United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle ParkNC 27711
EPA-450/3-83-005b
June 1990
Air
Distillation
Operations in
Synthetic Organic
Chemical
Manufacturing
Industry-
Background
Information for
Promulgated
Standards
             Final
             EIS

-------

-------
                        EPA-450/3-83-005b
Distillation Operations in

-------
                            EPA-450/3-83-005b
  Distillation Operations in
Synthetic Organic Chemical
 Manufacturing Industry  —
   Background Information
 for Promulgated Standards
           Emissions Standards Division
              U S F-virrr--^rrtal Protection Agency
              F^'.-vf^, I-b.cii-y (5?L-16)
              27."' b- Dearborn btreat, Room 167.0
              Chicago, IL  60604
        U.S. ENVIRONMENTAL PROTECTION AGENCY
            Office of Air and Radiation
         Office of Air Quality Planning and Standards
        Research Triangle Park, North Carolina 27711
              June 1990

-------

-------
                       ENVIRONMENTAL PROTECTION AGENCY

                            Background  Information
                   and Final  Environmental  Impact Statement
                 for Volatile Organic Compound Emissions from
                          Distillation Processes in
                   Synthetic Organic Chemical  Manufacturing
                                 Prepared by:
Director, Emission Standards Division
U. S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711

1     The promulgated standards of performance will limit emissions of
      volatile organic compounds from new, modified, and reconstructed
      distillation processes.  Section 111 of the Clean Air Act
      (42 U  S. C. 7411), as amended, directs the Administrator to establish
      standards of performance for any category of new stationary source of
      air pollution that ".  . . causes or contributes significantly to air
      pollution which may reasonably be anticipated to endanger public health
      or welfare."

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;  State  and Territorial Air Pollution Program  Administrators;
      EPA Regional Administrators; Local  Air Pollution Control Officials;
      Office of Management  and Budget; and  other  interested  parties.

 3.    For additional  information  contact:

      Mr. Doug Bell
      Standards Development Branch  (MD-13)
      U. S.  Environmental  Protection Agency
      Research Triangle Park,  N.  C.  27711
      Telephone:   (919)  541-5568

 4.    Copies of this  document may be obtained  from:

      U. S.  EPA Library (MD-35)
       Research Triangle Park, N.  C.   27711
      Telephone:   (919) 541-2777

       National Technical Information Service
       5285 Port Royal Road
       Springfield, VA  22161

-------

-------
This report has been reviewed by the Emission Standards Division of the Office
of Air Quality Planning and Standards, EPA, and approved for publication.
Mention of trade names or commercial products is not intended to constitute
endorsement or recommendation for use.  Copies of this report are available
through the Library Services Office (MD-35), U.S. Environmental Protection
Agency, Research Triangle Park, N.C. 27711, or from National Technical
Information Services, 5285 Port Royal Road, Springfield, Virginia 22161.
                                      ii

-------

-------
                              TABLE OF CONTENTS
                                                                      PAGE

TITLE PAGE	   1
DISCLAIMER  	  n
LIST OF TABLES	•'	1X
1.0 SUMMARY	1"1
     1.1  SUMMARY OF CHANGES SINCE PROPOSAL 	 I'1
          1.1.1     Affected Facility Designation  	  .1-2
          1.1.2     Applicability of the Standards	 1-2
          1.1.3     Exemptions	 1-3
          1.1.4     Negligibly  Photochemically Reactive
                    Compounds	1"3
          1.1.5     Monitoring, Testing, and Reporting/RecordKeeping
                    Requirements	-1"4
          1.1.6     Flare  Operating Specifications	1-5
          1.1.7     TRE Coefficients	!"5
          1.1.8      Incorporation  of  a  TRE  Index  Value Above
                    Which  Monitoring  and Recordkeeping Are
                    Not Required	l~6
           1.1.9      Definitions .......  	  I"6
           1.1.10    Net Heating Value Equation	1-7
           1.1.11    Miscellaneous  	  I"7
      1.2  SUMMARY OF  IMPACTS OF PROMULGATED ACTION	1-7
           1.2.1      Alternatives to Promulgated Action	1-7
           1.2.2      Environmental  Impacts  of Promulgation Action. .  .  1-8
           1 2 3      Energy and Economic Impacts of Promulgated
                     Action	1'8
           1.2.4     Other Considerations	I'8
                                       m

-------
                        TABLE OF CONTENTS (CONTINUED)
                                                                      PAGE
2.0  SUMMARY OF PUBLIC COMMENTS	2-1
     2.1  APPLICABILITY OF THE STANDARDS	2-1
          2.1.1     Exemption Request for Specific Chemicals and
                    Processes	2-1
          2.1.2     Exemption Requests for Low Level  Production of
                    Listed Chemicals	2-11
          2.1.3     Applicability to Petroleum Refineries 	 2-14
          2.1.4     Date of Initial  Performance Testing	2-16
          2.1.5     Low Vent Stream Flowrate and Design Capacity
                    Exemption	.2-17
          2.1.6     Applicability Date of Standards	2-19
          2.1.7     Batch Distillation Exemption  	 .2-20
          2.1.8     Request for Removal  of Chemicals  of Negligible
                    Photochemical  Reactivity From Regulation	2-20

     2.2  SELECTION OF AFFECTED FACILITY	2-21
          2.2.1     Designation of Affected Facility.  ...  	 2-21
     2.3  DEFINITIONS	2-26
          2.3.1     Definition of Vent Stream	2-26
          2.3.2     Definition of Distillation Operations 	 2-26
          2.3.3     Definition of Process  Heaters 	 2-27
          2.3.4     Definition of TOC	2-27
          2.3.5     Definition of Corrosive Vent Stream	2-28
     2.4  SELECTION OF BDT	2-28
          2.4.1     NOX Emissions  from BDT and BACT Review	2-28
          2.4.2     Flare Specifications	 . 2-29
          2.4.3     Catalytic Versus Thermal  Incineration 	 2-33
          2.4.4     Recovery Devices and Emissions  Reduction  .... 2-35

-------
                   TABLE OF CONTENTS (CONTINUED)
                                                                 PAGE

2.5  COST ESTIMATION	2"37
     2.5.1     Control  Cost for Flares and Incinerators  .... 2-37
     2.5.2     Control  Cost for Flare Systems	2-39
     2.5.3     Cost of Monitoring Incinerator Temperatures  . .  . 2-44
     2.5.4     Brine Disposal Costs   	 2~45
     2.5.5     Base Year Used in Costing Procedure	2-45
     2.5.6     Safety Costs for Boilers and Process Heaters  .  . 2-46
     2.5.7     Pipeline System Costs  	 2'47
     2.5.8     Economic Impact Analysis   .  	 2-48
     2.5.9     Costs for Low Heating  Value  Streams	  . 2-50
2.6  COST EFFECTIVENESS	•	 2-51
     2.6.1     TRE Approach  and Cutoff  Value	2-51
     2.6.2     Basis of TRE  Cutoff	•  •  • 2'57
     2.6.3     Derivation  of TRE  Equation  	 2'57
     2.6.4     Removal  Efficiency for Combined Vent  Streams  .  . 2-59
     2.6.5     Calculation of  TRE for Large Facilities 	 2-60
 2.7  FORMAT OF STANDARDS	 2-61
     2.7.1      Compounds  Included in TOC	 2-61
      2.7.2      Identification  of Chemicals of Low
                Photochemical Reactivity  .	*-*>*
      2.7.3      Water Vapor Included in  Low Flowrate Exemption.  .  2-64
 2.8  MODIFICATION/RECONSTRUCTION 	  2'64
      2.8.1      Effective Date of Modification and
                Reconstruction  	  2"54
      2.8.2     Clarification of Modification Definition	2-65
      2.8.3     Feasibility of Modifications to Meet Standards. .  2-66
      284     Clarification of Modification and
                Reconstruction Provisions  	 2-bb
      2.8.5     Replacement of Trays  and  Packing	2-68

-------
                   TABLE OF CONTENTS (CONTINUED)
                                                                 PAGE
2.9  MONITORING AND MEASUREMENT METHODS	2-68
     2.9.1     Cost and Complexity of Monitoring Requirements.  . 2-68
     2.9.2     Monitoring During Startup, Shutdown, and
               Malfunction	2-70
     2.9.3     Flow Meter and Flow Indicator Requirements. .  .  . 2-72
     2.9.4     Continuous Temperature Monitoring of Boiler .  .  . 2-74
     2.9.5     Waiver Request of Performance Tests;	2-74
     2.9.6     Distillation Unit Definition Effect; on
               Monitoring	  . 2-76
     2.9.7     Method for Sampling Small  Diameter Vents	2-76
     2.9.8     Use of Reference Method 2D	 2-77
     2.9.9     Measurement of Vent Stream Flowrate	2-77
     2.9.10    Flow Monitors for Incinerators	2-78
     2.9.11    Accuracy of Temperature and Flowrate Monitors  .  . 2-78
     2.9.12    Continuous Recording	2-79
     2.9.13    Continuous Monitoring Requirement 	 2-80
     2.9.14    Alternatives to Method 18  	 2-81
     2.9.15    Request for VOC Monitors on Incinerators	2-82
     2.9.16    Monitoring of Flare Performance ...  	 2-83
     2.9.17    Alternative Methods of Demonstrating Compliance  . 2-84
     2.9.18    Temperature Monitoring for Condensers  	 2-84
     2.9.19    Temperature Monitoring for Catalytic
               Incineration	„  . 2-85
     2.9.20    Product Recovery Systems and TRE Calculation.  .  . 2-86
     2.9.21    Scrubbers and Recovery Devices	2-87
     2.9.22    TRE Index Calculation Point .....  	 2-89
                                vi

-------
                   TABLE OF CONTENTS (CONTINUED)
                                                                 PAGE
2.10 REPORTING AND RECORDKEEPING 	 2-90
     2.10.1    Semiannual Reporting	2-90
     2.10.2    Low Flow and Low Capacity Facilities	2-91
     2.10.3    Length of Time Records Must be Kept ........ 2-92
     2.10.4    Recordkeeping for Changes in Production Rate.  .  . 2-92
     2.10.5    Reporting Variations in Steam Flowrate	2-92
     2.10.6    Reporting Condenser Temperature Increases .... 2-93
2.11 GENERAL	2-94
     2.11.1    Documentation of EPA Correspondence 	 2-94
     2.11.2    Safety of Combining Vent Streams	2-94
     2.11.3    CAS Numbers for Affected Chemicals	2-95
     2.11.4    Surface Condensers	2-95
     2.11.5    Significance of Energy Impacts	2-96
     2.11.6    Typographical Error in Regulation 	 2-96
     2.11.7    Typographical Error in Regulation 	 2-96
     2.11.8    Typographical Error in Regulation 	 2-97
     2.11.9    Typographical Error in Regulation 	 2-97
     2.11.10   Reordering of Sections in Regulation  ...... 2-97
     2.11.11   Position of "Note" in Regulation	2-97
APPENDIX A:    Federal Register Notices of Organic Compounds
               Determined to have Negligible Photochemical
               Reactivity.	A-l

            '   INTRODUCTION	A-l
               42 FR 35314	A-2
               42 FR 32042	A-5
               42 FR 32424	A-7
               45 FR 48941	....".. A-8
                               vn

-------
                   TABLE OF CONTENTS (CONCLUDED)

                                                                 PAGE

APPENDIX B:    TRE Equation and Coefficient Development for
               Thermal Incinerators and Flares  ... 	 B-1

     B.I       INTRODUCTION	B-1

     B.2       INCINERATOR TRE INDEX EQUATION	B-1
          B.2.1     Incinerator TRE Index Equation
                    Development	B-1

          B.2.2     Example Calculation of an Incinerator-
                    based TRE Index Value for a Facility .... B-4

     B.3       FLARE SYSTEM TRE DEVELOPMENT	 B-7
          B.3.1     Development of the Flare TRE
            v       Index Equation	.B-7

          B.3.2     Flare TRE Coefficients Verification	B-9
          B.3.3     Example Calculation of a Flare-based
                    TRE Index Value for a Facility	B-15
                               viii

-------
                               LIST OF TABLES

TITLE                                                                  PAGE

2-1  List of Commenters on the Proposed Standards of Performance
     for Distillation Operations 	   2-2

B-l  Distillation NSPS TRE Coefficients for Vent Streams
     Controlled by an Incinerator	   B-3

B-2  Maximum Vent Stream Flowrate and Net Heating Value
     Characteristics for Each Design Category	   B-5

B-3  Distillation Operations NSPS TRE Coefficients for Vent
     Streams Controlled by a Flare 	   B-10

B-4  TRE Index Values Generated Using TRE Coefficients and
     the Flare Cost Algorithm Net Heating Value Greater
     Than or Equal to 300 Btu/scf	   B-ll

B-5  Percent Difference Between TRE Index Values Generated
     Using TRE Equation and the Flare Cost Algorithm Net
     Heating Value Less Than 300 Btu/scf	   B-13

B-6  Percent Differences Between TRE Index Values Generated
     by the Cost Algorithm and the TRE Equation for Vent
     Streams with Heating Values Less Than 40 Btu/scf. .......   B-14

-------

-------
                                 1.0 SUMMARY

     On December 30, 1983, the Environmental Protection Agency  (EPA) proposed
standards of performance for distillation operations in the synthetic organic
chemical manufacturing industry  (48 FR 57538) under the authority of
Section 111 of the Clean Air Act.  Public comments were requested on the
proposal in the Federal Register.  There were 34 commenters, most of whom are
industry representatives.  Comments were also received from a vendor of
equipment used to control emissions from distillation operations and from a
representative of an environmental group.  On May 16, 1985, EPA reopened the
public comment period (50 FR 20446) for the purpose of allowing public
comment on the results of the Agency's reanalysis of the total resource
effectiveness (TRE) equation and coefficients, the costing procedures, and
the designation of affected facility.  The reanalysis resulted from the
acquisition of new information received in public comments and collected
since proposal.  In response to the reopening of the public comment period,
one comment was received.  The comments that were submitted, along with
responses to these comments, are summarized in this document.  The comments
and subsequent responses serve as the basis for the revisions made to the
regulation between proposal and promulgation.
1.1  SUMMARY OF CHANGES SINCE PROPOSAL
     Several changes and clarifications were made in the regulation as a
result of the review of public comments.  Changes and clarifications were
made in the following general areas:   (a) affected facility designation;
(b) applicability of the standards; (c) exemptions; (d) treatment of
negligibly photochemically reactive compounds; (e) monitoring requirements;
(f) flare operating specifications; (g) TRE coefficients; (h) incorporation
of a TRE index value above which monitoring and recordkeeping are not
required;  (i)  general definitions;  (j)  net heating value equation; and (k)
miscellaneous.
                                     1-1

-------
 1.1.1  Affected Facility Designation
      The EPA has decided to change the designation  of affected facility used
 in the standards.   The designation of affected  facility in  Section 60.660(a)
 has been changed from a single distillation  unit  with the associated recovery
 system to each recovery system and all  associated distillation units venting
 to that recovery system.   Each distillation  unit  not  feeding  offgas into a
 recovery system would constitute  a separate  affected  facility.   Because this
 designation  does not  require apportioning  of vent streams.  Sections 60.664(c)
 (l)(ii)  and  (iii),  which  specify  the method  of  apportioning where  distillation
 columns share  a common recovery system, have been deleted from the regulation.
 1.1.2  Applicability  of the Standards
      The natural  products   -pinene;  coconut  oil acids,  sodium  salt;  fatty
 acids;  tall  oil,  sodium salt;  tallow acids,  potassium salt; and  tallow  acids,
 sodium  salt; have been deleted  from  the list  of affected chemicals in
 Section  60.667.  The  manufacture  of  products  derived  from natural  sources
 such  as  these  is not  within  the scope of this new source performance standard
 (NSPS),  which  is  intended to  focus on synthetic organic chemical manufacturing
 processes.   Three fertilizer  chemicals, ammonium  carbamate,  urea ammonium
 nitrate  (UAN)  and urea have  also  been deleted from the list in Section  60.667.
 According to information available to the Agency,  the production of ammonium
 carbamate and  UAN does  not  involve the use of distillation operations and  no
 significant  volatile  organic  compounds (VOC)  are  emitted during the production
 of urea.
      Some commenters were concerned that trace amounts of listed chemicals in
 a product stream would  cause them to be subject to the standards.  This NSPS
would only be  applicable to distillation operations within process units
 producing as a product  any of the chemicals listed in Section 60.667.  A
definition of  "product" has been added to the regulation to  clarify that the
chemical is produced as a product when it is produced for sale or  use in
another process unit,  and that by-products, coproducts, arid  intermediates are
considered "products"  for the purposes of these standards.   This new defini-
tion clarifies what constitutes production of one  of the 211 high production
volume chemicals listed in Section 60.667.
                                     1-2

-------
1.1.3  Exemptions
     Batch distillation operations have been exempted from the standards.  The
control technologies considered in the development of the distillation NSPS
are appropriate for continuous processes but may not always be applicable to
batch processes because batch processes typically have intermittent vent
streams.  Therefore, EPA has decided to exempt batch processes explicitly in
Section 60.660(c).  In addition, to clarify what is meant by a "batch
distillation operation" a definition of this term has been added to Section
60.661.  This definition emphasizes the noncontinuous operation and the
discrete liquid feed nature of batch distillation operations.
     The low flow rate exemption has been changed from a design flow rate
basis to an operating flowrate basis in Section 60.660(c).  The EPA recog-
nizes that distillation facilities may be designed at the exemption level but
may operate at levels above or below.this design flow rate.  The record-
keeping (Section 60.665(i)) and the semiannual reporting (Section
60.665(1)(5)) requirements for the low flow rate exemption have been likewise
amended to reflect this change.
1.1.4  Negligibly Photochemicallv Reactive Compounds
     The distillation standards are intended to control VOC, i.e., compounds
that participate in atmospheric photochemical reactions to produce ozone.
Therefore, negligibly photochemically reactive compounds are permitted to be
subtracted from the total organic compound  (TOC) emissions measured for
calculation of the TRE index value.  However, when determining combustion
device emission reduction efficiency they should not be subtracted because
combustion devices are not compound-specific with respect to VOC destruction.
Furthermore, it is more costly and complex  to subtract the negligibly reactive
compounds during performance testing.
     In order to effect this change, Section 60.661 was amended to indicate
that the definition of "TOC" means TOC less all compounds that have been
determined by the Administrator to be negligibly photochemically reactive.
This definition is used only when applied to Sections 60.664(d)(2)(i) -
measuring molar composition; 60.664(d)(5),  and 60.664(e) - measuring hourly
                                      1-3

-------
emissions rate; and 60.665(b)(4) and 60.665(g)(4) - measuring TOC
concentration.  For all other quantifications of TOC under these standards,
VOC equals TOC less methane and ethane as indicated in Section 60.662(a).  The
Federal Register citations for the list of negligibly photochemically reactive
compounds that may be subtracted are presented in Appendix A and have also
been added to the definition of TOC in Section 60.661 of the regulation.
     The definition of "TRE index value" has been amended to incorporate the
change in the definition of "TOC."  The TRE is now defined as "a measure of
the supplemental total resource requirement per unit reduction of total
organic compounds . . ., emission rate of total organic compounds. . .  ."

1.1.5  Monitoring. Testing, and Reporting/Recordkeeping Requirements
     The required use of flow rate monitors for vent streams routed to thermal
incinerators has been changed.  The EPA has decided to require the use of flow
indicators rather than flow rate monitors for incinerators because the reason
for monitoring is to provide an indication that the vent: stream is being
routed for destruction and flow indicators adequately serve1 this function.
This requirement is consistent with vent stream monitoring requirements for
all other combustion devices that may be used under the standards.  Thus,
Section 60.663(A)(2) has been amended to indicate that at flow indicator is to
be used.  Likewise, the reporting and recordkeeping requirements for flow
monitoring in Section 60.665(c)(3) have been deleted and replaced with
reporting and recordkeeping requirements for flow indication that are
integrated into Section 60.665(d).
     Ultraviolet beam heat sensors have been included in addition to
thermocouples as suitable devices for the monitoring of the presence of a
flame.  Section 60.663(b)(l) has been amended to indicate this.
     The "continuous recording" requirements have been changed.  All
measurements such as firebox temperature, absorber liquid specific gravity,
carbon adsorber steam mass flow rate, and others are now required to be taken
at least every 15 minutes.  This allows use of computer-assisted monitors.
                                      1-4

-------
The frequency for measurement of these parameters during compliance testing
has also been revised to be made at least every 15 minutes.  Section 60.665(b)
has been amended accordingly.  An advantage of this change is that the
parameter measurement frequencies required for monitoring and continuous
recording are consistent with the frequencies required in compliance testing.
Furthermore, the same equipment may be used for both monitoring and compliance
testing.
1.1.6  Flare Operating Specifications
     Operating specifications for flares used to comply with requirements in
new source performance standards (NSPS) were recently added to Section 60.18
of the General Provisions (51 FR 2701, January 21, 1986).  The regulation has
been revised to refer all owners or operators of affected facilities which
use flares to comply with this NSPS to the requirements in that section.
1.1.7  TRE Coefficients
     Tables 1 and 2 of the regulation present the incinerator and flare
coefficients associated with the TRE index equation.  Some of the coefficients
in these tables were corrected to predict more accurately the TRE indexes
(and associated cost-effectiveness values) of facilities.  The modifications
of the coefficients resulted from changes in the costing procedures-on which
these coefficients are based.  The changes in costing procedures and TRE
coefficients are discussed in the Agency's notice reopening the public
comment period for the proposed distillation standards (50 FR 20446).
     Several modifications were also made in the format of Table 1 to provide
clarity to owners or operators of distillation facilities.  These modifica-
tions included: (a) the designation of Category Al and A2 streams was changed
from "chlorinated" to "halogenated"; (b) the designation for flow rate was
changed from "W" to "Q " so that the symbol would match the symbol in the EPA
Reference Methods discussion; (c) the term representing flow rate intervals
for selecting TRE coefficients was changed from "design standard flow rate" to
"vent stream flow rate" to indicate that actual operating flow rate should be
used in selecting TRE coefficients; (d) the first flow rate interval was
eliminated because all vent streams with flow rates below the'minimum
incinerator size of 500 scfm are assumed to have a flow rate of 500 scfm for
                                     1-5

-------
 the  purposes  of calculating capital and annual operating costs; and  (e)  the
 term representing  flow  rate for Category E streams was changed from  "design
 standard  flow rate"  to  "dilution flow rate =  (QsMHrVS.S"  to  indicate that
 dilution  flow rate should be used in selection of TRE coefficients.   Table 1
 presents  the  coefficients for the incinerator TRE equation.  Table 2  presents
 the  coefficients for the flare TRE index equation.
 1.1.8
Incorporation of a TRE Index Value Above Which Monitoring and
Recordkeemna Are Not Reauired
     Several changes were made in the regulation to provide for inclusion of a
maximum TRE  index value.  The maximum TRE index value of 8.0 represents the
value above  which monitoring and recordkeeping requirements would not be
imposed on a facility attempting to comply with the standards.  It is the
judgment of  the Agency that facilities with TRE index values above the maximum
would most likely not be able to make process changes that would cause the TRE
index value  to fall below the cutoff.  Thus, the Agency believes that the
monitoring and recordkeeping burden should not be imposed on such facilities.
However, if  a process change occurs, the facility should recalculate the TRE
index value  as required in Section 60.664(c) to determine whether the value
remains above the TRE maximum.  Sections 60.660 and 60.664 of the regulation
have been amended to incorporate the requirements associated with the maximum
TRE index value.

1.1.9  Definitions
     Both new and revised definitions have been included in Section 60.661.
New definitions for "batch distillation and operation," "continuous recorder,"
and "product" have already been discussed.   Revised definitions for
"distillation operation," "distillation unit," "total  organic compounds,"
"process heater," "process unit," "recovery system," "TRE index value," and
"vent stream" have also been included.
     To further clarify the applicability of the distillation NSPS,  the
definition of "distillation operations" has been amended to read,
"'distillation operation'  means an operation separating one or more feed-
stream(s)  into two or more exit streams.  .  ." instead  of two or more product
streams,  as stated in the proposed regulation.  The definition of "distilla-
                                     1-6

-------
tion unit" was broadened to include explicitly the accessories necessary for
some distillation operations, a vacuum pump and a steam jet.  Because process
heater tubes may be arranged in a number of configurations other than "tubular
coils," this definition was revised to refer simply to "tubes."  In order to
clarify what properly constitutes a "recovery system," this definition was
made more specific by referring to its purpose for recovering chemicals for
use in making other chemicals, reuse, or for sale."  "Vent stream" was
revised to exclude explicitly equipment leaks and relief valve discharges,
which are covered under VOC fugitive emission standards.  The reasons for
revising the definitions of "TOC" and "TRE index value" are explained in
Section 1.1.4 of this chapter.
1.1.10  Net Heating Value Equation
     The net heating value equation, Section 60.664(c)(4), was changed to
include the heating value associated with carbon monoxide since it can
contribute to the net vent stream heat content if present.  The concentration
of carbon monoxide in the vent stream must be determined according to
ASTM D1946-82 as required under Section 60.664 (c)(2)(ii).
     To be sure that the net heating value is calculated on a wet basis, the
definition of symbol "C." (Section 60.664(c)(4)) was amended to include "on a
wet basis."  The net heating value must be calculated on a wet basis because
the entire vent stream, including water vapor, would be combusted, and
therefore this is the heating value used in calculating a TRE value.
1.1.11  Miscellaneous
     Word and symbol changes were made in Sections 60.664(c) and 60.665(k)(2)
to correct typographical errors.
1.2  SUMMARY OF IMPACTS OF PROMULGATED ACTION
1.2.1  Alternatives to Promulgated Action
     The regulatory alternatives are discussed in Chapter 6 of the proposal
background information document (BID).  These regulatory alternatives reflect
the different estimated number of facilities required to reduce VOC emissions
by 98 weight-percent or to 20 parts per million by volume under a particular
                                     1-7

-------
 cost-effectiveness cutoff.   These regulatory alternatives were used for the
 selection of the best demonstrated technology (BDT),  considering the estimated
 cost impacts,  nonair quality health impacts,  environmental  impacts, and
 economic impacts associated with  each  alternative.  These alternatives have
 not been changed.
 1.2.2  Environmental  Impacts of Promulgated  Action
      The changes in  the  regulation described  above will  have  a minor effect
 on  the  estimated air quality impacts attributed to the  standards as originally
 proposed.   The  new estimated air  quality  impacts of the  standard are presented
 in  the  Agency's notice reopening  the public  comment period  for the  proposed
 distillation standards (50  FR 20446).  The changes in the regulation will
 have  a  negligible  impact on the water  quality and solid  waste  impacts
 attributed  to the  final  standards.  The analysis of environmental impacts  in
 Chapter 7 along with  the new air  quality  impacts presented  at  50 FR 20446  now
 constitute  the  final  Environmental  Impact Statement for  the promulgated
 standards.
 1.2.3   Energy and  Economic  Impacts of  Promulgated Action
      Section 7.4 of the proposal  BID describes the energy impacts and
 Chapter  9 describes the economic  impacts of the proposed standards.   The
 changes  in  the  regulation described above will have a negligible effect on
 these impacts.
 1.2.4  Other Considerations
      1.2.4.1  Irreversible  and Irretrievable Commitment of Resources.
 Chapter  7 of the proposal BID concludes that other than fuels required for
 the operation of control  equipment, there is no apparent irreversible or
 irretrievable commitment of resources associated with the standards.  The use
 of the TRE concept encourages the  use of  recovery techniques to recover
 chemicals for use, reuse or sale that would otherwise be disposed of  as
 pollutants.   The control  of VOC emissions using recovery techniques might be
 an alternative for some distillation facilities.   This would result  in the
conservation of both chemicals and fuels.   The changes in the regulation
described above will  have no impact on  the commitment of resources.
                                     1-8

-------
     1.2.4.2  Environmental and Energy Impacts of Delayed Standards.
Table 1-1 in the proposal BID summarizes the estimated environmental and
energy impacts associated with promulgation of the standards.   If the
standards were delayed, adverse impacts on air quality could result.  A delay
in promulgation would mean that affected facilities would be controlled to
the level specified in the appropriate State implementation plan.  Emission
levels would be higher than would be the case were the standards in effect
because these estimated impacts have not changed since proposal.
                                     1-9

-------

-------
                        2.0  SUMMARY OF  PUBLIC COMMENTS
     A total  of 34  letters  commenting  on  the proposed  standards  were
received.  A  public hearing on the proposed standards  was not  requested.  The
public comment period was reopened to  allow comment on the Agency's reanalysis
of the total  resource effectiveness  (TRE) equations and coefficients,  the
costing procedures,  and  the designation of affected facility.  One letter
commenting on the reanalysis was received.  The 35 comment letters have been
recorded and  placed in  the  docket.   The list of commenters, their affiliation,
and the Environmental Protection Agency (EPA) docket number for  each of the
comments are  shown  in Table 2-1.  The  docket reference is indicated in
parentheses in each comment.  Unless otherwise noted,  all docket references
are part of Docket  Number A-80-25, Category IV.  The comments  have been
organized into the  following 11 categories:

     2.1   Applicability of the Standards
     2.2   Selection of Affected Facility
     2.3   Definitions
     2.4   Selection of Best Demonstrated Technology
     2.5   Cost Estimation
     2.6   Cost Effectiveness
     2.7   Format of the Standards
     2.8   Modification and  Reconstruction
     2.9   Monitoring and Measurement Methods
     2.10  Reporting and Recordkeeping
     2.11  General

2.1  APPLICABILITY  OF THE STANDARDS
     2.1.1  COMMENT:  Several commenters (D-l,  D-2, D-3,  D-16, D-19, D-29,
D-31,  and D-34)  requested information on the possible  exemption  of various
organic chemicals and manufacturing processes from the proposed  standards.
One commenter (D-19) inquired if a distillation facility used to produce a
chemical  for use in the polymer manufacturing industry is exempt from the
standards when it is isolated from the polymers and resins process.   The
                                     2-1

-------
                                   TABLE 2-1

 List of Commenters on the Proposed Standards of Performance for Distillation
 Operations in the Synthetic Organic Chemical Manufacturing Industry

                           Docket Number A-80-25, IV

                                    Letters

      Addressee                                    Docket Reference

 Mr.  J.  F.  Cochrane                                     0-1
 Director,  Environmental  Affairs
  Department
 J.  R.  Simplot Company
 Post Office Box 912
 Pocatello,  Idaho  83201

 Mr.  D.  H.  Maybury                                      0-2
 Regional  Environmental Manager
 Regional Operations Services
 Hercules  Incorporated
 501  Glouchester Street
 Brunswick,  Georgia  31520

 Mr.  Finn Bohn                                          0-3
 Environmental  Engineer
 Allied  Chemical
 Post Office Box 1053R
 Morristown, New Jersey   07960

 Mr.  Edward  P.  Crockett                                  D-4
 American Petroleum Institute
 1220 L  Street,  Northwest
 Washington, D.C.   20005

 Dr. Thomas  A.  Robinson                                  D-5
 Director, Environmental Affairs
 Vulcan  Chemicals
 Post Office Box  7689
 Birmingham, Alabama  35253

Mr. Gary D. Myers                                       0-6
 President, The Fertilizer Institute
 1015 18th Street, Northwest
Washington, D.C.  20036
                                     2-2

-------
                             TABLE  2-1  (Continued)
     Addressee                                     Docket Reference

Mr. J. J. Moon                                          D-7
Manager,  Environment  &  Consumer  Protection
Corporate Engineering
Phillips  Petroleum Company
Bartlesville, Oklahoma   74004

Mr. David M.  Flannery                                  D-8
Counsel for Borg-Warner Chemicals,
  Incorporated
Robinson & McElwee Law  Offices
Post Office Box  1791
Charleston, West Virginia   25326

Mr. William J. Hague                                    D-9
Principal Process Engineer
Allied Chemical
Post Office Box  1139R
Morristown, New Jersey   07960

Mr. Robert M. Yarrington                                ti-10
Product Manager
Englehard Industries  Division
2655 US Route 22
Unfon, New Jersey  07083

Dr. Robert R. Romano                                    D-ll
Manager, Air Programs
Chemical Manufacturers  Association
2501 M Street, Northwest
Washington, D.C.  20037

Mr. G. L. Je.ssee                                        D-12
Director, Regulatory Management
Monsanto Company
800 North Lindbergh Boulevard
St. Louis, Missouri  63167

Mr. J. D. Reed                                          D-13
General Manager
Environmental  Affairs and Safety
Standard Oil  Company  (Indiana)
200 East Randolph Drive
Chicago,  Illinois  60601
                                     2-3

-------
                             TABLE 2-1 (Continued)


      Addressee                                    Docket; Reference

 Mr.  Peter W. McCallum                                  D-14
 Senior Corporate Environmental  Specialist
 The  Standard Oil Company
 Midland Building
 Cleveland,  Ohio  44115

 Mr.  J.  W.  Torrance                                     D-15
 Supervisor,  Environmental  Engineering
 Allied  Fibers and Plastics
 Technical  Center
 Post Office  Box 31
 Petersburg,  Virginia   23804

 Mr.  Gary D.  Myers                                      D-16
 President, The Fertilizer  Institute
 1015 18th  Street,  Northwest
 Washington,  D.C.   20036

 Mr.  Paul A.  Cammer                                     D-17
 Executive  Director
 Halogenated  Solvent Industry Alliance
 1612 K  Street,  Northwest
 Suite 300
 Washington,  D.C.   20006

 Mr.  U.  V. Henderson,  Jr.                                D-18
 Associate Director, Environmental Affairs
 Research, Engineering  and  Safety Department
 Texaco,  Incorporated
 Post Office  Box  509
 Beacon,  New  York   12508

 Mr.  Ronald F.  Black                                     D-19
 Environmental  Specialist
 Engineering  Division
 Rohm and Haas  Company
 Post  Office  Box 584
 Bristol, Pennsylvania  19007

 Dr.  V. J. Marchesani                                    D-20
 Director, Energy and Environmental Quality
 Health and Environmental  Affairs
 Post Office  Box 751
 ICI Americas,  Incorporated
Wilmington,  Delaware  19899
                                     2-4

-------
                             TABLE 2-1  (Continued)
      Addressee                                    Docket Reference

 Mr.  David  D.  Doniger                                   D-21
 Senior  Staff  Attorney
 Natural  Resources  Defense Council,
  Incorporated
 1725 I  Street,  Northwest
 Suite 600
 Washington, D.C.   20006

 Mr.  Ronald A.  Lang                                     D-22
 Executive  Director
 Synthetic  Organic  Chemical  Manufacturers
  Association,  Incorporated
 1612 K  Street,  Northwest
 Suite 300
 Washington, D.C.   20006

 Mr.  A.  G.  Smith                                         D-23
 Environmental  Affairs
 Shell Oil  Company
 One  Shell  Plaza
 Post  Office Box 4320
 Houston, Texas  77210

 Mr.  Barry  Christensen                                   D-24
 Environmental  Manager
 Diamond  Shamrock Chemicals  Company
 1149  Ellsworth Drive
 Pasadena,  Texas  77501

 Mr.  E. J.  Shields                             •          D-25
 Director,  Environmental Services
 Allied Chemical
 Post Office Box 1139R
 Morristown, New Jersey  07960

 Mr. Lawrence B. Gotlieb                                 D-26
Assistant General  Counsel
Distilled Spirits Council of the              *
 United States, Incorporated
425 13th Street, Northwest
Washington, D.C.  20004
                                     2-5

-------
                             TABLE 2-1  (Continued)


      Addressee                                    Docket: Reference

 Mr. A. H. Nickolaus                                    [l_27
 Chairman, CTG Subcommittee
 Air Conservation Committee
 Texas Chemical Council
 1000 Brazos, Suite 200
 Austin, Texas  78701

 Mr. Steven A. Correll                                  D-28
 Senior Environmental Engineer
 Georgia-Pacific Corporation
 133 Peachtree Street, Northwest
 Post Office Box 105605
 Atlanta,  Georgia  30348

 Mr. Gary  D.  Myers                                      0-29
 President,  The Fertilizer Institute
 1015 18th Street, Northwest
 Washington,  D.C.   20036

 Mr.  Mark  E.  Lowing                                     D_30
 Environmental  Specialist
 Dow Corning  U.S.A.
 3901  South  Saginaw Road
 Mail  #144
 Midland,  Michigan  48640

 Mr.  Robert  H.  Collom,  Jr.                               D-31
 Chief,  Air  Protection  Branch
 Department of Natural  Resources
 Environmental  Protection Division   •
 270 Washington Street,  Southwest
 Atlanta,  Georgia   30334

 Ms. Joyce M.  Wood                                       D-32
 Chief,  Ecology and  Conservation Division
 Office  of the Administrator
 United  States Department of Commerce
 National Oceanic  and Atmospheric Administration
 Washington, D.C.  20230

 Mr. Jerry M. Starkey                                    D-33
 Senior  Environmental Engineer
 Northern Petrochemical Company
 Post Office Box 459
Morris, Illinois  60450
                                     2-6

-------
                            TABLE 2-1  (Concluded)
     Addressee                                    Docket Reference

Mr. Gary D. Myers                                      D-34
President, The Fertilizer Institute
1015 18th Street, Northwest
Washington, D.C.  20036

Ms. Geraldine V. Cox, Ph.D.                            D-37
Vice President - Technical Director
Chemical Manufacturers Association
2501 M Street, Northwest
Washington, D.C.  20037
                                    2-7

-------
 commenter's question pertains to Section 60.660(a)  of the proposed regulation
 that states distillation units operating as part of a process unit which
 produces polymers and resins are not affected facilities.  Another commenter
 (D-33)  asked if the term "polymers and resins" used under Section 60.660(a)
 refers  to those specific polymers and resins which  were addressed in the
 proposed polymer manufacturing new source performance standard (NSPS).
      Three commenters (D-2,  D-3,  and D-31)  wanted to be sure  that the
 proposed standards are not applicable to coal  tar distillation facilities  and
 facilities that "produce"  the chemicals  listed in the regulation  by extrac-
 tion from natural  substances,  not by synthesis.   For example,  one commenter
 (D-31)  stated  that tall  oil  is a  natural  substance  but contains pinene,  one
 of  the  chemicals listed  in Section  60.667.   He asked if a distillation  unit
 would be subject to this NSPS if  the unit were to separate pinene from  tall
 oil.  One commenter (D-26)  agreed with EPA  for exempting  the  distillation
 operations in  the  beverage alcohol  industry  from  the proposed  standards.
      Four commenters  (D-l,  D-16,  D-29, and  D-34)  stated  that  urea,
 urea-ammonium  nitrate  (UAN),  and  ammonium carbamate  production should not  be
 covered  by the standards because  the  production of these  three chemicals
 neither  involves the  use of  a  distillation operation  nor  the  significant
 emission  of volatile  organic  compounds (VOC).  One commenter  (D-29)  further
 stated that  these  compounds  should  not even  be considered  synthetic  organic
 manufacturing  industry (SOCMI) chemicals  according to  the  EPA description  of
 how the SOCMI  list  was developed  (Proposed Standards  of Performance  for SOCMI
 Equipment  Leaks of  VOC,  46 FR  1136, January  5, 1981).  The commenter believes
 that only  one  of the five criteria  presented in EPA's description applies'  to
 urea, UAN,  and ammonium  carbamate.  The one common criterion is a high
 production volume,  and the commenter feels that it is not a sufficient reason
 for the three  chemicals  to be  included on the SOCMI  list.  One commenter
 (D-16) presented information to clarify that the processes their companies
use for the production of urea employ a flash evaporator which is unlike a
distillation unit.  Only water is removed by the flash evaporator with no
significant amounts of VOC being released during urea production,  in contrast
to a distillation unit that separates volatile organic materials.   Therefore,
                                     2-8

-------
the commenter stated that urea production as well as UAN and ammonium
carbamate production does not employ distillation units and negligible
amounts of VOC are emitted.  However, one commenter (D-29) indicated that
distillation units are used to produce urea and any of the organic material
emitted during production of urea is primarily in the form of particulates.
Another commenter (D-34) indicated that if formaldehyde based additives  (FBA)
(used in the production of urea) were present when urea is concentrated  (via
distillation) only insignificant amounts of formaldehyde would be vented to
the atmosphere.  Because formaldehyde reacts so rapidly with urea to form a
nonvolatile solid, it is unlikely that free formaldehyde would be emitted
during urea production.  For economic reasons, most existing facilities and
all new facilities use very efficient product recovery systems.  It was
pointed out that all affected facilities that produce urea, UAN or ammonium
carbamate would probably comply with the standards by maintaining a TRE index
value greater than 1.0 because the cost of controlling VOC emissions would be
above the TRE cutoff.  The reasons given for the expected high TRE index
values are the current and extensive use of recovery equipment and the high
amounts of fuel enrichment needed for the low heating value vent streams.
The commenter stated that exemptions based on the TRE cutoff would still be
costly because of the required monitoring, recordkeeping and reporting.

