United States
         Environmental Protection
         Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-450/3-92-006b
December 1992
         Air
&EPA    National Emission Standards
         for Hazardous Air Pollutants
         Compliance Extensions for
         Early Reductions -
         Background Information for
         Promulgated Standards


-------

-------
National Emission Standards
for Hazardous Air Pollutants
 Compliance Extensions for
      Early Reductions -
 Background Information for
   Promulgated Standards
            Prepared By
       Emission Standards Division
U. S. ENVIRONMENTAL PROTECTION AGENCY
        Office of Air and Radiation
   Office of Air Quality Planning and Standards
  Research Triangle Park, North Carolina 27711
           December 1992

-------
This  document 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, North Carolina 27711,
or from National  Technical Information Services, 5285 Port Royal
Road, Springfield,  Virginia 22161.
                              11

-------
             U.  S.  ENVIRONMENTAL PROTECTION AGENCY

                    Background Information
 for National Emission Standards for Hazardous Air Pollutants:
          Compliance Extensions for Early Reductions
                         Prepared by:
Bruce C. Jordan                                    (Date)
Director, Emission Standards Division
U. S. Environmental Protection Agency
Research Triangle Park, North Carolina  27711

1.   The promulgated standards establish rules for
     implementing the Early Reductions Program.
     Section 112(i)(5) of the Clean Air Act  (as amended  in
     1990)  ("the Act") allows existing sources to obtain a
     6-year extension of compliance from applicable
     Section 112(d) standards if the source achieves emission
     reductions of 90 percent or more of hazardous air
     pollutants  (95 percent or more for particulates) by
     certain dates specified by the Act.

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; and to the National Science Foundation; State
     and Territorial Air Pollution Program Administrators; the
     EPA's Regional Administrators; Local Air Pollution
     Control Officials; the Office of Management and Budget;
     and other interested parties.

3.   For additional information contact:

     Mr. David Beck, or
     Mr. Richard Colyer
     U.  S.  Environmental Protection Agency
     Emission Standards Division  (MD-13)
     Research Triangle Park, North Carolina  27711
     Telephone:   (919) 541-5421 or 541-5262
 4.
Copies of this document may be obtained from:
      U.  S.  EPA Library (MD-35)
      Research Triangle Park,  North Carolina
      Telephone:   (919)  541-2777

      National Technical Information Services
      5285 Port Royal Road
      Springfield,  Virginia  22161
                                        27711
                               iii

-------
                       TABLE OF CONTENTS
                                                          Page
CHAPTER 1.  SUMMARY       .......... 	  1-1

            1.1  SUMMARY OF  CHANGES  SINCE  PROPOSAL  . .  .  1-1
            1.2  SUMMARY OF  IMPACTS  OF                   "     .
                    PROMULGATED ACTION	  1-8
            1.3  REFERENCES	  .  1-8

CHAPTER 2.  SUMMARY OF PUBLIC  COMMENTS  	  2-1

            2.1  DEFINITION  OF SOURCE	2-1
            2.2  BASE YEAR EMISSIONS  	   2-31
            2.3  ENFORCEABLE COMMITMENTS	   2-39
            2.4  ALLOWABLE EMISSION  REDUCTIONS   ....   2-43
            2.5  DEMONSTRATION OF EARLY REDUCTIONS  . .   2-46
            2.6  HIGH-RISK POLLUTANTS  	   2-52
            2.7  STATE AUTHORITY   	2-100
            2.8  INTERFACE WITH TITLE  V PERMITS  . . . .  2-103
            2.9  INTERFACE WITH SECTION 112(G)
                   MODIFICATIONS	  2-111
           2.10  INTERFACE WITH TITLE  I PROVISIONS  . .  2-112
           2.11  TEST METHODS  AND PROCEDURES   	2-119
           2.12  MISCELLANEOUS COMMENTS  .........  2-130
           2.13  REFERENCES	2-143
                               IV

-------
                        LIST OF TABLES
1-1

1-2

1-3


1-4

2-1
LIST OF POLLUTANTS ADDED TO HIGH-RISK LIST   .

LIST OF POLLUTANTS DELETED FROM HIGH-RISK LIST

ADJUSTMENTS IN WEIGHTING FACTORS FOR
  SELECTED POLLUTANTS  .	•  -  •
LIST OF HIGH-RISK POLLUTANTS
LIST OF COMMENTERS ON THE PROPOSED
  EARLY REDUCTIONS RULE  	
Page

 1-3

 1-4


 1-4

 1-5


 2-2

-------

-------
                          1.0 SUMMARY

     The U. S. Environmental Protection Agency (EPA) proposed
rules governing compliance extensions for early reductions of
hazardous air pollutants (HAP's) on June 13, 1991
(56 FR 27338) under authority of Section 112(i)(5) of the
Clean Air Act (CAA) , as amended in 1990.1  Public comments
were requested in the Federal Register notice.  Sixty-seven
comment letters were received from industry, State and local
regulatory agencies, environmental groups, and other
interested parties.  The comments and subsequent responses
serve as the primary basis for revisions made to the rule
between proposal and promulgation.  The final rule was
promulgated December 29, 1992 (57 FR 61970).
1.1  SUMMARY OF CHANGES SINCE PROPOSAL
     The EPA has made several changes and clarifications to
the Early Reductions Program rule since proposal as a result
of comments received.  These changes and clarifications were
made in the following subject areas:
     •  High-risk pollutants;
     •  Definition of source;
     •  Averaging of gaseous and particulate emissions;
     •  Use of average emission factors for equipment leak
        emissions;
     •  Method 301; and
     •  Language in the regulation.
     A summary of these changes is given below.
 1.1.1  High-risk Pollutants
     The high-risk pollutant list now contains 47 pollutants,
 as opposed to 35 in the proposed rule.  The EPA revised the
 methodology  used to determine high-risk pollutants  and the
                              1-1

-------
weighting factors assigned to some pollutants.  As a result of
these changes, 17 pollutants were added to the list and
5 pollutants were deleted.  The weighting factors for five
pollutants were also adjusted.  Table 1-1 lists the pollutants
added to the high-risk list and Table 1-2 lists the pollutants
deleted from the list.  Table 1-3 shows the adjustments made
to the weighting factors for the five pollutants.  Table 1-4
is the final list of high-risk pollutants and their adjusted
weighting factors.
     Two other provisions concerning high-risk pollutants were
added to the final rule.  First, a provision was added to
ensure that there were no increases in radionuclide emissions
as a result of the emission reduction demonstration.  Second,
a provision was added that will "grandfather" sources that
already had approved enforceable commitments or alternative
emission limits specified by permit.  These sources will not
have to revise post-reduction demonstrations if a pollutant is
either newly-listed as a HAP under Section 112(b) or added to
the high-risk list.
1.1.2  Definition of Source
     Proposed paragraph (a)(4) of section 63.73 has been
deleted from the source definition.  Paragraph (a)(4) is a
subset of (a)(5), and thus redundant.  Selecting unrelated
emission points to constitute a source will still be allowed,
provided that the source created emits a significant quantity
of HAP's as defined in the regulation.
1.1.3  Averaging of Gaseous and Particulate Emissions
     The rule has been revised with respect to the combined
percent reduction that can be applied to sources with both
gaseous and particulate HAP's.  The provision in the final
rule only applies to any emission point, as opposed to the
entire source, that emits both gases and particulates.  An
emission point that emits solely a gaseous HAP can not be
combined with another emission point that emits particulate
                              1-2

-------
   TABLE  1-1... LIST OF POLLUTANTS ADDED TO HIGH-RISK LIST
CAS  NO.
     Chemical
Weighting Factor
   53963 .
  532274
  334883
   96128
   79447
  122667
  151564
   77474
       0
   60344
       0
  684935
   62759
   56382
  7803512
  7723140
  8001352
2-Acetylaminofluorene
2-Chloroacetophenone
Diazomethane
1,2-Dibromo-3-chloropropane
Dimethyl carbamoyl chloride
1,2-Diphenylhydrazine
Ethylenimine (Aziridine)
Hexachlorocyclopentadiene
Manganese compounds
Methyl hydrazine
Nickel compounds
N-Nitroso-N-methylurea
N-Nitrosodimethylamine
Parathion
Phosphine
Phosphorus
Toxaphene  (Chlorinated Camphene)
          100
          100
           10
           10
          100
           10
          100
           10
           10
           10
           10
          1000
           100
            10
            10
            10
           100
*Chemical Abstract Service
                              1-3

-------
  TABLE  1-2.   LIST OF  POLLUTANTS  DELETED FROM  HIGH-RISK  LIST
   CAS No.
    Chemical
            Weighting Factor
98077
126998
79345
584849
75354

Benzotrichloride
Chloroprene
1,1,2, 2-tetrachloroe'thane
2,4 -toluene diisocyanate
Vinylidene chloride
( 1 , 1-Dichloroethylene )
10
10
10
10

10
TABLE 1-3,
ADJUSTMENTS IN WEIGHTING FACTORS FOR SELECTED
                 POLLUTANTS
   CAS No.
    Chemical
Weighting Factor Adjustment
    107028
     57749
     76448
    118741
         0
    Acrolein
    Chlordane
    Heptachlor
    Hexachlorobenzene
    Mercury Compounds
             10  to  100
             10  to  100
             10  to  100
             10  to  100
             10  to  100
                              1-4

-------
TABLE 1-4.  LIST OF HIGH-RISK POLLUTANTS
CAS No.
53963
107028
79061
79107
107131
0
1332214
71432
92875
0
542881
106990
0
57749
532274
0
107302
0
334883
132649
96128
• 111444
79447
122667
106934
151564
75218
76448
Chemical Weighting
2-Acetylaminofluorene
Acrolein
Acrylamide
Acrylic acid
Acrylonitrile
Arsenic compounds
Asbestos
Benzene
Factor
100
100
10
10
10
100
100
10 .
Benzidine 1000
Beryllium compounds
Bis ( chloromethyl ) ether 1 /
1,3 -Butadiene
Cadmium compounds
Chlordane
2-Chloroacetophenone
Chromium compounds
Chloromethyl methyl ether
Coke oven emissions
Diazomethane
Dibenzofurans
1 , 2-Dibromo-3-chloropropane
Dichloroethyl ether
Dimethyl carbamoyl chloride
1 , 2-Diphenylhydrazine
Ethylene dibromide ( Dibromoethane )
Ethylenimine (Aziridine)
Ethylene oxide
Heptachlor
10
000
10
10
100
100
100
10
10
10
10
10
10
100
10
10
100
10
100
1-5

-------
  TABLE 1-4.  LIST OF HIGH-RISK POLLUTANTS  (Concluded)
CAS No.  Chemical
Weighting Factor
 118741  Hexachlorobenzene      .                  100
  77474  Hexachlorocyclopentadiene                 10
 302012  Hydrazine                                100
      0  Manganese compounds                       10
      0  Mercury compounds                        100
 101688  Methylene diphehyl diisocyanate (MDI)      10
  60344  Methyl hydrazine                          10
 624839  Methyl isocyanate                    .     10
      0  Nickel compounds                          10
  62759  N-Nitrosodimethylamine                   100
 684935  N-Nitrosp-N-methylurea                 1,000
  56382  Parathion                                 10
  75445  Phosgene                                  10
7803512  Phosphine                                 10
7723140  Phosphorus                                10
  75558  1,2-Propylenimine (2-Methyl aziridine)    100
1746016  2,3,7,8-Tetrachlorodibenzo-p-dioxin  100,000
8001352  Toxaphene (chlorinated camphene)          100
  75014  Vinyl chloride                            10
                           1-6

-------
HAP within a source for the purposes of averaging the emission
reduction.
1.1.4  Use of EPA Average Emission Factors for Equipment Leak
Emissions
    The EPA  has changed the rule to allow the use of the EPA's
average emission factors for equipment leaks to determine base
year emissions only if no reductions in equipment leak
emissions are claimed as part of the reduction demonstration.
The equipment leak source in Appendix B has also been
clarified to be specific to the individual component types.
1.1.5  Method 301
    A number of changes and corrections to Method 301 have
been made as a result of public comments.  The applicability
of the method has been broadened to cover other media.
1.1.6  Miscellaneous Changes to the Rule
    A number of minor changes  have been made  to  the  proposed
rule.  Two changes affect the determination of base year and
post-reduction emissions.  First,  the definition of  "actual
emissions" has been clarified to exclude startup and shutdown
emissions due to malfunctions.  Second, it is no longer a
requirement  to provide evidence of production curtailments or
shutdowns for post-reduction demonstrations.  Such reductions
will  become  conditions of the Title V permit.
     Other minor changes include:   (1)  the suggested statement
of commitment in the  enforceable commitment has  been revised
to be more consistent with the actual requirements;   (2) the
 length of time  allowed to provide  source test results to
 support the  post-reduction demonstration has  been increased  to
 120 days from 90  after the permit  submittal deadline;  (3)
 reference to Part 71  federal permitting  rules has been  deleted
 because they have not been issued; and (4)  the term "post-
 control" has been changed to  "post-reduction" throughout the
 rule to recognize that reductions  can be achieved through
 means other than control technology.
                               1-7

-------
 1.2  SUMMARY OF  IMPACTS OF PROMULGATED ACTION
     The Early Reductions  Program is a voluntary program.   The
 Program provides optional means  of  complying with future
 standards developed under Section 112 of the CAA.
     Because the Program is voluntary, no environmental,
 economic, or energy impacts are  directly associated with the
 rule.   However,  should an owner  or  operator choose to take
 advantage of the Early Reductions provisions, the public
.health  and the environment would gain from reductions in HAP' s
 before  such reductions would otherwise be required by the
 standard-setting process.
     In addition, any economic  and energy  impacts will occur
 sooner  at sources making early reductions of HAP's.  However,
 because of the greater flexibility  to apply cost-effective
 control measures than would be available under  a Section
 112(d)  standard, the  owner or operator will realize a net
 savings, with respect  to meeting  a Section 112(d) standard.
     At this  time,  it is not  possible to quantify early
 reductions benefits or impacts.  These will depend on how many
 source  owners  or operators are able to take advantage of the
 Early Reductions Program, the types of sources, which
 reduction measures will be implemented, and the requirements
 of otherwise applicable Section  112(d) standards.
 1.3  REFERENCES
 1.   Federal  Register. National  Emission  Standards  for
     Hazardous Air Pollutants for Source Categories:  Proposed
     Regulations Governing Compliance Extensions for Early
     Reductions of Hazardous  Air  Pollutants.  56  FR 27338-27374.
     June 13,  1991.
                              1-8

-------
                 2.0  SUMMARY OF PUBLIC COMMENTS
     A total of 67 comment letters were received.  In addition,
comments were received at a public hearing held following
proposal of the rule.  All of the comment letters and public
statements have been recorded and placed in the docket (Docket
No. A-90-47).  A list of commenters, their affiliation, and the
docket item number assigned to their correspondence is shown in
Table 2-1.  The comments have been organized into the following
categories:
       2.1
       2.2
       2.3
       2.4
       2.5
       2.6
       2.7
       2.8
       2.9
       2.10
       2.11
       2.12
            Definition of Source;
            Base  Year Emissions;
            Enforceable  Commitments;
            Allowable Emission  Reductions;
            Demonstration of  Early Reduction;
            High-Risk Pollutants.;
            State Authority;
            Interface with  Title V Permits;
            Interface with  Section  112(g)  Modifications;
             Interface with  Title I Provisions;
            Test Methods and Procedures;  and
            Miscellaneous Comments.
2.1  DEFINITION OF SOURCE
2.1.1  Comment t  A number of commenters contended that the
definition of source is too broad,  especially in paragraphs 4
and 5 (IV-D-06, IV-D-20, IV-D-22).   Other commenters felt the
definition should be consistent with the definition proposed in
Title V of the CAA to avoid confusion (IV-D-13, IV-D-14, IV-D-16,
IV-D-31).  One commenter suggested restricting the definition to
Section 63.73(a)(l) and  (3)(IV-D-60).  Finally a
                                2-1

-------
     TABLE 2-1.
LIST OF COMMENTERS'. ON THE PROPOSED
       EARLY REDUCTIONS RULE
Docket Item Number5


IV-D-2
IV-D-2
IV-D-3
IV-D-4
IV-D-5
IV-D-6
 IV-D-7
     Commenter and Affiliation
     Mr. Gordon D. Strickland
     Vice-President
     Chemical Manufacturers Association
     2501 M Street, NW
     Washington, DC  20037

     Mr. Edwin T. Still
     Kerr-McGee Corporation
     Kerr-McGee Center
     Oklahoma City, Oklahoma  73125

     Ms. Mary Anne Hunter
     Environmental Coordinator
     Arvin Industries, Inc.
     1531 13th Street
     Columbus, Indiana  .47201

     Mr. George Smith
     Chair, Air Quality
     Lone Star Chapter, Sierra Club
     6014 Woodbrook
     Houston, Texas  77008

     U.V. Henderson
     General Manager
     Environment and Product Safety
     Texaco, Inc.
     PO Box 509
     Beacon, New York  12508

     Ms. Kathleen F. Dalton
     NY State Department of
     Environmental Conservation
     50 Wolf Road
     Albany, New York  12233

     Mr. Roger Etter
     Koch Industries, Inc.
     PO Box 2256
     Wichita, Kansas  67201
                                2-2

-------
     TABLE 2-1.  LIST OF COMMENTERS'ON THE PROPOSED
                   EARLY REDUCTIONS RULE (continued)
Docket Item Numbera

IV-D-8
IV-D-9
IV-D-10
 IV-D-11
 IV-D-12
 IV-D-13
 IV-D-14
Commenter and Affiliation

Ms. Dorothy P. Bowers
Executive Director
Corporate Environmental Resources
Merck & Co., Inc.
PO Box 2000
Rahway, New Jersey  07065-0900

Mr. Michael C. Thompson
Manager, Government Relations
Whirlpool Corporation
Benton Harbor, Michigan  49022

Mr. George P. Ferreri
Director, Air Management Division
State of Maryland Dept. of the Environment
2500 Broening Highway
Baltimore, Maryland  21224

Mr. Charles D. Bennett
Environmental Coordinator
Ashland Petroleum Company
PO Box 391
Ashland, Kentucky  41114

Mr. John Mancini
American Electronics Association
 1225  Eye Street, SW
Washington,  DC   20005

Mr. Carl S.  Pavetto        >
Bureau Chief,  Bureau of Air Mgmt
 State of  Connecticut
 Dept. of  Environmental Protection
 165 Capitol Ave.
 Hartford,  Connecticut  06106

 Mr. Richard A.  Valentinetti, Director
 Air Pollution Control Division
 State of Vermont Agency of Natural
 Resources
 Montpelier, Vermont
                                 2-3

-------
     TABLE 2-1.  LIST OF COMMENTERSvON THE PROPOSED
                   EARLY REDUCTIONS RULE  (continued)
Docket Item Numbera

IV-D-15
IV-D-16
IV-D-17
IV-D-18
IV-D-19
IV-D-20
IV-D-21
Commenter and Affiliation

Mr. Robert P. Miller
Chief, Air Quality Division
State of Michigan
   Dept. of Natural Resources
PO Box 30028
Lansing, Michigan  48909

Mr. David R. Jordan, ALAPCO President
Mr. Richard E. Grusnick,
   STAPPA President
STAPPA/ALAPCO
444 North Capitol Street, NW
Washington, DC  20001

Mr. John L Wittenborn
Mr. William M. Guerry, Jr.
Collier, Shannon & Scott
3050 K,Street, NW
Washington; DC  20007

Mr. John A. Dege
Issue Manager, Clean Air Act
DuPont Chemicals
Wilmington, Delaware  19898

Mr. Thomas X. White
Assistant Vice-President
Pharmaceutical Manufacturers Assoc.
1100 Fifteenth Street NW
Washington, DC  20005

Mr. Dennis R. Lunderville
Director, Air Resources Division
State of New Hampshire
Dept. of Environmental Sciences
64 North Main Street
Concord, New Hampshire  03302-2033

Ashok K. Jain
National Council of the Paper Industry
   for Air and Stream Improvement,Inc.
   (NCASI)
PO Box 14483
Gainesville, Florida  32604
                                2-4

-------
     TABLE 2-1.
LIST OF COMMENTERS: ON THE PROPOSED
  EARLY REDUCTIONS RULE (continued)
Docket Item Numbera

IV-D-22
IV-D-23
IV-D-24
 IV-D-25
 IV-D-26
 IV-D-27
 IV-D-28
     Commenter and Affiliation

     Mr. Michael J. Bradley
     Executive Director
     Northeast States for Coordinated
     Air Use Management (NESCAUM)
     85 Merrimac Street
     Boston/ Massachusetts  02114
                                                  94120-7924
     R.L. Arscott
     Chevron Corporation
     PO Box 7924
     San Francisco, California

     Mr. Brian L. Taranto
     Exxon Chemicals Americas
     PO Box 3272
     Houston, Texas  77253-3272
      Mr.  Roger D.  Randolph
      State of  Missouri
      Department of Natural  Resources
      PO Box 176
      Jefferson City,  Missouri  65102

      Mr.  Steve Spaw,  Executive Director
      Texas Air Control  Board
      12124 Park 35 Circle
      Austin, Texas  78753

      Mr.  Ralph F.  Hall, Counsel
      Energy and Environmental Affairs
      Eli Lilly and Company
      Lilly Corporate Center
      Indianapolis, Indiana  46285
      Mr. David W. Dunn
      Sterling Chemicals
      PO Box 1311
      Texas City, Texas
                                          77592-1311
                                2-5

-------
     TABLE 2-1.
LIST OF COMMENTERS'rON THE PROPOSED
  EARLY REDUCTIONS RULE (continued)
Docket. Item Number3-

IV-D-29
IV-D-30
IV-D-31
IV-D-32
IV-D-33
IV-D-34
IV-D-35
     Commenter and Affiliation

     Mr. Lawrence C. Tropea
     Director
     Corporate Environmental Control
     Reynolds Metals Company
     PO Box 27003
     Richmond, Virginia  23261-7003

     Mr. Thomas J. Carr
     Vice-President, Technical Affairs
     Motor Vehicle Manufacturers Association
     7430 Second Avenue, Suite 300
     Detroit, Michigan 48202

     Mr. James D. Boyd
     Executive Officer
     State of California Air Resources Board
     1102 Q Street
     Sacramento, California  95812

     E. J. Mazeski
     Vice-President and Secretary
     PPG Industries
     One PPG Place
     Pittsburgh, Pennsylvania  15272

     Mr. Charles D. Malloch
     Director, Regulatory Management
     Monsanto Company
     800 N. Lindbergh Boulevard
     St. Louis, Missouri  63167

     Terry F. Yosie
     Vice-President
     American Petrolaum Institute
     1220 L Street, NW
     Washington, DC  20005

     Mr. Robert C. Kaufmann
     Director, API/NFPA Air Quality Program
     American Paper Institute
     &  National Forest Products Association
     1250 Connecticut Avenue, NW, Suite  210
     Washington, DC  20036
                                2-6

-------
     TABLE 2-1.
LIST OF COMMENTERS: ON THE.PROPOSED
  EARLY REDUCTIONS RULE (continued)
Docket Item Numbera

IV-D-36
IV-D-37
IV-D-38
 IV-D-39
 IV-D-40
 IV-D-41
 IV-D-42
     Commenter and Affiliation

     Ms. Mary J. Legatski
     Director, Government Relations
     Synthetic Organic Chemical
     Manufacturers Association, Inc.
     1330 Connecticut Ave, NW, Suite 300
     Washington, DC  20036-1702

     Ms. Sara Schotland
     Cleary, Gottlieb/ Steen & Hamilton
     1752 N Street, NW
     Washington, DC  20036

     Emil Romagnoli
     Director, Government Affairs
     ASARCO Incorporated
     1155 Connecticut Ave., NW
     Washington, DC  20036

     Mr. David Doniger
     Senior Attorney
     Natural  Resources Defense Council
      1350 New York Avenue, NW
     Washington, DC  20005

     Tipton R. Tyler
     Basic Acrylic Monomer Manufacturers
      1330 Connecticut Avenue, NW
     Washington, DC   20036-1702

      Mr.  Gordon D. Strickland
      Vice-President
      Chemical Manufacturers  Association
      2501 M Street,  NW
      Washington,  DC  20037

      Mr.  Gordon D. Strickland
      Vice-President
      Chemical Manufacturers Association
      2501 M Street, NW
      Washington, DC  20037
                                2-7

-------
     TABLE 2-1.
LIST OF COMMENTERS ON THE PROPOSED
  EARLY REDUCTIONS RULE (continued)
PocketItem Numbera

IV-D-43
IV-D-44
IV-D-45
IV-D-46
IV-D-47
IV-D-48
     Commenter and Affiliation

     Mr. Gordon D. Strickland
     Vice-President
     Chemical Manufacturers Association
     2501 M Street, NW
     Washington, DC  20037

     Mr. Walter Roy Quanstrom
     Vice-President
     Amoco Corporation
     200 East Randolph Drive
     Chicago, Illinois  60680-0703

     Mr. Gordon D. Strickland
     Vice-President
     Chemical Manufacturers Association
     2501 M Street, NW
     Washington, DC  20037

