
2-79
-------
the same PMPU. The EPA is not allowing the emission factor
format for this provision of multiple emission reduction formats,
because the emission factor format sets a maximum allowable
amount of emissions for process vents in the PMPU.
2.4.16 Uses and Emission Point Locations of Nonepoxides
Comment: One commenter (IV-D-07) stated that, in addition
to the examples of uses that the EPA included in the preamble
(such as use of a nonepoxide HAP as an initiator, catalyst, or a
reaction solvent), nonepoxide HAP may be used as a viscosity
adjuster in or downstream of the reactor, or to provide special
properties to the final product. The commenter asserted that,
contrary to the EPA's apparent understanding that all process
vents are manifolded or otherwise connected (Supplementary
Information Document (SID), page 20 of April 25, 1997 letter on
estimated impacts), these vents may and do emit at different
points in the process.
Response: This comment presented the idea that nonepoxide
HAP process vent emission points exist downstream from the
reactor, and that these emission points are probably not
manifolded to the process vents from the reactor. The owner or
operator is required to perform a group determination on these
emission points. As discussed previously, the EPA has decided
that group determinations for process vents from continuous unit
operations will be made on an individual vent basis. Therefore,
this should not cause a problem for group determinations for
process vents from continuous unit operations downstream of the
reactor.
For process vent emissions from batch unit operations, the
EPA is maintaining the requirement that group determinations be
made on an aggregated vent basis. As stated earlier, even though
the group determination is made on an aggregated vent basis, the
facility is not required to physically combine the downstream
process vent streams to the reactor vent streams for purposes of
emissions control.
2-80
-------
2.4.17 Possibility of Dual Controls For Nonepoxide HAPs from
Making or Modifying the Product
Comment: One commenter (IV-D-07) referred to the proposal
preamble statement that a process vent from which nonepoxide HAP
are emitted may also be subject to the epoxide emission reduction
requirements, and noted that this is only true in the event that
such a vent originates at a point in the process where it
contains epoxide and requires control. The commenter also noted
that the EPA's assertion in the proposal preamble (that if a
combustion, recovery, or recapture device is used to reduce
epoxide emissions from the vent, then that same device would also
reduce the emissions of the nonepoxide HAP) is only correct if
the vents are, or reasonably can be, combined, and if the device
is effective for the nonepoxide HAP in question. For instance,
the commenter stated that scrubbers may not be effective for some
nonepoxide HAP materials used to make or modify the product. The
commenter continued by stating that if extended cookout is the
control technique utilized by the facility to reduce epoxide
emissions, then the nonepoxide HAP emissions would not be
affected (unless they were initiators or possibly catalysts) and
would need to be addressed separately (indicating a possible
requirement for dual controls).
The commenter offered three options for alternate approaches
to those suggested by the EPA for addressing process vent
emissions of nonepoxide HAP from making or modifying the product,
to resolve the possible dual control requirement problem:
Option 1; Require a group determination for only nonepoxide
HAP emissions that are not controlled along with epoxide
emissions. The commenter believes that this option offers a
workable approach, so long as the triggers selected and emission
reductions required are consistent with section 112 requirements.
Option 2: Require all HAP emissions (epoxide and
nonepoxide) from making or modifying the product to be reduced by
the amount specified in the proposed rule for epoxide emissions.
The commenter did not encourage the EPA to choose this option.
The commenter claimed that in order to justify this approach, the
2-81
-------
EPA must consider that the floor for nonepoxide emission
reductions is much less than the floor for epoxide emission
reductions, and then must justify an emission reduction
requirement above the MACT floor. The commenter believed that
this demonstration would be very difficult for the EPA to make,
since not all controls in the floor are effective in reducing
nonepoxide HAP emissions, and since floors of zero and 39-percent
reduction were found for existing and new sources.
Option 3: The commenter mentioned the option of eliminating
the group determination provisions for nonepoxide HAP emitting
and requiring a specified percent emission reduction from all
vents above a "de minimis" nonepoxide HAP level. The commenter
concluded that the "de minimis" option, as proposed, is not
defined well enough for full comment. However, the commenter
noted that the standard for Polymers and Resins I, subpart U,
provides precedent for a de minimis level threshold of 225
kg/yr/vent emissions as part of the definition for a batch front-
end process vent.
The commenter suggested, as an alternate option, that the
EPA focus the final rule on epoxide and catalyst recovery
emissions only, because nonepoxide HAP process emissions from
making or modifying the product are insignificant, even if
uncontrolled.
Response: The EPA considered the commenter's points and the
options suggested by the commenter. The final rule requires that
the group determination for nonepoxide HAP emissions be made
after the stream has been controlled for the epoxide emissions
(commenter's option 1) . The EPA believes that this approach
addresses the situation regarding the possibility of dual
control. If the epoxide control device also reduces nonepoxide
emissions, then that control would impact whether the vent (or
group of batch vents) is Group 1. Therefore, control of
nonepoxide emissions along with the epoxides will impact whether
controls are required at all. If the vent (or group of vents)
still has sufficient nonepoxide organic HAP emissions after the
2-82
-------
epoxide control device to satisfy the Group 1 criteria, the EPA
does not believe it is unreasonable to require an additional
control device to achieve the specified percent reduction of the
nonepoxide emissions.
Therefore, using the example given by the commenter (of a
scrubber controlling epoxide emissions), the TRE index or the
Batch ACT equation would be applied to the stream at the outlet
of the scrubber. Likewise, if ECO is used as the control option,
the Batch ACT equations (assuming that ECO is applicable for
batch unit operations only) would be applied to the stream after
the ECO is completed.
2.4.18 Clarification of "Schedule for Compliance"
Comment: One commenter (IV-D-05) requested that the EPA
confirm, perhaps in the preamble, that a "schedule for
compliance" (as used in §63 .1425(f) (7) (i) (B) and (f) (7) (ii) (B),
§63.1430(1) (2) and elsewhere in the rule) is not the same thing
as a "compliance schedule" in the General Provisions. Two
commenters (IV-D-04, IV-D-05) requested that the EPA change the
term "compliance schedule" in §63.1439(e) (6) (iii) (D) (1), and in
various other locations in the rule, to say "schedule for
compliance." This will avoid use of a term (compliance schedule)
from subpart A, which is under litigation.
Response: Because the Notification of Compliance Status is
the report in which compliance (or non-compliance) is ultimately
documented, the EPA decided that it was not necessary for owners
or operators of affected sources to submit a compliance schedule,
or a "schedule for compliance". For this reason, the terms
"compliance schedule" and "schedule for compliance" have been
removed throughout the final rule (including in §63.1423,
§63.1425(f)(7)(i)(B), §63.1425(f)(7)(ii)(B),
§63.1425(f) (7) (iii) (B), §63 .1425(f) (7) (iv) (B), §63.1430(i) (2),
§63.1439(e) (6) (iii) (D) (1) , §63 .1439 (e) (6) (iii) (D) (2.) , and the
title of §63.1422), and all requirements to report information in
a "compliance schedule" or "schedule for compliance" have been
2-83
-------
removed. In particular, the owner or operator is no longer
required to submit a schedule for compliance with the applicable
provisions after process changes. However, please note that this
final rule does not override other regulations that might require
compliance schedules (e.g., Title V requirements, NSPS, or RACT
standards).
2.4.19 THE PMPU Exemption from Reporting
Comment: One commenter (IV-D-05) claimed that
§63.1425(f)(7)(ii),(iii) and (iv) under the "Requirements for
process vents at PMPUs that produce polyether polyol products
using tetrahydrofuran" should not require the submission of a
report if there is some other basis for exemption, such as a flow
rate below 0.005 scmm for paragraph (ii) or (iv), or a HAP
concentration less than 50 ppmv for paragraph (iii). Therefore,
the commenter recommended revising the text as follows:
"(ii) Whenever a process change, as defined in
§63.115(e) of subpart G, is made that causes a Group 2
process vent with a TRE greater than 4.0 to become a Group 2
process vent with a TRE less than 4.0, the owner or operator
shall submit a report within 180 days after the process
change is made or the information regarding the process
change is known to the owner or operator unless the flow
rate is less than 0.005 scmm. This report may be included
in the next Periodic Report. The following information
shall be submitted..."
"(iii) Whenever a process change, as defined in
§63.115(e) of subpart G, is made that causes a Group 2
process vent with a flow rate less than 0.005 standard cubic
meter per minute (scmm) to become a Group 2 process vent
with a flow rate of 0.005 scmm or greater, and a TRE index
value less than or equal to 4.0, the owner or operator shall
submit a report within 180 days after the process change is
made or the information regarding the process change is
known to the owner or operator unless the organic HAP
concentration is less than 50 ppmv. This report may be
included in the next Periodic Report. The following
information shall be submitted..."
"(iv) Whenever a process change, as defined in
§63.115(e) of subpart G, is made that causes a Group 2
process vent with an organic HAP concentration less than 50
parts per million by volume (ppmv) to become a Group 2
process vent with an organic HAP concentration of 50 ppmv or
greater and a TRE index value less than 4.0, the owner or
operator shall submit a report within 180 days after the
2-84
-------
process change is made or the information regarding the
process change is known to the owner or operator unless the
flow rate is less than 0.005 scmm. This report may be
included in the next Periodic Report. The following
information shall be submitted..."
Response: The EPA agrees with the commenter, and the
changes suggested have been incorporated into the final rule.
The EPA also incorporated these changes into paragraphs
§63.1430(j) (1) through (3) of the final rule, which contain
similar reporting provisions for Group 2 continuous process vents
associated with making or modifying the product.
2.4.20 Clarification to Condenser Identity
Comment: One commenter (IV-D-04) requested that the EPA
clarify whether all condensers are equal. Section
63.1426 (c) (1) (i) (A) (1.) provides that the inlet sampling site must
be at the exit from the continuous unit operation "before any
recovery devices." The commenter asked about reflux condensers.
Reflux condensers fit the definition of "recovery device," but
they are considered to be part of the process unit (either the
reactor or distillation unit). The commenter questioned whether
the inlet sampling site should be before, or after, reflux
condensers. Another commenter (IV-D-05) claimed that the word
"at" seems to allow only one location, while the word "before"
seems to allow more than one location. For clarity, the
commenter recommended that the phrases in
§63.1426(c)(1)(i)(A)(1), (c)(1)(i)(B), (d)(2), and (d)(3)(i)(A)
be changed to read "at or after the exit from the ... before any
recovery device."
Response: The reflux condenser is considered to be part of
the unit operation to which it belongs (be that a condenser or
reactor), and is not a recovery device. To clarify this in the
final rule, the EPA has redefined "recovery device" as follows:
".Recovery device means an individual unit of equipment
capable of and normally used for the purpose of recovering
chemicals for fuel value (i.e., net positive heating value),
2-85
-------
use, reuse, or for sale for fuel value, use, or reuse.
Reflux condensers are not recovery devices. Examples of
equipment that may be recovery devices include absorbers,
carbon adsorbers, condensers (except reflux condensers-r
because they arc part of the reactor unit operation), oil-
water separators or organic-water separators, or organic
removal devices such as decanters, strippers, or thin film
evaporation units. When this oubpart roquirco compliance
with For the purposes of the monitoring, recordkeeping, or
recording reporting requirements of this subpart, recapture
devices are considered to be recovery devices."
As stated in the definition of recovery device, a reflux
condenser is not a recovery device. Therefore, the inlet
sampling site is after the reflux condenser, and before the
control or recovery device. The EPA agrees with Commenter IV-D-
05's suggestion to use the word "after" in the context of "after
the exit from the...before any recovery device." However, the
EPA feels having both "at" and "after" is not necessary.
Therefore, the EPA revised the appropriate phrases in
§63.1426(c) (1) (i) (A) (1), (c) (1) (i) (B) (1), (d)(2), and(d)(3)(i)
in the final rule, accordingly.
2.4.21 Small Combustion Device Performance Testing
Comment: One commenter (IV-D-05) noted that
§§63.1426(c)(1)(i)(C), (C)(4)(iv), and §63.1430(b)(2)(iv)
describe how to collect samples during performance tests on
boilers or process heaters with a design capacity of less than 44
megawatts. The commenter recommended revising the text as
follows:
"If a process vent stream is introduced with the combustion
air or as a secondary fuel into a boiler or process heater with a
design capacity less than 44 megawatts and is not otherwise
exempt from performance testing under this subpart, selection of
the location of the inlet sampling sites shall ensure the
measurement of total organic HAP or TOC (minus methane and
ethane) concentrations in all process vent streams and primary
and secondary fuels introduced into the boiler or process
heater."
2-86
-------
Response: The EPA does not believe that the language
suggested by the commenter adds clarity to the exemption, since a
control device may be exempt based on its design capacity or for
other reasons. Further, the recommended addition has not been
added to the final rule because a process vent is not subject to
performance testing (only a control device can be subject to or
exempt from performance testing). For these reasons, the EPA did
not incorporate the additional language requested by the
commenter.
2.4.22 Concentration Compliance
Comment: One commenter (IV-D-05) recommended removing "in
Ib." as the unit of measure from §63.1426 (c) (3) (i) (B) (I) and (2.)
and §63.1426 (c) (3) (i) (C). The commenter stated that these
provisions should use the percentage without needlessly
specifying the unit of measure. In addition, the commenter
claimed that the first sentence in §63 .1426 (c) (3) (i) (B) (1) and
(2.) were incomplete. Therefore, the commenter requested revising
the text.
Response: The Agency agrees that the phrase "in Ib." was
not necessary for a percentage, and has deleted it from
§63.1426 (c) (3) (i) (B) (JJ . However, this same change was not made
in §63.1426(c) (3) (i) (B) (2.) or in §63 .1426 (c) (3) (i) (C) , because
there is no percent in these paragraphs and the "Ib/hr" is
needed. The EPA agrees with the commenter's claim that the first
sentence in §63 .1426 (c) (3) (i) (B) (1) and (2.) were incomplete.
This grammatical error was corrected when these sections were
replaced with the new "worst case" language.
2.4.23 "Organic HAP" versus "HAP of Concern"
Comment: One commenter (IV-D-04) supported the use of the
phrase "HAP of concern" in §63.1426. However, there are several
places in that section where the EPA mentions "organic HAP"
without saying "of the HAP of concern." This could be
misunderstood to mean that every organic HAP species must be
2-87
-------
considered, whether or not the species is relevant to the
regulatory provisions with which industry is demonstrating
compliance. Therefore, the commenter requested that the EPA use
the phrase "of the HAP of concern" consistently when "organic
HAP" is mentioned in §63.1426, unless there is a specific reason
not to.
Response: The EPA appreciates the commenter' s support of the
phrase "HAP of concern," and has made an effort to be sure that
this phrase was used in the final rule, where necessary.
However, in many instances in §63.1426, it was not appropriate to
follow the term "organic HAP" with the phrase "of the HAP of
concern," because the term "organic HAP" is used as a modifier for
another term (such as "percent reduction efficiency") , so the EPA
disagreed with the commenter about using the phrase "organic HAP
of the HAP of concern" universally throughout §63.1426. One
instance in which the EPA made a change in keeping with the
commenter's suggestion, however, was in §63.1426(c)(3)(i)(A),
where it was in fact appropriate to follow the term "HAP" with
the phrase "of the HAP of concern," because clarification was
necessary in that paragraph.
2.4.24 Rename "Product" as "Reactor Liquid"
Comment: One commenter (IV-D-05) stated that
§63.1427(h) (1) (iii) and (h)(2)(iii) refer only to direct
measurement of epoxide concentration in the reactor liquid at the
end of ECO. Changing "reactor liquid" to "product" would allow a
producer to drop the product from the reactor and sample it in a
tank rather than forcing them to sample it in the reactor itself.
This will improve reactor utilization, allowing prompt emptying
of the batch (so another may be started) , and will not affect the
accuracy of the epoxide sample, since the reaction essentially
stops once the product is cooled. Therefore, the commenter
recommended revising these paragraphs to use the term "product"
in place of the term "reactor liquid."
2-88
-------
Response: The coinmenters want to sample the mixture as, or
after it leaves the reactor. The EPA has added language to the
final rule [in §63.1427 (h) (1) (iii) and (h)(2)(iii)] explaining
when/where the sample may be taken. The EPA considers this to be
a more appropriate solution to the comment than using the term
"product" instead of the term "reactor liquid," since "product"
was already defined in the proposed rule. Further, the EPA has
added the definition of reactor liquid to the final rule and
defined it as follows:
"Reactor liquid means the compound or material made in the
reactor, even though the substance may be transferred to
another vessel. This material may require further
modifications before becoming a final product, in which case
the reactor liquid is classified as an "intermediate." This
material may be complete at this stage, in which case the
reactor liquid is classified as a "product."
2.4.25 §63.1425(e)(1)(i)
Comment: For clarity and to avoid unnecessary repetition,
one commenter (IV-D-05) recommended revising the text in
§63.1425(e)(1)(i), as follows:
"If an owner or operator chooses to comply with the control
efficiency provisions in paragraph (b)(1)(i) or (b)(2)(i) of this
section, the owner or operator shall comply with the provisions
of paragraph (b) (1) (i) or (b) (2) (i)—of thio aootion by
determining determine the epoxide emissions before and after
control."
Response: The EPA agrees with the commenter that
§63.1425(e)(1)(i) needs more clarity. The proposed
§63.1425(e)(1)(i) could have been interpreted to mean that
performing the determination of the controlled and uncontrolled
emissions would be equivalent to complying with §63.1425(b)(1)(i)
or (b)(2)(i), and this would have been an incorrect
interpretation. Further, the EPA agrees that, as proposed,
§63.1425(e) was redundant with many of the requirements in
§63.1425(b) through (d). Therefore, the EPA has "reserved"
§63.1425(e) in the final rule, and the appropriate process vent
control requirements are now contained in §63.1425(b) through (d)
only, as appropriate.
2-89
-------
2.5 EXTENDED COOKOUT AS A CONTROL OPTION
2.5.1 In Support of the Inclusion of ECO
Comment: Five commenters (IV-D-03, IV-D-04, IV-D-05, IV-D-
07, IV-D-10) fully supported the proposed rule's inclusion of the
concept of "extended cookout (ECO)," as a pollution prevention
technique. One commenter (IV-D-10) noted that their company had
worked closely with the EPA to define the calculation procedures
to demonstrate that equivalent epoxide emission reductions that
can be obtained using ECO compared to conventional control
technology.
Response: The Agency appreciates the commenter's support.
2.5.2 ECO Compliance Demonstration
Comment: One commenter (IV-D-04) noted that the proposed
rule would require ECO to reduce emissions by 98 percent;
however, the rule does not seem to require the owner or operator
to demonstrate that a 98-percent emission reduction (or any other
level of emission reduction) is actually being achieved. In a
follow-up telephone conference with the commenter (Docket Item
IV-E-1), the commenter explained that his first impression of
this requirement was that it was not objective enough, in
comparison with other rules. However, when the commenter
reviewed the section again, he realized that the rule provided
adequate steps for demonstrating compliance with the rule.
Further, commenter IV-G-02 submitted a late comment after
reading Commenter IV-D-04's comment, in which Commenter IV-G-02
disagreed with the above comment and stated that a demonstration
of the percentage emission reduction achieved by ECO is
specifically required in §63.1427(a).
Response; The EPA agrees with Commenter IV-G-02's late
comment, which stated that a demonstration of the percentage
emission reduction achieved by ECO is specifically required in
§63.1427(a). However, the word "demonstrate" is not specifically
2-90
-------
stated in §63.1427(a); therefore, the EPA decided to revise the
language in the final rule as follows:
"(a) Owners or operators of affected sources that produce
polyether polyol products using epoxides, and that are using
ECO extended cookout—(ECO)—as a control technique to reduce
epoxide emissions in order to comply with percent emission
reduction requirements in §63.1425(b)(1)(i) or (b)(2)(ii)
shall demonstrate that the specified percent emission
reduction is achieved by determining determine the batch
cycle percent epoxide emission reduction for each product
class in accordance with the provisions of paragraphs (b)
through (g) of this section...."
2.5.3 Flexibility of the Determination of a Site-specific
Onset of ECO
Comment: Three commenters (IV-D-04, IV-D-05 and IV-D-07)
supported defining the onset of ECO as the point in time when the
combined unreacted epoxide concentration in the reactor liquid is
equal to 25 percent of the concentration of epoxides at the end
of the epoxide feed. However, there was disagreement over
whether site-specific ECO onset determination should be allowed
in the final rule.
Commenter IV-D-05 supported the proposed provisions that
allowed individual producers the opportunity to provide their own
economic justification for the onset of ECO, and noted that the
mechanism for a request for a site-specific ECO onset was the
Precompliance Report, which is required to be submitted one year
before the compliance date. However, the commenter stated that
an owner or operator should be allowed the opportunity to change
the onset point. They recommended that the opportunity to
establish a different ECO onset point should not be limited to
the precompliance stage.
Commenter IV-D-07 added that an individual producer should
have the ability to prove a different starting point based on
that producer's economics. The commenter noted that the proposed
default definition of the onset of ECO was based on average
conditions, and that a producer may well have very different
economics from those presented in the model calculations. The
commenter maintained that the ability to request an alternative
2-91
-------
definition is appropriate and necessary. The commenter also
added that the mechanism provided in §63.1427 (c) (3) for an
alternate determination, and the reference to Item ll-B-7 of the
docket contained in the proposal preamble, are appropriate.
However, this commenter expressed concern that the basis
referenced in the proposal preamble will be lost over time, and
suggested that this reference be included in the actual rule at
§63.1427(c), or, at the very least, that the reference be
included in the preamble to the final rule.
In contrast, Commenter IV-D-04 was concerned that there was
too much flexibility in the requirements describing how to
establish the onset of ECO on a site-specific basis. The
commenter cautioned that selecting the onset based on the
economics of polyether polyol production has the potential for
being misused. The commenter explained in the comment that the
operator may select an earlier onset of ECO, making it easy to
demonstrate a 98-percent emission reduction and to justify this
ECO onset on the basis of economic factors. The commenter
cautioned that economics can be very subjective. The commenter
also cautioned that, in contrast to demonstrations of compliance
with other control devices, the determination of the onset of ECO
does not have standardized procedures. The commenter did state
that they had confidence in the default value for the onset of
ECO.
Further, the commenter pointed out that requiring an
emission reduction of 98 percent for ECO might still allow
emissions to exceed the applicable emission factors described in
§63.1425(b)(1) or (b)(2). As assurance that the ECO provisions
are used appropriately, the commenter requested that EPA require
that owners and operators using ECO with an onset other than the
default onset listed in the rule comply with the applicable
emission factor in §63.1425(b)(1) or (b)(2) and not have the
option of using the 98 percent emission reduction for
demonstrating compliance with the rule. According to the
commenter, this will serve two purposes: (1) it will assure that
2-92
-------
ECO achieves substantial "real" emission reductions; and (2) it
will provide for a compliance demonstration for ECO.
Commenter IV-G-02 submitted comments in response to the
issues raised by Commenter IV-D-04. This commenter pointed out
that the economic evaluation which was accepted by the EPA and
that formed the basis for the default ECO onset is very simple.
Provisions for investment in the unit or many other significant
costs that should be considered for true profitability were not
used in the analysis. The evaluation was simple and conservative
to avoid any significant debate over cost issues that producers
would be reluctant to divulge for competitive reasons or that
would be difficult for regulators to verify from outside sources.
The commenter believed that not allowing the site-specific ECO
onset would provide a disincentive to producers who elect to use
ECO as a control technology.
In response to Commenter IV-D-04's concern over site-
specific onset, Commenter IV-G-02 stated that establishing a
site-specific onset is not arbitrarily selected by the owner or
operator. The owner or operator must submit a request for an
alternative ECO onset, and this alternative must be approved by
the Administrator. The commenter points out that this is a
typical EPA "alternative standard-getting" process used in many
Regulatory settings. The commenter also points out that the data
required to be provided by the owner or operator for a site-
specific alternative ECO onset request are readily available.
In response to Commenter IV-D-04's request that the EPA
require ECO to achieve the applicable emission factor, Commenter
IV-G-02 stated that there is not need to "safeguard" ECO
reductions by requiring compliance with an emission factor. This
would remove one of the compliance options for these sources and
provide a potential competitive advantage for sources using an
add-on control technique. The commenter stressed that sources
that use ECO should not have to pay a penalty by complying with a
different standard that those who have elected to use
manufacturing methods which emit greater quantities of epoxide to
an add-on control device.
2-93
-------
Response: First, the EPA agrees with Commenter IV-G-02 that
sources using ECO should not be restricted to achieving
compliance with the applicable emission factor. In the final
rule, owners or operators using ECO may comply by using either
the applicable percent reduction or the applicable emission
factor.
However, the EPA also recognizes the concern raised by
commenter IV-D-04 that changing the ECO onset, and thus the point
for determining the uncontrolled emissions, could significantly
impact the percent reduction achieved by the ECO. The EPA also
shares the concern that economics can be very subjective,
although the EPA agrees with the Commenter IV-G-02 that the
proposed criteria for a site-specific request would not allow the
arbitrary selection of an alternative ECO onset.
However, the EPA disagrees with the statement by Commenter
IV-G-02 that the approval of a site-specific onset by the
Administrator would represent a "typical" EPA alternative process
used in many regulatory settings. The Agency does routinely
evaluate alternative control technologies and their equivalency
to the control technologies or levels specified in regulations.
Typical evaluations may involve economics, with respect to the
cost of the technology in relation to the corresponding emission
reduction (i.e., the cost effectiveness). However, these
decisions are made in assessing alternative requirements for
categories and/or subcategories of sources, and are not made for
individual sources. In reassessing the appropriateness of a
site-specific onset option, the EPA concluded that the subjective
nature of the option could result in different levels of control
between facilities in the same subcategory. That result would be
inconsistent with the concept of MACT. Therefore, in the final
rule, the EPA has removed the option of requesting and
establishing a site-specific ECO onset point.
2.5.4 ECO Requirements for Monitoring
Comment: One commenter (IV-D-04) requested that the EPA
specify in the final rule appropriate compliance demonstration
2-94
-------
requirements, monitoring requirements, bypass requirements,
parameter level requirements, etc., for emission control devices
that are used to supplement ECO. According to the commenter,
these requirements appear to be missing in the proposed rule.
The commenter also suggested that the EPA review the entire rule
to insure that all requirements apply appropriately to all
control devices, regardless of whether those devices are used
with (or without) ECO.