     RESPONSE:  It is not the intent of the distillation NSPS to regulate the
production of polymer chemicals because a polymer manufacturing NSPS is
currently being developed for this purpose.  However,  a distillation facility
that is not physically part of the polymer process line, as in the case cited
by one commenter,  would be covered by the distillation NSPS if chemicals
listed in the regulation are produced within the process unit that contains
the facility.  However,  when a distillation operation  is an affected facility
under both standards, only the polymer manufacturing standards are applicable.
In order to clarify the extent of coverage by the distillation standards,
                                          -»
Section 60.660 has been amended to defer to the polymer manufacturing
standards in cases of overlapping coverage.
                                     2-9

-------
      As noted in Section 60.660(a), coal tar distillation is not covered by
 these standards because these operations are covered by the national emission
 standards for hazardous air pollutants (NESHAP) regulating benzene emissions
 from coke by-product plants.
      There is a small portion of organic chemicals that are extracted from
 natural  sources.  The EPA does not consider the standards to be applicable to
 chemicals that are extracted from natural  sources because the production
 processes, emissions, and control alternatives have not been investigated by
 the Agency.   Therefore,  five chemicals listed in the proposed regulation that
 are produced by extraction from natural  sources have been removed from the
 list.  These five chemicals are:  (1)  coconut oil  acids,  sodium salt;
 (2) fatty acids, tall oil,  sodium salt;  (3)  pinene;  (4)  tallow acids,
 potassium salt;  and (5)  tallow acids,  sodium salt.
      In  considering the  applicability of the standards  to the production of
 the fertilizer chemicals mentioned  by the  commenters, EPA solicited informa-
 tion from many sources including  the  commenters  themselves;  (Docket  Item
 No.  IV-C-7).   According  to  the available information, including  that provided
 by  the commenters  (Docket  Entries IV-D-35  and  IV-E-14),  the Agency  concluded
 that distillation  operations  are  not  involved  in  the production  of  ammonium
 carbamate  and  UAN.   Therefore,  these  two chemicals have  been  removed from the
 list in  the distillation regulation because  the standards are not applicable
 to  the production  of these  chemicals.
     Although  distillation  operations are  involved in the production of  urea,
 EPA  has  removed  urea  from the  list because no VOC emissions are  expected to
 occur during distillation operations  involved with urea  production.  The vent
 streams  from these distillation units consist almost entirely of inert gases
 (e.g., nitrogen), water,  ammonia, and urea particulate.   The potential for
 VOC  emissions would exist when FBA and/or methanol are added before or during
 the  distillation operation.  However, available information indicates that at
 the majority of urea production facilities, FBA is injected into the product
 stream after the distillation operation.   Therefore, there is no potential
 for formaldehyde emissions from distillation units within this type of
production process.  For  process units where FBA are injected prior to or
during distillation, the  only organic emissions expected from the
                                    2-10

-------
 distillation unit would be particulate urea.  Any formaldehyde should be
 completely reacted with urea to form methylenediurea because urea is in molar
 excess of 125 to 1.  Methylenediurea is a chemically stable compound under
 the conditions encountered (Docket Item No.  IV-E-14).
      No information was available that indicates methanol  is added before or
 during distillation.  For these reasons,  the Agency has decided to remove
 urea from the list of chemicals covered by the final  standards.

      2.1.2  COMMENT:  Several  commenters  (D-8, D-ll,  D-12,  D-13,  and D-33)
 requested clarification on the applicability of the proposed standards  to
 distillation facilities that do not  produce  as the  main product any of  the
 chemicals listed in the regulation.   One  commenter  (D-12)  stated  that it is
 inferred  from the preamble that all  economic modeling  and  costing was done on
 distillation units that distilled as a product the  chemicals listed in  the
 regulation.   The commenter suggested that,  if this  were true,
 Section 60.660(a)  should be modified because it includes all  distillation
 units  that are part of a system used to produce one or more  of the regulated
 chemicals.
     Several  commenters (D-ll,  D-12,  D-13, and D-15) recommended  that
 distillation operations that produce any  of  the chemicals listed  in  the
 regulation at trace or low levels  should  be  excluded from coverage by the
 standards.   One  of the commenters  (D-15)  suggested  that  exclusion  should
 occur  for  distillation operations  that  separate inorganic or  aqueous  streams
 with only  trace  (less  than  1 percent)  organics  present.  Another  commenter
 (D-ll) believed  that the  proposed  regulation would  be  applicable  to distilla-
 tion facilities  that contain listed  chemicals  only  as  impurities  in a process
 stream.  Therefore,  he  suggested the  standards  state that if a  process stream
 could  be tapped  and  piped to a product  storage  vessel  (even  if  the process is
 not currently so operated) and such a vessel could  be  filled with  a product
 listed in Section 60.667, then the distillation process would not  be exempted
 from the standards.  The commenter indicated this wording would remove from
coverage by the  standards processes that only contain  listed chemicals as
impurities in the process streams.  Two commenters  (D-ll and D-12) suggested
that a minimum VOC level for the vent stream be specified to determine
                                    2-11

-------
 whether a distillation operation is an affected facility.   A breakpoint level
 of 10 percent VOC content was suggested as used in the promulgation of the
 standards of performance for equipment leaks of VOC in the SOCMI (48 FR 48328
 - October 18, 1983).
      A recommendation was made (D-13)  that only those distillation units that
 produce one or more of the chemicals listed in Section 60.667,  as products
 should be considered as affected facilities.   Contaminants,  by-products, and
 intermediates should not be included in the scope  of these standards.
      One commenter (D-8) requested  clarification on the term "intermediate
 product" as used in the proposed standards.   The commenter suggested the
 definition of "intermediate product" not include a manufacturing process
 waste or a feedstock component that is  recovered by the operator for subse-
 quent use by the operator.   Feedstock components may be recovered (not
 "produced")  during production for use  in a subsequent batch.  Another
 commenter (D-33)  requested  that  EPA clarify whether a facility would be  an
 affected facility if its process  stream contains listed chemicals that are
 either sold,  disposed  of, used in the production of other  chemicals  or
 recycled.

      RESPONSE:   The  EPA  developed the standards from  data  on distillation
 facilities within  process units that produce the chemicals listed in
 Section  60.667 as  a  product,  by-product, coproduct, or  intermediate.  The  EPA
 believes  it  appropriate  to  consider  by-products, coproducts and  intermediates
 as products,  as  indicated under Section  60.661 (Definitions), because the
 cost  of  controlling  emissions  from the  production of listed chemicals in any
 of these  forms is  similar.  Furthermore, the application of the  standards  to
 facilities producing any of the listed  chemicals was found to be  reasonable.
 Therefore, the Agency considers it appropriate for the standards  to apply  to
 any distillation facility within a process unit producing any of  the listed
 chemicals as a product.  All costing and economic impact analyses were based
 upon distillation facilities within  process units that produce the listed
 chemicals as a product.  However, sometimes process unit product  streams
contain listed chemicals that are not sold or used  in reactions to form other
chemicals.  These types of process units were not analyzed and the distilla-
                                    2-12

-------
 tion facilities within them are not covered by the standards.   Some examples
 of process units that do not produce a listed chemical  as a product are those
 in which the listed chemical is a contaminant, a feedstock that is recycled
 back to a process,  part of a carrier gas used in another distillation opera-
 tion, or a conveying gas for other process uses.  Distillation operations
 within these types  of process units would not be affected facilities.
      The main factor in.determining if a listed chemical  is produced as a
 product is the use  of the chemical.  The EPA considers  either  of the following
 downstream uses as  indicative of the production of a  listed chemical as a
 product:  (1) produced for sale as that listed chemical,  or (2)  used in
 another process that needs the listed chemical.   An example of (2)  is when
 methanol,  a listed  chemical,  is produced not for sale but to react  with
 carbon monoxide to  produce acetic acid.   However,  if  a  listed  chemical  is
 only part  of a mixed stream exiting a process unit and  cannot  be sold or used
 in another process  as the listed chemical,  then  that  listed chemical is not
 considered to be produced as  a product.   For example, cyclohexane,  a listed
 chemical,  may be present  in a product stream exiting  a  distillation facility
 within a process unit that only produces gasoline.  Because the  gasoline
 would not  be sold or used in  a downstream process  for the cyclohexane only,
 the  distillation facility would not be covered  by  the standards.  Therefore,
 a  distillation  operation  is an affected  facility only if  it is within a
 process unit that produces  a  compound to be  sold as a chemical listed in
 Section 60.667  or used  in  a process  that needs a chemical  listed  in
 Section 60.667  for  the  production of chemical(s) in another process  unit.
      Listed  chemicals can  be  formed  as contaminants from  side reactions  as a
 consequence  of  producing  other  chemicals  that are  not listed, or  by  frac-
 tionating  feedstocks with  small  amounts  of listed  chemicals.  These
 contaminants would  not  be  considered  to  be produced by the  process  unit  if
 they  are not fractionated  further to  be  sold or  used  in the  production  of a
 final product.
     Another case is when listed chemicals are separated  from a reactor
product stream  in a distillation operation and recycled back to the  reactor
as an unreacted feedstock.  For example,  a low-yielding reaction occurs where
the reactants A and B are listed chemicals and are reacted to form a product
                                    2-13

-------
 C that is unlisted.   A distillation  operation  is  used to  separate chemical  C
 from the unreacted chemicals  A and B.   Chemicals  A and B  are  then recycled
 back to the reactor.   The distillation  operations within  this process  unit
 would not be covered  by the standards because  chemicals that  are not pro-
 duced,  but are feedstock chemicals that are  recovered for reuse,  are not
 considered products.   However,  if these distillation  operations  were within a
 process unit manufacturing a  listed  chemical as a product.  it would  be
 covered.
      Another example  of a case  where the standards  would  not  apply is  when
 the  noncondensibles from a distillation overhead  are  used as  a carrier gas  to
 facilitate the operation of another  distillation  operation  or as  a conveyor
 gas  for other process  uses.   If the  carrier  gas contains  any  listed
 chemical(s)  and is a  process  stream  from a process  unit used  to  produce an
 unlisted  chemical, then  the standards would  not apply to  this  facility*
      The  EPA has decided that for the purposes of these standards  it is more
 appropriate  to determine applicability  according  to whether a  listed chemical
 is produced  as a product,  instead of setting a minimum concentration level  of
 a listed  chemical as  a means of defining what may constitute  production as  a
 product.   It is not feasible to set  any  one  concentration limit for listed
 chemicals  below which the  chemical is always an impurity  or waste.   It  is not
 feasible  because the necessary  concentration or purity for a  listed chemical
 to be considered a product can  vary  from site to  site.  For example, a
 chemical  that  is produced  as 90 percent pure from one process may  only  be
 80 percent pure to be considered as  a product for another process.  If  EPA
 attempted to establish different concentration limits for all  of the processes
 covered by the standards,  the complexity and resource requirements would be
 extremely prohibitive because of the diversity of the SOCMI.  Therefore, the
 applicability  of the standards is determined according to whether  a listed
 chemical  is  produced as  a product.   Section 60.661 has been amended to
 include more specific definitions of product, coproduct, and by-product in
order to clarify the applicability of the standards.

     2.1.3  COMMENT:  Several  commenters (D-4,  D-7, D-14,  and  D-18) requested
clarification of the proposed standards  with regard to the applicability of
                                    2-14

-------
petroleum refineries producing any of the chemicals listed in the regulation.
One commenter  (D-18) stated that because the information used to develop the
standards is limited to the organic chemical manufacturing industry, the
standards should not apply to distillation units in petroleum refineries.
Several commenters  (D-4, D-14, and D-18) pointed out that many refinery
product streams contain complex mixtures with trace amounts of the chemicals
listed in Section 60.667 of the regulation.  Two of these commenters (D-14
and D-18) stated that the standards should only apply to those processes that
produce the listed chemicals in pure or nearly pure form. It was recommended
the standards  specify a minimum level for the quantities of the listed
chemicals in a process stream or distillation unit that would cause a
distillation facility to be subject to the standards.  It was suggested that
streams having less than 10 percent of any of these listed compounds should
be exempt from the standards.
     Two commenters (D-7 and D-13) noted that the background information
document (BID) (page 9-43) indicates chemicals produced primarily by
refineries are assumed to incur no costs as a result of the standards.   Both
commenters stated that an increase in chemical cost would occur if petroleum
refineries must meet the standards.

     RESPONSE:  The standards were developed to apply to any distillation
facility within a process unit producing the chemicals listed in
Section 60.667 for sale or for use in the production of other chemicals (see
comment 2.1.2).  The primary purpose of most petroleum refineries is the
production of petroleum products such as motor fuels.  However, some
refineries are involved in producing one or more synthetic organic chemicals.
The EPA believes that the standards appropriately apply to distillation
facilities in the petroleum refineries that produce the listed chemicals.
     The Agency evaluated data on organic chemical  manufacturing including
that done at petroleum refineries, in developing this NSPS.   The EPA has
found that distillation facilities within petroleum refineries do not vent
directly to the atmosphere.   Instead, these facilities have their vent
streams recycled,  routed to recovery, used as a fuel, or combusted by flaring
or incineration.   In February of 1981, the American Petroleum Institute (API)
                                    2-15

-------
sent comments concerning Chapters 3 through 6 of the BID to EPA  (Docket  Item
No. II-D-182).  In these comments API agreed with the statement  in the BID
that refinery distillation vent streams are already well controlled, recycled,
or completely recovered.  The API further indicated that additional redundant
controls would not be necessary.  However, EPA considers it advantageous to
regulate a segment of the industry that is already well controlled in order
to ensure those controls are properly operated and maintained and to guarantee
they will be applied to all facilities yet to be constructed.  Another
advantage in regulating these facilities is to ensure that a consistent level
of control is maintained for all States.  The EPA recognizes that there will
be costs associated with compliance testing, monitoring, recordkeeping, and
reporting for verification of compliance with the standards but these costs
were found to be reasonable.
     Although the Agency believes that existing distillation facilities
within refineries are already well controlled, it is possible that a few new
refineries may be built with uncontrolled vent streams.  However, these vent
streams would only have to be controlled if.EPA determines the cost
effectiveness of combustion to be less than $l,900/Mg.   Therefore, the Agency
has determined that any control costs incurred by petroleum refineries as a
result of this NSPS would be reasonable.
     Commenter D-13 was contacted for clarification of a comment regarding an
increase in chemical  costs resulting from the NSPS.  She stated that the
comment assumed that relief valve venting would be subject to control under
this NSPS and, in that case, additional  costs would be incurred.  Relief
valve discharges are not covered by the distillation NSPS,.  For a discussion
of this issue refer to the response to comment 2.3.1.

     2.1.4  COMMENT:   One commenter (D-ll) stated that distillation
facilities starting up after promulgation of the standards; have 180 days to
attain compliance and that facilities starting up between the proposal and
promulgation should be given the same consideration.  The commenter requested
that the initial reporting, performance test,  and recordkeeping requirements
be required no sooner than 180 days after promulgation of these standards.
                                    2-16

-------
     RESPONSE:  The General  Provisions  (40  CFR  60.8)  specify  that  a
performance test be conducted  at  an affected  facility no  later  than  180  days
after the initial startup of the  facility.  Even  though the standards  apply
to facilities commencing construction,  reconstruction, or modification after
proposal (December 30,  1983),  EPA does  not  require an immediate performance
test and written report to the Administrator.   Once the standards  are
promulgated, all affected facilities that were  built  between  proposal  and
promulgation will be given a reasonable amount  of time of demonstrate
compliance.  Any affected facility starting up  after  promulgation  will have
180 days from their initial  startup to  conduct  a  performance  test.   A  written
report must then be submitted to  the Administrator.

     2.1.5  COMMENT:  Three  commenters  (D-9,  D-ll, and D-27)  inquired  about
the low vent stream flow rate and design capacity exemptions  specified in the
standards.  Two commenters (D-ll  and D-27)  requested  that a wording  change be
introduced into the existing low  flow rate  exemption  that would  allow
distillation facilities operated with a maximum vent  stream flow rate  of less
than 0.008 m /min to be exempt.  Section 60.660(d) of the regulation exempts
any distillation facility which is designed with  a maximum vent  stream
flow rate of less than 0.008 m /min from meeting  the  specifications  in this
subpart, except for recordkeeping requirements.   The  commenters  indicated the
designed vent stream flow rate is usually different from  the  operated  flow
rate.  With the requested wording change, distillation facilities  could
operate with vent stream flow rates below 0.008 m /min and not  become  subject
to the standards.  This could be done even  if the design  flow rate were
higher.   Since vents are typically designed to accommodate emergency condi-
tions,  the operating flow rates are usually lower than designed  flow rates.
The additional  wording would also clarify any ambiguity over  vents that  were
designed for 0.008 m /min but are operated at higher  rates.
     One commenter (D-9) noted that it is not clearly stated  in
Section  60.660(c) that the 1 Gg/yr design capacity exemption refers to the
total amount of product from the distillation facility or that portion of the
total products  which is a chemical listed in Section  60.667.
                                    2-17

-------
      RESPONSE;  The Agency recognizes that some distillation facilities may
 operate with vent stream flow rates  above  or below their designed level.
 Therefore,  EPA has decided to amend  the regulation to  exempt from coverage by
 the standards, except for recordkeeping and reporting  requirements,  those
 distillation facilities that operate with  a vent stream  flow rate less than
 0.008 m /min.   The Agency will  no  longer continue to exempt  distillation
 facilities  only because they were  designed for  a vent  stream flow rate below
 0.008 m /min.   To ensure the flow  rate  is  continuously operated  below
 0.008 m /min,  the owner or operator  must demonstrate it  to EPA.   The owner or
 operator of an affected facility that operates  with a  vent stream flow rate
              O
 below 0.008 m /min must notify  EPA and  demonstrate compliance with a per-
 formance test  that measures  the operating  flow  rate.   Furthermore, any
 operational  changes  in  the facility  that may cause the vent  stream flow rate
 to  no longer be below the exemptio»  level  must  be  recorded along  with  a
 flow  rate measurement after  the change  has  been  made.  No reporting  to EPA is
 required until  the low  flow  level  has been  exceeded and  the  report must
 contain  the  new flow  rate measurement.   The following are a  few examples of
 operational  changes that  could affect the  vent  stream flow rate:   increased
 production  or  production  capacity, use  of  a new  feedstock, use of a  new
 catalyst, or any  replacement, removal or addition  of recovery equipment, or
 changes  in  the  operating  characteristics of the distillation unit(s).   If  any
 of  these operational  changes result  in  a flow rate greater than the  exemption
 level, they  must  be reported to EPA semiannually along with the new  flow rate.
 Otherwise, there  are  no reporting requirements after the initial   performance
 test.  If the vent stream  flow rate exceeds the exemption level,   the owner  or
 operator must comply with  the provisions of Section 60.662.
     The low capacity exemption does  not pertain only to the production of
 chemicals listed  in Section 60.667.  The EPA exempts any distillation  facility
 from coverage by  the  standards, except for recordkeeping and reporting
 requirements, when the process unit has a total  design  capacity less than   •
 1 Gg/yr  (2.2 million Ibs/yr) of all products manufactured by that process
unit.  This  includes all listed and unlisted chemicals.  In order to clarify
the meaning of this exemption, Section 60.660(c) has been amended to read
 "... total design capacity less than  1 Gg/yr 	".
                                    2-18

-------
      As  required  under  Section  60.665(i),  recordkeeping  is  needed  only for
 changes  in  process  operation  that  increase the  design  production capacity of
 the  process unit  containing the affected facility.   Furthermore, semiannual
 reporting is required under Section  60.665 (1)(6) only when these  process
 changes  occur (see  comment 2.10.2).
      The EPA has  decided  to exempt process units with  production capacities
 lower than  1 Gg/yr  from this  NSPS  because  units of this  size that  produce a
 listed chemical are typically used for research and  development.   Because
 best demonstrated technology  (BDT) was analyzed with respect to industrial
 scale facilities  and because  the operation of research and  development scale
 facilities  is different,  BDT  may not be applicable.

      2.1.6   COMMENT:  Two commenters (D-12 and D-24) requested that the
 applicability date  of the standards cited  in Section 60.660(b) be  changed
 from December 30, 1983, to the  date that the final rule  is  promulgated.   It
 was  stated  that any plants which start construction  before  the rule is
 promulgated  do not  have a defined, final set of standards for design of their
 distillation  system.

      RESPONSE:  The EPA plans no change in  the applicability date  of the
 standards as  stipulated in Section 60.660(b).  (Section  lll(a)(2)  of the
 Clean Air Act (CAA) defines new sources as  those that  commence construction
 after proposal).  This date marks the time  after which a distillation
 facility may  be considered an affected facility.  However,  compliance  is  not
 required until after the promulgation date  to allow  for the  final   rule to
 first be established before the compliance period begins.  As noted in the
 response to comment 2.1.4, all facilities that start up between proposal  and
 promulgation have a reasonable time after promulgation to demonstrate com-
pliance with a performance test.  Furthermore,  although subject to change,
the proposed standards provide an indication of what will be contained in the
final standards.
                                    2-19

-------
     2.1.7  COMMENT:  Several commenters (D-8, D-ll, D-20, and D-22) stated
that the proposed standards do not seem to adequately treat the complex
nature of batch operations.  These commenters suggested the standards be
applied on a case-by-case basis for batch distillation systems because the
vent streams from these operations often have variable characteristics.  For
example, batch vacuum distillation may have an initial higher vent stream
flow rate as noncondensibles are removed from the products being purified
followed by very low emissions for most of the cycle.  The commenter stated
that application of a recovery system for compliance with a TRE greater than
1.0 is likely to represent greater difficulty and cost impact on batch
distillation facilities than anticipated by the Agency.

     RESPONSE;  The EPA has decided to exclude batch distillation operations
from this NSPS because BDT was evaluated with respect to continuous opera-
tions which have relatively constant vent stream flow rates and compositions
within a process unit.  Batch distillation typically emits a vent stream of
variable flow rate and composition during its operation.  As a result,  BDT may
not be applicable to vent streams from batch processes.  The Agency is
currently investigating the need for an NSPS for batch processes within the
SOCMI separately from this NSPS.
     Section 60.660 has been revised to specifically exempt batch
distillation operations.  The following definition has been added to
Section 60.661: "'Batch Distillation Operation'  means a noncontinuous
distillation operation in which a discrete quantity or batch of liquid feed
is charged into a distillation unit and distilled at one time.  After the
initial charging of the liquid feed,  no additional liquid is added during the
distillation operation."

     2.1.8  COMMENT:  Four commenters (D-9,  D-17, D-24, and D-25)  called for
the exclusion of certain chemicals listed in Section 60.667 that are con-
sidered to be low in photochemical  reactivity.   The commenters stated the
listing of five chemicals in the regulation  is  inappropriate because these
chemicals also appear in the list of chemicals  determined to be negligibly
photochemically reactive by EPA (48 FR 57541)  and thus are not considered to
                                    2-20

-------
 be VOC.   The  commenters  list the chemicals  in  question  as  the following:
 dichlorodifluoromethane;   methylene chloride;  trichlorofluoromethane;
 trichlorotrifluoroethane,  and 1,1,1-trichloroethane.
      In  addition to the  five nonphotochemically reactive chemicals listed by
 EPA in the regulation, two commenters  (D-17 and D-24)  requested the removal
 of perch!oroethylene from the list of  chemicals in Section 60.667.  The
 commenters indicated that EPA has previously designated this  chemical  as  a
 negligibly photochemically reactive compound that does  not contribute  to
 ambient  ozone formation.

      RESPONSE:   The production of chemicals of negligible  photochemical
 reactivity does not preclude the presence of photochemical^  reactive  com-
 pounds in the vent streams of these facilities.  For example, chlorinated
 organics*'(e.g., perchloroethylene) which are considered to be photochemically
 reactive can  be present  in the process vent streams from the  production of
 chlorofluorocarbons.  Thus, EPA plans  no change in the  list of chemicals
 covered  by the the standards.
      In  any event, with  respect to perchloroethylene,  the  Agency has only
 proposed that perchloroethylene be considered  as a negligibly photochemically
 reactive compound (48 FR 49097).  As discussed in responses to comments
 2.7.1 and 2.7.2, negligibly photochemically reactive compounds are allowed to
 be subtracted from the TOC emission rate used  to calculate the TRE index.
 However, perchloroethylene cannot be subtracted because its status has not
 been promulgated.

-2.2  SELECTION OF AFFECTED FACILITY

      2.2.1  COMMENT:  Seven commenters (D-8, D-ll, D-13, D-20, D-22, D-23,
 and D-27) disagreed with the designation of affected facility used in  the
 proposed standards.  At  proposal, the  Administrator designated the affected
 facility as a single distillation unit with the associated recovery devices
 located  within the process units which produce any of the  chemicals listed in
 Section  60.667.  Other possible designations were also mentioned at proposal
 and EPA  specifically solicited comments concerning the proposed designation
                                     2-21

-------
 of  affected  facility.  The commenters indicated that the proposed  designation
 of  affected  facility  is  unrealistic and unrepresentative of the way
 distillation units are operated within the industry.  According to the
 commenters,  distillation units are often physically linked together  and  to
 artificially separate them would be inappropriate.  The commenters pointed
 out that many processes  consist of a series of distillation units  and in  some
 cases, units receive, as feedstreams, the product streams of other units.
 Under these  conditions several units can use a common recovery device.  They
 concluded that in these  cases, application of VOC reduction requirements  and
 monitoring methods under the proposed designation of affected facility would
 be potentially confusing and costly.
     The commenters prefer an affected facility to be designated as  the
 recovery system with all associated distillation units.  The commenters
 indicated that this designation would best reflect the way distillation units
 are used within the industry.  Also, ambiguities in monitoring and testing
 requirements  would be minimized when more than one distillation unit is tied
 to a common  recovery system.  Overall, the commenters felt that this designa-
 tion would reduce monitoring, control, and hardware costs incurred by affected
 distillation  facilities.   Furthermore,  the commenters expressed that
 differences  between emission reductions  resulting from the EPA designation of
 affected facility and the designation preferred by the seven commenters would
 be small, if  any.
     The commenters stated that the broader affected facility designation
would allow  a facility the flexibility to change operating conditions or
equipment in  lieu of adding combustion devices.  For example,  the owner or
 operator of  an affected facility may reduce emissions in existing distilla-
tion units to offset increased emissions from new or modified distillation
 units that become part of the facility.   If this compensation is made, there
would be a zero net increase in emissions from the affected facility.
Therefore, no adverse environmental  impact would result, yet economic and
energy conserving changes at a facility  could be made.
     One commenter (D-15) is in favor of the selection  made by EPA because
this commenter believes the approach will  allow for continuous improvement of
the air quality which is the intent of the standards.   Another commenter
                                    2-22

-------
(D-21) indicated that the definition of affected facility contained in the
proposed regulation conforms with the requirements of the CAA insofar as it
does not allow two or more distillation units which are joined to a common
recovery system to be interpreted as one facility.

     RESPONSE:  Based on these comments, the Agency reevaluated the
designation of affected facility and presented the results of the
revaluation in the Federal Register on May 16, 1985 (50 FR 20446).  Based on
those results and the lack of adverse comment on that reevaluation, the
designation has been changed from a single distillation unit with the
associated recovery system to an individual recovery system and all
distillation units venting to that recovery system.  For the majority of the
industry, this change will have no effect because approximately 80 to 90
percent of the industry's distillation units exist as individual units.
However, the effect of the designation change will be seen for the remaining
10 to 20 percent of the distillation units that share recovery devices.  This
change was not made to provide industry with operating flexibility; rather
the new designation is estimated to result in greater emissions reduction.
Also, as discussed at the end of this response, the new designation would
facilitate the implementation of the standards because, in some cases, the
cost and complexity associated with determining a TRE index will decrease.
     The EPA estimates a greater reduction in national  VOC emissions because
of the change in designation of affected facility.  Greater reduction in
emissions will occur with the new designation because for facilities where
the TRE is less than 1.0, emissions from existing distillation units will
also be controlled when new units are combined with existing units sharing a
common recovery system.  The Agency believes that a widespread evasion of the
modification provisions would not occur under this designation.  For example,
a change that would be considered a modification is made to one distillation
unit within a group of existing units.  In order to avoid a modification, the
common recovery system is upgraded so that no emissions increase results.
However, operational  or physical  changes that would be considered modifica-
                                    2-23

-------
 tions  are rarely made to individual  distillation  units  sharing  a  recovery
 device.   Instead,  it  is  more  likely  that  new distillation  units would  be
 added  to  an existing  group  of units  resulting  in  increased emissions.
      In the event  an  owner  added  a distillation unit to  an existing  group of
 units  ducted to  the same recovery system,  it is unlikely the facility  could
 avoid  being considered a modification by  offsetting the  new distillation  unit
 emissions somewhere else within the  distillation  group.  This is  because  it
 would  likely be  technologically infeasible to  reduce emissions  sufficiently
 from the  other distillation units.   Although some VOC reductions  could occur
 through upgrading  recovery equipment, it  is unlikely that  this  reduction
 would  result in  a  full offset  of  the new  distillation emissions unit because
 the increased load on  the recovery device  (i.e.,  increased  flow and VOC)
 would  make  the needed  increase in VOC removal  efficiency difficult to achieve.
 Thus,  the likely result  is that addition  of a  distillation  unit to a group of
 joined units would bring the entire  set under  the coverage  of the standards
 as a modified facility.   Even though it is unlikely to occur, if the owners
 or operators of the facility could completely  offset emissions from a new
 distillation unit by upgrading the recovery system, it would be the equivalent
 of 100 percent VOC control for that new distillation unit.   This is 2 percent
 more than would be necessary if the  individual  units were designated as
 affected  facilities and  98 weight-percent control  were applied.
     The  EPA believes  that coverage of equipment through the reconstruction
 provisions will not be avoided under the broad designation  because major
 physical   changes (i.e.,  reconstructions) to individual  distillation units
within a group of units  rarely occurs within the industry.   It is not likely
 that an owner or operator of an affected facility made up of a group of
distillation units could replace one of the units  and avoid being considered
a reconstruction.  Available data show this situation would not  arise because
the replacement of the individual  distillation  units or pieces of recovery
equipment is rare within the industry.   This is because distillation units
are expensive pieces of equipment which are designed to last a long time
 (Docket Item No.  II-B-13).  Moreover, the Agency has concluded that those few
replacements which do  occur often result from process changes or catastrophic
events that would probably require replacement  of  most  of the group of
                                    2-24

-------
 distillation  units  joined  to  a  single recovery system.  These changes would
 likely  amount to  a  "reconstruction" of the facility as  it  is defined in  these
 standards.  Thus, in the small  percentage of cases where distillation unit
 replacements  occur, the facility would most likely fall under the coverage of
 the  standards.
     An  incidental  effect  of  this change in the affected facility designation
 is that  implementation of  the standards would be made significantly easier.
 When two or more  distillation units are joined to a common recovery system,
 determining a TRE index value is less complex and less  costly because fewer
 test sites are required to measure the vent stream characteristics needed to
 calculate a TRE index value.  Under the designation of  these standards set
 forth at proposal,  it was  required only that the portion of the combined vent
 stream contributed  by the  new,  modified or reconstructed distillation unit
 comply with the standards  when  it shares a recovery system with existing
 units.  Therefore,  the TRE index value was determined for the portion of the
 stream contributed  by that unit only.  This determination was complex and
 costly and was based upon  an apportioning method using  sampling sites located
 just downstream of  the new unit and sampling sites located upstream and
 downstream of the common recovery system.  These sampling data were to have
 been used to  determine the overall efficiency of the common recovery system.
 This efficiency was then to have been applied to the vent stream of the new
 unit to determine its contribution to the total  emissions from the common
 recovery system.
     Under the new designation,  the standards require only one sampling site
 located after the last recovery device to determine a TRE index value.   No
 determination of recovery  efficiency is needed in this case because the
 entire vent stream  is covered.  Therefore,  there is no need to determine
which portion of the final  vent  stream from a group of distillation units is
 attributable to new, modified, and reconstructed distillation units and which
portion is attributable to distillation units that have not been changed or
added.   This results in a  performance test  requirement that is less costly
and less complex because fewer sampling sites and subsequent analyses are
needed.
                                    2-25

-------
 2.3   DEFINITIONS

      2.3.1   COMMENT:  Three commenters  (D-8, D-13, and D-33) proposed
 modifications of the definition for "vent stream" in Section 60.661 where  the
 term  is designated to mean any gas stream released to the atmosphere from  any
 distillation unit.  Two of the commenters (D-8 and D-33) expressed that
 equipment leaks should be clearly excluded from the definition so that the
 wording would apply only to VOC-containing streams.  Two commenters (D-13  and
 D-33) requested that the definitions of "vent stream" be made less broad so
 as to exclude relief valve discharges.  The commenters added that equipment
 leaks and relief valve discharges are already covered by other NSPS.