     Ms. Barbara Harris Morin
     Principal Engineer, Division of Air and
     Hazardous Materials
     State of Rhode Island and
     Providence Plantations
     291 Promenade Street
     Providence, Rhode Island  02908-5767

     Mr. Jon Heinrich, Chief
     Planning Section, Bureau of Air Management
     State of Wisconsin, Department of
     Natural Resources
     101 South Webster Street
     Madison, Wisconsin  53707-7921

     Mr. Dale A. Duhon
     'Corporate Environmental Engineer
     Great Lakes Corporation
     PO Box  1958
     El Dorado, Arkansas  71731
                                2-8

-------
     TABLE 2-1.
LIST OF COMMENTERS :.ON THE PROPOSED
  EARLY REDUCTIONS RULE  (continued)
Docket Item Numbera

IV-D-49




IV-D-50
IV-D-51
IV-D-52
 IV-D-53
 IV-D-54
 IV-D-55
     Commenter and Affiliation

     Mr. Thomas A. Robinson
     Vulcan Chemicals
     PO Box 530390
     Birmingham, Alabama  35253-0390

     Mr. Alan T. Roy, Manager
     Pollution Control
     Allied Signal Inc.
     PO Box 831
     Hopewell, Virginia  23860

     Mr. Bryce Harthoorn
     Deere & Company
     John Deere Road
     Moline, Illinois  61265-8098

     Mr. J. David Thornton,  Section Manager
     Program Development & Air Analysis  Section
     Minnesota Pollution Control Agency
     520 Lafayette Road
     Saint Paul, Minnesota   55155-3898

     Mr. Vincent Lajiness, Director
     Environmental Legislative  &
     Regulatory Affairs
     The Coastal Corporation
     One Woodward Avenue
     Detroit, Michigan 48226

     Ms. Louise T. Noell
     Manager, Product Stewardship
     BASF  Corporation
      100 Cherry Hill Road
     Parsippany,  New Jersey  07054

      Mr.  Gordon D.  Strickland
      Vice-President
      Chemical Manufacturers Association
      2501  M Street,  NW
      Washington,  DC  20037
                                2-9

-------
     TABLE 2-1.
LIST OF COMMENTERS.: ON THE PROPOSED
  EARLY REDUCTIONS RULE (continued)
Docket; Item Numbera

IV-D-56
IV-D-57
IV-D-58
IV-D-59
IV-D-60
 IV-D-61
 IV-D-62
     Commenter and Affiliation

     Mr. Milton Feldstein
     Air Pollution Control Officer
     Bay Area Air Quality Management District
     939 Ellis Street
     San Francisco, California  94109

     Mr. Nick Nikkila
     Director of Engineering
     South Coast Air Quality Mgmt. District
     9150 Flair Drive
     El Monte, California  91731

     Mr. Dennis L. Arfman
     Mr. Jeffrey W. Schwarz
     Bradley, Campbell, Carney & Madsen
     1717 Washington Avenue
     Golden, Colorado  80401-1994

     Mr. Leonard D. Verrelli, Chief
     Air Quality Management
     State of Alaska Dept. of Environmental
     Cons ervation
     PO Box O
     Juneau, Alaska  99811-1800

     Mr. Richard E. Grusnick, Chief
     Air Division
     Alabama Dept. of Environmental
        Management
     1751 Dickinson Drive
     Montgomery, Alabama   36130

     Mr. Gordon D. Strickland
     Mr. Ron Van Mynen
     Chemical Manufacturers Association
     2501 M Street, NW
     Washington, DC  20037

     Ms. Deborah A. Sheiman
     Resource Specialist
     Natural Resources Defense Council
      1350 New York Avenue, NW
     Washington, DC  20005
                               2-10

-------
     TABLE 2-1.
LIST OF COMMENTERSr. ON THE PROPOSED
  EARLY REDUCTIONS RULE (continued)
Docket Item Numbera

IV-D-63
IV-D-64
IV-D-65
 IV-D-66
 IV-D-67
 IV-F-1
 IV-F-la
     Commenter and Affiliation

     Mr. John M. Clouse, Program Manager
     Stationary Sources Program, Air
        Pollution Control Division
     Colorado Department of Health
     4210 East llth Avenue
     Denver, Colorado  80220-3716

     Ms. Nancy Wittenberg, Director
     Department of Environmental Protection
     State of New Jersey
     CN 027
     Trenton, New Jersey  08625-0027

     Mr. H. Richard Seibert, Jr.
     National Association of Manufacturers
     1331 Pennsylvania Avenue,  NW
     Washington, DC  20004-1703

     Mr. David F. Zoll
     Vice-President
     Chemical Manufacturers Association
     2501 M Street, NW
     Washington, DC  20037

     Mr. Peter  P. Baljet, Chairman
     National Air Conservation Commission
     American Lung Association
      1726 M Street, NW
     Washington,  DC   20036-4502

      Public Hearing  in  the  matter of:
      Proposed Rule  for  the  Early Reductions
      Provision of Title III of the Clean Air
      Act.   Transcript of Hearing.   Research
      Triangle Park,  North Carolina.  July 11,
      1991,  with attachments (Items 1 and 2)

      Statement of J.  Hovious,
      on behalf of the Chemical Manufacturers
      Association, before the U. S. Environmental
      Protection Agency,
      Research Triangle Park, North Carolina.
      July 11, 1991.
                                2-11

-------
     TABLE 2-1.  LIST OF COMMENTERS; ON THE PROPOSED
                   •EARLY REDUCTIONS RULE (concluded)
Docket Item Number3-

IV-F-lb
Commenter and Affiliation

Testimony of M.W. Pucci, AT&T, before the
U. S. Environmental Protection Agency,
Research Triangle Park, North Carolina.
July 11, 1991.
 a The docket number for this project is A-90-47.  Dockets are
 on file at the EPA's Air Docket in Washington, D.C.

                                2-12

-------
number of commenters support the broad definition of source
(IV-D-02, IV-D-05,  IV-D-08, IV-D-12, IV-D-17, IV-D-19, IV-D-24,
IV-D-27, IV-D-28, IV-D-29, IV-D-30, IV-D-32, IV-D-33, IV-D-34,
IV-D-35, IV-D-36, IV-D-38, IV-D-49, IV-D-57, IV-D-58).
     Response;  The definition of source for purposes of
Section  112(i)(5) is consistent with the definition of source
under Section  112 of the CAA.  Nothing precludes different
definitions  or interpretations of the term  "source" in other
parts of the CAA.   The  Title V permitting authority will be
required to  establish permit limits for a facility during the
permit process required by the Title V permit program.  It may
well be  that one permit for a contiguous facility may contain
'permit limits  for multiple Section  112(d) sources.  Whether
participating  in the Early Reductions Program or not, many
industrial  facilities will be subject to more than one
Section  112(d) standard and thus may have more  than one
Section  112(d) source as  part of its permit.
     The commenters' conclusion that the EPA should change the
definition  of  "source"  to match that of a Section  112(d)  source
proceeds from an  incorrect  assumption.  The commenters assume
that at  the time  a facility owner  or operator undertakes  to  make
early  reductions,  or enters into an enforceable commitment to do
 so, a  determination of  the  "source" can be  made for purposes of
 "the otherwise applicable Section  112(d)  standard."
      However,  in many instances, this will  not  be  possible
 because of the structure of Section 112(i)(5).   The Early
 Reductions provision requires a facility  owner  or  operator to
 make the reductions prior to proposal  of  a Section 112(d)
 standard, or enter into an enforceable commitment  to  do so  before
 proposal provided the reductions are demonstrated before January
 1, 1994.  If the otherwise applicable Section 112(d)  standard has
 not yet been proposed and "source" has not been defined for that
 standard, it would be impossible for an owner or operator to know
 whether the necessary reductions had been made or could be made
 to achieve the 90  (95) percent reduction for gaseous and
                                2-13

-------
particulate HAP's, respectively.
     As recognized by Senator Durenberger and others, the EPA
will have broad discretion in defining "source" for any
particular Section 112(d) standard.  In some cases, a source will
be a process unit, in others it will be an entire plant, (see 126
Cong. Rec. S16927 [3rd col. daily ed. Oct. 27, 1990]1). .Congress
could not have meant to so restrict the Early Reductions Program
to only those companies _who owned or operated emission points
that the EPA has clearly defined as a source before proposal of a
particular Section 112(d) standard.  Moreover, because it would
be impossible at this time to identify all possible sources that
may be subject to Section 112(d) standards, it is impossible for
the EPA to list each and every type of source for the Early
Reductions Program.
     The EPA's definition of "source" recognizes the flexibility
set forth in the statute.  The EPA, or the State authority if
operating under a delegated program, must review the applicant's
Early Reductions submittal to ensure that it is consistent with
one of the proposed regulatory definitions under the Early
Reductions Program.  Given the limitation described above, the
EPA developed a definition of source for this provision that is
designed to encompass the broad definition of source contained in
Section 112(a), which incorporates the broad definition of
Section 111(a).
2.1.2  Comment:  The multipart definition of source is too broad
and will allow for arbitrary groupings of sources.  The commenter
suggests that the definitions be narrowed to. eliminate arbitrary
groupings and require functional groupings.  In addition, it is
suggested that the EPA  establish source definitions for
Section 112(d) standards as early as possible, and that these
definitions be used for the Early Reductions Program  (IV-D-46).
Another commenter suggested that the definition of source in
Section 63.73(a)(1) be  expanded to cover all the sources that the
EPA is now  actively developing standards for and that are
expected to be included in the first phase of standard proposals
                               2-14

-------
expected by the end of 1991, including the HON package (IV-D-50,
IV-D-64).
     Response:  The EPA has not yet determined which groups of
emission points will constitute a source for any particular
Section 112(d) standard.  The definition of source is set forth
in a manner consistent with the broad statutory definition.
Section 112 will establish standards for hundreds of different
types of industrial processes and other sources that emit HAP's.
It is impossible to determine at this time how source might be
defined for any particular source category to be regulated by a
Section 112(d) standard.  The list of potential source categories
to be subject to regulation does not necessarily aid in the
definition of source.  In some instances, an industrial source
category  (e.g., a particular type of manufacturing process) may
have one source for purposes of a Section 112(d) emission limit.
In other instances, multiple sources may be included within any
source category, e.g*, tanks in a process could be one source and
wastewater emissions  from the same process may constitute another
source within the same source category.  The EPA has not yet
developed any specific definitions of source for all of the
categories listed in  57  FR  31576  (source category listing).2
2.1.3   Comment;  One  commenter contended that the EPA's
Section 63.73(a).(l) definition in of source proceeds from an
error in equating  "source"  and "source  category," and interprets
sources listed  in Appendix  B as  "source categories"  (IV-D-39).
     Responses  The commenter fundamentally misunderstands the
EPA's definition of source  and how it relates to the  "source
category."   As  the commenter  suggests,  the EPA  is to identify
source  categories  that emit HAP's.  The EPA is  not  saying,
however,  that it must know each  category before it  can make the
linkage to Section 112(i)(5).  A "source category"  for the
purpose of defining  industry  types to be regulated  and the
 "source" to which  a  standard  specifically  applies may be  quite
different (e.g.,  in  the new source performance  standard  for
 electric utilities the applicable "source"  is  the  steam  boiler).
                               2-15

-------
     The single "source" listed in Appendix B also exemplifies
this distinction; this source, equipment leaks, is contained
within a source category—the synthetic organic chemical
manufacturing industry (SOCMI).  Furthermore, equipment leaks are
a special situation that the EPA believes should be addressed
separately, as equipment leak emissions arise from the pumps,
valves, connectors, and other equipment that is integrally
associated with a more traditional "source," i.e., building,
structure, facility, or installation.
     The outcome of a regulatory negotiation involving equipment
leaks published by the EPA designated equipment leaks as a
separate source (56 FR 9318; March 6, 1991).3  The EPA proposed
the equipment leak rule as part of the hazardous organic national
emission standard for hazardous air pollutants (NESHAP)(HON),
which will be promulgated under Section 112(d).  The publication
of the outcome of the regulatory negotiation identifies equipment
leaks.as a sufficiently distinct source to  separate them for the
purposes of early reductions.  The EPA will determine the
definition of source for the HON in that rule.
     The EPA is allowing the option of including equipment leaks
as part of a designated Early Reductions source, leaving
equipment leaks out of the source or identifying equipment leaks
solely .as the source.
     Indeed, once the EPA defines a source  category, the
issue of the source definition may still remain.  In some
instances, individual facilities within a source category may
constitute an entire source.  In other instances, each source
subject to a Section 112(d) standard may only constitute a
portion of the facility within a source category.  Thus, it is
the regulated source that is important for  the purposes of
determining otherwise applicable Section 112(d) standards, not
the source category.  Because the EPA has not defined source for
particular Section  112(d) standards  (except in one instance), the
EPA's definition reflects the range of possibilities contemplated
by the definition of source in the statute.
                               2-16

-------
2.1.4  Comment:  One commenter objected to proposed paragraph
63.73(a)(2) and another argued that a plantwide definition of
source was expressly prohibited, citing a floor statement by
Senator Durenberger (IV-D-39, IV-D-67).
     Response:  Section lll(a)(l) (incorporated into the
definition of "existing source" in Section 112) defines a
"stationary source" as "any building, structure, facility, or
installation which emits or may emit any hazardous air
pollutant."  This terra is obviously broad enough to encompass an
entire plant or facility.  In some instances, a Section 112(d)
standard may encompass an entire facility, which certainly
includes the concept of a plantwide source.
     Moreover, the definition of major source, which covers an
entire contiguous facility under common ownership or control, may
consist of a single stationary  source.  "Major source" is defined
as  ".  . . any  stationary source or group of stationary sources
  .  .  ." (see Section 112  (a)(l) of the CAA).  Thus, contrary to
the commenter's assertions, the definition  of source under
Section 112 expressly encompasses a plantwide definition of
source, among  other configurations.   The proposed and final
regulations reflect that  statutory language.
      Finally,  a closer reading  of the complete referenced
Durenberger statement expressly confirms that the source to which
a Section 112(d)  standard might apply could be a specific portion
of a facility or  encompass  an entire  contiguous facility  (see  126
Cong.  Rec. S  16927  [3rd.  col.]1).  If the  EPA  has the discretion
to establish  a Section  112(d)  standard for a source that
encompasses an entire contiguous facility,  then it  follows that
an entire contiguous facility may be a source  under
Section 112(i)(5).
 2.1.5  Comment:   One commenter objected to the proposed
 Section 63.73(a)(3)  because it allows an applicant  to define a
 grouping of points that constitutes  a building,  structure,
 facility, or installation as a source.  The commenter cites
 proposal preamble language that explains that emission points
                                2-17

-------
having a functional or geographical relationship could be defined
as a source under this part of the definition and concludes that
there is no statutory basis for this type of grouping (IV-D-39).
     Response;  Subparagraph 63.73(a)(3) directly follows the
definition of source in the statute.  As noted, a "stationary
source" is any "building, structure, facility or installation
which emits, or may emit, any hazardous air pollutant" (see
Section lll(a)(3) of the CAA) .  The phrase "functional or
geographical relationship" merely gives meaning to the statutory
terms.  For example, a "building" suggests a geographical
grouping of emission points (it may also have a functional
relationship).  Likewise, an "installation" suggests some type of
unit that undertakes a particular function, such as a wastewater
treatment system.  As the EPA develops Section 112(d) standards,
it will define "source" for particular standards considering
these types of logical groupings of emission points.  However,
because the EPA has not yet defined what will constitute a
"source" for a particular Section 112(d) standard prior to its
proposal (with the single exception noted earlier), the language
of this rule reflects the range of options available within the
statutory definition.
2.1.6  Comment:  Several commenters contended that subparagraphs
(a)(4) and (a)(5) of the proposed definition are contrary to the
statute because they allow "bubbling" across two or more sources.
One commenter noted that previous guidance from the EPA precluded
the use of "bubbling" to avoid Section 112(d) standards (IV-D-06,
IV-D-39,. IV-D-67) .
     Response;  As discussed in the proposed rule, the EPA
believes that the provisions of Section 63.73(a)(4) and (5) are
consistent with the statute and the underlying purposes of the
Early Reductions provisions.  However, the EPA is deleting
subparagraph (a)(4) because (a)(4) is redundant with the
                               2-18

-------
provisions of (a) (5),. as any combination of emission points that.
would meet the language of proposed paragraph (a)(4) would also
fit with the language of (a)(5).
     The EPA has concluded that unrelated emission points may be
considered as a single source for purposes of the Early
Reductions Program, provided that the emission points are all
under common ownership or control and are located "within a
contiguous area" such as a common plant site and constitute a
significant level of emissions.  The individual emission points
to be aggregated do not have to be located next to one another or
be functionally related in order to be grouped as a source,
provided that they are all located within the same contiguous
facility.
     The conclusion that aggregating unrelated emission points is
permissible for purposes of the Early Reductions Program is
confirmed not only by the statutory language and case law
discussed above, but also by the statutory policies and
legislative history of the Early Reductions Program.  In
Section 112(i)(5)(E), Congress  authorized a form of emissions
trading by allowing offsetting  reductions of one HAP against
reductions of other HAP's for purposes of the Early Reductions
Program.  Aggregation of emission points is merely  another  form
of  emissions trading.a
     Nevertheless, the rule  is  not inconsistent with the general
emissions  trading policy referenced by the commenter  (51 FR
43814;  December 4,  1986),4  i.e., that  "bubbling" may not be used
to  avoid a Section 112(d)  standard.  The Early Reductions
      aThis  conclusion  is  limited  to  the  context  of the  Early
 Reductions Program for several reasons.  In  particular,  Congress
 specifically contemplated netting one hazardous pollutant against
 another for purposes of  the Program.  Congress indicated  in the
 legislative  history  that  it  wishes  the   EPA   to  encourage
 participation  in this Program to obtain  early reductions  (see
 discussions below);  and  the statutory requirement  of  90 percent
 reductions raises  quite  different  factual  and policy issues than
 might be applicable elsewhere.
                                2-19

-------
provision does not provide a mechanism for a source to avoid
applicability of a Section 112(d) standard.  Rather, it provides
for compliance extension in the form of an alternative emission
limit for the source for a period of 6 years in exchange for the
source having achieved a certain emissions reduction level by a
specified date (either by January 1, 1994, or before proposal of
an otherwise applicable Section 112(d) standard).  The source
must meet the Section 112(d) standards when the compliance
extension has expired.
     Moreover/ the legislative history indicates that Congress
wanted the EPA to encourage participation in the Program (see
H.R. Rep. No. 101-490, 101st. Cong., 2d Sess. 332 [May 17,
1990]^.  "In the administration of this provision the EPA thus
should strive to encourage companies to take advantage of this
incentive to reduce emissions early.").,  In addition, the
significance threshold (i.e., 10 tons per year [tpy], or 5 tpy at
a contiguous plant site less than or equal to 25 tpy) is designed
to ensure that substantial real reductions are achieved and that
the Program is not trivialized.  The flexibility afforded through
aggregating emission points offers greater incentive to
participate in the Early Reductions Program.  Greater overall
participation in the Program, combined with the minimum threshold
for early reductions achieved, will help ensure that the
reductions achieved under the Program are both real and
substantial and the concomitant environmental benefits maximized.
     The commenter argues further that the random pooling
approach that allows for credits undertaken in the past
demonstrates that subparagraph  (a)(5) is inconsistent with the
purposes of the statute.  The commenter argues that the
justification for random pooling—to encourage greater
participation in the Program and thereby reduce air toxics
emissions—is not present if a  source can include past reductions
in the pool.  The commenter concludes that any pollution
reduction measure undertaken prior to the enactment of the law
cannot have been dependent on the incentive of pooling., The
                               2-20

-------
commenter acknowledges that past reductions can be credited, but
that they should not be pooled with unrelated, uncontrolled
emission points to demonstrate an early reduction.
     The statute affords a 6-year extension for those sources
that achieve a 90 (95) percent reduction in emissions by a
specified time.  Those emission reductions may have occurred at
any time back to 1987, and in limited instances 1985 or 1986.
The commenter assumes that the only time applicants will use the
grouping of emissions points under proposed paragraphs (a)(4) and
(a)(5) is to get credit for past actions, while making no
additional reductions, and obtaining a 6-year extension for
remaining uncontrolled emission points within the Early
Reductions source.  This will not be the case.  The EPA has
discussed tentative Early Reductions plans with many companies
and has received several enforceable commitments that indicate
that the flexible definition of source is encouraging prospective
emission reduction projects.
    In many instances, the EPA expects that an applicant will
group some recently well-controlled emission points with others,
some or all of which  may have controls installed on them.   The
ability to apply credits in excess of 90  (95) percent reduction
from one point to a point unable to achieve 90  (95) percent by
itself provides an incentive to control the second point enough
to cover the  shortfall.
     For example,  suppose the applicant previously controlled a
 100-ton emission  (point A) to a 98-percent control level, i.e.,
 it now emits  2 tons of HAP's.  The applicant  also has a second
 100-ton uncontrolled  emission  (point B).  Existing control
measures  for  point B  indicate that maximum achievable control
 technology (MACT)  will likely require at  least  90-percent
 reduction.   However,  a much  less  expensive technology can reduce
 emissions  from point  B by 85  percent, i.e., to  15 tons.
      Furthermore,  the owner  has  determined that a net savings
 would be realized if  85  percent  control could be implemented now
 and MACT implemented 10  years later,  compared with meeting  MACT
                               2-21

-------
in 4 years.  The extra time gained from a compliance extension
would also allow development of process modifications that would
completely eliminate HAP emissions from the source.  Without
proposed paragraph (a)(5), the applicant will have no incentive
to install any controls on point B prior to a Section 112(d)
standard because that point by itself will not be reduced by 90
(95) percent, short of installing MACT control anyway.  However,
if an additional 5 tons can be credited from emission point A,
the source will have an incentive to participate in the Program
by including points A and B, and thereby substantially reduce
overall air toxics emissions from that source.
     Without the option of proposed paragraph (a)(5), the source
would not seek a compliance extension for point B and would have
no incentive to enter the Program, and the emissions from the two
points will be 102 tons.  By entering both points in the Program
using proposed paragraph  (a)(5), the source's emissions will be
17 tons, which is greater than a 90 (95) percent reduction from  ,
the original 200 tons being emitted by the two points.  Moreover,
there is no basis in the law for penalizing those that made
reductions between 1987 and the passage of the CAA Amendments of
1990 in defining a source.
2.1.7  Comment;  A number of commenters believe that
Section 63.74(b) of the regulation, which defines significant
emissions, unnecessarily restricts participation in the Program
by small  (but major) stationary sources.  The cost and burden of
the reduction demonstration will prevent trivial submissions
(IV-D-05, IV-D-12, IV-D-23, IV-D-28).  Other commenters f.urther  •
suggest that the limits be set at 3 tpy for 25-tqn-or-greater
facilities and 1 tpy for  less-than-25-ton facilities  (IV-D-34,
IV-D-44).  One commenter  suggested adding "20 percent of the HAP
emitted in the base year  from the entire contiguous facility" to
the significance definition (IV-D-38).  Another commenter agrees
that the Early Reductions Program should focus on major emitters
(i.e., 10 tpy or 5 tpy at 25-ton facilities)  (IV-D-08).  Another
commenter feels that the  definition of significant should include
                               2-22

-------
only 10 tpy sources.  Allowing sources that emit 5 tpy at 25 tpy
sources allows resources to be wasted on small sources (IV-D-26).
Finally, one commenter supports the concept of significant
emission reduction as it applies to the definition of source, but
pointed out that the multiple definition of source may make
review time.longer than anticipated by the EPA because of it's
complexity  (IV-D-47).
     Response:  The EPA's definition of source is designed to
provide broad flexibility in defining source for purposes of the
Early Reductions provisions.  The goal is to encourage widespread
participation in this Program to achieve early reductions of air
toxics emissions.  The significance threshold included in
Section 63.73(b) is designed to ensure that significant
reductions  occur.  It is important to point out, however, that
the 5 to 10 ton threshold of significant emissions applies to
only subparagraph  (a) (4) (formerly (a) (5). in the proposed rule;
proposed paragraph  (a)(4) has been deleted in the final rule).
Thus, any process unit or other entity that meets any definition
in  (a)(l) through  (a)(3) is eligible for an Early Reductions
extension,  even if it emits less than the 5 or  10 ton threshold.
Thus, the EPA anticipates that, in most instances, even smaller
facilities  will be able to participate in the Program.
     Although EPA could have considered other thresholds, such as
3 tons/25 tpy or 20  percent of any plant's emissions, as
suggested by  some commenters, the levels chosen are  a reasonable
attempt to  balance  the various factors involved,  including
allowing participation and ensuring  significant reductions.
2.1.8   Comment;  Within the source definition developed by
Section 63.73(a)(3),  it is not clear if an applicant has  the
choice  of designating two or more common facilities  (those
servicing more  than one production unit) as two or more  sources
or  whether  it is  necessary  to define all common facilities  as a
 single source.   The commenter  requests clarification of  this
 issue (IV-D-50).
                                2-23