Commenter (IV-G-02) provided comments after the end of the
comment period, in response to Commenter IV-D-04's input.
Commenter IV-G-02 did not see any "gap" in the requirements for
ECO demonstrations, monitoring, etc. This commenter interpreted
§63.1427 as providing for compliance demonstrations, monitoring,
etc. for the ECO portion of the compliance alternative, and
believed that §63.1427 clearly refers the source owner or
operator to other portions of the rule for demonstrations and
monitoring for using other control devices.
Response: The Agency called Commenter IV-D-04 on April 23,
1998 (See Docket Item IV-E-2). The commenter explained that
compliance demonstration requirements, monitoring requirements,
bypass requirements, and parameter level requirements are given
for ECO, but not for those emission control devices that are used
to supplement ECO. The final rule includes compliance
demonstration requirements, monitoring requirements, bypass
requirements, and parameter level requirements for those emission
control devices that are used to supplement ECO. The following
language was added to §63.1427(a):
"If additional control devices are used to further reduce
the HAP emissions from a process vent already controlled by
ECO, then the owner or operator shall also comply with the
testing, monitoring, recordkeepinq, and reporting:
requirements associated with the additional control device,
as specified in §§63.1426. 63.1429, and 63.1430.
respectively."
2-95
-------
2.5.5 Accuracy "Buffer" for ECO Emissions Calculations
Comment: One commenter (IV-D-07) recoinmended that guidance
on an acceptable accuracy range be added to the provisions in
§63.1427(a)(2), which requires measurement of the percent
reduction for each product to verify the accuracy of the
estimation method selected, in order to resolve conflicts when
measurements and calculations are not exactly equal. The
commenter suggested that kinetic models are likely to be accurate
to within +/- 10 percent, for typical situations. The commenter
noted that this request is valid only if engineering calculations
are not allowed as requested in a previous comment (Section
2.12.8) or if a compliance test is conducted for ECO. The
commenter recommended that §63.1427(a)(2) be modified as follows:
"The owner or operator may determine the batch cycle
percent epoxide reduction by directly measuring the
concentration of the unreacted epoxide, or by using
process knowledge, reaction kinetics, and engineering
knowledge. If the owner or operator elects to use any
methods, other than direct measurement, the percent
reduction must be determined by direct measurement for
one product for each PMPU to verify the accuracy of the
estimation method selected. The alternate method of
estimating the concentration of unreacted epoxide is
acceptable if it is within +/- 25 percent of the result
of direct measurement."
Response: The Agency agrees with the concept of needing an
"accuracy buffer," but, does not believe that 25 percent is an
appropriate value. The EPA believes that 10 percent is a more
reasonable buffer to allow use of the calculations without
adjustment. For situations where the difference between the
calculation and the measurement is between 10 and 25 percent, the
calculated values may be used, but need to be adjusted. The EPA
revised the text suggested by the commenter and incorporated it
into §63.1427(a)(2) in the final rule, as follows:
(2) The owner or operator may determine the batch
cycle percent epoxide emission reduction by directly
measuring the concentration of the unreacted epoxide, or by
using process knowledge, reaction kinetics, and engineering
knowledge, in accordance with paragraph (a)(2)(i) of this
section.
2-96
-------
(i) If the owner or operator elects to use any method
other than direct measurement, the percent reduction must
epoxide concentration shall be determined by direct
measurement for one product for each PMPU to verify the
accuracy of the catimation method selected from each product
class and compared with the epoxide concentration determined
using the selected estimation method, with the exception
noted in paragraph (a)(2)(ii) of this section. If the
difference between the directly determined epoxide
concentration and the calculated epoxide concentration is
less than 25 percent, then the selected estimation method
will be considered to be an acceptable alternative to direct
measurement for that class.
(ii) If uncontrolled epoxide emissions prior to the
end of the ECO are less than 10 tons per year (9.1 meqaqrams
per year) . the owner or operator is not recruired to perform
the direct measurement required in paragraph (a)(2)(i) of
this section. Uncontrolled epoxide emissions prior to the
end of the ECO shall be determined by the procedures in
paragraph (d)(1) of this section.
The proposed rule contained an error in the second sentence
in §63.1427(a)(2). That sentence should have read; "If the owner
or operator elects to use any methods other than direct
measurement for one product for each PMPU product class to verify
the accuracy of the estimation method selected." The EPA
intended that the comparison of the engineering calculations to
the direct measurement be conducted on a product class basis,
since each product class behaves differently.
2.5.6 Clarification of the Definition of "Emissions" in ECO
Destruction Efficiency Calculation
Comment: One commenter (IV-D-04) cited three locations in
§63.1427(c) that use the word "emissions," but the intended
meaning of the word is unclear: (1) "the uncontrolled emissions
for the batch cycle,-" (2) "the epoxide emissions prior to the
onset of the ECO;" and (3) "the epoxide emissions at the onset of
the ECO." The commenter requested clarification of the word
"emissions" in these cases. In the first case, the commenter was
unsure whether "emissions" meant the emissions that would occur
if there were no ECO, or emissions that remain after ECO and are
either sent to a control device, or released (without "control"
after ECO) to the air. In the second and third instances, the
2-97
-------
commenter was unsure about whether "emissions" meant actual
emissions that really occur, or emissions that would have
occurred if there were no ECO. Commenter IV-D-05 suggested
replacing the language in the second case with the following
language: "...calculate the uncontrolled epoxide
emissions... calculating the epoxide emissions, if any, prior to
the onset of the ECO."
As a rebuttal to Commenter IV-D-04's comment, another
Commenter (IV-G-02) stated that Equation 7 in §63.1427(c)(1)
describes how to calculate "the uncontrolled emissions for the
batch cycle." These are the calculated epoxide emissions that
would occur in the absence of ECO if the reactor were opened at
the time that ECO commenced plus any epoxide emissions that
actually occur prior to that time.
Commenter (IV-G-02) also stated that the "epoxide emissions
prior to the onset of the ECO" are those emissions that actually
occur from process vents (for example, those that occur as a
result of initial reactor charge). The commenter added that the
method for calculating these emissions is specified in
§63.1426(d) in conjunction with the definition of "Epoxide" in
§63.1423(b).
Commenter (IV-G-02) noted that the "epoxide emissions at the
onset of the ECO" are described in §63.1427 (c) and are calculated
according to §63.1427(b). These are the emissions that would
occur from the reactor contents if there were no ECO. They are
calculated as 25 percent of the reactor epoxide contents unless
the owner/operator justifies an alternate starting point.
Commenter (IV-G-02) ended by saying that the relation between
uncontrolled epoxide emissions and the epoxide emissions at the
onset of ECO is described in Equation 7.
Response: The Agency defines ECO as a control option;
therefore, uncontrolled emissions are emissions that would have
occurred had there not been an ECO. This definition is in accord
with the interpretation offered by Commenter IV-G-02. Further,
the "epoxide emissions at the onset of the ECO" that the Agency
2-98
-------
is referring to are any actual emissions that occur before the
beginning of the ECO, which is again in agreement with Commenter
IV-G-02's interpretation. Therefore, no changes have been made
to the final rule as a result of Commenter IV-D-07's concern.
2.5.7 First Order for Epoxide Reactions
Comment: One commenter (IV-D-07) responded to the EPA's
request for documentation to support or refute the first order
reaction rate equation used in the ECO calculation (62 FR 46814).
The commenter has developed proprietary reaction models based on
extensive laboratory and commercial plant operation data, but
this information is not public information. However, they
included a copy of a paper on reaction kinetics by K. Nagase and
Y. Sakaguchi, which was presented at the 12th annual meeting of
the Japanese Chemical Society, April 1959, which the commenter
considered to be a good source of information. The commenter
noted that their proprietary models do use a first order reaction
rate with respect to epoxide concentration to describe the
reaction.
Response: The Agency appreciates this input and has
retained the assumption that the polyol production reaction is
first order with respect to the epoxide.
2.5.8 Update of Product Class List
Comment: One commenter (IV-D-04) stated that in situations
where a change in operating conditions causes a product to move
from one product class to another, §63.1427(1)(3)(i)(A) requires
industry to "update the list of products for the product class,"
and then §63.1427(1)(3)(i)(B) requires submission of a report
updating the list for "the product class." The commenter
requested that these sections be revised so that the owner or
operator knows which updated list of products within a product
class to update: the class that this product has just left, or
the class that this product has just entered, or both lists.
2-99
-------
Commenter IV-D-05 requested the following change to the text
in §63.1427(1) (3) (i) (B) , as follows:
"Within 180 days of the change in operating conditions
for the polyether polyol product, the owner or operator
shall submit a report updating the product list originally
submitted for the product class. This information may be
submitted along with the next Periodic Report."
Response: The EPA agrees that the proposed rule did not
make it clear which product lists need to be updated when
operating conditions cause a product to move from one product
class to another. The EPA has revised what was
§63.1427(1) (3) (i) (B) , which is §63 . 1427 (m) (3 ) (i) (B) in the final
rule, to read as follows:
"Within 180 days after e£ the change in operating
conditions for the polyether polyol product, the owner or
operator shall submit a report updating the product lists.
originally previously submitted for the product class. This
information may be submitted along with the next Periodic
Report . "
2.5.9 Product Class Definition
Comment : One commenter (IV-D-07) requested the following
technical correction in proposed §63.1427(1) (1) (i) (E) :
"The group product classification should depend on EO/PO
ratio at the end of the batch, not the ratio during the batch. "
Further, Commenter IV-D-05 commented that he understood that
the concentration in §63. 1427 (i) (1) (iv) is to be determined at
the "onset of the ECO" rather than at the "end of the onset of
ECO." Therefore, the commenter recommended revising the text
accordingly .
Response : In the proposed rule, §63. 1427 (i) (1) (i) (E) did not
specify when in the batch the EO/PO ratio should be determined.
The EPA believes that the time in the batch at which the EO/PO
ratio is to be determined should be specified, and proposed
§63.1427(1) (1) (i) (E) (§63.1427(j) (1) (i) (E) in the final rule) has
been revised to specify that this ratio shall be determined at
the end of the epoxide feed, because, according to some
2-100
-------
coinmenters (docket item no. IV-G-10) , the EO/PO ratio changes
during a batch, but has leveled out by the time that extended
cookout begins. Further, §63.1427(j)(1)(iv) in the final rule
has been revised as requested by commenter IV-D-05.
2.5.10 Pressure Measurements
Comment: One commenter (IV-D-05) referred to §63.1427(g),
which requires the calibration of pressure measurement devices
"in accordance with manufacturer's recommendations." The
commenter claimed that this has the unintended effect of
transferring the setting of enforceable regulatory standards to
third parties. In addition, some manufacturer's recommendations
may be inappropriate for purposes of this rule. Therefore, the
commenter recommended revising the text using language adopted
into the amended HON, as did §63.1429(a) at proposal.
Response: Since the Agency has determined that the language
used in the HON and §63.1430(a) is also appropriate in
§63.1427(g), the final rule has incorporated similar language
into §63.1427(g), as follows:
"(g) Determination of pressure. The owner or operator
shall determine the total pressure of the system using
standard pressure measurement devices calibrated ift
accordance with according to the manufacturer's
specifications or other written procedures that provide
adequate assurance that the ecruipment would reasonably be
expected to monitor accurately."
2.5.11 ECO Recordkeeping
Comment: One commenter (IV-D-05) asserted that in
§63.1427(i) (1) (i)(C) it makes no sense to include the number of
-OH groups in the catalyst feed. The metal ion is the active
species for these reactions, not the base. In addition, some of
these reactions may be self catalyzed or acid catalyzed, and
there will be no -OH groups in these systems. The appropriate
measure should be the number of -OH groups (or, better, the
number of reactive sites) in the starting material that is being
2-101
-------
reacted with the epoxide. Therefore, the commenter recommended
revising the text, as follows:
"(i) Operating conditions of the product class, including
(A) Pressure decline decay curve
(B) Minimum reaction temperature
(C) Number of -OH groups in the catalyot food raw material
(D) Minimum catalyst concentration
(E) The EO/PO ratio
(F) Reaction conditions, including the size of the reactor
or batch"
Response: The EPA agrees that the term "raw material" is a
better description of where the reactive hydrogens are than the
phrase "catalyst feed," and has made the suggested change to what
was §63.1427(1)(1)(i)(C), and is §63.1427(j)(1)(i)(C) in the
final rule. Further, in §63.1427(j) (1) (i) (C) , the EPA has
replaced the term "number of -OH groups in the catalyst feed"
with the term "number of reactive hydrogens in the raw material,"
because the phrase "reactive hydrogens" is consistent with the
terminology used in the definition of polyether polyols. Also,
as discussed in an earlier response in this section,
§63.1427 (j) (1) (i) (E) has been revised to specify that the EO/PO
ratio shall be determined at the end of the epoxide feed.
2.6 MONITORING REQUIREMENTS FOR PROCESS VENTS
2.6.1 Monitoring Requirement Exemptions for the Process Vent
Requirements
Comment: One commenter (IV-D-05) stated that, according to
§63.1422 (j), if the unit is otherwise exempt from performance
tests because it complies with another law, such as the Resource
Conservation and Recovery Act (RCRA) requirements, the owner or
operator will comply with the monitoring requirements of that
law. To clarify that §63 .1429 (a) (3) does not impose requirements
on equipment otherwise exempt from testing, the commenter
recommended adding the following wording to §63.1429(a)(3), such
that it reads:
"(3) Where a boiler or process heater of less than 44
megawatts design heat input capacity and not otherwise
2-102
-------
exempt from performance testing under this subpart, is
used...."
Response: The EPA does not, in subpart PPP, exempt a boiler
from performance testing because it complies with another law.
Section 63.1422(j) allows the owner or operator the choice of
whether to comply with the monitoring, recordkeeping and
reporting requirements of this rule or the RCRA rule, if both
apply. Therefore, the EPA believes the suggested change is
inappropriate, and has not made that change in the final rule.
2.6.2 Scrubber Monitoring Requirements
Comment: Three commenters (IV-D-03, IV-D-05, IV-D-07)
referred to §63.1429(a)(4) of the proposed rule which specifies
scrubbing liquid temperature and specific gravity as appropriate
monitoring parameters. The commenters stated that specific
gravity is not an appropriate parameter for "once through"
scrubber systems and should be deleted. It was recommended that
scrubbing liquid flow rate, which is a key parameter for most
scrubber systems, be added to this section in the final rule.
Two other commenters (IV-D-05, IV-D-07) agreed, and also
suggested monitoring pH, if the scrubber is an assisted scrubber.
Response: The Agency realizes that the scrubbers operated
at the polyether polyol production facilities are once-through
scrubbers, which are similar to those added after combustion
devices to reduce emissions of halogens. Therefore, the final
rule requires the same monitoring parameter requirements as for
halogen reduction scrubbers in the HON. Specifically, in the HON
[§63.114(a) (4) (i) and (ii)] , a pH monitoring device to monitor
the pH of the scrubber effluent and a flow meter are required to
be used for scrubbers used with an incinerator, boiler, or
process heater in the case of halogenated vent streams.
In the final rule, §63.1429(a)(4) reads:
"(4) Where an absorber is used, a scrubbing licruid flow
rate meter or a pressure temperature monitoring device and a
specific gravity monitoring device arc required,—each
2-103
-------
equipped with a oontinuouo recorder is required and should
be equipped with a continuous recorder. If an acid or base
absorbent is used, a pH monitoring device to monitor
scrubber effluent is also recruired. If two or more
absorbers in series are used, a scrubbing liquid flow rate
meter, or a pressure monitoring device, equipped with a
continuous recorder, is recruired for each absorber in the
series. An owner or operator may submit a request to
instead install the scrubbing licruid flow rate meter, or a
pressure monitoring device, equipped with a continuous
recorder, on only the final absorber in a series, in
accordance with the alternative parameter monitoring
reporting requirements in §63.1439(f).
2.6.3 Monitoring of Multiple Absorbers
Comment: One commenter (IV-D-05) maintained that
§63.1429(a)(4) does not indicate how monitoring is to be
performed where two or more absorbers are used, and requested
that EPA clarify this provision.
Response: In general, the EPA believes that if multiple
scrubbers are used in series to achieve the required percent
efficiency, then it is important that the performance of each
scrubber be monitored. However, the EPA recognizes that there
could be circumstances where monitoring only one scrubber would
be adequate. Under this circumstance, the EPA believes that a
request for alternative monitoring can be submitted in accordance
with §63.1439(f). The following language was added in
§63.1429(a)(4):
"If two or more absorbers in series are used, a scrubbing
liquid flow rate meter, or a pressure monitoring device.
eguipped with a continuous recorder, is required for each
absorber in the series. An owner or operator may submit a
request to instead install the scrubbing liquid flow rate
meter, or a pressure monitoring device, equipped with a
continuous recorder, on only the final absorber in a series,
in accordance with §63 .1439(f)."
2.6.4 Alternative to §63.1429(a)(7)
Comment: One commenter (IV-D-05) claimed that, in
§63.1429(a)(7), an organic monitor with a continuous recorder
should be acceptable as an alternative to §63.1429(a)(1) and
2-104
-------
(a)(3) through (a)(6). This commenter recommended revising the
text in §63.1429(a)(7) as follows:
"(7) As an alternate to paragraphs (b) (4)—through—(b) (6)
(a) (1) and (a) (3) through (a) (6) of this section, the owner or
operator may install an organic monitoring device equipped with a
continuous recorder."
Response: The proposed rule allowed the monitoring of
organic compound concentration as an alternative to monitoring
operating parameters for absorbers, condensers, and carbon
adsorbers (as was required in paragraphs (a)(4), (5), and (6)).
The commenter is requesting that the monitoring of organic
compound concentration also be allowed as an alternative to
monitoring operating parameters for incinerators and boilers or
process heaters (paragraphs (a)(1) and (3)). The EPA agrees with
the commenter that an organic monitor with a continuous recorder
could be an acceptable alternative to §63.1429(a)(1) and (3)
through (6) in some instances. However, the commenter did not
provide sufficient data or rationale on the monitoring of organic
compounds at the exit of combustion devices to convince the EPA
that such an alternative should be generally allowed in the final
rule. However, the EPA believes that organic compound
concentration monitors can be used on a site-specific basis.
Owners or operators wishing to monitor organic compound
concentration may submit a request to monitor these parameters in
accordance with the alternative monitoring parameters provisions
in §63.1439(f).
The EPA appreciates the commenter's pointing out the cross-
referencing error in the proposed rule. The cross-referencing
error was corrected in the final rule.
2.6.5 Flow Indicator Specifications
Comment: Two commenters (IV-D-04, IV-D-05) requested that
the EPA delete the provisions of §63.1429(c)(3), which implied
that computer monitoring of a bypass line damper or valve
position [§63.1429(c)(3)] was different than operating a flow
indicator [§63.1429 (c) (1)]. In addition, one commenter (IV-D-04)
2-105
-------
recommended deleting two other paragraphs (or portions thereof)
that relate to §63.1429(c)(3) because, once §63.1429(c)(3) was
deleted, they would have no meaning:
(1) In §63.1430(d)(4), delete the words "or where computer
monitoring of the bypass damper or valve is used to comply with
§63.1429(c) (3) . "
(2) Delete paragraph 63.1430(d)(4)(ii).
Response: The EPA removed the proposed paragraph
63.1429(c)(3). The EPA believed that §63.1429(c)(1) and (2) were
sufficient for specifying the monitoring requirements associated
with bypass lines, since the definition of a flow indicator in
subpart G includes computer monitoring. The other changes
mentioned by commenter IV-D-04 have also been made in the final
rule, as a result of the removal of §63.1429(c)(3), in order to
eliminate references to that paragraph which no longer exists in
the final rule.
2.7 GROUP DETERMINATION FOR NONEPOXIDE HAP EMISSIONS
2.7.1 Group Determination Records When the TRE Index is
Between 1.0 and 4.0
Comment: One commenter (IV-D-04) referred to
§63.1428(h) (2) (ii), which requires a report if the TRE index
value after a process change is between 1.0 and 4.0. According
to the commenter, this seems to presume that, before the process
change, the TRE index value was greater than 4.0. Instead of
presuming it, the section should be specific. The commenter
claimed that industry should not have to submit a report if the
TRE index value was already between 1.0 and 4.0 before the
process change, and the value is still in that range after the
process change. The commenter suggested the following revisions
to §63.1428(h)(2)(ii):
"(ii) Where the recalculated TRE index value is less than or
equal to 1.0, or where the TRE index value before the process
change was greater than 4.0 and the recalculated TRE index value
2-106
-------
is less than or equal to 4.0 but greater than 1.0, the owner or
operator shall submit a report ...."
Response: The Agency agrees with the commenter's revision
and has incorporated it into the final rule.
2.8 PROCESS VENT RECORDKEEPING AND REPORTING REQUIREMENTS
2.8.1 "Up-to-date" Process Vent Records
Comment: One commenter (IV-D-05) stated that §63.1430(b) and
§63.1430(f)(5) require records to be kept "up-to-date," but
stated that this term has no meaning for one-time records and is
unnecessary for recurring records, since owners or operators are
already required to obtain and retain each record when the time
arises. Therefore, the commenter is recommending that this term
be deleted.
Response: The EPA has removed the phrase "up-to-date" from
these recordkeeping requirements in the final rule, because that
phrase did not actually state the frequency with which records
were to be "up-dated." The EPA feels that the proposed
regulatory text, minus the phrase "up-to-date", is sufficient to
convey the EPA's intent, which was that the owner or operator
keep these records current.
2.8.2 Flares Compliance Demonstration
Comment: Two commenters (IV-D-04, IV-D-05) expressed
concern that §§63.1430(b)(1)(ii) and (b)(1)(iii) imply that a
compliance demonstration for flares is required by §63.11(b) of
subpart A. They maintained that they have searched §63.11(b)
very carefully and cannot find anything in that section that
specifically requires a compliance demonstration. Therefore, the
commenters requested that, if EPA wants a compliance
demonstration for flares, that the requirement be included in the
final rule, specifying what elements are included in the
demonstration and what the deadline is.
2-107
-------
Response: At proposal, subpart PPP referred to §63.11(b) for
determining compliance with the flare requirements. However, as
the commenters have pointed out, §63.11(b) does not actually
require a compliance demonstration. To remedy this situation, in
the final rule the EPA has added a requirement to perform the
compliance demonstration for flares to §63.1437(c). Appropriate
changes have also been made in §63.1430(b)(1)(iii) and other
parts of the rule, to replace the HON reference to §63.11(b) with
a reference to the provisions in §63.1437(c).
In the final rule, the EPA has added §63.1437(c) to make it
clear that a compliance demonstration for flares must be
conducted using the provisions found in §63.11(b). Specifically,
the owner or operator is required to (1) conduct a visible
emission test, (2) determine the net heating value of the gas
being combusted, and (3) determine the exit velocity. In each
case, the provisions specify that these parameters be determined
in accordance with specific paragraphs in §63.11. Section
63.1437(c) also specifies that an owner or operator is not
required to conduct a performance test to determine percent
emission reductions or outlet organic HAP or TOC concentrations
for flares. In addition, the final regulatory language specifies
that a previously conducted flare compliance demonstration may be
used to demonstrate compliance for the purposes of subpart PPP,
provided that no deliberate process changes have been made since
the compliance demonstration, or that the results of the
compliance demonstration reliably demonstrate compliance despite
process changes.
2.8.3 Records for Start-ups, Shutdowns, Malfunctions, and
Periods of Non-operation
Comment: Two commenters (IV-D-04, IV-D-05) maintained that
§63.1430(d)(2)(i) borrows some, but not all, of the relevant HON
amendments, and it should be revised to address start-ups,
shutdowns, malfunctions, and periods of non-operation of a
relevant portion of the process. Additionally, it should refer
to cessation of the monitored emissions, rather than emissions
2-108
-------
generally. While it may be appropriate to record all data, the
commenters asserted that averages should exclude periods of
system start-up, shutdown, breakdowns, repairs, calibration
checks, and zero (low-level) and high-level adjustments. To be
consistent with the HON, the commenters recommended changing the
text. Commenter IV-D-05 provided revised wording and commenter
IV-D-04 supported the text suggested by commenter IV-D-05.
Response: The EPA agreed that §63.1430(d)(2)(i) should
apply to "specific portions" of processes, and has amended the
language in §63.1430(d)(2)(i) accordingly. In addition, the EPA
has added a provision stating that monitoring data collected
during periods of start-up, shutdown, or malfunction are not to
be included in the daily average. The EPA does not agree with
the commenters that this paragraph should depend only on the
cessation of "monitored emissions," rather than on the cessation
of all emissions.
2.8.4 Records Related to Group Determination
Comment: One commenter (IV-D-05) supported not having to
develop unnecessary information in order to determine Group
status. Sources should have the discretion to designate
individual vents as Group 1 and skip the determination process.
To reflect the EPA's intent in this regard, the commenter
recommended that a new sentence be added at the end of
§63.1430(e)(1) as follows:
"The owner or operator may elect Group 1 status for process
vents without making a Group 1/Group 2 determination. In such
event, no determination records are recruired."
Response: It is the EPA's intent to allow sources to
designate individual vents as Group 1 and skip the group
determination process. Since the commenter did not think that
§63.1430(e)(1) was clear enough in allowing for this option, the
Agency has added the modified version of the sentence that the
commenter suggested to the final rule, as follows:
2-109
-------
"The owner or operator may elect Group 1 status for process
vents without making a Group 1/Group 2 determination. In
such event, none of the records specified in this paragraph
(e) are required."
2.8.5 Recordkeepinq Exemption
Comment: One commenter (IV-D-05) stated that
§63.1430(e)(1)(vi) and (e)(2) do not require certain records if
batch vents are in compliance with §63.1425(c)(1) (aggregate 90
percent HAP reduction) and the relevant compliance device is
operating at all times. This means that, if a source ever fails
to get 90 percent, or if the device ever goes "down," there would
be a violation not only for failing to meet the 90 percent limit
but also for not having certain records that the source was
previously exempt from having to retain. Therefore, the
commenter recommended revising the text by adding a new sentence
at the end of each of §63.1430(e)(1)(vi) and (e)(2), as follows:
"(vi) ...This subparaqraph (vi) applies even if the
affected source temporarily fails to meet the recruirements
of §63.1425(c)(1) or the device does not function
temporarily."