     RESPONSE:  This NSPS is not concerned with regulating equipment leaks or
 relief valve discharges of VOC.  These sources are regulated by an NSPS for
 fugitive emissions in the SOCMI (48 FR 48328).  In order to clarify the
 meaning of "vent stream" as used in this regulation,  the Agency has amended
 Section 60.661 as follows:  "vent stream" means any gas stream released to
 the atmosphere from any distillation facility excluding equipment leaks and
 relief valve discharges.

     2.3.2  COMMENT;  One commenter (D-19) asked if the definition of
 distillation operations excludes those distillation units that have one
 product stream.  The definition of "distillation operations" in
 Section 60.661 of the proposed regulation states that one or more feed-
 stream(s) are separated into two or more product streams during a
 distillation operation.   The commenter also inquired  about still  bottoms
 being considered as a product stream according to this definition.

     RESPONSE:  The number of product streams exiting a distillation unit
would not be used to determine if it would be an affected facility.   As
 indicated in Section 60.660(a), the standards are applicable to any
distillation facility operating as part of a process  unit producing any of
the chemicals listed in  Section 60.667 as a product.   Even if a listed
chemical  were not produced as a product by a distillation unit,  that unit
                                    2-26

-------
would  be affected  by the  standards  if  it were part of'a process  unit  pro-
ducing a listed chemical  as a product.  A discussion concerning  the meaning
of  "product"  for the purposes of this  NSPS  is given in the response to
comment 2.1.2.  To further clarify  the applicability of the distillation
NSPS,  the definition of distillation operations in Section 60.661 has been
amended to read "'distillation operation' means an operation  separating one
or  more feedstream(s)  into two or more exit streams .  . . within the
distillation  unit."
     In order for  still bottoms or  any other stream exiting a distillation
unit to be considered  a product, the stream must contain a chemical listed in
the regulation.  Furthermore, the stream must either be sold  as that listed
chemical or used in another process requiring that listed chemical.

     2.3.3  COMMENT:   Two commenters (D-ll and D-12) stated that since some
process heaters may transfer heat to process fluids not contained in tubular
coils,  any definition  of process heater should not include a reference to
tubular coils.  Therefore, both commenters suggested that the reference to
"tubular coils" be eliminated in the definition of "process heaters" in
Section 60.661 of  the  regulation.

     RESPONSE:  The Agency agrees that some process heaters do not have
tubular coils, but instead have straight-tube or U-tube arrangements.   Thus,
to  further clarify the definition of "process heaters," Section 60.661 has
been amended  so that "tubular coils" is replaced by "tubes."

     2.3.4  COMMENT:   One commenter (D-27)  stated that "TOC" as used in the
preamble is a poor abbreviation for total  organic compound.  The meaning for
"TOC"  is more commonly taken to be total  organic carbon.

     RESPONSE:  Because there is no reference to total  organic carbon  in the
preamble or regulation, the Administrator feels that "TOC" is not an
ambiguous abbreviation for the purposes of this NSPS.   Furthermore,  the term
total  organic compound is clearly defined  in the beginning of the regulation.
                                    2-27

-------
      2.3.5  COMMENT:   One commenter  (D-19)  stated  that  the  definition of
 "corrosive vent  stream"  in Section 60.661  is  not clear.   The  commenter was
 uncertain  whether EPA means streams  containing  the equivalent of 20 parts per
 million  by volume (ppmv)  halogens or 20  ppmv  halogen-bearing  compounds are
 corrosive.   The  commenter stated that the  use of a 20 ppmv  halogen-bearing
 compound limit could  result in a significant  range of halogen concentrations
 in  the incinerator flue gas depending on the  percentage of  halogen  in the
 compound.

      RESPONSE:   The EPA considers a  vent stream to be corrosive when  the
 stream concentration  of halogen-bearing compounds  is 20 ppmv  or greater.  The
 Agency is  aware  that  even  small amounts of halogenated compounds may  be
 corrosive  and can  necessitate the use of an incinerator with  a scrubber
 system.  It  was  judged that  this 20  ppmv concentration is low enough  to
 account  for  this  situation,  even if many halogen atoms are  attached to one
 compound (e.g.,  carbon tetrachloride).  Therefore,  the control cost and TRE
 were  not underestimated for  any facility.
      In order to  be consistent throughout the regulation,  the phrase
 "corrosive vent  stream" in Section 60.661 (Definitions)  has been changed to
 "halogenated vent  stream."  However,  the definition for halogenated vent
 stream will  remain the same as the earlier corrosive vent  stream definition.

 2.4   SELECTION OF  BDT

      2.4.1  COMMENT:  One commenter (D-24)  requested clarification on whether
 BDT for the proposed standards would  qualify as  a best available control
 technology (BACT) or lowest achievable emission  rate (LAER).
     The  commenter also stated that NO  emission increases resulting from the
                                      A
 combustion of process vent streams  could bring combustion  sources under BACT
review in attainment areas.  In light of this, the  commenter suggested that
EPA review the impact on costs associated with NO  emission  increases from
                                                 A
flares, boilers,  and incinerators.
                                    2-28

-------
      RESPONSE:  The BDT for the standards will be considered  in defining  BACT
 or LAER in each plant-specific evaluation.  The evaluation encompasses a
 consideration of emissions such as S0? and NO  as well as VOC emissions.
                                      fc       A
 There are situations, however, where the emissions reduction required for a
 pollutant under BACT or LAER may be greater than that which results from the
 NSPS.  The BACT is defined in Section 169(3) and LAER is defined in
 Section 171(3) of the CAA.
      Any NOX emissions increase resulting from the combustion of distillation
 vent streams under the NSPS are not expected to be great enough to bring the
 combustion sources under BACT review in attainment areas.  A distillation
 facility under BACT review would be required to control  its NO  emissions if
    f                                                           ^
 an increase in NOX of 40 tons per year or more occurred  as required under
 40 CFR 51.24(B)(23)(i).   In order to estimate the likelihood that a 40 ton
 per year increase in NOX emissions would  occur as a result of vent stream
 combustion of VOC emissions,  two analyses were performed (Docket  Item
 No.  IV-B-16).   Typically,  distillation vent  streams  in SOCMI  contain non-
 nitrogenous  VOC compounds.   One  analysis  was  done for a  facility  with
 non-nitrogenous VOC  compounds  in  the  vent stream.   This  analysis  maximized
 potential  NOX  emissions  from an  example facility  by  using the  highest  flow
 rate  vent  stream found in  the  available emissions  data along with  a high VOC
 concentration.   The  total  amount of NOX as NOp  estimated  by this  analysis  is
 less  than  10 tons  per year.  An  additional analysis was  performed  for  the  few
 cases where the  vent stream does contain  nitrogenous  compounds.   For this
 analysis the highest vent  stream flow  rate was  also assumed and it  was assumed
 that this  vent  stream contained a  high concentration  of nitrogenous compounds.
 The total  amount of NOX as N02 estimated  for this case is less than 30 tons
 per year.  Thus, analyses show that the combustion of vent streams  from the
 production of a listed chemical would not be expected to  require the control
 of NOX under a BACT review.  Therefore, costs associated with NO  control
will not be examined for this NSPS.

     2.4.2  COMMENT:   Eleven commenters (D-5, D-7, D-ll,  D-12, 0-13, D-14
D-19,  D-22, D-23, D-24,  and D-27) are concerned that the  flare specifications
listed in the proposed standards for combustion of process vent streams are
                                    2-29

-------
 too  strict.   The  specifications  set  a minimum heating value  at  either
 11.2 MJ/scm  (300  Btu/scf)  or  7.45 MJ/scm  (200 Btu/scf)  (depending  upon  the
 type of  flare)  and  an  exit velocity  of less than  18 m/sec  (60 ft/sec).
 Commenters (D-5 and D-22)  suggested  that  whatever changes  are made to the
 flare specifications under the standards  of performance for  SOCMI  equipment
 leaks of VOC  (48  FR 48328  October 18, 1983) should also be applied to this
 proposed NSPS to  maintain  consistency.
      Various  reasons were  given  by the commenters why EPA  should reconsider
 using the flare specifications now listed as part of the distillation
 standards.  Four  commenters (D-7, D-12, D-22, and D-24) stated  that little
 data and not  a  broad enough range of test conditions were  used  to  establish
 the  specifications.  It was also stated that the technical basis for  the •
 specifications  should  be explained.  Commenters (D-7, D-ll, D-12,  and D-23)
 asserted that flares at higher velocities provide highly effective VOC
 control.  One commenter (D-27) stated that the vent stream minimum heating
 value  requirement for  flame stability should be 150 Btu/scf, instead  of the
 300  Btu/scf in  the  proposed standards.  Another commenter  (D-19) remarked
 that  the EPA  flare  specifications on flare gas velocity and Btu content are
 not  reasonable  and  should  be shifted to a performance based standard.  The
 commenter felt  that the presence of a suitable flame should be the main
 indicator of an efficient  flare.
      Several  commenters were concerned that the specifications would  have
 negative effects upon the  cost and operation of flares.   Two commenters (D-ll
 and D-13) noted that the use of advanced technology flares such as some
 Linear Relief Gas Oxidizers (LRGO) would be precluded by the velocity
 restrictions.  Other commenters (D-13 and D-23)  stated that normal  and
 emergency venting for pressure relief would be hindered by the flare maximum
 velocity specification.  Both commenters requested a provision for the waiver
 of the maximum velocity of 60 ft/sec during periods of emergency pressure
relief.  Three commenters  (D-7, D-ll, and D-33)  stated that the proposed
flare specifications would have a large economic impact on some existing
facilities that would come under  the regulation  of these standards.  These
plants already use flares designed and operated  at velocities considerably
                                    2-30

-------
 higher than the 60'ft/sec specification.   Two of the commenters (D-ll and
 D-33)  stated that construction of new flares to meet the velocity
 restrictions will  be costly and in many cases completely infeasible.
     One commenter (D-14) remarked that it should be possible to show that a
 velocity lower than  60 ft/sec combined with a lower Btu content than  either
 300 or 200 Btu/scf (depending upon the flare type)  would lead to a reduction
 efficiency of at least 98 percent.  It was noted the standards as written do
 not give'proper credit for control devices prior to flares,  such as vent gas
 scrubbing systems  that cause the vent stream heating value  to decrease due to
 the removal  of organics.   The commenter indicated that  it may be unnecessary
 to  add natural  gas to scrubbed vent streams if an exit  velocity less  than 60
                                                                   «
 ft/sec were used.
     One commenter (D-27) requested an extension of the comment period so
 they could review  the Energy and Environmental  Research report on flares
 prepared for EPA.

     RESPONSE:   Because of the technical  infeasibility  of testing for the VOC
 reduction  efficiency  of flares,  EPA determined  it necessary  to set  opera-
 tional  specifications  to  ensure  98 weight-percent reduction  efficiency.   The
 original  specifications were based upon the best data available at  the time
 of  proposal.   Based on new data  on flare  performance obtained  since proposal,
 EPA  developed  revised  flare operating  specifications and  proposed these
 specifications  on  April 16,  1985  (50  FR 1494).   After receiving  public
 comments on  the  revised specifications, they were finalized  and  incorporated
 into 40  CFR  60.18  of  the  General  Provisions  (January 21,  1986,  51 FR  2699).
 During this  period the commenters  were given opportunity  to  review the new
 study on the efficiency of  flares  and  to comment  on the study  and the  revised
 operating  specifications.
     Two commenters noted that LRGO flares could  not be used with the
 60 ft/sec exit velocity limitation.  The EPA does not have sufficient  data to
evaluate the reduction efficiency  of LRGO flares  at exit velocities greater
than 60 ft/sec.  However,  the Agency will  evaluate any data  submitted  by  the
                                    2-31

-------
 SOCMI to demonstrate the 98 percent reduction efficiency of LRGO flares.  If
 these flares are judged to be capable of achieving 98 percent reduction
 efficiency then they could be used in complying with the standards.
      Two commenters requested a provision for the waiver of the maximum
 velocity of 60 ft/sec during periods of emergency pressure relief.   Emergency
 venting is considered a malfunction and allowed under these standards and
 there is no need for a waiver.   However,  records must be kept of each
•occurrence and its duration as  required under Section 60.7(b) of the General
 Provisions.  Recordkeeping is required so that EPA will  be able to  determine
 how many malfunctions are occurring and then be able to  determine if the
 device is being properly operated and maintained.
                                                          it
      Two commenters stated that the exit  velocity limitation of 60  ft/sec
 when the gas stream heating value is less than 1,000 Btu/scf could  necessi-
 tate the replacement of existing flares designed and operated with  velocities
 greater than 60 ft/sec.   However,  EPA believes existing  facilities  coming
 under the regulation of the standards would  not have to  completely  replace
 these flares.   Instead of building a new  flare,  only the existing flare tip
 and some auxiliary equipment would have to be changed to accommodate the
 60 ft/sec limitation.   Considering that many components  of the existing flare
 could still be used,  EPA has judged the cost of modifying an existing flare
 to be well  below the cost estimated to construct an  entire flare.
 Furthermore,  the cost  of constructing a new  flare  was found to be reasonable,
 and the TRE equations  are based on new flare costs.   Therefore,  the cost
 impacts for facilities which choose to control  VOC by modifying  an  existing
 flare should also be reasonable.
      The requirement that vent  streams to  be flared  have a heating  value of
 at least 300 Btu/scf,  if air or steam assisted flares are used,  is  necessary
 because available data show that  for  some  flares,  flame  stability and 98
 percent emissions reduction cannot be consistently maintained  below a heating
 value of 300 Btu/scf.  Furthermore,  it is  important  to maintain  the flare
 specifications  and not only rely  on  the presence of  a flame,  as  suggested by
 one commenter.   Data  indicate that some flares with  a visible  flame can  not
 achieve a reduction  efficiency  of  98  percent  (Docket  Item No.  IV-A-1).
      Another commenter requested  that  specifications  for velocities lower
 than  60 ft/sec  be provided  and  that.lower  fuel requirements  be specified for
 these velocities.   However,  the current available  data do not  indicate  that
                                     0 1O

-------
 an emissions reduction of 98 percent or greater can be constantly maintained
 at heating values below 300 Btu/scf or 200 Btu/scf depending on flare type
 and exit velocities lower than 60 ft/sec.

      2.4.3  COMMENT:   One commenter (D-10) stated that catalytic oxidation
 can be an attractive  alternative to thermal  incineration.   He indicated that
 although catalytic oxidation can be designed for high VOC  reduction
 efficiencies,  economic factors dictate whether these levels are practical.
 The commenter  added that the 98 percent reduction efficiencies associated
 with the proposed standards would possibly require the use of uneconomically
 large catalyst volumes in catalytic incinerators or require the use of
 thermal  incinerators.   The commenter pointed out that 98 percent reduction
 efficiency may not result in a measurable  improvement in the environment over
 the case of catalytic  oxidation at an efficiency slightly  lower than
 98 percent.  Furthermore,  the use of thermal  incineration  may 'potentially
 entail the following detrimental  effects:   (a)  higher energy usage  by the
 affected facilities;  (b)  an increase in NO  emissions from the affected
                                           A
 facilities using  thermal  incineration;  and (c)  a decrease  in the inter-
 national  competitive position of domestic  chemical  producers with respect to
 foreign  competition.

      RESPONSE:  The EPA  has  determined  that, where  catalytic oxidation  units
 are  applicable, they can  achieve  98  weight-percent  reduction efficiency.
 However,  since  using catalytic  oxidation for VOC emissions  reduction  has  not
 been  demonstrated  to be  universally  applicable  for  all distillation  process
 vent  streams, catalytic oxidation was not  evaluated  by the  Agency.   Catalytic
 oxidizers  are limited  by their  inability to handle  streams  with  high  heating
 values because  deactivation  of  the catalyst occurs  at high  temperatures.
 Catalysts  can also be deactivated by compounds present in some waste  streams,
 such  as  arsenic, sulfur, mercury, lead, zinc, or tin.
     The Agency evaluated two control techniques universally  applicable
within the industry, therma-1  incineration and flaring.  The Agency examined
emissions data  from incinerators and flares already operating within  the
 industry, as well  as incinerator and flare tests conducted by the'Agency  and
by chemical companies.   All the new, well-operated incinerators and flares
                                    2-33

-------
 were found to achieve 98 percent reduction  efficiency.   From the available
 data the Agency determined many facilities  could  achieve 98 percent reduction
 efficiency at a reasonable cost using  thermal  incineration  and flaring.   The
 Agency has bounded  the cost impacts  associated with  combustion devices  by
 costing the most expensive devices:  thermal  incinerators and flares.  Costs
 associated with these devices were found  to  be reasonable for facilities  with
 TRE  index values less than 1.0.   Therefore,  if a  stream  does exist  whose
 constituents are such that a catalytic oxidizer is not applicable or  if the
 cost of using a catalytic  oxidizer is  too high for a particular situation,  a
 flare or thermal  incinerator can be  used.
      When evaluating  the economic effects of this NSPS,  EPA calculated the
 maximum chemical  price increases that-could  be expected  as  a direct conse-
 quence  of the standards.   Because these "reasonably worst-case"  price
 increases were quite  small, EPA  can  be  sure  there will be no significant,
 unexpected,  harmful economic effects associated with 'the  standards.  To be
 sure, price  changes are not the  only economic  effects of  an  NSPS, but if
 potential  price  changes are small, there is  no reason to  suspect that the
 NSPS  might  trigger significant changes  in profits, interest  rates,  industry
 growth,  employment, production,  competition, foreign trade,  and  related
 economic  variables.   In general,  firms will   not build, modify,  or reconstruct
 production  facilities  until a reasonable return can be expected  on  the
 investment,  including  the  investment in pollution control.   If  the  cost of
 pollution control is  not offset  by improvements in reactor catalysts,
 marketing economies of scale, etc.,  then firms may delay  construction plans
 for a short  period until prices  rise to cover the pollution  control  costs.
     The  commenter is particularly concerned that EPA, by riot allowing
 catalytic incinerators to substitute for thermal  incinerators, will  create  a
 competitive disadvantage for U.  S. plants vis-a-vis foreign competition.
 However,  the Agency is not preventing the use of catalytic incinerators.
They may  be used if the emission reduction requirements are satisfied.
Furthermore, if an owner or operator chose to use a thermal   incinerator,  no
economic disadvantage should be created vis-a-vis  foreign competition.  Using
reasonable control estimates,  chemical  price increases are expected  to range
from 0 to about 4 percent.
                                    2-34

-------
       The energy and environmental  effects mentioned by commenter D-10 were
  considered in selecting thermal  incineration as one of the control  techniques
  upon which the impacts of the standards  are based.   The energy consumption
  and cost associated with thermal  incineration were  considered and found to be
  reasonable because these can  be  offset by the use of recuperative heat
  exchangers.   The potential  for increased NOX emissions was also examined,  but
  the rate of N0x formation is  expected to be low due to relatively low
  combustion temperatures  and relatively short residence times  associated with
  thermal  incineration.

       2.4.4   COMMENT;   A  request was made  concerning  the  use of  control
  techniques other than  combustion.  One commenter (D-12)  stated  that industry
  should be  allowed  to use  switching condensers for control  of  VOC  emissions  if
  98 percent reduction of VOC is achievable.  A switching  condenser operates
  such that the condensate  is removed by freezing on the heat exchanger
  surface,  and would probably work effectively in general VOC control service.
      Another commenter (D-21)  stated that the consideration of emission
 control  technologies should be expanded  to include the evaluation of
 technologies other than flaring or incineration.  The commenter recommended
 that technologies which involve lower costs and energy requirements than
 flaring  or incineration be evaluated for  application to distillation
 facilities for which flaring or incineration has been determined to have too
 high a cost-effectiveness value.   Specifically,  the  commenter  pointed  to
 product recovery devices,  such as  carbon  adsorption,  or other  devices  such  as
 catalytic oxidation as  technologies  which should be  examined further by the
 Agency for application  to sources which are currently exempted from  the
 emission  reduction requirements.

      RESPONSE:   Several VOC  control  technologies  such  as  recovery  devices
 (including  condensers)  and catalytic oxidizers were  also  examined  but were
not  included  in  the  impacts  analysis because  the Agency was unable to
identify subcategories  for which those devices would always apply.  The
applicability and effectiveness of adsorption, absorption, or  condenser
devices for VOC emission control is sensitive to several physical
                                    2-35

-------
 characteristics of the organics  in  a vent stream  and  other stream
 characteristics.   Organic characteristics such  as solubility,  molecular
 weight and liquid/vapor equilibrium are important to  the  success  of using
 specific  recovery devices for  VOC control.   Vent  stream moisture  content is
 also  important.   However, these  characteristics are variable  from vent  stream
 to  vent stream making  it difficult  to  identify which  particular recovery
 device would  be appropriate  for  application  to categories  or  subcategories of
 distillation  vent streams.
      In the case  of catalytic  oxidizers,  these control devices are limited
 by: (1) their inability to combust  all  streams equally as  efficiently,  (2)
 the high  capital  cost  of the catalyst,  and (3) their  inability to handle
 streams with  high heating values.   In  addition, deactivation of the  catalyst
 in these  systems  occurs at high  temperatures.  This is likely to  occur  during
 incineration  of streams with high heating values, which is  a common  situation
 for distillation  facilities.   Catalysts can  also  be deactivated by compounds
 present irr some waste  streams, such  as  sulfur, bismuth, phosphorus,  arsenic,
 antimony,  mercury,  lead,  zinc, or tin.
      Because  of these  susceptibilities  to individual waste  stream
 characteristics,  using  catalytic oxidation or recovery devices (including
 condensers) for VOC  emissions  reduction has  not been demonstrated  to be
 universally applicable  to any  identifiable subcategories of distillation
 process vent  streams.   The VOC reduction efficiencies may vary among
 processes  and  among  plants.  Although catalytic oxidizers and recovery
 devices were  not  included in the impacts analysis, these devices  can be  used
 to meet the requirements  of these standards  if they can achieve the
 98 percent VOC reduction  requirement on streams to which they are  being
 applied.
     Since the Agency is  unable with available information and resources to
 identify subcategories  of distillation operations for which other  VOC control
 techniques have been demonstrated to always apply, there is no way to evaluate
 techniques for application to distillation units for which thermal incinera-
 tion and flaring have been determined to be too costly.   Even with greater
 resources, this approach would be infeasible  because it would require a
 stream-by-stream characterization,  ultimately resulting in the need for a
 separate standard for each vent stream from a process  used to produce a
distillation chemical.   The number  of standards required to regulate the same
                                    2-36

-------
 number of sources would increase by several  hundred.   The Agency believes
 that such an approach to regulating the SOCMI distillation industry would be
 administratively infeasible and therefore environmentally counterproductive.
 In any event,  proceeding now with this generic regulation based on thermal
 incineration and flaring at least represents an important first step in
 regulating distillation emissions and does not preclude later regulation of
 subcategories  of distillation facilities should that  become feasible.   The
 EPA believes it  has  the authority to take this step by step approach under
 Section 111.   See, e.g..  Group Against Smog  and Pollution v.  EPA.  665  F.2d
 1284 (D.C Cir.  1981).

 2.5  COST ESTIMATION

      2.5.1   COMMENT:   One commenter (D-15) suggested  that the cost analyses
 presented in the preamble and BID that were  used  in developing the TRE  index
 should  be reevaluated.   Several  expenses were not  included in the  EPA cost
 equations.   New  cost  equations should include:
      (a)   the cost of  pipeline required to retrofit a  distillation column in
           an existing  plant  to control  vent  stream emissions;
      (b)   the cost of  locating combustion  units  in relatively remote areas
           to avoid a  safety  hazard;
      (c)   the cost of  supplemental  fuel  required by flares  for distillation
           vent streams  with  low  net  heating  values; and
      (d)  the cost of monitoring  equipment and recordkeeping.

      RESPONSE:   In response  to this  and  other comments, the Agency reexamined
the costing procedures  used  for the  distillation operations NSPS.   In addition
to reexamining the costing procedures  in light of these comments,  many other
cost assumptions not commented upon  by  industry were examined  and,  in some
cases, changed to include more up-to-date information.  When making  revisions
to the costs, the goal  has been to provide an updated estimate of  total
annualized costs that are representative of the actual costs experienced by
the majority of facilities in the industry.  Industry comments, particularly
those dealing with recommended equipment additions, were first carefully
                                    2-37

-------
 evaluated  by  EPA  to  determine  if  inclusion of such  items were  reasonable.   If
 the  addition  of equipment was  found to be valid, costs for that  equipment
 were estimated and included.   This approach was used for recommended
 operating  cost revisions as well.
      This  commenter  stated that new cost equations  should include  the  cost  of
 pipeline required to retrofit  a distillation column in an existing plant.
 The  EPA determined that since  use of a new combustion device at  a  new  plant
 should be  the basis  of costing, duct length criteria should be based on the
 most probable location of a new combustion device.  Based on criteria
 established by the National Fire Protection Association (NFPA) and the
 Industrial Risk Insurers, EPA  revised its design criteria to include 200 feet
 of ducting between the edge of the process unit and an incinerator and 300
 feet  of ducting between the edge of the process unit and a flare.   An  addi-
 tional 100 feet of ducting was included in the costing to route the vent
 stream from the distillation unit vent to the edge of the process  unit.
 Installation factors are included in the costing.
     The costs of locating combustion units in relatively remote areas to
 avoid a safety hazard are implicit in the costing assumptions described
 above.  A combustion unit is expected to be located as close as possible to
 the  process unit,  but far enough away to provide safety.   For this  reason,
 the Agency used insurance underwriter criteria for locating combustion
 sources at a safe distance from process equipment as a basis for costing.
     The cost of supplemental   fuel required by flares for distillation vent
 streams with low heating values was reevaluated by EPA in its operating cost
 revisions.   For those distillation vent streams having heating values less
 than 300 Btu/scf (11.2 MJ/scm), gas enrichment is  necessary to comply with
 the flare requirements of the proposed standards.   It is  likely that most
 vent streams would be enriched through the addition of natura'i  gas.
 Therefore,  the cost algorithm has been changed to  incorporate the  cost for
 adding enough natural gas to achieve a vent stream heating value of
300 Btu/scf (11.2  MJ/scm).
     The cost of monitoring equipment  and recordkeeping is not included in
the cost analyses  that were used in developing the TRE equations.  However,
these costs were estimated  and determined to  be reasonable to ensure proper
                                    2-38

-------
 operation and maintenance of either the recovery system or the control
 device.   The TRE equations are used to determine BDT for each affected
 facility.  All costs associated with reducing VOC emissions by 98 weight-
 percent  or to 20 ppmv using either a flare or a thermal  incinerator have been
 included in the TRE equations.  These are the costs that are the basis of the
 BDT determination.   Once this determination is made,  the costs associated
 with monitoring the recovery system or control device to ensure proper
 operation and maintenance were estimated.   The costs  were then evaluated and
 used to  determine which monitoring and recordkeeping  requirements were
 necessary and reasonable.  However,  the TRE equations which are used to
 determine BDT for an affected facility do  not include monitoring or
 recordkeeping costs.

      2.5.2  COMMENT:  Three commenters (D-ll,  D-27, and  D-37)  stated that the
 cost equation for flare systems result in  costs that  are too low.   One
 commenter (D-27)  reevaluated a specific case for flare costing,  and felt that
 EPA has  underestimated the total  installed capital  (investment)  cost by a
 factor of 3  and the annual  cost by a factor of 2.   The reason  for the low
 estimate of  capital  cost,  according  to commenters D-27 and  D-37,  is that EPA
 did not  adequately  include all  the services and auxiliaries needed to make
 the system operate.   The  major capital  cost-items identified as  being either
 overlooked or not adequately treated  by EPA include the  following:
 (a)  500  feet  of new pipe  bridge;  (b)  all steam,  natural  gas and  electrical
 services  to  the flare;  (c)  knock-out  drum  and  fluidic  seal;  and  (d)  a TV
 camera to  observe flare operations for smoke  and a TV  monitor  in  the central
 control  room.   Major annual  cost components  identified as being  either
 overlooked or not adequately treated  by  EPA include the  following:
 (a)  general plant overhead  estimated  as  50  percent of  operating  and  main-
 tenance  labor costs;  (b) engineering  and environmental oversight costs
 estimated  as  about 40 percent of maintenance and operator labor costs;
 (c)  labor  and  supervision  for maintenance estimated as 3 percent of  the  total
 installed capital costs; and  (d) operating  supplies estimated as  15  percent
of maintenance costs.  Miscellaneous cost components that the commenter  said
were overlooked by EPA include the following:  (a) a general contingency
                                    2-39

-------
 allowance;  (b)  a labor allowance  for  bad weather;  (c)  the  cost  of a strip
 chart  recorder  for  temperature  monitoring;  and  (d)  electrical control  room
 equipment.
     The  same commenter (D-27)  indicated EPA  has made  two  basic flare  design
 errors  that  would result  in  additional material and operating expenses.   One
 flare design error  results in an  unsafe continuous  heat  intensity at ground
 level.  A much  lower  heat intensity should  be used, resulting in  a greater
 flare height, and therefore  greater capital and annual costs.   However,  the
 commenter did not recommend  an  alternative  value for ground level  heat
 intensity.   The  other design error indicated by the commenter is  that  the
 flare tip pressure  drop of 27 inches of water is too high  for the 60 ft/sec
 exit velocity limitation in  the proposed standards.  The 27 inch  pressure
 drop is also inconsistent with  smokeless flare operation.  The  commenter
 suggested the pressure drop  be  specified at less than  1  inch of water, which
 would require a  greater flare tip diameter  and an increase in the  estimated
 capital cost.
     One  commenter  (D-ll) stated that the EPA flare specifications would
 result  in higher capital and operating costs than the  costs estimated by  EPA.
 The increased costs would be due to the larger diameter flares  needed to  meet
 the flare specifications of  60 ft/sec exit velocity and a 300 or  200 Btu/scf
 (depending on flare type) minimum heating value.  Examples of these increased
 costs include:   (a) an increase in pilot gas requirements; (b)  shortening  of
 flare tip life due to higher operating temperatures when exit velocities  at
 or below  60  ft/sec are used;  and (c)  added purge gas requirements during
 shutdown  and periods of no flow to the flare.
     RESPONSE: In response to commenters'  concerns regarding flare costing
 assumptions, the Agency has  reevaluated all  costing assumptions  and revised
 them where it was justified.   Throughout the development of the  distillation
 NSPS the Agency has made efforts to ensure that the cost algorithm resulted
 in estimates that adequately  represent control costs anticipated to be
 incurred by the majority of  facilities in  the  industry.  Prior to proposal,
 industry members were given the  opportunity  to provide substantial input  into
the development of the cost algorithm.   Preliminary costing assumptions were
reviewed by industry and subsequently  revised  based upon industry input.
                                    2-40

-------
 After proposal, costing assumption revisions were prepared and presented in a
 supplemental Federal Register notice (50 FR 20446) on May 16, 1985.  This
 notice solicited further comments on costing procedures.  The bases for these
 revisions are documented in Docket Entry IV-B-8.  The Agency feels confident
 that the revised cost procedures represent accurate estimates for typical
 facilities.
      Revisions to the capital cost assumptions included the addition of the
 costs of a new pipe support system for use in routing flare services and the
 vent stream to the base of the flare.   The 500-ft pipe bridge referred to by
 the commenter is apparently a heavy-duty support structure capable of
 supporting several  large diameter pipes or ducts.  Based on vent stream data,
 the Agency judged that the flare-system would need a structure capable of
 supporting relatively small  diameter pipes or ducts and the number of these
 pipes or ducts would be few.   Therefore,  the cost for a lighter weight pipe
 rack was included in the cost algorithm instead  of a pipe bridge.
      Flare services (steam,  air,  and natural  gas) were not included in the
 original  cost analysis for the proposed distillation standard because  EPA
 originally thought  that existing  flare  services  would be available.  Upon
 review of the requirements for flares dedicated  to the control  of  distillation
 vent streams,  EPA determined  that  it was  appropriate to include  the capital
 costs for flare  services  because many new flares  would be built  at locations
 remotely  located from existing flare services.   Estimated capital  costs  for
 flare services  include  pipe suspension  and  installation costs.   The  commenter
 included  electrical  services  in the  flare  services  item.   Capital  costs
 associated with  electrical services  are accounted for in  the  installation
 factors built  into  the  costing procedures.  These include a 0.01 electrical
 factor under  installation, a  0.03 contingency factor  under indirect  costs and
 a 0.10 construction  and field  factor under  indirect  costs.  These  installa-
 tion  factors  are meant  to  account for any variability  of  costs beyond the
 actual costs for the  flare itself and are consistent with  engineering cost
 estimation practices.
     The knock-out drum and fluidic seal are included  in  the estimated cost
for the flare system.  Furthermore, EPA believes that these two devices have
been adequately accounted for  in the costing procedure developed at proposal.
                                    2-41

-------
      The Agency has  reviewed  and,  where  appropriate,  revised  costing assump-
 tions for annual  cost  components.   The costing  procedures;  used  at proposal
 did include the calculation of operating labor,  maintenance labor,  and
 overhead,  but the maintenance labor and  overhead calculations were  not
 separate,  easily identified line  items,  making  comparison  of  labor  rates
 difficult.   Since procedures  used  at  proposal were  not well defined,  and
 because  it  is reasonable  to assume supervisory  labor  will  be  needed in
 operating  the flare  system, EPA decided  to  include  supervisory  labor costs
 and to use  an alternative method for  calculating overhead  costs.  Supervisory
 labor cost  is estimated to be 15 percent  of the  operating  labor cost.
 Overhead  labor cost  is calculated  as  80  percent  of  the sum of operating,
 supervisory,  and  maintenance  labor costs.  Maintenance labor  cost is
 calculated  as 3  percent of the  total  installed capital cost of a  new  flare
 system.  Previously, the  cost of maintenance labor  and parts  had  been
 estimated together.  Maintenance parts costs are  now  calculated as  3  percent
 of  total installed capital cost.
      Engineering  and environmental  oversight costs were cited by the
 commenter as  being inadequately represented.  However, EPA believes these
 costs are adequately accounted  for  in its revised costing  procedure.  As
 stated above,  the Agency  has  considered supervisory labor  in  its revised
 costing procedures.  In addition,  the costs of maintenance labor and
 materials to  ensure the proper operation and maintenance of control devices
 are  included  in the costing procedures.   The Agency considers these costs to
 be representative of the costs associated with proper operation and
maintenance of the control devices.  Because any additional labor needs as a
 result of bad weather will be one  time costs and are unpredictable, no
 allowance was made in developing the costs associated with normal  operation.
     The Agency has reviewed flare design criteria to ensure they represent
the safe practices for workers in  facilities complying with the standards.
The maximum ground level  thermal radiation intensity design criterion used at
proposal  (1,200 Btu/hr-ft2 from the flare alone) was determined to be too
high.  A lower maximum ground  level intensity (140 Btu/hr-ft2  from the flare
alone) has been selected.   The EPA agrees that for vent streams having high
heating values and high flow rates, the  revised  maximum thermal  radiation
                                    2-42

-------
  intensity at ground level  lead to a taller flare design.   However,  the cost
  for controlling most distillation vent streams would be unaffected  by this
  change because they typically contribute much  less  than 140 Btu/hr-ft2 at
  ground level  even  when  combusted through the shortest flare commercially
  available (30 ft).
       In  the  cost algorithm used  at  proposal, the pressure  drop  associated
  with  the flare tip  was  at  least  27  in.  water column,  depending  on the vent
  stream characteristics.  However,  in reviewing the  previously constructed
  cost  algorithm and  the  associated flare design criteria, EPA has determined
  that  a pressure  drop  of this  magnitude  is  representative of a typical
  emergency flare  operation,  not a flare  designed  for  low velocity, continuous
  vent  stream such as for a  distillation  operation.  A  flare  tip pressure drop
  that  is more  representative of a dedicated flare designed to handle a con-
  tinuous flow  vent stream was  calculated to be  slightly lower than 0.5 in.
 water column.  The Agency does not believe that the flare tip design pressure
 drop would promote smoking in an operating flare.
      The EPA's revision of the flare design criteria to reflect the velocity
 limitation included in the proposed standard resulted in an increased
 diameter for the flare tip, and,  therefore, a slightly increased capital cost
 for some of the higher flow rate  vent streams.   For  many distillation vent
 streams,- the diameter calculated  according to the revised  design criteria
 will  be less than 2 inches, the smallest available flare diameter.   These
 streams would have  required a  2-inch diameter flare  according to the
 previously used design criteria as well.  For those  distillation vent
 streams,  the  cost of the flare would remain unchanged.   However,  for streams
 with  flow rates sufficiently high  to require larger  flare diameters  using  the
 revised design criteria, the revision will  cause  a slight increase in  capital
 cost.
     Pilot gas and purge gas costs are  included in annual direct operating
 and maintenance costs  and have been  calculated  based  on flare design
 parameters.  The  commenter  cited costs resulting  from  the shortening of flare
 tip life due to higher operating temperatures.  The Agency is aware that
 combustion control equipment is subject  to maintenance, repair,  and replace-
ment.   In the  costing procedures,  maintenance labor and maintenance parts  are
                                    2-43

-------
each  included  as 3 percent of the total installed capital cost of the
combustion system.  The costing procedures also include annualized equipment
capital costs  which are based on a 15-year life for flares.
      The commenters suggested the inclusion of the cost of monitoring
equipment in the cost analysis and TRE equations.  The items suggested
include a TV camera and monitor, a strip chart recorder for temperature
monitoring, and electrical control room monitoring equipment.  While the cost
of the monitoring and recordkeeping requirements has been estimated and
determined to  be reasonable, this cost was not included in the development of
the TRE equations.  For further information, see response to Comment 2.5.1.