-------
     Response;  Under Section 63.73(a)(3), it is possible to
define a unit, such as a wastewater treatment system that serves
multiple parts of an entire plant, as a single source.
2.1.9  Comment;  The definition of "process or production unit"
should be expanded to state that a process or production unit can
contain many sources or can be a part of a larger single source
(IV-D-02).
     Response;  The current regulation does not use the terms
"process or production unit."  The current source definition
allows source to be defined as a group of emission points that
can be identified as a facility, building, structure, or
installation for the purposes of establishing standards under
Section 112(d) of the CAA.  Under this definition, a process unit
would certainly qualify.  The definition of source also allows
for any combination of emission points, provided that the
emission reductions are significant, i.e., greater than 10 tpy at
a facility that emits greater than 25 tpy or 5 tpy at a facility
that emits 25 tpy or less.
2.1.10  Comment;  The source definition should specify that an
"entire facility under common ownership or control" may include
facilities with different Standard Industrial Classification
(SIC) codes at the same location  (IV-D-02).
     Response;  The CAA, and now the Early Reductions regulation,
describe a contiguous "area" under common ownership or control.
The definition of source allows a facility to include different
SIC codes, as long as the portions of the facility are under
common ownership or control.  In addition, proposed paragraph
(a)(5)  (now paragraph (a)(4)) of Section  63.73 of the rule allows
combinations  of emission points, provided that the emission
reduction from such combinations is significant.  Therefore,
emission points with different  SIC codes  may be combined as a
source, as long as the  source,  as defined by the owner or
operator, meets one of  the  source definitions contained in the
regulation.
                               2-24

-------
2.1.11  Comment;  Research operations should not be included in
the source definition for early reductions because the pollutants
and emissions from these operations are highly unpredictable
(IV-D-08).
     Response;  The Early Reductions Program allows the owner or
operator the option of including or excluding emissions from
research facilities in the definition of source.  The
predictability of emissions from research operations may vary
from source to source.  There may be a research operations source
that has predictable emissions.  If a source includes research
operations and the base year or post-reduction emissions prove
not to be sufficiently reliable or well-documented,
Section 63.78(b)(3) of the regulation allows the permitting
authority to deny the application.  If the emissions are reliable
and meet the requirements of Section 63.74, research facilities
defined as part of a source may be granted an extension.
2.1.12  Comment:  Once a group of emission points has been
accepted as a source by the EPA, the source definition should
remain constant throughout the Program.  When a Section 112(d)
standard is proposed, the definition of source in the standard
should not affect the already defined source  (IV-D-27).
       Response:  Once an individual source is defined and
accepted by the EPA as part of the Early Reductions Program, the
Early Reductions source will not change as a  result of a
Section  112(d)  standard.  Any particular Section  112(d) standard
may apply to  the entire Early Reductions source or to a subset of
emission points within that source.  All emission points included
in the defined  Early Reductions source are entitled to a six-year
extension from  all otherwise applicable Section 112(d) standards.
At the end  of the  6-year period, emission points within the Early
Reductions  source  affected by the  earliest applicable
Section  112(d)  standard must meet  those standards.  Emission
points not  affected by the earliest  Section  112(d) standard, must
continue to meet the  alternative emission limitation until  six
years after the compliance date of any Section 1.12 (d)  standard
                               2-25

-------
applicable to the emission point.  Any emission points not
included in the Early Reductions source definition, but included
in the source definition for a Section 112(d) standard must
achieve reductions according to the compliance schedule in the
applicable standard.
2.1.13  Comments. An owner or operator should have the
flexibility to redefine the source during the 6-year extension
period if a decision is made to restart a shutdown portion of the
source or if significant process changes are made.  The new
source could exclude the restarted or changed portion but must
continue to achieve 90 (95) percent reduction from base year
emissions (IV-D-29).
     Response;  Once an individual source is defined and accepted
by the EPA as part of the Early Reductions Program and the source
has been granted a 6-year extension by permit, the source cannot
be redefined.  If .a portion of the original source restarts or
expands, those emissions must.be accounted for in the post-
reduction emissions.  If part of a source shuts down or curtails
production, and production of the same product is increased at
another unit or is replaced by a new unit at the plant site that
was not included as part of the original source, the HAP
emissions from the increased production outside of the Early
Reductions source must be accounted for as well.  This is to
ensure that emissions from the source are not simply moved
elsewhere in the plant.  Owners and operators must give careful
consideration to the designation of "source" when applying for
the Early Reductions Program in light of potential future
expansions.  .
     If an owner or operator anticipates increased production in
certain areas of a facility, several options are available.  The
flexibility of the source definition allows the applicant to
define a source that does not include the portion of the facility
that may expand.  The applicant may also reduce emissions from
the base year by greater than 90 (95) percent, thereby allowing
for subsequent increases in emissions that would still meet the
                               2-26

-------
required reduction.  If the owner or operator can anticipate what
changes will be made to the source, the owner or operator may
coordinate with the State to determine the most appropriate type
of alternative emission limitation (i.e., a numerical emissions
limitation for each emission point or other requirement) that may
allow more .flexibility for expanding one portion of the source.
The EPA recognizes that process changes are necessary, but to
participate in this Program, overall source emissions must remain
at a level that is 90  (95) percent of the base year emissions.
The permitting authority may revise or adjust alternative
emission limits through permit modifications, as appropriate,
provided that the overall 90 (95) percent reduction for the
source is maintained during the extension.
2.1.14  Comment;  The  source definition is confusing.  The
paragraph states that  "emission reductions from a source are
considered significant if..." but proceeds to define the base
year emissions of the  facility and not the reductions.  Change
the language to establish a size threshold (IV-D-59).
     Response:  The purpose of the Early Reductions Program is
the reduction of emissions.  The purpose of the significance test
is to  focus resources  on  significant emission reductions.  Such
reductions are quantitatively linked to the base year emissions
because the emission reduction is  90  (95) percent of the base
year.  Therefore,  in order to determine if "significant"
•reductions can result  from a source, base year emissions must  be
established.
2.1.15 Comment:   The  regulation needs to clarify what  happens
when  the  Section  112(d)  source definition is different  from
the Early Reductions Program source definition  (IV-D-59).
      Response:  Emission points  that  are part of the  source
defined  for  the Early  Reductions Program are granted  a  6-year
extension from compliance with an  applicable Section  112(d)
 standard. Any emission point that is not part  of  an  Early
Reductions source must comply with appropriate  Section  112(d)
 standards.   At the end of the  6-year  extension,  emission points
                               2-27

-------
that are part of an Early Reductions source must then meet
Section 112(d) or Section 112(f) standards if the standards apply
to those emission points.
     Emission points within the Early Reductions source that are
not covered by a Section 112(d) standard will be required to
continue meeting the 90 (95) percent reduction specified by the
original alternative emission limitation.  This emission
limitation must be met until 6 years after the compliance date of
the last applicable Section 112(d) standard that affects the
source.
     This scenario will create a staggered reduction of emissions
from the Early Reductions source as various Section 112(d)
standards are promulgated and the 6-year extensions to each
standard expire.  Eventually, all emission points at a facility
will be in compliance with any applicable Section 112(d) or
Section 112(f) standard.  It should be noted that after the
6-year extension, at which time some emission points of the
original Early Reductions source will apply the Section 112(d),
standard, the source may not average increased reductions from
these emission points to achieve 90 (95) percent reduction from
the Early Reductions source.  The emission points not covered by
the Section 112(d) standard must comply with the alternative
emission limitations set by the permit as if the other emission
points now covered by Section 112(d) were continuing to emit at
the level prior to installation of the control required to meet
the Section 112(d) standard.
2.1.16  Comment;  In the source definition,  (a)(2) and a(5) do
not include references to Section 112(d).  Is it assumed that
only sources  listed in accordance with the provisions of the CAA
are to be included in the early reductions calculations
(IV-D-52)?
     Response;  Any source  submitted that meets the definition of
"source" as described in Section  63.73 of the Early Reductions
regulations will be accepted as a source under the Early
Reductions Program.  The paragraphs in the regulation do not
                               2-28

-------
refer to the CAA because it is not necessary for an Early
Reductions applicant to be on the list of source categories to be
covered by Section 112(d) standards.
2.1.17  Comment;  The final rule should include guidelines that
would allow emission increases to occur during the extension
period without jeopardizing the extension.  The commenter feels
this flexibility is needed to encourage participation in the
Program without impeding possible growth and expansion of
facilities  (IV-D-50).  Another commenter requested that the rule
specify how equipment moves or process changes within a source
with an approved permit under the Early Reductions Program will
be handled  (IV-D-27, IV-D-28).  Another commenter suggested that
the EPA allow emission increases with production increases in one
portion of the source if HAP emissions from other portions of the
source are  further reduced to offset the increases  (IV-D-45).
     Response:  If an owner or operator anticipates increased
production  in certain areas of a facility, several options are
available.  The flexibility of the  source definition allows the
applicant to define  a source that does not include the portion of
the  facility that may expand.  The  applicant may also reduce
emissions from the base  year by greater than 90  (95) percent,
thereby  allowing  for subsequent increases in emissions that would
still meet  the required  reduction.   If the owner or operator can
anticipate  what changes  will  be made to the source, the  owner or
operator may coordinate  with  the'state to determine the  most
appropriate type  of  alterative emission  limitation (i.e.,  a
numerical emissions  limitation for each  emission point or other
requirement)  that may allow more  flexibility  for expanding one
portion of  the source.  The EPA recognizes that process changes
 are necessary, but to participate in this Program,  overall source
 emissions must remain at a level that is 90  (95)  percent of  the
 base year emissions.  The permitting authority may revise or
 adjust alternative emission limits through permit modification,
 as appropriate, provided that the overall 90  (95)  percent
 reduction for the source.is maintained during the extension.
                                2-29

-------
2.1.18  Comment;  The proposed rule should clarify what happens
to an existing Early Reductions permit in the event that the
source adds a new line identical to those included in the source
that has been given a 6-year extension.  Must the new line comply
with Section 112(d) or can it be added to the source (IV-D-33)?
     Response;  In most cases, new emission points will not be
able to participate in the Early Reductions Program.  The CAA
specifies that a compliance extension may be granted only to an
existing source.  New emission points would thus be outside the
scope of the Early Reductions source and, therefore, would be
subject to Section 112(d) standards, if previously proposed.
     The only exception occurs when the entire plant site or an
entire enclosed building is defined as the source by the
applicant and a "new" process unit is constructed within the
plant site or building, respectively, and emits less than 10 tpy
of" HAP's.  The new unit must then be included as part of the
Early Reductions source and the source must maintain the original
90 (95) percent emission reduction from base year emissions.
Emissions from the existing points in the original source must,
therefore, be reduced to compensate for the additional emissions
from the new points.  If, however, emissions from the new line
exceed 10 tpy of a single HAP or 25 tpy of total HAP's, the new
unit is considered, for the purposes of Early Reductions only, a
source by itself and thus would not be part of the original Early
Reductions source, regardless of the source definition.
     The EPA will distinguish between existing sources and new
sources in other rules.  However, the Early Reductions Program
will occur before proposal of Section 112(d) standards and before
other rules that will distinguish between existing and new
sources are finalized.  Furthermore, the Early Reductions Program
applies only to existing sources.  Therefore, the EPA had to make
an administrative determination for the purposes of Early
Reductions that new units emitting greater than 10 tpy of total
HAP's are new sources.  The EPA may distinguish between new
sources and existing sources differently in other rulemakings.
                               2-30

-------
2.2  BASE YEAR EMISSIONS
2.2.1  Comment;  The commenter suggests that 1990 be used as the
base year in order to be consistent with Title I.  There is
concern that the use of earlier base years may result in many
companies qualifying because of reductions in emissions that have
resulted from economic downturns or other external factors
(IV-D-46).
     Response;  The CAA specifically states that the base year
not be earlier than 1987.  However, the owner or operator may
choose 1985 or 1986 as the base year if the source can
demonstrate that EPA received the supporting HAP emissions data
pursuant to an information request under Section 114 the EPA
prior to November 15, 1990.  The EPA is allowing the owner or
operator to choose any base year after 1986 to provide maximum
flexibility.  However, emission data developed during years other
than the base year, including years prior to 1985, can be used
for determining base year emissions if the data are
representative of operating conditions in the base year.  Owners
or operators will have to demonstrate the applicability of these
data to base year conditions.
2.2.2  Comment:  Two commenters were concerned with potential
delays in  the  review process and suggest that base year emissions
be automatically accepted if the EPA does not complete the review
in a  specified time frame  (IV-D-02, IV-D-45).
      Response;  The EPA's goal  is  to review all  enforceable
commitments  and submittals within  the  times identified in the
rule.  However, the EPA intends to adequately review  all
submittals,  and will take whatever time  is necessary  to ensure
that the applications  are complete and verifiable.  The EPA will
not  automatically accept base  year emissions  that have not  been
reviewed because  of  time constraints.  The  purpose of reviewing
base year emissions  early  is to provide  some  assurance to the
 submitter that the base year emissions for  early reductions will
be approved at permit review.
                                2-31

-------
2.2.3  Comment;  Some facilities reported data to the EPA in 1984
and would like these data to be acceptable as base year emissions
(IV-D-05).
     Response;  The CAA specifically states that the base year
not be earlier than 1987.  However, as noted above, an owner or
operator may choose 1985 or 1986 as the base year only if the
source owner or operator can demonstrate that EPA received the
supporting data pursuant to an information request under
Section 114 prior to November 15, 1990.  However, emission data
developed during years other than the base year, including years
prior to 1985, can be used for determining base year emissions if
the data are representative of operating conditions in the base
year.  Owners or operators will have to demonstrate the
applicability of these data to base year conditions.
2.2.4  Comment;  Once base year emission data have been reviewed
and accepted by the EPA, no further auditing of such data should
be allowed (IV-D-09, IV-D-19, IV-D-34, IV-D-35).  Another
commenter requested that if early base year review is retained,
any presumption of "clearance" for such base year be dropped
(IV-D-39).
     Response;  Acceptance of base year emission data does not
provide an absolute shield against revision should the data later
be found incorrect.  Reviewers and commenters are urged to
present any criticisms during the early review period regarding
base year data.  Applicants can make appropriate modifications at
that point and proceed with reasonable confidence that their
emissions are. acceptable.  Base year emissions that have been
reviewed and approved are still subject to additional review if
errors or fraud are discovered at a later date.  Discovery of
incorrect or fraudulent information in the emission data or
supporting materials after initial approval could potentially
invalidate the base year and/or require the applicant to make
revisions.
                                                       [
2.2.5  Comment;  Base year emissions that are substantially
greater than other years should be acceptable as long as the
                               2-32

-------
emissions were within permit levels because production rates vary
(IV-D-29, IV-D-35).
     Response;  The CAA specifically states that base
year emissions cannot be "artificially or substantially greater
than emissions in other years prior to implementation of
emissions reduction measures."  Permit levels are simply a
maximum acceptable emission limit and may not reflect "actual and
verifiable" emission levels_as required by the CAA. . Therefore,
comparison of base year emissions to permit levels does not
determine the relationship of base year emissions to emissions in
other years.  Even emissions well within permit levels, but
substantially greater than other years, would not be allowable.
For example, a permit may be written to allow 100 tpy, but if the
source was emitting 30 tpy in years other than the base year,
base year emissions of 60 tons would not be allowable.  It is
clearly the intent of the CAA to avoid the use of unusually high
emission levels to count towards the base year, which would lead
to overstating reductions achieved and minimize the benefits to
the environment achieved by the Early Reductions Program.
2.2.6  Comment;  The commenters believe that use of the EPA's
average emission factors for estimating base year equipment leak
emissions should be allowed  (IV-D-02, IV-D-05, IV-D-08, IV-D-11,
IV-D-19, IV-D-28,  IV-D-34, IV-F-la).
     Response;  Average equipment leak emission factors developed
by the EPA  generally may not be used in establishing base year
emissions.  Use of the EPA's average emission factors  for base
year emissions would artificially inflate  the base year estimate.
Subsequent  use of  a more source-specific method for the post-
reduction demonstration would  result in lower estimated
emissions,  but,  in fact this would  only be a  "paper" reduction.
As mentioned  in  the preamble  to the proposed  regulation,  source
owners  or  operators can establish  emission levels  for  equipment
leaks  using any  procedure  except for average  emission  factors
established in the document entitled  "Protocols  for  Generating
                               2-33

-------
Unit Specific Emission Estimates for Equipment Leaks of VOC and
VHAP," EPA-450/3-88-010, October 1988.6
     The EPA has considered allowing use of average emission
factors to establish equipment leak emissions only in the case
where no equipment leak emission reduction will be claimed; i.e.,
when equipment leak emissions are the same in the base year and
post-reduction.  The EPA is allowing this exception because it
will result in greater actual emission reductions from the
source.  The source will have to control nonequipment leak
emission points to a greater extent to compensate for the over-
estimate of the post-reduction equipment leak emissions.  In
addition, it is noted that equipment leaks must be less than 10
percent of base year source emissions, or the source could not
achieve a 90 (95) percent overall reduction.
     Appendix B has also been clarified to better describe the
equipment leak "source."  The negotiated regulation for equipment
leaks  (56 FR 9315; March 6, 1991)7 requires that certain
equipment in HAP service within a process unit to which the
equipment leak standards apply must be viewed as a whole.  This
means that valves, pumps, or connectors within a process unit
must be considered together because the regulation is written in
terms of percent leaking components across a process unit.  In
other words, valves, pumps, or connectors cannot be split up such
that, for example, some of the valves in a process unit have an
Early Reductions alternative emission limitation and the others
meet the Section 112(d) standard.  For example, it must be that
either all valves within a process unit are in the Early
Reductions Program or none are.  A "process unit" is a collection
of  equipment associated with a unit process operation, including
storage and transfer of feed materials to the operation and final
or  intermediate product from the operation, and operations
treating wastewater  from the unit process operation.
     The logic for requiring inclusion of all the valves,
pumps, or connectors from a process unit would not extend to the
other  equipment covered by the equipment leak rule, such as
                               2-34

-------
pressure relief devices or product accumulator vessels, which
will be subject to individual standards applicable to each device
or vessel.  Equipment subject to such "piece-specific" standards
could individually be assigned alternative emission limits as
part of an Early Reductions source or meet the Section 112(d)
standard, and are not constrained by the process unit coverage.
2.2.7  Comment;  The commenter would like to narrow the scope of
HAP emissions to cover only major emission points that are more
quantifiable (versus numerous small emission points that may not
have data) (IV-D-09).
     Response;  Hazardous air pollutants are produced by both
major and minor emissions points.  Although it would be easier to
cover only the major points, minor points may =be numerous and
contribute significantly to total HAP emissions or be responsible
for many high-risk pollutants that have a great impact on human
health.  Therefore, both major and minor emission points must be
accounted for in determining base year and post-reduction HAP
emissions.
2.2.8  Comment;  The commenters would like Toxics Release
Inventory (TRI) data to be acceptable for submission as base year
emissions data if the source for which the compliance extension
is being requested is consistent with the source definition for
which the TRI data apply  (IV-D-12, IV-D-19, IV-D-27, IV-D-30,
IV-F-lb).
     Response:  The TRI data alone are not sufficient to support
base year emissions.  The Early Reductions Program requires more
rigorous  support for emissions data than is required by TRI.   If
the supporting information for the TRI data meets all the
requirements of the Early Reductions Program, then that
information will be acceptable.  Supporting documentation for  the
Early Reductions Program must stand on its own merits.
2.2.9  Commentt  Concerning base year emissions, the regulation
should clarify the term "substantially greater" than other years
 (IV-D-27,  IV-D-52).
                               2-35

-------
     Response;  The determination of "substantially greater" is
necessarily subjective.  "Substantially greater" will be
determined in large part by historical emissions variations of
the source and the reasons for the variations.  If an application
is denied because base year emissions are found to be
"substantially greater" than other surrounding years, the
reviewer will provide rationale for this determination.  Each
base year review will be judged on a case-by-case basis.  The
enabling document ("Enabling Document for Regulations Governing
Compliance for Early Reductions of Hazardous Air Pollutants,"
EPA-450/3-91-013, November, 1992)8 provides guidance to reviewers
and submitters to better understand what is "substantially
greater."  To avoid later problems with this issue, it is
recommended that the applicant request a preapplication
conference to discuss the specific details surrounding base year
emissions.  The preapplication meeting should.be held with the
appropriate EPA Regional Office and the State.
2.2.10  Comment;  Early Reductions applicants should not be
required to submit evidence that base year emissions are not
artificially high or that emission reductions, due to lower
production rates or shutdown are permanent.  The seriousness of
violating Section 114 and being subject to enforcement provisions
of Section 113 make it unnecessary that an owner or operator be
required to submit evidence in the permit application.  Evidence
should be provided only at the specific request of the reviewing
agency  (IV-D-08).
     Response;  Early Reductions applicants will be required to
submit evidence demonstrating that base year emissions are not
artificially high.  Submission of this information provides
reasonable assurance to the applicant that base year emissions
will be acceptable for an Early Reductions demonstration, except
in the case of error or submittal of fraudulent information.
Prior approval of base year emissions facilitates the Title V
application review process where the actual reduction
demonstration is made.
                               2-36

-------
     The regulation has been changed with regard to providing
evidence of lower production rates and shutdowns.  Sources will
not be required to show evidence that lower production rates or
shutdowns are permanent for the extension duration.  This will be
accomplished by making the shutdown or lower production rate
conditions of the Title V permit.
2.2.11  Comment:  The proposed rule should provide guidance for
the assessment of trace pollutants and impurities.  There should
be a minimum amount required to make an assessment (IV-D-32,
IV-D-50, IV-D-60).
     Response:  Trace pollutants and impurities will be
handled on a case-by-case basis.  General policy would be
difficult to develop for all pollutants.  For example, small
quantities of dioxin may be considered significant while the same
quantity of another non-high-risk pollutant would be considered
an inconsequential trace amount.
     Based on process and product knowledge, owners or operators
should be aware of the possible presence of HAP's in their  source
emissions.  If trace quantities of HAP's are expected or known to
be emitted, owners or operators should account for their
presence.  The actual determination of the quantity of emissions
may be similar to the determination for small, insignificant
emission points.  Generally, testing will not be required.  The
source owner or operator may make conservative assumptions  in the
calculations used to determine the quantity that may be emitted.
If trace HAP's  are neither  expected nor known to be present, then
there is no need  to make an assessment.  The EPA expects
participants  in the Program to quantify all HAP's  reasonably
expected to be  present.
2.2.12  Comment:  The  EPA  should provide guidance  regarding:
 (1)  how sources should measure  fugitive emissions  covered by the
Section 112(d)  standard and those not covered by the
Section 112(d)  standard,  (2) what supporting documentation will
be required to be submitted for sources  relative to their
measurements of fugitive emissions and,  (3) whether the
                               2-37

-------
measurements for fugitive emissions sources that are covered by
the Section 112(d) standard will be more or less quantitative
than the fugitive emissions sources that are not covered by a the
Section 112(d) standard (IV-D-32).
     Response;  As mentioned in the preamble to the proposed
regulation, source owners or operators can establish emission
levels for equipment leaks using any procedure except for average
emission factors established in the document entitled "Protocols
for Generating Unit Specific Emission Estimates for Equipment
Leaks of VOC and VHAP," EPA-450/3-88-010, October 1988.6  Average
EPA equipment leak emission factors may generally not be used in
establishing base year emissions.  Use of EPA average emission
factors for base year emissions would artificially inflate the
base year estimate.  Subsequent use of a more source-specific
method for the post-reduction demonstration would result in lower
estimated emissions, but, in fact, this would only be a "paper"
reduction.  As discussed in a previous comment, EPA will allow
use of average emission factors to establish equipment leak
emissions only in the case where no equipment leak reduction will
be claimed; i.e. equipment leak emissions are the same in the
base year and the post-reduction year.
     The same extent of documentation is required for equipment
leak emissions as is required for all other emission points
outlined in Section 63.74(d) of the regulation.  Acceptable
techniques to establish equipment leak emissions are described in
the "Protocols" document mentioned earlier.  If an owner or
operator would like to use a different emission estimation
technique based on like sources or some other information, these
techniques will be discussed on a case-by-case basis at a
preapplication meeting.  Owners or operators who wish to submit
equipment leak estimates based on another technique are
encouraged to schedule a preapplication meeting.  Finally, the
Early Reductions Program has no bearing on equipment leak sources
that are affected by standards other than those in
                               2-38

-------
Section 112(d).  These equipment leak sources must meet the
measurement requirements of whatever standards apply to them.
2.3  ENFORCEABLE COMMITMENTS
2.3.1  Comment;  The only penalty for a source that fails to
achieve an enforceable commitment should be reversion to the
applicable Section 112(d) standard.  The source should be treated
as though it never submitted an Early Reductions application or
enforceable commitment (IV-D-02, IV-D-24). _    •
     Response;  The regulations give applicants submitting an
enforceable commitment ample time to reconsider their commitments
and allow them to rescind commitments by December 1, 1993,
without penalty.  However, once that deadline has passed,
applicants must show.that they have achieved the required
reductions by January 1, 1994, or be subject to penalties.  If a
source does not achieve reductions by January 1, 1994, the source
will not receive a 6-year extension for compliance with
Section 112(d) standards.  Failing to meet the enforceable
commitments without penalty would defeat the intended purpose of
Congress.
2.3.2  Comment;  The proposed rule needs to clarify the apparent
inconsistency for submittal of revised data.  In one paragraph
the rule reads  "EPA is not proposing a time limit for submittal
of revised  data."  In another it states "If the revised base year
emissions data are not received by the reviewing agency within 90
days  of notifying the applicant of disapproval  of base year
emission data,  the source will be considered to have withdrawn
the enforceable commitment and a notice to that effect will  be
sent  to the applicant"  (IV-D-30).
      Response;  Base year emissions that are  submitted as  part of
an enforceable commitment must be  revised and resubmitted, or
notification of intent  to revise and resubmit must  be
accomplished within  90  days  after  notification  of disapproval or
the application will  be considered by  the EPA to be withdrawn.
This  allows the reviewing  agency to recognize in a  timely fashion
which commitments will  be  revised  and  which will be rescinded.
                               2-39