"(2) This paragraph (2) applies even if the affected
source temporarily fails to meet the recruirements of
§63.1425(c)(1) or the device does not function temporarily."
Response: The EPA understands the commenter's concern about
being held liable for a double violation under the situation
described above. However, the EPA felt that the suggested
language was too vague (e.g., "temporarily" is not defined, and
no time frame is specified). The final rule resolves this
problem by replacing the phrase "[the] process vent is in
compliance with §63.1425(c) (1)" with the phrase "is subject to
§63.1425(c) (1)," in §63.1430(e) (1) (vi). This change will prevent
the owner or operator from being in a double penalty situation,
while still offering the group determination recordkeeping
exemption to the owner or operator who is subject to
§63.1425(c)(1) or (c)(3). A similar change has been made to the
language in §63.1430(e)(2), such that the revised language
appears as follows:
2-110
-------
(2) Process vents from continuous unit operationsVcntG
from Continuous Unit Operations. EachThe owner or operator
of an affected source that uses nonepoxide organic HAP to
make or modify a polycthor polyol the product in continuous
unit operations shall keep records regarding the
measurements and calculations performed to determine the TRE
index value of tho combined each process vent stream. The
owner Owners or operators- of combined atrcamp Group 1
continuous process vents that are in compliance withsublect
to the Group 1 requirements of §63.1425(c)(3) is arc not
required to keep these records.
2.8.6 Process Change Resulting in a Change in the TRE
Comment: One commenter (IV-D-05) requested that sources be
exempted from the reporting requirements of §63.1430(k) if there
is some other basis for exemption, such as a flow rate below
0.005 scmm or a HAP concentration less than 50 ppmv. Therefore,
the commenter recommended adding the following new paragraph at
§63.1430(k). Proposed §63.1430(k) would then become §63.1430(1).
§63.1430(k) would read:
(k) The owner or operator is not required to submit a
report of a process change if one of the conditions listed in
paragraphs (k)(1) through (k)(4) of this section is met.
(1) The process change does not meet the definition of a
process change in subpart G; or
(2) The vent stream flow rate is recalculated according to
subpart G and the recalculated value is less than 0.005 standard
cubic meter per minute, or
(3) The organic HAP concentration of the vent stream is
recalculated according to subpart G and the recalculated value is
less than 50 parts per million by volume, or
(4) The TRE index value is recalculated according to subpart
G and the recalculated value is greater than 4.0.
Response: The EPA does not believe that the "exemptions"
cited by the commenter, (i.e., flow rate below 0.005 scmm or a
HAP concentration less than 50 ppmv) are appropriate for
§63.1430(k) as proposed. Since proposal this section was
renumbered §63.1430 (j), where the owner or operator is subject to
the process vent reporting and recordkeeping requirements in
§63.1430 when a process change has occurred that causes a Group 2
continuous process vent with a TRE greater than 4.0 to become
Group 2 with a TRE less than 4.0. By definition (see §63.1423) a
Group 2 process vent from continuous unit operations is defined
2-111
-------
as "...not classified as a Group 1 continuous process vent," and
a Group 1 continuous process vent is defined as "(1) has a flow
rate greater than or equal to 0.005 standard cubic meters per
minute, and (2) has a total organic HAP concentration greater
than or equal to 50 parts per million by volume, and (3) has a
total resource effectiveness index value, calculated in
accordance with §63.1428(h)(1), less than or equal to 1.0." The
exemptions cited by the commenter define the process vent as
being Group 2 and are not needed in §63.1430 (j). Therefore,
§63.1430 (j) of the final rule does not include the revisions
suggested by the commenter.
2.9 EMISSION FACTOR
2.9.1 Emission Factor Plan
Comment: One commenter (IV-D-05) requested, for purposes of
clarity, that §63.1431(b)(1) be revised to refer to the use of a
combustion, recovery, or recapture device "without ECO," and that
§63.1431(b)(2) be revised to refer to the use of ECO "without a
combustion, recovery, or recapture device," since §63.1431(b)(3)
provides requirements for when both ECO and a combustion,
recovery, or recapture device are used.
Response: The EPA has made this distinction in the final
rule.
Comment: One commenter (IV-D-05) recommended, for purposes
of clarity, that the provisions of §63.1431(c)(2) be revised to
include the phrase "unless exempted by the provisions of this
subpart" as follows:
"The owner or operator shall conduct a performance test in
accordance with §63.1426 to determine the epoxide control
efficiency of the combustion, recovery, or recapture device
unless exempted by the provisions of this subpart...."
Response: Section 63.1426 lists the exemptions that apply,
so by citing §63.1426, the exemptions are also cited. Therefore,
2-112
-------
the EPA does not find it necessary to make the requested revision
to §63.1431(c)(2).
2.10 STORAGE VESSEL
2.10.1 Storage Vessel Provisions
Comment: One commenter (IV-D-05) referred to the language in
§63.1432(e), (f), and (h), which says that certain HON language
should be "replaced with" other language. For clarity, the
commenter requested that the text be changed to state that the
referenced language "applies," instead of saying that it
"replaces" language in the HON.
Response: The EPA appreciates the commenter's input, and
has changed the referenced language in accordance with the
commenter's suggestions, in the final rule.
2.10.2 Previous Performance Testing of Storage Vessel Control
Device
Comment: Two commenters (IV-D-04 and IV-D-05) supported the
provisions in §63.1432(g) that state that if a storage vessel has
a control device that has been performance-tested for other
reasons, the prior performance test would satisfy the storage
vessel requirements of this rule. However, Commenter IV-D-04
stated that the provisions are unclear in the following two
areas, and should be clarified:
(1) The text mentions control devices that are used to
comply with "§63.1425 through (sic.) §63.1433," which is
impossible. For one thing, the range of §63.1425 through
§63.1433 includes §63.1432, which is where the storage vessel
provisions are located. A control device on a storage vessel
cannot "also" be used for storage vessel compliance. Also, it is
unlikely that any single control device would be used
simultaneously for the entire range of sections from §63.1425
through §63.1433. The commenter believed that the EPA meant to
say that, if the control device had already been performance-
2-113
-------
tested under any one of those sections, no further demonstration
is required.
(2) The provisions accept only performance tests "required
by" specified sections of the rule. However, those sections do
not always "require" performance tests; sometimes they "allow"
performance tests, such as when a test that is conducted to
comply with other regulations promulgated by the EPA is allowed
to be used. If the control device has been performance-tested
under some NSPS or NESHAP using the same reference methods, and
if the results are still reliable, the commenter claimed that the
previously conducted performance test should be acceptable.
Response: The EPA agrees with the commenters on both of
these points, and has made changes to §63.1432(g), accordingly.
The regulatory language in the final rule reads as follows:
(g) EachThe owner or operator of an affected source
shall comply with this paragraph instead of
§63.120(d)(1)(ii) of oubpart G for the purposes of this
subpart. If the combustion, recovery, or recapture device
used to comply with §63.119 (e) is also used to comply with
any of the requirements found in §§63.1425 through 63.1431
and/or §63.1433, the performance test required in or
accepted by for §§63.1425 through 63.1431 and/or §63.1433 is
acceptable for demonstrating compliance with §63.119(e) of
aubpart G, for the purposes of this subpart. The owner or
operator will not be required to prepare a design evaluation
for the combustion, recovery, or recapture device as
described in §63.120(d)(1)(i) of Qubpart G, if the
performance test meets the criteria specified in paragraphs
(g) (1) and (g) (2) of this section.
2.11 WASTEWATER PROVISIONS
2.11.1 Wastewater Cross-Referencing with the HON
Comment: One commenter (IV-D-07) asserted that the cross-
references provided in §63.1433 of the proposed rule are very
complex, make compliance difficult, and might cause inadvertent
violations. The commenter urged the EPA to develop a
comprehensive set of requirements for this section of the rule.
2-114
-------
Response: In the final rule, the EPA has maintained the
cross-references to the HON wastewater provisions. The EPA
recognizes that a complex system of cross-referencing can be
confusing to the owner or operator. However, the EPA considers
the benefit to industry [which is derived from the fact that a
company producing products subject to several rules (e.g., the
HON, subpart JJJ, and subpart U) can rely on the fact that their
wastewater equipment will be subject to the same basic control
requirements] to outweigh the negative effects that this
structure might have on those same owners or operators. This is
particularly important in consideration of the fact that
wastewater may originate at process units producing several
different products, and yet be combined into an individual stream
prior to treatment. Each NESHAP should spell out the specific
requirements for such a wastewater stream, but the owner or
operator's job will be simplified if the basic control
requirements applying to wastewater streams from the different
types of process units are the same, or at least very similar.
In addition, the EPA has received positive feedback regarding
this strategy for dealing with process wastewater from other
industry representatives.
2.11.2 Definition of Wastewater in the Basis and Purpose for
Proposed Standards
Comment: One commenter (IV-D-07) requested that the EPA
clarify one statement that was made in section 2.5.3, Wastewater
Operation, of the Basis and Purpose for Proposed Standards
(Document no. EPA-453/R-97-010a, May 1997) . The commenter
claimed that one statement in that section omitted an important
part of the wording from the definition of wastewater that was
included in subparts F and G (the wastewater model for this
rule), that wastewater exists after it is discarded, not after it
exits a piece of equipment. Furthermore, the commenter
maintained that a wastewater stream is considered to be discarded
after it exits the last recovery device in a PMPU. The commenter
requested that the EPA clarify in its response to comments
2-115
-------
document that wastewater exists after it is discarded from the
PMPU.
Response: The commenter is correct, and the EPA would like
to make note of the fact that the definition of "wastewater" in
subpart PPP continues to mirror the definition of "wastewater"
in §63.111 of the HON, which contains the concept of the fluid
having been "discarded" from a process unit. The "discard"
concept is fundamental in determining which fluids exiting the
PMPU are subject to the wastewater provisions in §63.1433. In
addition, in the final rule the definition of "wastewater" has
been corrected to refer to Table 4 of subpart PPP (instead of
Table 5, which it incorrectly referenced at proposal).
2.11.3 Classification of Wastewater Receiving Facilities
Comment: Three commenters (IV-D-04, IV-D-05, IV-D-07)
requested that the EPA clarify that when facilities use the HON's
third-party wastewater treatment provisions, the receiving
facility does not become subject to the Off-Site Waste and
Recovery Operations MACT rule (subpart DD). One commenter (IV-D-
04) claimed that there is a disconnect between the proposed rule
and subpart DD, because subpart DD's exemption mentions only the
HON, not other rules that make slight changes to the HON (such as
different compliance dates, slight changes in definitions, and
different chemical lists). Therefore, the commenter recommended
adding a paragraph (d) to §63.1433 as follows:
(d) The owner or operator of a facility which receives
a Group 1 wastewater stream, or a residual removed from a
Group 1 wastewater stream, for treatment pursuant to
§63.132(g) of subpart G as referenced in paragraph (a) of
this section, is subject to the recruirements of §63.132 (a)
with the differences identified in this section, and is not
subject to subpart DD of this part with respect to that
material.
Response: The EPA agrees that this clarification needed to
be added to the final rule. The final rule contains language
very similar to that suggest by the commenters, in
2-116
-------
§63.1433(a)(20), with the exception that the cross references are
omitted. The new language reads as follows:
(20) The owner or operator of a facility which
receives a Group 1 wastewater stream, or a residual removed
from a Group 1 wastewater stream, for treatment pursuant to
§63.132(a) is subiect to the requirements of §63.132(a).
with the differences identified in this section, and is not
subnect to subpart DP of this part, with_respect to the
received material.
2.11.4 Clarifying the Definition of Residuals
Comment: One commenter (IV-D-04). requested that the EPA
clarify that residuals are not process wastewater. The commenter
referenced the proposal preamble (62 FR 46810, col. 3), which
stated that one example of a process wastewater stream is
"residuals recovered from waste management units." The commenter
considered this statement problematic because a residual is not a
process wastewater stream and not every residual is regulated.
The commenter asserted that the regulations should only apply to
residuals from Group 1 process wastewater streams; residuals from
Group 2 streams should not be regulated.
Response: The Agency agrees with this commenter. The
preamble to the final rule clarifies that residuals recovered
from waste management units do not necessarily constitute an
example of process wastewater.
2.11.5 Wastewater Control Options
Comment: One commenter (IV-D-04) requested that the EPA
clarify that, for an individual emission point, emission
suppression does not always require the combination of a cover,
an enclosure, and a closed-vent system to a control device. This
is the commenter's interpretation of the following statement in
the proposal preamble (62 FR 46811, col. 2): "Suppression of
emissions from the point of determination to the treatment device
will be achieved by using covers and enclosures and closed-vent
systems to collect organic HAP vapors from the wastewater and
convey them to treatment devices."
2-117
-------
Response: The commenter is correct, in that individual
emission points do not require a cover and an enclosure and a
closed-vent system to a control device. The final preamble
states that individual emission points require a control device,
or a seal/cover/enclosure that may be routed to a control device,
depending on the source.
2.11.6 "Inprocess" Aqueous Streams Should not be Identified as
Wastewater Streams
Comment; One commenter (IV-D-04) claimed that there is a
mismatch in §63.1433(a) and §63.1433(a)(10). Section 63.1433(a)
refers to "each process wastewater stream" and requires
compliance with HON §63.132 through §63.149. However, §63.149
does not deal with process wastewater streams, it deals with
certain in-process aqueous streams in open equipment. The
commenter asserted that nothing is wastewater until it leaves the
process. Similarly, §63.1433(a)(10) says certain definitions
apply whenever HON §63.132 through §63.149 refer to a Group 1
wastewater stream or to a Group 2 wastewater stream. However,
§63.149 applies to certain in-process (non-wastewater) aqueous
streams in open equipment. The commenter recommended that
§63.1433(a) and (a)(10) be revised as follows:
" (a) For each process wastewater stream originating at
an affected source, the owner or operator shall comply with
the requirements of §63.132 through §63.145-7. of subpart G,
with the differences noted in paragraphs (a)(1) through
(a) (19) , and (b) and (c) of this section, for the purposes
of this subpart. The owner or operator shall comply with
the requirements of §63.149 of subpart G, with the
differences noted in paragraphs (a)(1), (a)(2), (a)(7),
(a) (8) , (a) (11) and (a) (12) of this section, for equipment
meeting the criteria of §63.149 of subpart G as modified by
those differences."
"(10) Whenever §63.132 through §63.149 §63.147 of
subpart G refer to a Group 1 wastewater stream or a Group 2
wastewater stream, the definitions of the terms contained in
§63.1423 shall apply, for the purposes of this subpart."
Response: The EPA agrees with the general concept behind
commenter IV-D-04's comment. However, the EPA does not feel that
it is necessary to list each paragraph that corresponds to either
2-118
-------
wastewater streams or "in process" equipment, as the commenter
suggested. Instead, the EPA has added language clarifying that
the owner or operator must comply with the requirements in
§63.148 for leak inspection provisions, and with the requirements
of §63.149 for equipment that is subject to §63.149. The final
rule also clarifies that the owner or operator must comply with-
the requirements in §63.105(a) for maintenance wastewater. The
EPA has also made the distinction between the §63.132 through
§63.147 requirements and the §63.149 requirements, as the
commenter requested.
2.11.7 Selection of Compounds Sublect to the Wastewater
Provisions
Comment: One commenter (IV-D-07) supported regulating
wastewater for only the list of organic HAP in Table 4 of Subpart
PPP; however, the commenter requested that the EPA establish a
Henry's Law Constant "de minimis" value and delete non-volatile
organic HAP, such as methanol, from Table 4 as it applies to the
wastewater provisions. The commenter stated that EPA has deleted
other organic HAP from lists of regulated organic HAP under the
wastewater provisions in the HON and other rules, based on the
insignificant potential of those organic HAP to volatilize from
wastewater, and based on the knowledge that those organic HAP are
highly biologically degradable. The commenter noted that a "de
minimis" value of 1.8 x 1CT6 atm mVgmole fraction at 25 °C was
used in the HON. The commenter provided examples of chemicals
that have been deleted from the list of regulated organic HAP
under the HON or other rules, including diethanolamine, ethylene
glycol, and formaldehyde. The commenter stated that comments
submitted by the pharmaceutical industry on the proposed MACT
standard affecting their industry showed that methanol has a
higher function of biological degradation than that estimated by
the EPA. The commenter recommended that the EPA (1) reevaluate
the emissions estimates and significance of the Table 4 organic
HAPs to volatilize from wastewater, (2) determine those organic
HAP that should be excluded because of their high degree of
2-119
-------
biological degradation, and (3) delete the organic HAP meeting
those two criteria from the list of regulated organic HAP in
Table 4.
Response: The commenter incorrectly stated that the
compounds in table 4 of subpart PPP are subject to the wastewater
provisions. As stated in §63.1433(a)(2), when subpart G (the HON
wastewater provisions in subpart PPP) refers to table 9 or table
36, the owner or operator is only required to consider organic
HAP listed in table 9 or table 36 of subpart G that are also
listed in table 4 (actually, the proposed rule stated "table 5,"
which was a typographical error) of this subpart. In other
words, only compounds that are both on Table 4 of subpart PPP and
on Table 9 of subpart G are subject to the wastewater provisions.
By doing this, the EPA has eliminated (from the wastewater
provisions) those organic HAP that were eliminated from the HON
tables due to their low volatility. Therefore, the EPA did not
find it necessary to delete any compounds from table 4 of the
final rule.
2.11.8 Wastewater Nationwide Baseline Emissions Estimate:
Steam Jet System's Condensate Estimate
Comment: One commenter (IV-D-07) questioned the data and
wastewater emission estimates in the Supplementary Information
Document (SID) for the national baseline emissions, and stated,
in particular, that the EPA had overestimated the nationwide
emissions from wastewater from polyether polyols production. The
commenter cited two issues related to the data presented in Table
9 of the April 29, 1996 memorandum (Docket Item II-B-5),
regarding nationwide baseline emissions.
First, the commenter stated that the "average organic HAP
concentration" of 70,000 ppmv of PO is inconsistent with the
commenter's operating and engineering experience of steam jet
ejector systems, and noted that the volatile organic HAP
concentration in steam condensate tends to be in the range of 10
to 500 ppmv. The commenter noted that the memorandum in the
2-120
-------
docket (II-B-5) was unclear about whether the data gathered from
"vacuum systems" to determine the average concentration of organic
HAP were from a single stream, or from a combination of several
streams. The commenter recommended that the EPA gather more
accurate information on steam jet system's condensate, and base
the estimate on an adequate representation of the systems.
The commenter's second concern was whether the EPA had
relied on one data point to inappropriately extrapolate emissions
from the model stream to nationwide emissions. The commenter
also noted that the nationwide emissions from this one model
stream accounted for 68 percent of wastewater emissions, and 42
percent of the total nationwide emissions from polyether polyols
production.
Response: The commenter's first concern was that the PO
concentration in the vacuum system model stream (70,000 ppmv) was
inconsistent with the commenter's knowledge of the industry,
which would indicate that the PO concentration in these streams
is in the range of 10 tp 500 ppmv. It is the EPA's understanding
that the commenter's experience has been with vacuum systems with
low HAP concentrations, which were not included in the EPA's
database. Data provided by industry in responses to the
questionnaire were used to create the database. The
questionnaire stated that it applied to HAP emission sources
generated by polyether polyol production processes, except for:
(1) wastewater streams with an annual average flow rate less than
0.1 gallons per minute, or a total HAP content less than 10 ppmv,
or (2) waste with a total HAP content less than 10 ppmv or a
generation rate less than 220 pounds per month. Therefore, the
concentrations reported in response to the EPA/SPI questionnaire
were only for the more concentrated streams. The EPA took this
bias into account when estimating nationwide baseline emissions,
by multiplying the percentage of facilities with more
concentrated wastewater streams (58 percent) by the emissions
from the model stream and the estimated number of facilities in
the nation. The concentration, of 79,000 ppmv, cited in the
2-121
-------
memorandum (Docket Item II-B-5) was based on actual data provided
from 7 facilities' responses to the wastewater section of the
EPA/SPI questionnaire. Therefore, the EPA believes that the PO
concentration (70,000 ppmv) in the vacuum system wastewater model
streams is representative.
2.11.9 Wastewater Nationwide Baseline Emissions Estimate: The
Fraction Emitted Factors Used
Comment: One commenter (IV-D-07) voiced two concerns over
the fraction emitted (Fe) factors used to determine an emission
estimate. The commenter's first concern was that it was unclear
if an "average" Fe was used or if the Fe specific to each
chemical was applied to each specific organic HAP identified in
the wastewater stream. The commenter's second concern was that
the EPA did not update the fraction emitted factors (Fe) from
those used in the HON Table 34. Therefore, the commenter
requested that the EPA determine the emissions for each specific
organic HAP to which subpart PPP applies, before averaging the Fe
factors, or averaging the organic HAP concentrations, and that
the EPA update the Fe factor to reflect these new calculations.
Response: In response to the commenter's question, for each
model wastewater stream, an emission rate was calculated for the
seven facilities in the database, from which weighted average
emission rate, flow rate and concentrations were calculated. An
individual Fe (from Table 34 of the HON) was assigned to each
model wastewater stream, corresponding to the predominant organic
HAP identified for that model wastewater stream (i.e., toluene
with an Fe of 0.8, EO with an Fe of 0.5, and PO with an Fe 0.6) .
No revisions were made to the model wastewater streams, the
calculations, or the baseline nationwide wastewater emissions,
because these emission estimates used the Fe's from table 34 of
the HON, which has not been amended since promulgation of the HON
on April 22, 1994, and which the EPA considers to be consistent
with current organic HAP information.
2-122
-------
2.11.10 Inclusion of Biological Treatment as Wastewater Control
Option
Comment: One commenter (IV-D-07) noted that the EPA had
recently (at the time that the commenter's statement was made)
proposed changes for wastewater compliance provisions using
biological treatment under the HON. After the HON was
promulgated, the commenter requested that the EPA allow industry
representatives to comment on how the final changes to the HON
rule would affect subpart PPP.
Response: The EPA did propose clarifications to the HON on
August 22, 1997 (62 FR 44608). These proposed clarifications,
which are slated to go final in the near future, dealt with a
very narrow portion of the HON wastewater provisions. In
particular, the EPA proposed to revise the definition of "enhanced
biological treatment systems or enhanced biological treatment
processes," in order to clarify the meaning of the term, and it
proposed to revise appendix C of part 63 to reflect the
clarification of the definition of "enhanced biological treatment
systems or enhanced biological treatment processes." The EPA
involved industry representatives in the revision of the
definition of "enhanced biological treatment systems or enhanced
biological treatment processes," and the promulgated clarification
will incorporate public comments on the proposed clarification,
as necessary. For these reasons, the EPA does not believe that
it is necessary to request further comments on how those changes
to the HON wastewater provisions might affect owners and
operators of polyether polyols affected sources. The EPA would,
however, like to clarify (here and in the preamble to the final
rule) that the provisions of subpart PPP that cross-reference the
HON or any other regulation refer to the most recent, promulgated
versions of those rules, and that commenters are encouraged to
provide comments on any future proposed changes to those rules
cross-referenced in subpart PPP, with regard to how those
proposed changes might affect subpart PPP sources.
2-123
-------
2.11.11 PEPO Chemical List Versus HON's Table 8
Comment: Commenter IV-D-04 noted that §63.1433(b) should
have a "heading," to call attention to the fact that it contains
the requirements for maintenance wastewater, and noted that
§63.1433(b) mentions only one difference from the HON. That
difference was that owners and operators of polyether polyols
affected sources should use the proposed polyether polyols rule's
definition of "organic HAP," instead of the HON definition of
that term. However, according to the commenter, other
differences from the HON should be listed, such as the fact that
maintenance wastewater is a subset of "wastewater," and
"wastewater" has criteria that depend on chemical lists.
Therefore, the commenter requested that the EPA clarify that, for
the purposes of subpart PPP, the PEPO chemical list (Table 4)
applies, rather than HON Table 8 and/or Table 9 chemicals, and
also that PEPO focuses only on a subset of HON Table 9 chemicals,
while Table 8 in the HON does not apply to polyether polyols
affected sources.
Response: The EPA agrees that there is a need for further
clarification in §63.1433(b), through the use of a "heading," and
with regard to how the maintenance wastewater requirements in
§63.105 of subpart F apply to polyether polyols affected sources.
Therefore, in the final rule, §63.1433(b) is amended to read as
follows:
(b) Maintenance wastewater. The owner or operator of
each affected source shall comply with the requirements for
maintenance wastewater in §63.105 of subpart F, except that
when with the exceptions noted in paragraphs (b)(1), (2),
and (3) of this section.
(1) When the HON wastewater provisions in §63.105(a)
refers- to "organic HAPs, " the definition of "organic HAP" in
§63.1423 shall apply, for the purposes of this subpart.
(2) When the term "maintenance wastewater" is used in
§63.105, the definition of "maintenance wastewater" in
§63.1423 shall apply, for the purposes of this subpart.
(3) When the term "wastewater" is used in §63.105, the
definition of "wastewater" in §63.1423 shall apply, for the
purposes of this subpart.
2-124
-------
2.12 EQUIPMENT LEAK PROVISIONS
2.12.1 Method 21 for Equipment Leak Detection
Comment: One commenter (IV-D-04) maintained that the
statement in the proposal preamble (62 FR 46812, col. 2) that
"The equipment leak standards require the use of Method 21 of
Appendix A of part 60 to detect leaks" is not quite correct. The
commenter stated that the equipment leak standards require
compliance with subpart H, which requires a slightly modified
version of Method 21, and anyone using Method 21 in its
unmodified state might not comply with subpart H.
Response: The EPA agrees with the commenter. The proposal
preamble incorrectly stated that the HON requires the use of
Method 21, without mentioning that the detection instrument
response factor criteria in section 3.1.2(a) of Method 21 must be
for the average composition of the process fluid, instead of for
each individual VOC in the stream. The EPA apologizes for this
oversight in the proposal preamble, and wishes to clarify that
all of the equipment leak standards in subpart H apply to
polyether polyols affected sources, with the exceptions noted in
§63.1434.
2.12.2 "Delay of Repair" Example in the Preamble
Comment: One commenter (IV-D-04) questioned the fourth
example given in the proposal preamble (62 FR 46812, col. 1) for
situations where "delay of repair" may be allowed. The example
seems to say that single seals have better performance, and that
the EPA is replacing them with less desirable dual seals. The
commenter believed the opposite would be more accurate.