      2.5.3  COMMENT:  One commenter (D-9) stated that EPA did not include in
its cost estimates of combustion control systems the cost associated with
continuous recording of temperature measurements for incinerator fireboxes
and for boilers or process heaters.   The commenter indicated that continuous
recording can  be quite costly.  A temperature range of 2,500 - 3,200°F is
common to the  fireboxes of incinerators and boilers.  Typically, platinum/
rhodium thermocouples are required to handle these temperatures, and such
systems commonly cost approximately $15,000 per installation.

      RESPONSE:  The EPA believes that the cost of an adequate temperature/
recording system would cost well below $15,000.  In fact, vendor data show
the equipment  cost could be approximately $4,500 even if a platinum/rhodium
thermocouple were needed (see Docket Entry IV-E-9).  Furthermore, peak
temperatures exceeding 2,800°F are expected to occur for boilers where
temperature monitoring is required while the maximum temperature for an
incinerator firebox is expected to be around 2,000°F.   Therefore, less
expensive materials than the commenter suggests could be used to monitor
incinerator temperatures.  The Administrator has determined that the costs
associated with the monitoring requirements of the final  standards are
reasonable.   However,  these costs were not included in the development of the
TRE equations  (see response to Comment 2.5.1).
                                    2-44

-------
       2.5.4  COMMENT:  Two commenters (D-12 and D-27)  disagreed with the
  assertion in the preamble that annualized costs associated with the disposal
  of sodium chloride from scrubbing incinerator flue gases are insignificant.
  Both commenters stated that disposal  will  be expensive unless the plant is
  located near salt water and can get  a permit to dump  its brine.   One
  commenter (D-27)  indicated that the  operating costs for deep-well  disposal of
  sodium chloride range  from $3  to $6  per  1,000 gallons  of brine.

       RESPONSE:   The  Agency has  conducted  an  investigation  into  the methods of
  brine disposal  available  to  the chemical  industry  (see  Docket Entry IV-B-11).
  This  investigation included  a review  of available  literature  and discussions
  with  State regulatory  agencies,  brine disposal  companies, chemical  manu-
  facturers, and  consultants.  Alternatives available for  brine disposal
  include:
      o    Direct discharge.
      o    Evaporation.
      o    On-site deep well injection.
      o    Off-site deep well injection.
      o    Existing disposal method if the plant is  already disposing of
           brine solutions.
      Of the disposal  methods shown above,  off-site  deep well  injection is
 considered to be the  most expensive and least popular.   The option is
 generally viewed as impractical  by industry representatives due  to its costs.
 Based on available data,  there  appears to  be  no reason  why any facility
 covered by the  distillation NSPS would not  be able  to  select  one  of the other
 lower cost options.   It is assumed that all facilities  will  use  the lowest
 cost  option available.  Therefore,  the Agency has no reason  to believe that
 any distillation  facility  will incur a significant  brine  disposal cost as a
 result  of  this NSPS.

     2.5.5 COMMENT:  One commenter  (D-27) recommended that all costs
discussed  in the preamble be inflated  from 1978 dollars to first quarter 1984
dollars.  This commenter indicated that the cost-effectiveness cutoff would
then become $2,600/Mg, which is  significantly higher than $l,900/Mg.
                                    2-45

-------
      RESPONSE:  The  EPA  agrees that the cost effectiveness of $l,900/Mg
 (December  1978 dollars)  would be about $2,600/Mg in first quarter  1984
 dollars.   However, EPA maintains that this would not change the analysis or
 the requirements of  the  standards.  When the analysis for the distillation
 NSPS  was begun, it was decided that 1978 would be the appropriate  base year
 for costs  because more recent data were not available.  If the implicit price
 deflator for the gross national product is applied, the cost-effectiveness
 cutoff inflates 40 percent over the 5-year period.  However, regardless of
 whether it is expressed  in 1978 or 1984 dollars, the cost-effectiveness
 cutoff has the same  impact.  If the cost-effectiveness cutoff is increased by
 40 percent by an inflation factor to $2,600/Mg, the reference cost
 effectiveness will also  increase by 40 percent, since both values  are
 calculated using the same cost assumptions.  Furthermore,,  in considering an
 inflated TRE cutoff, it  should also be realized that the value of  the
 benefits associated with the standards are also inflated accordingly.  Thus,
 the ratio will remain the same, and the TRE index cutoff value will still be
 1.0.  Inflation does not affect the validity of the TRE index.  Thus, there
 is no need to revise the cost-effectiveness cutoff.

     2.5.6  COMMENT:  Two commenters (D-7 and D-25) noted that EPA did not
 adequately address the capital  costs of precautionary safety measures
 necessary when introducing a vent stream into a boiler or process heater as a
 fuel.   According to one commenter,  the NFPA may require safety closure valves
 and vent systems,  as well as appropriate containment when  vent streams are
 combusted as fuel.  These safety measures would increase the capital cost of
 using a boiler or process heater to combust vent streams.   The other
 commenter indicated that to prevent explosions, liquid hydrocarbons must be
 kept out of the vent streams going  to these units.   The extra cost for surge
 tanks, knock-out drums and extra control  instruments would cancel  out most of
 the economic benefits associated with using the vent stream as a fuel.

     RESPONSE:  Boilers and process heaters can sometimes  be attractive
candidates for the control  of vent  streams from distillation facilities
because they can provide at least 98 percent reduction efficiency of VOC and
nearly the complete recovery of the heat content of a vent stream.  There are
                                    2-46

-------
  three plants in the available data that use boilers as a control device on
  vent streams with heating values ranging from 449 to 1,258 Btu/scf.  Because
  these streams have high energy recovery potential,  there is an economic
  incentive to use boilers in order to recover heat energy.
       The EPA did not include boilers or process  heaters as control  devices in
  the regulatory analysis and as such,  specific boiler costs such as  the ones
  outlined by the commenter do not play a role in  analyzing  cost impacts
  associated  with the standards.   Since flares and incinerators  are more widely
  applicable  for VOC  control  than  are  boilers  or process  heaters,  EPA selected
  flares  and  incinerators  as  the control  techniques upon  which the standards
  are  based and  estimated  the impacts  associated with  their  application.
  Because  the  impacts  associated with the  use  of flares and  thermal
  incinerators were found  to  be reasonable for  all distillation  operations that
 would be required to reduce VOC emissions by  98 weight-percent or to 20 ppmv
 according to the standards, there was no need  to evaluate the  impacts
 associated with other less expensive control techniques.

      2.5.7  COMMENT:  One commenter (D-25)  stated EPA did not include the
 cost of more expensive compressor and pipeline systems in its cost  estimates
 for combustion control.   It was noted that  for each  of the  combustion control
 options considered,  EPA assumes in its cost equations that  the  same  type of
 pipeline system is used  to recover vent gases and deliver them  to the com-
 bustion device.   This approach may be both  impractical  and  unsafe for some
 vent  streams containing  potentially explosive compounds, or compounds which
 present  erosion  problems  during liquefaction.   Such  compounds (e.g.,
 chlorotrifluoroethylene)  are not  necessarily  included in the list of products
 to  be regulated,  but  may  be  included  in  the emission  streams containing
 compounds that are listed.

      RESPONSE:   The commenter noted that  the  vent stream mover  system
 included  in the  costing procedures at proposal would present erosion problems
due to liquefaction.  The EPA believes that the relatively low pressure drops
expected for flare and incinerator mover systems [i.e., 3.0  kPa (12  in. w.c.)
                                    2-47

-------
 for  an  incinerator  and  5.2  kPa  (21  in. w.c.) for a flare] would not  cause
 significant  liquefaction problems with pipeline components.  Thus, erosion
 problems would be minimal.  However, EPA is aware that minor problems with
 erosion, corrosion, and the like over an extended period of time may cause
 maintenance  expenses.  To account for these costs, the costing procedures for
 flares  and incinerators include maintenance labor at 3 percent and maintenance
 parts at 3 percent  of the total installed capital cost of the combustion
 system.  In  addition, since the combustion systems will eventually have to be
 replaced, annualized capital costs are calculated based on reasonable
 expected system lifetimes.  The costing procedures use a a 10-year life for
 incinerators and a  15-year life for flares.  It should be noted that
 incinerators are used for combusting more corrosive (i.e., halogenated)
 streams.
     The design of  flare systems include various safeguards against
 explosion.  These include knockout drums, water seals, fluidic seals, and
 purge gas systems.  Major equipment purchase costs for incinerators  include
 equipment for fire  protection.  Furthermore, the capital costing for com-
 bustion systems includes a contingency factor to address site-specific needs
 for special combustion system design.  The unique requirements of individual
 distillation units  are expected to be addressed by the contingency factor
 allowance.

     2.5.8  COMMENT:  One commenter (D-32)  stated that the BID description of
 price increases that might follow from the NSPS is deceptive because EPA
 assumed that control costs will automatically pass through to consumers.  The
 competitive environment of the chemical  industry,  the commenter continued,
makes pass-through opportunities rare.   As  evidence,  the commenter cited the
 experience of producers who are unable to pass Superfund costs through to
 consumers.   Therefore,  the commenter concluded, EPA should emphasize the
 separation of costs and prices, and should  modify the analysis of rolled-
through costs,  particularly the statement on p. 8-39 of the BID that reads
 "producers of intermediates will roll  through the entire cost of control to
other SOCMI producers."
                                    2-48

-------
      RESPONSE:  Standards may affect prices,  profits,  production levels,
 capacity utilization,  construction of new facilities,  employment, foreign
 trade,  and a myriad of other economic variables.   Confronted with so many
 variables, and with standards that will  affect the production of 211
 chemicals, EPA focuses its analysis on projecting worst-case, or reasonably
 worst-case price effects.  This focus simplifies  the economic analysis and
 yet enables EPA to be  sure a standard will  not have any significant,
 unexpected, harmful  economic effects.  However, a worst-case analysis does
 not lead to predictions about most probable effects.  In the case in point,
 an  effort was made to  find the upper limit  of price increases that could
 result  from this distillation NSPS.   Finding  the  upper limit is  quite small,
 less than 5 percent, EPA went on to analyze other variables.  No effort was
 made to predict most likely short- or long-term price  effects.
      The EPA did not intend to convey the  impression that all control  costs
 are rolled along a chemical  process  chain and then passed along  to consumers.
 Some are,  but,  especially in the short run, some  are shared  by stockholders,
 taxpayers,  and suppliers.   The commenter's  finding that some firms cannot
 pass through Superfund  costs is  not  applicable to an NSPS.   Superfund  taxes
 are assessed to existing  operations.   The NSPS costs apply only  to future,
 new facilities,  and to  some  existing  facilities after  they are modified or
 reconstructed.   It is not in the economic interest of  a firm to  build,  to
 modify,  or to  reconstruct  facilities  until  and unless  a reasonable return  can
 be  expected  on  the investment,  including the  investment in pollution control.
 Thus, in  the very  long  run,  when  all  facilities come under the standard, all
 control  costs  probably  will  be passed  through.  However,  it  should be
 remembered  that  the modeling  assumptions EPA  used  to compute price increases
 are  very conservative.   This  means that the actual  long-term price increases
 should  be  less  than EPA's very conservative estimates.
     The quotation given by  the commenter is  incomplete.  The full  sentence
 from p. 8-39 of  the BID is "The analysis here  assumes that all facilities  --
existing as well as the new  -- are subject to  the  standards,   and that
producers of intermediates will roll through the entire costs of control to
other SOCMI producers."  These are two conservative  assumptions in the  sense
                                    2-49

-------
 that  they  lead  to  an  overestimation of  price  increases.  Clearly,  the
 distillation  NSPS  does  not  apply to the majority of existing  facilities.   It
 is  also  clear that all  producers will not roll through all costs  in  the  first
 5 years  of the  standard.   (Five years is the  period of analysis used in  most
 studies  of this nature.)  To replace these and other conservative  assumptions
 with  most-probable assumptions would require  a substantial commitment of
 EPA's time and  resources.   Furthermore, prediction of expected, most-probable
 price changes would not influence the final form of the standard.
      For these  reasons, EPA does not think it is necessary to revise the
 price increase  analysis.

      2.5.9 COMMENT:  One commenter (D-15) stated that EPA should  consider
 the cost of fuel oil  in estimating the  cost of controlling vent streams with
 extremely  low heating values.  It was indicated that some distillation
 facilities may  choose to use fuel oil instead of natural gas to supplement
 the low  heating value of vent streams if flaring were used as a control
 device.  Fuel oil  may be used in some cases because it is relied upon more
 than  natural  gas at some facilities where both fuels are available.  The
 commenter  believes that using fuel  oil  instead of natural gas could  increase
 the cost of controlling VOC emissions.

      RESPONSE;  The Agency  chose to include the most typical  equipment and
 operating  procedures  in estimating the cost of controlling VOC emissions.
 The use of fuel  oil as a supplemental  fuel  for flaring was not considered
 because the Agency believes that the vast majority of producers have access
 to natural   gas  and the vast majority of producers would use natural  gas as a
 supplementary fuel  for flaring.   The costs were developed to  be represen-
 tative of  the costs anticipated to  occur at the majority of facilities in the
 industry.
     The commenter was contacted for clarification of the statement  regarding
the need to consider the cost of fuel  oil.   The commenter indicated  that
natural  gas is  not always available and fuel  oil  is required  for flaring for
 about 10 percent of the operating schedule (Docket Item No.  IV-D-38).
                                    2-50

-------
  However,  because there is a diverse group of chemical  producers it would be
  infeasible to evaluate every process and  producer on  an  individual basis.
  Therefore,  the Agency must use  a  method of estimating  control  costs that are
  typical of the vast  majority of SOCMI  producers.   The  EPA  believes that  it  is
  reasonable to incorporate into  the  flare  costing  algorithm the cost of
  natural gas  as the only  supplemental fuel  for flaring.

  2.6  COST  EFFECTIVENESS

      2.6.1   COMMENT:   Five  commenters  (D-ll, D-20, D-23, D-24,  and  D-27) are
  in favor of  using the  TRE approach  in  the proposed standards.   Reasons given
  include: (a)  it gives  a standardized cost estimate that can serve  as a
  rational basis for developing standards;  (b) it encourages product  recovery
  and considers  economic viability;  and  (c)  it shows a consideration  of the
  large incremental costs associated with diminishing air quality benefits in
 the standards development process.  One commenter (D-2) is in disagreement
 with the use of cost effectiveness as the  sole determinant for which
 facilities have to use add-on control devices.
      Two commenters  (D-21 and D-27)  indicated that the $l,900/Mg
 cost-effectiveness cutoff is unreasonable.  One  commenter (D-21) suggested
 that  cost-effectiveness ratios are not  a valid basis for  not requiring  the
 use of  controls that  EPA's analysis  shows  are affordable.   This commenter
 then  suggested that,  even  if cost  effectiveness  were relevant,  EPA chose  the
 wrong cutoff because:  (1)  it is  below the  cost-effectiveness  level  associated
 with  a  substantial number  of past  standards;  and (2) even  if the cutoff were
 above those  past  figures,  the higher costs  are justifiable  as means to
 protect the  public health  from the potentially hazardous pollutants in  these
 streams.
     One commenter (D-27)  considers  the TRE cutoff as too high  and  stated
that the $l,900/Mg cutoff  is  not adequately justified by the Agency.  The
commenter suggested that a $500/Mg cutoff is more than adequate  to  secure all
meaningful  reductions.  A plot of the data from Table 1 at 48 FR 57545 shows
                                    2-51

-------
a point of sharply diminished VOC emission reduction in the $200 to $600/Mg
range with essentially no additional  reduction in emissions achieved at any
cost above that range.
     Commenter 0-27 stated because EPA has underestimated the total installed
capital cost of control devices by a  factor of 3 and the annual  operating
costs by a factor of 2, the cutoff of $l,900/Mg (1978 dollars) for VOC
abatement is understated.  The actual cutoff figure, as calculated by the
commenter in first quarter 1984 dollars,  it about $5,400/Mg.   The commenter
indicated that this exceeds the $l,000/Mg limit used in most other NSPS.
     Two commenters (D-27 and D-37) asserted that the $l,900/Mg cutoff cannot
be justified based on the presence of toxic constituents in the vent streams
from distillation facilities.  One of the commenters pointed out that control
of toxic pollutants is the objective  of standards developed under Section 112
of the CM (NESHAP) and not standards such as these which are being proposed
pursuant to Section 111 of the Act (NSPS).  The commenter also stated the
preamble does not adequately demonstrate that the presence of toxic pollutants
in the emissions from distillation facilities are sufficiently different from  .
the emissions from other VOC sources  to justify a special consideration of
their hazards.  Concerning VOC emissions generally, the commenter stated that
EPA has decided (48 FR 628) an ambient air standard to protect public health,
etc., from hydrocarbons is unnecessary.

     RESPONSE:  The EPA believes that its decision to consider cost
effectiveness when determining the cutoff for applying the percent reduction
standards reflects a reasonable interpretation of Section 111 of the CAA.  In
analyzing the question of whether the consideration of cost effectiveness is
appropriate, EPA looked to see whether Congress had "directly spoken to the
precise question."  Chevron. U.S.A..  Inc. v NRDC. 467 U.S. 837, 104 S.Ct.
2778, 2782 (1984).  Section  111 requires EPA to promulgate NSPS limiting
emissions to the level that  reflects the best system of emission reduction
"which (taking into consideration the cost of achieving such emission
reduction, any nonair quality health and environmental impact and energy
requirements) the Administrator determines has been adequately demonstrated."
Section lll(a)(l).  Nothing  in either Section 111 or elsewhere in the Act
defines "the cost of achieving such emission reduction."  The plain meaning of
the  phrase, however,  is quite broad.  This indicates that; Congress implicitly

                                     2-52

-------
 delegated to EPA the authority to interpret the phrase to encompass a range of
 impacts, including costs of control in relation to the emission reduction
 achieved.  Further, Congress did not specify any particular manner in which
 EPA was to take these costs "into consideration."  Thus, absent a clear
 Congressional direction to the contrary discernible from the Act's history,
 Chevron, 104 S.Ct. at 2783, Section 111 gives EPA authority to reject NSPS
 control options on the ground that their costs are unreasonably high in light
 of the emission reductions they achieve.  I/
      The EPA has reviewed the legislative history of Section 111 and concluded
 that no contrary intent is discernible.  Most important, the history contains
 no express  repudiation of the use of cost effectiveness as one mechanism in
 considering costs when setting an NSPS.
      For these  reasons,  EPA believes that Congress implicitly delegated the
 Agency the  authority  to decide how best to "take into consideration...  cost-
 in setting  NSPS and,  if the Agency concluded it  was  appropriate,  to  consider
 cost effectiveness.
      Further, in Portland Cement  Association v.  Train.  513 F.2d 506,  508 (D.C.
 C1r-  1975),  cert,  denied.  416  U.  S.  1025  (1975)  ("Portland II"),  the  Court
 stated that  EPA may reject  control  options that  result  in  a  "gross
 disproportion between  achievable  reduction in  emissions  and  cost  of the
 control  technique."  Since  the  purpose  of cost-effectiveness  analysis is  to
 highlight such  disproportion,  this passage supports  EPA's  approach.
      In  selecting  cutoffs related  to applicability of NSPS,  EPA looks at  a
 variety  of factors including:   (1) the  technical feasibility  of additional
 control;  (2) the economic feasibility associated with different control
 alternatives; (3) the magnitude of emission  reductions associated with a
 control  alternative (e.g.,  a slightly higher cutoff could  be  selected if  it
 led to a substantial increase in the emission reduction achieved by the NSPS);
I/  For instance, Congress provided a more specific restriction on the
consideration of costs on Part C of the Act.  Section 169(3) defines "best
available control technology" as "an emissions limitation based upon the
maximum degree of reduction... taking into account energy, environmental, and
economic impacts and other costs, determines is achievable for such
facility..."  (Emphasis added).  Here it is more likely that Congress intended
to ensure the maximum control considering case-by-case economic impacts but
regardless of cost effectiveness.
                                     2-53

-------
(4) the cost effectiveness (C/E)  of the control  alternative in terms of annual
cost per megagram ($/Mg) of emissions reduced;  (5)  the quality of the cost
estimates (e.g., worst case versus realistic estimates);  (6)  potential
reductions in other air pollutants not specifically regulated by the NSPS
resulting from a control alternative; and (7) the location of the sources
(e.g., urban versus rural).  Because these factors  vary from industry to
industry and, in some cases, within the same industry, decisions on the
appropriate level of control are made on a category by category basis.
     In evaluating the above factors, EPA found that the following
considerations were key to the selection of the appropriate cutoff for SOCMI
distillation operations:  (1) the cost effectiveness of NSPS for VOC emissions
previously promulgated by the EPA; (2) the fact that distillation vent streams
contain compounds that are considered potentially toxic by EPA and that many
of the facilities are located in urban areas; and (3) the likelihood that
these maximum costs will not be incurred by industry.
     A survey of the VOC standards for other source categories shows that the
cost effectiveness of those control requirements has sometimes ranged as high
as $2,000/Mg.   (See Docket Item No. IV-B-17.)  The Agency's experience in
implementing these standards reveals that NSPS requiring this level of control
have proved a useful tool in bringing about the installation of much emissions
control technology, significant reductions in emissions and corresponding
improvements in air quality, yet have not imposed costs that appear "grossly
disproportionate" to the emission reduction achieved.  Portland II. 513 F.2d
at 508.  Such an approach simply makes this NSPS consistent (as to dollars
spent per metric ton of VOC removed) with the existing body of NSPS
regulations, all of which have either been promulgated without legal challenge
or have been judicially upheld.
     EPA also considered evidence that distillation streams include compounds
that may be toxic. 2/  Although that evidence has not yet resulted 1n a
2/  The Agency has adequately documented that this is the case.   (See Wehrum,
W. et a!.,  "Air Toxics Emission Patterns and Trends", Docket Item No. IV-A3,
and Registry of Toxic Effects of Chemical Substances, Docket Item
No. IV-J-9).  Moreover,  it  is apparent that combustion of those  streams will
reduce those compounds proportionately.  (See, e.g., "Thermal  Incinerator
Performance for NSPS", Docket Item No. II-B-3.)  The Agency received no
comment questioning this documentation.

                                      2-54

-------
determination that those compounds should be listed as hazardous under
Section 112, EPA considered this potential toxicity along with other relevant
factors when choosing the cutoff.  As stated in EPA's Air Toxic Strategy
published in July 1985, the Agency will consider the likely toxic pollutant
control benefits in the course of carrying out its responsibilities under
Section 111.  This strategy reduces emissions of potentially toxic compounds
from new sources and from industries as their facilities are reconstructed or
modified.  This approach achieves significant reductions in these compounds of
concern while the Agency evaluates them for regulation under Section 112.  The
Agency disagrees with the argument that EPA has no authority to do this.  The
EPA is not attempting here to regulate streams based on a decision that they
contain hazardous air pollutants within the meaning of Section 112.  Rather,
the Agency is simply considering all available evidence within the framework
of Section 111.  Section 111 does not attempt to restrict EPA's discretion to
consider all relevant factors in making that decision, and certainly the
potential toxicity of a stream is relevant to the control requirement
selected.  Many SOCMI facilities are located in urban areas and, as a result,
many people will be exposed to any hazardous air pollutants emitted from these
facilities.
     A third consideration in setting the cutoff at $l,900/Mg is the
likelihood that no facility will actually have to incur the costs implied by
that cutoff.  The reasons are:  (a) less expensive control systems may be
used, thus reducing the costs and cost effectiveness incurred by individual
facilities; (b) the cost estimates for thermal incinerators and natural gas
prices are overstated; and (c) the inherent flexibility within the regulation
encourages the use of product recovery modifications that will significantly
reduce the cost incurred by individual facilities that may have otherwise had
to add a combustion device.  The regulatory analysis assumes that each
distillation operation process vent would have its own combustion device and
would need separate ducting and support structures.  It is expected, however,
that some operations will share control systems with other process vents.  The
analysis also assumes that incinerators or flares will be used to reduce VOC
emissions by 98 weight-percent.   However, many facilities will opt to use
boilers, process heaters or catalytic oxidizers.   When these devices are used,
the cost of control  will be significantly reduced over the cost of thermal
incineration.   Data on current capital costs of thermal  incinerators indicate
                                     2-55

-------
Incinerators Indicate that units are now available at substantially reduced
costs compared to the costs used In developing these standards.   Lower capital
costs would reduce the annualized costs estimates, also,  but not as
significantly.  This 1s an Important consideration in selecting  the
appropriate cost-effectiveness  cutoff.   Another consideration is the fact that
natural gas prices used to calculate the cost effectiveness for  each stream
are overstated by about 40 percent, even though they were updated after
proposal (see following section on "Costing Revisions").   These  conservative
assumptions have resulted in higher cost and cost-effectiveness  estimates than
will actually occur.  Finally,  the standard encourages pollution prevention by
not requiring 98 weight-percent reduction if a TRE index  greater than 1.0 is
maintained.  The EPA believes that many facilities having a TRE  index just
below the 1.0 cutoff (equivalent to $l,900/Mg) will upgrade product recovery
to reduce VOC and raise their TRE index above 1.0.  This  will significantly
reduce the cost of control incurred by the industry while reducing emissions
and will also minimize the national energy impacts.  A preliminary examination
of the national statistical profile shows that because many facilities have
the potential to reduce VOC emissions sufficiently to raise their TRE values
above 1.0, the highest cost effectiveness that a facility will actually incur
as a result of installing a combustion device is estimated to be approximately
$l,400/Mg.
     The EPA believes that this process reflects a reasoned interpretation of
the phrase "taking into consideration the cost of achieving such emission
reduction," especially given the lack of clear Congressional guidance.  The
commenters' arguments that EPA should have selected either a higher cutoff to
provide for a greater degree of protection of the public  health, or a lower
cutoff because most VOC standards have lower costs in relation to the
resulting emission reduction, fail to provide a more reasoned methodology for
selecting the appropriate level.   Instead, they merely reflect each of the
competing goals reflected in Section Ill's history, as described above.
     Consideration of all of the above factors confirmed EPA's belief that a
TRE value of  1.0  (i.e., $l,900/Mg) represents an appropriate cutoff for
determining which facilities must  reduce VOC emissions by 98 weight-percent or
to 20 ppmv.  The  cutoff is specific to the SOCMI distillation operations
source category and would not necessarily be appropriate for other source
categories; therefore, it should not be viewed as a benchmark for other
                                     2-56

-------
standards.  It is not surprising, however, that the cost-effectiveness cutoff
for distillation would be the same as for the air oxidation NSPS for a number
of reasons.  First, the same pollutant, (i.e., VOC) is being regulated.
Second, the same general class of VOC emitters (i.e., SOCMI) and similar types
of process equipment (e.g., recovery equipment) are affected.  Third, the same
types of control techniques are applicable.
     A commenter was concerned that because EPA has underestimated the total
installed capital cost and the annual operating costs of control devices, the
TRE cutoff is underestimated.  As a result of this and other comments, the
Agency performed a complete review of all costing procedures.  Based on that
analysis, several changes to the costing methodology were made and presented
in a supplemental Federal Register notice (50 FR 20446) on May 16, 1985.
Changes made to costing procedures were based on information presented in two
memoranda entitled "Revisions to the Incinerator Costing Algorithm" (Docket
                                    2-56a

-------
Entry  IV-B-7), and "Revisions to the Flare Costing Algorithm"" (Docket Entry
IV-B-8).  The commenter is referred to Section 2.5.2 for a discussion of the
costing methodology.

     2.6.2  COMMENT:  Two commenters (D-21 and D-27) indicated that EPA had
used the same $l,900/Mg cutoff in the recently proposed air oxidation NSPS.
They suggested this figure results from an EPA policy decision not from a
specific evaluation of the costs and impacts associated with the distillation
NSPS.

     RESPONSE:  The Agency has based the $l,900/Mg cutoff on a specific
evaluation of the costs and impacts associated with the distillation NSPS.
It is not surprising however that the cost-effectiveness cutoff for
distillation would be the same as for the air oxidation NSPS for a number of
reasons.  First, the same pollutant (i.e., VOC) is being regulated.  Second,
the same general class of VOC emitters (i.e, SOCMI) and similar types of
process equipment (e.g., product recovery) are affected.  Third, the same
types of control techniques are applicable.

     2.6.3  COMMENT:  One commenter (D-24) requested information on how the
TRE equation was derived.  The commenter stated that the derivation of the
TRE equation coefficients is not explained in an understandable manner.
     The commenter also requested clarification on when the TRE index value
would need to be recalculated.  The commenter presented two suggestions
regarding when the TRE index value should be recalculated.   Section 60.664(d)
of the regulation states:  "Each owner or operator of an affected facility
shall recalculate the TRE index value for that affected facility whenever
changes are made in production capacity,  feedstock type, or catalyst type, or
whenever there is replacement, removal,  or addition of product recovery
equipment."  One of the suggestions made by the commenter is that recalcula-
tions of TRE index values should be done as part of the permit process only
in situations where increases in emissions or changes in chemical constituents
                                    2-57

-------
  are likely to occur.   The other recommendation  presented by the commenter is
  that an  exemption from monitoring or recordkeeping  requirements should be
  allowed,  perhaps  if the TRE exceeds  10.0.

       RESPONSE:  A description  of the development of the  TRE index  equations
  is  presented  in Appendix B of  this document.  An explanation of the
  derivation  of the TRE  equation  coefficients is  presented in Docket Item  No.
  IV-B-15.
       Since  proposal of the standards, the Agency has revised the TRE equation
  and  derived new coefficients for  the  equation.  These revisions resulted  from
  changes in  the costing  procedures.  The results of the Agency's reanalysis of
  the  TRE equations, coefficients,  and  costing procedures  are discussed  in  the
  supplemental notice reopening the public comment period  (50 FR 20446).
      The EPA requires that the TRE be recalculated for the changes listed in
 Section 60.664(d)  because  the changes have the potential  to lower the TRE
  index below 1.0 indicating the cost of control for that facility is below
 $l,900/Mg of VOC removed.  In some cases it may be possible to  make process
 changes that would require the recalculation of a TRE index but would not
 come.under the permit review process.  Therefore,  it would be improper to use
 the  permit review  process as a trigger for initiating this determination  as
 suggested by the commenter.  In some  cases it  may  be possible to make changes
 in the process which could result in  a TRE index less than 1.0,  but which
 would not  come under the permit review process.  Therefore,  EPA believes  that
 it is necessary to require a TRE calculation as  a result  of the changes
 listed in  the  regulation in order to  ensure  that the  standards  are  being  met
 by all  facilities.
      The commenter suggested  that  EPA exempt affected facilities  showing  a
 TRE  index  value greater than  10.0  from monitoring or  recordkeeping  require-
 ments.  Several changes  were made  in  the regulation to provide  for  inclusion
 of a  maximum TRE index  value above which monitoring and recordkeeping
 requirements would not  be  imposed  on a facility  attempting to comply with  the
 standards.   It is the judgment of the Agency that facilities with TRE index
values above 8.0 would most likely not be able to make process changes that
would cause the TRE index value to fall below the cutoff.   Thus, the Agency
                                    2-58

-------
 believes  that  the  monitoring  and  recordkeeping  burden  should  not be imposed
 on  such facilities.   However,  if  a  process  change  occurs,  the facility would
 have  to recalculate  the  TRE index value  as  required  in Section 60.665(f)(l)
 to  determine whether the value remains above the TRE maximum.   Sections
 60.664 and  60.665  of the regulation have been amended  to  incorporate the
 requirements associated  with  the  maximum TRE index value.,  The basis for  the
 determination  of the 8.0 TRE  index value is discussed  in  Docket  Item No.
 IV-B-14.

      2.6.4  COMMENT:  One commenter (D-27)  requested clarification  on
 determining removal  efficiency for a recovery device when  vents  that are
 affected  and vents that  are unaffected by the proposed  standards  are routed
 through the same recovery device.  The commenter presented an  example  where
 the vent  streams from an  affected distillation column  and  a distillation
 column not covered by the standards are combined and fed to a  single con-
 denser.   The commenter stated that for this example two different TRE  values
 could be  calculated  depending on the assumption used in estimating  the
 removal efficiency of the condenser.  One TRE value  is  based on the condenser
 removal efficiency for all of the VOC removed from both vent streams  (Case
 1).  The  other value  is  based only on the removal efficiency associated with
 VOC present in the vent  stream affected by the proposed standards (Case 2).
 For Case  1, the TRE  value calculated with the condenser removal efficiency
 based on  the combined stream was found to be below the $l,900/Mg cutoff.  For
 Case 2, however, the TRE  value calculated with the condenser removal
 efficiency based on only  the affected vent was found to be above $l,900/Mg.
The commenter is unsure of the correct procedure for estimating removal
efficiency but suggested  that the efficiency be based on only the affected
vent.
     The commenter indicated the designation of affected facility that
 includes all distillation units vented to a common recovery device  as one
affected facility would eliminate the confusion over which removal  efficiency
to use for the TRE calculation.
                                    2-59

-------
      RESPONSE:  With the change in designation of affected facility  (see
 comment 2.2.1) there will be no need to use the removal efficiency for a
 product recovery device in calculating the TRE when an affected distillation
 vent stream is combined with existing distillation vent streams.  As a result
 of the change in designation, all  of the distillation units vented into a
 common product recovery device will be the affected facility.  To calculate a
 TRE index value the vent stream characteristics of the total  flow exiting the
 final  recovery device will  be used and no apportioning of removal efficiency
 to the affected vent stream will be necessary.