-------
Because of the large number of applications expected to be
received by the EPA, a deadline for resubmittal or notification
of intent to continue in the Program is necessary to expedite the
review process and to identify those sources that do not wish to
pursue reduction demonstration.  A withdrawal alerts the
reviewing agency not to pursue penalties after January 1, 1994 if
the source fails to meet the required reductions.  Base year
emissions that are submitted for early review and not as part of
an enforceable commitment do not have a deadline for resubmittal.
The regulation has been clarified to distinguish the differences
between resubmittal schedules.
2.3.3  Comment;  A grace period should be allowed for sources
that cannot make commitment deadlines because of external forces
beyond their control (e.g., delay by the permitting authority to
approve a modification in the control plan)  (IV-D-19).
     Response; The CAA requires that the reductions specified in
the commitment must be achieved by January 1, 1994 in order for a
source to receive a Section 112(d) compliance extension.  The
Early Reductions regulations allow test data to demonstrate that
the reductions have been achieved to be received as late as March
31, 1994.  The reductions, however, must have been achieved by
the date specified by the CAA.  If a source cannot achieve the
reductions required by the enforceable commitment, the applicant
may rescind the commitment as  late as December 1, 1993 without
penalty.  Therefore, the applicant should have adequate time to
assess whether or not the source can achieve the reductions and
to take whatever action he deems appropriate.
2.3.4  Comment;  The EPA should delay the January 1,  1994
deadline, if possible, to allow completion of major reduction
projects  (IV-D-28).  .
     Response; The  CAA specifically states that  enforceable
commitments must be achieved  before January  1, 1994.  This date
cannot be changed without Congressional amendment of  the CAA.
2.3.5  Comment;  The commenter suggests that once an  enforceable
commitment  is  accepted, the only demonstration a source  should
                               2-40

-------
need to make to be issued a Part 70 permit is that everything
outlined in the enforceable commitment has been completed.  The
permitting agency should not be given the opportunity to second-
guess the plan presented in the enforceable commitment (IV-D-44).
     Response;  The control plan outlined in the enforceable
commitment is a general, nonbinding strategy that the facility
will implement to reduce emissions and is not enforceable.  The
facility develops the plan with the goal of reducing emissions by
90  (95) percent.  As the facility implements the plan, however,
it may discover that the emission reduction can be better met
with another control strategy.  The control plan presented in the
enforceable commitment must, therefore, remain somewhat flexible.
The general plan is required to assure the reviewing authority
that the reductions can be achieved and that the source owner or
operator has given serious consideration to achieving early
reductions.  The control plan will not likely be questioned or
denied unless the plan indicates that the required reductions
will not be met or the plan does not appear reasonable.  The
ultimate test will not be whether or not the described control
plan will be  followed, but a demonstration that the reductions
have actually occurred.
2.3.6  Comment:  The deadline  for submittal of enforceable
commitments  should be delayed  as much as possible.  The early
dates will cause a  flood of nonachievable commitments because
facilities did  not  have  time to evaluate the situation before  the
deadline  (IV-D-28).
     Response;  The  CAA specifies that  enforceable commitments
must submitted prior to proposal of  an  applicable
 Section 112(d)  standard.  This date  cannot be  changed without
 Congressional amendment of the CAA.   Nonachievable commitments
 can be rescinded up to December,  1,  1993.
 2.3.7   Comment;  In addition to base year emissions,  public
 comments should be solicited on the  proposed control plan
 (IV-D-04).
                                2-41

-------
     Response;  The public will be able to review control plans
that are included as part of the base year review and enforceable
commitment applications.  However, approval or disapproval of the
application will not be dependent on the proposed control
strategy.  Facilities will have some flexibility in meeting their
enforceable commitment as long as the reductions are 90 (95)
percent of the base year emissions.
2.3.8  Comment:  How do _spurces submit an enforceable commitment
before proposal of a Section 112(d) regulation if the regulation
is proposed before the Early Reductions regulations are
promulgated (e.g., the SOCMI/HON sources) (IV-D-27)?
     Response;  Sources covered by Section 112(d) regulations
scheduled for proposal prior to promulgation of the Early
Reductions regulations should submit an enforceable commitment
according to guidelines provided in the proposed regulation.  If
changes in the promulgated Early Reductions regulations affect a
submittal based on the proposed regulations, the applicant will
be given the opportunity to make appropriate changes- to the
enforceable commitment.  The changes may include altering the
definition of the Early Reductions source.
2.3.9  Comment;  There is a discrepancy in the preamble language
and the language in the proposed regulation regarding enforceable
commitments.  The regulation states that a company will be held
to the emission limitations contained in the enforceable
commitment, while the preamble does not.  The EPA should clarify
that the emission reductions in the enforceable commitment are
enforceable (IV-D-64).
     Response;  The proposed preamble and regulation are
consistent regarding the enforcement of enforceable commitments.
The preamble to the proposed regulation states on page 27356 that
"...once a source has made a commitment to achieve early
reductions, an enforcement action may be brought against that
source if it fails to achieve the reductions...."  Later on the
same page the preamble  states that "Any source that does not
rescind the commitment  prior to December 1, 1993, and fails to
                               2-42

-------
achieve a 90 (95) percent reduction before January 1, 1994 will
be subject to penalties for the emission reductions not achieved
until compliance is achieved with either the enforceable
commitment or the Section 112(d) standard for all affected
sources."  This is consistent with the language in
Section 63.80(c) of the regulation, which states that "If a
source subject to an enforceable commitment fails to achieve
reductions before January 1, 1994, sufficient to qualify the
source for an extension under this subpart, the source shall be
considered to be in violation of the commitment and shall be
subject to enforcement action under Section 113 of the CAA."
2.4  ALLOWABLE EMISSION REDUCTIONS
2.4.1  Comment;  Hazardous air pollutant reductions achieved as a
result of New Source Performance Standards  (NSPS), Prevention of
Significant Deterioration  (PSD), VOC reductions, State
requirements, or shutdowns should not be included in an Early
Reductions demonstration.  Only reductions  achieved for toxic
control  should be used for this Program  (IV-D-06).
     Response;   The  statute  provides that the Administrator shall
issue a  permit to a  source that achieves a  90  (95) percent
reduction in emissions of  HAP's that allows the  source to meet  an
alternative emission limit reflecting the 90  (95) .percent
reduction in  lieu of the otherwise  applicable  standard issued
under Section  112(d)  if the  reduction is made  before the proposal
of a Section  112(d)  standard or if  the  source  enters into an
enforceable commitment before proposal  of the  standard and  to
make such reductions before.January 1,  1994.   The  statute further
provides that the  reduction  will  be determined with  respect to
verifiable  and actual emissions in a base year not earlier  than
calendar year 1987.   There is no  limitation in the statute  as  to
the reasons those reductions were made.
 2.4.2   Comment;  Several commenters expressed concerns  that the
 requirements for extensions are too flexible,  thereby  allowing
 many industries to qualify without making any additional
 reductions beyond those realized for unrelated reasons.
                                2-43

-------
Conunenters believe that only truly voluntary reductions should be
credited toward extensions.  Specifically, the commenters contend
that credit,should not be given for shutdowns due to economic
conditions, compliance requirements, and other CAA requirements
(IV-D-13, iv-D-14, IV-D-22, IV-D-26, IV-D-46).  Other commenters
suggest changing the base year emission deadline to 1990 to
eliminate the problem (IV-D-06, IV-D-16, IV-D-20, IV-D-60,
IV-D-62, IV-D-64, IV-D-67).
     Response;  It is clearly the intent of the CAA to credit
sources for emission reductions accomplished by any means, as
long as the emission reductions are actual and verifiable and
made prior to proposal of an applicable standard or, in the case
of an enforceable commitment, by January 1, 1994.  According to
the legislative history of the CAA, the Early Reductions Program
initially was limited to voluntary reductions.  Some commenters
on the draft legislation suggested that limiting qualified
reductions was unfair.  For example, facilities located in States
with extensive air toxics programs would not be able to benefit
from early reductions.  After considering this and other
comments, there was a deliberate change to the CAA amendments to
allow any reductions in emissions after the base year.  The EPA
interprets this change as the intent of Congress to expand the
Program to allow all early reductions, regardless of how they are
achieved.
2.4.3  Comment:  The commenter agrees that average fugitive
equipment leak factors should not be used but would like the EPA
to establish the maximum percent reduction that can be claimed by
standard options.  For example, a plant with an NSPS program that
proposed a 500 parts per million (ppm) leak definition could use
3£ percent control.  This would greatly reduce the burden on the
reviewer (IV-D-2 6).
     Response;  The NSPS factors are based on modeling of
emissions from hypothetical plants given a number of assumptions
such as frequency  of monitoring, leak rate, repairability, etc.
The factors were developed to provide nationwide estimates of
                               2-44

-------
emissions and should not be applied to specific sources for the
purposes of this Program.  The CAA requires "actual and
verifiable" emissions and verification of this requirement can be
obtained only through screening at individual sources.
2.4.4  Comment;  The Early Reductions Program should accept all
reductions of actual emissions regardless of whether the base
year emissions are in compliance with regulated or permitted
levels.  It was the intent of Congress to reduce current
emissions (IV-D-27).
     Response:  Sources should not be rewarded for being out of
compliance with any existing regulations.  Sources can take
credit for reductions achieved as a result of compliance with
regulations after the base year, but will not be allowed to count
base year emissions that are in excess of those allowed.
2.4.5  Comment:  The definition of "malfunction" should clarify
whether excess emissions incurred during startup and shutdown are
considered malfunctions or routine.  The commenters have been
treating startups and shutdowns as routine (IV-D-33, IV-D-52).
     Response;  Scheduled startups and shutdowns are considered
routine.  Emissions created by these routine operations can be
counted toward base year emissions.  However, startups and
shutdowns associated with malfunctions are not routine and cannot
be  included with base year emissions.  "Actual emissions," as
defined in the proposed rule, do not include excess emissions
from a malfunction.  Likewise, other emissions from startups and
shutdowns directly attributable to the malfunction should not be
included.  In order to clarify this exclusion, the definition of
actual emissions in the Early Reductions regulation has been
modified to  read "... does not include excess emissions from a
malfunction  or any  startups and shutdowns  associated with a
malfunction."
2.4.6  Comment:  The commenter is concerned with the definition
of  malfunction.  As  it  is written in the proposed regulation,
 "failures  that are  caused entirely or in part by poor
maintenance,  careless operation, or any other preventable upset
                               2-45

-------
conditions or preventable equipment breakdown shall not be .
considered malfunctions."  This will allow facilities to earn
emission reductions by performing maintenance on a unit that has
been neglected (IV-D-39).
     Response;  Facilities will not be allowed to credit emission
reductions if the base year emissions were in violation of
emission standards.  However, poor maintenance, careless
operation, or any other preventable upset conditions or
 •                           s»,
preventable equipment breakdown that result in emissions that are
not in violation of any emission standard are not considered
malfunctions and can be credited toward base year emissions.
Thus, the Early Reductions Program provides an incentive for
facilities to improve their maintenance and operation practices
(which would not otherwise be required), resulting in a net
benefit to the environment.
2.5  DEMONSTRATION OF EARLY REDUCTIONS
2.5.1  Comment;  Several commenters urge the EPA to require
particulates to -be controlled to 95 percent, with no offsets of
gaseous pollutants allowed (IV-D-06, IV-D-14, IV-D-16,
IV-D-20, IV-D-25, IV-D-31, IV-D-46, IV-D-57, IV-D-64).  Other
commenters support the weighted average approach for gaseous and
particulate mixtures.   (IV-D-30, IV-D-32, IV-D-34).
     Response;  At proposal, the EPA recognized that there may be
sources that have already achieved 95 to 99 percent reduction in
particulates through control measures employed prior to the
earliest allowable base year, but that have significant potential
reductions from relatively uncontrolled gaseous HAP emissions.
If, for an Early Reductions demonstration, such sources were
required to reduce gaseous HAP's by 90 percent as well as reduce
the remaining particulate emissions by another 95 percent, the
source may find further particulate emission reductions are not
possible or that further reductions likely would exceed
requirements of an applicable Section  112(d) standard for
particulate emissions.  In either case, the source would seek a
                               2-46

-------
compliance extension and thus would not make early reductions of
the gaseous HAP' s.
     A telephone query from a potential applicant exemplified
this problem.  The facility coats metal parts and has vents that
release both gases and particulates.  In this case, particulates
are .very tightly controlled (prior to the earliest possible base
year) but gases are not.  The caller was interested in making
early reductions by reducing the gases by 90 percent, but if the
source also had to achieve an additional 95-percent reduction in
the already tightly controlled particulates, the source would not
be able to enter the Program.
     The EPA recognized that it would be beneficial from an
environmental standpoint to allow a reduction in the gases
without requiring an additional 95-percent reduction in
particulates; i.e., without some allowance for this situation,
the source would not enter the Program and no reductions in
gaseous HAP's would be achieved until required by a
Section 112(d) standard.  The EPA therefore proposed to allow a
source to average the required reductions for gases and
particulates; particulates could be reduced by less than 95
percent as  long as compensating reductions were achieved in
gaseous HAP's.-  The overall target reduction would be between 90
and 95 percent and would be determined by the relative amounts of
gases and particulates emitted from the source in the base year.
This averaging method still would require the same amount of
overall HAP emission reduction as would the separate 90 and 95
percent requirements.
     The proposed averaging method could have been used at any
source, whether the gases and particulates were emitted from
different emission points or the same emission point.  However,
based on commenters' concerns, the EPA has reconsidered the
applicability of  the averaging method.
     The  statutory  language  in Section  112(i)(5) explicitly
requires  95 percent  reduction of particulate HAP's and 90 percent
reduction  of  gaseous HAP's.  The EPA, however, feels that
                               2-47

-------
Congress did not consider, or was not aware of, emission points
that emit both gases and particulates and that different controls
are required for each, and therefore did not provide for this
situation.  Furthermore, the EPA believes that it would be
beneficial to allow a weighted average for gases and particulates
when they are emitted from the same points, especially where only
small quantities of one are emitted with respect to the other, in
light of the alternative of no early emission reduction.
     Therefore, the final rule allows the use of the averaging at
sources in which the gases and particulates are emitted from the
.same points.  For  example, if three emission points at the source
emit both gases and particulates, these emission points could be
aggregated to establish a combined percent reduction between 90
and 95 percent, e.g., the three points could achieve 94 percent
reduction overall. The combined percent reduction would depend
on the relative amounts of gases and particulates emitted by the
points.  By restricting the use of averaging gaseous and
particulate emission reductions, the EPA notes that some sources
emitting both types of pollutants, but from different points may
now be unable to achieve the required early reductions if
averaging were not allowed.  In these cases, because emissions
are emitted from different points, the source should be defined
to exclude those points that create the difficulty in emission
reduction.  However, the source must still conform to one of the
allowable definitions  in the regulation.
2.5.2  Comment;  Currently,  sources that can only achieve a
reduction of  89 percent may  not participate in the Early
Reductions Program.  The commenter  suggests that these sources be
allowed  to participate but that the length of the extension be
determined by the  achieved reductions.  For example, the source
should be given a  1-month extension for each 1.25 percent
reduction in  HAP's (IV-D-02).
      Response;   Participation  in  the Early Reductions Program is
only possible if  reductions  of 90  (95) percent are achieved.
Granting an extension for sources  demonstrating  89 percent
                               2-48

-------
reduction, including extensions, would not be allowed.  The
language in the CAA specifically calls for 90 (95) percent
reductions of HAP's or an enforceable commitment prior to
proposal of an applicable Section 112(d) standard in order to the
included in the Program.
2.5.3  Comment:  The 95-percent reduction requirements should be
limited to PM-10 rather than all particulate matter because the
smaller particles pose a greater health threat (IV-D-59).
     Response;  The CAA specifies a 95-percent reduction of
particulates, without regard to size.  Therefore, the Early
Reductions Program requires reduction of total HAP particulates
only.
2.5.4  Comment:  The Early Reductions Program should allow
partial reduction of HAP on a case-by-case basis.  For example, a
source might substitute one raw material for another, resulting
in a 90  (95) percent reduction of some but not all of the HAP's
generated by the process.  It was suggested that Paragraph
63.72(g) be added that reads:  "A source may be granted an
alternative emission limitation based on the reduction of 90  (95)
percent of one or more of the hazardous air pollutants, if it is
determined on a case-by-case basis that the partial reduction
will result in an improvement in air quality"  (IV-D-29).
     Response:  The CAA specifies that  90  (95) percent reduction
of all HAP's from the defined source must  be achieved.  Selective
HAP reduction is  not allowed, although  some HAP's may be greater
than 90  percent and some HAP's  less thatn  90 percent as long  as
the overall reduction is 90 percent.  The  Early Reductions
Program  allows flexibility in defining  the source, but that .
flexibility cannot be extended  to specific pollutants within  the
source,  to the exclusion of others.
2.5.5  Comment;   Review of base year  emissions should be extended
to 90  days  for at least the first year  of  the  Program to allow
evolving programs to  adequately review  the emissions  (IV-D-26).
     Response;   The EPA recognizes  that 60 days is  a  compact
 schedule.  However,  the EPA will continue  to retain this time
                               2-49

-------
limit in order to expedite the review as much as possible.  In
some instances, the 60-day review period may be impossible to
achieve, but the EPA prefers to limit longer review times to
these exceptional cases rather than extend all review periods to
90 days.  As a practical matter, a 30-day completeness review
combined with a 60-day acceptance review will provide an overall
90-day review time.
2.5.6  Comment!  One commenter stated that the EPA should not
expect emissions testing as the basis for demonstrating
post-reduction emissions.  The testing presumption should at
least be limited to the pollutants that have an accepted test
method  (IV-D-35).
     Response;  Sources will be required to adequately document
post-reduction emissions before being granted an alternative
emission limit.  The EPA requires the best available data to
demonstrate emissions.  Direct measurement of emissions is the
presumed best demonstration for many cases.  However, test data
will not be required to demonstrate all emissions.  Calculations
based on engineering principles, emission factors, or material
balances may be acceptable if the applicant demonstrates to the
satisfaction of the permitting authority that:   (1) no test
method  exists;  (2) it is not technologically or economically
feasible to perform source tests; (3) it can be demonstrated that
the accuracy of a calculated estimate is comparable to source
testing; or  (4) emissions from oife or more set of points are
insignificant compared to total source emissions.  The EPA agrees
that in some situations  (e.g., storage tank emissions or batch
operations using- solvent HAP's) calculations or mass balances
generally will  be acceptable and may be the preferred method of
establishing emissions.
2.5.7   Comment!  The EPA should clarify what is meant by
"sufficiently variable" in Section 63.78(b)(4) of the proposed
regulation.  Batch manufacturing produces inherent process
variations.  The commenters  agree that clearly unreliable
emissions control measures -should be grounds for denial of an
                               2-50

-------
extension, but variations typical of batch processes should not
be grounds for denial (IV-D-19, IV-D-27).
     Response;  The term "sufficiently variable" has been removed
from the language of the regulation.  The EPA recognizes that
batch processes are inherently variable, and that variability
alone should not be a reason for denial.  The reason for denial
will now read:  "The emission of hazardous air pollutants or the
performance of emission control measures is unreliable so as to
preclude determination that the required reductions have been
achieved or will continue to be achieved during the extension
period."
2.5.8  Comment;  It was suggested that a sixth reason be added to
the reasons why calculations may be acceptable:  it is
"physically impossible to perform the testing due to
configuration  of the process equipment"  (IV-D-30).
     Response;  "Physically impossible to perform the test-ing due
to configuration of the process equipment" is the.same as  "it is
not technologically or economically feasible to perform source
tests."   Therefore, this would be a valid reason not to perform
source testing.
2.5.9  Comment;  Commenters would like  emissions to be quantified
as  emissions  per pound of product versus a specific quantity per
year.  This would  assist in overcoming  problems with fluctuating
production rates,  especially with the base year  (IV-D-33,
 IV-F-1).
      Response;  The statute requires  90 (95) percent reduction
 from base year emissions.  Thus, the  Early Reductions Program is
 concerned with absolute  reductions  from base year  emissions, not
 a percent reduction from what  current emissions would have been.
 The alternative emission limit will reflect  a  specific  emission
 level that is 90 (95)  percent  less  than base year.   Sources with
 fluctuating production must consider future  increases  in
 emissions and make adjustments accordingly.
 2.5.10  Comment;   The commenter opposes any efforts by the EPA
 to require continuous emissions monitoring (GEM) to document
                                2-51

-------
early reductions.  The commenter states that the CAA clearly  ;
allows for options other than continuous monitoring (IV-D-65).
     Response;  Continuous emission monitoring is not a general
requirement of the EPA to document that the source has achieved
sufficient reductions.  General monitoring requirements for
permit provisions, including but not limited to, alternative
emission limitations established for the Early Reductions
Program, will be established according to the requirements of
Title V of the CAA.  Section 504 of Title V states that
"continuous emissions monitoring need not be required if
alternative methods are available that provide sufficiently
reliable and timely information for determining compliance."
Therefore, if other acceptable means for determining compliance
are available, GEM will not be required.  Nothing in the CAA,
however, prevents the requirement of GEM for the Early Reductions
Program, and GEM may be required if no other acceptable methods
are available.
2.6  HIGH-RISK POLLUTANTS
2.6.1  Methodology for Selecting High-risk Pollutants
2.6.1.1  Comment:  One commenter questioned the use of the three-
tier methodology for selecting the high-risk pollutants.  The
commenter believes that the EPA should encourage early reductions
of pollutants associated with actual hazards to human health
(IV-D-36).
     Response;  The EPA devised a three-tier approach for
selecting high-risk pollutants in order to systematically screen
the 189 HAP's listed in Section 112(b) of the CAA.  Health
effects data were examined first to rank pollutants based solely
on health effects.  Consideration of health effects alone
however, was judged insufficient to address Section 112(i)(5)(E)
which describes high-risk pollutants as those for which high
risks of adverse public health effects may be associated with
exposure to small quantities.  Accordingly, tier 2 of the
analysis was exposure modeling using the EPA's Human Exposure
Model  (HEM).  As described in the preamble to the proposed rule,
                               2-52

-------
the ambient concentration predicted by the model was compared to
selected risk levels to screen for pollutants likely to have an
adverse effect on public health at the exposure level that was
modeled.
     Tier 3 of the analysis was an examination of actual
emissions data using the 1989 TRI.  Tier 3 was intended as a
"reality check" to determine whether or not the pollutants were
actually emitted at levels that could reasonably be expected to
adversely affect public health.  Pollutants with a low exposure
potential were consequently eliminated from the high-risk list.
After consideration of public comments, the EPA has decided to
eliminate tier 3 from the analysis.  (This decision and its
impact is discussed separately below.)  The EPA believes that
tiers 1 and.2 adequately identify those pollutants that could
reasonably be expected to adversely affect public health, and
thus limited the offsetting of these pollutants in the Early
Reductions Program.
2.6.1.2  Comment;  Two commenters stated that insufficient data
were presented on health effects and that the documentation of
criteria for excluding or including a pollutant were not
specified  (IV-D-22, IV-D-46).
     Response;  The health data and selection criteria used to
select  the proposed high-risk pollutants can be found in the May
30,  1991 memorandum  (Docket No. A-90-47, Item No. II-B-10).9.
2.6.1."*3 Comment;  One commenter  stated that all HAP's with the
EPA's approved potency factors should be included on the
high-risk  list  (IV-D-14).
     Response:   A potency  factor  is a measure of the probability
of an individual's developing  cancer as a  result of exposure to a
specified  unit  ambient concentration of a  pollutant.  Potency
factors for the  carcinogens  listed in  Section  112(b) of  the CAA
vary by several  orders of  magnitude.   Those  that have  low potency
 factors have a low probability of causing  cancer and thus are  not
considered to pose a high  risk under  the criteria used for the
purposes of the Early Reductions  provisions.   The mere existence
                               2-53