Response: The EPA agrees that the fourth example could be
misleading: the confusion results from the apparent omission of
a phrase from example number four. The fourth example of an
acceptable delay of repair beyond the required period should have
read "When equipment is being replaced by equipment with better
2-125
-------
leak performance, such as when a pump with single mechanical
seals is being replaced with a pump with dual mechanical seals. "
2.12.3 Cost-Effective Alternatives to the RON LDAR
Comment: One commenter (IV-D-08) urged the EPA to consider
more cost effective alternatives to the HON leak detection and
repair program (LDAR) in this and future MACT rules. The
commenter had previously submitted material to the EPA,
demonstrating that the initial leak rate assumptions in the HON
considerably overestimate actual leak rates in many instances.
Since cost effectiveness assumptions of the HON LDAR program are
based in part on initial leak rate estimates, the commenter is
concerned that the HON LDAR requirements do not represent the
most effective method of achieving reductions from fugitive
sources.
Response: The EPA appreciates the comment, but would like
to point out that the MACT floor level of control for subpart PPP
was determined to be the HON level of control. Because the cost
effectiveness of this level of control is not relevant in setting
the MACT floor level of control, the EPA did not consider all
methods of achieving reductions from fugitive sources. Further,
the EPA would like to point the commenter to §63.177(e) of part
G, which address obtaining approval from the Administrator to
utilize alternative means of emission limitations for equipment
leaks.
2.12.4 De minimis Ecruipment Count
Comment: One commenter (IV-D-09) has provided an analysis
to EPA that they believe shows that it is not cost-effective to
do equipment leak survey programs for less than 100 components,
unless a source has other facilities on site which can offset the
cost of purchasing a monitoring device.
Response: The EPA evaluated the cost effectiveness of
performing the HON level of LDAR to a facility with 100 equipment
2-126
-------
components, for facilities with and without monitors on-site.
The documentation is provided in docket item number IV-B-01. The
cost-effectiveness values ranged from $2,900 to $3,300 for
facilities that previously controlled to the CTG level of
control, and ranged from $3,000 to 3,500 for facilities that were
uncontrolled. The EPA concluded that controlling facilities with
100 equipment components is cost effective, and made no revisions
to the rule.
2.12.5 Applicability of Subpart I
Comment: One commenter (IV-D-05) maintained that §63.1434(c)
should discuss "resetting the clock" for sources in a quality
Improvement Program (QIP) under subpart I, as do other part 63
rules.
Response: The EPA agrees, and has incorporated language
that parallels that used in other part 63 rules, by adding the
following sentence to the end of §63.1434(c):
"However, sources subject to 40 CFR part 63, subpart I
that have elected to comply through a quality
improvement program, as specified in §63.175 or §63.176
or both of subpart H, may elect to continue these
programs without interruption as a means of complying
with this subpart. In other words, becoming subject to
this subpart does not restart or reset the "compliance
clock" as it relates to reduced burden earned through a
quality improvement program."
2.12.6 Inclusion of Phase-in Option
Comment: One commenter (IV-D-05) stated that §63.1434 does
not address the phase-in issue, and that failure to address this
issue created a compliance problem in the implementation of
subpart U, which resulted in a subsequent change to that rule.
To avoid a similar problem in this rule, the commenter suggested
adding language similar to newly proposed §63.502(m) from subpart
U as §63.1434(h), as follows:
(h) The owner or operator of each affected source shall
substitute the phrase "the provisions of subparts F, I, or PPP of
this part" for both the phrases "the provisions of subparts F or
I'of this part" and the phrase "the provisions of subpart F or I
2-127
-------
of this part" throughout §63.163 and §63.168, for the purposes of
this subpart. In addition, the owner or operator of each
affected source shall substitute the phrase "subparts F, I, and
PPP" for the phrase "subparts F and I" in §63.174(c)(2)(iii), for
the purposes of this subpart.
Response: The EPA agrees that the suggested text would
clarify how the requirements in subpart H of the HON apply to
owners and operators of polyether polyols production affected
sources. A slightly modified version of the suggested text has
been added as §63.1434(h), as the commenter requested.
2.12.7 Exemption for Heat Exchanger Units not Using HAP
Comment: One commenter (IV-D-05) requested that §63.1435(a)
be revised to exempt PMPUs that do not produce or use any organic
HAPs. This commenter also requested that §63.1435(b), in both
parts of the sentence where it occurs, be modified as follows:
"...the term 'polyether polyols manufacturing process unitj_
except those that do not manufacture or use any organic HAP, '
shall apply for purposes of this subpart."
Response: The EPA agrees with the intent of the commenter's
suggestion, but finds it simpler and more accurate to refer back
to the exemption in §63.1420(b), for PMPU without organic HAP.
Therefore, in the final rule §63.1435(b) has been revised so that
the last sentence reads:
"Further, when the phrase "a chemical manufacturing
process unit meeting the conditions of §63.100(b)(1)
through -f&4-(3) of this subpart, except for chemical
manufacturing process units meeting the condition
specified in §63.100(c) of this subpart" is used in
§63.104(a) of oubpart F, the term "polyether polyols
manufacturing procoao unit PMPU, except for PMPUs
meeting the conditions specified in §63.1420(b)" shall
apply for the purposes of this subpart."
2-128
-------
2.13 TESTING
2.13.1 Consistency Between the Test Method Required in the
Proposed Rule and the Test Method Used to Establish the
New Source Process Vent MACT Floor
Comment: One commenter (IV-D-04) stated that the EPA has
proposed a 99.9 percent emission reduction for new sources based
on a performance test for Facility M. However, the proposed rule
also requires performance tests using Method 18. According to
the commenter, the performance test for Facility M did not rely
entirely on Method 18. Method 18 was used at the inlet, but
Method 25A was used at the outlet. The proposed rule seems to
say that Method 25A may not be used, unless it is validated under
Method 301. Thus, the commenter interpreted that Facility M has
not conducted an acceptable performance test to be used as the
basis for the proposed standard. The commenter maintained that
if the performance test already conducted for Facility M was good
enough to be the basis for the rule, it should be good enough to
satisfy the performance testing requirements of the rule. The
commenter requested that the EPA clarify that Facility M is not
required to conduct another performance test.
Response: As discussed in the next response, the EPA has
revised the test methods allowed in the final rule. Based on the
review of the test report for Facility M's test, the EPA believes
that the test was conducted in accordance with the revised
testing procedures. However, paragraph §63.1426(b)(3) of the
final rule contains an exemption from performance testing for
process vents based on the use of previous tests. Whether
Facility M would be required to conduct another performance test
would ultimately depend on whether the conditions of
§63.1426(b)(3) are met.
2.13.2 Use of Method 25A Without Method 301 Validation
Comment: Three commenters (IV-D-04, IV-D-05, IV-D-08)
requested that the EPA clarify that owners or operators may use
Method 25A without the need for validation under Method 301.
2-129
-------
According to §63.1437(b), any analytical method, other than
Method 18, used for performance tests would have to be validated
according to the protocol in Method 301. Two of the commenters
(IV-D-04, IV-D-05) presented the following reasons why Method 301
validation should not be required if the owners or operators uses
Method 25A for the performance test.
1. In establishing MACT for new sources, the EPA relied on
a performance test for Facility M, which used Method 25A on the
outlet emissions. If the EPA found the data from Method 25A
sufficiently credible and valid to make those data the basis for
the rule, then the EPA cannot now say Method 25A is
insufficiently credible for use in performance tests.
2. Validation under Method 301 is incredibly and needlessly
burdensome. One commenter noted that they have been validating
analytical methods for years, without Method 301, quickly and
inexpensively. In contrast, validation under the Method 301
protocol is so burdensome and time-consuming as to be nearly
impossible.
3. The proposed rule relied on Method 25A in other
contexts. For example, §63.1433(a)(19) specified conditions
under which Method 25A may be used, as an alternative to Method
18, for wastewater. If Method 25A is appropriate for wastewater,
it should be appropriate for performance tests.
4. Commenter (IV-D-05) noted that Method 25A is allowed in
the Polymers and Resins IV rule without validation through Method
301. This commenter recommended revising §63.1426(c)(3) and
(c)(4) to allow the use of Method 25A without Method 301
validation.
One commenter (IV-D-08) requested that if the EPA keeps the
validation requirement, then the "abbreviated" version of Method
301 that was allowed in the HON wastewater provisions should also
be allowed under subpart PPP.
Response: The EPA agrees that allowing of the use of Method
25A would provide more flexibility, and potentially provide the
opportunity for less costly testing. However, the EPA believes
2-130
-------
that Method 25A should be used only after an accurate response
factor has been determined. The importance of calibrating a
flame ionization detector (FID) reading obtained using Method 25A
with respect to a certain compound (adjustment by response
factor) depends on how the Method will be used to demonstrate
compliance with the standard. In general, the EPA believes that
an accurate response factor is necessary in cases where Method
25A is used to demonstrate control efficiency across a device
where the composition of the stream may change, or in situations
where multiple components, including non-HAP VOC, are present.
Because the relative proportion of organic compounds may change
across the control device, appropriate response factors are
needed to accurately quantify TOC at the inlet and outlet of a
control device. In addition, the EPA believes that owners and
operators should have the opportunity to demonstrate compliance
at the outlet of a control device by measuring 20 ppmv TOC or
less. Therefore, the final rule does allow the use of Method 25A
under certain conditions. The following describes the choices of
test methods allowed in the final rule: (1) Method 18 (40 CFR
part 60, appendix A) to determine HAP concentration in any
control device efficiency determination; (2) Method 25 (40 CFR
part 60, appendix A) to determine total gaseous nonmethane
organic concentration for control efficiency determinations in
combustion devices; (3) Method 25A (40 CFR part 60, appendix A)
to determine the HAP or TOC concentration for control device
efficiency determinations under the conditions specified in
Method 25 (40 CFR part 60, appendix A) for direct measurement of
an effluent with a flame ionization detector, or in demonstrating
compliance with the 20 ppmv TOC outlet standard. As an
alternative, any other method or data that have been validated
according to the applicable procedures in Method 301 (40 CFR part
63, appendix A) may be used.
2.13.3 Allow any Testing to Demonstrate Compliance
Comment: One commenter (IV-D-06) referred to §63.1426(b)(3),
which allows for an exception to the performance test
2-131
-------
requirements if testing was previously conducted for determining
compliance with a regulation promulgated by EPA. The commenter
requested that this provision be amended to allow the use of any
representative testing conducted using methods specified in this
standard, and not just the use of compliance testing conducted
for compliance with promulgated EPA regulations. The commenter
pointed that the cost per facility for retesting is approximately
$100,000.
Response: The EPA has agreed to grant the commenter's
request. If the facility used the appropriate standardized EPA
method, the EPA agrees that it should not matter whether the
method was used to determine compliance with a. regulation already
promulgated by the EPA or for some other purpose. Therefore, the
provision has been revised to allow the use of any representative
testing conducted using methods specified in this standard. The
revision does include a 5-year limit on the age of the test
report, along with assurances that the process is still operating
under similar conditions as those that it was operating under
during the test. Further, the original test would need to have
monitored operating parameters that could be used to comply with
the parametric monitoring requirements in subpart PPP.
2.13.4 Notice for Rescheduling of a Test
Comment: One commenter (IV-D-05) recommended, for
consistency with other rules, that the text in §63.1437(a)(4) be
changed to add the following sentence: "If the owner or operator
reschedules the test for any reason, it must provide the
Administrator 7 days' notice."
Response: The EPA has added language to §63.1437(a)(4) in
the final rule, in order to specify that the owner or operator
needs to give the Administrator at least 7 days notice (prior to
the originally scheduled performance test) if a performance test
needs to be rescheduled. The changes to this paragraph also
allow the performance test to be rescheduled by mutual agreement
2-132
-------
between the Administrator and the owner or operator, if
necessary.
2.13.5 Engineering Calculations for Worst-case Recruirements
Comment: One commenter (IV-D-03) recommended that
engineering calculations be allowed to establish "worst case"
parameter monitoring requirements when conducted in association
with an EPA approved stack test.
Response: The use of engineering calculations, coupled with
an EPA-approved stack test, to establish parametric monitoring
requirements was already allowed in the proposed rule. The
proposed language in §63.1438(a) was very explicit about which
procedures (i.e., those contained in §63.1438(b), (c), or (d))
were permissible under varying circumstances. Specifically,
§63.1438(b) and (c) could be used by owners or operators to set
their parameter monitoring levels for a combustion, recovery, or
recapture device, if a performance test was required by subpart
PPP for that device. At proposal, it was not clear that
§63.1438(b) [and only §63.1438(b)] applied to owners or operators
desiring to set their parameter monitoring levels based
exclusively on parameter values determined during the performance
test. This has been clarified in §63.1438(b) in the final rule.
As §63.1438(c) read at proposal (and continues to read),
parameter monitoring levels established under this paragraph are
to be based on the parameter values measured during a performance
test, supplemented by engineering assessments and (or)
manufacturer's recommendations. In addition, §63.1438(b), (c),
or (d) may be used by owners or operators to set their parameter
monitoring levels for a combustion, recovery, or recapture
device, if a performance test is not required by this subpart for
that device. As §63.1438(d) read at proposal (and continues to
read), parameter monitoring levels may be established, under this
paragraph, based solely on engineering assessments and/or
manufacturer's recommendations.
2-133
-------
However, in reviewing these requirements, the EPA did notice
one discrepancy in the proposed version of §63.1438(a), which
needed correcting in the final rule. At proposal, §63.1438(a)
required that owners or operators using §63.1438(c) or (d) to
establish their parameter monitoring levels submit the
information specified in "§63.506(e)(3)(vii)" for review and
approval, as part of the Precompliance Report. This reference to
a reporting requirement in subpart U of part 63 was a mistake,
which has been remedied in the final rule. The correct reference
is to §63.1439(e)(4)(viii) in the final rule, which has been
added to that section and lists the following information to be
submitted in the Precompliance Report: (1) identification of
which procedure (i.e., §63.1438 (c) or (d)) is to be used; and (2)
a description of how the parameter monitoring level is to be
established, using those procedures.
The request to use §63.1438(c) or (d) for the establishment
of parameter monitoring levels is subject to review and approval
(or disapproval) by the Administrator; however, as the final rule
states in §63.1439(e)(4)(i), unless the Administrator objects to
a request submitted in the Precompliance Report within 45 days
after its receipt, the request shall be deemed approved. This
means that the amount of time that the owner or operator would
have to wait for a response to the request to use §63.1438(c) or
(d) is limited to a maximum of 45 days.
2.13.6 Daily Averages versus Compliance
Comment: One commenter (IV-D-04) noted that §63.1438(a)(1)
requires that the daily average value of monitored parameters be
kept within the established limit, which the commenter believes
is generally appropriate. However, the commenter gave some
examples of situations where the daily average value may be
outside the established limit without this being a noncompliance
concern. The commenter requested that §63.1438(a)(1) be amended
to require that industry keep the daily average value within the
established limit "except as otherwise provided in this subpart."
2-134
-------
In addition, the coimnenter interpreted §63 .1438 (a) (1) to
mean that the owner or operator is allowed to operate with a
daily average parameter value "at or above" the minimum, or "at
or below" the maximum. Therefore, it was requested that
paragraph (a)(1) be revised to reflect this interpretation as
follows:
(1) The owner or operator shall operate control and
recovery devices such that the daily average value of monitored
parameters remains, at or above the minimum established levelj_ or
remains at or below the maximum established level, except as
otherwise provided in this subpart.
Response: The EPA agrees with the commenter, and has added
the language requested by the commenter to §63.1438(a)(1) .
2.13.7 Method 1 or 1A
Comment: One commenter (IV-D-04) noted that §63 .1426(c) (1)
requires the use of Method 1 or 1A of 40 CFR part 60, Appendix A,
as appropriate, to select sampling sites. However, the commenter
claimed that, in many instances, neither method would be
appropriate. Method 1 is only for pipes or stacks with a
diameter of 12 inches or more, and it cannot be used when flow is
cyclonic or swirling or when there is a flow disturbance within
specified distances from the sampling site. Method 1A can be
used for smaller diameter pipes, but is only for particulate
matter sampling. Therefore, the commenter concluded that neither
method will be appropriate for determining sampling sites.
Response: First, the rule was restructured by separating the
paragraph addressing the use of Method 1 or 1A for sample or
velocity traverses from the paragraphs specifying the sampling
site location. In other words, if the owner or operator conducts
a sample or velocity traverse, the final rule requires that
Method 1 or 1A be used. However, it does not require that these
methods be used to select sampling sites. Second, the EPA has
decided to add text that states that references to particulate
matter in Method 1A do not apply for the purposes of subpart PPP.
2-135
-------
Because Method 1A can be used for smaller diameter pipes, it does
not have the problem that would exist if only Method 1 could be
used (since Method 1 is only for stacks or pipes that are greater
than 12 inches in diameter). By saying that "references to
particulate matter in Method 1A do not apply for the purposes of
this subpart," in §63.1426(c)(4)(i) of the final rule (test
method requirements), the EPA is making sure that owners and
operators can use Method 1A to select a sampling site.
2.13.8 Engineering Calculations as an Alternative Compliance
Demonstration to Performance Testing
Comment: Three commenters (IV-D-05, IV-D-07, IV-D-08)
stated that industry representatives are concerned with the
feasibility, accuracy, and safety of taking sample emissions from
process vents in batch unit operations. The commenters stated
that a performance test on these short duration, variable vents
is likely to be very inaccurate and potentially dangerous as
well. Therefore, Commenters IV-D-07 and IV-D-08 suggested that a
material balance based on common engineering calculations, which
the commenter felt would provide a more accurate, less costly,
and significantly safer means to verify compliance, should be
included in the final rule as a compliance demonstration option.
Commenter IV-D-05 said engineering calculations or other
alternatives, such as pilot plant data or manufacturer's
recommendations, should be permitted for compliance testing.
Commenter IV-D-05 reasoned that otherwise, demonstration of
emission reduction efficiency based on testing will be extremely
burdensome to the owners or operators of PMPUs that are designed
for multi-product operation and that employ a batch process with
very short venting times (the most typical processes, according
to the commenter). Commenter IV-D-05 recommended revisions in
§63.1426 through §63.1428, §63.1431, and §63.1438 to clarify that
these other alternatives are available to affected sources to
demonstrate compliance.
2-136
-------
Response: The EPA recognizes that there are issues related
to the feasibility, accuracy, and expense of testing process
vents from batch unit operations. The EPA would refer readers to
Section 7.3 of EPA's "Control of Volatile Organic Compound
Emissions from Batch Processes - Alternative Control Techniques
Information Document" EPA-453/R-94-020 for a detailed discussion
of these issues. However, the EPA does believe that accurate
emission tests can be conducted for these process vents.
One reason that the EPA has historically required
performance testing for control devices that reduce emissions
from process vents, when engineering analyses is allowed for
other emission sources (such as storage vessels), is that
emissions from process vents are typically significantly larger
than those from other emission sources. When emissions are
larger, the EPA believes that it is important that the
effectiveness of the control device be accurately determined by a
performance test.
Given that the magnitude of the emissions was a part of the
basis for requiring performance tests, the EPA believes that it
is reasonable to allow an alternative to performance testing for
a process vent control device if emissions being routed to the
device are comparable to the emissions that would be vented to
control devices for other emission sources for which performance
tests are not required. Therefore, the EPA decided that
engineering assessments could be allowed in lieu of performance
testing for "small" control devices that reduce HAP emissions from
process vents. For the Pharmaceutical Production NESHAP, the EPA
also determined that it was appropriate to allow engineering
calculations as an alternative to performance testing for small
control devices, where a small control device is defined as one
with uncontrolled annual HAP emissions of less than 10 tons per
year. The EPA believes that this level of uncontrolled emissions
is also appropriate to define a small control device for the
polyether polyols industry. Therefore, the final rule allows the
use of a design evaluation instead of a performance test if the
2-137
-------
control device receives less than 10 tons per year uncontrolled
emissions from one or more PMPUs.
The exemption from performance testing for small control
devices discussed above should help to alleviate some of the
concerns raised by the coiranenters. Many of the concerns related
to the feasibility, accuracy, and expense of testing these batch
vents are due to the short duration, variable nature of batch
venting episodes. The EPA believes that if a control device
receives more than 10 tons per year of uncontrolled HAP
emissions, it is likely that the vent streams being routed to the
device are of longer duration and less variable, thus making it
easier to conduct the performance test.
However, the EPA also recognizes that the small control
device exemption will not totally eliminate the concerns raised
by the commenters. Therefore, the EPA made other changes to the
testing requirements to address potential problems related to the
testing of batch process vents, which are briefly discussed
below.
Since batch emission episodes can be less than one hour, the
rule was changed to specify that test runs be conducted for the
complete duration of the batch venting episode or one hour,
whichever is less. Other references to one-hour periods were
also removed.
The changes discussed in the previous comment relating to
the use of Method 1 or 1A to select sampling sites were also
made.
With regard to the safety issue, the final rule states that,
in cases where it is imperative to limit any leakage of emissions
into the work atmosphere, a sampling port with a double seal
should be installed so that the probe can be inserted and removed
without any leakage of exhaust gas into the work atmosphere.
Further, the final rule requires that permanent sampling ports be
installed at the inlet to the control device during a period when
it is most convenient (or least disruptive) to shut the process
down (e.g., during a scheduled maintenance outage). In addition
to these specific requirements, a general requirement was added
2-138
-------
that allows owners or operators to eliminate potential testing
scenarios if the test could create a situation which could cause
plant or testing personnel to be subject to unsafe conditions.
In conclusion, the EPA acknowledges that issues exist with
regard to the testing of emissions from batch process units.
Changes have been made to the final rule to address these issues.
However, the Agency maintains that numerous other industries that
utilize batch processes are regulated by MACT standards, and are
able to conduct performance tests. The EPA believes that the
commenters did not provide sufficient rationale why the polyether
polyols industry presents unique testing problems that are not
present in these other industries that utilize batch processes.
Therefore, the final rule requires that control devices that
receive more than 10 tons per year of uncontrolled organic HAP
emissions conduct tests to demonstrate control device
performance.
Comment: Commenter IV-D-07 requested that, at a minimum,
the EPA should provide better technical guidance before making
flow measurement mandatory for these variable, and potentially
high organic content, vent streams. The commenter stated that a
performance test on these short duration, variable vents is
likely to be very inaccurate and discussed the safety concerns as
well.
Response: The EPA feels that the technical guidance for
measuring flow measurements are sufficient, and did not provide
additional guidance in the final rule.
2 .13 . 9 Recruest for Exemption from Testing Multiple Similar
Controls
Comment: One commenter (IV-D-03) maintained that although
many of the facilities in the EPA's polyether polyol database may
use a single control device for process vent emission control,
this is not necessarily true for all current or future
facilities. For example, the commenter's facility operates
2-139
-------
several separate control devices within a PMPU which have similar
designs and operating rates. They may also operate separate
PMPU's which have similar control system designs and operating
rates. They requested that engineering calculations be permitted
in lieu of testing where it can be demonstrated that the process
vents are similar sources, and the commenter suggested adding
this exception to §63.1426(b) in the final rule.
Response: The EPA does allow engineering calculations in
the case of a control technique that receives less than 10 tons
per year uncontrolled emissions (see Section 2.12.8). However,
for control techniques receiving more than 10 tons per year of
HAP emissions, the EPA requires performance testing, regardless
of whether there are separate, similar control devices on-site.
The EPA believes that the application of this cutoff, as well as
the allowances for direct measurement of condenser exhaust gas
temperature, have decreased the testing burden associated with
the rule and contends that such large control devices should be
tested.
2.13.10 Worst-Case Testing
Comment: Four commenters (IV-D-04, IV-D-05, IV-D-06 and IV-
D-08) expressed concern that the proposed rule requires that
performance tests for process vents be conducted during worst-
case operating conditions for the process. The commenters
requested that this requirement be deleted from the rule for the
following reasons:
1. Commenter IV-D-04 stated that there is no definition of
"worst-case" conditions for the process. The EPA has provided
criteria for determining worst-case emission episodes from batch
process vents, but no criteria for determining worst-case
operating conditions for the process. The commenter maintained
that owners or operators will have to guess what the phrase
means, and they will have inconsistent interpretations.
2. Two commenters (IV-D-04, IV-D-06) noted that many
companies have large, integrated manufacturing sites where a
2-140
-------
control device may be shared by more than one process. In those
cases, the validity of the performance tests depends not so much
on the operating conditions of "the process" (the PMPU), but on
the operating conditions for the control device.
3. Commenter IV-D-06 explained that batch reactor vents to
the control device are typically at low flow rates and of short
duration, making testing of such derived "worst case" episodes
difficult, if not impossible. Commenter IV-D-08 added that these
measurements may be technically unfeasible.
4. Commenter IV-D-04 pointed out that other regulations for
continuous processes have allowed performance tests during
representative operating conditions, and there is no obvious
technological difference that would require a different approach
to performance testing in this rule.
5. Commenter IV-D-04 stated that performance tests always
have a deadline. The commenter was concerned that, if the
industry must achieve "worst-case operating conditions" for a
specific process during that deadline, then they would have to
change the production rate for the PMPU. This would cause
problems, because the production rate would otherwise be dictated
by demand for the product of that PMPU. Commenter IV-D-06 also
noted that, in batch operations, staging such a scenario would
result in additional manpower cost and the manufacturing of
products for which a market demand may not exist.
6. Commenter IV-D-04 noted that, in most cases, the organic
HAP reduction efficiency of a control device is fairly stable
across a wide range of HAP concentrations. Since control devices
are designed to have some excess capacity, operating any single
process unit at its worst-case rate, rather than a representative
rate, would not be expected to make any significant difference in
the performance of the control device.
7. Two Commenters (IV-D-06 and IV-D-08) indicated that, due
to process design limitations, monitoring of these "worst-case"
scenarios could result in unsafe operating conditions.
For these reasons, the commenters encouraged EPA to revise
§63.1437(a)(1) to delete the clause "except that performance
2-141
-------
tests shall be conducted during worst case operating conditions
for the process."
Response: Worst-case testing requirements were not deleted
from the final rule, but were revised. The EPA's reason for
requiring compliance testing under worst case conditions is so
that the reduction efficiency of the control device is documented
under the most challenging conditions for that control device,
especially since commenters noted how difficult it is to
represent a typical venting episode. The phrase "worst-case" in
the proposed rule referred to the operating conditions of the
process (or PMPU). The worst-case testing requirement has been
revised to require testing during the worst-case conditions with
respect to the combustion, recovery, or recapture (i.e., control)
device.