      2.6.5  COMMENT;  One commenter (D-30) believes that the  TRE index
 calculation contains no provision  for the high efficiency of  product recovery
 devices used in large distillation facilities.  It was pointed out that
 generally,  larger facilities operate more efficiently than smaller facilities
 in the production of synthetic organic chemicals  and subsequent recovery  of
 organics from process vent  streams.   The commenter stated that even though a
 large  facility is more efficient than a small  one,  the TRE index calculation
 would  allow a small  facility to be exempt from coverage by the standards,
 while  a larger facility would not  be exempt  only  because of its larger vent
 stream flow rate  and higher  VOC emission rate.  Therefore,  the commenter
 suggested  that a  factor be  included  into the TRE  index calculation that
 relates total  facility  production  rate  to  the  final  vent VOC  emission rate.

     RESPONSE:  The  TRE equation was  developed  to determine if the cost
 effectiveness  of  reducing VOC emissions  is reasonable  for  a particular
 facility regardless  of  size.   No facility will  have  to spend  more  than
 $l,900/Mg to meet the emission  reduction requirements.   The TRE equation
 estimates the  total  cost of constructing, operating  and  maintaining com-
 bustion equipment sized according  to  the flow rate and heating  value  of the
 vent stream from the facility.  In general,  vent streams from  facilities
 using highly efficient  product recovery would have relatively  low  VOC
 concentrations.  These  types of streams are expensive  to control because
 substantial amounts of supplemental fuel must be added prior to combustion.
This fact is taken into account in  the TRE calculation.  Conversely, vent
                                    2-60

-------
 streams  emanating  from  less  efficient product recovery devices  usually  have
 higher VOC  concentrations.   In these cases, combustion control  is  more  cost
 effective and  likely  could be accomplished for less than $l,900/Mg.   There-
 fore, it is not  the size  of  the facility or the vent stream that
 predominantly  determines  which streams are required to be combusted  but
 rather the  vent  stream  VOC concentration (and associated heating value).

 2.7  FORMAT OF THE STANDARDS

     2.7.1   COMMENT:  Eight  commenters (D-5, D-ll, D-12, D-13,  D-17,  D-20,
 D-22, and D-27)  disagreed with the regulation of VOC emissions  by  controlling
 TOC emissions  less methane and ethane.  These commenters stated that  the
 standards should not  list methane and ethane as the only chemicals to be
 subtracted  from  TOC emissions'.  Instead, they suggested that the standards
 should allow for all  compounds listed in 48 FR 57542 to be subtracted,  if
 present  in  the vent stream,  from TOC emissions.  They stated that  the com-
 pounds listed  in 48 FR  57542 have been determined by the Administrator  to be
 negligibly  photochemically reactive.
     One commenter (D-5)  stated that it is not proper to subject distillation
 facilities  that  emit  negligibly photochemically reactive compounds to
 standards specifically  designed to limit the emissions of VOC which are
 photochemically reactive  compounds and which contribute to ozone formation.
     One commenter (D-27) stated that TOC's (minus methane and ethane)  are
 not the best demonstrated surrogate to regulate VOC.  The commenter indicated
 that Reference Method 18  gives specific compound identification and measure-
 ment capabilities.  Thus, the compounds listed in 48 FR 157542 can  be
 accurately measured and subtracted from TOC emissions.   Therefore, the
 commenter requested that  EPA regulate VOC emissions directly and allow the
 subtraction  of negligibly photochemically reactive compounds from TOC
 emissions.    In order  for  VOC emissions to be directly regulated the
 commenters suggested that the phrase: "less those compounds that are not VOC
 as determined by the Administrator" be used whenever the phrase: "less
methane and  ethane" is  used in the standards.
                                    2-61

-------
      RESPONSE:  The NSPS for SOCMI distillation operations are intended to
 cover distillation facilities that emit VOC (i.e.,  compounds which
 participate in atmospheric photochemical  reactions  to produce ozone).  Since
 compounds with negligible photochemical reactivity  do not appreciably con-
 tribute to the production of ozone,  the Agency believes that it is
 appropriate to exclude these compounds in determining a TRE index.  The owner
 or operator of an affected facility  should subtract the negligibly photo-
 chemically reactive organic compounds  from the TOC, when quantifying the
 hourly emissions rate for input into the  TRE equation.   For example, if the
 vent stream of a facility contains 90  percent negligibly reactive organic
 compounds and 10 percent reactive  organic compounds only 10 percent of the
 organic compounds emitted from that  facility would  be considered  for
 calculating a TRE index.   Although subtraction of negligibly reactive organic
 compounds is permitted,  it is  expected that  no significant  change in national
 impacts will  occur since less  than 5 percent of the distillation  vent streams
 represented in the available emissions data  contain these compounds.
      However,  subtraction of negligibly reactive compounds  applies to hourly
 emission rate  only and when  determining the  vent stream flow rate and heating
 value of a  stream,  these  compounds must not  be ignored.  The TRE  value
 incorporates  the cost of supplemental  fuel and if any of these  compounds  have
 a  heating value  associated with them they must be included  to avoid  over-
 estimating  supplemental  fuel costs.  The TRE value  also  incorporates  the  vent
 stream  flow rate because  the size  of the total  flow rate entering  the
 combustion  device  influences the cost  of control.   The  larger the  total flow
 rate  the  greater it will  cost to construct, operate  and maintain  a combustion
 device.   Therefore, to properly evaluate the cost effectiveness for
 controlling a  particular  vent stream, the total  flow  rate and total net
 heating value  are needed  for the TRE calculation.
     To allow  for subtraction of compounds with negligible photochemical
 reactivity  in  calculating a TRE index,  the definition of TOC in Section 60.661
 has been modified.  The new definition  indicates that "Total Organic Compound"
means those compounds measured according to the procedures in Section 60.664.
 For the purpose of determining the molar composition as required in
Section 60.664(c)(l)(i)  and the hourly  TOC emissions rate as required in
                                    2-62

-------
Section 60.664(c)(5), Section 60.664(c)(7), and Section 60.665(g)(4), the
definition of TOC excludes or allows the subtraction of those compounds which
the Administrator has determined do not participate in photochemical reactions
to produce ozone.  The compounds to be subtracted are identified in EPA
statements on ozone abatement policy for SIP revisions (42 FR 35314;
44 FR 32042; 45 FR 32424; 45 FR 48941)."  These compounds are methane;
ethane; 1,1,1-trichloroethane; methylene chloride; trichlorofluoromethane;
dichlorodifluoromethane; chlorodifluoromethane; trifluorornethane; trichloro-
trifluoroethane; dichlorotetrafluoroethane; and chloropentafluoroethane.  An
appendix to this document contains the complete copies of these notices.
     Combustion devices destroy TOC by at least as great an efficiency as
when only VOC are in the stream.  Furthermore, it is less costly and less
complex not to subtract the negligibly reactive compounds during performance
testing.  Therefore, these compounds should be included in determining the
removal efficiency of thermal incinerators (performance te>sts).

     2.7.2  COMMENT:  Two commenters (D-8 and D-29) stated that EPA has given
no method in the proposed standards for determining that the organic
compounds listed in the preamble are not photochemically reactive organics.
The commenters suggested that the owner or operator of an affected facility
should be allowed to demonstrate that the particular organic compounds(s)
involved in a given operation are not photochemically reactive organics.  One
commenter (D-8) further suggested that this demonstration could be submitted
with the notification of initial startup required by Section 60.7(a)(3).

     RESPONSE:  There are no EPA-approved procedures on determining the
photochemical reactivity of organic compounds.  The EPA established the
current list of negligibly photochemically reactive compounds through a broad
research effort.  During this effort, EPA has not developed a standard
procedure by which anyone outside the Agency could demonstrate an organic
compound to be of negligible photochemical reactivity.
     As indicated in the response to comment 2.7.1, the Agency has agreed to
allow for the subtraction of negligibly photochemically reactive compounds
from the hourly emissions rate of TOC in determining a TRE index.  The only
                                    2-63

-------
   negligibly  reactive  compounds  that  will  be allowed  to  be  subtracted are
   identified  in  EPA  statements on  ozone  abatement  policy for  SIP revisions
   (42  FR 35314,  44 FR  32042; 45  FR 32424;  45 FR 48941) along  with the rationale
   for  determining that they are  negligibly reactive.  These citations are also
   given in the definition of "TOC" in the  regulation, and copies  of the complete
   notices are given  in Appendix A  to this  document.

       2.7.3  COMMENT:   In Section 60.662(a) the regulation specifies that the
  98 weight-percent reduction requirement  and the 20 ppmv concentration limit
  must be  met  on a dry basis.   One commenter (D-12) suggested that the low flow
  exemption  of 0.008  nT/min  should also be on a dry basis.
                                               a

       RESEQNS£:   The EPA  will  continue  to require  the flow  rate measurement
  for  the low  flow exemption to be  based  upon all of the  vent  stream
  components,  including water vapor.   Because the flow rate  used in the TRE
  equation includes the concentration  of  water  vapor and  because the low flow
  exemption was developed  from flow rate  data measured on  a  "wet basis," it is
  appropriate  for the low  flow exemption  to be  consistent with the TRE^equation
  and data.

 2.8  MODIFICATION/RECONSTRUCTION

      2.8.1   COMMENT.:  Four  commenters (D-8,  D-ll,  D-15,  and D-22) recommended
 a wording change in  Section  60.660(b) to indicate  that affected facilities
 beginning modification or reconstruction after the date  of  proposal  would be
 covered by  the  standards.  This  addition was recommended because of  a dis-
 crepancy between the preamble  and  the regulation.   The preamble (48  FR 57549)
 states:  "The proposed standards would apply to all affected  facilities,
 which  commenced  construction, reconstruction.  Or modification after the'date
 of proposal  of the NSPS."  However, Section  60.660(b) of  the  regulation now
 refers only to affected facilities commencing  construction after  the  proposal
date.   Another commenter  (D-12) did not  suggest any wording change  but
requested  clarification on the initial startup date for cases when existing
facilities  become subject  to the  NSPS.
                                    2-64

-------
     RESPONSE;  As stated 1n the General  Provisions*(40 CFR 60.1),  standards
of performance for new stationary sources apply to new, modified,  and
reconstructed facilities.  This statement was reiterated in the preamble to
the proposed standards.  To avoid misinterpretation, EPA has clarified the
wording in Section 60.660(b) of the regulation.

     2.8.2  COMMENT:  One commenter (0-19) Inquired whether a change in the
production of one chemical listed in the  proposed standards to another listed
chemical would result in a modification if emissions do not increase.
     The same commenter requested that Section 60.660 of the regulation
include a statement exempting process improvements from the standards.  The
requested statement is as follows:  the addition or replacement of equipment
for the purpose of process improvement that is accomplished without a capital
expenditure shall not by itself be considered a modification under this
subpart.

     RESPONSE:  Section 60.14 of the General  Provisions of 40 CFR Part 60
defines "modification" for purposes of NSPS generally.  If a change in the
production of one listed chemical to another listed chemical were done with a
resulting increase in emissions then that change could be considered a
modification.  However, if the distillation operation were designed prior to
the applicability date of the standards to accommodate the production of this
different listed product then the change would not be a modification
regardless of changes in emissions.
     The commenter requested that process improvements that are accomplished
without a capital expenditure not be considered as modifications.   An exemp-
tion similar to what the commenter requested was included in the standards of
performance for equipment leaks in the SOCMI (48 PR 48328).  However, as
discussed in the preamble when those standards were proposed (46 FR 1139), the
reason for the exemption was that routine changes and additions of fugitive
emission sources in an existing SOCMI process unit could result in
                                     2-65

-------
   the  unit's  being  modified and,  therefore,  subject  to  the  standards of
   performance.   In  this  situation,  the  addition  of a small  fugitive emission
   source to the  process  unit would  result  in all of  the  fugitive  emission
   sources within the entire process unit being covered by the standards.
   However, under the standards of performance for SOCMI  distillation  opera-
   tions, if a change or  addition is made for process  improvement  reasons to an
   existing distillation  facility that results in an  increase in VOC emissions,
   only that facility is  considered modified and subject to the standards.  The
   impacts associated with covering this type of situation have been analyzed
  and determined by the Administrator to be reasonable.   Therefore, process
  improvements will  not be exempted  from the standards.

       2.8.3   COMMENT:   One commenter  (D-24)  stated that  it  may  not be
  technically  or  economically feasible  to meet the standards  for a f«w
  modifications of existing facilities.  In  these circumstances,  adequate space
  or  land may  not be available for construction of the combustion  or recovery
  equipment which may be  required to meet the NSPS.

      RESPONSE:  Before  any modifications are begun,   the owner or operator
  should plan for the possibility of complying with these standards  and he
 should consider such things as space limitations when planning construction
 of new facilities  or the modification or reconstruction of existing
 facilities.   The EPA believes  these requirements are reasonable because it is
 possible that the  owner or operator of an  affected facility can comply with
 the  standards by showing a cost  of  VOC  control to be above  the  established
 cutoff.

     2.8.4  CQMMENI:   One  commenter (D-12)  indicated that Chapter 5 of the
 BID  does not  address the subject of modification and reconstruction with
 specific reference to  distillation facilities.
     The same commenter  requested an explanation of what would constitute a
modification or reconstruction for an existing distillation facility  and how
the affected facility would be expected to initiate its compliance obliga-
                                    2-66

-------
tions.  The commenter also requested that EPA distinguish between the General
Provisions of the proposed standards that apply to new facilities and those
that apply to modified or reconstructed facilities.

     RESPONSE:  The EPA acknowledges that Chapter 5 of the BID does not
address the subject of modification and reconstruction with specific
reference to distillation facilities.  An errata sheet has been prepared and
sent to all individuals who received the incorrect copy of Chapter 5 of the
BID.
     As stated in the preamble, the proposed standards would apply to all
affected facilities which commenced construction, reconstruction, or
modification after the date of proposal of the NSPS.  According to the
definition of modification given in 40 CFR 60.14, a modified distillation
facility would occur when any physical change in, or changf in the method of
operation of, an existing distillation facility increases the amount of VOC
emitted into the atmosphere by that facility, or results  in the emission of
VOC not previously emitted.  Some examples of a potential modification
include:  (a) replacement of column  internals (e.g., trays, packing); (b)
replacement of column accessories  (e.g., reboiler, condenser, vacuum
systems);  (c) feedstock or catalyst changes; and  (d) equipment changes for
the purpose of energy conservation.   Exceptions to the definition  of
modification  are presented in paragraph  (e)  of Section 60.14.  A
reconstructed distillation facility,  based on the definition  of
reconstruction in  40  CFR  60.15, occurs when  components of an  existing
distillation  facility are  replaced  to such an extent that;   (1)  the  fixed
capital  cost  of  the  new components  exceeds 50 percent  of the  fixed capital
cost  that would  be required to  construct a comparable  entirely new facility,
and (2)  it is technologically  and  economically  feasible  to meet  the
applicable standards. Any of the previous examples  of  a  potential
modification  could also  represent  a potential  reconstruction  if  these  two
criteria are  met.
      An existing distillation  facility that  undergoes  a  modification or
reconstruction would be  expected to initiate its compliance  obligations by
                                     2-67

-------
   notifying  the Administrator as  required  by 40 CFR 60.7 and by completing a
   performance  test  as  required by 40  CFR 60.8.

        2.8.5   COMMENT:  One commenter (D-33)  requested  that  the replacement  of
   trays and packing in distillation columns  not  be  considered a reconstruction
   when the cost of replacement exceeds 50 percent of the  facility replacement
   cost and no  increase in emissions results.  The commenter  indicated that
   custom fabricated trays and packing are subject to wear, corrosion, fouling,
   and damage and can be very expensive to replace.

       RESPONSE:  The  EPA states in the BID for the proposed standards that the
  replacement of internals (trays  and  packing) for the  most part involves a low
  percentage  capital cost  relative to  a new facility.   Generally, individual
  trays  and packing  are replaced one at a time in response to maintenance needs
  (e.g., cracked tray).  Since replacement  costs considered for  reconstruction
  are  only accumulated  over a  2-year period,  these  types of replacements  would
  not  likely result  in  a facility  being considered  as reconstructed.   It  is
  possible that  the  internals  replacement cost could approach the 50  percent
  criteria for  large capacity  columns  with a  large number  of  trays if the  trays
  and  all the packing were all replaced at the same  time.   However,
  replacements of this extent would not be regarded  as a routine maintenance or
 repair activity.  Rather, total replacement of  internals would be done as a
 result of a process change either to  increase efficiency or to make a new
 product.

 2.9  MONITORING AND MEASUREMENT METHODS

      2.9.1  COMMENT;   Two commenters  (D-8  and D-15) stated that the
monitoring requirements will  be costly and  cumbersome  for both  the  industry
and EPA.  The  industry will be  burdened by  the  constant vigilance and upkeep
required to maintain the  recording and sensing  devices.   One commenter (D-15)
stated that the monitoring requirements would make  it difficult for  the
Agency to properly evaluate the measurement data provided  by the affected
                                    2-68

-------
facilities because of the large amount of information involved.   The other
commenter (D-8) urged EPA to reduce the monitoring  requirements  to a more
reasonable level.

     RESPONSE;  In establishing the monitoring requirements for  this NSPS,
EPA had to reconcile the need to ensure the effective operation  of control
and product recovery devices with the degree of complexity and economic
burden of the monitoring systems on the industry.   The Agency believes that
it has chosen the most reliable and fair methods of monitoring.   Considering
the complexity of the processes involved and the burden on both  EPA and
industry, the monitoring and reporting methods specified are believed to be
the least burdensome way of providing EPA with reliable information about
changes in combustion device operation that could lead to increased VOC
emissions and to ensure the proper operation and maintenance of product
recovery devices.
     Two monitoring methods were considered for this NSPS.  One was to
require the continuous monitoring of selected parameters at the final product
recovery or control device.  Selected parameters for an incinerator, for
example, would  be the inlet and outlet organics concentrations monitored
continuously.   This procedure would provide a continuous direct indication  of
actual emissions  and control device efficiency.  However, two monitors would
be required in  order to determine the  incoming and outgoing organic concen-
trations.  Therefore, it was decided that this type  of  system would be too
complex,  labor intensive and relatively  expensive even  if only one monitor
were  required on  the outlet.
      The  other monitoring  method considered was to rely on more easily
measured  process  operating parameters  that  could be  related to control device
efficiency.   These  monitored values  could  then  be compared to the values
obtained  during the most recent performance test to  ensure the required  VOC
removal  was  still  being  achieved.   For example, the  owner or  operator of an
affected facility using  an incinerator to  comply with  the standards  could
monitor  the  temperature  of the firebox because  it  has  been  shown  to  have a
                                     2-69

-------
 profound effect on the efficiency of VOC reduction.  Changes in temperature
 from the original test could indicate that 98 percent VOC reduction
 efficiency is not being achieved.
      The same approach was investigated and found to be feasible for all
 recovery devices and combustion devices that may be used to comply with these
 standards.  This method has the advantages of a lower cost while still having
 a reasonable reliability compared to the continuous monitoring of organics
 concentrations.   Furthermore,  EPA believes the cost of the monitoring
 requirements to be reasonable  and necessary to ensure the proper operation
 and maintenance of control or  recovery systems used to comply with the
 standards.  Thus,  the monitoring method relying on easily measured parameters
 was selected for this NSPS. Also,  EPA has decided to allow for the use of
 computerized data control  systems to monitor product recovery and combustion
 control  equipment at  a frequency of at least 1 percent of the compliance
 period  (see comment  2.9.12).

      2.9.2  COMMENT:   Two  commenters (D-ll  and D-13)  stated  that monitoring
 requirements  should not  apply to startups,  shutdowns  and  malfunctions.
 Because  the standards do not exempt  affected distillation  facilities  from
 monitoring requirements  during  irregular operations,  the  commenters requested
 this  exemption be  included in the monitoring section  of the  regulation.
 Also, three commenters  (D-5, D-12, and  D-13)  stated that  reporting  and
 recordkeeping requirements are  not appropriate for this NSPS for periods when
 there is no vent stream flow rate such  as during shutdowns and  malfunctions.
 One of the  commenters  (D-12) indicated  that  periods of no  flow  rate into
 combustion  devices wil.l occur often  during shutdowns.  He  believes that the
 semiannual  reporting  required under  Section  60.665(k) should  not  include
 these shutdowns.  Another  commenter  (D-5) stated that the  reporting require-
ments would apply to distillation facilities that normally have  no vent
stream flow rate because of the use  of pressure relief valves on the
accumulator or reflux drums.  There would be a flow rate of the  vent stream
only when the pressure at the relief valve were great enough to cause a
release of the gas.
                                    2-70

-------
     One of the commenters  (D-13) requested clarification on the  requirement
that smokeless flares not have visible emissions (except for periods of no
more than 5 minutes  in 2 consecutive hours).  It is the commenter's
impression that startups, shutdowns, and malfunctions are not covered under
these standards and  monitoring should not be required.
     A request was made by one commenter (D-9) for an allowance of increased
emissions during periods when boilers and process heaters are shutdown.  Such
devices are generally shutdown for periods from 5 to 10 days per year to
conduct safety and operational inspections, and perform preventative
maintenance work.

     RESPONSE:  The  General Provisions (40 FR 60.8(c)), do specify that
emissions in excess  of the level of the applicable emission limit during
periods of startup,  shutdown, and malfunction are not considered a violation
of the applicable emission limit.  This means that emission levels during
these periods are not counted as violations if they exceed the levels
specified in the standards.
     However, monitoring is still necessary during these periods.  The
General  Provisions require under Section 60.11(d) that at all times,
including periods of startup, shutdown, and malfunctions, owners and
operators shall to the extent practicable,  maintain and operate any affected
facility including associated air pollution control equipment in a manner
consistent with good air pollution control  practice for minimizing emissions.
Determination of whether acceptable operating and maintenance procedures are
being used will be based on information available to the Administrator which
may include, but is  not limited to,  monitoring results, opacity observations,
review of operating  and maintenance procedures,  and inspection of the source.
     According to the definitions given in  the General  Provisions, "startups"
and "shutdowns" refer only to the affected  facility and not to control
devices.   However, "malfunctions" are any sudden and unavoidable failure of
the affected facility or control  devices to operate in a normal  manner or
usual  manner.  Therefore,  emissions  beyond  the limit due to scheduled
maintenance work on  boilers or process heaters are not allowed by EPA.
                                    2-71

-------
       2.9.3   COMMENT:  One  commenter  (D-27)  considered  it  unnecessary to
  monitor the  inlet of each  incinerator within an affected  facility  by using  a
  flow meter and continuous  recorder according to Section 60.663(a)(2).  The
  commenter further stated that EPA indicated in the air oxidation NSPS
  preamble (48 FR 48945) that a flow meter and continuous recorder are not
  needed because there is no meaningful relationship between flow rate and VOC
  reduction efficiency.
       Two commenters (D-27 and D-33)  stated that flow indicators are not
  necessary for flares,  boilers and process heaters, as required under para-
  graphs  (b)(2) and (c)(l)  of Section  60.663.   They  indicated that although the
  requirement  is  intended  to  ensure vent  streams  are being  routed for
  destruction,  the  chance  of  a vent stream flowing anywhere  unintended is
  unlikely.   In the case of flares,  one of the commenters (D-33)  felt that  the.
  only way to ensure  flow to  the flare  is  to hard pipe  the flow to the flare
  and only the  flare.  The  other commenter (D-27) indicated  that  vent streams
  must be actively  directed for safety  reasons.  To  avoid explosion hazards,
  the commenter explained, pipes and headers will have  to be  operated at  least
  a little above atmospheric  pressure to prevent air from leaking  in.   Vents
  from these pipes and headers will have to be pressurized and purged  with
 either nitrogen or natural gas to remove oxygen from the vents.  An  oxygen
 monitor will  also be required.  Because the system is under pressure, it is
 unlikely that leaks of VOC to the atmosphere  will  occur.
      One commenter also requested that reporting and recordkeeping
 requirements  for vent stream flow rates  into  combustion devices be deleted
 from  the standards according to the same  reasoning  provided about monitoring
 requirements.  He  added that flow rate monitoring is  not required for the  same
 equipment  in the SOCMI equipment  leaks NSPS and  the air oxidation NSPS.
     One commenter (D-33)  requested that  there be no  requirement to  monitor
 flow of  individual vent streams at points  before the  streams are combined  for
 routing to the flare.  He saw no  need  to  use  monitors  on individual  vent
 stream flow which, at his proposed plant,  could result  in costs  of up to
 $115,000.  The commenter stated that if flow  had to be monitored, EPA should
 require monitoring of the combined flare from all sources and not individual
vents  from all  pipes fed to the flare.
                                    2-72

-------
      RESPONSE:  The  EPA  has  amended the regulation  (Section 60.663(a)(2))  to
 require flow  indicators  instead of flow meters for monitoring  a  distillation
 vent  stream flowing  into an  incinerator.  The Agency has determined  that  flow
 indicators are sufficient for ensuring the vent stream  is; being  routed  for
 destruction and that firebox temperature alone provides an adequate  indication
 of  incinerator performance.
      As stated above, flow indicators and recordkeeping and reporting,  are
 required for  incinerators, as well as boilers, process heaters and flares  in
 order that EPA can be assured vent stream emissions are routed to an  appro-
 priate control device.   The  EPA found that the cost associated with  these
 requirements  is reasonable.  Even though vent streams may be directed to  a
 control device for safety reasons, the Agency requires a demonstration  that
 each  vent stream is directed to a properly functioning control devjce.  In
 the event flow to the combustion device is interrupted, such as  emergency
 venting due to over pressure in the lines or a combustion device shutdown,
 the Agency must have a means of identifying when this occurs and how  often.
 A discussion  of the cost  resulting from monitoring requirements  is presented
 in the response to comment 2.9.1.
      One commenter requested that vent stream flow indicators be required  for
 flares at a point after  all streams have been combined.  The EPA will
 continue to require flow  indicators in the vent stream from each distillation
 unit  within an affected  facility before each stream is combined with  any
 other vent stream being  routed to a combustion device.  The Agency has
 determined the cost of this monitoring requirement to be reasonable.  If
 indicators were not placed in each affected vent stream before it were
 combined,  then it would  not be possible for the Agency to be certain  that
 each  vent stream is routed to the combustion device.  Furthermore, if
 nondistillation vent streams were combined with distillation vent streams
 before combustion and a  flow indicator were placed in the combined stream,
 then  the monitoring requirements would be applied to process streams  that
were not intended to be covered by the standards.
                                    2-73

-------
      2.9.4  COMMENT:   One commenter (D-13)  suggested that only periodic
 thermocouple monitoring plus recordkeeping  should be required to determine
 whether a boiler is in service or not in service.  Section 60.663(c)(2)
 requires the use of a temperature measurement device equipped with a con-
 tinuous recorder for  boilers or process  heaters  of less than 44 MW
 (150 million Btu/hr)  heat input design capacity.   It was indicated that
 thermocouples are routinely placed in process heaters where temperatures run
 about 1,600 - 1,700°F.  However,  boilers do not  usually have firebox thermo-
 couples because the firebox temperature  runs between 2,000 - 2,700°F.   The
 commenter concluded that continuous temperature  monitoring would be difficult
 because of high temperatures in boilers  and the  related problems with  thermo-
 couple maintenance.   A decreased  exposure time of the thermocouple to  the
 firebox,  as with periodic sampling,  would decrease the heat stress upon the
 thermocouple and thus prolong  its  reliability.

      RESPONSE:   The EPA believes  that  continuous  temperature monitoring of a
 boiler firebox  with a design heat  input  capacity  of less  than 44 MW can be
 accomplished at a  reasonable cost.   In order to protect the thermocouple from
 heat  stress,  a  relatively inexpensive  ceramic insulator can be  used (see
 Docket Item No.  IV-J-16).   The  thermocouple  and ceramic insulator can  be
 purchased  for less than  $400 and can be  permanently  installed to the boiler.
 Thus,  the  continuous  temperature monitoring  of boilers  as  required  under
 Section 60.663(c)(2)  has  not been changed.   The EPA  has  changed  the
 definition  of "continuous"  to mean  a time interval of  at  least  every
 15 minutes.   This change  will allow for  the  use of computer-assisted systems
 for monitoring  requirements  (see comment  2.9.12).

     2.9.5  COMMENT:  One commenter (D-28) agreed with  EPA  and  stated that it
 is appropriate to waive performance tests and monitoring requirements for
 sources combusting process vent streams  in steam generating devices that have
 heat input capacities of 44 MW or greater.  The commenter suggested that
performance testing and monitoring requirements be waived under the following
conditions:   (a)  boilers with heat input capacities of 44 MW or greater;
 (b) all combustion devices maintaining a  combustion temperature of  1,100°C
                                    2-74

-------
and 1 second residence time; or (c)  incinerators maintaining a temperature of
870°F and 0.75 second residence time if no halogenated organic compounds are
in the vent stream.

     RESPONSE:  The EPA believes that condition (a) mentioned by the
commenter is sufficient for exemption from performance tests.  The Agency
believes that boilers and process heaters with a design heat input capacity
of 44 MW or greater would achieve a reduction efficiency of 98 percent or a
reduction to 20 ppmv, so long as the distillation vent stream is introduced
into the flame zone.  These boilers and process heaters are typically
operated at temperatures and residence times greater than 1,100 C and
1 second, respectively.  A firebox temperature of 1,100°C and 1 second
residence time represent the conditions that might in the worst case be
necessary to achieve 98 percent reduction, even if the organics were
chlorinated.  Furthermore,  it is to the economic advantage of the owners of
facilities  using boilers or process heaters to design and operate them with
adequate mixing of gases to maximize the extent of combustion, thereby
maximizing  the steam or heat generation rate.
     However, conditions (b) and (c) mentioned by the commenter are not
sufficient.   Even  though an incineration device may be operated at tempera-
tures  greater than 1,100°C  and  1 second residence time  (870°C and 0.75  second
residence  time for nonhalogenated streams) the distillation  column offgas,
combustion  gases,  and  supplemental  air must  be well mixed  in order to achieve
complete combustion.   The  EPA  has determined  that  proper mixing is,  in  fact,
as  important  as temperature and residence  time  in  determining  incinerator
efficiency.  This  concept  is explained in  an  EPA memorandum (Docket  Item
No.  II-B-3).   Improperly mixed gases may  actually  offset  the increases  in
efficiency generated by  raising the combustion  temperature.   This  is  due to
the fact that increases  in temperature only  increase  the  destruction
efficiency for VOC within  the  well-mixed  portion of the waste gas.   In  an
 improperly mixed  stream the increase in temperature does  not greatly affect
 combustion efficiency.  Temperature would be a poor indicator of  system
 efficiency in such a case.
                                     2-75

-------
      Unfortunately, mixing is a variable which cannot be measured.  Proper
 mixing is generally achieved through a trial-and-error process of adjusting
 the incinerator after startup.   There is no practical method of ensuring that
 proper mixing occurs except by  conducting a performance test and making the
 necessary adjustments.   For this reason,  incinerators operating at the
 temperatures and residence times expressed by the commenter in conditions (b)
 and (c) are not exempt  from the performance test  requirements.

      2.9.6  COMMENT:  One commenter (D-12)  stated that the definition  of
 "distillation unit" in  Section  60.661 should  include  a vacuum pump or  steam
 jet,  if present,  as part of the distillation  unit.  The commenter noted that
 it  is impractical  to  sample upstream of  these pumps or jets in vacuum
 distillation systems  and wants  to  be sure that EPA would not require sampling
 at  this location.

      RESPONSE:   Vacuum  pumps  and steam jets attached  to a distillation  column
 are necessary components for  the distillation  operation to occur  and are
 considered  to be examples  of  the accessories  referred  to  in  the definition  of
 "distillation unit" in  Section  60.661.   In order  to clarify  that  sampling
 downstream  of a vacuum  pump or  jet  is considered  to be  appropriate by EPA,
 the definition of  "distillation unit" has been amended  to  specifically
 include these accessories.

      2.9.7   COMMENT:  One  commenter  (D-15) requested that  a method for
 locating sampling sites  for vents less than 4  inches in diameter be provided
 in Section 60.664(c)(l).  Methods 1 or 1A are those specified for selecting a
 sample site.  Method 1  is applicable to vents greater than 12 inches in
diameter.  Method 1A applies to  vents from 4 to 12 inches  in diameter.
However, no method is given for  locating sampling sites on vents less than
4 inches in diameter.
                                    2-76

-------
     RESPONSE:  Method 2D - Measurement of Gas Volume Flow rates in Small
Pipes and Ducts (48 FR 48957) can be used to sample vents less than 4 inches
in diameter.  The method is listed in Section 60.664 as a. suitable method for
this NSPS.
     Because a pitot tube and sampling probe, as used in larger diameter
vents, would be disruptive to the flow rate of a small  diameter vent,
Method 2D requires that the entire vent stream be directed through a
measurement device such as a rotameter.  Therefore, a specific site location
is not required.  Section 60.664(c)(l) has been amended to clarify that no
site selection method for the vent cross section is needed for vents smaller
than 4 inches.  Method 2D is appropriate in this case.

     2.9.8  COMMENT:  One commenter (D-27) requested the addition of
Reference Method 2D to Section 60.664(f).  This section indicates that only
Methods 2A and 2C are appropriate to determine vent stream volumetric
flow rates.

     RESPONSE:  Section 60.664(f) has been amended so that Reference
Methods 2 and 2D will  be allowed in order to be consistent with
Section 60.664(a)(2).

     2.9.9  COMMENT:  One commenter (D-14) stated that  the measurement of
vent stream velocities according to the proposed standards may present a
safety hazard.  He indicated that when an orifice is placed in a vent in
order to measure stream velocity, a constriction is made in the piping.   This
constriction would hamper the ability of the piping to  handle high vent
stream loads in upset conditions, possibly leading to unsafe pressure buildup
in upstream equipment.  Therefore,  the commenter suggested deleting this
requirement.

     RESPONSE:  If the owner or operator feels the use  of an orifice meter is
an unsafe procedure, there are several  ways to ensure the safety of the
procedure.  For example,  a T-joint with a rupture disc  can be placed in the
vent upstream of the orifice meter so that the disc is  parallel to the
                                    2-77

-------
 direction of  flow.  The rupture disc would provide  for  emergency  venting  due
 to any potential pressure buildup.  Because these and other  safety  procedures
 can be implemented at a relatively low cost and because methods other than
 orifice meters are available, EPA has not deleted the requirement to use  a
 pipe constriction as one of several alternatives to measure  the flow rate in
 small vents.

      2.9.10  COMMENT:  One commenter (D-13) suggested that rather than
 requiring flow monitors to be installed on all small vent streams affected by
 the standards, the option of tank car seals should be allowed in the
 regulation.   It was pointed out that these seals can be used to assure that
 all connections on a vent line,  other than to a combustion or recovery
 device,  are  closed.  The commenter recommended that quarterly monitoring of
 the tank car seals could be used to ensure seal closure.  The commenter
 stated that  the use of these seals on  vent streams fed to boilers  or process
 heaters  would allow for regulatory flexibility.

      RESPONSE:   As  discussed in  the response  to comment  2.9.3,  EPA has
 changed  the  monitoring  requirement (Section 60.663  (a)(2))  for  a vent  stream
 combusted  by an incinerator  from the use  of .flow meters  to  flow indicators.
 Flow  indicators are required to  ensure  that the vent stream is  directed  to
 the combustion  device used to  control VOC  emissions.   These indicators
 provide a  record  and can be  quickly and regularly checked to  determine  if  the
 vent stream  is  reaching  the  combustion device.   The  EPA  decided  to require
 the use of flow indicators because of this reliability and  th.eir low cost.
     As listed  in the General  Provisions  (Section 60.13(1)),  EPA allows  the
 owner or operator of an affected facility to apply to the Administrator  to
 use alternatives to any monitoring procedures or requirements listed in  the
 distillation regulation.  The owner or operator can  pursue  this avenue for
 alternative methods, including tank car seals.

     2.9.11  COMMENT;  One commenter (D-19) expressed uncertainty about the
basis for the specified accuracy of temperature recorders and flow rate
recorders.   The commenter stated that it is unreasonable to attain an
                                    2-78

-------
accuracy of 1 percent or +2.5 C at firebox temperatures of 1,100 C.  The
commenter also questioned the need for an accuracy of 1 percent or +2.5QC for
temperature readings on scrubbing liquids, condenser coolants, and carbon
beds (Section 60.663(d)(l), (2) and (3)).  The commenter indicated that this
is too restrictive in light of the accuracy of current technology.
     The commenter stated that for continuous flow rate recorders the accu-
racy of 5 percent at flow rates of 100 to 500 scfm (Section 60.663(a)(2))
seems restrictive.  He feels that current recorders are not capable of
providing this level of accuracy.