-------
of a potency factor is an insufficient reason for a chemical to
be added to the high-risk list.
2.6.1.4  Comment;  The commenter suggested that all carcinogens
classified as Group A or B be listed, whether or not there is an
approved potency factor.  This commenter also recommended
including pollutants identified by the National Toxicology
Program (NTP) or the International Agency for Research on Cancer
(IARC) because these organizations have made determinations of
cancer for chemicals that may not have been reviewed by the EPA
(IV-D-39).
     Response;  The EPA did list all carcinogens with a Group A
weight-of-evidence classification because known human carcinogens
were considered to be of sufficient concern to be termed "high
risk."  For Group B and C carcinogens (i.e., probable and
possible carcinogens, respectively), the weight-of-evidence
classification alone was not considered compelling enough to
warrant listing a pollutant as high-risk.  For these pollutants,
potency and potential for exposure were given careful
consideration. In addition, pollutants without a unit risk
estimate, but listed as potential carcinogens of high concern
under the Comprehensive Environmental Response Compensation and
Liability Act (CERCLA), were also evaluated for potency or
exposure potential which resulted in additional pollutants being
listed.  These pollutants have received extensive review by the
EPA as well as EPA external peer review.
     The EPA is aware that other cancer classifications and unit
risk estimates are available  (such as the NTP and IARC
classifications- mentioned by the commenter), but for consistency,
the EPA review was a primary criterion for  selecting the high-
risk pollutants.  However, if  additional information becomes
available that indicates that  the EPA has overlooked an important
carcinogen that meets the high-risk  list criteria, the EPA may
amend the list.
                               2-54

-------
2.6.1.5  Comment;  One commenter was concerned about the health
effects data bases used to make the initial selection of the
high-risk pollutants.   The commenter is of the opinion that many
data contained in the Instructional Resources Information System
(IRIS) and the data used for CERCLA Section 102 determinations
are outdated, fail to incorporate the latest pharmacokinetic
findings, and are more conservative than they should be because
of linearized multistage (LMS) modeling.  The commenter maintains
that the LMS modeling (used to develop unit risk estimates)
overestimates the potency of carcinogens.  For noncarcinogens,
the commenter says the quality of the study used to calculate the
inhalation reference concentration or oral reference dose
(RfC/RfD) must be considered and the evaluation must be updated
with new data (IV-D-35).
     Response:  The information on chemicals contained in the
IRIS and CERCLA data bases has received review both by the EPA
and external review and is considered to be of the highest
quality.  For this reason, these data bases were relied upon,
particularly for data on carcinogens.  The EPA agrees that
improvements to the  science are being made, such as in the area
of pharmacokinetics, but these types of data for specific
chemicals are still  scarce.  When new information for a
particular chemical  becomes available, the EPA's Carcinogen Risk
Assessment Verification Endeavor Work Group  (which meets
regularly and is  responsible  for development of unit risk
estimates) reviews  this information and updates the EPA's
findings accordingly.  A recent example would be the EPA's
revision of  the  unit risk  estimate  for methylene chloride.
      Typically,  new information  is  released  as an addendum to a
health assessment document, followed by  an update of the IRIS
files.  Because  the IRIS  files are  updated later, for this
analysis the EPA also relied  on health  assessment documents that
have been  reviewed by the  EPA's Science Advisory Board.  It
 should be  noted as well that  the EPA maintains  a  "high  hurdle"
                               2-55

-------
for data quality, which is the reason that IRIS files are changed
only after thorough review and documentation of new information.
     With respect to LMS modeling, the LMS procedure provides a
"plausible upper limit" of the risk that is consistent with some
proposed mechanisms of carcinogenesis (i.e., nonthreshold
response).  Such an estimate, however, does not necessarily give
a realistic prediction of the true risk which may be as low as
zero.  In this sense, the commenter is correct in stating that
LMS modeling is conservative.  However, the EPA believes it has
the best, though limited, scientific basis of any of the current
mathematical extrapolation models.
     With regard to the development of RfC's/RfD's, the quality
of the scientific studies is given careful and thorough
consideration.  In addition, although a limited number of RfC's
have been developed, several have already been updated as new
information becomes available.  If new RfC's/RfD's become
available that would indicate a pollutant should be added to or
deleted from the high-risk list, the EPA may amend the list.
2.6.1.6  Comment;  Two commenters strongly urged the EPA to
confine the selection of carcinogens for the high-risk list to
consideration of potency and not make use of weight-of-evidence
classification.  On the other hand, two other commenters
(IV-D-30, IV-D-33) recommended that the EPA prioritize the high-
risk list using a qualitative rather than a quantitative approach
(i.e, use weight of evidence reviews rather than potency factors
for identifying substances)(IV-D-23, IV-D-31).
    'Response;  The EPA believes it appropriate for
identification of high-risk pollutants to consider both the
weight-of-evidence and the carcinogenic potency of a pollutant.
The listing of several Group A carcinogens was based primarily on
weight-of-evidence because the EPA judged that emissions of Group
A carcinogens are of sufficient concern to warrant limitations on
offsetting under the Early Reductions provisions.  Group A
carcinogens are those that are known to be carcinogenic to
humans.  For other carcinogens on the HAP list  (i.e., probable or
                               2-56

-------
possible human carcinogens), the EPA felt that the weight-of-
evidence classification alone was insufficient to list the
pollutant as high-risk.  For these pollutants, the three-tier
analysis, which considered both potency and exposure potential,
was employed.
2.6.1.7  Comment:  Three commenters objected to the addition of
four human carcinogens to the list after the pollutants had been
excluded by the three-tier selection process.  One of these
commenters further claimed that data for human carcinogens come
primarily from workplace exposure studies conducted prior to
1960, and that these exposures have decreased significantly
(IV-D-23, IV-D-30, IV-D-33).
     Response:  The commenters are correct in noting that four
Group A carcinogens were added to the list after they had been
excluded by the three-tier process.  However, the EPA stands by
the decision to list these pollutants as high-risk based on the
fact that they are known human carcinogens.  .The EPA judged that
emissions of known human carcinogens are of  sufficient concern
that offsetting reductions in other less hazardous pollutants
under an Early Reductions demonstration should be limited.  Three
of the pollutants  (benzidine, bis(chloromethy1)ether, and
chloromethyl methyl ether) are extremely potent carcinogens and
were originally excluded from the three-tier process at tier 3.
This means that, according to the 1989 TRI,  these pollutants were
not emitted  by any reporting facility in quantities sufficient to
exceed the risk benchmark level  proposed by  the EPA.  In spite of
this, the EPA  felt it  prudent to list them as high-risk in case a
source's emissions changed  in the future.
      Since proposal  of the  high-risk list, the EPA has eliminated
tier  3  from  the  screening analysis.  As a  result, even if the  EPA
had decided  not  to use weight-of-evidence  as a criterion, these
carcinogens  would still be  listed as high-risk.   Benzene was also
 included on  the  list even though the risk  benchmark level was  not
 exceeded when a  10 tpy emission rate was modeled.  The rationale
 for this decision was that  in addition  to  being  a known human
                               2-57

-------
carcinogen, benzene is a very high volume chemical and emissions
in excess of 10 tons per year are not uncommon.  The EPA believes
that offsetting emissions of this known human carcinogen with
less hazardous pollutants should also be limited.
     With respect to the comment that data used to classify
carcinogens as Group A come primarily from workplace exposures:
the source of the data does not alter the conclusion that these
pollutants are known human carcinogens.  Human exposure data as
well as data from animal studies lend further support to the
Group A classification.
2.6.1.8  Comment:  Two commenters suggested that the EPA revise
the criteria to include only known human carcinogens (Group A),
probable human carcinogens (Group Bl)f and noncarcinogens for
which chronic human health effects can be expected to occur at
extremely low levels of exposure (IV-D-44, IV-D-49).
     Response;  In selecting pollutants for the high-risk list,
the EPA considered weight-of-evidence, potency, .and exposure
potential of the pollutant.  A weight-of-evidence classification
of Group A (known human carcinogen) was deemed sufficient to list
a pollutant as high-risk.  The EPA does not agree with the
commenters1 suggestion that Group B2 and Group C carcinogens
should not be considered for, or included on, the high-risk list.
In order to be listed, however, carcinogens other than Group A
carcinogens had to pass tier 2 of the analysis, which considered
potency and exposure potential.  Once the pollutants were
initially identified for the high-risk list by the tier screening
process, they were assigned a weighting factor.  The weighting
factor system decided on by the EPA gives additional emphasis to
the weight-of-evidence classification of carcinogens.  Group C
carcinogens are therefore assigned the lowest weights.
     The EPA also disagrees with the commenters1 suggestion that
only chronic human health effects should be considered.
Section 112(b) of the CAA describes HAP's as those pollutants
that present, or may present, a threat of adverse human health
effects including substances that are carcinogenic, mutagenic,
                               2-58

-------
teratogenic, neurotoxic, that cause reproductive dysfunction, or
that are acutely or chronically toxic.  Thus, the EPA believed it
appropriate to assess (and include on the high-risk list, where
appropriate) pollutants that are acutely as well as chronically
toxic.
2.6.1.9  Comment;  Two comment letters questioned the data used
to select noncarcinogens for the high-risk list.  The commenters
maintain that RfCs and RfDs can be based on different health
endpoints and that the values do not reflect the severity of the
health effects.  Similarly, the lowest observed effect level
(LOEL) (used to screen noncarcinogens that do not have an RfC or
RfD) may not necessarily be an adverse or toxic effect.  Also, a
chemical may be extremely hazardous upon high level, short-term
exposure, but be essentially nontoxic at very low levels of
exposure.  The commenters recommend that health effects be
grouped according to  severity at anticipated levels of exposure
in order to differentiate between noncancer effects (IV-D-30,
IV-D-33).
     Response:  In developing the high-risk list, the EPA
considered  several health effects endpoints, including
carcinogenicity, reproductive and developmental toxicity, acute
lethality,  and  systemic effects other than acute lethality  (e.g.,
neurologic  disorders).  The commenters are correct in noting that
the  selection of noncarcinogens for the  high-risk list was  based
on different  health  endpoints.  The EPA  feels this approach is
necessary,  however,  because of the large number of noncarcinogens
on the HAP  list and  the variety of noncancer health effects they
can  potentially cause.  The EPA did not  feel it was appropriate
to  judge whether one health endpoint  was more severe than
another,  e.g.,  whether a  reproductive effect is  "worse"  than a
systemic effect.
      It is true that the  LOEL may not necessarily  be an  adverse
or toxic effect.   If a pollutant's LOEL was low enough to  cause
 the pollutant to be listed as high-risk, the EPA reviewed  the
 critical study for scientific quality.   The EPA found,  in  all
                               2-59

-------
cases, that the study was flawed and, in fact, did not list any
pollutants on the basis of a LOEL.  By comparison, the inhalation
RfC's are typically based on the lowest observed adverse effect
level.  In establishing the RfC, the EPA has made a judgment that
the observed health effect is sufficiently severe to be  .
considered adverse.  Thus, the selection of several
noncarcinogens for the high-risk list was based on inhalation
RfC' s rather than LOEL' s.    „                        •:_...
     The commenters' suggestion that health effects be grouped
according to anticipated levels of exposure in order to
differentiate between rioncancer effects is interesting, but
unworkable for the Early Reductions Program.  The facilities
eligible to participate in the Early Reductions Program represent
all types and sizes of source categories that could emit any
number of HAP's at a variety of emission rates.  For this reason,
the EPA is unable to anticipate exposure levels.
2.6.1.10  Comment:  One comment letter suggested that all HAP's
having potentially serious health effects other than cancer be
listed as high-risk.  The commenter suggested using facility-
specific information to determine whether a health-effects
threshold would be exceeded.  The weighting factor could then be
customized for that facility by relating the weighting factor to
the degree to which a health-effects threshold is exceeded.  The
commenter believes this approach would emphasize reductions in
cases where health-effects thresholds may be exceeded and assigns
less weight to high-risk pollutants in cases where the
health-effects threshold is not likely to be exceeded  (IV-D-31).
     Response;  The high-risk list includes several pollutants
that cause health effects other than cancer.  The listing
criteria that the EPA devised were meant to select a subset of
pollutants from the list of 189 HAP's in Section 112(b), that the
EPA judged to be toxic enough in small quantities to warrant
being listed as high-risk.  The EPA believes these criteria were
used appropriately to identify the high-risk noncarcinogens.
There are several reasons why the EPA prefers that weighting
                               2-60

-------
factors not be customized for each participating facility. .The
primary reason is that the weighting factors reflect the
differences in the toxicity of the pollutants regardless of
emission rates.  For example, if facility X emits 50 tons of
benzidine, a 1,000-tb-l trade is necessary with a HAP not on the
high-risk list.  If facility Y emits 1 ton of benzidine, a
1,000-to-l offset is also required.  The EPA recognizes this is a
conservative approach, but believes it will serve to always
emphasize reductions in the high-risk pollutants even if these
emissions are small.
     In addition, participation in the Early Reductions Program
is available to all facilities throughout the Nation.  For
consistency in program implementation, the EPA prefers the
weighting factors for the high-risk pollutants not be customized
for each participating facility.  In addition, some facilities
and also the reviewing agencies may not have the expertise or
resources to either perform or review site-specific modeling
analyses and determine appropriate weighting factors.
2.6.1.11  Comment:  Two commenters requested documentation
regarding the  selection of the critical study for HAP's that do
not have an RfC or RfD  (IV-D-30, IV-D-33).
     Response:  The critical  study selected was the study that
documented an  LCso for methyl isocyanate and phosgene.  Since
proposal, diazomethane, hexachlorocyclopentadiene, phosphine, and
parathion have also been  added to the list on the basis of an
LC5o  (LD50 in  the case  of parathion).  The critical studies may
be found  in a  memorandum  to  the docket entitled  "Background
Information for Noncarcinogens Recommended for Inclusion  on the
High-Risk List Since  Proposal"  (Docket No. A-90-47, Item  No.
IV-B-8).10
2.6.1.12  Comment;  The comment letter stated that  relying on
cancer potency factors  to weight  toxicity  and screen  unacceptable
risk resulted in some illogical choices  for  the  high-risk list.
For example,  benzene, a Group A carcinogen emitted  in large
 amounts,  did not make the list  (it was added on  even  though  it
                               2-61

-------
failed the  screening criteria).  In contrast, chlordane, a
Group B2 carcinogen with  lower emissions did make the list.
Also, compounds  such as carbon tetrachloride, chloroform, and
methylene chloride, which are  B2 carcinogens, have noncancer
effects, and  are emitted  in  significant quantities, did not make
the  list.   Comment letter IV-D-30 also stated that the selection
of high-risk  carcinogens  should include consideration of cancer
classification in selecting jthe pollutants and in their weighting
(IV-D-26).
     Response;   In screening the list of 189 HAP's, the weight-
of-evidence of a carcinogen  was a criterion to the extent that
Group A carcinogens were  added to the list.  The EPA added Group
A carcinogens to the list to ensure that exposure to these
compounds would  be limited to  the extent possible.  Since
proposal of the  high-risk list, the EPA has revised the weighting
system in such a way that the  weight-of-evidence classification
•is given greater emphasis.   Carcinogens such as those mentioned
in comment  letter IV-D-26 that were not listed on the basis of
carcinogenicity  were also evaluated for effects other than
cancer.  The  specific  chemicals mentioned by the commenter did
not  meet the  screening criteria for these effects.
2.6.2.  Comments Concerning  Tier 2  (Exposure Modeling)
2*.6.2.1  Comment; Five commenters recommended that the EPA not
include tiers 2  or 3 of the  screening analysis.  These commenters
felt that selection of high-risk pollutants should be based only
on toxicity (IV-D-14,  IV-D-22, IV-D-31, IV-D-39, IV-D-46).
     Response;  In selecting the high-risk pollutants, health
effects data  were examined first to rank the pollutants on the
basis of health  effects.  Consideration of health effects alone
however, was  judged insufficient to address Section 112(i)(5)(E),
which describes  high-risk pollutants as those for which high
risks of adverse public health effects may be associated with
exposure to small quantities.  Accordingly, tier 2 of the
analysis was  exposure  modeling.  The EPA believes this step of
the  analysis  to  be appropriate.  However, after reconsideration,
                               2-62

-------
the EPA has decided to drop tier 3 from the analysis (this tier
considered emissions data as a criterion for listing).
2.6.2.2  Comment:  Many comments were received concerning the
parameters chosen for the tier 2 exposure analysis (IV-D-14,
IV-D-16, IV-D-22, IV-D-33, IV-D-39, IV-D-47, IV-D-61).  Several
commenters felt that the exposure assumptions were not
conservative for several reasons, including the meteorological
data, the stack parameters,, and the limitations of the HEM.
(Comments received concerning the presumptive risk benchmark
level are addressed separately below [IV-D-14, IV-D-16,
IV-D-22, IV-D-33, IV-D-39, IV-D-47].)
     One commenter was of the opinion that 1 year of
meteorological data is not representative and downwash was not
considered  (IV-D-47).  Two commenters suggested a 3.5-meter stack
height and a 20-meter distance to the nearest residence  (IV-D-39
and  IV-D-47). According to one commenter, the modeling should
have used more realistic assumptions about pollutants and
emission sources  (IV-D-16).  Another commenter felt  that the HEM
is too simplistic because it does not consider actual exposure,
short-term  exposures, or population activity patterns  (IV-D-33).
On the other hand, two commenters felt the 500-meter distance to
the  nearest residence was too close  (IV-D-42, IV-D-61).  One
commenter recommended a 3,000-meter distance and a  10-year
population  residence time  (rather than the 70-year  lifetime
exposure associated with  the cancer potency factor)  (IV-D-61).
     Response;   The second  tier  of the screening analysis to
select the  high-risk pollutants  was  a generic exposure modeling
exercise.   The  EPA intended that the modeling results be used to
determine which pollutants  merited further  analysis based  on
exposure potential.  The  EPA did not use the modeling results to
estimate  actual risk levels, but did use the modeling results to
differentiate between  pollutants.
      The  EPA still considers  the use of  generic modeling
parameters appropriate for this analysis because  a  wide  range  of
 emission release and exposure scenarios  across many varied source
                               2-63

-------
categories are anticipated under the Early Reductions Program.
     The HEM was used to estimate a theoretical downwind
concentration of a typical HAP.  Because the pollutants are
released from various types of sources with variable stack
parameters (stack height, emission velocity, distance to nearest
residence, etc.), no attempt was made to model site-specific
conditions.  Meteorological data used were representative of
between 2 and 10 years of data from the selected meteorological
station.  A 10-meter stack height was chosen to represent a
chemical plant.  For the proposed rule, the modeling parameters
also assumed a 70-year continuous exposure and a 500-meter
distance to the nearest residence.  Upon re-evaluation, the EPA
has made changes to these two parameters.
     First, the 70-year period over which exposure to a pollutant
was assumed to occur continuously has been changed for the
purposes of this rulemaking to a 33-year period.  Thirty-three
years represents the 95th percentile for the number of years an
individual would remain at the same residence, based on
population mobility and mortality data.  The 33-year period was
selected for use in this analysis as representative of a
reasonable worst-case approach to assessing exposure duration.
     Second, the EPA reconsidered the choice of 500 meters to the
nearest residence and has revised the analysis using a 200-meter
distance.  The 200-meter distance to the nearest residence is the
distance that has been most often used by the EPA for analyses
estimating exposure to emissions from point sources.  This
distance represents a reasonable worst case for facilities that
are anticipated to participate in the Early Reductions Program.
The net impact of these changes was the addition of
hexachlorocyclopentadiene, phosphine, parathion, and manganese
compounds to the list.
2.6.2.3  Comment;  One commenter stated that the EPA modeled
emissions  (hydrazine in particular) as if emitted from point
releases rather than as fugitive releases, which resulted in
higher predicted concentrations  (IV-D-42).
                               2-64

-------
     Response:  The EPA's screening model does not address
fugitive or dispersed releases.  The EPA agrees that modeling
fugitive emissions as a point source release will generally
result in higher predicted concentrations.  On the other hand,
for this analysis, the EPA used a 10-meter stack.  If a lower
stack height such as 3.5 meters had been modeled to account for
fugitive ground level releases, the predicted concentration would
have probably been higher than that predicted by modeling a point
release from a height of 10 meters.  The generic modeling
parameters chosen by the EPA for this analysis are believed to
represent the majority of facilities that are anticipated to
participate in the Early Reductions Program and the dispersion
and emission characteristics of most pollutants.
2.6.2.4  Comment;  Two commenters questioned the benchmarks
chosen for health effects other than cancer.  These commenters
felt that neither the LOEL divided by a safety factor of 100 nor
the LC5o divided by a safety factor of 1,000 were conservative
enough.  The commenters also thought that using a value one order
of magnitude above the RfC was not conservative enough  (IV-D-22,
IV-D-39).
     Response:  As stated in the preamble to the proposed rule,
uncertainty factors were applied to the health effects benchmarks
because  the LOEL and LCsg values represent levels at which health
effects  are known to occur.  These uncertainty factors were meant
to account for variables such  as interspecies variations and
sensitive subpopulations, but  did not incorporate all the factors
typically used in developing RfC's.  The  uncertainty factors  used
to develop an RfC are typically applied to a  lowest observed
adverse  effect level  (LOAEL) or a no observed adverse effects
level  (NOAEL).  The  resulting  RfC represents  a level below which
the  potential for adverse public health effects  is considered to
be negligible.  However, the purpose of this  analysis was to
identify pollutants  whose emissions could potentially present a
high risk,  not a  de  minimis risk.   Therefore,  the uncertainty
 factors used in this analysis  were not meant  to  account for  all
                               2-65

-------
the factors typically used in developing an RfC.  For this
analysis, the EPA made a judgment that if a source's emissions
were such that a LOEL or LC50 could potentially be exceeded
(after application of the uncertainty factors and under the
modeling scenario used), then the pollutant could be considered
high-risk.  In this application,, the EPA believes uncertainty
factors of 100 and 1,000 to be appropriate.  Similarly, with
respect to using a value one order of magnitude above, the RfC,
the EPA believes this to be an appropriate estimate of a level at
which health effects could potentially be significant enough to
be regarded as high-risk.
2.6.3  Comments on One in Ten Thousand Risk Benchmark
2.6.3.1  Comment;  Four comment letters were received that stated
that the one in one thousand presumptive risk benchmark was not
conservative enough  (IV-D-16, IV-D-22, IV-D-39, IV-D-47).
Commenter IV-D-47 recommended that the EPA include every
carcinogen with a potency factor greater than 1.35 X
10~8[ (ug/mS)""1] based on modeling a 3.5-meter stack.  Commenter
IV-D-16 argued that Congress intended that sources with risk
levels of one in one million be controlled.  On the other hand,
one comment letter recommended a risk level of one in 1,000.
This commenter believes that the EPA's one in ten thousand risk
benchmark is inconsistent with the "Savings Provision" in
Section  112(q)(l), where standards promulgated under Section  112
in effect prior to the  1990 amendments are to remain in force
unless revised by the EPA  (IV-D-33).
     Response;  The  exposure modeling exercise conducted as part
of tier  2 of the screening analysis was used as a tool to
identify pollutants  that could potentially cause a high risk  to
public health  (under the exposure scenario that was modeled).
The EPA  has decided  that for carcinogens a one  in ten thousand
presumptive risk level  is  appropriate for this  analysis.  This
decision is based primarily on guidance concerning acceptable
risk  discussed  in the NESHAP for benzene promulgated on September
14,  1989 (54 FR 38044J.11
                               2-66

-------
     In the preamble to the benzene decision, the use of the
one in ten thousand benchmark is described:
     "EPA will consider the extent of the estimated risk were an
     individual exposed to the maximum level of pollutant for a
     lifetime.  The EPA will generally presume that if the risk
     to that individual is no higher than approximately one in
     ten thousand that risk level is considered acceptable and
     the EPA then considers the other health and risk factors to
     complete an overall judgement on acceptability.  The
     presumptive level provides a benchmark for judging the
     acceptability of maximum individual risk, but does not
     constitute a rigid line for making that determination."
     In the case of the screening analysis for high-risk
pollutants under the Early Reductions provisions, the EPA
believes the one in ten thousand presumptive risk level has been
used appropriately.  The commenters are reminded that the generic
exposure analysis was not meant to estimate the actual risk from
any one type of source category; rather it was used to
differentiate between pollutants based on relative toxicities.
The EPA recognizes that other modeling assumptions could have
been used that would have resulted in either more or fewer
pollutants exceeding the presumptive risk level.  However, the
EPA does not agree with the commenter's suggestion that every
                                                       -•ft
carcinogen with a potency factor greater than 1.35 X 10""°
[(ug/mS)"1] should be considered high risk.
     With respect to the comment that Congress intended that
sources posing risks of one in one million be controlled, the
commenter is referring to Section 112(f)(A), the so-called
"residual risk" provision.  In this provision, it is stipulated
that additional emission standards must be promulgated if a
source's emissions pose a risk greater than  one in one million
after  application of control technology  [as  stipulated in
Section  112(d)].  The commenter  is reminded  that the Early
Reductions Program provides an extension of  the compliance time
for applying  control technology  that  is  stipulated by standards
promulgated under Section  112(d).  After the compliance time
extension,  facilities participating  in the Program must still
                               2-67