Presumably, the control device should function as well or
better under conditions that are not as challenging. By revising
the rule to require testing during the worst-case conditions with
respect to the control device, continuous monitoring of operating
parameters established during the test provides a reasonable
measure of continuous compliance with the efficiency requirement
under all conditions.
The commenters asserted that there is no obvious
technological difference that would require a different approach
to performance testing in this rule as from other regulations
have allowed performance tests during representative operating
conditions. The EPA disagrees with the commenters' rationale.
The EPA believes that there are obvious technological differences
from the polyether polyols industry to industries previously
regulated (particularly SOCMI type industries) since polyether
polyols are produced on a batch basis. There is much more
variance in the process vent parameters (i.e., flow and
concentration) for process vent streams from batch unit
operations, compared to process vents from continuous unit
operations. In fact, this point was stressed by commenters. The
EPA believes that it is more appropriate to compare the
2-142
-------
requirements of this rule with other rules that also regulate
industries that operate on a batch basis. For this rule the EPA
not only compared the worst-case testing conditions with other
rules regulating batch processes, but adopted similar language to
that which is used in the Pharmaceutical Production NESHAP (40
CFR 63, subpart GGG).
The EPA would like to clarify a misconception related to
these worst-case testing provisions. It is not the intent that
production schedules be significantly altered, or that
impractical scenarios be created for testing that would never
occur in actual production. In other words, the EPA intends that
testing be conducted for the worst-case situation that can
reasonably be expected to occur during normal production. In
order to clarify this intent, the EPA has added language in
§63.1438, the general testing section of the rule. This new
language specifies that absolute worst case testing conditions
does not include situations that could cause damage to equipment,
situations that necessitate that the owner or operator make
product that does not meet an existing specification for sale to
a customer, or situations that necessitate that the owner or
operator make product in excess of demand.
The added language in §63.1438 also specifies the time
period in which the worst-case conditions are to be determined.
This time period is either the 6-month period that ends 2 months
before the Notification of Compliance Status is due, or the 6-
month period that begins 3 months before the performance test and
ends 3 months after the performance test. By limiting the worst-
case conditions to one of these 6-month periods, the rule
eliminates the need for an owner or operator to consider endless
possible production scenarios, and allows them to focus on those
production scenarios in the 6-month period selected by the owner
or operator.
In conclusion, the EPA believes that requiring that
performance tests for process vents from batch unit operations
during absolute worst-case conditions is necessary to ensure that
the emission limitations in the rule are achieved. The EPA also
2-143
-------
believes that, with the modifications to the rule made after
proposal, that the worst case provisions are reasonable and
workable for the polyether polyols industry.
Comment: One commenter (IV-D-04) maintained that if EPA
keeps the concept of "worst-case" scenarios, that the EPA should
clarify that a "simulated" scenario, as described in
§63.1426(c) (3) (i) (B) (4J , involves modeling or calculations,
rather than actual production. It was suggested that someone
might interpret "simulated" to mean that industry must produce an
artificial worst-case scenario by actually running all its
production units at top capacity simultaneously, which would not
be practical. The commenter requested that EPA clarify this
point by adding a parenthetical phrase, "(i.e., modeling or
calculations) " to §63 .1426 (c) (3) (i) (B) (.4) . It was also suggested
that EPA clarify that "worst-case" is limited to the maximum
production allowed in a State or Federal permit or regulation.
Response: The EPA agrees with the commenter that, at
proposal, the concept of "worst-case" scenarios was not clear.
The EPA has clarified the requirements in the final rule.
The EPA did not incorporate the specific language requested
by the commenter (i.e., modeling or calculations). However, the
EPA believes that the changes discussed in response to the
previous comment address the concerns raised by this commenter.
Finally, the commenter's suggestion that the EPA also
include language stating that "worst-case" is limited to what is
allowed under State or Federal rules or permits was included in
the final rule.
2.13.11 Determination of Emission Profile for Worst-Case
Testing
Comment: One commenter (IV-D-05) noted that
§63.1426(c) (3) (i) (C) allows "either process knowledge or test
data" to be used to determine the emission profile. The
commenter recommended that the section be changed to allow both
2-144
-------
process knowledge and test data to be used. This paragraph also
allows previous test results only if the results are still
representative of current conditions. The commenter also
recommended that previous test results be allowed if they can
readily be adjusted to account for changes in conditions, which
will avoid unneeded, costly additional tests. Therefore, for
clarity, the commenter recommended that the text be revised
accordingly.
Response: Due to the total re-working of the worst-case
testing provisions discussed earlier, the paragraph cited by the
commenter does not exist in the final rule. Therefore, no
changes were made by the EPA in response to this comment.
2.14 PARAMETRIC MONITORING
2.14.1 Operating Permit Recruirements
Comment: One commenter (IV-D-05) maintained that the
information in §63.1438(a)(2) is too detailed to be included in
an operating permit and recommended revising the text to delete
"or operating permit" as follows:
"As specified in §63.1439(e)(6), all established
levels, along with their supporting documentation and the
definition of an operating day, shall be submitted as part
of the Notification of Compliance Status. Once approved,
this information shall be incorporated into the affected
source's Notification of Compliance Status or operating
permit."
Response: The EPA has decided to remove the entire last
sentence of this paragraph because it is redundant and
unnecessary.
2.14.2 Average versus Maximum Value for Monitoring
Comment: One commenter (IV-D-05) maintained that the
mandatory parameter limit should not be set at the "average"
minimum (or maximum, where appropriate) value from the three test
runs. According to the commenter, using an average takes away
some legitimate leeway to use a broader parameter range for which
2-145
-------
compliance has been demonstrated. Therefore, the commenter
recommended revising the text in §63 .1438 (b) (2) as follows:
(2) Process vents from continuous unit operations. During
initial compliance testing, the appropriate parameter shall be
continuously monitored during the required 1-hour runs for
process vents from continuous unit operations. The maximum (or
minimum) monitoring levels(s) shall then be established aa the
average of based on the maximum (or minimum) point value from the
three one-hour test runs.—The average of the maximum values
shall bo used when establishing a maximum level,—and the average
of the minimum values shall be used when establishing a minimum
level.
Response: The EPA does not agree with this comment. The EPA
has reevaluated the parametric monitoring for this rule, as well
as other recent NESHAP (namely the Pharmaceutical NESHAP) and has
determined that the—operating parameter level must be established
as the average of the maximum (or minimum) point values obtained
during the three one-hour (continuously monitored) test runs.
However, if the owner or operator wishes to adjust the parametric
levels established during the test runs because the test results
indicated a higher control efficiency than is required by the
regulation, then the owner or operator has this option.
Specifically, provisions in §63.1438(c) and (d) allow the
parametric monitoring levels to be adjusted based on engineering
assessments.
2.14.3 Compliance Determination
Comment: Two commenters (IV-D-04 and IV-D-05) requested
that the EPA clarify the appropriate use of parameter monitoring
data for enforcement purposes. Commenter IV-D-04 noted that the
provisions of paragraph 63.1428(h), which require parameter
monitoring data to be used to demonstrate continuous compliance
with the emission limit, are inaccurate and unnecessary.
Commenter IV-D-05 had similar concerns regarding §63.1427(h).
The commenters stated that parameters such as temperatures or pH
readings are only surrogates that indicate proper operation of a
control device. They do not prove compliance or noncompliance
2-146
-------
with an emission standard because they do not measure emissions
or emission reductions.
Commenter IV-D-04 claimed that the EPA had a more valid
enforcement tool available in §63.1438(a)(1) which requires
industry to keep the daily average value of monitored parameters
within the approved limit. The commenter recommended that the
EPA revise §63.1427(h) to state that parameter monitoring data
will be used "to demonstrate continuous compliance with
§63.1438(a)(1)." The commenter concluded that EPA could then
assess exactly the same penalties, while maintaining a connection
with fact.
Commenter IV-D-05 stated that the rule should have a
requirement to keep the parameter data within a specified range
or limit, and excursions (appropriately defined) should, if not
excused, be violations of that operating requirement. Therefore,
the commenter recommended revising the text in §63.1427(h), as
follows:
(h) ECO Monitoring Requirements. The owner or
operator using ECO shall comply with the monitoring
requirements of this paragraph to demonstrate continuous
compliance with the cmiaaion limitation §63.1438(a)(1).
Paragraphs (h)(!) through (h)(3) address monitoring of the
ECO.
Similarly, Commenter IV-D-05 also recommended that the text
in §63.1438(e)(1) be revised to add the reference for determining
compliance, by inserting the phrase "with §63.1438(a)(1)" after
the word "compliance." The commenter also requested that
§63.1438(e)(2) be revised to add appropriate references as
follows:
"Except as provided in paragraphs, (e) (3) and (g) of this
section, for each excursion, as defined in paragraphs (e)(3) and
(f) of this section, the owner or operator shall be deemed out of
compliance with the provisions of thio oubpart §63.1438(a)(1)."
Response: First, the EPA agrees that the proposed language
in §63.1429(h) regarding compliance with the emission limitation
was not appropriate. However, the EPA does not believe that the
2-147
-------
specific reference to §63.1438(a)(1) is appropriate in either
§63.1427(h) (Note: the proposed §63.1427(h) is §63.1427(i) in the
final rule) or §63.1438(e)(1).
Paragraph §63.1438(a)(1) requires the owner or operator to
operate combustion, recovery, and recapture devices so that the
daily average value of monitored parameters remains at or above
the minimum established parameter level, or remains at or below
the maximum established monitoring level.
The EPA maintains that the requirement to maintain the daily
average monitored parameter within the established limit is only
one aspect of compliance with the monitoring provisions of
subpart PPP. In order to comply with the monitoring
requirements, the owner or operator must accomplish a number of
activities, from the installation of proper monitoring equipment
to the establishment of parameter monitoring levels, to the
proper operation of the combustion, recovery, or recapture device
and monitoring equipment.
For ECO, the owner or operator has similar requirements from
the establishment of parameter monitoring levels to ensuring that
each batch is accomplished in accordance with the established
levels. Therefore, the EPA revised what used to be §63.1427(h)
in the proposed rule and is §63.1427(i) in the final rule, to
state that owners or operators using ECO "shall comply with the
monitoring requirements of this paragraph to demonstrate
continuous compliance with this subpart ..."
While the EPA disagrees with the commenter that the
relationship between compliance and paragraph §63.1438(a)(1), the
EPA agrees with the argument that exceedances of operating
parameters should not be classified as violations of the emission
standard.
To assure that control devices used by the owner or operator
are properly operated and maintained so that continued compliance
with the applicable requirements is accomplished, the EPA has
adopted the approach in part 63 standards that monitoring be used
as a method for directly determining continuous compliance with
the applicable requirements. Further, the Agency is committed to
2-148
-------
following this approach whenever appropriate in future
rulemakings.
When determining appropriate monitoring options, the EPA
considers the availability and feasibility of the following
monitoring strategies in a "top-down" fashion: (1) Continuous
emission monitors (CEMs) for the actual HAP emitted, (2) GEMS for
HAP surrogates, (3) monitoring operating parameters, and (4) work
practice standards. In this standard, monitoring of control
device operating parameters is considered appropriate for all
emission sources. However, the EPA has allowed the option of the
continuous monitoring of organic compounds, which could mean
monitoring of the actual organic HAP or an organic surrogate.
The EPA believes that if organic compounds are monitored,
exceedance of the established value represents a violation of the
emission limitation. Similarly, because the exit gas temperature
of a condenser is so closely correlated with emissions, the EPA
believes that an exceedance of the established condenser
temperature should also represent a violation of the emission
limit. The EPA agrees with the commenters that exceedance of
other monitoring parameters is not necessarily an exceedance of
an emission limit.
Paragraph §63.1438(e) of the final rule has been rewritten
to add specificity regarding what the owner or operator is out of
compliance with when an excursion occurs (that is not an excused
excursion). If an organic monitoring device is used to monitor
HAP or TOC concentration at the outlet of a recovery or recapture
device, the final rule clarifies that each excursion where the
daily average value of monitored parameters is above the maximum,
or below the minimum established parameter level, represents a
violation of the emission limit. Similarly, an excursion where
the daily average temperature is above the maximum established
temperature for a condenser represents a violation of the
emission limit. Other excursions where average values are above
the maximum, or below the minimum established parameters
represent violations of the operating limit, rather than
violations of the emission limit. Also, excursions due to
2-149
-------
insufficient monitoring data are violations of the operating
limit.
2.14.4 Excursion Provisions for Storage Vessels Exempt from
Continuous Monitoring, Process Vents from Batch Unit
Operations, and Extended Cookout
Comment: One commenter (IV-D-04) suggested that EPA add
provisions in §63.1438 for storage vessels that are not required
to conduct continuous monitoring. The commenter noted that not
all storage vessels are required to conduct continuous
monitoring. The PEPO standard requires a monitoring plan, which
must specify what will be monitored and how often. For example,
some storage vessels may be monitored only while they are being
filled, which could be for 2 hours. Consequently, the concept of
a "daily average" parameter value will not apply to those storage
vessels. Also, in some cases the monitored value may not be for
a "parameter," in the strict sense of the word. The commenter
noted that because of the significant differences in emission
patterns and controls among different storage vessels, MACT
standards do not specify a "one size fits all" approach to
monitoring. Therefore, there cannot be a "one size fits all"
definition of "excursion," even though §63.1438(f) attempts to do
just that. Section 63.1438(f) combines all storage vessels with
process vents and says the daily average is the measure of
compliance for all of them, which will not work. The commenter
recommended that the EPA revise §63.1438(f) so that it refers
only to storage vessels required to conduct continuous
monitoring. Also, it was recommended that the EPA add a new
paragraph "(h)" with appropriate excursion definitions for
storage vessels that are not required to conduct continuous
monitoring.
Response: The EPA agrees with the commenter that separate
monitoring requirements should be established for storage vessels
that are required to be continuously monitored and for storage
vessels that are not required to be continuously monitored.
2-150
-------
However, instead of adding a new paragraph as §63.1438(h), the
EPA has renumbered §63.1438(f) so that the proposed language is
now (f)(!) and the new paragraph is §63.1438(f)(2). A
parenthetical has been added in §63.1438(f)(1) to clarify that
the provisions apply to storage vessels where the applicable
monitoring plan specifies continuous monitoring.
In addition, consideration of this comment caused the EPA to
realize that the proposed excursion definitions related to
insufficient monitoring data in §63.1438(f)(2), (f)(3), and
(f)(4) were not always suitable for process vents from batch unit
operations. For these batch process vents, venting episodes may
be less than one hour, which makes the "valid hour of data"
concept unworkable. Clearly the EPA did not intend that an
excursion occur when the entire emission episode is controlled
and monitored in accordance with the rule, but the episode is
less than one hour. Therefore, paragraph §63.1438(f) (3) was
added to address excursions for process vents from batch unit
operations.
Also, excursions were defined in the proposed §63.1427(h)(3)
for owners or operators using ECO to comply with the epoxide
emission limitations in §63.1425(b). The EPA determined that a
reference to those excursion definitions was needed in the
parameter monitoring levels and excursions section (§63.1438).
Therefore, paragraph §63.1438(f)(4) was added referring to the
excursion definitions for ECO in §63.1427(h)(3).
2.14.5 Continuous Monitoring
Comment: One commenter (IV-D-04) strongly supported
§63.1438(e) and (f) of the proposed rule. However, proposed
§63.1438(e) stated that its subparagraphs only applied to
emission points and control or recovery devices for which
continuous monitoring was required, and the commenter requested
that EPA revise §63.1438(e)(3) so that it would apply regardless
of whether continuous monitoring was required. Paragraph
§63.1438(e)(3) lists the situations in which no excursion is
considered to have occurred, even though parameters strayed
2-151
-------
outside of their limits, or data were not collected. According
to the commenter, these situations are universal; they apply to
any emission point or control or recovery device, regardless of
whether continuous monitoring is required. The commenter stated
that the equivalent paragraph in the HON (§63.152(c) (2) (ii) (E) )
was specifically revised for the same reasons as those cited
above. The commenter recommended the following revision to
§63.1438(e):
" (e) Compliance determinations. The provisions of this
paragraph, except (e)(3) of this paragraph, apply only to
emission points and control or recovery devices for which
continuous monitoring is required under this subpart."
Response: The EPA is in general agreement with the concepts
raised by the commenter. However, the EPA decided to more
significantly alter the structure of §63.1438(e) and (f), as
described below. In the final rule, §63.1438(e) describes
"violations" to the rule. As discussed in response to an earlier
comment (2.14.3), §63.1438(e) has been revised to address the
relationship between excursions and violations.
The EPA has made changes to §63.1438(f) to more clearly
provide all of the necessary information about the definition of
excursions. First, as discussed above in response to comment in
section 2.14.4, excursion definitions were added for storage
vessels where the applicable monitoring plan does not specify
continuous monitoring, for batch process vents, and for ECO.
Basically, there are two ways an excursion can occur. The
first is if the average parameter value measured is above a
maximum, or below a minimum, established value. The second is if
insufficient monitoring data are collected. The final rule makes
clarifications of the data to be used in both of these
circumstances.
With regard to calculating averages, §63.1439(d)(7) of the
final rule specifies that monitoring data collected during
periods of monitoring system breakdowns, repairs, calibration
checks, and zero (low-level) and high-level adjustments; start-
2-152
-------
ups; shutdowns; malfunctions; and periods of non-operation of the
affected source that result in the cessation of emissions to
which the monitoring applies are not to be included when
calculating any average.
Language has also been added to §63.1438(f) to clarify when
monitoring data are insufficient. An excursion due to
insufficient monitoring data occurs if measured values are
unavailable for a specified percentage of time the control device
is in operation. First, the rule now clarifies the situations
that cause measured values to be unavailable: monitoring system
breakdowns, repairs, calibration checks, or zero (low-level) and
high-level adjustments. Second, the final rule clarifies that
periods of start-ups; shutdowns; malfunctions; and periods of
non-operation of the affected source that result in the cessation
of emissions to which the monitoring applies are not to be
included in defining the period of control device operation.
The EPA believes that the clarifications discussed above address
the commenter's concern over the provisions of the proposed
§63.1438(e)(3) applying to all situations, whether or not
continuous monitoring is required.
2.14.6 Clarification to Text
Comment: One commenter (IV-D-05) recommended, for clarity,
that the text in §63.1438(c) be revised as follows:
"Establishment of parameter monitoring' levels based on
performance tests, supplemented by engineering assessments,
and/or manufacturer's recommendations. Parameter monitoring
levels established under this paragraph shall be based on the
parameter values measured during the performance tests
supplemented by engineering assessments and/or manufacturer's
recommendations...."
Response: The EPA agrees with the commenter, and has made
the suggested change in the final rule.
2-153
-------
2.15 GENERAL RECORDKEEPING AND REPORTING
2.15.1 Elimination of Initial Notification
Comment: One coitimenter (IV-D-05) stated that the Initial
Notification in §63.1439(e)(3) should not be required, in order
to reduce the regulatory burden, and to be consistent with the
Polymers and Resins MACT. The commenter recommended that
§63.1439(e)(3) be deleted, the subsequent sections renumbered
accordingly, and that the reporting cross-reference in Table 5 be
modified. In addition, the commenter recommended that
§63.1432(n) and §63.1434(d) be changed accordingly.
Response: The Agency's enforcement personnel and the State
representatives involved in this regulatory process consider the
Initial Notification requirement in §63.1439(e)(3) a necessary
tool for enforcement and compliance purposes. Moreover,
completion of the Initial Notification should not take more than
a few hours, since the information requested is very basic; i.e.,
the name and address of the owner or operator; the address
(physical location) of the affected source; an identification of
the emission points and affected source; and an identification of
whether the affected source can achieve compliance by the
relevant compliance date. Therefore, the Agency has not made any
of the requested changes.
2.15.2 Provide Examples of Emission Points
Comment: One commenter (IV-D-05) maintained that the
addition of some examples in §63.1439(e)(3)(i)(C) of the "kinds"
of emission points to be identified in the Initial Notification
would be helpful.
Response: The emission points subject to this rule and
required to be identified in the Initial Notification include the
emission points and equipment specified in the definition of
affected source under §63.1420(a). Section 63.1420(c) of the
rule describes emission points not subject to the provisions of
this rule. Since other provisions of the rule describe the
2-154
-------
emission points to be regulated, the Agency does not believe it
is necessary to provide examples of emission points to be
identified in the Initial Notification in §63.1439(e)(3)(i)(C).
2.15.3 Periodic Report
Comment: One commenter (IV-D-05) noted that §63.1434(f)
should say the Periodic Reports under subpart H "may" (rather
than "shall") be submitted with the Periodic Reports under this
rule.
Response: The Agency agrees that the owner or operator has
the option to submit the Periodic Reports for equipment leaks as
specified in subpart H at the same time as the Periodic Report
for this subpart or at another time. The final rule has been
changed accordingly.
2.15.4 Include all Records in One Section
Comment: One commenter (IV-D-05) recommended that in order
to make the recordkeeping and reporting section (§63.1439) and
Table 5 in the proposed rule as useful as possible, the EPA
should include all relevant requirements in §63.1439. For
example, the ECO recordkeeping requirements are in §63.1427(1)
and the reporting requirements are in §63.1427(j), but neither
citation appears in Table 5 or in §63.1439. The commenter
maintained that with a rule as complex as this, it is especially
important that all of the monitoring, recordkeeping, and
reporting requirements be included in a single section to ensure
that inadvertent non-compliance through failure to prepare a
report or maintain a record that is not included in the
"Recordkeeping" section or table does not occur.
Another commenter (IV-D-04) was also concerned that the
recordkeeping and reporting requirements were scattered
throughout the proposed rule, plus various recordkeeping and
reporting requirements from the General Provisions. The
commenter believed that this approach is very likely to cause
inadvertent noncompliance. The commenter recommended that all
2-155
-------
recordkeeping and reporting requirements be in one place in the
rule. Alternatively, if that cannot be done, the commenter
requested that EPA include every recordkeeping and reporting
requirement in a single table, with the specific citation to
where that requirement is found.
Response: The Agency believes it is more logical to include
the specific recordkeeping and reporting requirements related to
each emission source type in the applicable sections of the rule
because not all facilities have every source type. Therefore,
the suggestion to include all recordkeeping and reporting
requirements in one place has not been adopted. The Agency,
however, has added a table in the final rule for all routine
recordkeeping and reporting requirements, including the specific
citation in the rule for the requirement, and the due date for
the specific report as recommended (as Table 7 of the final
rule). The Agency believes that this table will be useful to
owners and operators of affected sources in complying with the
various reporting requirements of the rule.
2.15.5 Parametric Monitoring During Periods of Start-up,
Shutdown or Malfunction
Comment; Three commenters (IV-D-05, IV-D-07 and IV-D-09)
provided revised language for §63.1438(e)(3) regarding monitoring
during periods of start-up, shutdown, and malfunction (SSM) .
One commenter (IV-D-05) maintained that the provisions of
§63.1438(e)(3) do not quite follow the revised HON, and that the
differences cause problems. For example, the proposed rule text
literally says there is no excursion "if the daily average value
of a monitored parameter is above the maximum level or below the
minimum level established." In other words, there can never be
an excursion. This comes about from adding a comma after the
word "established" (which was not in the HON). Also, this
paragraph (unlike the HON) does not discuss monitoring data
collected during start-ups and shutdowns. To be consistent with
2-156
-------
the HON, Commenter IV-D-05 recommended deleting the proposed text
in §63.1438(e)(3) and replacing it with the following:
(3) If a monitored parameter is outside its
established range or monitoring data are not collected
during periods of start-up, shutdown, or malfunction (and
the source is operated during such periods in accordance
with the source's start-up, shutdown, and malfunction plan
as required by §63.6(e)(3) of subpart A), or during periods
of non-operation of the PMPU portion thereof (resulting in
cessation of the emissions to which the monitoring applies),
then the excursion is not a violation and, in cases where
continuous monitoring is recruired, the excursion does not
count toward the number of excused excursions for
determining compliance.
Another commenter (Commenter IV-D-07) maintained that
monitoring records from periods of SSM would not be expected to
provide additional information as to whether plans are followed
in many cases, nor would they provide the Agency with information
regarding the adequacy of the plans. Commenter IV-D-07 provided
a few examples to support this claim. Commenter IV-D-07 stated
that the proposed rule simply requires that the source collect
data for periods when it is not required to comply with the
standard resulting in a potential noncompliance status (for not
collecting data), with essentially no benefit to the environment.
The commenter recommended that this position be dropped in the
final rule and that §63.1438(e)(3) of the final rule be revised
to conform with the HON as follows:
(3) If the daily average value of a monitored
parameter is above the maximum level or below the minimum
level established, or if monitoring data cannot be collected
during monitoring device calibration check or monitoring
device malfunction, or if monitoring data are not collected
during periods of start-up, shutdown, or malfunction, or if
monitoring data are not collected during periods of non-
operation of the affected source or portion thereof
(resulting in cessation of the emissions to which the
monitoring applies), but the affected source is operated
during the period of start-up, shutdown, or malfunction in
accordance with the affected source's Start-up, Shutdown,
and Malfunction Plan, then the event shall not be considered
a monitoring parameter excursion.
Response: As discussed in section 2.14, the EPA made
significant revisions to paragraphs §63.1438(e) and (f). The
2-157
-------
paragraph cited by the commenters no longer exists in the final
rule. However, the EPA believes that most of the concerns raised
by the commenters are addressed in the final rule.
First, §63.1439(d)(7) of the final rule specifies that
monitoring data collected during periods of monitoring system
breakdowns, repairs, calibration checks, and zero (low-level) and
high-level adjustments; start-ups; shutdowns; malfunctions; and
periods of non-operation of the affected source that result in
the cessation of emissions to which the monitoring applies are
not to be included when calculating any average. This paragraph,
which was in the proposed rule, clearly states that monitoring
data collected during start-ups, shutdowns, malfunctions, are not
to be included in an average. Therefore, the EPA believes that
it is unnecessary to additionally state that the exceedance of an
average value due to data collected during a start-up, shutdown,
or malfunction is not an excursion, when the data collected
should not be used to calculate an average.