     RESPONSE:  The EPA has acquired a great deal of experience in the
capabilities and limitations of various test methods and monitoring/
measurement equipment.  Based on this experience, the Agency has reevaluated
the temperature monitor accuracy requirement and has determined that the
accuracy should be 1 percent or +0.5°C, whichever is greater.  Therefore, the
final regulation will require this accuracy instead of the +2.5°C in the
proposed regulation.  For an incinerator operating at 870°C an accuracy of
+8.7°C would be required.  These accuracies can be achieved at a reasonable
cost through the use of readily available equipment.  Flow meters are no
longer required for the measurement of vent stream flow rate into an
incinerator, instead, flow indicators are now required.  Flow indicators have
no accuracy requirements because their function is only to indicate the
presence or absence of flow to the incinerator.

     2.9.12  COMMENT:  Two commenters (D-15 and D-27.) requested the standards
allow the use of computer-assisted systems to monitor product recovery and
combustion control equipment.  One commenter (D-15) stated that only inter-
mittent recording at 6 to 10 minute intervals (not a totally continuous
recording) is possible with certain computer control/recording systems.  The
other commenter (D-27) noted the maintenance, storage of spare parts and
understanding of an analog system with a strictly continuous recording system
would present many problems since they are rarely Used in the industry
anymore.  The commenter recommended that "continuous" be defined in the
regulation such that data should be collected at a frequency not less than
                                    2-79

-------
 1 percent of the compliance period.  For example, if the compliance period
 were 180 minutes, then the frequency of sampling would be at least every
 1.8 minutes.  The commenter believes that this definition of "continuous"
 would serve the EPA purposes equally well while still allowing industry to
 use a monitoring/recording system compatible with current computer control
 systems.  The other commenter (D-15) recommended that continuous monitoring
 be defined as systems capable of continuously recording parameters in
 increments of 10 minutes or less.

      RESPONSE:   The EPA has agreed to add a. definition of "continuous" to
 Section 60.661  as follows:   "Continuous recorder" means a data recording
 device capable  of recording data at time intervals of at least every
 15 minutes.   This will  enable industry to use existing computerized data
 control  systems attached to a measurement device.   Furthermore,  this time
 interval  has been found to  be an adequate time period for providing EPA with
 sufficient data to ensure proper operation and maintenance of VOC control
 equipment.   The measurement device will  actually be  sensing on a constant
 basis,  but the  data will  only be sampled at  certain  intervals.

      2.9.13   COMMENT:   One  commenter (D-12)  wanted to be  certain  that  the
 standards  provide for continuous monitoring.   Therefore,  he  suggested  a
 wording  change  for  Section  60.665.   Where  the  phrase  "every-15-minutes"  is
 used,  a wording  change  should  be made  so the  phrase reads  "at  least  every
 15 minutes."  The  same  commenter requested that  EPA discuss  the potential
 difficulties  in  making  continuously  recorded measurements.   He also wanted to
 know  if reliable devices  are readily available.

      RESPONSE:   In Section  60.665 of the proposed  regulation, the phrase
 "every 15 minutes" refers to the measurement of parameters during performance
 testing to establish parameter boundaries for use during monitoring pro-
 cedures to ensure that the  required VOC emission reductions occur.  However,
 based on a review of performance test requirements and monitoring
requirements the regulation has been amended to require measurement and
monitoring be done on a consistent basis.  The measurements of temperature,
                                    2-80

-------
specific gravity and steam mass flow rate made during performance tests are to
be made at least every 15 minutes.  This interval is the same as the
monitoring interval discussed in the response to comment 2.9.12.
     A discussion of the practicality of making continuously recorded
measurements was given in the preamble to the proposed standards
(48 FR 57550).  Two of the criteria used by EPA to establish the monitoring
requirements are reliability and accessibility.  For example, continuous
monitors to measure TOC were not required because they are expensive and
require much maintenance.  Instead, EPA chose to require monitoring of
temperature and flow rate as a more reliable and less cumbersome measure that
would involve readily available equipment.

     2.9.14  COMMENT:  One commenter (D-28) stated that Method 18 (a gas
chromatography procedure for measuring organic compounds) is not always the
best method to measure specific organic compound concentrations.  According
to the commenter, gas chromatography is not sensitive enough to detect
concentrations of certain compounds in the  range of 20 ppmv.  It was
suggested that the standards contain a discussion of alternate test
procedures, such as wet chemical methods, that have greater sensitivity than
gas chromatography.

     RESPONSE:  The EPA has judged this method to be accurate within
10 percent, and to have a lower limit of detectability to about 1 ppmv.
Although Method 18 cannot be used in a limited number of situations such as
in measuring compounds that can polymerize  before analysis, the vast majority
of organics emitted from industrial sources can be analyzed using this
method.  Therefore, Method 18 has been judged to be an applicable and
dependable method for measuring emissions from distillation facilities.
However, under Section 60.13(i), the General Provisions allow an owner or
operator to apply to the Administrator for  the approval of alternatives to
any of the measurement requirements or procedures listed in the distillation
regulation.
                                    2-81

-------
        2.9.15  COMMENT:   One commenter (D-10)  stated that the actual
   measurement of VOC emissions from incinerators would provide a more accurate
   assessment  of performance than  what  is  currently  required  in the proposed
   regulation.   He stated  that  in  addition  to internal  operating temperature,
   residence time and turbulence are also  important  indicators  of thermal
   incineration  performance.  He indicated  that many  incinerators  have heat
   exchangers  to  preheat the  incoming vent  stream  by  using  incinerator exhaust
   Because of  the  very high operating temperatures of the incinerator,  leaks may
   develop within  the heat exchanger.  According to the commenter, these leaks
   can permit uncombusted VOC in the incoming vent stream to migrate  into the
   incinerator exhaust.  He further pointed out  that this decrease in the amount
  of VOC that is combusted would not be detected by monitoring internal
  temperature alone.   The  commenter stated that measurement of VOC emissions
  would enable the SOCMI and EPA to judge  the relative merits of various
  control  equipment designs  on  a consistent basis.

       RESP°NSE:   The A9ency analyzed the  combustion  device operating
  parameters that  affect incinerator performance.  Included in  these  variables
  are temperature,  mixing, type of  compound combusted,  residence  time  inTet
  concentration,  and  flow regime.   The last two variables were judged'to have
  only a small impact on incinerator performance.  Residence time  is
  essentially set  after incinerator construction unless vent stream flow rate
  is changed.   Compound type has little effect  on combustion efficiency at
 temperatures above 760°C.  Mixing (turbulence) was  judged to be as  important
 as temperature and residence time in determining incinerator efficiency (see
 Docket Item  No. II-B-3).   Unfortunately,  mixing is  a variable that cannot  be
 measured.  Given the large  effect  of temperature on  efficiency and the low
 cost of temperature  monitors,  this variable is  clearly an  effective  parameter
 to monitor.
     The EPA  recognizes that monitoring temperature  alone  is not  a sufficient
means of determining  incinerator performance.   It is for this  reason that
temperature and vent stream flow rate are measured during  the performance
test and then monitored to determine if their values deviate from the  values
measured during the performance test.  This is  true for both catalytic and
                                    2-82

-------
thermal incinerators.  Considering the complexity of the processes involved
and the burden on both EPA and industry, the monitoring methods specified are
believed to be the least burdensome way of ensuring that the control
equipment is properly operated and maintained.
     Development of leaks in the heat exchanger is an example of deterioration
problems that can occur in all control equipment over a period of time.  The
General Provisions (Section 60.11(d) require owners and operators to maintain
and operate control equipment in a manner consistent with good air pollution
control practices for minimizing emissions.  General enforcement inspections
may be made to check for any deterioration in control equipment and to
determine the need for performance tests.

     2.9.16  COMMENT:  Two commenters (D-ll and D-13) requested that the
proposed standards include the option to monitor flare performance visually
or by ultra-violet beam sensor instead of using a thermocouple heat sensor on
the pilot flame of the flare.  One of the commenters (D-ll) stated that when
a temperature device is used for flare monitoring, increased emissions may
result due to shutdowns for monitoring instrument repair.  Therefore, the
commenter requested devices such as an ultra-violet beam sensor be allowed by
these standards.  The other commenter (D-13) noted that normal flare design
assures a flame presence at all  times because the pilot flame is supplied
with gas by an independent and reliable source.  It was pointed out that
experience shows thermocouples to present a major and unnecessary maintenance
problem.  This commenter recommended flame and smoke detection to be done by
a remote video camera.

     RESPONSE:  The EPA has decided that use of a ultra-violet beam sensor is
suitable to indicate the presence of a flame.  Therefore, the regulation will
be amended to allow for ultra-violet beam sensors or thermocouples to be used
as a heat sensing device at the  pilot light to indicate the continuous
presence of a flame.
     The detection of flame presence by visual  means or by remote video
camera is not a suitable method.   If a flare is operating smokelessly it can
be difficult to determine if a flame is present.
                                    2-83

-------
      2.9.17  COMMENT:   One commenter (D-28)  suggested that EPA provide for
 alternative methods of demonstrating compliance when distillation process
 emissions are combined with other emission sources within the plant.   For
 example,  vent streams  from distillation processes may be combusted in wood-
 fired boilers.   Since  wood-fired boilers inherently generate VOC emissions,
 demonstration of compliance with the proposed standards may be difficult.

      RESPONSE:   The General Provisions  (40 CFR 60.8) state that the
 Administrator may approve the use of "an alternative method (of demonstrating
 compliance),  the results  of which he has determined to be adequate for
 indicating whether a specific source is in compliance" with a standard.   This
 is  applicable to all NSPS and need not  be specified in the regulation.
      When distillation vent streams are combined  with nondistillation offgas
 streams within  the plant,  compliance of the-combined stream may be demon-
 strated using Reference Method 18,  using an  alternative method approved  for
 the  particular  facility by the Administrator or waived because the owner or
 operator  has  demonstrated by other means to  the Administrator's  satisfaction
 that  the  affected  facility is in compliance  with  the standards.   For  example,
 the  offgas  stream  from a  reactor may be routed  through the product recovery
 device or directly to  the  control  device of  a distillation affected facility.
 The  EPA has determined that if compliance  is demonstrated  with  the combined
 stream, compliance would  also be achieved  when  routing the distillation  vent
 stream alone.
      In the commenter's example  wherein  a  wood-fired  boiler  is  used to
 combust distillation vent  gases,  VOC will  be generated by  the  combustion
 device itself.   In  this case,  the  total  VOC reduction  would  still  have to be
 98 percent.   If  the VOC generated  by the wood-fired  boiler prevents this, it
will not  be considered  in  compliance with  the standards.

     2.9.18  COMMENT:  One  commenter (D-ll) requested  the  option to use other
methods to monitor condenser  performance in lieu of the current requirement
to monitor the exit (product  side) temperature of the  offgas.  The commenter
recommended methods such as the cooling water differential temperature or the
                                    2-84

-------
on-line  analysis of the exit VOC concentration of the condenser on the
product  side.   Correlations developed during a performance test were also
recommended  as  an  indicator of compliance during routine operation.

     RESPONSE:  The EPA has determined that the monitoring of the exit
(product side)  temperature provides the Agency with sufficient information
and is a relatively inexpensive indicator of condenser performance.  Although
measuring the cooling water differential temperature may also provide
sufficient information, EPA does not require this measurement because it
would call for  the use of two temperature monitors instead of one monitor.
This differential measurement would be more costly and would provide little
additional information than the exiting product side temperature alone.
However, as  indicated in the General Provisions under Section 60.13(i) and in
the distillation regulation under Section 60.663(d), the owner or operator
may apply to the Administrator to use alternative monitoring methods, such as
cooling water differential temperature.
     A continuous monitor to measure the TOC (minus methane and ethane)
exiting the condenser would be much more expensive and complex than the
current monitoring requirements.  Furthermore,  the reliability and accuracy
of these devices may be poor in some situations.   Therefore, EPA has decided
that,  generally, the exiting product side temperature is the least burdensome
way for an owner or operator to effectively monitor condenser performance.
However, as indicated above, an application to  the Administrator may be made
to use this device in lieu of the current requirements.

     2.9.19  COMMENT:  One commenter (D-12) requested clarification on
temperature monitoring during catalytic incineration.  The commenter stated
that the temperature of the incinerator after the catalyst bed is critical
for compliance, not before the bed as stated in Section  60.665(c)(2).

     RESPONSE:  According to Section 60.665(c)(2),  temperature measurement is
required immediately after the catalyst bed as  well  as before the bed.   The
inlet  and outlet temperatures must be monitored because  a temperature
differential  that is within 80 percent of the differential  measured during
                                    2-85

-------
   the most recent performance test indicates  that catalyst performance and
   subsequent  destruction  efficiency is  similar  to those  during performance test
   conditions.   When  only  the  outlet temperature is monitored,  a preheated vent
   stream  could  be sent  through  an  inactive catalyst bed  at a desired  tempera-
   ture so it would appear as  if the catalyst were  functioning  properly.
   However, monitoring a temperature rise from the  inlet  to outlet of  the
   catalyst bed would indicate that  the temperature increase was due to
   emissions destruction.  Therefore, no changes will  be made to the regulation
   concerning the monitoring requirements for catalytic incinerators.

       2.9.20  COMMENT:   One commenter (D-20)  expressed confusion over what
  constitutes "recovery" with  regard to  the  use  of the TRE index.   The
  commenter stated that  the term "recovery system" is  defined  in Section 60.661
  as  an individual unit  or series  of material  recovery units used  for  the
  purpose  of recovering  TOC from a  vent  stream.   The commenter  pointed out that
  the  BID  (pp. 4-8, 4-10)  indicates  "recovered"  VOC may be disposed.   The
  commenter  prefers the  definition  of "recovery"  to allow for the disposal  of
  collected VOC.   Therefore, he  requested the definition  of recovery system be
  revised  to allow for the disposal  of VOC.
      The commenter stated that if  recovery means  reclaimed for beneficial
  reuse only, then the point in the  recovery system where the distillation  vent
  stream is selected for a TRE index calculation  is inappropriate.  If any  of
 the organics collected from recovery equipment  are reclaimed  for beneficial
 reuse, then the TRE  value is  calculated for the vent  stream after the
 recovery  equipment.   But, if  organics exiting the same recovery equipment
 were  disposed of, then  the TRE  value for the  vent stream would be  calculated
 before that equipment.  The commenter noted that the  same emissions would
 result in either  case.

      RESPONSE:  According  to Section 60.661 of  the regulation,  in order for
 the TRE to be determined  after a recovery device,  the VOC  exiting that device
•ust be either used,  reused (e.g.,  recycled),  or sold.  The Agency maintains
that disposal of VOC collected by a device  does  not constitute recovery of
that VOC.   Therefore, it is not appropriate  to designate that  as a recovery
                                    2-86

-------
device for the purposes of these standards.   The wording in the BID indicated
by the commenter does not convey the proper intent of the standards.  The
outflows from the recovery devices diagrammed on pp.  4-8 and 4-10 should not
have been designated as routed for disposal.   Instead, the label  should have
indicated "for use, reuse, or sale."
     In order for a control device which disposes of collected VOC to be used
in complying with these standards, a 98 weight-percent VOC reduction or
20 ppmv emission limit must be maintained.   However,  the owner/operator of an
affected facility using such a device must  satisfy existing regulations
concerning the disposal of the collected VOC.  When using such a device to
comply with the 98 weight-percent or 20 ppmv emission limit, compliance must
be demonstrated as described in Section 60.664(b)(4).  Furthermore, as
described in Section 60.663(e), the owner/operator must provide to the
Administrator information describing the operation of the control device and
the process parameter(s) which would indicate proper operation and maintenance
of the device.

     2.9.21  COMMENT;  One commenter (D-30)  requested a change in the
monitoring requirements for the case when two or more distillation streams
containing dimethyldichlorosilone, a chemical listed Section 60.667, share a
common recovery device [Section 60.664(c)(ii)].  The commenter requested an
allowance for these distillation streams to be monitored after the outlet of
the last methylchlorosilone hydrolysis control device.  It was pointed out
that in the production of dimethyldichlorosilone, as much as 50 percent of the
TOC's emitted from distillation columns are methylchlorosilones.   The
commenter stated that these chemicals readily hydrolyze upon contact with
water to form nonvolatile polysiloxane gels,  water soluble silanols, and
hydrogen chloride gas.  The commenter indicated EPA does not recognize that
VOC is removed and converted to nonvolatile forms through the hydrolysis
control of methylchlorosiTones.  The commenter believes that a properly
designed water scrubbing device provides a VOC emission control that meets the
EPA standards for the production of dimethyldichlorosilone.  The commenter
proposed that the method for locating sites to sample flow rate and
                                     2-87

-------
 molar composition as listed under Section 60.664(c)(l)(ii) be changed to
 include "the outlet of the last methylchlorosilone hydrolysis control
 device."
      The same commenter stated that Section 60.664(c)(l)(111) does not
 contain sufficient information to allow evaluation of the sampling method
 requirements when an affected distillation column shares a common recovery
 device with one or more existing distillation  columns.   The commenter
 requested that EPA reword the paragraph.

      RESPONSE:  According to  the definition of  recovery system in
 Section 60.661 the methylchlorosilone hydrolysis  control  device would be a
 product recovery device if it were  used to recover VOC  exiting the device for
 beneficial  reuse.   The  term,  beneficial  reuse,  refers to the  fate of VOC
 after it exits the recovery device.  If the VOC is either sold,  recycled
 within  the  process unit or used  in  another process unit,  but  not disposed of,
 then  the VOC is  beneficially  reused.  Therefore,  the TRE determination  and
 monitoring  would be associated with the last such  recovery device.   If  the
 VOC were not beneficially  reused and thus  disposed of,  then the  vent  stream
 characteristics  must be measured before that device for the TRE  calculation.
 As discussed in  the response  to comment 2.9.20, the only  way  a device which
 disposes  VOC can  be used to comply with these standards  is  if a
 98 weight-percent  VOC reduction or 20 ppmv  outlet  concentration  is
 maintained.
     The  commenter  also  requested more  information on the  sampling
 requirements when  an affected distillation  column  shares  a recovery device
with one  or  more existing distillation columns.  The EPA  has  decided to
change the designation of affected facility such that all of  the columns
sharing a common recovery device would constitute  a single affected facility
 (see comment 2.2.1).  Therefore, there will be no need to apportion the vent
stream between affected and nonaffected distillation columns  in this case
since all columns would constitute one affected facility.
                                    2-i

-------
     2.9.22  COMMENT:  Commenter D-37 stated that the TRE index should be
calculated including all equipment up to the point at which the process
stream vents to the atmosphere to ensure that the TRE equation can be used
with respect to some developing technologies.  In addition, the commenter
said that the TRE equation should be modified to allow use of the charac-
teristics of the vent stream emitted to the atmosphere.  The owner/operator
should then be required to demonstrate that no negative environmental impact
will result from disposal of liquids or solids from the control device.  The
commenter stated that the proposed method for determining TRE could prevent
the cost-effective use of some control devices without any commensurate
benefit to environmental quality.  The commenter further stated that without
the modifications suggested, some control devices would be able to show
compliance under the "98 percent reduction" alternative standard but not show
compliance using the TRE index.

     RESPONSE:  As stated in the preamble to the proposed standards, a
distillation column generally does not release emissions directly to the
atmosphere since all offgas from it is vented to recovery devices.  In
developing the final standards, EPA designated the recovery system and its
associated distillation columns as the affected facility because it led to
greater emission reductions than alternative designations (see Comment
2.2.1).  Because recovery systems are part of the process and also very
effective in reducing VOC emissions, it was decided that all decisions on the
need for additional VOC control should be based on the stream characteristics
of the vent stream exiting the recovery system.  The impacts associated with
requiring additional VOC control after product recovery were then estimated
and evaluated.  The Administrator determined that when the cost of control of
the recovery system's vent stream was greater than $l,900/Mg of VOC removed,
BDT is no additional control.  It was also determined that when the cost was-
less than $l,900/Mg of VOC removed, BDT is the reduction of VOC emissions by
98 weight-percent or to 20 ppmv.  Recovery equipment normally is operated at
the most economically efficient level which may not correspond to maximum
reduction of VOC emissions.  However, the Administrator determined that where
the cost of control is less than $l,900/Mg of VOC removed, requiring an
                                    2-89

-------
 additional 98 weight-percent reduction is reasonable.  The change the
 commenter suggested would allow the use of control devices that are not as
 effective as BDT.  Because the impacts associated with requiring BDT are
 reasonable, no change has been made to the final standards.

 2.10  REPORTING AND RECORDKEEPING

      2.10.1  COMMENT:  Three commenters (D-5, D-12, and D-13) stated that
 annual  reporting of changes in normal  operations would be sufficient instead
 of the  semiannual reports currently required under Section 60.665(k).   One
 commenter (D-24) agreed with the waiving of the semiannual reporting
 requirements for affected facilities in States where EPA, in the course of
 delegating the enforcement programs, approves alternative reporting
 requirements or means of source surveillance.

      RESPONSE:   Semiannual  reporting is only required for the following
 circumstances:   (a)  when the monitored parameters of combustion  and recovery
 devices  exceed  the latest performance  tests;  (b)  when a vent  stream is
 diverted from a control  device  or  does not  have a flow rate;  (c)  for periods
 when  a boiler or process heater is not operating; (d)  when the pilot flame of
 a  flare  is  absent;  (e)  for  any  changes in a  process  operation that  cause  an
 increase in  the maximum design  vent stream  flow rate  for  affected facilities
 with  a flow  rate  less than  0.008 m /min;  (f)  for  any  changes  in  a process
 operation that  increases the  design production  capacity of the process  unit
 for affected  facilities  with  a  design  production  capacity less than  1 Gg/yr;
 and (g)  for  any recalculation of the TRE index  value.   Reporting of  these
 occurrences  are required on a semiannual, instead of  annual,  basis because
 any situation which has  the potential  to result in an  increase in pollution
 should not go unreported  for  as long as a full year.   In  addition, because
 changes  to the process can affect which control requirements  the affected
 facility must comply with, it is important that these changes  be reported  at
 least semi annually.
     The Agency also agrees that semiannual  reporting should  be waived when
the enforcement program has been delegated to the States and  alternative
                                    2-90

-------
requirements have been approved.   The Administrator has determined semiannual
reporting of the above occurrences not to be overly burdensome to the
industry to ensure compliance of each affected facility with the appropriate
section of the standards.

     2.10.2  COMMENT;  One commenter (D-19) stated that the recordkeeping
requirements are excessively burdensome for distillation facilities having a
design capacity of less than 1 Gg/year or a designed maximum flow rate of
0.008 m3/nrin.  The commenter suggested that those facilities meeting the
design capacity or vent stream flow rate exemption should only be required to
notify EPA of its exemption status and when the status changes.  The
commenter requested that EPA specify under Section 60.665 what these reports
should contain.

     RESPONSE:  The EPA has decided to exempt from coverage by the standards
distillation facilities that operate with  a vent stream flow rate less than
0.008 m3/min, even if the  facility is designed for a vent stream flow rate
above that  level.   However, Section 60.665(i) has been amended to require
the  owner or operator of exempted facilities to record any  changes in process
operation that may cause the vent stream flow rate to  exceed the 0.008 m /min
level and to measure  and record  the  flow rate after the change has been  made.
The  EPA  requires  semiannual  reporting of process changes  and the new vent
stream  flow rate  measurement only when changes have been  made  that cause the
vent stream flow  rate to exceed  the  exemption value.   Examples of changes  in
process  operation are given in comment 2.1.5.
      The Agency also requires the recordkeeping  and semiannual  reporting of
changes  in  process operation that  increase the design  production  capacity  of
facilities  with design production capacities  below 1  Gg/yr.  If a  change in
process  operation does not increase  the  total design  capacity then  no record-
 keeping or reporting are necessary.   The EPA does  not feel  that semiannual
 reporting is excessive.
                                     2-91

-------
        2.10.3   COMMENT:   One commenter (D-5)  requested clarification on the
   length  of time  that  records must  be  maintained  by the owner or operator   The
   commenter suggested  that  only  the most  recent performance  test data be kept
   It was  also  suggested  that the continuous records  of the equipment operating
   parameters as well as  the  type of information mentioned in  comment 2.10.1  be
   kept during  the interval between  reporting periods.

       RESPONSE:  The preamble to the proposed standards states  that  "all
  records would be required to be kept up to date and in readily accessible
  files for 2 years."  Two types of records are required to be kept for this
  period.   One type is  the operations  parameters that are monitored during the
  operation of control  devices or product recovery devices (Section 60 663
  Section  60.7(d)  of the  General  Provisions).   The other type is  the tabulation
  of periods when  the measurements of  the  control  device or product recovery
  device operating parameters significantly deviate  from measurements of the
  same  parameters during  the  most recent performance  test [Section  60 665(c)
  through  (g)].

      2.10.4   COMMENT.:   One  commenter  (D-5) requested clarification  on whether
 Section 60.665(g)(5)(i)  requires recordkeeping for any  affected distillation
 facility that changes its production  rate or for only those facilities where
 the design capacity is exceeded.

      RESPONSE:  The EPA  requires under Section 60.665(g)(5)(i) recordkeeping
 of any changes in  production rate.  The Agency may need this information to
 be certain that an  affected facility  demonstrating  compliance with a TRE
 index  value greater than 1.0 is  still  in  compliance after a  change in
 production rate has  been made.

     2.10.5  COMMENT.:  One commenter (D-12) stated that  practical  "parameter
 boundaries" are needed for the total mass  steam flow during carbon bed
 regeneration cycles  [Section 60.665(g)(3)(i)].  A parameter boundary  is a
 li.it set during the most recent performance test.  Any  future measurements
that fall  below or exceed (depending on the parameter)  this limit must be
                                    2-92

-------
recorded as specified under Section 60.665(g).  Another commenter  (D-ll)
requested that the same section allow for the exclusion of insignificant
steam flow rate changes that are within the variability of the steam meter.
The commenter recommended that Section 60.665(g)(3)(i) be modified by adding
the following wording:  "within the accuracy of the steam meter."

     RESPONSE:  It is inappropriate for an accuracy limit to be used as a
parameter boundary because it allows for no minor deviation from the
parameter value measured during the last performance test.  Therefore, the
regulation has been amended so that reporting is now required under
Section 60.665(g) (3)(i) when the mass steam flow is more than 10 percent
below the total mass flow during the most recent performance test.  When the
amount of steam used to regenerate a carbon bed has decreased beyond the
parameter boundary, the carbon adsorber may not adequately be serving as a
product recovery device and thus the TRE value may be less than 1.0.

     2.10.6  COMMENT;  One commenter (D-15) requested an allowance in the
reporting requirements for condensers [Section 60.665(g)(2)] to account for
increased water temperature during summer conditions.  If the performance
test is done in winter when a cooling tower or once-through river water is
used for cooling, the incoming water temperature will increase by greater
than 10°C in the summer.  This temperature change will cause the summer
operating temperature of the condenser to continuously exceed the previous
winter operating temperature only because of the summer conditions.

     RESPONSE:  When the cooling water entering a condenser increases in.
temperature, the effectiveness of the condenser as a product recovery device
is likely to decrease.  The requirement to record and report exceedances of
the operating parameters is needed for the owner or operator and the
enforcement agency to know when the condenser may no longer be operating
properly, even if the exceedance is due to seasonal  temperature changes.
     Therefore, in this case,  the owner or operator should either cool the
water to a temperature below the exceedance level  or demonstrate to EPA that
the TRE value is still above 1.0.
                                    2-93

-------
   2.11   GENERAL
       2.11.1  COMMENT:  One commenter  (D-21) requested better documentation  of
  contacts between EPA and the Office of Management and Budget (OMB)
  especially in regard to the cost-effectiveness cutoff used in the proposed
  standards.  To substantiate this request, the commenter cited Sierra Cluh .
  Costle, 657 F.2d 298 (D.C.  Cir. 1981), in which the court accepted the
  practice of reducing oral  communications to memoranda and inserting them in
  the docket.   This commenter also cited the CAA Section 307(d)(4)(B)(1i)
  which requires  written  communications  to be placed in the public docket!

       BESEQNSE:  All  correspondence  between EPA and OMB directly  related to
  the proposed  NSPS for SOCMI distillation  operations  are  contained in  Docket
  No.  A-80-25,  which is available  for public inspection.   The correspondence
  can  be  found  under Docket Item Numbers II-F-1  and  II-F-2.  The policy of how
  any  communication between EPA and any other Federal  agency is treated by EPA
  has  been clearly described in a  letter from a  previous Administrator to  the
  commenter's organization.  (Docket No.  IV-C-6).

      2.11.2  COMMENT:  One commenter (D-15) stated that ducting the vent
 streams from several  distillation units into a  common recovery device is
 potentially dangerous.    It  was  further pointed out that  using a  manifold to
 jom several  vent  streams  results in a  system that is difficult to control
 and  does not  meet  the safety requirements  of insurance underwriters.

      RESPONSE:  The Agency does not  require ducting vent  streams  into  common
 recovery  devices.  However, some  distillation facilities  operate  by using
 common ducting.  The  EPA realizes that caution  must be taken when  vent
 streams from several  distillation units are ducted  into a common  recovery
device.   It is the responsibility of facility owners or operators  to meet  the
safety requirements established by the  company or its underwriters
                                    2-94

-------
     2.11.3  COMMENT:  One commenter (D-12) stated that chemical names,
mixtures, and generic terms are intermingled for the chemicals listed in
Section 60.667.  Also, he indicated that no Chemical Abstracts Registry (CAS)
numbers were supplied.  The commenter requested that the chemicals in
Section 60.667 be listed according to the format used for the standards of
performance for equipment leaks of VOC in SOCMI (48 FR 48342 to 48344).

     RESPONSE:  In order to facilitate the identification of chemicals
affected by the distillation NSPS, the Agency has included the appropriate
CAS numbers in Section 60.667 in the final rule.  The chemicals are now
listed according to the format used for the VOC equipment leaks standard.

     2.11.4  COMMENT:  One commenter (D-12) stated that the discussion of
condensation in the BID should contain a treatment of surface condensers that
do not require dehumidification equipment.  It was pointed out that EPA
should not assume water will  be present in all distillation vent streams.
The commenter indicated that presence of water vapor requires the use of
dehumidification equipment.   According to the commenter, this equipment would
not be suitable for the production of some organics such as benzene that
freeze at the temperature needed for proper water vapor control.

     RESPONSE:  The discussion in the BID on surface condensers is intended
to provide general description of the product recovery techniques that can be
used to reduce and recover VOC emissions from distillation facilities.
Dehumidification equipment is included in the description of surface con-
densers because water must be removed from many product streams from
distillation operations.   However, EPA recognizes that some vent streams
contain no water vapor and this is reflected in the data used to analyze the
standards.  Furthermore,  neglecting to indicate that some surface condensers
do not use dehumidification  equipment does not affect the applicability of
the proposed standards to the production of the listed chemicals.  Therefore,
no further discussion of surface condensers is planned for the BID.
                                    2-95

-------
      2.11.5  COMMENT:   One commenter (D-12)  stated that the potentially
 detrimental  impacts from energy demand as a  result of the standards may
 offset the benefit to  air quality noted in Table 1-1  of the BID.   The
 numerical  values given in Table 1-1  indicate that energy demand is expected
 to  have minimal  impacts.   However, the discussion of  this table in the text
 indicates  that  energy  impacts  could  be significantly  detrimental.   The
 commenter  suggested that  because the BID text discusses possible  negative
 energy impacts,  Table  1-1 should quantify an energy impact that could be
 significantly detrimental.

      RESPONSE:   The numerical  values given in Table 1-1 do not  contradict the
 discussion  of the table given  in the text.   The  text  does not indicate that
 the energy  impacts are very detrimental.   It indicates  that  the impacts are
 reasonable  even  under  a worst-case scenario  where in  the fifth  year
 1.2 billion MJ/yr (190 thousand  barrels  of oil equivalent) would be  used  for
 a flare  preference on  vent streams with  no halogenated  compounds.
 Furthermore,  EPA believes that the energy  impacts will  be substantially less
 than  the worst-case scenario.  The impacts from  this  worst-case scenario
 would  be lessened because of heat recovery with  combusting vent streams in
 boilers  and process  heaters, and the  upgrading of product  recovery equipment
 to raise the  TRE value above the cutoff.

     2.11.6   COMMENT:  Two commenters  (D-9 and D-14)  stated that the
 conversion constant  "K" in Section 60.664(c)(4)  should  be  1.740 x 10"7
 instead  of 1.740  x  10  as it is presently written.

     RESPONSE:   Section 60.664(c)(4)  will be amended to read "K » constant,
 1.740 x  10"   ..."  instead of 1.740 x 107.  A typographical error had been
made.

     2.11.7  COMMENT:  Seven commenters (D-8, D-9, D-ll, D-14,  D-22, D-27,
and D-33) indicated that there is an  error in the first column  of Table 1  in
Section 60.664(c)(6).  Four of the commenters (D-9, D-ll, D-22,  and D-27)
recommended that the second line of Table 1 contain a  "greater  than" symbol.
                                    2-96

-------
     2.11.8  COMMENT;  One commenter (D-9)  recommended that the coefficient
on the second line under heading "f" of Table 1 [Section 60.664(c)(6)] should
be -0.0036.  The coefficient is now given as 0.0036.

     RESPONSE:  The commenters in 2.11.7 and 2.11.8 correctly identified
errors in Table 1 of Section 60.664 in the proposed regulation.  Because the
coefficients and tables have been completely revised,  the specifics of the
comments are no longer relevant.  However,  the errors  cited have been
corrected and the corrections are incorporated into the final regulation.

     2.11.9  COMMENT:  Two commenters (D-ll and D-12)  suggested that the word
"steam" in Section 60.665(k)(2) should be changed to "stream."

     RESPONSE:  Section 60.665(k)(2) will be amended such that the word
"steam" will be replaced by "stream."

     2.11.10  COMMENT:  One commenter (D-27) indicated that Section 60.663
(c)(3) should be moved to Section 60.665.

     RESPONSE;  Section 60.663(c)(3) is needed in order to indicate that no
monitoring is required for a boiler or process heater when the design heat
input is 44 MW (150 million Btu/hr) or greater.  Records indicating periods
of operation are required in lieu of monitoring.  The same requirement is
also given under Section 60.665(e).

     2.11.11  COMMENT:  One commenter (D-9) suggested that the "note" at the
end of Section 60.660 should be placed at the end of Section 60.662.

     RESPONSE;  The  "note" at the end of Section 60.660 is in an appropriate
position because it  indicates that numerical emission limits are expressed in
terms of TOC's, less methane and ethane.  This information is useful when
reading the definitions in the  next section.  Three of the 13 definitions
refer to TOC's.
                                    2-97

-------
             APPENDIX A:   FEDERAL REGISTER NOTICES OF ORGANIC
                  COMPOUNDS DETERMINED TO HAVE NEGLIGIBLE
                         PHOTOCHEMICAL REACTIVITY

 INTRODUCTION
      As indicated by the Federal  Register notices included in  this
 appendix,  the  following chemicals have  been  determined  to  be negligibly
 Photochemically  reactive compounds:  methane; ethane; 1,1,1-trichloro-
 ethane; methylene chloride, trichlorofluoromethane; dichlorodifluoro-
 methane; chlorodifluorotnethane; trifluoromethane;
 trichlorotrifluoroethane-; dichlorotetrafluoroethane; and
chloropentafluoroethane.
                                 A-l

-------
35314

  ENVIRONMENTAL PROTECTION
              AGENCY
             ina. 739-8]
            MR QUALITY
                                                    NOTICES
              PUWOSB

              of this notice Is to rec-
       . This
be approvable. However.
be followed by EPA whenever it
aulred to draft  State
Flans for the  control- of
                                  re-
•tuiamea wu» lormuUte la an _
proved rule for national UM.