-------
apply the required technology if their emissions exceed specified
levels.  In addition, facilities that participate in the Early
Reductions Program are not exempted from a residual risk test
under Section 112(f), so the commenter.'s contention that the
intent of Congress is not being fulfilled is unwarranted.
     With respect to the Savings Provision of Section 112(q)(l),
the commenter is reminded that the Early Reductions Program is
not an emissions standard.  The use of the presumptive risk
benchmark of 1 in 10,000 in the screening analysis for selecting
high-risk pollutants has no bearing on NESHAP that have already
been promulgated.
2.6.4.  Comments on Tier 3
2.6.4.1  Comment;  Five comment letters found it inappropriate to
eliminate pollutants from the high-risk list based on emissions
data, as was done in tier 3 of the screening analysis.  The
commenters contend that use of the TRI for emissions data is
inappropriate because not all pollutants are reported. Only large
facilities in the manufacturing sector report to the inventory,
and not all types of source categories are covered.  The
commenters feel that the EPA should not eliminate substances
because they are not currently in common use and that the tier 3
screen should be dropped from the analysis (IV-D-14, IV-D-16,
IV-D-22, IV-D-33, IV-D-39).
     Response; The EPA agrees with the commenters1 concerns
regarding use of the TRI and has consequently dropped the tier 3
screen from the analysis.  The EPA also agrees that chemicals not
currently in common use should still be listed as high-risk in
case production of that chemical is resumed at a later date.  The
impact of this decision is that the following pollutants have
been added to the high-risk list as a  result of dropping tier 3;
2-acetylaminofluorene; 1,2 diphenyl hydrazine; toxaphene; nickel
compounds; N-nitrosodimethylamine; ethylenimine  (aziridine);
dimethylcarbamoyl chloride; phosphorus; and diazomethane.
                               2-68

-------
2.6.5  Comments on Offsetting the High-risk Pollutants
2.6.5.1. Comment; . Several commenters believe that the indexed
weighting system for the high-risk pollutants conflicts with the
CAA because it allows offsetting of high-risk pollutants with
non-high-risk pollutants.  The commenters state that high-risk
pollutants should not be allowed to be offset.by pollutants not
on the high-risk list, or only be allowed when high-risk
pollutants are emitted in trace amounts (IV-D-14, IV-D-16,
IV-D-20, IV-D-22, IV-D-57).  Other commenters agree with the
weighting factor concept and the ability to offset high-risk
pollutants (IV-D-24, IV-D-36).
     Response;  Contrary to several commenters' contentions, the
high-risk pollutant strategy does not conflict with
Section  112(i)(5)(E).  That provision requires the Administrator
to limit, by regulation, the use of offsetting reductions of
other HAP's in counting towards the 90-percent reduction of
high-risk pollutants.  The statute does not say or even imply
that the regulation must prohibit such offsetting reductions.
     In response  to the comment that these offsets should only be
allowed when these pollutants are emitted in  trace amounts, the
EPA notes that the weighting system effectively restricts the
offsetting of  emissions of the high-risk pollutants. The amount
of reduction of  a pollutant not on the high-risk list  needed to
offset  a high-risk pollutant  (up to 10,000 to 1) is so large that
only relatively  trace  amounts could be traded.  Even in
situations where the pollutant concentrations are relatively
large  (i.e., not trace amounts), it is appropriate to  allow
offsetting as  long as  the weighting factor system is followed.
If  1 ton of  a  hazardous  pollutant with a weighting factor of  10
is  offset with a 10-ton  reduction in  a pollutant with  a weighting
factor of  1, the offset  should generally provide an equivalent
risk reduction.   Consequently, the EPA believes  such offsets
 should not  be  prohibited.   Note  that  these relative trades  deal
with decreases in the high-risk  pollutants.   It  is not envisioned
 that any successful  Early Reductions  submittal will include any
                               2-69

-------
increase in high-risk pollutants other than incidental trace
amounts.
2.6.5.2  Comment;  Several comments stated that trading among
pollutants with different types of health effects is
inappropriate.  The commenters recommend using critical ;health
effects and toxic potency information to place the listed
pollutants into categories, with trading allowed within a
category.  Trading across categories would be allowed only for
categories based on similar health effects.  Weighting factors
based on relative potencies would be used in such tradeoffs
(IV-D-14, IV-D-20, IV-D-22, IV-D-31, IV-D-39, IV-D-46).
     Response;  The development of the weighting system and the
decision to allow trading between carcinogens and noncarcinogens
is intended to provide flexibility to the participating
facilities in achieving their emission reduction goals.  As such,
it is a policy decision and not one based on purely scientific
grounds.  If trading between pollutants with different health
endpoints was restricted, this could potentially exclude some
facilities from participating in the Early Reductions Program.
The EPA feels that such restrictions would not benefit the goals
of the Program as a whole.
2.6.6.  Comments on Development of the Weighting Factors
2.6.6.1  Comment;  Several comment letters made reference to the
development of weighting factors for the high-risk pollutants.
One•commenter recommended the weighting factors be based on
weight-of-evidence classification  (IV-D-45).  On the other hand,
three commenters recommended the weighting factors be based only
on potency as described fay the cancer potency factor (IV-D-31,
IV-D-33, IV-D-44).  One commenter  further suggested that the
cancer potency factor be multiplied by 1 million to arrive at the
weighting factor  (IV-D-31).  Another commenter suggested a
combination approach where both weight-of-evidence and potency
would be considered.  In this approach, pollutants with the same
weight-of-evidence classification  and having potency factors
within an order  of magnitude would be grouped together  (IV-D-39).
                               2-70

-------
     Response;  The weighting factor system has been revised
slightly since the high-risk list was proposed.  In the final
rule, carcinogens with a Group C classification (possible human
carcinogens) have been assigned a lower weight.  Greater
uncertainty exists regarding the evidence for a Group C
classification than for chemicals classified as Group A (known
human carcinogen) or Group B (probably carcinogenic to humans).
A Group C classification is defined by positive carcinogenicity
in a single experiment, a tumor response of marginal statistical
significance, or finding benign tumors,only.  A Group B
classification is defined by carcinogenicity in two or more
animal species, strains, or experiments, either with or without
limited human evidence.  Group A is reserved for known human
carcinogens.
     The greater uncertainty regarding a Group C classification
is reflected in a lower weighting factor.  As a result, two Group
C carcinogens  (1,1,2,2 tetrachloroethane and vinylidene chloride)
have been assigned weighting factors of 1 rather than 10 as
proposed.  The EPA does not feel, however, that Group B
carcinogens should necessarily be weighted lower than Group A
carcinogens.  With respect to Group B chemicals, a sound
foundation exists regarding carcinogenicity in animals, but the
human data are either  inconclusive or of limited value, most
likely reflecting the  difficulty of obtaining quality
epidemiologic  data*  The EPA is currently revising the cancer
classification scheme  to account for these data gaps.  Therefore,
for  this weighting system, chemicals classified as Group A or  B
carcinogens are  grouped together and ranked by potency.
     For carcinogens,  the actual weighting factors are based on
the  potency of the high-risk carcinogens relative to carcinogens
not  on the  high-risk list.  This approach is the same as was
proposed.   Multiplying the potency  factors by  one million to
calculate weighting  factors as  one  commenter suggested, would
result in the same relative ranking of  the pollutants, but higher
weighting factors.   The EPA believes that basing the weighting
                               2-71

-------
factors on the differences between the potency of the high-risk
carcinogens and the geometric mean potency of the group of
carcinogens not on the high-risk list is more justifiable.  The
weighting factors for. noncarcinogens are discussed below.
2.6.6.2  Comment;  Four commenters believe the EPA should
establish more weighting factor categories to reflect more
accurately the different toxicity levels of the various
pollutants (IV-D-22, IV-D-23, IV-D-31, IV-D-44).  On the other
hand, two commenters believe more pollutants should be grouped
together so that there is not -such a wide range in weighting
factors (IV-D-36, IV-D-45).
     Response;  The EPA has reviewed the weighting factor
categories and still finds four separate categories to be
appropriate.  As described in the comment above concerning
weighting factors, there were some adjustments made to the
placement of.certain pollutants.  However, the general categories
with weighting factors of 100,000, 1,000, 100, and 10 are still
believed to be appropriate.  The difference between the most
potent carcinogen and the least potent carcinogen with a
weighting factor of 10 is about 35 times  (with the exception of
one pollutant).  The EPA does not feel that this is enough of a
difference to warrant another weighting factor category.
2.6.6.3  Comment;  One commenter stated that several high-risk
pollutants that may be emitted in trace amounts as a byproduct of
combustion are given relatively low weights of  10.  Because these
pollutants are emitted in small quantities, the commenter
believes it  is reasonable to require  a relatively large offset
 (IV-D-39).
     Response;   The final weighting factor system does not
incorporate  consideration of emissions, but was based ori  a
combination  of cancer classification  (weight-of-evidence) and
cancer potency.   The EPA believes that requiring a large  offset
because the  pollutant is emitted in trace amounts, without
consideration of the relative toxicity of the pollutant,  is not
 justified.
                               2-72

-------
2.6.6.4  Comment;  One comment letter stated that the EPA needs
to devise a system that requires a certain percentage reduction
in the target pollutant (i.e., the high-risk pollutant of
concern).  A goal of, for example, 75 percent reduction of the
target pollutant should be required before offset credits of
reductions of other compounds could be used to round the
reductions up to the desired 90 percent (IV-D-04).
     Response:  The EPA carefully considered the merits of this
suggestion.  However, the EPA has decided that imposing this type
of requirement would add significantly to both the administrative
burden of reviewing and approving an Early Reductions application
and the technological burden of actually achieving the
reductions.  The EPA believes that the weighting system published
in the final rule adequately addresses emissions of the high-risk
pollutants and is consistent with the overall goal of achieving
early emissions reductions.
2.6.6.5  Comment!  One commenter suggested1 that the EPA include
consideration of emissions in determining the weighting factor.
This commenter believes that the weight-of-evidence
classification and amount of emissions of a compound may give  a
truer picture of which compounds pose the greatest risk
 (IV-D-26).
     Response;   The weighting factors do not incorporate
emissions  because small amounts of one pollutant may pose a much
higher  risk  than greater emissions of a different pollutant.
Also, the  Early  Reductions determinations will be done on a case-
by-case basis.   If  a weighting  factor, which would apply
 industry-wide, was  biased by emissions data, it could lead to
 inappropriate offsetting at  a facility having  significant
 emissions  of a very toxic pollutant  that  had been assigned a  low
weighting  factor on the basis of  low nationwide emissions.
 2.6.6.6  Comment;   Three commenters  questioned the EPA's  policy
 decision concerning weighting factors for noncarcinogens.  The
 commenters believe that the  EPA must assess many  additional
 factors when ranking the  hazards  of  noncarcinogens.   These
                               2-73

-------
factors include the severity of the health effect, its
reversibility, and chemical properties such as the half-life of
volatile compounds in air, vapor pressure, persistence, and
bioaccumulation potential (IV-D-33, IV-D-45, IV-D-47).
     Response;  In assigning weighting factors to the
high-risk pollutants, the EPA attempted to base the factors on
the relative toxicity of the compounds.  For the carcinogens, the
cancer potency factor is a straightforward measure of relative
toxicity and was used in conjunction with the weight-of-evidence
classification to develop the weighting factors.  For the
noncarcinogens, however, there is not a comparable measure of
toxicity that can be used consistently for pollutants with
different health effects.  In the absence of such a measure, the
EPA decided to assign a weighting factor of 10 to the
noncarcinogens on the high-risk list.  Many of the carcinogens
were also assigned a weighting factor of 10.
     The EPA chose a weighting factor of 10 for the
noncarcinogens in recognition that noncancer health effects can
be as seriously debilitating as the effects of cancer.  Another
alternative would have been to assign these pollutants a
weighting factor of 1 because other indices were not available.
The EPA rejected this option because only by assigning a
weighting factor greater than 1 is offsetting limited between
these noncarcinogens and other pollutants not on the high-risk
list.
     In the final rule, three noncarcinogens were assigned
weighting factors greater than 10.  The rationale for increasing
the weighting factor from 10 to 100 for mercury is based on
consideration of persistence in the environment and
bioaccumulation.  After a review of the scientific data, two
other noncarcinogens  (acrolein and 2-chloroacetophenone) have
been assigned weighting factors of 100 rather than 10 in the
final rule to provide an adequate margin of safety from adverse
health effects.  This decision is explained fully in a memorandum
to the docket entitled  "Need for Additional Weighting of Two
                               2-74

-------
Chemicals." (Docket No. A-90-47, Item IV-B-4)12
2.6.7  Comments Concerning Bioaccumulation/Persistence Factor
2.6.7.1  Comment;  Four comment letters discussed pollutants that
persist in the environment and/or bioaccumulate.  These
commenters recommended that any pollutants identified as
persistent or bioaccumulative be added to the list.  Specific
chemicals such as polychlorinated biphenyls (PCB's), metals, and
some pesticides were specifically recommended for listing.  The
commenters also recommended that an additional weighting factor
of 10 be applied to these pollutants (IV-D-14, IV-D-16, IV-D-39,
IV-D-47).
     Response;  After consideration of public comments, the EPA
has decided to adjust the weighting factors for chlordane,
heptachlor, hexachlorobenzene, mercury compounds, and toxaphene
from 10 to 100 based on persistence in the environment and
bioaccumulation.  The weighting factors for these particular
pollutants were increased based on persistence and
bioaccumulation data compiled for the EPA in support of the CAA
Great Waters Study.
2.6.8  Comments on Specific Chemicals
2.6.8.1  Comment;  One comment urged EPA to add radionuclides to
the high-risk list  (IV-D-39).
     Response;  The EPA did not add radionuclides to the
high-risk  list because radionuclide emissions are measured in
term's of activity rather than mass and it would be extremely
difficult  to equate the two for the purpose of offsetting.  The
EPA recognizes, however, that radionuclides could potentially be
present  in trace amounts from some combustion sources.  To
account  for this, language has been added to the final rule that
stipulates that  if  a radionuclide source is included in the
emissions  pool,  the EPA will not allow increases in radionuclide
emissions  under  any post-reduction scenario.
2.6.8.2  Comment;   One comment  letter questioned why lead is not
on the high-risk list  (IV-D-26).
                               2-75

-------
     Response;  Airborne lead emissions are currently regulated
by a National Ambient Air Quality Standard (NAAQS).   The EPA is
currently reviewing the data the NAAQS is based on and expects to
propose a revised standard in the near future.  Given this and
other policy considerations regarding identifying a criteria
pollutant as high risk, at this time EPA believes it is
appropriate to leave lead compounds off the high-risk list.
2.6.8.3  Comment;  One comment letter questioned the development
of the inhalation RfC for methylene diphenyl diisocyanate (MDI).
The commenter claimed that the EPA should only use the thoracic
data (lung effects) and not the extrathoracic data (nasal passage
effects) to calculate an RfC for MDI (IV-D-41).
     Response;  The RfC/RfD Verification Work Group judged that
the nasal effects observed in the Reuzel study are of concern
with respect to chronic exposure situations.  Analysis of the
incidence and severity of these lesions at necropsy at 1 and 2
years of exposure indicate that both increased with continuing
exposure. In addition, lesions described as degenerative occurred
in 3/6 nasal sections at 2 years compared with lesions in 2/6
sections at 1-year examination.  Basal cell hyperplasia in the
2-year study also was accompanied by atrophy of the olfactory
epithelium. The Work Group considers that the spectrum of effects
in the nasal cavity is more extensive than minimal irritation.
The increase in severity and incidence suggests that similar and
more extensive damage may occur in humans as exposure continues.
     The Work Group agrees that MDI exposure in the Reuzel study
resulted in adverse lung effects.  Accumulation of pigment by
macrophage was accompanied by tissue damage in areas where
pigment-laden macrophage accumulated.  The Work Group considers
the lung effects to be biologically significant.
2.6.8.4  Comment;  Commenter believes the EPA's use of several
uncertainty factors to calculate the RfC for MDI is not warranted
 (IV-D-41).
     Response;  The RfC/RfD Verification Work Group has reduced
the data base uncertainty  factor from  10 to 3 because two  species
                               2-76

-------
were evaluated. This action reduced the total uncertainty from
300 to 100, which results in an RfC that is somewhat higher.  The
data base factor is considered justified since there are no
developmental/reproductive studies in animals for MDI and there
is only limited information pertaining to human exposures.  In
any case, the revised RfC is still sufficiently low to meet the
high-risk screening criteria.
2.6.8.5  Comment;  One.commenter stated that the EPA did not take
into account the low vapor pressure of MDI in the dispersion
modeling performed in tier 2 of the analysis.  Also, the
commenter subsequently submitted additional data to the EPA that
showed that MDI emissions have been over-reported to the TRI.
According to the commenter, there are no facilities emitting MDI
that exceed the "significant source" benchmark the EPA
established in tier 3 of the analysis.  (See memorandum to the
docket entitled "Diisocyanates as High-Risk Pollutants for the
Early Reductions Program - Meeting between the Environmental
Protection Agency and the Chemical Manufacturers Association
[CMA]."  (Docket No. A-90-47, Item No. IV-E-13)13(IV-D-41).
     Response:  As discussed above in the response to
Comment  2.6.1.1, the EPA has decided to drop tier 3 from  the
analysis and  list pollutants on the basis of the health effects
and potential  for exposure.  Therefore, emissions information no
longer has any bearing on the pollutant selection process.  The
EPA agrees with the commenter that MDI has a low vapor pressure
and is a solid at ambient temperatures.  However, the EPA still
believes that the dispersion model used was appropriate for MDI  •
even  if  it were to be  emitted as an aerosol.  It is likely that
MDI is emitted at temperatures higher than ambient and so would
disperse much like any other pollutant.
2.6.8.6  Comment;  One comment letter questioned the RfC
developed  for 2,4 toluene diisocyanate  (TDI).  The commenter
asserts  that developmental  and reproductive  studies are available
 for  TDI, so that  an  uncertainty  factor  of  3  is not warranted.   In
 addition the commenter claims that existing  data indicate that
                               2-77

-------
the NOAEL for TDI is above the value used by the EPA.  The
commenter submitted data in support of these comments (IV-D-66).
     Response;  As a result of additional data becoming
available, the EPA has decided to revise the RfC for TDI.  As a
result, TDI is no longer listed on the high-risk list.  Responses
to the commenter's specific concerns are addressed below.
     The EPA RfD/RfC Verification Work Group agreed at its
December 1990_meeting that an uncertainty factor of 3 is not
needed for developmental and reproductive effects because
experimental data indicate that little material reaches the
bloodstream.  The Tyl (1988)14 and Tyl and Bradley (1989)15
studies are being evaluated and discussion has been incorporated
into a revised coversheet.  Because the study results are
negative, it is reasonable to expect that an uncertainty factor
for these endpoints will not appear in the final coversheet.
     With respect to the NOAEL for TDI, at the December 1990
meeting of the Work Group, the study by Diem et al.  (1982)16 was
selected as the critical study and chronic pulmonary function
loss was selected as the critical effect.  At the time of these
deliberations, neither the Bugler et al.  (199I)17 nor the Jones
et al.  (1991)1,8 study was available.  In  spite of study drawbacks
(e.g., additions to cohort, lack of exposure characterization in
the first 2 years, and the cumulative exposure characterization),
the Work Group agreed that the Diem et al.  (1982)16  study was
well-conducted, the lung deficits were biologically  significant,
and the data were extensive enough to identify a conservative
LOAEL. The question of whether time-weighted-average (TWA)  levels
or peak levels of exposure were appropriate for exposure
characterization was extensively discussed.  The Work Group,
after  consultation with two consultants and one of the authors  of
the study, felt that TWA measurements were more appropriate.
There  is  no a priori information to suggest that peak exposures,
in spite  of high correlation with  lung deficits, are better
measures  of exposure than TWA  levels, which also show high
correlation.  It was agreed that instantaneous measurements below
                               2-78

-------
5 parts per billion (ppb) are unreliable because of
methodological considerations.  However, TWA levels derived from
many measurements offer more reliable data.  For this reason, TWA
levels were selected to characterize .the group that experienced
lung function decrements.
     Decrements were noted in only the high-cumulative-exposure
group of 21 never-smokers.  Although the job categories for this
high group had a geometric mean TWA of 4.5 ppb, the exposures of
these 21 individuals were not identified in the study findings.
Thus, it is possible that their true TWA may have approximated
more closely the TWA of the low-cumulative-exposure category, a
geometric mean of 1.1 ppb. In the absence of exposure data and
pre-study exposure history for these 21 individuals, the Work
Group felt it prudent and conservative to select 1.1 ppb as a
"minimal" LOAEL for lung function decrements and no NOAEL was
identified.  It is acknowledged by the Work Group that this LOAEL
is equivocal and may actually be higher.  Studies by
Peters19'20'21'23, Wegman22'23'24, and Musk25'26 suggested that
effects observed by Diem and colleagues were real and the studies
together provide supportive evidence of a concentration-response
relationship.
     The analytical techniques used in the Diem et al.  (1982)
study involved state-of-the-art technology, a technology (paper
tape personal monitors) that was used also in the Jones et al.
 (1991)18 and Bugler et al.  (1991)17 studies of TDI-exposed
individuals.  The limitations of this technology were recognized
by Diem and colleagues and their use of personal monitors is
considered to accurately reflect breathing zone concentrations of
TDI vapors.
     The revisions to the RfC coversheet will also include an
evaluation of the recently available Bugler et al.  (199I)17
 study.  It is anticipated that a discussion of the Jones
et al.  (1991)18  study will also be  included in the coversheet
revised by the EPA  for TDI,  assuming that publication of this
 study will take  place in a timely manner.  The summary  of the
                               2-79
16

-------
findings of the Jones et al. (1991)18 study included in the
commenter's submission appears to lack the level of detail for
proper evaluation.  Data from the Bugler et al. (199I)17 and
Jones et al. (199I)18 studies may necessitate modifications to
the confidence statement in the RfC coversheet with respect to
the database.  Thus, the final RfC may be different.
2.6.8.7  Comment;  Two commenters stated that chromium compounds
[chromium IV] should be listed only as hexavalent chromium
compounds [chromium IV] because it has been shown that trivalent
chromium does not pose the same risk as hexavalent chromium.
According to the commenters, trivalent chromium is not a
carcinogen.  The EPA's 1984 Health Assessment Document27
concludes that trivalent chromium is not reported to be
carcinogenic by any route of administration.  Within the last
year, IARC has concluded that "metallic chromium and chromium
(III) compounds are not classifiable as to their carcinogenicity
to humans (Group 3)."  The U. S. Department of Health and Human
Services also concluded that there is "inadequate evidence" for
carcinogenicity of chromium compounds other than hexavalent
chromium (IV) .  According to the commenters, concentrations of
chromium (III) 20 to 50 times higher than chromium  (IV) are
needed to cause an allergic skin reaction.  In addition, the 1989
ATSDR Toxicologic Profile for Chromium28 only calculates adverse
human exposure levels with regard to the inhalation of chromium
(IV) "and not chromium  (III.) (IV-D-03, IV-D-17).
     Response;  The EPA reviewed health data on chromium in a
1990 Federal Register  (55 FR 650)29 denying a petition to exempt
chrome antimony titanium buff rutile  (CATBR) from reporting
requirements under Section 313 of the Emergency Planning and
Community Right-to-Know Act of 1988.  The Federal Register notice
provided information on both carcinogenic and noncarcinogenic
health effects.  On carcinogenicity, the notice indicates that:
      "If CATBR, which  is a chromium compound, is retained in the
      lungs  and made available to the cells, then it is of
      carcinogenic concern by inhalation....The National

                               2-80

-------
     Toxicology Program (NTP) in their 1985 Fourth Annual Report
     concluded that "if any chromium compounds are carcinogenic,
     then all compounds containing chromium are potentially
     carcinogenic."...chromium (III) is believed to be the
     ultimate carcinogenic form of chromium (VI) after its entry
     into target cells and subsequent intracellular metabolic
     reduction."
For these reasons, chromium compounds as a group will continue to
be listed as high risk.
     The commenter also noted data on noncancer effects.  The
selection of chromium for inclusion as a high-risk pollutant,
however, was based on chromium as a carcinogen.
2.6.8.8  Comment;  One comment letter stated that listing
chloroprene as a high-risk pollutant is inappropriate because its
toxicological properties are well-defined and do not suggest that
the material is extremely hazardous.  According to the commenter,
the approximate lethal concentrations of chloroprene vapor for
rats and hamsters suggest only mild to moderate acute toxicity.
There is no evidence of mutagenicity in in vivo tests in rats at
concentrations up to 100 ppm.  Chloroprene does not induce
dominant lethal mutations in either rats or mice.
Epidemiological studies have demonstrated that chloroprene is not
carcinogenic to man.  A study of active chloroprene workers at  a
Du Pont Company plant provided no evidence of biochemical or
hematologic effects  (IV-D-18).
     Response:  In the final rule,  chloroprene has been deleted
from the high-risk list.  The focus of the testing referred to  by
the commenter was on the carcinogenicity and mutagenicity of
chloroprene, not  noncancer effects.  The EPA believes that any
determination of  the cancer  potential of chloroprene must be
contingent upon the  inhalation study now underway by the NTP.
However, a recently-completed NTP subchronic  study30 indicated •
damage  to nasal tissue.  Assuming that the RfC  Verification Work
Group  selects this study as  the  critical study  for RfC
development,  it  is still very unlikely that the resultant RfC
would warrant  listing  chloroprene as  high-risk.
                               2-81