Also, the EPA has added paragraphs §63.1438(f)(1)(v)(A)
through (D) and §63.1438(f)(2)(i)(B), which describe the periods
that are not to be included when determining the period of
combustion, recovery, or recapture device operation for the
purpose of determining whether an excursion has occurred due to
insufficient monitoring data. Under these paragraphs, the
periods that should be left out when determining the period of
combustion, recovery, or recapture device operation include
start-ups; shutdowns; malfunctions; and periods of non-operation
of the affected source that result in the cessation of emissions
to which the monitoring applies.
The EPA does not agree with Commenter IV-D-07's opinion that
monitoring during start-ups, shutdowns, and malfunctions results
in "essentially no benefit to the environment." It is the EPA's
position that requiring monitoring during these periods will
provide the EPA with more information concerning whether or not
Start-up, Shutdown, and Malfunction Plans were followed, and will
provide the EPA with valuable information for assessing the
adequacy of a source's Start-up, Shutdown, and Malfunction Plan
2-158
-------
for future situations. Therefore, the final rule continues to
require that monitoring data be collected during periods of SSM.
Comment: One commenter (IV-D-04) maintained that EPA should
not impose a blanket requirement to monitor during start-ups,
shutdowns and malfunctions. There should be exceptions to this
requirement. For example, the industry cannot keep monitoring if
the monitoring device itself has the malfunction. Similarly, it
may sometimes be necessary to "valve off" a monitoring device
(isolate the device from the monitored stream) in order to keep
the device from being damaged. The commenter requested adding a
new paragraph, §63.1438(e)(4), to address these instances:
(4) Failure to collect monitoring data shall not be
considered an excursion during periods of monitoring system
malfunction, or when the monitoring system must be isolated
or otherwise rendered nonoperational in order to prevent
damage to the monitoring system.
Also, the commenter recommended that EPA clarify that
parameter data gathered during start-ups, shutdowns and
malfunctions are excluded from daily averages for the purpose of
determining excursions and referred to another commenter's (IV-D-
05) suggested revisions to §63.1438(e)(3) to address this.
In addition, the commenter noted that although the rule,
§63.1438(e)(3), is clear that once a shutdown is complete and
emissions have ceased, monitoring is not required during the
ensuing period of non-operation, the preamble was not clear and
asked that EPA reaffirm this point.
Further, Commenter (IV-D-09) stated that during the General
Provisions litigation, they discussed reasons for needing the
provisions for start-up, shutdown, and malfunctions and why
modifications were needed. They stated that EPA has agreed to
revised language; e.g., depending on flow, concentration, etc., a
control device may need to be diverted during start-up, shutdown,
or malfunction to prevent explosions, etc. and requested that
this language be incorporated into the Polyether Polyols MACT.
2-159
-------
Response: The EPA is in general agreement with commenters
IV-D-04 and IV-D-09 on these points. As discussed in the
previous response, the changes to §63.1438(f) and §63.1439(d)(7)
clarify that parameter monitoring data gathered during start-ups,
shutdowns, malfunctions, and periods of non-operation of the
affected source resulting in cessation of the emissions to which
the monitoring applies, are to be excluded from daily averages,
and, in fact, all averages computed under subpart PPP or the
subparts that it references.
In addition, changes to §63.1439(b)(1) allow owners and
operators to "cease" collecting monitoring data from a particular
monitor (e.g., by shutting off the monitor, or diverting flow
away from it) during a start-up, shutdown, or malfunction if the
owner or operator can show that the monitor would be damaged or
destroyed as a result of the start-up, shutdown, or malfunction.
This provision should satisfy the concerns expressed by
commenters IV-D-04 and IV-D-09. Such a provision must be
included in the Start-up, Shutdown, and Malfunction Plan.
Getting such a provision in the Start-up, Shutdown, and
Malfunction Plan requires is accomplished by submitting a
request, and rationale defending the request, in the
Precompliance Report or in a supplement to the Precompliance
Report, as described in the new language in §63.1439(e)(4). If
the request is not denied by the Administrator within 45 days
after receiving the request, it can then be incorporated into the
Start-up, Shutdown, Malfunction Plan.
These changes are meant to strike a balance between the
EPA's concern that monitoring data are collected at all relevant
times and industry's concern that valuable monitoring equipment
could be damaged during a start-up, shutdown, or malfunction.
The changes are intended to provide protection for monitoring
equipment during those periods, while providing the EPA with
assurance that monitoring equipment is not being "shut off"
indiscriminately. '
2-160
-------
Comment: One cornmenter (IV-D-05) maintained that
§63.1439(b)(1)(i)(C), which pertains to continuous monitoring
systems records of calibration checks, was not part of the Start-
up, Shutdown, and Malfunction plan and should be deleted.
Response: The Agency has not deleted §63.1439(b)(1)(i)(C),
as requested by the commenter; however, the regulatory text in
that paragraph has been moved to §63.1439 (d) (8), because the
Agency agrees that it is more of a global recordkeeping
requirement than a record specifically associated with the Start-
up, Shutdown, Malfunction Plan.
2.15.6 Exclusion of Monitoring Data from Daily Averaging
Comment: Three commenters (IV-D-02, IV-D-04, IV-D-05)
claimed that §63.1439(d) should include other situations where
monitoring data should not be included in the daily average
recorded. For example, data collected during start-ups,
shutdowns, malfunctions, and periods of non-operation (of the
affected source or a portion thereof), resulting in cessation of
the emissions to which the monitoring applies. Commenter IV-D-04
also mentioned "data collected during calibration checks" as
another example. According to the commenters, these concepts are
captured in the HON, which has a revised paragraph structure for
greater clarity. Commenter IV-D-05 recommended, for consistency
with the HON, that proposed §63.1439(d)(7) be replaced with the
revised wording:
"Monitoring data recorded during periods of monitoring
system breakdowns,—repairs,—calibration checks,—and zero
(low'level)—and high level adjustments identified in
paragraphs (d)(7)(i) through (d)(7)(v) of this section shall
not be included in any average computed under this subpart.
Records shall be kept of the times and durations of all such
periods and any other periods during process or combustion,
recovery,—or recapture control device operation when
monitors are not operating.
(i) Monitoring system breakdowns, repairs, calibration
checks, and zero (low-level) and high-level adiustments;
(ii) Start-ups
(i i i) Shutdowns;
(iv) Malfunctions;
2-161
-------
(v) Periods of non-operation of the PMPU (or portion
thereof), resulting in cessation of the emissions to which
the monitoring applies.
Commenter IV-D-04 supported the revised wording.
Response: For consistency with decisions made on other
rules [e.g., the HON, §63 .152(c) (2) (ii) (C)], the Agency has
revised §63.1439(d)(7) as suggested by the commenters to clarify
that data recorded during periods of start-up, shutdown,
malfunction, etc. should not be included in averages of monitored
data, including daily averages.
2.15.7 Retention of Superseded Start-up, Shutdown, and
Malfunction Plan
Comment: One commenter (IV-D-05) stated that §63.1439(b)(1)
does not say how long a superseded start-up, shutdown, and
malfunction plan must be retained. To be consistent with the HON
provisions, the commenter recommended revising the text in
paragraph (b)(1) to add the following sentence after the fourth
sentence:
"... In addition, if the start-up, shutdown, and malfunction
plan is revised, the owner or operator shall keep previous
(i.e., superseded) versions of the start-up, shutdown, and
malfunction plan for a period of 5 years after each revision
to the plan...."
Response: For purposes of clarification and consistency,
the Agency has added the commenter's suggested language into
§63.1439(b)(1) of the final rule.
2.15.8 Exclusion to Continuous Recordkeepinq Requirements
Comment: Two commenters (IV-D-04 and IV-D-05) stated that
§63.1439(d) says that anyone subject to §63.1438 is required to
keep continuous records, which may not be true. For example, the
owner or operator of a storage vessel (closed-vented to a control
device) may be subject to §63.1438 (required to establish
parameter levels for the control device) but not be required to
keep continuous records. According to one commenter (IV-D-04),
2-162
-------
the monitoring plan will specify the type and frequency of
required monitoring. Therefore, the commenters suggested
revising the text in §63.1439(d) to delete the words "required to
comply with §63.1438 and, therefore," in paragraph (d).
Response: The EPA agrees that continuous record keeping is
not required for all emission points, and the EPA has revised
§63.1439(d) accordingly.
2.15.9 Delete Redundancy in Recordkeepino; Reoruirements
Comment: One commenter (IV-D-05) maintained that the source
should not be required to retain records if the daily average
value is within the limits and, therefore, the commenter
recommended deleting paragraph §63.1439(d)(5). Another commenter
(IV-D-04) requested that EPA delete paragraphs §63.1439(d)(4) and
(d)(5), which describe records to keep when there are excursions
and when there are not excursions, respectively. According to
the commenter, they both require exactly the same records. Since
there is no difference, and since other portions of the rule
already require a record of this information, the commenter
maintained that paragraphs (d)(4) and (d)(5) serve no purpose and
should be deleted.
Response: The EPA agrees with Commenter IV-D-04 that
§63.1439(d)(4) and (5) are redundant with other portions of the
rule that already require a record of this information.
Therefore, §63.1439 (d) (4) and (5) have been "reserved" in the
final rule.
2.15.10 Recordkeepinq Waiver
Comment: One commenter (IV-D-04) claimed that paragraph
§63.1439(d)(9) is inappropriate and should be deleted. It
requires that records be kept if industry has obtained a waiver
of recordkeeping requirements, which defeats the purpose of
having a waiver of recordkeeping requirements. The commenter
referred to the General Provisions, subpart A, §63.10(f) (on
2-163
-------
which paragraph §63.1439(d)(9) relies), which allows a waiver in
any of three circumstances. The commenter maintained that
§63.10(f)(5) of subpart A, which provides that a waiver may be
conditioned on other recordkeeping or reporting requirements
deemed necessary by the Administrator, already provides for any
necessary records, and, therefore, paragraph (d)(9) is
unnecessary.
Response: The EPA disagrees that §63.1439(d)(9) is
unnecessary, but has revised §63.1439(d)(9) to be consistent with
§63.10(f)(5). This change is being made as a further measure to
reduce the recordkeeping burden imposed by subpart PPP on owners
and operators, by overriding, in Table 1 of subpart PPP,
§63.10(b)(2) (§63.10(b)(2)(xii) requires the information that was
required in the proposed version of §63.1439(d)(9)). In the
final rule, §63.1439(d)(9) reads as follows:
(9) The owner or operator of an affected source
granted a waiver of recordkeeping or reporting recruirements
under the General Provisions' recordkeepinq and reporting
recruirements in §63.10(f) shall maintain the information, if
any, specified by the Administrator as a condition of the
demonstrating whether an affected source io meeting the
rcquircmcnto for a waiver of recordkeeping or reporting
requirements.
2.15.11 "Document on Demand"
Comment: One commenter (IV-D-05) referred to §63.1439(d)(8)
which says exempt flexible operation units must "maintain the
documentation" required by §63.1420(e)(7). The commenter
recommended that this provision be revised to include an option
that does not involve constantly maintaining documentation of the
unit's exempt status (i.e., allowing the "document on demand"
option) and provided the following recommended revised wording:
(8) For each flexible operation unit in which the
primary product is determined to be something other than a
polyether polyol product, the owner or operator shall either
maintain the documentation specified in §63.1420(e)(7) or be
able to document upon recruest that the primary product is
not a polvether polvol."
2-164
-------
Response: The EPA agrees and the final rule allows the
owner or operator the option of providing documentation on
demand, showing that the primary product of a flexible operation
unit is not a polyether polyol. Therefore, the proposed
requirement under §63.1439(d)(8), referred to by the commenter,
has been deleted, and §63.1420(e) (8) now allows the owner or
operator of a flexible operation unit with something other than a
polyether polyol as its primary product to maintain documentation
of that fact or produce documentation on demand.
2.15.12 Notification of Compliance Status
Comment: One commenter (IV-D-05) referred to
§63.1439(e) (5) (i) which says to include any other information
"required to be included" in the Notification of Compliance
Status (NOCS) under a variety of HON sections. The commenter
requested that EPA provide a list of the specific data this
section requires.
Response: The EPA appreciates the comment. However, the EPA
provides different compliance choices, and it would be lengthy
and confusing to describe every bit of information for every
compliance option in subpart PPP. For example, §63.1439(e)(5)(i)
references §63.1422 (j) of this rule, which provides for overlap
with other regulations for monitoring, recordkeeping, or
reporting with respect to combustion devices, recovery devices,
or recapture devices. Under these overlap provisions, the owner
or operator has compliance choices, but he must notify the
Administrator of his choice in the NOCS required by
§63.1439(e)(5).
In addition to specific information required in
§63.1439(e)(5)(i), this section requires "any other information
required to be included" in the NOCS under other sections, as
applicable. If any information in the referenced provisions is
applicable, then that information must also be included, as
appropriate. Therefore, the owner or operator of an affected
source must review the referenced provisions and submit any
2-165
-------
information that is required to be reported in the NOCS, as
applicable.
2.15.13 Addition of a Provision for Reporting Updates
Comment: Two commenters (IV-D-04 and IV-D-05) stated that
throughout the proposed rule, there are various places where the
EPA says certain information must be submitted in a specific
(named) one-time report. However, there are circumstances where
the same type of information may need to be submitted later. For
example, there may be changes to the process, or additional
emission points, etc., which could justify either a new report,
or an update to a previous report. The commenter requested that
EPA add a provision that specifies how and when to report such
information. Commenter (IV-D-05) provided suggested language.
Response: The Agency has considered these comments and
agrees that there will be circumstances where certain information
already reported may need to be supplemented or updated.
Therefore, in order to allow the submittal of such information,
the Agency has amended paragraphs 63.1439(e), (e)(1), and (e)(4)
as follows:
(e) Reporting and notification. 4t) In addition to
the reports and notifications required by 40 CFR part 63,
subpart aubparta A and H of thin part, as specified in this
subpart, the owner or operator of an affected source shall
prepare and submit the reports listed in paragraphs (e) (3)
through (o)(9)(8) of this section, as applicable. All
reports recruired by this subpart, and the schedule for their
submittal, are listed in Table 8 of this subpart.
(1) Violation of reporting requirements. Owners and
operators shall not be in violation of the reporting
requirements of this paragraph (e) for failing to submit
information required to be included in a specified report if
the owner or operator meets the requirements in paragraphs
(e)(1)(i) through (iii) of this section. Examples of
circumstances where this paragraph may apply include
information related to newly-added eguipment or emission
points, changes in the process, changes in equipment
recruired or utilized for compliance with the reoruirements of
this subpart, or changes in methods or equipment for
monitoring, recordkeeping, or reporting.
(i) The information was not known in time for
inclusion in the report specified by this subpart.
2-166
-------
(ii) The owner or operator has been diligent in
obtaining the information.
(iii) The owner or operator submits a report according
to the provisions of paragraphs (e) (1) (iii) (A) through (C)
of this section.
(A) If this subpart expressly provides for supplements
to the report in which the information is required, the
owner or operator shall submit the information as a
supplement to that report. The information shall be
submitted no later than 60 days after it is obtained, unless
otherwise specified in this subpart.
(B) If this subpart does not expressly provide for
supplements, but the owner or operator must submit a request
for revision of an operating permit pursuant to the State
operating permit programs in part 70 or the Federal
operating permit programs in part 71, due to circumstances
to which the information pertains, the owner or operator
shall submit the information with the request for revision
to the operating permit.
(C) In any case not addressed by paragraph
(e) (1) (iii) (A) or (B) of this section, the owner or operator
shall submit the information with the first Periodic Report,
as recruired by this subpart, which has a submission deadline
at least 60 days after the information is obtained.
In §63.1439(e)(4), paragraph (e)(4)(i) has been amended and
a new paragraph ((e)(4)(vii)) has been added in response to
comments, as follows:
" (i) Submittal dates. The Precompliance Report shall
be submitted to the Administrator no later than 12 months
prior to the compliance date. Unless the Administrator
obiects to a recruest submitted in the Precompliance Report
within 45 days after its receipt, the recruest shall be
deemed approved. For new affected sources, the
Precompliance Report shall be submitted to the Administrator
with the application for approval of construction or
reconstruction required in paragraph (b)(2) of this section.
Supplements to the Precompliance Report may be submitted as
specified in paragraph (e)(4)(vii) of this section."
"(vii) Supplements to the Precompliance Report may be
submitted as specified in paragraph (e)(4)(vii)(A) or
(e)(4)(vii)(B) of this section. Unless the Administrator
objects to a recruest submitted in a supplement to the
Precompliance Report within 45 days after its receipt, the
recruest shall be deemed approved.
(A) Supplements to the Precompliance Report may be
submitted to clarify or modify information previously
submitted.
(B) Supplements to the Precompliance Report may be
submitted to recruest approval to use alternative monitoring
parameters as specified in paragraph (e)(4)(iii) of this
2-167
-------
section; to use alternative continuous monitoring and
recordkeeping, as specified in paragraph (e)(4)(iv) of this
section; to use alternative controls, as specified in
paragraph (e)(4)(v) of this section; or to include a
provision for ceasing to collect monitoring data during a
start-up, shutdown, or malfunction, in the start-up,
shutdown, and malfunction plan, when that monitoring
ecniipment would be damaged if it did not cease to collect
monitoring data, as specified in paragraph (e)(3)(vi) of
this section,"
2.15.14 Revise Exemptions from Recordkeepinq
Comment: One commenter (IV-D-04) requested that EPA revise
§§63.1430(e)(1)(vi) and (e)(2) to avoid unfair double penalties
and retroactive violations. Section 63.1430(e)(1)(vi) provides
an exemption from certain recordkeeping requirements if batch
vents meet two requirements: (1) the batch vents are in
compliance with the aggregate 90 percent HAP reduction
requirement of §63.1425(c)(1); and (2) the control device is
operating at all times. Similarly, §63.1430(e)(2) provides an
exemption from recordkeeping if certain continuous process vents
are "in compliance with" the Group 1 requirements of
§63.1425(c)(3). The commenter cited the example of a facility
that was relying on this exemption from recordkeeping, and then
(perhaps 10 years later) the control device went "down" or there
was an instance of noncompliance with the emission control
requirements. This could result in a penalty for failure to meet
the required level of control. Also, it appears that the owner
or operator would immediately lose the exemption from
recordkeeping. For these reasons, the commenter requested that
EPA revise these two paragraphs in §63.1430(e) as follows:
"(1)* * *
(vi) If the combination of all process vents from batch
unit operations associated with the use of an organic HAP to
make or modify a polyether polyol product process vent ic in
compliance with is subject to §63.1425(c)(1), and the
combustion, recovery, or recapture device is operating
intended to operate at all times, none of the records in
paragraphs (b)(1)(i) through (b)(1)(v) of this section are
required.
(2) Process Vents from Continuous Unit Operations.
. . .Owners or operators of combined streams that are -i
2-168
-------
compliance with subject to the Group 1 requirements of
§63.1425(c)(3) are not required to keep these records."
Response: The Agency has reviewed the provisions in
§§63.1430(e)(1)(vi) and (e)(2) and agrees with the commenter that
some clarification is needed to avoid the possibility of double
penalties and retroactive violations. The appropriate
clarifications have been incorporated into the final rule.
2.15.15 Recordkeepinq Retention Revision
Comment: Two commenters (IV-D-04, IV-D-05) expressed
concern over the wording in §63.1439(h)(1)(vi)(D) which requires
industry to keep certain records for a specified period beginning
when the records are "last employed." The commenters requested
that this section be revised to base all mandatory retention
periods for records on the date when the record was "created,"
not on the date when the record as last employed. This revision
would be consistent with the HON litigation settlement
amendments. As an alternative, these provisions could allow
owners or operators to send the superseded documents to EPA
instead of requiring industry to keep them. Commenter IV-D-05
provided the following wording for §63.1439(h)(1)(vi)(D), which
is consistent with the HON litigation amendments.
"Owners and operators subject to paragraph
(h)(1)(vi)(B) of this section shall retain the current
description of the monitoring system as long as the
description is current, but not less than 5 years from the
date it was created last employed...."
Response: For purposes of consistency with other rules, the
Agency has amended §63.1439(h)(1)(vi)(D) to base retention
periods for records on the date when the record was created.
Therefore, paragraph (D) has been amended as follows:
(D) Ownoro and opcratora The owner or operator subject
to paragraph (h)(1)(vi)(B) of this section shall retain the
current description of the monitoring system as long as the
description is current-;—but not Icon than 5 years from the
date it was laat employed. The current description shall,
at all times, be retained on-site or be accessible from a
central location by computer or other means that provides
2-169
-------
access within 2 hours after a- request. The owner or
operator shall retain the moat roacnt all superseded
descriptions for at least until 5 years from after the date
it WQO last employed of their creation. Superseded
descriptions shall be retained on-site (or accessible from a
central location by computer or other means that provides
access within 2 hours after a recruest) for at least 6 months
after their creation. Thereafter, superseded descriptions
may be stored off-site."
The vague time frame indicated by the use of the term "last
employed" has been replaced with the more concrete date (the date
of the document's creation) as requested by the commenter, but
the new language is more specific than the HON, in that it
requires that all descriptions less than 5 years old be
maintained. This ensures that there will always be a record of
the past five years, no matter how often the descriptions are
replaced.
2.15.16 Exclusion of Ecruipment Leaks to the Periodic Reporting
Requirements
Comment: One commenter (IV-D-04) stated that
§63.1439 (e) (6) (iii) (D) (2.) should not require a report every time
a valve or connector is installed and requested that EPA
expressly exclude equipment leaks, because it is routine to add
new equipment leak points. Therefore, the commenter requested
that §63.1439(e) (6) (iii) (D) (2J be revised as follows:
"Notification if one or more emission points (other than
equipment leaks), or one or more PMPU is added to an
affected source."
Response: The Agency agrees that a report should not be
required every time a routine valve or connector is installed.
Therefore, the Agency has amended §63.1439 (e) (6) (iii) (D) (2.) in
the final rule, but changed the wording to be consistent with the
rest of the rule. Section 63 .1439 (e) (6) (iii) (D) (2.) in the final
rule reads:
"Notification if one or more emission points (other than
equipment leak components subject to §63.1434), or one or
more PMPU is added to an affected source. The owner or
operator shall submit the information contained in
2-170
-------
paragraphs (c) (6) (iii) (D) (2) (ii) (e) (6) (iii) (D) (2) (i) and
(ii) through (o) (6) (iii) (D) (3) (iii) of this section."
2.15.17 Record of Preparation of Standards
Comment: One commenter (IV-D-05) maintained that the
"record of preparation of standards," cited in
§63.1439(e)(5)(i)(B), should not be required for standards not
prepared by the source, such as standards that are obtained from
EPA or that are obtained as certified standards.
Response; The Agency agrees that the "record of preparation
of standards" should not be required for standards not prepared
by the owner or operator of the source. Therefore,
§63.1439(e) (5) (i) (B) has been amended to incorporate this change.
2.15.18 Periodic Reports
Comment: One commenter (IV-D-04) stated that
§63.1439(e)(6)(ii) requires a certification that the affected
source was "in compliance" for the previous 6-month period. The
commenter believed that this section requires a certification
even if there were compliance exceptions. If so, then the
commenter requested that the paragraph be revised to say that.
Commenter IV-D-04 further claimed that §63.1439(e)(6)(i)
appears to have the timing backward. Paragraph 63.1439(e)(6)(i)
states that, after the first Periodic Report, subsequent reports
must cover each "preceding" 6-month period. This seems to be
backward. To the commenter, "preceding" means "earlier." So,
taken literally, this means the second Periodic Report must cover
the 6-month period before the period that was covered in the
first Periodic Report and so on. The commenter recommended that,
instead of "preceding," the paragraph should use the word
"subsequent." Another commenter (IV-D-05) provided the following
revision of paragraph §63.1439(e)(6)(ii) for clarity:
"If none of the compliance exceptions in paragraphs
(e) (6) (iii) through (e) (6) (vii) of this section
occurred during the 6-month period, the Periodic Report
required by paragraph (e)(6)(i) of this section shall
be a statement that the affected source waa in
2-171
-------
compliance there were no compliance exceptions as
described in this paragraph, for the preceding 6-month
period and that none of the activities specified in
paragraphs (e)(6)(iii) through (e)(6)(vii) of this
section occurred."
Commenter IV-D-04 supported edits to this section that were
suggested by commenter IV-D-05.
Response: Because §63.1439(e)(6)(viii)(A) discussed
reporting requirements for time periods including compliance
exceptions, Commenter IV-D-04 is mistaken in thinking that
§63.1439(e)(6)(ii) needed to be edited to included periods with
compliance exceptions. The purpose of §63.1439(e)(6)(ii) is to
minimize the periodic reporting burden on owners or operations
with no compliance exceptions. However, the EPA agrees with the
commenters that the proposed language in §63.1439(e)(6)(ii) was
unclear, both regarding what must be reported, and what time
period the report is expected to cover. The EPA agrees that "the
affected source was in compliance" should be replaced with the
term "there were no compliance exceptions." However, the EPA does
not agree that replacing the word "preceding" with the word
"subsequent" would correctly express the intent of the paragraph.
Rather, the EPA has revised §63.1439(e)(6)(ii) to read as
follows:
"If none of the compliance exceptions in paragraphs
(e)(6)(iii) through (o)(6)(vii) of this section occurred
during the 6-month period, the Periodic Report required by
paragraph (e)(6)(i) of this section shall be a statement
that the affected aourcc waa in compliance there were no
compliance exceptions as described in this paragraph, for
the preceding 6-month period covered by that report and that
none of the activities specified in paragraphs (e)(6)(iii)
through (o)(6)(vii) of this section occurred during the
period covered by that report."
The EPA feels that §63.1439(e) (6) (i) is clear about when Periodic
Reports are due, and what time period each covers, and has made
no changes to the proposed language for §63.1439(e)(6)(i).
2-172
-------
2.15.19 Predominant Use Reporting
Comment: Two commenters (IV-D-04, IV-D-05) requested that
§63.1439(e)(6)(vi) be clarified. This section requires a report
for any change to the predominant use determination for a storage
vessel. The commenters believed that this meant a storage vessel
that belongs to the affected source after the change. The
commenters requested that the paragraph be revised to clarify the
intent, but each had different suggestions. Commenter IV-D-05
suggested the following language, noting that the reference
should be to paragraph (f)(8) and not (f)(6):
"The results for each change to a predominant use
determination for a storage vessel belonging to an affected
source subject to this subpart after the change that is made
under §63.1420(f)(8)."