   	SUMMARY		

  Analysis of available data and tof or- .
nation show that very few volatile or-.
Sanic compounds are of «<£low photo-
chemical reactivity that  they can be-.
Snored  in  oxldant  control programs.
For this reason. EPA's recommended
policy reiterates the  need for positive
reduction techniques (such as the reduc-
tion  of  volatile organic  compounds  in
surface  coatings, process. changes, and
 the use of  control equipment) rather
 than the substitution of compounds «f •
 low  (slow)  reactivity to the place of
 more highly (fast)  reactive compounds.
 There are three reasons for this. First
 nn.ny of the VOC that previously have
 been* designated as having low reactnW
 .are.now known to  be moderately or
        reactive to urban atmospheres.
                                     Preon 114. and Preon 115. which are cur.
                                     rentiy used ae aerosol propellants. The
                                     Agency la planning to Investigate control
                                     systeiiu and substitutes for nonpropei-
                                     • lent •uses nn^1* TSCA. as announced m
                                     May 13. Methyl chloroform is not a fully
                                     naSognnated chloronuoroalkane. Rather,
                                     it is among the chlorine-containing com-
                                     Dountls for which, the Agency has not
                                     completed its analysis: EPA has not ret
                                     ccncliided whether it is or is not a threat
                                     to th« stratospheric ozone. Therefore, it
                                     has been placed on this list as an accept-
                                      able exempt compound. As new informa-
                                      tion  becomes available on these com-
                                     jwuniis. EPA will reconsider the recom-
                                      mendation.                        —•
                                        •The -volatile organic compounds listed
                                      to Table 2. while more photochemicaUy
                                      reactive than, those to Table 1.  never-
                                      theless do not contribute large quantities
                                      of ozidant under many atmospheric con-
  Photochemical  oxidants result  from,
sunlight acting on volatile o**»n]*eom-
pounds (VOC) and oxides of nttrogen.
^^VOC. by then-nature, start to form
     •~* after only a short period  of ff-
     bidfr to ttM* atmosphere. Other VOC
     "undergo irradiation for a  longer
period  before  they yield-measurable
  . toits guidance.to State. le*ttuprep-.
 •ration, adoption, and submittal of State
 implementation planspuWtohed toan.
 the Environmental  Protection Agency
 —Tph..*^ reduction of total organic
 compound emissions, rather than^nb-
 BJ However, to  ..
 that substitution  of one «~—
 another might *e-useful wbw«^^
 —suit to a  clearly evident decrease in
 teacttvity and thus tend to reduce photo-
 chemical  oxldant  formation.  Subse-
 mently  Tp«nT  State. Implementation
 Plans  were promulgated  with *>»«£
 substitution provisions similar to Rule
 88 of the Los Angeles County Air Pollu-
 tion Control   ""'  """•" *""*' * ""*
   Speared in  the  PES«AL  RxoOTa on
   ££ruary5.197«<41FR5330).
     The 1978 policy statement ««*fd»°
   that the reactivity concept was useful
   .Tan interim measure only, and would
   Sotbe considered a reduction to organic
   emissions for purposes of estimating at-
   tainment of  the ambient air quality
   rtandard for oxidants. The  document
    also included the following statement:
     Although the lubttltutton Po
    M  and  rtfflllir rules represent
      ttw«*        regulations eta
    veloped. one* on current knowledge of re-
 nigf"Y mcww 4** *•*•!"   •    ^  _ _
 Second, even compounds.that are pres-
 ently known to Have low reacttvttj can
 form appreciable  amounts  of oxldant-
 under multiday stagnation cortrtlttnna
 such as occur during summer to many
 areas. Third, some compounds of  low
 or negligible reactivity may have other
 deleterious effects.
   Of the small number of VOC which
'have only negligible photochemical re-11
 activity, several. (benzene. *cetonilzile.-
 chlomform. carbon tetrachtorlde.ethyl-
 ene dichlortde. ethylene dibromide, and.
 methylene chloride) have been identlfled,
 or implicated as being carcinogenic, mu-
 tagenic. or teratogenlc. An  additional
 compound, benzaldehyde. while produc-
 togno appreciable ozone. nevertheUas.
 forms a strong eye irritant under irradia-
 tion. In view of these circumstances, it
 would be inappropriate for EPA to en-
 courage or support increased utilization
. of these compounds. Therefore, they are
 not recommended for  exclusion  from
 control. Only the four compounds listed
 to Table  1 are recommended for exclu-
 sion from SIP regulations and. therefore.
  It is not necessary that they be inven-
 toried or controlled, to determintoii; re-
  ductions   required  to  meet  oxidant.
  NAAQS.  these VOC should  not be In-
  cluded to the base line nor should reduc-
  Uonsto their emission be credited toward
  achievement of the NAAQS.
   • rt 1*  —~™««d  that the  two  halo-
  -enated^oWMnds listed to Table 1
   (methyl chloroform and Preon 113J may
  cauMdeterioration of the earth's ultra-
  violet  radiation shield  since they  are
  neSy-unreactlve to the lower atmos-
   phere and all contsJn appreciable frac-
   tions  of  chlorine.  The  Agency has
   reached conclusions on the effects of only
   Selully halogenated chloronuoroal-
   kanes. The Agency on May  13.1977 (42
   FR 24542)  proposed  rules  under the
   Toxic Substances Control Act (TSCA) to
   prohibit the nonessential use of fully
   halogenated chlorofluoroallcanes as aero-
   sol propellants. The restrictions were ap-
   plied  to all  members of this class In-
    cluding Freon 113. since they are poten-
    tial substitutes for Freon 11.. Freon 12.
                                                                                   U—VolatU*  Orjante. Compound! of
                                                                               KKritafUf motaehcnrical Beactttity That
                                                                               ShESuB* Zttmp* from Regulation
                                                                              rt&tttt ImpUiwntatton Plan*
                                                                              vSSwchKroethene (Methyl Chloroform*'
                                                                              PJ^IT-J tfBjBMMAflMAf^Kfftfch^AA (FrooQ 113)

                                                                                                h»v«
                                                                                       tueretore.
                                                                             jtareoootzoU.
                                                                                 r Xr-ralotUt Organic  Compo-.-«Jj
                                                                               Low Motocntmic* Reactivity
                                                                              Acdone •
                                                                              Methyl Ethyl Ketone
                                                                              Uethenol   '••
                                                                              bopropiaot
                                                                              Uethyl Pencn«te
                                                                              TtnttaryAlkyl Alcohols
                                                                              UrtbylAoet***  •-
                                                                              phenyl Aceteta .
                                                                              Ethyl Amtoe*   '"  -
                                                                              •
                                         V. IT-dlmethyl fornumide
                                           Only during multiday stagnations do,
                                         Table 2 VOC yield significant oxidamt;
                                         Tfaerefore. if resources are limited or U.
                                         the sources are located  to areas where;
                                         prolonged atmospheric stagnations  are
                                        -uncommon, priority should be given to
                                         controlling more reactive VOC first ana
                                         Table 2 organies later. Table 2 VOC art-
                                         to be included to base line emission In-
                                         ventories and reductions to them wul B«-
                                         credlted toward achievement of  tae.
                                         NAAQS. Reasonably  available  control
                                         technology should be applied to signin-
                                         eaiatsourees of Table 2 VOC where neca-
                                         sary toattoto the NAAQS for oxidants.
                                         SZ s^ceTof these compounds wlU a^
                                         be subject to new source renew require^

                                         "'p^hloroethylene. the principal soU
                                         ventmployed to the dry cleaning todus-
                                          toy UaJso of low reactivity. comparaW
                                          to VOC listed to Table 2. It was not Itt-
                                          SBdeTtoTable 2 because of reported adj



                                          tenslvely by occupational healthi aumor _
                                          «:hylene currently are being^tudiefl'.^
                                          vistigations may have major impact o»
                                           IMinn. VOl. «». NO. ,31_«IOAt. JULY •.
                                                         A-2

-------
industrial users. In designing control reg-
oUtions for perchloroethylene sources.
oartlculariy dry cleaners, consideration
should be gtven to these findings as well
as industry requirements and the cost of
applying controls. Available control tech-
nology is highly cost effective forjarge
Derehloroethylene  dry clfimmg opera-
tions. However, for corn-operated and
irnT«n dry cleaners,  the same equipment
would  represent   a  heavy -economic
burden.                          	
  As part of Its continuing program. EPA.
will review new information relative to
the photochemical reactivity, toxlcity, or
effects on stratospheric ozone of volatile
organic compounds. Where appropriate;
additions or deletions will be made to the
lists of VOC la Tabels 1 and 2.  .
              Dsccssmf
  Most air pollution control regulations
applicable to stationary sources of VOC
to the United States are patterned after
Rule 06 of the tow Angeles county Air
pollution control  District  (presently
Begutetton 443 of the> Southern Califor-
nia Air PoUutton Control Ettstrictt. Bole
66 and similar regulations Incorporate
two basic strategies to reduce ambient
oxidant levels,  f-e,  positive VOC reduc-
tion and selective  solvent substitution
based on pboteehendeal reectWtr. Posi-
ttvet reduction, schemes such as Incinera-
tion, absorption, and the'use of low-sol-
vent coatings are acknowledged means of
reducing ambient oxidant levels; they
should be retained m future VOC control
programs, m contrast, the utility of sol-
vent  substitution  strategies has  been
questioned as more  information on pho-
to chemical reactivity has emerged.
  EPA  acknowledged the shortcomings
of solvent substitution based on Rule 66
reactivity criteria hi a 1976 policy state-
ment (41 FR 5350). Findings were cited
which indicated that almost an  VOC
eventually  react in the  atmosphere to
form some oxidant. Concurrently..EPA
initiated an Investigation to consider Im-
plications of revising the solvent eubsti-
tuton aspects of Rule 86. Three separate
forms were conducted with representa-
tives  of State end local ah-  pollution
cnntiul egencies. university professors.
end  Industrial  representatives   with
knowledge  and expertise  in-the fields of
Atmospheric chemistry  and industrial
polvent applications. In addition, nu-
merous discussions were held with ac-
knowledged experts in the field. Topics
of particular concern were:
           Bui* 06 substitution  criteria
eould be renied consistent with available
*»acttTlty data am yet be compatible with
industrial  processes and wltn product  re-
quirements.
  Whether  eome compounds an of suffi-
ciently low reactinty that ther-are not oxl-
«ant preeunors and can be exempted from
control under State Implementation Plane.
  Whether the Imposition  of reactmty re-
e&lctlons In addition to positive emission
reductons  will delay the  development or
implementation  o* promising  technologies,
particularly  «he  use at  water-borne  end
high-solids  eurfaos
                                                     NOTICES

                                         investigation showed that:
                                         1. Solvent substmtion based on Role
                                       66 ha* been dlrecUonally correct in the
                                       aggregate and probably effects some re-
                                       ductions in peak oxidant levels. How-
                                       ever. because of  the relatively high re-
                                       activity of  most  of the substituted sol-
                                       vents. the reduction is small compared to
                                       that which can  be accomplished with
                                       positive reduction techniques. Revision
                                       of Rule 66 consistent with current knowl-
                                       edge of. reactivity would eliminate  the
                                      •solvent substitution option, for  most
                                       sources la which  substitution is new em-
                                       ployed. Many of the organic solvents
                                       which  ham been categorized as having.
                                       low photochemical reactivity are, la fact.
                                       tmyforfttoiy or I^f^y reactive; they yield
                                       ^ynm/.ant  oxidant when, subjected to
                                                 I
                                       nlm"!?** **"? urban, atmosphere.
                                         2. A few VOC yield only negligible
                                                    Ha«^ •««< ethane* e> yfM*tp
                                                   	be so-classified. Tf"**
                                       	.	react Terr alowly  yielding
                                       little  ozone- during, .the first few  days.
                                       followinc fe*1**^ release to *F»* alny?fp**£T^i
                                       Available data, suggest that none of the.
                            etitZniflCeVLLv -
 oxidanteven during extended Irradiation
 under multiday stagnation conditions.  ' '
   The broad group "halogenated paraf-
 fins"   includes  important  Industrial
 solvents, most of which are chlorinated
 methanes and ethanes and ehlorofiuoro-
 ethanes. They *"* use as metal cleaning
 and dry cleaning solvents and as paint
 removers.  Halogenated paraffins  also
 serve as building blocks in the manufac-
 ture  of other halogenated  organlcs:
 these processes do not necessarily release
 significant VOC to  the atmosphere.
  *3.  Besides  focusing  on  VOC  of
 negligible  reactivity,  smog  chamber
 studies show that a few additional VOC
 generate oxidant at a relatively stow rate.
 Under favorable atmospheric conditions."
 these VOC releases may not form oxidant
 until they have been transported sub-
 stantial distances  and  become-greatly
 diluted. However, under multiday stag- -
 nation conditions such as occur during •
 summer in "^"y areas of the mf*1*^<> and
' eastern United  States,  there is  the •
 potential for these organlcs to undergo
 appreciable conversion to oxidant. The
 more important VOC In this category are
 acetone, methyl ethyl ketone. parebloro-
 ethylene. methanol, Isopropanol. and
 propane. All except propane are indus-
 trial solvents. The latter, a gas under.
 normal  conditions;4 Is associated  prin-
 cipally  with crude  oil and  liquefied
 petroleum gas- operations.
  ' 4. The vast number of volatile organic
 compounds—particularly nonbalogenat-
 ed VQC—yield appreciable ozone when
 irradiated in the presence of oxides of
 nitrogen. While  there are measurable
 variations in then* rates of ozone f orma-
                                       «**" VOC listed in Table X Quickly re-
                                       active VOC include almost all aliphatic
                               35315

and aromatic  solvents,  alcohols,  ke-
tones.  glycols. and ethers.
  5. Low photochemical reactivity is not
synonymous with tow biological activity.
Some of the negligible or slowly reactive
compounds have adverse effects on hu-
man health. Benzene, acetonltrile. car-
bon tetrachloride. chloroform,  perchlo-
roetbxlene, ethylene dlchloride. ethylene
dlbromlde. and methylene chloride have
been Implicated  as  being carcinogens.
teratogens,  or  mutagens.  In. addition.
benzaldehyde. which produces no ap-
preciable ozone,  nevertheless  forms  a
strong eye  irritant  under Irradiation
While their use might  reduce ambient
oxidant levels. It would be unwise to en-
courage their uncontrolled release. Ad-
ditional halogenated organlcs are being
UTvestigatedfor possible toxlcity.    " .
  Most of the related health inform*-'
Hnn iTtillaMm*1'thu time concerns acute
toxlcity. Threshold limit values (TLVs)
have been  developed  for many VOC.
They are appropriate- for the healthy.
adult work force nrrmtnt eight hour* *
day. five day* a, week. Experts suggest.
that more, stringent levels-should be
^flMuh^t foe the general population.
Hazards-represented by cfarnnVT. aM sub-.
chronie rumm^ are rmv* more diffi-
cult to- quantify than acute toxlcity. Ad-
verse health effect* of the VOC cited
above are g»q»~iiy- recognized although
not completely quantified. Chlorinated
solvents currently are  under intensive
study.              -        '-,   •
  6. Some VOC are of such low photo-.
chemical reactivity that they persist in
the atmosphere for several years, even-
tually  migrating to the stratosphere
where they are suspected of reacting and
destroying  ozone. Since  stratospheric
ozone Is.the principal absorber of ultra-
violet  (UV)  light, the  depletion could
teed to an  Increase in UV penetration
with a resultant worldwide increase in
skin cancer. The only in-depth analysis
of this potential problem has focused on
the chlorofluoromethanes (CFM). Freon
11 and Freon 12. because of their known
stability and widespread use in aerosol
containers.  A  report  of the National
Academy of Sciences concerning  envi-
ronmental effects of CFM's concluded
jShat: -  .--.- -   -   "     -.
  "•«'•  • seletclve  regulation of CFM uses
and releases Is almost certain to be necessary
at some time and  to some extent of com-"
ptotenees.  '          • .      : ^  -'

In response-to the report of the National
Acadcr^y of  Sciences *^rf other studies.
. EPA on May 13.1977 (42 FR 24542). pro-
posed rules to prohibit nonessential use-
age of fully halogenated chtorofluoroal-
kanes  as areosol propellents.  The re-
strictions, were applied to all members
of  this class including  Freon 113 since
they are potential substitutes for Freon
11. Freon 12. Freon 114. and Freon 115
which  are  currently  used. as  aerosol
propellents.
  Other  stable  halogenated  solvents
which  are released In volumes compare- <
ble to the cnlorofluoroalkanes also are
suspected of depleting  tba earth's UV
shield.  Of major concern is the wide-
                                 KOUAL U6ISTEI. VOL 42. NO. 131—EIIOAT, JULY  (. 1977


                                                      A-J

-------
 35316

 spread substitution of methyl chloroform
 (U.1  trlchloroethane)  for the photo-
 rhemtcallr  reactive degreasing solvent
 trlchloroethylene. Such substitution un-
 der Rale 66 generation regulations has
 already influenced industrial decreasing
 operations  to  the extent that methyl
 chloroform  production  has  surpassed
 that of tnchloroethylene in the United
 States. Any regulation  in the area .will
 have a marked effect on the production
 and atmospheric «••«•«<««««  of both sol-
 vents. Endorsing methyl chloroform sub-
 stitution would increase emissions, par-
 ticularly in industrial States that have-
 not, heretofore. Implemented Bole 66. On
 the other hand, disallowing methyl chlo-
 roform as a substitute or h"w««g it alto-;
 gather would significantly increase emis-'
 slons of trlcnloroethylene  even  if de-
 greasers were controlled to the limits of
 available'technology. Presently;  tech-
•nology is only able to reduce emissions by
 approximately 50 percent. Itt metropoli-
 tan areas which have  already 'imple-
 mented Rule 66. a return to trlchloro-
 ethylene would have an advene effect
-on «»»)M»nt oxidant levels. In addition to
•being highly reactive, trlchloroethylene
rhas been implicated as a carcinogen.
 _  -Alternatives to the above-cited choices
-would be (1)-development and appuea-
 .tion of highly  efficient degreaser control
 systems and  (2) •replacement with  an
               NOTICES

 intermediate sol vent which Is neither re-
 active nor detrimental to the upper at-'
 mosphere. Major  revisions would  be
 needed to degreaser designs to improve
 vapor capture above the current best
 level. Antlripatfri design changes could
 add materially to degreaser costs. No. al-
 ternative solvent is  clearly acceptable •
 from the standpoints of photochemical
 oxidant Brnl stratospheric prone deple-
 tion.  Neither  metbylene  chloride  nor
 trtchlorotrifluoroethane are reactive, but.
 like methyl chloroform, are suspected of*
 causing damage  to  *•-***  stratospheric
 flypflft layer. In ftdditlff", methyleno chlo-
 ride -is- a' suspect  mutagen. Perchloro-
 ethylene. the principal dry. cleaning sol-
 Tent, does "not present  a hazard to the
•stratosphere but has  been Implicated as.
 being a«arcinogen and also reacts slowly
 In the- atmosphere to form oxidant.
 .-  7.-Organic-solvents of low or negligible
 photochemical.' reactivity- have  only
 limited use in many Industries. Most are
 chlorinated- organlcs  that find principal
 applications as cleaners for metals and
 fabrics. Atew nonhalogenated VOC such
 as acetone,  methyl  ethyl ketone.  and
 isopropanol  are of low reactivity but
 these, can't possibly satisfy all the myriad
 needs of the paint, plastics; pharmaceu-
 tical,  or many other-industries.. While
 users of reactive VOC usually can employ
 effective control equipment to recover or
destroy VOC •»««"*""•, they seldom have
the option  of applying  reactivity con-
siderations in choosing solvents. Applying
reactivity restrictions to the surface coat-
ing industry would be especially disad-
vantageous since It would greatly inhibit
the development of low-solvent coatings;
essentially all  of the organic solvents
used  to  constitute high-solids coatings
and waters-borne coatings are. in fact..
highly reactive.
  8. It Is recognized that smog chamber'
studies conducted to date are incomplete
because*  many organic compounds have
not been examined and It has been  un- -!
posslbltt to duplicate all atmospheric sit-'.
nations.  For example., there has been ,
only limited fTftmlnp+l"" of oxidant for- .:
mation under relatively  high ratios of:
VOC to NO, (30:1 and greater) . compar-
able to rural conditions. Any policy on
.photochemical reactivity necessarily has
to be open to revision as new Information
is developed which may show  specific
organic:  compounds to  be more  or  less,
photoeliemlcally. reactive than indicated
by current data.!   .....       ' * -
          June 29. 1077.
  .   ..•     .   EDWUD F. Tens.
    '  Acting .Assistant AdmtiHttrator
  •J.f?z10*4i' and Watte Management.
  ^IFBf Doe.TT-lBSSS FIM 7-7-77:8:45 am|
                                       ttGISTEt, V01. 41, NO. 131—HMOAY, JUIY ». 1977


                                                       A-4

-------
                    Federal Register /  VoL 44. No. 108 / Monday. June 4. 1979 /  Notices
Review under 42 U.S.C ! 719(b) (1977
Stiff.] from en order of the Secretary of
Energy.
  Copies of the petition for review have
been served on the Secretary.        y
Department of Energy, and all
participants in prior proceedings before
the Secretary.
  Any person desiring to be heard with
reference to such filing should on or
before June 12, 1979. file a petition to
•intervene with the Federal Energy
Regulatory Commission. 825 North
Capitol Street Ni. Washington. D.C
20429. in accordance with the
Commission's rules of practice and
procedure (18 CFR 1.8). Any person
wishing to become a party or to
participate as a party must file a petition
to intervene. Such petition must also be
served on the parties of record in this
proceeding and the Secretary of Energy
through Gaynell C Methvia, Deputy
General Counsel for Enforcement and
Litigation. Department of Energy. 12th
and Pennsylvania Ave, N.W,
Washington, D.C. 20481. Copies' of the
 petition for review are on file with the
Commission and are available for public
 inspection at Room 1000, 825 North
 Capitol SU RE, Washington. D.C
 20428.
 KnaMhF.Ptas*.
 (Ft OH.
 [Docket No. RI79-M1

 Triton OH * Gas Conn; Petition for
 Declaratory Order
 and 598-A or show cause why such
 refunds were not due. Triton's position
 is that because sales under these rate
 schedules were authorized by
 permanent certificates of public
 convenience and necessity which
 contained no refund conditions.'there is
 no refund obligation. Triton
 acknowledges that the Commission may
 order refunds and reductions in rates
 after August 1.1971—the effective date
 of Opinion No. 598. However, it asserts
 that the Commission is without
 authority to order such adjustments
- prior to the effective date where rates
 were not collected subject to a
 suspension order or under a temporary .
 certificate.
  , Any person desiring to be heard or to
 make any protest with reference to said
 petition should file a petition to
 intervene or a protest with the Federal
 Energy Regulatory Commission, 825
 North Capitol Street N.E, Washington.
 D.C 20428. in accordance with
 requirements of the Commission's rules
 of practice and procedure (18 CFJL U
 or 1.10). All such petitions or protests
 should be filed on or before June-20.
 1979. AH protests filed with the
 Commission will be considered by it in
 determining'the appropriate action to be
 taken but will not serve to make the
 protestants parties to the proceeding.
•- Any person wishing to become a party
 to a proceeding, or to participate as a
 party in any hearing therein, must file  a
 petition to intervene in accordance with
 ttui Commission's rules.
 KeanetaKPhnnb.
   Take notice that on April 5, 1979.
 Triton-Oil and Gas Corporation (Triton).
 One Energy Square. 4925 Greenville
 Avenue. Dallas. Texas 75208 filed in
 Docket No. RITB-M a petition for
 declaratory order pursuant to Section U
 of the* CommissioB/s Rules -OK Practica*
 and ProcedurevTriton requests-a
obtigatioa- under Southern Looisiana
Area Rat* Opimon No. 59» for rates it
cofieeted for certain sales of gas. Ths
gas is produced from four fields in th*
                 i Area-siid sold tar
          Gas Pipeline Company,
      CM Transmission Company and
                       npany under.
 Southern Natural Gas Co
 Triton's Rat* Schedules land 8.6. and 7
 respectively.
   On June 6.1978, the Comntission
 directed Triton, among other producers.
 to disburse refunds for the period from
 October 1988 to January 1971 pursuant
 to the Commission's Opinion Nos. 598
  [Docket Net. IW7B-M1

  UnttsdGas Pte« Un« Co; informs*
    Takejwoce that on jtm*7,197arat~
  f*yy p nil aii.faifti-Hff'fli conference *yf»it
  Interested persons* will be convened tot
  the* purpose of cootnmed settlement.
  discussions fat -^iw proceeding. The*
  conference- will be held in Room 3300 of
  the Federal Energy Regulatory
  Commission at 941 North Capitol Street.
 -KE-. Washington. D-C 20428.
    Customers and other interested
  persons will be permitted to attend, but
  if such persons have not previously been
  permitted to intervene by order of the
  Commission, attendance will not be
  deemed to authorize intervention as a
  party in mis proceeding.
  All parties will be expected to come
fully prepared to discuss the merits of
the issues arising in this proceeding and
to make commitments with respect to
such issues and any offers of settlement
or stipulation discussed at the
conference.
Lois D. CubcO.
Acting Stcntarr-
                        soon and
                                                                            Office of Energy Co
                                                                            Solar Applications
                                                                           *
                                                                            Meeting Regarding Emergency
                                                                            Building Temperature) Restrictions
                                                                            Program

                                                                              Notice is hereby given that the
                                                                            Department of Energy (DOE) will hold a
                                                                            meeting with the National Governors'
                                                                            Association on Friday. June 8.1979. at It
                                                                            ajn. in Room 285.444 North Capitol
                                                                            Street Washington, D.C
                                                                            -  The purpose of the meeting will be to
                                                                            discuss the rote of the States in
                                                                            implementing the Emergency Building
                                                                            Temperature Restrictions Program. This
                                                                            program is authorized by-the President's
                                                                            "Standby Conservation Plan No. 2:
                                                                            Emergency Building Temperature
                                                                            Restrictions." which recently was
                                                                            approved by the Congress.
                                                                              Issued la Wssttngton. D.C on May n.
                                                                            1879.
                                                                            MaxmeSevmc,     .
                                                                            Deputy AMOtantSfCTttarf. Cotatrratioa
                                                                            andSoiarAppiicatiam.
                                                                            (fin
                                        ENVIRONMENTAL PROTECTION
                                        AGENCY
 [FRI.12JS-41
                                                             i of Agency
                                        PoOcyConcsjrningOzone-SlP
                                                                             ACT*
                                                                                              LIS nnniisnm n******
 the anthoritrof section lOl(b) and
 section 103 of the Clean Air Act The
 notice darlffas EPA's "Recommended
 Policy on Control of Volatile Organic
 Compounds." 42 FR 35314 duly 8,1977).

 STATZMCNT: The July 1977 Policy
'Statement noted that only reactive
 volatile organic compounds participate
 in the chemical reactions that form
 photochemical oxidants. Currently
• available information suggests that
 negligibly pbotochemically reactive
 volatile organic compounds as defined
 in that Statement including methyl
                                                      A-5

-------
                                                                                          32043
      	Federal Register
======      \.  -j  A*    these compounds under the Clean Air

                                 a           ——
         Cementation plan or plan
                                                                 MJJMO DAT! «M»«1-II
                                                                 ^~~-
                                27711(919)541-5204.
ha controlled under state           ,
SptaStation plan, for *•££• •«
sss«S=aSr-   --

 \gn and March 8. 1979. there*
     stive evidence that .both
          are
                                                                  SnmiW.eonBTforc«n«it;Ooen
                                                                  Meotina
                                                                  issuer. Eavironmental Protection
                                                                  Agency CEPA). Office of Pesticide
                                                                  programs.
                                                                   .-^.i. Notice of Open Meeting.
                                                                   ^MMA^TTie«willb«atwo-day
           .                .,
                                    •w«wi«~j	                  RCi9earch and Evaluation Group




                                                  irinkina Water Act   w w Atlanta. Georgia. Teiepnone.««/
                                               ^JonS-           S^oaandwulbeopentothepubha
                                              . Of the Environmental          FURTMBI INMUMATION CONTACT:
                                                                    Kfr William BuSaloe. North CaroUna
                                                                    Department of Agriculture. Rale.'^;
                                                                    NorthCarolina. Telephone: 919/733-
                                                ULu«niu»fc-	  3556: or Mr. Anthony Dellavecchia,
                                                Cumberland and Cape   I»esticida and Toxic Substances
                                                   "-  	J         Enforcement Division. EPA. 401M
                                                                     Street S.W, Washington. D.C,
                                                                     telephone: 202/755-0914.
                                                                     wftunmnuet WWOHMATKJN: This the
                                                                     Mcond meeting of the Working
                                                                     Conumtt«oBEnfo«m«t^.meeting
                                   sfe^S5;j-=:s: .»====••=*•
                                                                      1. Plan for future recall and
    nogrim.aainiarfJ7.w^^^^   arfWimsdstopiwrtdemei
      Contej^j^^^0^^    ^«»^thecmnn.ent^
                                                   imments. data and
      jW
             both methyl chloroform and
             duorfd. «« P

     ..  uo               .
     hamfaL BPA recommenda ttat tnea«
     etenicala not be snbstftuted for other
     aohvot* in efforts to reduce otone
     ooneentrationa, EPA further
     Meammends that the states control
     th,,. compounds underthe authority
     re^rred to-them in section 118 of &•
     Clean Air Act Moreover, there is a
     stiong possibility for future regulation of
                                        o              .  .Y. 10007,
                                       Attention: Coastal Plain Aquifer.
                                       Information concerning &e Coastal
                                       Plain Aquifer System will be available
                                       for inspection at the above address.
                                                  A-6
Section 28 and 27 of HFRA:
  3. Status of State-primacy use
enforcement notice
  .ijjea of raconanendations «*
aaricnltaral extension publications by
•pesticide sales representatives;
  5. Discussion of definition of "non
cropland:'*    '
  8. FFRA Section 7—producers of
 active ingredients: and
  7. Other enforcement matters wmch
 may arise.
  Dated! May 25.1979.
 EdwinMohn»oa,   .             .
          Want Administrator for Pestiad
 ire Doe. rvtra* raw ••*->««• «-l
 Mima coo« «•*«>«•

-------
                                           '--•
T-——*•««*,   sr"*"1**'1^^
'F**SttS&1SSZ*   *i*.»*1****'»**»
SScttt^Sraarm  *^» «££
SSfeasn-tt.*   Mgr
•ssffls^^   tsssssr
^I?mSSP^.t^Sir        ^^^r
 drfnMra ofa"sen«ittve are*. The
 «l.M.ofanyp~ticid.spr.y..not      ^ ^^ ^ AcTTh. no'tic.
 permitted over a sensitive;area or m ™   fj"^"..-^ clarification of a policy

 =saiS.vsS& "^"   fe^^Tfssrti-.
 only spray drift fallout from th.          g2£da?"«ra 3S314 (July 8.1977)
 —'—«-« «»•«-                     "jSnTriScation of Agency Policy
                            EPA approval. If a state chooses to
                            controKssions of these compounds.
                            TOchmeaswes will be considered as
                            aiate regulations only and not as part of
                            en ozone SIP. EPA will not enforce
                            Sn^ok on emissions of eiUier methyl
                            chloroform or methylene chloride
                            adopted by the state as part of an
                            implementation plan for ozone.
                               States retain authority to control
                             emissions of these compounds under the
                             authority reserved to them under
                             Section 118 of the Dean Air Act For
                             further information relevant to the
                             exercise of this authority see the July 8.
                             ^and June 4.1979 policy statements.
                             This policy notice should not be read as
                             a statement of EPA's views on the
                             desirability of controls on these
                             nihstances.            .  .
jSKSssssra-
be buffered against direct application.
Hnwcver many of these dwellings are
ne« aquatic sites listed in Table H
which will be buffered.
  To minimize operational errors.
overflights of the treatment area pnor to
the actual spray operation are
encouraged. The purpose of these
overflights is to locate visually all
sensitive areas and bufferzones
 designated on the spray block auP*-
 Particular attention should be given to
 identifying ephemeral streams and
 oonds visible from an aircraft flying at
 an altitude of UOO feet or less above me
 t.rrain at the time of treatment, which
 maynot be designated on the spray
 block map due to their seasonably.
 Authority
   This Advisory Opinion governing the
 use of certain insecticides for the
 .uppression of the spruce budworm in
 Maine through July. 1980. is issued
 pursuant to the authority granted to the
  Administrator by Section 2^et'^Lgv
                                   4.1979).
                                                     us policy
                              "FteffiBPA wishes to point outthat
                              this policy notice addresses only the
                              ATeicVslack of authority to include in
                              fcderaUy approved SIPs controls on
                              .ubSes whose emissions do not
                              contribute, either directly or indirectly.
                              £ «ncentrations of po^'ants for which
                              NAAQS have been established under
                              ..ction 109 of the Act This poUcy notice
                              does not address the question of SIP
                              measures which control substances
                                    reactive and do not appreciably
                                    !X>ute to the formation of ozone.
                                    Consequently, controls on emissions ol
                                    these two compound would not
                                    contribute to the attainmen and
                                    maintenance of the national ambient.air
                                    quality standards for ozone. In the June
                                    1979 policy statement EPA explained
                                    mat it would not disapprove any state
                                    toplementation plan ISIP) or plan
                                    ^vision for its failure to contain
                                    regulations restricting emissions of
                                    ^Sfchlorofonn and/or methylene
                               bVmore strict than absolutelynecessary
                                 ittain and maintain the NAAQS. EPA
                                 . „„ authority to exclude such
                            «--
    See) defines this terminology"
    prohibiting the use of registered
      estiddTln a n«»n«-no»Pnm?
     pes                      ,
     {he labeling." However. »^°° *">
     .1,0 provides that mis prohibition does
     not apply with respect to "any use of a
     pesticide In a manner that the
     Administrator determines to be
•^Scft's statement clarifies EPA
oolicv reaarding state implementation
SttlnKtals which do contain
mulations restricting emissions of the
KSSrnd^ StetoilOMei) of th«
dean Air Act limits state
Implementation plans to measures
designed to achieve and maintain the
Mttonal «nbient air quality ^^Q
fNAAQS). Because current information
indicates that emissions of methyl
chloroform and methylene chloride do
not appreciably affect ambient ozone
tavels. EPA cannot approve measures
 .oeciflcaily controlling emissions of
- either or both com^unds as partof a
 SderaUy enforceable ozone SIP. EPA
                                to attain ana mamwu» u« ^ - -
                                has no authority to exclude such
                                measures from SIPs.
                                ran rowTHW INFOHMATIOM CONTACT
                                G T Helms. Chief. Control Programs
                                G^eration, Branch (MD-15). Research
                                Triwgle Park. North Carolina 27711.
                                S£)541-S228. FTS 829-5226.   -

                                  Dated: May 9.1980.
                                 David G. Hawkins.
                                 AssittantAdmifiituatorforAir. NO>M a"d
                                 Radiation.
                                                                            -
                                                                           cooe •*»+*-*
 deraUy enforceable ozone   .
wUl take no action on any measures
•necifically controlling emissions of the
S^mpounds which are submitted by
the states as ozone SIP measures for
im. 14S1-7; W 601807/12391

Extension oi a Temporary Tolerance
AGENCY: Environmental Protection
Agency (EPA).
ACTION; Notice.	

SUMMARY: EPA has extended the
temporary tolerance for residues of the
herbicide thidiazuron (JV-phenyl-N -1O.3-
thiadiazoM-ylurea) and its aniline-
containing metabolites in or on the raw
agricultural commodities cottonseed at
O2 part per million (ppm). milk O.OS ppm.
 eggs 0.1 ppm. meat fat and meat
 byproducts of cattle, goats, hogs, horses.
 poultry, and sheep at OJ ppm.