-------
      The proposal of  chloroprene  as  high-risk was based pn
 results of a study that reported  effects on reproductive
 parameters at exceedingly low concentrations  (see Docket No. A-
 90-47,  Item No.  II-B-4).10  After further review, the EPA has
 judged  this study to  be flawed because  of the lack of detailed
 chemical information, contaminants,  and storage conditions.  This
 is  especially critical for this compound, as some evidence
 suggests that toxic materials are generated in chloroprene  stored.
 under oxidative  conditions and it may be these products that
 caused  the effects seen in the study.
 2.6.8.9  Comment;  One comment letter stated that the proposed
 dioxin  multiplier should be reevaluated in light of new research.
•According to the commenter, recent research suggests that dioxin
 poses far less (if any) risk to human health than previously
 thought (IV-D-35).
      Response;  The EPA's Office  of  Research and Development has
 begun a comprehensive new assessment of the toxicity of dioxin.
 This assessment  will  incorporate  information  from new
 epidemiologic investigations and  laboratory studies and will
 reflect the thinking  that is emerging from recent scientific
 discussions.  The assessment will include a quantitative
 dose-response analysis.  The commenter  is incorrect, however, in
 asserting that the new assessment is being done because dioxin
 "poses  far less   (if any) risk ... than  previously thought." This
 •statement is speculative and does not characterize the full body
 of data that is  being evaluated.   The EPA has  not yet reached any
 conclusions.  In the-meantime, the EPA's present characterization
 will be used.  It would be premature and contrary to the EPA's
 policy to change the dioxin multiplier  at this time.
 2.6.8.10  Comment;  One comment letter  claimed that  studies of
 smelter communities have failed to support the hypothesis  that
 arsenic in smelter emissions is responsible  for  any  higher
 incidence of lung cancer.  The commenter  also questioned the
 appropriateness  of the weighting factors  for arsenic  and cadmium.
 The commenter submitted supporting data (IV-D-38).
                                2-82

-------
     Response:  The commenter believes that the. epidemiology
studies of smelter communities by Greaves et al.31 and Rom
et al.32 raise a question as to the appropriateness of the
weighting factors for arsenic and cadmium.  The study by Rom et
al.,32 which was addressed in the EPA's 1984 "Health Assessment
Document on Arsenic",2^ found no association between cases of
lung cancer and distance from a nonferrous smelter.  The authors
were unable to control for such critical factors as smoking,
occupation, and migration, however, and were unable to obtain
environmental measurements over most of the years studied.  The
study by Greaves et al.3-'- also found no association between
having lung cancer and distance from a smelter.  Again, however,
the authors were unable to control for such factors as smoking,
occupation, and migration and had no environmental data to
evaluate exposure other than distance from the smelter.
     Even had Greaves et al.31 and Rom et al.32 addressed the
factors mentioned above, however, the lack of a community effect
would not necessarily be inconsistent with the current lung
cancer dose-response estimate for inhaled arsenic.  The EPA's
unit risk estimate of 4.3 X 10"3 [(ug/mS)"1] is a lifetime risk
estimate that .assumes a lifetime of exposure.  To adequately
evaluate this estimate in community studies would require
detailed knowledge of amount and length of exposure to the
individuals in the study.
     The lack of an  association between distance from the
smelters and having  lung cancer has no impact on the unit risk
estimate.  Neither study was designed to  evaluate the ri-sk
estimate, and both studies have limitations that make it
difficult to  evaluate, even in a qualitative manner, whether or
not  an  elevated  lung cancer risk exists in communities
surrounding  smelters.
     The designation of arsenic and cadmium as high-risk  and the
subsequent assignment weighting factors were based  on the
EPA-approved cancer  potency factors and consideration of  the
weight-of-evidence that these compounds cause  cancer.  The  EPA
                               2-83

-------
believes that this designation and the weighting factors are
appropriate.
2.6.8.11  Comment;  One commenter suggested that chemicals such
as phosgene and methyl isocyanate, which pose potential severe,
acute hazards should not be included because they are generally
not emitted over long durations, nor do they demonstrate a
potential for chronic health hazards from lifetime exposure.
They do, however, pose an acute hazard under the conditions of a
catastrophic accident or large release, and should be handled
under Section 112(r) (IV-D-44).
     Response;  Section 112(i)(5)(E) specifies that the
Administrator shall limit the use of offsetting reductions in
emissions of other HAP's from the source as counting toward the
90-percent reduction in such high-risk pollutants qualifying for
an alternative emissions limitation.  No provision is made in
this mandate for distinguishing between pollutants producing
effects from acute exposure and those producing effects from
chronic exposure.
2.6.8.12  Comment;  One commenter questioned the EPA's cancer
unit risk estimate for 1,3 butadiene.  The comments had a number
of aspects, including an inventory and analysis of new health
data available since the EPA's 1985 assessment, and
reconsideration of known issues or areas of uncertainty related
to butadiene.  Part of the comments also referenced ongoing
assessment and regulatory activity by the Occupational Safety and
Health Administration.  The commenter took issue with the use of '
the data from a mouse bioassay rather than data, from a rat
bioassay.  The commenter claims the rat is a more appropriate
model and that the EPA neglected to consider all available
metabolic and species differences when developing the unit risk
estimate.  Therefore, the commenter believes the unit risk
estimate for butadiene greatly overestimates the potential risk
to humans  (IV-D-43).
     Response;  There has been new  information  on butadiene since
1985 in several areas including short- and long-term animal test
                               2-84

-------
systems, evidence from human studies, and some information on
metabolic/pharmicokinetic considerations.  For these reasons, the
1985 EPA risk characterization (as reflected in the IRIS file) is
outdated relative to information now available.  This should not
be taken to mean, a priori, that the 1985 assessment findings are
inferior, but rather that the EPA's risk characterization
warrants updating.
     The public comments largely assert that the EPA's risk
characterization should be revised according to the perspective
of the commenter and with a different perspective for some
fundamental assumptions.  The specific merits of these assertions
and critique of the EPA's past evaluations cannot be addressed
until the EPA can initiate and complete it's own evaluation of
the newer data.  The completion of a new evaluation by the EPA is
12 to 18 months in the future according to current plans.
     The question of how the unit risk estimate might change
given new data or other considerations cannot be estimated.  The
question is not whether butadiene is carcinogenic in animals, it
is how to extrapolate the animal data to humans with appropriate
knowledge of species-specific considerations and a host of
dosimetry considerations.  If the human data base now provides a
more robust set of data for quantitative risk analysis, this
would also be a new factor relative to the 1985 assessment.  In
the meantime, however, the EPA's present characterization of 1,3
butadiene will be used.  Therefore,  1,3 butadiene remains on the
high-risk list.
2.6.8.13  Commentt  One commenter submitted extensive comments
and  supporting data regarding acrylic acid.  The commenter took
issue with the RfC that has been developed for acrylic acid  and
upon which the high-risk listing is  based  (IV-D-40).
     Response: The EPA does not believe  that the RfC  for acrylic
acid is in error.  Therefore, acrylic acid continues  to be listed
as a high-risk pollutant.  Responses to  the commenter's specific
concerns about the acrylic acid RfC  are  addressed  separately
below.
                               2-85

-------
2.6.8.14  Comment;  The IRIS RfC for acrylic acid is based on a
90-day inhalation study in mice reporting a "very slight" focal
degeneration of the nasal mucosa at an exposure concentration of
5 ppm.  Other researchers have reported that mammalian olfactory
epithelia completely regenerate if the basal cell layer remains
intact.  Thus, the slight focal degeneration is considered a
reversible lesion (IV-D-40).
     Response;  Regarding the discussion of the possible
reversibility of the nasal lesions due to acrylic acid exposure,
there is some evidence that nasal epithelial lesions are
reversible to some extent after cessation of the insult.
However, the RfC is designed to be protective for a lifetime
exposure.  There is no evidence that these lesions are reversible
with a continuing exposure, therefore the proposition that the
lesions are reversible following a suitable recovery period is
not relevant to the derivation of the RfC.  Additionally, in a
study of chronic exposure of rats to butyl acrylate, nasal
lesions similar to those found after acrylic acid exposure
persisted after a 6-month recovery period following a 2-year
exposure (Reininghaus et al. 1991, Fd. Chem. Toxicol.
29:329-339).33  In the Miller et al. (1985) study,34 cited by the
commenter, rats and mice exposed to ethyl acrylate for 6 months
followed by a 21-month recovery period showed nasal lesions.
These studies suggest that the lesion is not necessarily
reversible.
   •
2.6.8.15  Comment;  The EPA classified acrylic acid as a
high-risk air pollutant by rounding the RfC value in the IRIS
data base to one decimal place.  The actual RfC value as listed
in IRIS is 0.33 ug/m3, not 0.3 ug/m3 (IV-D-40).
     Response;  The commenter is in error in stating that the
actual RfC value listed in IRIS is 0.33 ug/m3  The RfC's are
generally presented to only one significant figure in recognition
of the interpretation and definition of the RfC as an estimate
with precision to an order-of-magnitude.  When data are available
that mitigate the need for a full uncertainty factor of 10 for a
                             '  2-86

-------
 particular area of uncertainty,  such as  for extrapolation from a
.LOAEL to a NOAEL,  a reduced uncertainty  factor is  applied.   The
.reduced uncertainty factor is typically  the geometric  half  of the
 normal factor of 10, which is the square root  of  10.   This  is
 represented as a 3 to avoid the  implication of precision beyond a
 single significant, figure.  Thus, the use of a factor  of 3  for
 two areas of uncertainty results in a composite uncertainty
 factor of 10.  The correct uncertainty factor  for  the  acrylic
 acid RfC derivation on IRIS is 1,000. The correct RfC is 0.3
 ug/m^.
 2.6.8.16  Comment;  The adverse  effect upon which  the  acrylic
 acid RfC is based is a very slight, localized, reversible sign of
 upper respiratory tract irritation.  A threshold-dependent effect
 such as this certainly does not  warrant  the excessive
 conservatism employed in this classification process  (IV-D-40).
      Response;  The nasal epithelial lesion on which the acrylic
 acid RfC is based is degeneration of the epithelial tissues of
 the nose, which occurred in 40 percent of the  exposed  animals at
 the LOAEL.  The incidence of the effect  increases  in an exposure-
 concentration-related manner and the severity  of  the effect
 increases at increasing concentrations.   These factors led the
 EPA RfD/RfC Work Group to the conclusion that  the effect is
 adverse.  The EPA agrees that the effect is of minimal severity,
 and this is reflected in a reduction in  the uncertainty factor
 •for extrapolation from a LOAEL to a NOAEL  (i.e.,  an uncertainty
 factor of 3 rather than 10 was used here).  Because a  NOAEL was
 not demonstrated, some uncertainty remains regarding the amount
 of reduction necessary to reach a level  that is not likely to
 produce the effect.  Therefore,  an uncertainty factor  is applied
 for extrapolation from a level producing an adverse effect to a
 presumed NOAEL.  The EPA does not believe that the derivation of
 the RfC for acrylic acid employs  "excessive conservatism."
 2.6.8.17  Comment;  Available chronic inhalation studies for
 methyl acrylate, ethyl acrylate,  and butyl acrylate should be
 considered as  supporting data.   Data from these chronic
                                2-87

-------
inhalation studies of acrylate esters, when combined with uptake
and hydrolysis data from the rat upper respiratory tract, are
directly applicable to the determination of potential effects
from chronic inhalation of acrylic acid (IV-D-40).
     Response:  The chronic studies on acrylate ester were not
reviewed during the development of the RfC for acrylic acid.  As
indicated by the commenter, the nasal lesions produced by acrylic
acid and by several acrylate and methacrylate esters are very
similar.  The studies on acrylate ester may therefore provide
relevant information on the toxicity of acrylic acid.  Acrylate
esters are deposited in the nose and metabolized to acrylic acid
and the corresponding alcohol.  Acrylic acid is likely to be the
toxic metabolite for methyl acrylate, ethyl acrylate, and butyl
acrylate.  Acrylic acid is deposited directly in the nasal
epithelium and does not require metabolic activation.  Because of
this difference in the mechanism of dose delivery, the acrylate
esters cannot be directly used to evaluate acrylic acid toxicity
quantitatively.
     The chronic studies of methyl acrylate and butyl acrylate
show adverse nasal effects in rats at 15 ppm, the lowest
concentration tested.  The chronic study of ethyl acrylate shows
effects in the nasal epithelium at 25 ppm in a high percentage of
the exposed animals, and no effect at 5 ppm.  The occurrence of
an effect at lower concentrations of acrylic acid than of the
acrylate esters could reasonably be expected to result from
differences in delivered dose resulting from differences in
solubility,-deposition, or distribution of the chemicals.
     The studies of the acrylate esters may provide information
on the extrapolation of subchronic to chronic data, as suggested
by the commenter.  Only subchronic data are available for acrylic
acid, and uncertainty exists as to the extent of progression of
the lesion  leading to increased incidence and/or  severity of the
lesion with a chronic or lifetime exposure.  Comparison  of the
subchronic  acrylic acid study directly with the chronic  studies
on acrylate esters, as suggested by the commenter, is not
                               2-88

-------
appropriate because the comparison is confounded by differences
in dosimetry.  However, a comparison of the subchronic and
chronic studies on .the acrylate esters might provide information
on the extent of the progression of the lesion that could be
extrapolated to acrylic acid because of the similarity of the
lesions.  The results of subchronic studies of methyl acrylate
and butyl acrylate done by BASF and on ethyl acrylate by the Dow
Chemical Company are not available to the EPA at this time.
2.6.8.18  Comment:  In chronic studies of acrylate esters, the
esters caused no injury to internal organs.  No tumors developed
in the nasal turbinates with any of the esters; thus, the lesions
cannot be considered preneoplastic.  Furthermore, there was no
evidence of a carcinogenic effect in any organ system or tissue
(IV-D-40).
     Response;  The RfC derivation employs an uncertainty factor
for lack of a two-generation reproductive toxicity study.  It is
the practice of the RfD/RfC Work Group to require reproductive
and developmental toxicity studies because functional effects may
occur in the absence of any tissue damage that would be seen in a
typical subchronic or  chronic toxicity study.  The absence of
either a developmental toxicity or a two-generation reproductive
toxicity study is an area of uncertainty that requires
application  of an uncertainty factor.  This uncertainty may be
reduced if a reproductive toxicity study were available on any of
the acrylate esters  showing reproductive toxicity endpoints to be
less  sensitive than  the respiratory tract effect.  At this time,
the EPA is unaware of  any two-generation reproductive toxicity
study on  acrylic  acid  or on methyl acrylate, ethyl acrylate, or
butyl acrylate.
2.6.8.19   Comment:   The  incorporation of data  from chronic
inhalation studies  for acrylate esters eliminates the need  for
the uncertainty  factor of  10  for extrapolation from  subchronic to
chronic studies.   Furthermore,  these studies complement the
 studies already  cited in the IRIS data base and should
                                2-89

-------
significantly increase the confidence in the inhalation data base
from "MEDIUM" to "SIGH" (IV-D-40).
     Response;  The commenter is in error in stating that the RfC
on IRIS uses an uncertainty factor of 10 for extrapolation for
subchronic to chronic exposure.  The IRIS text states that:  "A
factor of 10 is used for extrapolation from subchronic to chronic
and for limited reproductive studies."  As discussed above, the
data available to the EPA at this time are not sufficient to
reduce or eliminate these areas of concern.  However, data might
be available that would cause the EPA to re-assess the RfC for
acrylic acid.  Such data would have to be submitted through the
IRIS Submission Desk for consideration.
     The minimum data base for high confidence in the RfC data
base is two chronic inhalation bioassays in different mammalian
species, two developmental studies in different mammalian
species, and a two-generation reproductive study.  The confidence
in the acrylic acid data base is medium due to the lack of
reproductive studies and chronic studies.  Because of the
dosimetric difference described above, and the lack of
reproductive studies for any of the acrylate esters, the analogy
of acrylic acid with acrylate esters is not sufficient to raise
the confidence in the RfC.
2.6.8.20  Comment;  Two long-term acrylic acid drinking water
studies have been conducted in rats.  Decrements in drinking
water consumption and body weight gain were the only effects
noted.  When these studies are included in IRIS, the present oral
RfD value of 80 mg/kg/day should be increased by a factor of 10
because the extrapolation from subchronic to chronic studies will
no longer be applicable  (IV-D-40).
     Response;  The EPA agrees that the data submitted on oral
exposures to acrylic acid should be reviewed and incorporated
into the RfD file.  However, this review would not change the
classification of acrylic acid as high-risk because the
high-risk designation is based on the RfC.
                               2-90

-------
2.6.8.21  Comments  Extensive comments concerning vinylidene
chloride (VDC) were received from two commenters.  The
commenters' primary contention lies in the designation of VDC as
a carcinogen.  (Vinylidene chloride has previously been
designated as a Group C [possible human carcinogen] by the EPA.)
The letters included comments regarding metabolism and
pharmacokinetics and possible species variations, as well as
comments about the "large number of 'negative1 studies" compared
to the number of "positive" studies.   The commenters believe
that Group C carcinogens should be treated differently than Group
A or Group B carcinogens-similar to the treatment of Group C
carcinogens by the EPA's corrective action program for solid
waste management units.  The commenters maintain that by failing
to consider distinctions in cancer classification in the tier
analysis, the EPA acted contrary to its own guidance as set forth
in the "Guidelines for Carcinogen Risk Assessment"35 and the
information presented in the 1985 "Health Assessment Document for
VDC."36  The commenters also questioned the specific study by
Maltoni et al.,37 which is the basis for the unit risk estimate
for VDC  (IV-D-49, IV-D-55).
     Response;  In response to public comment, the EPA has
revised the weighting factor system in such a way that cancer
classification and not just cancer potency is given
consideration.  As a result, Group C carcinogens that were
proposed for  the  high-risk list have now been assigned a
weighting  factor  of 1.  This decision affects VDC and
1,1,2,2 tetrachloroethane.  The commenters1 concerns about the
cancer classification of VDC are addressed below.
     The EPA has  classified VDC as a Group C carcinogen based on
animal data.  This classification is based primarily on the
finding  of. kidney adenocarcinomas in male mice,  as well as other
considerations that contribute to placing VDC into Group C.
Although a number of  studies have been conducted in different
strains  and species that might provide information on  the
potential  carcinogenicity  of VDC, many of these  studies were not
                             •  2-91

-------
designed as cancer bioassays, or they were flawed in some ways
that lessened their usefulness in determining the carcinogenic
potential of the chemical.  Vinylidene chloride was administered
by different routes in different studies—by inhalation, orally
by gavage and in drinking water, and by skin application and
subcutaneous injection..                      .
     Unfortunately, most of these studies were not adequate for
proper and conclusive evaluation of cancer-causing activity
because of a number of factors, including less-than-lifetime or
otherwise inadequate exposure regimens, insufficient numbers of
animals used in the studies, and/or inadequate number of
dose/exposure levels.  Only some of the studies were actually
designed and conducted as cancer, bioassays.  Therefore, the fact
that increases in tumor incidence were not observed in most of
the studies referred to by the two commenters does not weigh
nearly as heavily in the overall weight-of-evidence determination
as their comments imply.
     There were positive findings in the studies conducted by
Maltoni et al., in spite of certain shortcomings.  Kidney
adenocarcinomas, a rare tumor in mice, were induced in male mice
exposed by inhalation.  This data set was the one used to
calculate an upper-bound risk estimate for VDC.  In addition to
the kidney adenocarcinomas, there were statistically significant
increases in mammary carcinomas observed in female mice and
pulmonary adenomas in mice of both sexes, although there was not
a dose-response relationship.  In a two-stage skin bioassay,
there was some evidence that the chemical acted as an initiator.
2.6.8.22  Comment:  One comment letter was received concerning
ethylene oxide (EO).  The commenter's contention is that the
EPA's 1985 Health Assessment Document  (HAD)  (EPA-600/8-84-009F)38
on EO is out of date because:   (1) it fails to consider the array
of new epidemiological data,  (2) the endpoints observed in animal
studies are not necessarily relevant to humans, and .(3) the
cancer potency factor used is dated and inadequate to represent
the potential risk to humans  (IV-D-61).
                               2-92

-------
     Response;  The commenter's submission cites four recent
epidemiology studies: Steenland et al.(1991);39 Greenberg et al.
(1990);40 Gardner et al. (1990);41 and Kisselbach et al. (1988)42
in  support of the commenter's claim.  These studies plus
additional studies by Hogstedt et al. (1986);43 Hogstedt
(1988);44 Bisanti et al. (1988)45 and two unpublished reports
(Stolley et al. [1988]46 and Hagmar et al. [1986]47) form the
basis of response.
     The Hogstedt (1986, 1988)43'44 studies showed significant
increases in mortality from both digestive system cancers and
leukemia.  This confirmed the results of their earlier paper
cited in the 1985 HAD.  Stolley et al. (1988)46 reported a
statistically significant increase in female breast cancer ambng
EO sterilant workers.  The cohort had limited latency and power.
A small increase in lymphatic and hematopoietic cancer was noted
also but the results were not significant.  Hagmar et al.
(1986)47 reported a statistically significant increase in the
incidence of malignant lymphoma and myelomatosis in male workers
exposed to EO at a chemical factory.
     The Steenland (199I)39, study of  18,254 workers in 14
U. S. plants is by far the largest of the four studies submitted
by the commenter.  Although there were no significant increases
in mortality from any cause in this cohort, there was an excess
of hematopoietic cancers in male workers, especially among
workers in the longest latency category.  This was not observed
in females.  The report concludes, and the commenter concurs
although with a different emphasis, that the results are
inconclusive with respect to human carcinogenicity at workplace
exposure levels.
      The Gardner  (1989)41 study reported  slight increases in
deaths  from leukemia (3 observed versus 2.09 expected) in men who
were  not exposed to  leukemogens other than  EO.  Additionally,
there was  an  increase  in deaths due  to non-Hodgkin's lymphoma  (4
observed versus  1.62 expected).  There were no  increases in
deaths  due to stomach  cancer  (5 observed  versus 5.95 expected).
                               2-93

-------
Overall, the authors report only a slight increase in cancer
mortality over the expected.  Although they state that
differences in exposure levels may explain the discrepancies with
the.literature results, they conclude that their findings do not
exclude the possibility that EO is a human carcinogen.
     The third study, Greenberg et al.(1990)40 was a
retrospective cohort study of 2,174 men who either produced or
used EO.  While there was increased mortality from leukemia and
pancreatic cancer, the authors attributed these to exposures from
the chlorohydrin department, where EO exposures were considered
low.
     The Kiesselbach (1988)42 study reported on the mortality of
2,658 subjects from eight different plants exposed to EO for 1
year  (no exposure level was given).  Of the 63 cancer deaths
reported, only 2 were from leukemia (2.35 expected), while deaths
from esophageal and stomach cancers were higher but not
significantly so..
     In response to the commenter's second point, the EPA's
guidelines for carcinogen risk assessment hold that site
concordance for tumors between animal and human studies adds to
the weight of evidence, but some known human carcinogens, e.g.,
arsenic, benzene, chromium, have not shown site concordance.
Therefore, lack of concordance should not detract from the weight
of  evidence.
      To  support the claim that the rat tumor response was an
inappropriate model for human risks, the commenter referenced the
work  of  Dr. Leon Goldberg  (1986),48 which contains his analysis
as  well  as those of Drs. Susan Austin and Robert Sielkin.  This
book,  in report form, was submitted by the commenter  to the EPA's
Science  Advisory Board in its 1984 review of the draft HAD.49
Therefore, this material has  already been reviewed and the
appropriate parts  incorporated into the June  1985 Final Report.
Additionally,  in  its  currant  submission, the commenter failed to
note  the existence or significance of a major  inhalation  study  in
mice  conducted by the National Toxicology Program (Toxicology and
                               2-94

-------
Carcinogenesis studies of EO, NTP TR 326, NIH #88-2582).49  The
results of this study showed clear evidence of carcinogenic
activity in B6C3F mice.  The EPA's use of total number of animals
with at least one EO-caused tumor site would lead to a revised
cancer potency factor very close to, if not higher, than the
current value.
     In summary, the commenter's conclusion that EO should not be
designated a high-risk pollutant is not supported by the health
data submitted.  While it is quite possible that new
pharmacokinetic data will aid in a reassessment, the submission
does not justify waiting,for those data to become available
before listing EO as high-risk.
2.6.8'.23  Commenti  The weighting factor for hydrazine should be
changed from 100 to 10 because the potency factor for hydrazine
is only a factor of 2 greater than the highest potency of other
chemicals with a weighting factor of 10.  Also, the commenter
feels the short half-life of hydrazine (a few minutes to a few
hours, depending on ambient concentrations of other reactive
species) is inconsistent with a weighting factor of 100.  The
EPA's exposure model did not account for the short half-life
(IV-D-42).
     Response:  In the final rule, the weighting factor for
hydrazine remains at 100.  The commenter is correct in stating
that the exposure model used does not take into account the
volatility of pollutants.  The half-life of hydrazine is greatly
influenced by the presence of ozone, the hydroxyl ion, and other
reactive species in the ambient air.  As a result, hydrazine is
likely to degrade faster in polluted areas where ozone and
hydroxyl ion  levels are higher.  Depending on these conditions,
it  is estimated that the half-life of hydrazine ranges up to
several hours.  The EPA believes this would be ample time for
hydrazine to  be present  in the ambient air at a distance 200
meters  from the emission point.  The commenter submitted
information on  the distances to the nearest residence from the
                               2-95