Response: For the purposes of the Periodic Report
requirements [§63.1439(e)(6)(vi)], the EPA is interested in any
subsequent action that may change the predominant use of a
storage vessel. However, the EPA does believe it is appropriate
to specify the changes that must be reported. Therefore, the
final rule requires that the results for each reevaluation of
predominant use of a storage vessel be reported if the vessel
begins receiving material from (or sending material to) a process
unit that was not included in the initial determination, or if
the storage vessel ceases to receive material from (or send
material to) a process unit that was included in the initial
determination. Also, because the EPA is only interested in
changes in the predominant use of the storage vessel, only the
paragraphs in §63.1420(f) that apply to changes in predominant
use should be referred to in §63.1439(e)(6)(vi). For this
reason, the EPA agrees with commenter IV-D-05 about the cross-
reference, and has changed the reference in §63.1439(e)(6)(vi) so
that it refers to §63.1420(f)(8).
Comment: Commenter IV-D-04 believed that the citations to
§63.1420(f)(6) in §§63.1439(e)(5)(v) and (e)(6)(vi) should
probably refer to §63.1420(f)(3) and not (f)(6), because
2-173
-------
paragraph (f)(3) deals with assigning storage vessels to process
units on the basis of predominant use.
Response: Section 63.1439(e)(5)(v) pertains to initial
predominant use determination, whereas §63.1439(e)(6)(vi)
pertains to a change in predominant use. For these reasons,
§63.1439(e)(5)(v) should cite §63.1420(f)(1) through (7), which
specify procedures to follow for initial predominant use
determination. Further, §63.1439(e)(6)(vi) should cite
§63.1420(f) (8), which addresses a change in the utilization of
the storage vessel.
2.15.20 Alternative Continuous Monitoring and Recordkeepina
Comment: One commenter (IV-D-04) agreed with
§63.1428(g) (3) (ii), which states that if process changes do not
result in a change in Group status at a Group 2 process vent from
a batch unit operation, no reporting is required. However, the
commenter requested that a similar paragraph be added specifying
that, if the group status of a Group 2 process vent from a
continuous unit operation is unchanged, no report would be
required.
Response: The Agency has added a new paragraph as
§63.1428(h)(2)(iii), clarifying that if, after the TRE index
value recalculation, it is determined that a Group 2 process vent
from a continuous unit operation has a TRE index value of 4.0 or
greater, no report is required for that Group 2 process vent.
However, the EPA is still requiring a report to be submitted if
the TRE index value changes from being above 4.0 to less than 4.0
but greater than 1.0, despite the fact that the process vent
remains Group 2. Therefore, the requirements in
§63.1428 (h) (2) (ii) and §63.1430 (j) have not been changed based on
this comment.
2.15.21 Consolidation of Periodic Reporting
2-174
-------
Comment: Two commenters (IV-D-04, IV-D-05) requested that
§63.1439(e)(6) be clarified to allow a single Periodic Report
instead of three different Periodic Reports for existing and new
affected sources, one for the general requirements, one for
equipment leaks, and one for heat exchange systems. The
commenters suggested adding the following sentence to the end of
§63.1439(e)(6) to accomplish this: "All of the information
required to be reported in this subsection may be submitted in
one report."
Response: The Agency agrees that the reports required under
§63.1439(e)(6) may be combined into one report. Therefore, the
EPA has added the phrase "as part of the Periodic Report required
by this paragraph (e)(6)" to the end of §63 .1439 (e) (6) .
2.15.22 Group Status Change Reporting
Comment: One commenter (IV-D-04) expressed concern about the
requirement in §63.1439(e) (6) (iii) (D) (I) to report any process
change if "the group status of any emission point changes." This
would seem, literally, to require reporting even if the status
changed from Group 1 to Group 2. The commenter claimed that
reports should be required only if the status changes from Group
2 to Group 1. The commenter requested that this paragraph be
revised to say "the group status of any emission point changes
from Group 2 to Group 1. "
Response: The intent of §63.1439(e)(6)(iii)(D)(1) in the
proposed rule was to require reporting only if the status changes
from Group 2 to Group 1. However, the EPA would like to remind
owners and operators that, until notification is made that a
Group 1 emission point has become a Group 2 emission point, the
owner or operator will be required to comply with the Group 1
requirements for that emission point. Therefore, to clarify this
intent, the Agency has amended the first sentence of paragraph
§63.1439(e)(6)(iii)(D)(1) as follows:
2-175
-------
"Notification if a process change is made such that the
group status of any emission point changes from Group 2 to
Group 1. The owner or operator is not required to submit a
notification of a process change if that process change
caused the group status of an emission point to change from
Group 1 to Group 2. However, until the owner or operator
notifies the Administrator that the group status of an
emission point has changed from Group 1 to Group 2, the
owner or operator is recruired to continue to comply with the
Group 1 recruirements for that emission point. This
notification may be submitted at any time. The information
submitted shall include a compliance schedule,—as specified
in paragraphs—(c) (6) (iii) (D) (2J (±) and
(o) (6) (iii) (D) (2) (ii)of this section,—for emission points
that change from Group 2 to Croup 1 as specified
§63.1420(g) () ;—or for process vents under the conditions
listed in §63.1429(g) (3) (i) . This information may be
submitted in • a separate report,—as specified in §63.1430(i).
(•±3 The owner or operator shall submit to the
Administrator for approval a compliance schedule and a
justification for the schedule.
(4-jr) The Administrator shall approve the compliance
schedule or request changes within 120 days of receipt of
the compliance schedule and justification.
2.15.23 Excursions
Comment: One commenter (IV-D-04) noted that neither emission
points nor process sections have excursions. The commenter
stated that only control devices or recovery devices have
excursions. Therefore, it was requested that
§63.1439(e)(6)(viii) be revised to specify "control devices" or
"recovery devices" instead of "emission point" or "process
sections." Another commenter (IV-D-05) expressed the same
concern and suggested that "control device" be used in lieu of
"process section" in §63.1439(e)(6)(viii).
Response: The Agency agrees that neither emission points
nor process sections have excursions. Paragraph
§63.1439(e)(6)(viii) of the final rule refers to "A control or
recovery device for a particular emission point or process
section" that has more excursions, as defined in §63.1438(f), than
the number of excused excursions allowed under §63.1438(g). In
the final rule, §63.1439(e)(6)(viii) also provides more specific
2-176
-------
guidelines for when the quarterly reports are due, and what they
must contain.
2.15.24 Alternative Continuous Monitoring for Storage Vessels
Comment: One commenter (IV-D-04) requested that §63.1439(f)
be clarified to exclude storage vessel monitoring plans. This
section establishes a detailed procedure for situations where an
owner or operator has been directed to set unique monitoring
parameters. According to the commenter, this detailed procedure
(apparently borrowed from the HON) was not intended to apply, and
is not appropriate, in situations where the rule directs an owner
or operator to establish a monitoring plan for storage vessels.
The corresponding paragraph of the HON was amended to make that
clear, by saying paragraph (f) applies only when specifically
referenced. The commenter requested that EPA do likewise in this
rule. Another commenter, IV-D-05, recommended revising the
provisions to avoid the misimpression that alternative monitoring
parameters must be requested for every storage vessel monitoring
plan and provided the following revision of paragraph (f):
"The owner or operator who has been directed by any section
of this subpart that specifically references this paragraph or
any section of another subpart referenced by this subpart that
specifically references this paragraph to set unique monitoring
parameters...."
Commenter IV-D-04 supported this recommended language
change.
Response: The Agency agrees that this detailed procedure
was not intended to apply in situations where the rule directs
the establishment of a monitoring plan for storage vessels.
Therefore, in order to clarify that these provisions apply only
when specifically referenced, §63.1439(f) has been amended as
follows:
" (f) Alternative monitoring parameters. The owner or
operator of an affected source who has been directed by any
section of this subpartj. or any section of another subpart
referenced by this subpart, that specifically references
2-177
-------
this paragraph to set unique monitoring parameters, or who
requests approval to monitor a different parameter than
those listed in §63.1432 for storage vessels, §63.142-6-7. for
ECO, §63.1429 for process vents, or §63.143 of aubpart G for
process wastewater shall submit the information...."
2.15.25 Cross-referencing in Reduced Recordkeepina
Comment: One commenter (IV-D-04) requested that EPA either
add more cross-references, or delete the current cross-
references, from §63.1439(h). This section provides that a
"reduced recordkeeping" program may be implemented as an
alternative to the continuous operating parameter monitoring and
recordkeeping requirements in the following three locations:
§63.1432 for storage vessels; §63.1429 for process vents; and
§63.1433 for wastewater. According to the commenter, the HON (in
which the reduced recordkeeping program originated) has a total
of ten cross-references to locations where the "normal" operating
parameter monitoring and recordkeeping requirements are located.
Since the proposed rule is as complex as the HON, the commenter
questioned whether the proposed rule omits any necessary cross-
references from this paragraph. The commenter suggested that the
easiest way to resolve this concern is to revise the first
sentence of §63.1439(h) as follows:
"For any parameter with respect to any item of
equipment, the owner or operator may implement the
recordkeeping requirements in paragraph (h)(l) or (h)(2) of
this section as alternatives to the continuous operating
parameter monitoring and recordkeeping provisions listed in
§63.1432 for storage voooolo, §63.1429 for process vcnto,
and §63.1433—for waatowatcr that would otherwise apply under
this subpart...."
Response; The Agency agrees that the proposed language may
lack clarity. Therefore, the Agency has revised the first
sentence of §63.1439(h) as suggested by the commenter.
2.15.26 Operating Permit Application
Comment: Two commenters (IV-D-04, IV-D-05) stated that
§63.1439(e)(8) is unclear. Commenter IV-D-05 claimed that there
2-178
-------
are many other things in addition to the information listed in
paragraph (e)(4) that must be submitted in an operating permit
application and strongly recommended that EPA provide a
comprehensive list of the monitoring, recordkeeping, reporting,
and operating permit requirements applicable to this rule.
Therefore, at a minimum, for clarity, both commenters suggested
the following revision:
"(8) Operating permit application. An owner or operator
who submits an operating permit application instead of a
Precompliance Report shall submit the information specified in
paragraph (e) (4) of this section, Prccomplianco Report,—as
applicable, with the operating permit application."
Response: The Agency recognizes that the information
required in an operating permit application goes beyond that
specified in §63.1439(e)(4). Therefore, the EPA has revised
§63.1439(e)(8) as suggested by the commenters.
2.15.27 Manual Reading
Comment: One commenter (IV-D-05) stated that
§63.1439(g)(2)(i) refers to "manual" reading and recording of
parameter values, but instead should say "visual" or "sensory."
Response: For clarity, the Agency has substituted the word
"visual" for "manual" in §63.1439(g)(2)(i) in the final rule.
2.15.28 "Any Other Information" in Sections 63.1439(e)(5)(i)(B)
and 63.1439(e)(6)(iv)(B)
Comment: Two commenters (IV-D-04, IV-D-05) expressed concern
over the use of the words "any other information" in
§§63.1439 (e) (5) (i) (B) and 63 .1439 (e) (6) (iv) (B) , as well as the
words "any information" or "any information required" that appear
elsewhere in the proposed rule. The commenter stated that
compliance with such vaguely worded provisions is impossible and
requested that EPA avoid using these words or carefully describe
where industry must look for the referenced requirement.
Commenter IV-D-05 recommended amending these paragraphs to read:
2-179
-------
"... any other information required by the test method to be in
the test report."
These two commenters also questioned the need for the words
"and any other required information" in the last sentence of
§63.1439(e)(5)(i)(A) which reads, "For additional tests performed
for the same kind of emission point using the same method, the
results and any other required information shall be submitted,
but a complete test report is not required." The commenter
believed that these words were confusing because they have no
idea what "any other required information" means. They requested
that EPA delete these words from this paragraph. They maintained
that if there is some specific information that EPA wants
submitted, this paragraph should either name it, or give the
specific citation where it is identified.
Response: The commenter's request that the EPA change "any
other information" or "any information" to "any other information
required by the test method to be in the test report" clarifies
the EPA's intent. This revision was incorporated throughout the
final rule, whenever those phrases referring to test methods were
used.
2.15.29 §63.1439(g)(3)
Comment: One commenter (IV-D-05) claimed that
§63.1439(g)(3), allowing the recording of hourly averages instead
of 15-minute data points, is no longer necessary as the regular
provisions of the rule allow the same averaging. The commenter
therefore recommended deleting subsection (g)(3) and renumbering
subsection (g)(4) accordingly.
Response: The EPA agrees with the commenter that it is not
necessary to state that hourly averages are allowed in
§63.1439(g)(3); however, the EPA chooses to retain this language
in the final rule as a clarification. The EPA feels that this
clarity is warranted, since many other rules require 15-minute
data points in their definition of a continuous recorder.
2-180
-------
2.16 EDITORIAL
2.16.1 "The Owner or Operator" Versus "Each Owner or Operator"
Comment: One commenter (IV-D-05) maintained that usually the
text requires "the owner or operator" or "an owner or operator"
to comply. If "each" owner or operator or "all" owners and
operators were required to comply, in cases where there are two
or more companies involved (e.g., a joint venture), then the
burdens of the rule would be duplicated for no reason. The
commenter did not believe that is EPA's intent. The commenter
requested EPA to change "each owner or operator" to "the owner or
operator" and change "owners and operators" to "the owner or
operator" (and change the following verb agreement as
appropriate) in the following sections: 63.1420(b);
63.1420(e) (5) (i); 63 .1420(e) (5) (ii); 63.1429(a); 63.1430(b);
63.1430(d); 63.1430(e) (1); 63.1430(e) (2); 63.1430 (f) (2) ;
63.1430(f) (3); 63.1430(f) (4); 63 .1430(f) (5) ; 63.1420(g)(1);
63.1439(a); 63.1439(b)(2); 63.1439(e)(3); 63.1439(e) (5) ;
63.1439(e)(6).
Response: The EPA appreciates the comment, and agrees with
the commenter that the EPA did not intend to duplicate the
responsibilities of the rule to joint venture partners.
Therefore, the EPA has revised the final rule as requested.
2.16.2 Section 63.1420(e)(3)
Comment: One commenter (IV-D-05) recommended deleting the
word "considered" from the first sentence in §63.1420(e)(3).
According to the commenter, the process either is a PMPU, or is
not a PMPU. The commenter also suggested moving the phrase "if
the plant site is a major source" from the end of the second
sentence to the beginning of that sentence to clarify that the
whole sentence applies only to major sources. For grammatical
reasons, it was also suggested that the word "is" in the second
sentence be deleted.
2-181
-------
Response: The EPA appreciates the comments and has made
these revisions in the final rule at §63.1420(e)(1)(iv), which
was §63.1420(e)(3) at proposal.
2.16.3 Section 63.1420(e)(5)(i)
Comment: One commenter (IV-D-05), for clarity, recommended
changing the first sentence in §63.1420(e)(5)(i) to read:
"... shall determine the applicability of the
provisions (e.g., the Group status) for each emission point
that is part of that flexible operation unit ...."
The commenter, for clarity, also recommended changing the
last sentence in paragraph (i) to read:
"... Based on this finding, the owner or operator shall
comply with the applicable standards of this subpart for each
emission point, as appropriate, at all times, rcgardlooo of what
as though the primary product is being produced."
Response: The EPA has rewritten §63.1420(e), and has
eliminated §63.1420(e)(5)(i) and the phrases "regardless of what"
and "shall determine the applicability of the provisions" from
this section of the final rule.
2.16.4 Section 63.1420(e)(5)(iii)
Comment: One commenter (IV-D-05) stated that paragraph
63.1420(e)(5)(iii) states that the owner or operator shall comply
"only with either" (b)(1) or (b)(2). That could be interpreted
as forbidding compliance with both. The commenter recommended
changing the text to delete "only":
"The owner or operator shall comply with either paragraph
(b) (1) or (b) (2) "
Response: The EPA did not intend for §63.1420(e) (5) (iii)
to be interpreted as forbidding compliance with both (b) (1) and
(b) (2) . Therefore, in the final rule [as §63 .1420 (e) (5) (i) ] ,
this language has been clarified to avoid any possible
misinterpretation.
2-182
-------
2.16.5 Sections 63.1427(h)(1) and (h)(2)
Comment: One coimnenter (IV-D-05) stated that §63 .1427 (h) (1)
and (h)(2) allow the owner or operator to monitor or establish,
respectively, "one" of the parameters in paragraphs (i) through
(iii) . However, at the end of §63.1427 (h) (1) (ii) and (h) (2) (ii)
the proposed rule said "and" instead of "or." Therefore, the
commenter recommended revising the text in §63.1427(h)(1)(ii) and
(2)(ii) to say "or."
Response: The Agency agrees with this comment. However,
since there is an "or" at the end of both §63.1427(i)(1)(iii) and
(i) (2) (iii) (which were §63 .1427 (h) (1) (iii) and (h)(2)(iii) at
proposal) , the word "and" was simply removed at the end of
§63.1427(1)(1)(ii) and (i)(2)(ii), in the final rule.
2.16.6 Section 63.1427(1)(iv)
Comment: One commenter (IV-D-05) understood that the
concentration in §63.1427(i)(1)(iv) is to be determined at the
"onset of the ECO" rather than at the "end of the onset of ECO."
Therefore, the commenter recommended revising the text
accordingly.
Response: The phrase "end of the onset of ECO" has been
corrected in the final rule to use the correct phrase, "onset of
ECO" .
2.16.7 Section 63.1433(a)(9)
Comment: One commenter (IV-D-05) requested that
§63.1433(a)(9) be revised to add "applicable" to read "the
applicable compliance dates specified in §63.1422 shall
apply...."
Response: The Agency agrees with the commenter's suggestion
and has revised the final rule accordingly.
2.16.8 Section 63.1433(b)
2-183
-------
Comment: One commenter (IV-D-04) requested that EPA give
§63.1433(b) a heading, i.e., "Maintenance wastewater" to call
attention to it, because it is brief and could be considered part
of the preceding text for process wastewater.
Response: The EPA appreciates the suggestion to provide a
title for §63.1433(b) such as "Maintenance wastewater."
Therefore, the EPA titled that section as requested and, for
consistency, titled §63.1433(a) "Process wastewater."
2.16.9 Section 63.1431(f)(1)
Comment: One commenter (IV-D-05) recommended, for clarity,
that the text in §63.1431(f)(1) be revised to delete the word
"control" at the end of the second sentence as follows:
"The owner or operator shall notify the Agency of the
intent to comply with the epoxide emission factor limitation
in §63.1425(b)(1)(iii) or (b)(2)(iii) without the use of
extended cookout or a combustion, recovery, or recapture
device. The owner or operator shall prepare an estimate of
the annual epoxide emissions control."
Response: The Agency agrees with the commenter that the
word "control" is inappropriate at the end of this sentence. The
Agency has revised the final rule to delete this word.
2.16.10 62 FR 46812, col. 3
Comment: One commenter (IV-D-04) requested that the EPA
correct the statement in the proposal preamble (62 FR 46812, col.
3) that referred to the Start-up, Shutdown and Malfunction Plan
as a "report" that must be submitted to the Administrator. The
commenter noted that the proposed rule, §63.1439(b)(1) ,
accurately considers the start-up, shutdown, and malfunction plan
to be a record that must be retained on-site.
Response: The Agency realizes that the reference to the
Start-up, Shutdown and Malfunction Plan as a report was incorrect
2-184
-------
in the preamble. The preamble for the final rule does not
include this erroneous information.
2.16.11 Section 63.1438(b)(3)
Comment: One commenter (IV-D-05), recommended revising
§63.1438(b)(3) to be consistent with other parts of the rule
where "are" is replaced with "shall be," as follows:
"Process vents from batch unit operations. For process
vents from batch unit operations, during initial compliance
testing, the appropriate parameter shall be monitored
continuously during the entire test period. The monitoring
level(s) aro shall be those established during the
compliance test demonstration."
Response: The EPA intended to have consistent language
throughout the rule. Therefore, §63.1438(b)(2) of the final rule
(which was §63.1438(b)(3) at proposal) was revised as suggested
by the commenter.
2.16.12 Heading in §63.1420(c)
Comment: One commenter (IV-D-05) requested that, for
purposes of clarity, the EPA change the heading in §63.1420(c) as
follows:
"Emission points included in the affected source but
not subject to the provisions of this subpart."
Response: The EPA appreciates the comment, and revised the
final rule to add this language to the title.
2.16.13 Section 63.1420(c) (2)
Comment: One commenter (IV-D-05) suggested, for clarity that
EPA revise §63.1420 (c) (2) as follows:
"(2) Stormwater managed in from segregated sewers."
Response: The EPA agrees with this revision and has
incorporated it into the final rule.
2-185
-------
2.16.14 Section 63.1420(d)
Comment: For grammatical reasons, one commenter (IV-D-05)
requested that EPA revise §63.1420(d) as follows:
"...and are exempted from the requirements of both this
subpart and from tho provisions of subpart A."
Response: The EPA agrees that the commenter's suggested
language is grammatically correct. Therefore, the EPA
incorporated the commenter's suggested language into the final
rule.
2.16.15 Section 63.1420(e)(1)
Comment: One commenter (IV-D-05) stated that the word "only"
needs to come after (not before) "manufactures" in
§63.1420(e)(1). In the current wording, it could be interpreted
to mean that a process unit "only manufactures" one product
(instead of manufacturing, processing, and shipping it). It
should read "manufactures only" one product. Therefore, for
clarity, the commenter recommended changing this sentence as
follows:
"If a process unit only manufactures only one product..."
Response: The EPA agrees that the phrase quoted may be
incorrectly interpreted. Therefore, the EPA has changed the
language in the final rule to the language suggested by the
commenter at §63.1420(e)(1)(i), which was §63.1420(e)(1) at
proposal.
2.16.16 Section 63.1422(d)(2)(iv) and (d)(5)
Comment: One commenter (IV-D-04) stated that in
§63.1422(d)(2)(iv) and (d)(5), there are bracketed comments
saying to insert a date that is a certain amount of time "from"
the date of publication of the final rule. The commenter
requested that these bracketed words say "after" the date of
publication of the final rule, as the commenter believes EPA
intended.
2-186
-------
Response: In the proposed rule, these bracketed statements
were included to show the time frame when compliance would be
required. The final rule contains actual dates, so the
commenter's concern no longer exists.
2.16.17 Section 63.1420(f)(7)(i)
Comment: One commenter (IV-D-04) requested that "raw
materials" in §63.1420(f)(7)(i) be revised to read "raw
material."
Response: The commenter's request pointed out the
possibility that someone reading the rule could interpret this
section to pertain only if the process had more than one raw
material. This was not the EPA's intent. Therefore, the EPA
revised the final rule as requested by the commenter.
2.16.18 Section 63.1424(a)
Comment: One commenter (IV-D-05) stated that §63.1424(a)
refers to paragraph (c) or (d), which do not exist. Therefore,
the commenter recommended changing the text as follows:
" (a) Except as provided in paragraphs- (b) through—(-eVf
of this section "
Response: The EPA appreciates the commenter bringing this
typographical error to the EPA's attention. Section 63.1424(a)
of the final rule was revised to only cite paragraph (b).
2.16.19 Section 63 .1425(c) (4) (i)
Comment: One commenter (IV-D-05) maintained that
§63.1425(c) (4) (i) references the wrong subsection. The commenter
recommended revising the text, to refer to §63.1428(h)(2) instead
of §63.1428(h)(4).
Response: The commenter correctly stated that
§63.1425(c) (4) (i) should reference §63.1428(h) (2) and not (h) (4).
2-187
-------
Therefore, the EPA revised the language in the final rule as
suggested.
2.16.20 Section 63.1426(c)
Comment: One commenter (IV-D-05) noted that §63.1426 (c) has
an "either" in the wrong place. Therefore, the commenter
recommended changing the text to read:
"...may be measured either as cither total organic HAP
or as TOC minus methane and ethane...."
Response: The EPA agrees and has revised the final rule,
putting the word "either" in the position suggested by the
commenter.
2-16.21 Section 63.1426(c)(1) fi)(C)(ii). (c)(3)(ii), and (e)(1)
Comment: One commenter (IV-D-07) requested that the
following editorial corrections be made in the final rule:
First, §63.1426(c)(1)(ii) has two improper cites:(1)
§63.1425(b)(1)(ii) is "reserved" and does not require any ppmv
limit; and (2) the reference to §63.1425(b)(2)(iii) should be to
§63.1425(b) (2) (ii) , which is the correct reference to the ppmv
standard.
Second, in §63.1426(c)(3)(ii), "of the HAP of concern"
should be "or the HAP of concern".
Third, in §63.1426(e) (1) , there is a typo just above
equation 6: the word "determination" should be "determined".
Response: The EPA intended to cite §63.1425(b)(1)(ii) as a
concentration cutoff limit for new sources. However, at the time
of proposal, the EPA had not decided on a value for this
concentration limit, and instead simply "reserved" this paragraph.
Between proposal and promulgation, the EPA decided on a 20 ppmv
limit, which was inserted into §63.1425(b)(1)(ii) of the final
rule. Therefore, the EPA disagrees with the commenter's
2-188
-------
interpretation of the first citation for a ppmv HAP limit in
§63.1426(c)(1)(ii).
However, the conunenter correctly stated that the second
citation in §63.1426(c)(1)(ii) should be to §63.1425(b)(2)(ii),
instead of to §63.1425(b)(2)(iii). Therefore, the EPA revised
§63.1426(c)(1)(ii) in the final rule accordingly.
With regard to the commenter's second comment, the EPA
disagrees: §63.1426(c)(3)(ii) was intended to say "of the HAP of
concern." The phrase "of the HAP of concern" refers to the HAP
listed in Table 4 of subpart PPP, which limits the owner or
operator to having to take into consideration the concentrations
of the 6 HAP listed in that table, rather than having to consider
all 188 of the HAP currently listed under §112(b) of the Clean
Air Act. With regard to the commenter's third comment, the EPA
agrees that the word "determination" in §63.1426(e)(1) should have
been "determined," and has made the change suggested by the
commenter, in the final rule.