-------
                     Federal Register / Vol. 45. No. 142  / Tuesday. July 22. 1980 / Notices
                                                                       48941
 Dempasco Service Sta. U.S. 1 and Hwy A1A.
  Juno Beach. FL 33408—S-14-80
 Par Mobil 324 Par Avenue. Orlando. FL
  32804—3-18-80
 John Gibson. 1-65 and KY 90, Cave City. KY
  42127—3-18-80
 Bellmeade Shell 5315 S. Harding. Nashville,
  TN 37205—3-19-80
 Corner Store. 1401 No. Main Street.
  Kissimmee. FL 32741—3-19-W
 Kopper KetUe. Highway 100 & 1-65. Franklin.
  KY 42134  « 7-80
 Bueehel Terrace Chevron. 4219 Bardstown
  Rd, Louisville. KY 40218-4-10-60
 UPorte Exxon. 1829 a Federal Hwy,
  Hollywood. FL 33020—4-24-40
 Risner's Chevron. 3420 Lebanon Road.
  Hermitage. TN 37076—5-13-80
 Douglas Amoco Service. 583 Donaldson Pike,
  NashviUe. TN 37214—9-14-80
 Town & River Texaco. 1024 Cypress Lakes
  Rd, Ft Meyers. FL 33907—5-14-80
 Trail Sunoco. 8168 So. Tamimiami. Ft Meyers,
  FL 33907—8-14-80
 Villas Chervron. 8180 So. Tamamiami. Ft
  Meyers. FL 33907—6-14-80
 Port Comfort Box 105, Rt 24. Ft Meyers. FL
  33908—5-15-80
 Cantrell's Exxon. 1910 Dickerson Rd,
  NashvittVTN 37207—5-18-80
 Barker Westgate Standard 2510 Pio Nono
  Ave. Macon. CA 31206—S-19-80
 Seminole Exxon. 1949 W. Tenn, Tallahassee.
  FL 32304—5-19-80
 Fred Hulse^s Chevron, 5012 Romeiser Road,
  Macon.  CA 31204—5-20-80
 Winston Chevron. 825 Madison Street.
  Huntsville. AL 35501—4-22-80
 H & A Fuel Service. P.O. Box 449. Hardeville.
  SC 29929—*^—23—80
 Chancy'i Standard, P.O. Box 1701 St Sunona
  Island. GA 31S23—5-28-80
 Norman's  Standard. 3304 Glynn Avenue,
  Brunswick. CA 31520—5-28-80
 Plaza Standard. 1965 Glynn Avenue.
  Brunswick. GA 31520—5-28-80
 Coley's Exxon. Rt 11-85 and SC 290. Duncan.
  SC 29334—5-28-80
 Bingham's Texaco. Rt 11-85 and  SC 290.
  Duncan. SC 29334—5-28-80
 White's Exxon. Hwy MS and SC-8.
  Spartanburg. SC 29303—5-28-80
Mauldin Chevron. 804 N. Main. Mauldin, SC
  29662—5-29-80
Wade Hampton Mall Exxon. 1035 Wade
  Hampton Blvd. Greenville. SC 29609—5-
  29-80
Harris Standard, P.O. Box 405. Nahunta. GA
  31553—5-29-80
Pittman's Standard. 1-75 and Juliette Rd,
  Forsyth. GA 31029—5-30-80
Trout's Texaco. 108 N A1A Hwy. Satellite
  Beach. FL 32937—5-30-80
Magnolia Plantation. P.O. Drawer. Tifton. CA
  31794—5-30-80  '
M & M 76.1100 SR 324 Rt 1. Cocoa. FL
  32922—5-30-80
  Issued in Atlanta. Georgia on the llth day
of July 1980.
James CEatterday.
District Manager.
  Concurrence:
Leonard F.Bittner.
Chief Enforcement Counsel.
[FH Doe a>-ZUM Flbd 7-a-Hk Ml «•)
eajJNO COM (4M-OVM
ENVIRONMENTAL PROTECTION
AGENCY
[FHL1545-7]

Air Quality; Clarification of Agency
Policy Concerning Ozorw SIP
Revision* and Solvent Reactivities
AGENCY: Environmental Protection
Agency (EPA).
ACTION; Notice.	

BACKGROUND: This notice is published
under the authority of section 101(b) and
section 103 of the Clean Air, Act The
notice provides further clarification of a
policy announced in EPA's
"Recommended Policy on the Control of
Volatile Organic Compounds." 42 FR
35314 (July 8.1977) and  "Clarification of
Agency Policy Concerning Ozone SIP
Revisions and Solvent Reactivities." 44
FR 32042 (June 4.1979) and 43 FR 32424
(May 16,1980).
DISCUSSION: The previous policy
statements on the control of volatile
organic compounds (VOCa) noted that
despite concerns about  their potential
toxicity 1.1,1-trichloroethane (methyl
chloroform) and methylene chloride are
negligibly photochemically reactive and
do not appreciably contribute to the
formation of ozone. Today's statement
expands the list (45 FR 32424) of organic
compounds (VOCs) of negligible
photochemical reactivity to include the
following chlorofluorocarbons (CFC) or
fluorocarbons (FC):
trichlorofluoromethane  (CFC-11):
dichlorodifluoromethane (CFC-12);
chlorodifluoromethane (CFC-22);
trifluoromethane (FC-23);
trichlorotrifluoroethane (CFC-113);
dichlorotetrafluoroethane (CFC-114);
and chloropentafluoroethane (CFC-11S).
  EPA has determined that these
halogenated compounds are no more
photochemically reactive than methyl
chloroform and methylene chloride and
do not appreciably contribute to the
formation of ambient ozone.
Consequently, controls on emissions of
these compounds would not contribute
to the attainment and maintenance of
the national ambient air quality
standards for ozone. EPA cannot
approve or enforce controls on these
compounds as part of a Federally
enforceable ozone State Implementation
Plan (SIP). EPA will take no action on
any measures specifically controlling
emissions of these compounds which
are submitted by the States as ozone SIP
measures for EPA approval. (See 45 FR
32424.)
  However. EPA would like to reiterate
its continuing concern over the possible
environmental effects from emissions of
these compounds. As such, EPA is not
precluding the possible future regulation
of these compounds.
  It should be recognized that the two
halogenated compounds, methyl
chloroform and CFC-113. stated to be of
negligible photochemical reactivity in
the July 8,1977 Federal Register, have
been implicated in the depletion of the
stratospheric ozone layer. This layer is a
region of the upper atmosphere which
shields the earth from harmful
wavelengths of ultraviolet radiation that
increase the risk of skin cancer in
humanii.
  In response to this concern, the
Agency promulgated on March 17.1978
(43  FR 11318). rules under the Toxic
Substances Control Act (TSCA) to
prohibit the nonessential use of fully
halogenated chlorofluoroalkanes as
aerosol propellants. Restrictions were
applied to all members of this class,
including CFC-113. since they are
potential substitutes for CFC-11, CFC-
12. CFC-114, and CFC-115, which are
currently used as aerosol propellants.
The Agency i' investigating control
options and restitutes for
nonpropellai;  ases.
  EPA lias prr-Tosed new source
performance standards under Section
111 for organic solvent cleaners (45 FR
39768, June 11,1980). These proposed
standards would limit emissions of the
reactive volatile organic compounds
trichloroethylene and perchloroethylene
as well as methyl chloroform, methylene
chloride, and trichlorotrifluoroethane
(CFC-113) from new. modified, or
reconstructed organic solvent
degreasers. If these standards are
promulgated, EPA will develop a
guideline document for States to us*  in
developing regulations required under
Section lll(d) for existing organic
solvent cleaners that use any of the
designated compounds.
  Whether, and to what  extent, methyl
chloroform and methylene chloride are
human 'Carcinogens or have other toxic
effects, and to what extent methyl
chloroform. CFC-113. and other CFCs
deplete the ozone layer, are issues of
considerable debate. Detailed health
assessments of methyl chloroform.
methylene chloride, and  CFC-113 are
being prepared by EPA's Office of
                                                           A-8

-------
                --- i===^—

      Research and Development These
      assessments will be submitted for
      external review, including a review by
      the Science Advisory Board, prior to
      promulgation of the regulations and the
      proposal of EPA guidance to States for
      developing existing source control
      measures. The extent to which the
      preliminary findings are affirmed by the
      review process may affect the final
     rulemaking for new as well as existina
     sources.                         ^
     . J?"'!1 the«j8»UB» of environmental
     impact are fully resolved. EPA remains
     ««Ce7!f f*8t " theae «*"*•!• are
     exempted from regulation, the
     substitution of exempt for nonexemnt
     solvents could result in large increases
     of emissions of pollutants that may have
     adverse health impacts.
      The emissions of CFC-22 and FC-23.
    also of relatively low photochemical
    reactivity, are of continuing concern
    with regard to possible environmental
    effects. Consequently. EPA is not
    precluding the possible future regulation
    of these compounds as well
      Finally. EPA wishes to point out that
    this notice addressee  only the Agency's
    lack of authority to include in Federally
    approved SIPs controls on substance,
    whose emissions do not contribute.
   either directly or indirectly, to

          '"0"8
         r-?* Cajifo"»a Air Resources
  Board (GARB) notified EPA of two
  recent amendments to California's
  emission standards and test procedures
  for motor vehicles produced by certain
  small-volume manufacturers, and
  requested a waiver of Federal
  preemption for each amendment EPA
  will consider these waiver requests
 among other issues, at a public hearing
 already scheduled for July 24, 1980 at
 EPA's San Francisco office, as

                          ** "0tiCe
f — "j ™«» **#ww*
AOORKSSU: EPA will consider the

I?n cr ?q"MtS at a public heari«S held
at U.S. Environmental Protection

NS IT™*- °P«6 (Re8ion ^
                                                                           ^^
                                            that the State standards will be. in the
                                            aggregate, at least as protective of
                                            public health and welfare as applicable
                                            Federal standards. The Administrator
                                            mll?l8ra/lt a waiver unless he finds that-
                                            (1) The determination of the State is
                                            arbitrary and capricious. (2) the State
                                            does not need the State standards to
                                            meet compelling and extraordinary
                                            conditions, or (3) the State standards
                                            and accompanying enforcement
                                           procedures are inconsistent with section
                                           202(a)oftheAct
                                             Pursuant to these provisions, the
                                           Administrator of EPA granted California
                                           waivers of Federal preemption allowing
                                           the State to enforce its exhaust emission
                                           standards for 1979 and subsequent
                                           model year passenger cars ' and for 1979

                                           Ste!E^.^J^*
                o  Pofcnts   r which
   NAAQShave been established undeT
   Section 109 of the Act This policy notice
   does not address the question ofSlP
   measures which control substances
   contributing to concentrations of
 ---»- -~. i laucisco. California. Conies
 of all materials relevant to the hearing
 are available forpublic inspection
 dunng normal working hours (8:00 a.m.
 to 4:30 p.mj at- U.S. Environmental
 Protection Agency. Public Information
 Reference Unit Room 2922 (EPA
 Library). 401M Street  SW,
 Washington. D.C. 20460.
 FOR FURTHER INFORMATION CONTACT:
 Glenn Unterberger, Chief, Waivers
Section. Manufacturers Operations
Division (EN-340), U.S. EnVironmental
                                                               u*-«y veice
                                                 ). • In American Motors Corp. v

                                                 lfeDf u0*011 held that "ction
                                          202 b)(l)(B) of the Act entitled American
                                          Motors Corporation (AMQ to two
                                         -additional years of lead time to meet
                                          certain California oxide of nitrogen
                                          (NOJ emission standards for passenger

                                           As a result in a Federal Register
                                         notice issued July 3. 1980. the
                                         Administrator modified his passenser •
                                               *e' diion "^         *
            ,       c  are contended to
  be more strict than absolutely necessary
  to attain and maintain the NAAQS. EPA
  has no authority to exclude such
  measures from SIPs.

  ro*FUHTHra INFORMATION CONTACT!
  C. T. Helms. Chief. Control Programs
  Operations Branch (MD-is) Research
  Triangle Park, North Carol*, ™
  (919) 541-5226. FTS 629-5228.
   Dated: July 10,1980.
  David C. Hawkins,

                         r- *** «*
 IF* Doe. ohitwi RM 7-a^a tu ln|
 •HUNG COM «€
ACTION: Notice of public hearing on
   — -™» \****i i* *-trt«.i.
   SUPPLEMENTARY INFORMATION:

   I Background and Discussion
    Section 209(a) of the Clean Air Act as
  amended. 42 U.S.C. 7543{a) ("Act")
  Pro.^!de?in part: "No state or any
  political subdivision thereof shaU adopt
  or attempt to enforce any standard  P
  relating to control of emissions from   "
  new motor vehicles or new moto?
  vehicle engines subject to this part*  •  •
  for] require certification, inspection, or
  any other approval relating to the
  control of emissions*  • * as  condition
  precedent  to the initial retail sZtiitag
  (if any), or registration of such motor
 vehicle.motor vehicle engine, or
 equipment"
   Section 209(b)(l) of the Act  requires
 the Administrator, after notice and
 opportunity for public hearing, to waive
 application of the prohibitions of section
 209 to any State which had adopted
 standards (other than crankcase
 emission standardsj.for the control of
 emissions from new motor vehicles or
 new motor vehicle engines prior to
March 30.1988. if the State determines


              A-9
                                          ,      fion "   re*pect to
                                     and 1981 model year AMC passenger
                                     cars, and announced a public hearing to
                                     reconsHier the earlier LDT/MDV waiver
                                     decisions in light of AMC v. Slum The
                                     notice further provided that EPA wo"ld
                                     consider at the public hearing anv T  .w
                                     waiver requests filed by California   , or
                                     before July 7. 1980 to cover amended
                                     NO, standards and enforcement
                                     procedures for 1980 and later model
                                     year passenger cars and 1981 and later
                                     year passenger cars and 1981 and later
                                     year model year iDTs and MDVs
                                     manufactured by AMC
                                      In a June 13. 1980 letter to the
                                    Adminisfrator, CARS notified EPA that
                                    it had taken several actions to revise   '
                                    t^lifornia s new motor vehicles
                                             m     Pre8ram ta resP°n»
                                           v. Blum. GARB requested a
                                   waiver of Federal preemption for the
                                   following items:
                                   s ,3nrfm,enU> t0 
                                   standards and test procedures for 1980

-------

-------
             APPENDIX B:  TRE EQUATION AND COEFFICIENT DEVELOPMENT

                           FOR THERMAL INCINERATORS  AND FLARES


 B.I  INTRODUCTION













B.2  INCINERATOR  TRE INDEX EQUATION


     This section presents the method used to develop the incinerator TRE

-dex elation and an exaraple calculation of the  incinerator"E lex
B.2.1   Inci
                             uation Devplnpmont
c    nno  h      t                          e   COS  «' "«• -I- y
co^b n,ng  the equat.ons for each component of the annuaHzed costs   The

  ua ;r, °;nr r;alized cost component  -  sh°- •« ^   ».
  bo cos"    elT   I a"nUal1Zed CaPUa1  C°StS' "W1-"^ «« costs
                                                               costs,
                                               •  scrubber
             "
     results in an equation with the following  form:



     1
     ETQC [a


     f (Ys)0-]
                                          + ^^           Q g


          0-5                                  S        T
               treafOWate                                   of
                                 B-l

-------
     TRE  = Total resource effectiveness index value.

      Q   = Vent stream flowrate (scm/min), at a standard temperature of
            20 C.

       HT = Vent stream net heating value  (MJ/scm), where the net enthalpy
            per mole of vent stream is based on combustion at 25 C and
            760 mm Hg, but the standard temperature for determining the
            volume corresponding to one mole is 20 C, as in the definition
            of Qs.

     ETQC = Hourly emissions of total organic compounds reported in kg/hr
            measured at full operating flowrate.

       Y  = Q  for all vent stream categories listed in Table B-l except for
            Category E vent streams where Y  = (Q )(HT)/3.6.
                                           SSI'
where for a vent stream flowrate (scm/min) at a standard temperature of 20°C
that is less than 14.2 scm/min:

     TRE = Total resource effectiveness index value.

      Qs = 14.2 scm/min

      HT = (FLOW)(HVAL)/14.2

where:

     FLOW = Vent stream flowrate (scm/min), at a temperature of 20°C.

     HVAL = Vent stream net heating value (MJ/scm), where the net enthalpy
            per mole of vent stream is based on combustion at 25 C and
            760 mm Hg, but the standard temperature for determining the
            volume corresponding to one mole is 20 C,  as in the definition
            of Qs.

            Hourly emissions of total  organic compounds reported in Kg/hr
            measured at full  operating flowrate.

       Y  = Q  for all vent stream categories listed in Table B-l except for
            Category E vent streams where Y  = (Q_)(HT)/3.6.
                                           S     SI'

     The coefficients a through f are functions of incinerator design

parameters,  such as temperature,  residence time,  supplemental  fuel  require-

ments,  etc.   There are six different design categories of incinerators used
                                     B-2

-------
        Vent Stream Flowrate (scm/min)
                                                                                                   '  '
       14-2 < Q  < 18.8
       18.8 < Qs  < 699
       699  < Q*  < 1400
      1400  < Q  < 2100
      2100  < Q*  <  2800
      2800  < Qs  <  3500
 14'2
 18.8 <
 699
1400
2100
2800
              «  <  18-8
             Q  <  699
           <  Q3  <  uoo
          < Q  < 2100
          < q* < 2800
          < Qs < 3500
     14-2 1  Q,  < 1340
    1340  <  Q5  < 2690
    2690
Qs - Vent Stream Flowrate (,c./mtn>
    u-2 < Q  < 1340
   1340  < Q  £ 2690
   2690  < Q* < 4040
  	_

DESIGN CATEGORY D.

   - Vent  Stream Flowrate (scm/min)
19.18370
20.00563
39.87022
59.73481
79.59941
99.46400
0.27580
0.27580
0.29973
0.31467
0.32572
0.33456
0.75762
0.30387
0.30387
0.30387
0.30387
0.30387
                                                                                        -0.13064
                                                                                        -0.13064
                                                                                        -0.13064
                                                                                        -0.13064
                                                                                        -0.13064
                                                                                        -0.13064
18.84466
19.66658
39.19213
58.71768
78.24323
97.76879
0.26742
0.26742
0.29062
0.30511
0.31582
0.32439
0 . 20044
-0.25332
-0.25332
-0.25332
-0.25332
-0.25332
                                   0
                                   0
                                   0
                                   0
                                   0
                                   0
                                             8.54245
                                            16.94386
                                            25.34528
  0.10555
  0.11470
  0.12042
  0.09030
  0.09030
  0.09030
                                                 9.25233
                                               18.36363
                                               27.47492
                                                         0.06105
                                                         0.06635
                                                         0.06965
               0.31937
               0.31937
               0.31937
               -0.16181
               -0.16181
               -0.16181
                                                0
                                                0
                                                0
                                                0
                                                0
                                                0
                                 0
                                 0
                               , 0
                                 0
                                 0
                                 0
                                        0.01025
                                        0.01025
                                        0.01449
                                        0.01775
                                        0.02049
                                        0.02291
                       0.01025
                       0.01025
                       0.01449
                       0.01775
                       0.02049
                       0.02291
 DESIGN CATEGORY B.

 Qs - Vent Stream Flowrate (scm/min)
-0.17109
-0.17109
-0.17109
                                                                                                                0.01025
                                                                                                                0.01449
                                                                                                                0.01775
                     0.01025
                     0.01449
                     0.01775
   14-2 < Q  < 1180
  1180  < Qs < 2370
  2370  < QS < 3550
                                           6.67868
                                          13.21633
                                          19.75398
0.06943
0.07546
0.07922
                                                           „
0.02582
0.02582
0.02582
                    0.01025
                    0.01449
                    0.01775
                                               6.67868
                                              13.21633
                                              19.75398
                                                                                    0.00707
                                                                                    0.00707
                                                                                    0.00707
                                       0.02220   0.01025
                                       0.02412   0.01449
                                       0.02533   0.01775
                                                     B-3

-------
in the costing algorithm.  These design categories and category parameters are
discussed in Chapter 8 of the BID for the Air Oxidation Processes in SOCMI
(EPA-450/3-82-001a).  Table B-2 presents the updated heating values and
flowrate intervals associated with each category.  Substituting the design
values into the general equation allows values for coefficients a through f to
be derived for each design category.  This derivation is included in Docket
item No. IV-B-15.
     The results of this deriviation are summarized in Table B-l.  As shown,
the coefficients are divided into six incinerator categories.  Under each
design category listed in Table B-l, there are several intervals of vent
stream flowrate.  Each flowrate interval is associated with a different set of
coefficients.  The first flowrate interval in each design category applies to
vent streams with a flowrate corresponding to the smallest control equipment
system easily available without special custom design.
     The remaining flowrate intervals in each design category apply to vent
streams which would be expected to use two, three, four, or five sets of
control equipment, respectively.  These flowrate intervals are distinguished
from one another because of limits to prefabricated equipment sizes.

B.2.2  Example Calculation of an Incinerator-based TRE Index Value for a
       Facility

     This section presents an example of use of the TRE index equation.  The
example distillation vent stream has the following characteristics:

     1.   Qs    - 284 scm/min
     2.   HT    = 0.37 MJ/scm
     3.   ETOC  =76.1 kg/hr.
     4.   Ys    - 284 scm/min.
     5.   No halogenated compounds in the vent stream.

     Based on the stream heating value of 0.37 NJ/scm, Category B is the
applicable incinerator design category for this stream.  The flowrate is
284 scm/min,  and therefore the coefficients for the first flowrate interval
under Category B are used.  The coefficients for Category B, flow interval #1
are:
                                     B-4

-------
       TABLE B-2.  MAXIMUM VENT STREAM FLOWRATES AND NET HEATING VALUE
                    CHARACTERISTICS FOR EACH DESIGN CATEGORY
                ... .                                    Maximum Process Vent
                Minimum Net         Maximum Net         Stream Flowrate at
               Hewing Value       Heating Value        Incinerator Inlet
Category         (MJ/scm)*           (MJ/scm)*            (io5 scm/min)
Al
A2
B
C
D
E
0
3.5
0
0.48
1.9
3.6
3.5
-
0.48
1.9
3.6
-
0.70
0.70
1.34
1.34
1.18
1.18
These values are based on process vent stream conditions.
                                   B-5

-------
     a = 8.54
     b = 0.106
     c = 0.090
     d = -0.171
     e =  0
     f =  0.010

The TRE equation is:

       1
TRE = ETQC [a + b(Qs)°'88 + c(Q$) + d(Qs)(HT) + e(Q$ °'88)(HT °'88)
      f (Qs)°'5]

     TRE = (.013)[8.54 + 0.106 (284)°*88 + (0.090)(284)(-0.171)

           (284)(.37) + 0 + 0.010)(284)°'5]

     TRE = 0.111 + 0.199 + 0.332 - 0.236 + 0 + 0.002

     TRE = 0.408

Since the calculated TRE index value of 0.408 is less than the cutoff value
of 1.0, this facility would be required to reduce VOC emissions by 98 weight-
percent or to 20 ppmv because the cost of incineration is considered to be
reasonable.  Because the TRE index is a ratio of two cost-effectiveness
values, it is possible to calculate cost effectiveness for controlling any
vent stream given its TRE index value.  The TRE index value of the facility
is multiplied by the reference cost effectiveness $l,900/Mg as follows:
     TRE = 0.408
     Reference cost effectiveness = $l,900/Mg
     Cost effectiveness for example stream = (0.408)(1,900) = $775/Mg of
      VOC removed
                                     B-6

-------
  If  the TRE  index value for this example were above 1.0, the flare-based TRE
  equation  (see Section B.3) would be used to calculate the flare-based TRE
  index because flares can be applied to nonhalogenated vent streams.  If the
  flare-based TRE index were less than 1.0, this facility would have to reduce
  VOC emissions by 98 weight-percent or to 20 ppmv, whichever is less stringent.
  If the flare TRE index were also above 1.0, or if the stream contained
  halogenated compounds so a flare could not be used, then no further controls
  would be required.

  B.3  FLARE SYSTEM TRE DEVELOPMENT

      This section presents the development of the flare TRE index equation,
 verification of the equation,  and an example calculation of the flare TRE
 index.
                                     t
 B-3.1   Development of the Flare TRE Index Equation

     The  flare TRE index  equation was  developed  by selecting a  general  form
 for  the equation  which  contained  the stream characteristics of  flowrate,
 heating value,  and  VOC  emission rate as the  independent  variables,  and  the TRE
 index as  the dependent  variable, and fitting this  equation to the TRE index
 values calculated  from  the annualized cost equations.  The form of  the TRE
 index equation  for  flaring had  to be selected so that an accurate prediction
 of the TRE index could  be obtained  for a given set of vent stream
 characteristics.  The form of the flare TRE index  equation selected was the
 same as the  form used in the proposed standards for Distillation Operations
 (50  FR 20446).

     The general form of the equation is as follows:

             [a(QJ  + b(Qsr + c(QJ(HT) + dfE^) + e]
where:
      TRE - total resource effectiveness index value
      Qs   = vent stream flowrate (scm/min)  at a standard temperature of 20°C
                                     B-7

-------
           = vent stream net heating value (MJ/scm) where the net enthalpy
             per mole of vent stream is based on combustion at 25°C and
             760 mm Hg, but the standard temperature for determining the
             volume corresponding to one mole is 20°C as in the definition
             of Qs.
             hourly emission rate of total organic compounds reported  in
             kg/hr measured at full operating flowrate.
a, b, c, d, and e are coefficients.
     The coefficients for the flare TRE index equation were developed with a
regression analysis procedure.  The regression analysis procedure used is the
General Linear Model (GLM) procedure of the Statistical Analysis System
Institute, Inc., Raleigh,.North Carolina.  The development of the
coefficients involved three steps: (1) formation of an appropriate data base
for the regression; (2) calculating a TRE index value for each set of vent
stream characteristics in the data base with the revised flare costing
procedure described in Docket Entry IV-B-8; and (3) using the GLM procedure
to regress TRE index values against the vent stream characteristics.
     The distillation NSPS National Emissions Profile (NEP) and the reactor
processes Emissions Data Profile (EDP) were used as the data base for the
regression analysis.  Adding the reactor processes EDP was judged to be
appropriate because of two significant similarities with the distillation
NEP: (1) the vent stream characteristics represented in the two data bases
are similar; and (2) identical or similar synthetic organic chemicals are
produced by both reactor processes and distillation operations.
     After the data base was formed,  the cost of controlling VOC emissions
using flares was calculated from the annualized cost equations for each
facility with nonhalogenated vent streams in the NEP and EDP.   These costs
were divided by the amount of VOC emissions reduced by flaring (i.e.,  98
weight-percent) to obtain a value for cost of control  per megagram of VOC
reduced.  Next, these values were divided by a TRE cutoff of $l,900/Mg to
obtain a TRE index value for each facility.  The TRE index value and vent
stream characteristics for each facility were then input to the GLM
regression program.
                                     B-8

-------
  TRE inH    7ntS       e"e0Ped f°r "Ch te™ fn the TRE """"on "'Ing the
    E ,  dex value as the dependent variable and  the vent stream character s
  «  depen en   ar1 ables.  The flare TRE coefficients are shown  in Tab,   "

  wi   a  i     r    ;  7 deveioped for each °f
  (0 Btu s f,   r?V       WUh  heatjn9 Va'U
   lues  for  b"    ' C°mbUSt10n WUh  3 fUre f°r
  values at or above  11.2 MJ/scm (300 Btu/scf).  The
           w,th vent stream heating values at  or above 11.2


    ar:StChf)heatThereT ' """ "* °? «  ^"'^
 streams W1th heating values at or above 11.2 MJ/scm.

 B'3'2  ^are TRF Coefficients •tfa>.1f1catinn


 capabnitv'oT TRE T*10" and C°effiCie"ts "" «-1»- to ensure their
 capability of accurately predicting the TRE Index value for a facility fr™

  E6 cVoelStr"; Cha^iSt1"-  * -'fication  procedure for^,

u  no  th      i   lnV°1Ved ^^ Ste"S: U) "'culati- " • TRE index
   n   he  newly derived TRE elation for each  facility 1n the data ^.

             „ . „.„ ,,, „„„.   „.
                                                    „
                                 B-9

-------
TABLE B-3   DISTILLATION OPERATIONS NSPS TRE COEFFICIENTS FOR VENT STREAMS
                                CONTROLLED BY A FLARE
MJ/scm
                        a          b          c          d          e

         MJ/scm        2.25      0.288      -0.193    -0.0051      2.08
                                     B-10
                       0.309     0.0619     -0.0043   -0.0034      2.08

-------
TABLE B-4.   TRE INDEX VALUES GENERATED USING TRE COEFFICIENTS AND THE
            ALGORITHM NET HEATING VALUE GREATER THAN OR
Flowrate
(scf/min)
— • 	
70.00
1.45
1.20
2.04
1.39
.20
0.30
6.60
— ^— — _ — _^^____
Heat Content
(Btu/scf)
	 • 	
323.00
903.00
1024.00
1024.00
966.00
2778.00
4978.00
1286.00
VOC
db/hr)
	 ___^_
6.60
1.60
3.81
6.47
6.04
2.00
4.90
3.00
» TRE
INDEX VALOE
Algorithm Coefficients
	
0.88
2.90
1.22
0.72
0.77
2.31
0.95
1.57
	 — — 	
0.91
2.91
1.22
0.72
0.77
2.31
0.94
1.57
Percen
D if fere
• 	 .
3.39
0.43
0.03
-.08
-.06
-.01
-.49
0.08
                           B-ll

-------
values calculated with the TRE equation and those calculated using the cost
algorithm for the same facility as described above.
     For vent streams with heating values below 11.2 MJ/scm there was poor
agreement initially between the algorithm and TRE equation.  Therefore, those
data points resulting in very higji TRE indexes were  removed after the initial
verification procedure was performed because they caused the poor agreement
at TRE index values near the cutoff.  After removal  of those data points, the
TRE coefficients for vent stream heating values less than 11.2 MJ/scm were
recalculated and the verification procedure was undertaken again.  The
percentage difference in the recalculated TRE index  values near the cutoff
ranged from 2.38 to -7.39.  Thus, it was concluded that the recalculated TRE
coefficients for vent streams with heating values below 11.2 MJ/scm provided
good agreement with the actual TRE index values.   Table B-5 presents a
comparison of TRE indexes near the cutoff for vent streams with heating
values below 11.2 MJ/scm.
     As a final verification step for vent streams with heating values below
11.2 MJ/scm, the recalculated TRE coefficients were  used to determine a TRE
index value for those data points which were removed after the initial
verification procedure was performed.  The percentage difference between the
TRE index values determined using the recalculated coefficients and the TRE
index values determined using the flare cost algorithm ranged from 2.29 to
-6.24.  Thus, it was concluded that the coefficients enable accurate
estimation of even those facilities with high TRE index values.  Table B-6
presents a comparison of TRE index values for those  vent streams with high
TRE index values.
     In summary, the flare TRE equations developed for this NSPS allow for
the calculation of TRE index values that are highly  correlated with the TRE
index values obtained from the costing algorithm.  The TRE equations do not
necessarily result in the best statistical fit between TRE values and vent
stream characteristics.  This is because the primary concern in developing
the equation and coefficients is to ensure very good agreement between the
TRE equation and cost algorithm for TRE's at or around the cutoff.
                                    B-12

-------
               TABLE B-5.  PERCENT DIFFERENCE BETWEEH TRE IMDEX VALUES GENERATED USIHG IRE EQUATION AND THE
                                  FLARE COST ALGORITHM NET HEATING VALUE LESS THAN  300 Btu/scf

Flowrate
(scf/mln)
17.00
75.00
50.40
4.40
22.60
11.30
68.70
7.57
27.30
4.20
88.00
7.50
2.40
17.90
15.00
80.00
Heat Content
(Btu/scf)
181.00
102.00
70.00
190.00
92.00
168.00
72.00
18.00
47.00
18.00
47.00
47.00
260.00
69.00
149.00
9.00
VOC
(Ib/hr)
16.00
6.10
16.90
4.00
10.50
5.23
26.30
5.00
8.50
28.50
2.50
4.00
4.00
8.00
6.60
19.60
TRE
Algorithm
0.38
2.12
0.65
1.22
0.73
1.16
0.50
1.14
1.02
1.04
0.57
2.23
1.17
0,92
0.96
0.87
INDEX VALUES
Coefficients
Recalculated
.37
2.12
0.65
1.24
0.69
1.07
0.50
1.16
0.99
1.06
.58
2.28
1.18
0.87
0.91
0.86
Percent Differences
Compared to Algorithm
Coefficients
Recalculated
-2.23
- .10
0.00
1.84
-5.16
-7.39
0.21
2.06
-3.63
1.29
0.68
2.38
1.05
-5.71
-5.2-9
-1.53
aTRE coefficients  derived from vent  streams with a heating value greater than 40 Btu/scf but less than 300 Btu/scf.
                                                   B-13

-------
      TABLE B-6.  PERCENT DIFFERENCES BETWEEN TRE INDEX VALUES GENERATED BY THE COST
                 ALGORITHM AND THE TRE EQUATION FOR VENT STREAMS WITH. HEATING
                                 VALUES LESS THAN 40 Btu/scf

Flowrare
(scf/min)
99.00
822.00
16.67
0.05
39.20
6.60
2.00
6.25
12.40
13.53
Heat Content
(Btu/scf)
0.00
0.00
4.00
36.00
4.00
8.00
0.00
9.00
0.00
0.00
VOC
(Ib/hr)
0.10
0.10
.37
0.10
0.18
.60
.003
.40
0.14
0.03
TRE INDEX VALUE
Algorithm
203
1325
21
46
61
9
1640
14
51
242
Coefficient.!*
202
1290
20
46
60
9
1658
14
48
228
Percent
Difference
- .91
-2.65
-5.16
0.19
-2.28
2.24
1.14
2.29
-6.24
-5.96
Equation coefficients were developed after excluding vent streams with heating values
less than 40 Btu/scf,
                                          B-14

-------
   B.3.3
                              f a Flare-Rased TRF TnH
                                                        Value for a Far*Tft
                             an example calculation for the
       3.
       4.
         .2.2.  The vent stream
  Qs   = 284 scm/min
  HT   =0.37 MJ/scm
  ETOC m7*-1 k9/hr
                                                          are as fonows:
      a
      b
      c
      d
      e
  No halogenated compounds  in  vent  stream.
   on the  stream h.^™  value Qf Q 3?  MJA(;m
        as  '-"
2.25
0.288
-0.193
-0.0051
2.08

     .
emlsslons by 98 weight-percent or below 20 ppmv
                                  B-15

-------

-------
|l. REPORT NO.
   EPA-450/3-83-005b
   ton          Operations  1n Synthetic Organic Chemical
   Manufacturing - Background Information for Final
   standards
7. AUTHOmSI
"' P6"?°?MING ORGANIZATION NAME AND ADDRESS'
   Office of Air Quality Planning and Standards
   U.  S.  Environmental Protection Agency
   Research Triangle Park, North Carolina  27711
12. SPONSORING AGENCY NAME AND ADDRESS
   Office of Air and Radiation, U. S. EPA
   Research Triangle Park. North Carolina
IS. SUPPLEMENTARY NOTES~ ThlS~ Ul

  S2M5^^ra^*^5^^;^SBr
  operations in the synthetic organic chemical  manufacturing industry.
  AasTSA/r          • - . .   i_n.   , .              _ ..            	J
16. ABSTRACT
                                            EPORT DATA
                                             rertnt be fan completing
                                                         9. RfCIfliNT-S ACCESSION NO!
                                                         B. ftiFOftT OATB
                                                                     _
                                                         ». PERFORMING ORGANIZATION CODE
                                                         •. PERFORMING ORGANIZATION REPORT NO
                                                           68-02-3058

                                                         3. TYP£ Of (WORT ANO PEBlOO COVEBEo
                                                           EPA/200/04
                                         27711

                                  	                i
                             KEY WORDS AND DOCUMENT ANALYSIS
                             ^	n—• i   	 _
                                                         _-___..

                                           b.lDENTIFIERS/QPEN 6NO6D TERMS
                                                                       COSATI f-ifld/Croup
                                           Air Pollution Control
	DESCBIfTOBS
 Air pollution
 Distillation unit operations
 Pollution control
 Standards of performance
 Organic chemical industry
 Volatile organic compounds  (VOC)
IB. O'STBisunoN STATEMENT"               	
  Unlimited -  available to the public free
  of charge from U. S. EPA Library (MD-35)
  Research  Triangle Park. N.C.   ?77JV   ^
 PAF«n,2220.UR...4.77,   P.cv.ou, CC.T.ON „ ,
                                          19 56CUHITY CLASS <
                                            Unclassified
                                          20 SECURITY CLASS tTiltspafti'
                                            Unclassified
                                                                    j21. NO. OF PAGES
                                                                       142
                                                                    (22. PRICE

-------

-------