2.16.22 Section 63.1426(e)(1)
Comment: One commenter (IV-D-05) cited §63.1426(e)(1) which
requires owners or operators to determine the emission reduction
for each group of process vents subject to "the same paragraph"
of §63.1425. According to the commenter, this may work for
paragraphs (b), (c) and (d), but it will not .work for other
paragraphs of §63.1425. Therefore, the commenter recommended
that §63.1426(e)(1) be revised as follows:
"The owner or operator shall determine the organic HAP
emission reduction for process vents in a PMPU using
Equation 6. The organic HAP emission reduction must be
determined for each group of process vents subject to the
same paragraph (i.e., paragraph (b), (c) or (d)) of §63.1425
of this subpart. For instance, process vents that emit
epoxides are subject to paragraph (b) of §63.1425.
Therefore, the organic HAP (i.e., epoxide) emission
reduction must be determinationed for the group of vents in
a PMPU that are subject to this paragraph."
2-189
-------
Response: The EPA agrees that some of the citations in the
proposed §63.1426(e)(1) were incorrect. Therefore, the revisions
suggested by the commenter were incorporated into the final rule.
2.16.23 "Pressure Decline Curve" Versus "Pressure Decay Curve"
Comment: One commenter (IV-D-05) requested, for consistency,
changing "pressure decline curve" in §63.1427(i)(1)(i)(A) and (C)
to "pressure decay curve," a term that is used elsewhere. The
commenter has provided a definition for "pressure decay curve" in
§63.1423(b).
Response: The EPA intended to consistently use the term
"pressure decay curve." Therefore, the EPA replaced the phrase
"pressure decline curve" in §63.1427(j)(1)(i)(A) and (C) (in the
final rule) with "pressure decay curve." As discussed in section
23.17 of this document, the same change was made in the
definition of "product class."
2.16.24 Section 63 .1427(1) (3) (ii)
Comment: One commenter (IV-D-05) recommended, for clarity,
revising the text in §63 .1427(1) (3) (ii) to read:
"If the new operating conditions of the polyether
polvol product do not conform with the operating
characteristics of an existing product class, the owner or
operator shall establish a new product class and shall
comply with provisions of (1)(3)(i)(A) through (C) of this
section."
Response: The EPA inadvertently omitted the word "polyol"
from the sentence quoted by the commenter. Therefore, the EPA
revised the text in §63 .1427 (1) (3) (ii) (§63 .1427 (m) (3 ) (ii) in the
final rule) to insert the word "polyol."
2.16.25 Section 63.1428(c)
Comment: One commenter (IV-D-05) recommended adding the
phrase "as selected by the owner or operator" in the first
2-190
-------
sentence of §63.1428(c) to avoid ambiguity and for accuracy as
follows:
"If the annual emissions of TOC or nonepoxide organic
HAP, as selected by the owner or operator, from the
combination of process vents from batch unit operations
Also, the commenter recommended changing the reference in the
same sentence from §63.1425(c)(4) to (c)(2) for accuracy.
Response: The EPA agrees with the commenter's first point
that a clarifying phrase after the first phrase in §63.1428(c)
would avoid ambiguity. However, the EPA does not believe that
the phrase "as selected by the owner or operator" is appropriate.
Instead, the EPA clarified the first phrase in §63.1428(c) by
adding "as applicable." The commenter also recommended changing
the reference in the same sentence from §63.1425(c)(4) to (c)(2).
The EPA agrees with this recommended change and has incorporated
it into the final rule.
2.16.26 Section 63.1435(d)
Comment: One commenter (IV-D-05) suggested that §63.1435(d)
be changed to use the word "may" instead of the word "should."
Response: The Agency agrees that "should" is inappropriate
but disagrees that "may" is the correct term to use in
§63.1435(d). In the final rule, "should" has been replaced with
"shall."
2.16.27 Section 63.1439(b)
Comment; One commenter (IV-D-04) requested that two minor
errors in §63.1439(b) be corrected:
1. this section requires industry to keep a start-up,
shutdown and monitoring plan "onsite," which should be corrected
to "on-site" or "on site."
2. this section also requires industry to keep the plan "on
record" after it is developed, for a specified amount of time.
The commenter suggested deleting the words "on record" because
2-191
-------
they add no meaning to the sentence and could be misunderstood.
The industry typically keeps their plans on paper or on computer,
not "on record."
Response: The EPA agrees with the commenter on both points.
The EPA revised "onsite" to "on site." However, the entire
sentence containing "on record" has been removed. Therefore, no
additional change is needed.
2.16.28 Section 63.1439(e)(6)(iii)(A)
Comment: One commenter (IV-D-05) referred to the provisions
of §63.1439(e)(6)(iii)(A), which apply to owners or operators
complying with §63.1432 (storage vessels) through §63.1433
(wastewater). The commenter recommended revising this paragraph
to delete the references to process vents and heat exchange
systems because they do not apply.
Response: The EPA agrees that paragraph
63.1439(e)(6)(iii)(A) does not need references for information
required for process vents or heat exchange systems since
paragraph 63.1439(e)(6)(iii) refers to periodic reports for
storage vessels and wastewater. Therefore, the EPA deleted the
references for process vents and heat exchanger systems.
2.16.29 Section 63.1439(e)(6)(iii)(D)(1) and (f)(3)
Comment: One commenter (IV-D-04) pointed out that
§63.1439 (e) (6) (iii) (D) (1) refers to §63 .1429 (g) (3) (i) , and there
is no such section. The last major division in §63.1429 is
"(d)."
Also, §63.1439(e)(6)(iii)(D) has a reference to
(e) (6) (iii) (D) (4) , and there is no such paragraph. They noted
that the reference should probably be changed to
(e) (6) (iii) (D) (3)
In addition, the commenter noted that the word "recording"
on the first line of §63.1439(f)(3) should be "reporting."
2-192
-------
Response: Regarding §63.1439(e)(6)(iii)(D)(1), the EPA
agrees with the coitunenter, and deleted the second part of the
sentence, which at proposal read "or for process vents under
conditions listed in §63 .1429(g) (3) (i)." The EPA also agrees
with the second comment about the incorrect cross-reference. The
EPA changed the final rule to refer to paragraph
(e) (6) (iii) (D) (1) . Further, the EPA agrees that "recording"
should be "reporting" on the first line of §63.1439(f)(3), and
the EPA revised the wording accordingly.
2.16.30 Section 63.1426(c)(4)(iv), (c). and (c)(1)(i)(A)
Comment: One commenter (IV-D-04) requested that the
following editorial changes be made:
§63.1426(c)(4)(iv): Move "respectively" and place after
"total organic HAP."
§63.1426(c): Correct placement of commas as follows:
"...an owner or operator using a combustion, recovery,
or recapture device to comply with an organic HAP
percent reduction efficiency requirement in section
$63.1425,—(b) (2) (i), (c) (1), (c)(3), or (d)^ an
organic..."
§63.1426(c)(1)(i)(A): The word "vent" should be "vents."
Response: The EPA agrees with all of the commenter's
editorial changes, and incorporated them into the final rule.
2.16.31 Equation 9
Comment: Commenter IV-D-05 recommended that Equation 9 for
determining the percent epoxide emission reduction for the batch
cycle in §63.1427(e)(1) be corrected. The commenter recommended
that the EPA revise the numerator, and provided the revised
equation.
Response: The EPA agreed with the commenter that Equation 9
needed to be revised. In the final rule, the Equation 9 reads as
follows:
2-193
-------
R
batchcycle
R ,, . /?,.."
V (F \ C\ addon, i^ ,p \ (] addon, j x
e,u e,E 1QO e, o JQQ
Ee, u
* 100
2.16.32 Section 63 .1425(e) (2) (i)
Comment: Commenter IV-D-05 pointed out that the regulatory
language in §63.1425(e)(2)(i) states that continuous processes
should be controlled according to §63.1428(b). Paragraph
63.1428(b), however, deals with emissions from batch processes.
The only provision in that section for continuous processes is
§63.1428(h). The commenter believed that the regulatory citation
should be changed to §63.1428(h).
Response: The commenter is correct. However, §63.1425(e)
has been reserved in the final rule, and §63.1428(h) is cited at
other places in the rule (e.g., the definition of Group 1
continuous process vent and §63.1425(c)), correctly.
2.16.33 Section 63.1425(f)
Comment: For clarity, one commenter (IV-D-05) recommended
changing the text in §63.1425(f) as follows:
"For each process vent in a PMPU that is, or is part
of, an affected source and that uses tetrahydrofuran (THF)
to produce one or more polyether polyol products...."
Response: The introduction paragraph to the process vent
control requirements section, §63.1425(a), states in the last
sentence: "The owners- or operator of an affected source where
polyether polyol products are produced using tetrahydrofuran
shall comply with paragraph (f) of this section." Therefore, the
EPA believes that the language requested by the commenter is
redundant and was not incorporated into the final rule.
2-194
-------
2.16.34 Section 63.1429(d) (3)
Comment: One commenter (IV-D-04) requested that
§63.1429(d)(3) be clarified. It requires the industry to specify
the "times" when an operating day begins and ends. Actually,
there are no "times;" there is only a "time." For example, if
the operating day runs from midnight to midnight, then "midnight"
is the time when one operating day begins and the previous
operating day ends. Under the proposed wording of this section,
the commenter believed that they would be subject to an
enforcement action, with penalties of up to $27,500, if they
specify only one time instead of two or more. The commenter
recommended that the word "times" be revised either to "time" or
to "time(s)."
Response: The EPA agrees with the commenter's request and
changed "times" to "time(s)" in the final rule.
2-16.35 Section 63 .1439 (e) (6) (v)
Comment: One commenter (IV-D-05), for clarity, suggested
that §63.1439(e)(6)(v) be changed to substitute "PMPU" for
"polyether polyol product" as follows: "...the results for each
change made to a primary product determination for a PMPU...."
Response: The EPA agrees that the primary product
determination is made on a PMPU basis. Therefore,
§63.1439(e)(6)(v) in the final rule was revised accordingly.
2.16.36 Section 63.1439(h)(1)
Comment: One commenter (IV-D-04} stated that unlike other
paragraphs of §63.1439(h)(1), which specify that certain
monitoring systems must alert the owner or operator "by alarm or
other means," paragraph (h)(1)(iv) says to alert the owner or
operator "by an alarm." The commenter sees no apparent reason
for this inconsistency and suggested that §63.1439(h)(1)(iv) be
revised as follows:
2-195
-------
"(Iv) The monitoring system will alert the owner or
operator by an alarm or other means, if the running average
parameter value ...."
Response: The EPA agrees with the commenter, and has
revised §63.1439(h)(1)(iv) accordingly.
2.16.37 Section 63.1427
Comment: One commenter (IV-D-04) requested that the EPA
correct the following minor typographical errors in §63.1427 to
avoid confusion. The comments are numbered, as follows:
(1) (c): "... by calculating the epoxide emissions, if any,
prior to the .onset of the ECO, if any, plus the epoxide emissions
at the onset of the ECO...."
(2) (h)(1)(ii): The epoxide partial pressure in the
reactor; aael or ....
(3) (h)(2)(ii): The reactor epoxide partial pressure at the
end of the ECO; aa4 or ....
(4) (i)(2): Continuouo Ongoing records....
(5) (k)(1)(ii): Within 180 days e£ after the production
of the new polyether polyol product, the owner or operator shall
submit a report updating the product list originally previously
submitted for the product class ....
(6) (1)(2): The owner or operator shall only update the
records specified in paragraphs (i)(1)(i)(A) through (G) of this
section for the product.
(7) (1)(3)(i)(A): The owner or operator shall update the
list of products for the product class required by paragraph
(i) (1) (ii)—of thio section that the product is leaving, and for
the product class that the product is entering, and shall record
(8) (1)(3)(i)(B): Within 180 days e£ after the change in
operating conditions for the polyether polyol product, the owner
or operator shall submit a report updating the product list
originally previously submitted for the product class ....
Another commenter (IV-G-02) supported the changes suggested
by Commenter IV-D-04, in particular the suggested changes to
§§63.1427 (h) (1) (ii) and (h)(2)(ii).
Response: The EPA agrees with all eight of the commenter's
suggested corrections; and they were incorporated into the final
rule.
2.16.38 Section 63.1439(e) (5) (ii) (D)
2-196
-------
Comment: Two commenters (IV-D-04, IV-D-05) provided
comments on §63.1439(e) (5) (ii) (D). One commenter (IV-D-04)
stated that paragraph (D) appears to have been borrowed by
mistake from another rule, perhaps the Group I Polymers and
Resins standard. It refers to monthly measurements of residual
organic HAP, a topic not germane to this rule, and it is ignored
in cross-references, so there is no way to reach it. The
commenter recommended that the paragraph be deleted from this
rule.
On the other hand, another commenter (IV-D-05) requested
that paragraph (D) be revised, for purposes of clarity, by adding
the words "where applicable" at the end of the provisions to read
as follows:
"The required information shall include a definition of the
affected source's operating month for the purposes of determining
monthly average values of residual organic HAP, where
applicable."
Response: Section 63.1439(e)(5)(ii)(D) was inadvertently
included in this rule. The EPA deleted §63.1439(e)(5)(ii)(D)
from the final rule.
2.16.39 Section 63.1439(b)
Comment: One commenter (IV-D-04) maintained that in
§63.1439(b), one paragraph should be moved and two cross-
references should be corrected as follows:
1. Paragraph 63.1439(b)(1)(i)(C) should be moved because it
has nothing to do with records of start-up, shutdown and
malfunction.
2. Section 63.1439(b)(1)(i) mentions a paragraph
(b)(1)(i)(D); there is no such paragraph. Nor should there be a
paragraph (b)(1)(i)(C) as noted above. Therefore, the reference
to (b) (1) (i) (D) should be changed to (b) (1) (i) (B) .
3. Section 63.1439(b)(1)(ii) also refers to paragraph
(b)(1)(i)(C) and, as noted above, does not belong here.
Therefore, once it is moved, the reference to (b)(1)(i)(C) should
be changed to (b) (1) (i) (B) .
2-197
-------
Response; The EPA appreciates the comments regarding the
cross-referencing errors. Paragraph 63.1439(b) (1) (i) (C) was
moved to §63.1439(d)(8), and the other cross-references were
corrected, as recommended.
2.16.40 Actual Date Versus "Date of Promulgation"
Comment: One commenter (IV-D-04) asserted that the EPA
should request the Office of the Federal Register to insert the
actual date or deadline in §63.1439(e)(3)(ii)(A) through (C)
instead of referring to the date of promulgation or the date of
publication. The commenter believed that this would avoid
confusion and could reduce the workload on owners or operators.
The commenter stated that some people might not know what
"promulgation" means. The commenter added that even for owners
and operators that do know what "promulgation" means, the language
in proposed §63.1439(e)(3)(ii)(A) through (C) creates extra work
and increases the chance for mistakes or misinterpretations.
The commenter provided revisions to paragraphs
§63.1439(e)(3)(ii)(A) through (C) which provide for the insertion
of the actual date, and the commenter requested that EPA revise
those paragraphs accordingly in the final rule.
Response: The EPA appreciates the comment, and the final
rule includes specific dates in those paragraphs.
2.16.41 Changes to Several Equations
Comment: One commenter (IV-D-04) requested that EPA make
minor changes to several equations in the proposed rule:
1. Throughout the rule's equations, wherever the term
"weight percent" is used, the weight fraction (or weight percent
expressed in decimal notation) should be used.
2. In Equation 9, there should be another "(" after the
minus sign and a ")" right after the second term in the
numerator.
3. Equation 13 should be: AEcontrol=(AEuncontrolled)[(100-
R)] -
2-198
-------
Response: Regarding the commenter's first suggestion,
Equation 11 is the only equation where a weight percent is used.
The EPA agrees that a weight fraction can be used, but does not
find it necessary to change weight percent to weight fraction.
The EPA disagrees with the second comment, which requests
that a set of parenthesis be added to Equation 9 to subdivide the
second term, since an extra set of parenthesis around the second
term is not needed. The EPA's reasoning is that, according to
algebraic rules, multiplication is conducted before addition or
subtraction, and addition or subtraction can be conducted in any
order.
Finally, the EPA agrees with the commenter regarding
Equation 13. Equation 13 has been revised as suggested by the
commenter in the final rule.
2.16.42 Section 63.1420(h)(3)
Comment: One commenter (IV-D-05) stated that the first
sentence in §63.1420(h)(3) refers to the wrong subparagraph. For
clarity, the commenter recommended revising it as follows:
"During start-ups, shutdowns, and malfunctions when the
requirements of this subpart do not apply pursuant to paragraphs
(h) (1) through (h)-(-3-M2) of this section, the owner or operator
shall implement, to the extent reasonably available ..."
Response: The EPA agrees with the commenter; however, as
was explained earlier in this document, proposed §63.1420(h)(3)
is §63.1420(h)(4) in the final rule, and this paragraph
(§63.1420 (h) ) in the final rule does refer to "paragraphs (h) (1)
through (3)".
2.16.43 Table 1
Comment: One commenter (IV-D-07) offered several corrections
to the subpart PPP cross-references (citations) with subpart A
(Part 63 General Provisions), as well as some editorial
corrections to Table 1 of the proposed rule.
Response: The EPA appreciates these comments, and revised
Table 1 accordingly.
2-199
-------
2.16.44 Table 1
Comment: One coimnenter (IV-D-04) requested that EPA make the
following corrections to Table 1 in the proposed rule:
1. In the entry for §63.1(a)(3), change the comment to say
"... which overlap with the requirements of subparts PPP and H
and specify how compliance shall be achieved." Also, if EPA
accepts this comment to add paragraphs dealing with overlaps for
wastewater and heat exchanger provisions, the comment should
begin as follows: "Section 63.1422(f) through (k) of this subpart
M
2. In the entry for §63.6(e)(3)(i), instead of saying
"combustion, recovery, or recapture devices," the comment should
say "control devices." This would be consistent with subpart U,
and "control devices" would automatically include all combustion,
recovery or recapture devices that are used for emission control.
3. In the entry for §63.7(a)(2), the phrase "compliance
demonstration test results" should be changed to "compliance
demonstration results," deleting the word "test." Subpart H does
not require performance tests.
4. The entry for §63.7(e)(l) currently says performance
tests must be conducted at "maximum representative" operating
conditions, which is inconsistent with the wording of the
proposed rule, which specifies "worst-case" operating conditions.
5. In the entry for §63.7(g), the comment should say
"equipment leaks" subject to §63.1434, rather than "emission
points" subject to §63.1434. This would be consistent with other
entries in the table.
6. In the entry for §63.7(h), the comment should refer to
§63.7(c)(2), not 63.7(c)(3).
Response: All six of the commenter's suggested corrections
were incorporated into the final rule.
2.16.45 Table 2
Comment: One commenter (IV-D-07) offered several corrections
to the subpart PPP cross-references (citations) to subparts F, G
2-200
-------
and H of the HON, as well as some editorial corrections to Table
2 of the proposed rule.
Response: The EPA appreciates these comments, and has
revised Table 2, as necessary.
2.16.46 Table 2
Comment: One commenter (IV-D-04) requested that EPA make the
following changes in Table 2 of the proposed rule:
1. At the end of the comment for subpart H §§63.160-63.193,
add the phrase, "with the differences noted in §63.1434."
2. In the "reference" column of the entry for subpart H,
instead of saying §§63.160-63.193. it should say §§63.160-63.182.
There are no sections after §63.182 in subpart H.
Response: The Agency appreciates the comments, and made the
revisions to the final rule.
2.16.47 Table 4
Comment: One commenter (IV-D-05) stated that the footnote in
Table 4 of the proposed rule should read "CAS No. = Chemical
Abstracts Service Registry Number."
Response: The EPA agrees and the final rule was corrected,
as suggested.
2.16.48 Table 5
Comment: One commenter (IV-D-04) requested that EPA make the
following changes in Table 5 of the proposed rule:
1. In the entry for "Thermal Incinerator," in the third
column, the first item should not say section "63.1429b;" the
letter "b" should be a superscript, referring to a footnote.
2. All entries should be subdivided into requirements for
"continuous" process vents, and requirements for "batch" process
vents. Otherwise, one could interpret that continuous process
vents are exempt from all requirements, which is not consistent
with the rule.
2-201
-------
3. In the entry for "Carbon Adsorber," column 3, item #3
currently says "or volumetric mass flow" but it should say "mass
or volumetric flow."
Response: The EPA agrees with all three comments, and has
made the suggested revisions. However, in lieu of "subdividing"
the requirements in Table 5 into those for process vents from
continuous unit operations and those for process vents from batch
unit operation, the final rule contains a separate table (Tables
5 and 6) for each set of requirements. Table 5 lists the
monitoring, recordkeeping, and reporting requirements for process
vents from batch unit operations, while the new Table 6 lists the
monitoring, recordkeeping, and reporting requirements for process
vents from continuous unit operations.
2.17 LEGAL CONSIDERATIONS
2.17.1 Executive Order 12866
Comment: One commenter (IV-D-04) disagreed with EPA's
determination that the proposed polyether polyols rule was not a
"significant regulatory action" for purposes of OMB review under
Executive Order 12866. The commenter stated that a regulatory
action is "significant" if it raises "novel legal or policy
issues arising out of legal mandates, the President's priorities,
or the principles set forth in Executive Order 12866." The
commenter claimed that the proposed rule raises novel legal or
policy issues, including at least the following:
1. Whether Facility M should be considered a "similar"
source, and thus be counted as the floor for new-source MACT,
despite having a very different manner of operation from other
sources that the EPA has considered (with resulting differences
in the amount and pattern of emissions and in the achievable
degree of emission reduction).
2. Whether a "new source" HAP reduction of 99.9 percent
should be required, when previous MACT standards have uniformly
made a policy decision to consider all combustion devices as no
more than 98 percent efficient.
2-202
-------
3. Whether to require monitoring (and associated
recordkeeping and reporting) during start-ups, shutdowns and
malfunctions, even though the "normal" emission control
requirements, which the monitoring is intended to track do not
apply during those periods.
4. Whether to require group determinations for
"combinations" of process vents, despite the fact that the
criteria for determining Group 1 or Group 2 status were tailored
to the characteristics of individual process vents.
5. Whether to incorporate future changes to the General
Provisions (and perhaps future changes to other standards on
which various portions of this rule are based) without further
rulemaking, in apparent violation of the Administrative
Procedures Act and its Clean Air Act counterpart.
6. Whether the rule may permissibly omit a lower
concentration cutoff (such as 20 ppmv) when compliance with a
percentage HAP reduction limit cannot be demonstrated or
achievable at some HAP feed concentrations.
7. Whether the rule may classify parameter monitoring
excursions as violations of "the emission limitation," rather
than as violations of an operating requirement, even though
parameter monitoring data are incapable of directly demonstrating
compliance or noncompliance with an emission limitation.
For these reasons, the commenter requested that EPA classify
this rule as a "significant regulatory action" for purposes of
OMB review.
Response: The EPA stands by its original determination that
the proposed polyether polyols rule was not a "significant
regulatory action" since it does not raise "novel legal or policy
issues." The actions raised by the commenter are technical in
nature, and do not introduce any novel legal or policy issues.
In the first issue, the commenter questioned whether
Facility M (the facility upon whose process vent control the MACT
level of control for new sources was originally based) is a
"similar" source to others in the source category. The commenter
2-203
-------
stated that Facility M has a "very different manner of operation
from other sources that the EPA has considered, with resulting
differences in the amount and pattern of emissions and in the
achievable degree of emission reduction." This comment was
addressed in more detail in Section 2.4.1.
The commenter's second comment addressed whether a new
source HAP reduction of 99.9 percent should be required, when
previous MACT standards "have uniformly made a policy decision to
consider all combustion devices as no more than 98 percent
efficient." This was not a policy decision, but a technical
issue, addressed in Section 2.4.1.
The third "novel legal or policy issue" brought up by the
commenter pertained to the EPA's decision to require monitoring
during start-ups, shutdowns and malfunctions. The EPA does not
believe that monitoring requirements are not legal or policy
issues, which was addressed in Sections 2.4.4 and 2.4.16.
The fourth issue that the commenter considered to be a
"legal or policy issue" discussed the appropriateness of basing
the group determinations on a combination of process vents,
despite the fact that the criteria for Group 1 or Group 2 status
in the proposed rule based on the characteristics of individual
process vents. The EPA also maintains that this is a technical,
and not a legal or policy issue, which was addressed in Section
2.4.4.
The commenter's fifth "novel legal or policy issue"
addressed whether to incorporate future changes to the General
Provisions into subpart PPP, without further rulemaking. This is
not a novel legal or policy decision. The commenter contended
that the automatic incorporation of future changes to the General
Provisions into the subpart PPP requirements would be a violation
of the Administrative Procedures Act (APA) and its Clean Air Act
(CAA) counterpart. The EPA disagrees with this statement because
the any changes made to the General Provisions would have a
public comment period during which parties subject to subpart PPP
would be able to comment.
2-204
-------
The conunenter' s sixth "legal or policy issue" was whether
the rule may permissibly omit a lower concentration cutoff (such
as 20 ppmv) when compliance with a percentage HAP reduction limit
cannot be demonstrated or achievable at some HAP feed
concentrations. The EPA revised the rule to include an
applicable concentration cutoff. Section 2.4.2 of this document
discusses this issue in more detail.
In issue number 7 the commenter questioned whether the rule
may classify parameter monitoring excursions as violations of
"the emission limitation," rather than classifying them as
violations of an operating requirement. Again, this is not a
legal or policy issue and is addressed in section 2.14.4.
2-205
-------
TECHNICAL REPORT DATA
(Please read Instructions on reverse before completing)
1. REPORT NO.
EPA-453/R-99-002b
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
National Emission Standards for Hazardous Air Pollutants
(NESHAP) for the Polyether Polyols Manufacturing Industry —
Summary of Public Comments and Responses
5. REPORT DATE
May 1999
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Emission Standards Division (Mail Drop 13)
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
1(X PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-D6-0010
12. SPONSORING AGENCY NAME AND ADDRESS
Director
Office of Air Quality Planning and Standards
Office of Air and Radiation
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
13 TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This document contains a summary of public comments received on the NESHAP for Polyether Polyols
Production (40 CFR 63, subpart PPP), which was proposed on September 4,1997 (62 FR 46804). This
document also provides the EPA's response to each comment, and outlines the changes made to the
regulation in response to public comments.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Held/Group
Air Pollution
Hazardous air pollutants
Emission reduction
Polyether Polyols
Hazardous air pollutants
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (Report)
Unclassified
21. NO. OF PAGES
238
20. SECURITY CLASS (Page)
Unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION IS OBSOLETE
------- |