-------
largest hydrazine producers, which showed that residences were
further away than 200 meters. However, other types of facilities
that emit hydrazine may also participate in the Early Reductions
Program and could be closer to residences.  Therefore, no
adjustment to the weighting factor was made on this basis.
     The weighting factor of 100 for hydrazine is based on the
potency of hydrazine relative to the geometric mean potency of
the group of carcinogens not on the high-risk list.  By this
method of comparison, hydrazine is about 200 times more potent
than carcinogens not on the high-risk list.  Because the EPA had
previously decided that rather than establish unique weighting
factors for each pollutant it would be appropriate to combine
pollutants of similar relative toxicities in groups that differ
by an order of magnitude, hydrazine was assigned a weighting
factor of 100.
2.6.8.24  Comment;  One comment letter concerned the scientific
study used as the basis of the potency factor for hydrazine.  The
potency value for hydrazine was calculated on the basis of
results of the MacEwen et al.  (1981)50 study.  The commenter
maintains there are several problems with the.study, including
(1) the animals may-have been compromised by a disease state, (2)
the animals were probably exposed to higher concentrations of
hydrazine than the measured amounts,  (3) higher concentrations
led to an underestimate of the NOAEL, resulting in an
overestimate of the cancer potency, and  (4) the rat may not be an
appropriate model for humans  (IV-D-42).
     Response;  While the number of rats  showing an inflammatory
response increased somewhat at the lowest dose, tumors were not
observed in this group.  Moreover, most  of the incidences of
inflammatory response were  in  the trachea and larynx, while
tumors in the high dose animals were mostly in the nasal  region.
There was little increase in the numbers of animals  showing
inflammatory response  in the  0.25 and 1.0 ppm dose groups even
though the  latter group did show an increased number  of  animals
with tumors.  Finally, respiratory tract tumors were  seen in
                               2-96

-------
hamsters with little evidence for inflammation, suggesting that
inflammation was not necessary for the tumorigenic response.  For
these reasons/ the acute inflammatory response seen in the lowest
exposure group is not considered to be a significant factor
influencing responses at the two highest dose groups.
     The possibility that there were errors in the concentration
measurements cannot be considered at this time.  The results were
published in a peer-reviewed journal and are considered valid
unless there is definitive evidence to the contrary.  If an Air
Force reanalysis of this study indicates significant errors and
the degree of error can be determined, the EPA will then consider
adjusting cancer potency factors.  Also, any potential error in
dose and, therefore, the NOAEL does not affect the estimation of
cancer potency because the latter value is based on tumor
incidence and not noncancer endpoints.
     The data are inadequate to show that the rat is more
sensitive to inhaled hydrazine than other laboratory species or
humans.  Adenomatous polyps were seen in exposed hamsters, even
though this species is considered to be quite resistant to the
induction of respiratory tract tumors.  While mice gave only a
marginal response at 1 ppm, they were not exposed to 5 ppm, as
were rats and hamsters.  While the rat may be quite sensitive to
the irritating effects of hydrazine, adenomatous polyps developed
in the hamster without apparent inflammation.
2.6.8.25  Comment:  The cancer potency of hydrazine has been
overestimated by the inhalation route (IV-D-42).
     Response;  Although tumors have been induced by hydrazine at
a'variety of sites, the upper respiratory tract is the most
sensitive.  Potency by one route may be much different than
potency by another route because of pharmacokinetic factors.
Therefore, low potency by the oral route does not necessarily
mean low potency by the inhalation route.
2.6.8.26  Comment:  Epidemiology studies show that hydrazine is
not a human carcinogen  (IV-D-42).
                               2-97

-------
     Response;  The EPA has determined that the human
epidemiological data are inadequate to refute or demonstrate a
carcinogenic hazard.  Hydrazine has been classified as a Group B
carcinogen (probable human carcinogen) based on evidence of
animal carcinogenicity.
2.6.9  Miscellaneous                    .
2.6.9.1  Comment;  Commenter requested that five or so optional
spaces be included on the list for State agencies to add
compounds that are of high priority for reduction in their States
(IV-D-26).
     Response;  The authority to designate high-risk pollutants
has not been delegated to the States.  Section 112(i)(5)(E)
stipulates that the EPA Administrator shall, by rule, designate
high-risk pollutants under the Early Reductions Program.
2.6.9.2  Comment;  One letter urged the EPA to drop the inorganic
grouping concept for such pollutants as arsenic, asbestos,
beryllium, cadmium, chromium, and mercury, and develop an
alternative based on individual toxicity  (IV-D-51).
     Response;  At the current time, the EPA does not plan to
split the pollutant groups specified by Congress into individual
pollutants.  However, in Section 112(b)(3), individual pollutants
within pollutant groups will be considered for deletion from the
list of HAP's if a petitioner can make a  showing that the unique
chemical substance meets the deletion requirements of
Section 112  (b)(3)(C).  Exceptions to this are the compound
groups of polycyclic organic matter, mineral fibers, and coke
oven emissions.
2.6.9.3  Comment;   One commenter was  of the opinion that sources
should be encouraged to reduce the amount of any high-risk
pollutant released  (IV-D-25).
     Response;  The EPA agrees fully with the commenter.  Nothing
in this rule should be construed to  discourage control  of
high-risk pollutants.  The purpose of establishing the  high-risk
list was to  ensure  that in achieving the  required emission
reductions under  the Early Reductions Program, high-risk
                               2-98

-------
pollutants would be adequately controlled.
2.6.9.4  Comment;  When pollutants are added to the high-risk
pollutant list, a source should not be exempt from further
reducing those pollutants, but the source should be given up to 3
years to adjust its emissions to continue to qualify for an Early
Reductions extension or comply with Section 112(d) standards
(IV-D-16, IV-D-25).  Other commenters believe that changes to the
high-risk pollutant list should not affect the status of a
previously submitted commitment or application (IV-D-19, IV-D-30,
IV-D-34, IV-D-47).
     Response;  The EPA has decided that sources with an approved
enforceable commitment or an approved permit specifying an
alternative emission limit will not be affected when -a HAP is
either added to the list in Section 112(b) or is newly-designated
as a high-risk pollutant.  The EPA expects to update the list of
high-risk pollutants as the science requires.  It is conceivable
that a source would have to revise a post-reduction demonstration
with each revision of the list.  This requirement could add
significantly to the administrative burden for both the source
and the reviewing agency.  Given that the source will be in full
compliance with the emission standards under Section 112(d) at
the end of the 6-year extension, the EPA feels this additional
administrative burden is unnecessary.
2.6.9.5  Commentt  The proposed regulation does not provide a
means of addressing high-risk HAP's if the facility chooses to
use the weighted average approach for reducing gaseous and
particulate emissions  (IV-D-06).
     Response:  According to-Section 63.74(g), the applicant may
demonstrate that the total base year emissions adjusted for
high-risk pollutants have been reduced by at least 90  (95)
percent  or a  combined weighted percentage between 90  (95) percent
for sources emitting both gaseous and particulate HAP.  The
equation for  calculating the weighted percent  is  provided in
paragraph  63.74(f)  of  the regulation.  The same weighted percent
will be  used  for comparison with the total base year reductions
                               2-99

-------
and with the total base year adjusted for high-risk pollutants.
2.7  STATE AUTHORITY
2.7.1  Comment;  One commenter suggests that States be consulted
during early permit approvals and have the authority to require
greater than 90 (95) percent reduction to protect public health
even when the EPA is the reviewing Agency (IV-D-04).  Several
commenters suggested the EPA consult States during early review
and receive firm commitments from the States that the 90 (95)
percent reduction is acceptable and that new State air toxics
requirements will not burden facilities with additional
requirements after Early Reductions requirements have been met.
The EPA should also discourage States from requiring greater than
90 (95) percent reduction (IV-D-12, IV-D-19,' IV-D-30, IV-D-44).
Another commenter suggested that the EPA require each State to
make a determination on the percentage reduction that it will
accept, and establish a deadline for making this determination
(IV-D-3.8, IV-D-64).
     Response:  The CAA states in Section 112(i)(5)(A) that:
"Nothing in this paragraph shall preclude a State from requiring
reductions in excess of those specified in this subparagraph as a
condition of granting the extension..."  Although not
specifically stated, it is implied that the excess reductions can
only be required as a condition of the State granting the
extension.  The CAA, therefore, implies that when the State is
the permitting authority, the State may require greater than 90
(95) percent reduction as a condition for granting the extension.
As long as the EPA  (administered through the Region) remains the
permitting authority, the CAA specifies 90  (95) percent
reduction.
2.7.2  Comment;  The regulation itself should provide for
coordination with the States prior to the State having permitting
authority  (IV-D-22).
     Response;  The EPA intends to coordinate all efforts with
State  agencies.  The amount of involvement  for the  States may
vary depending on the interest and knowledge of the  State
                              2-100

-------
concerning a specific source, or the resources available to a
State for the Early Reductions Program.  The EPA is committed to
State participation, and will include States in the review and
approval of permits and enforceable commitments.  It is not
necessary to put implementation details such as these in the
regulation.                                  ,          .
2.7.3  Comment:  The commenter suggests that States receive all
copies of permit applications, not just base year emissions
evaluations.  States may have valuable information about the
source that can affect Early Reductions proposals (IV-D-52).
     Response:  The regulation does not specify any of the review
and disposition requirements of permit applications because these
requirements will be contained in Part 71 regulations.  The
regulation does state that the EPA Regional offices will closely
coordinate with States concerning all submittals prior to the
States being delegated permitting authority.
2.7.4  Comment;  State and local agencies need guidance and
training prior to submittal of applications (IV-D-06).
     Response;  The EPA has worked closely with State and local
agencies in the development of the Early Reductions regulations.
The EPA has conducted and will continue to conduct training to
assist State and local agencies whenever requested in the
implementation of the Early Reductions Program.  Through this
training and independent mailings, the EPA has provided State and
regional offices with a draft of the Enabling Document, which
will be updated according to the final regulation.  The EPA has
also compiled a question-and-answer document based on questions
raised by regulatory agencies and industry representatives.  In
accordance with the schedule of source categories, the EPA will
identify references to assist State and local agencies with the
review of calculated emissions.  The EPA will continue to provide
support and guidance as necessary to State and local agencies.
2.7.5  Comment;  A number of commenters suggests delegating
approval authority for the Early Reductions Program to the States
independently of approval for State permit programs under Title V
                              2-101

-------
(IV-D-02, IV-D-10, IV-D-14, IV-D-16, IV-D-20, IV-D-25, IV-D-26,
IV-D-28, IV-D-31, IV-D-46, IV-D-47, IV-D-57, IV-D-59).  One of
these commenters  (IV-D-46) also believes that language in the
regulation implies that States will have sign-off authority on
all extensions, regardless of whether States are delegated to
administer the Program.  Several other commenters suggest that
the EPA retain responsibility for approving enforceable
commitments and permit applications until the States have an
approved Program.  States already have an enormous burden in
developing an approvable Title V permit program (IV-D-08,
IV-D-19, IV-D-28, IV-D-32).
     Response:  The EPA specifically solicited comment on whether
this Program should be delegated to the States.  Based on the
comments received, the EPA is proceeding to establish the
criteria for delegating the Early Reductions Program to those
States that seek  delegation authority even in advance of a State
having a Title V  operating permit program.
     Section 112(1) of the CAA authorizes the Administrator to
approve a State program for the implementation and enforcement of
the emission standards and other requirements of the  section,
including partial delegation of the Program.  That is, the EPA
anticipates that  it will, if a State so desires, delegate to a
State the authority to develop and  implement the Early Reductions
Program.  The EPA anticipates that  it will publish delegation
guidance in the  near  future that will be useful to the States  in
developing programs for submittal.
     Of  course,  until such time as  a State  has an approved State
program under Title V, it cannot issue  a permit  [see  CAA
Section 112(1)(9)].   Therefore, the Administrator cannot delegate
his  authority to issue a  permit establishing an alternative
emission limit  under  Section  112(i)(5)  until, such time as a  State
has  an approved Title V permit program.
2.7.6   Comment;   Several  commenters requested  that  63.72(f)  be
deleted.   States should not be allowed  to  require greater than 90
 (95) percent reduction (IV-D-02,  IV-D-08,  IV-D-27).
                               2-102

-------
     Response;  Section 112(1)(5)(A) of the CAA Amendments
specifically allows States to require reductions greater than 90
(95) percent.  It is clear from the language of the CAA that
Congress intended States to have the power to promulgate and
enforce stricter standards.
2.7.7  Comment;  States should have the option of not allowing
facilities in their State to participate in the Early Reductions
Program (IV-D-14, IV-D-22).
     Response;  The Early Reductions Program is a national
Program mandated by Congress through the CAA and States do not
have the authority to disallow participation.  However, States do
have the power to require additional or stricter State standards
or require greater than 90 (95) percent reduction under their
Title V State permitting program.
2.8  INTERFACE WITH TITLE V PERMITS
2.8.1  Comment;  EPA should establish a policy for dealing with
applications under review during the time that States receive the
final approval of a State Title V permitting program.  Commenter
suggests that the reviewing agency remain the same throughout the
review and issuance of the permit (IV-D-13).
     Response;  The EPA generally agrees with the commenter that
Title V permit applications undergoing review by the EPA for an
Early Reductions demonstration and alternative emission limits
should not have to be reviewed twice—once by the EPA and then by
the State.  In most cases, this should not occur.  However, there
may be some situations where a State has received approval of its
Title V permitting program during the time that the EPA is
reviewing a permit application submitted by a source to the EPA
for an Early Reductions demonstration.  The EPA will address this
situation in an upcoming Federal Register notice that will
describe a specialty permit rule for the Early Reductions
Program.
     Briefly, the EPA does not believe that permit issuance would
be  substantially delayed as a result of the transition to State
responsibility for permitting.  Owners or operators must know in
                              2-103

-------
a timely fashion whether their sources will be granted a
compliance extension.  If an extension denial occurs, the source
will have to meet the applicable Section 112(d) standard on
schedule, and if substantial delays occur, it will mean less time
for the source to meet the 112(d) standard.
     The anticipated Early Reductions specialty permit rule will
address when the EPA will retain permit issuance even after a
State receives approval for its comprehensive Title V permit
program, as well as when such review and specialty permit
issuance will be transferred to the State after submittal to the
EPA by the source owner or operator.  It is envisioned that in
the event of such transfer, the State will adhere as closely as
possible'to the original review timetable established by the EPA.
2.8.2  Comment;  Most applications for early reductions will be
submitted before State programs are approved.  Commenters would
like language added to the regulation and/or assurance that any
Early Reductions application that is approved by the EPA (versus
the State) will be included in any future operating permit
without  further negotiation  (IV-D-02, IV-D-38, IV-F-la).
     Response:  Once the EPA has issued a Title V operating
permit providing for alternative emission limitations for Early
Reductions  sources, they remain in effect for the duration of the
6-year compliance extension.  Because permits must be renewed
after 5  years, the State would be the reviewing agency  at that
point.   However, the alternative emission limitations,  once
granted,  continue to be permit conditions until the  compliance
extension expires, in which  case applicable  Section  112(d)
standards then take  effect.
2.8.3  Comment;  Permits  for the Early Reductions Program should
not be revoked for any  reason prior  to the  end of the  6-year
extension,  even  pursuant  to  Section  112(f)  of  the CAA  (IV-D-29).
      Response;   Permits issued  during the 6-year  compliance
 extension may be revoked only in those circumstances that warrant
permit revocation under the Title V permit regulations.  If  the
compliance extension-were revoked,  then  the source would have to
                               2-104

-------
achieve timely compliance with any applicable Section 112
standard.  Section  112(i)(5) does not shield a source from
standards issued under Section 112(f).                .
.2.8.4  Comment:  Sources covered by Section 112(d) standards
prior to 1994 will  be discouraged from participating in the
Program because proposed permit rules will not be completed..  The
commenter based this comment on the statement in the preamble:
"Sources subject to early 112(d) standards but which nonetheless
have achieved qualifying reductions before proposal, are not
required to  submit  enforceable commitments to participate in the
Early Reductions Program. " Such sources need only file a permit
application  demonstrating that qualifying reductions have been
achieved."  (56 FR 27356)51  (IV-D-36).
     Response;  The Early Reductions Program provides the
"enforceable commitment" vehicle to allow participation by
sources subject to  the Section 112(d) standards expected to be
proposed by  January 1, 1994.  Sources that submit an enforceable
commitment may submit a permit application as late as December 1,
1993.  Emission reductions do not have to be achieved until
January 1,  1994.  Sources unable to achieve the required
reductions may rescind their commitment by December 1, 1993
without penalty.
     The statement  quoted by the commenter was not intended to
discourage  sources  from submitting an enforceable commitment but
was  intended to clarify that sources have the option of achieving
emission reductions and submitting a permit application according
to the deadlines  imposed on sources covered by Section 112(d)
standards proposed  after January  1, 1994.  However, the
enforceable commitment option allows sources additional time to
achieve reductions, additional time to submit the permit
application, and  the  option of rescinding the commitment without
penalty.  The enforceable commitment mechanism, along with the
permitting  program, will give facilities maximum opportunity to
participate in the  Program.
                               2-105

-------
2.8.5  Comment;  The public should only comment during the
Title V permit issuance process to reduce the time burden and
simplify the process (IV-D-27).
     Response;  Acceptance of base year emissions is essential to
participants in the Early Reductions Program.  If base year
emissions are accepted by both the EPA and the public, facilities
will know how to proceed in achieving the actual reductions.  In
many cases, facilities will need to devote resources to
installing control equipment.  Therefore, industry will have the
benefit of a complete base year review prior to making pollution
control expenditures and prior to submission of the permit
application.  Any necessary changes or adjustments can be made
following base year review, therefore facilitating participation
in the Program.
     Submittal of base year emissions for early review is
optional where the reductions are made before proposal.  Any
source that believes that this submittal is burdensome may simply
submit a permit application according to the guidelines in the
regulation.   It is, however, in industry's best interest to
submit base year emissions for early review and to have the
public review the emissions.
     It should be noted that acceptance of base year  emissions
does not provide an absolute shield against changes required
during subsequent permit review.  However, these  changes should
be minimal.
2.8.6  Comment;  Items  (b) and (c) in Section  63.77 need to be
clarified.   Does the  "additional  information"  refer only to a
defined source (IV-D-52)?
     Response;  All permit applications will require  the
information outlined  in  Section  63.74 of  the regulation.   The
 "additional information" wording in the  regulation refers  to the
information required  specifically by the  permitting agency over
and above that needed to demonstrate early reductions.   The
information required  to  complete a  permit application will be
                               2-106

-------
specified in the Part 70 and Part 71 permitting regulations (when
issued).
2.8.7  Comment;  Commenter believes that the "Alternative
Emissions Limitation" permit should be a separate permit that is
not included on regular Title V operating permits because of
conflicting requirements of the two programs.  If the permits
were separate, the alternative emission limit permit could be
written for a 6_-year period, thereby eliminating the unnecessary
review after 5 years (IV-D-13).
     Response;  The CAA specifically requires that an Early
Reductions alternative emission limitation be established by
permit under Title V.  Section 112(i)(5)(D) specifies that this
permit be issued under Title V by stating: "For each source
granted an alternative emission limitation under this paragraph
there shall be established by permit issued pursuant to Title V
an enforceable emission limitation..."
2.8.8  Comment;  Using a Title V permit for a source that is less
than an entire facility is inappropriate.  A separate means of
tracking legally enforceable commitments and reductions is
necessary.  The reductions may then be worked into the plantwide
Title V permit (IV-D-26).
     Response;  Title V permits apply to an entire facility.  All
parts of the permit, however, do not have to be issued at the
same time.  The Early .Reductions regulation specifies that a
permit will be issued upon approval of the application.  Although
the permit for the alternative emission limitation will be part
of the Title V permit for the entire facility, each part may be
issued as it is approved.
2.8.9  Comment;  Fees should be collected for the Early
Reductions Program or the Title V permit program but not both.
Language in the proposed rule should be changed to state that the
review fees would be covered under the Title V fee program
 (IV-D-30).  Another commenter also suggests that it would be
unfair to collect permit fees for pollutants covered by two
permitting programs.  Because companies not participating in the
                              2-107

-------
Early Reductions Program and complying with Section 112(d) will
only pay one permitting fee, companies participating in the Early
Reductions Program should also pay only one permit fee (IV-D-45).
     Response;  The CAA under Section 112(i)(5) does not
establish a separate fee collection system for the Early
Reductions Program.  However, the CAA states that the Early
Reductions Program will be administered by the Title V program
and according to Section 502(B) of the CAA, the permitting
authority will collect fees under the Title V program.
Therefore, fees will be collected but only as prescribed by the
Title V program.  Applicants who submit enforceable commitments
will not be subject to fees until the source applies for a
permit.                                                 , .
2.8.10  Comment;  One commenter suggests eliminating the base
year review for enforceable commitments but requiring submittal
of permit applications by January 1, 1993 so that permits may be
issued by January 1, 1994.  The base year review is truncated and
will inappropriately stress the capabilities of the EPA and State
personnel.  A permit is required to enforce the reduction
achieved by January 1, 1994  (IV-D-39).
     Response;  Base year review will expedite the permit review
and assist industry in making  "acceptable" reductions.  Early
review of base year emissions  is essential before a source
commits the resources to reduce emissions.  This review will
provide assurance of acceptable base year prior to major
exp enditure s.
2.8.11  Comment;  The EPA must provide consistency between the
operating permit program and the Early Reductions Program.  The
EPA needs to  give  applicants at least  30 days  to correct
deficiencies  with  Early  Reductions  applications  (IV-D-45,
IV-D-52).                           .
      Response;   The Early  Reductions  Program will be  administered
by the Title  V permitting  program.  Therefore, Early  Reductions
permit applications are  for a  Title V permit and  the  same
deadlines and requirements, including the  deadlines  for
                               2-108

-------
 correcting deficiencies, apply.  The Early Reductions regulations
 do not specify these deadlines because these will be specified in
 the final permitting regulations (i.e., Part 70 and Part 71).
= Enforceable commitments are given 90 days to correct
 deficiencies.
 2.8.12  Comment;  The commenter is concerned about the provision
 that a compliance extension be in the form of a Title V permit.
 The regulations for operating permits will not be finalized until
 the first set of Section 112(d) standards are proposed and it
 would be too late to apply for the compliance extension for all
 sources covered by the proposed Section 112(d) standards.  The
 commenter suggests allowing entry into the compliance extension
 program within 3 months after issuance of the final rule on
 operating permits (IV-D-48).
      Response:  The regulation allows sources covered by
 standards proposed prior to January 1, 1994 to submit an
 enforceable commitment prior to proposal of the standard.  The
 source is then allowed until January I, 1994 to achieve the
 reductions and may submit a permit application as late as
 December 1, 1993.  Sources that have already achieved the
 required reduction prior to proposal of an applicable
 Section 112(d) standard must submit a permit application
 containing the reduction demonstration prior to such proposal; or
 if a Federal or State permitting program is not yet in place, the
 permit application for the Early Reductions source may be
 submitted up to 120 days after promulgation of Federal permitting
 regulations or  120 days after approval of a State permitting
 program under Title V, whichever comes first.  In the latter
 situation, even though the permit application would have to be
 submitted after proposal, the source owner or operator must still
 document that the required early reductions were achieved prior
 to proposal of the applicable Section  112(d) standard.  In the
 event that such documentation cannot be provided, the source's
 Early Reductions demonstration would be disallowed and the source
 would have to meet the Section 112(d)  standard.
                               2-109

-------
     It has become apparent recently that no comprehensive
Title V permitting mechanism, either a federal rule for issuing
Title V permits or an approved State program, will be available
in time to process permit applications for enforceable
commitments that are due by December 1, 1993.  In addition, it
may be some time before any source achieving reductions prior to
proposal of an early Section 112(d) standard could submit a
permit application.  In both cases the owner or operator may not
know if he would be granted a compliance extension in time enough
to meet the Section 112(d) standard if their Early Reductions
demonstration were disapproved.  Therefore, the EPA is
considering options to deal with this developing problem.
     One option is for the EPA to promulgate quickly a rule for
interim federal issuance of Title V specialty permits.  The rule
would apply only to Early Reductions permit applications that
could not be processed due to lack of an applicable comprehensive.
Title V program.  Permits issued under the specialty program
would encompass only the Early Reductions source at a facility
and only the hazardous air pollutant emissions from that source.
All other Clean Air Act requirements applicable to the Early
Reductions source would be handled through the comprehensive
Title V permitting process as it becomes available.  A future
Federal Register notice will describe EPA's proposed action on
this matter.
2.8.13  Comment;  There may be a gap between the time that an
application  is approved and the time a Title V permit is issued.
Reductions made under the Early Reductions Program should be
considered federally enforceable immediately upon approval of the
application,  regardless of whether a Title V permit program
exists  (IV-D-27).
      Response;  Permits do not have separate approval and
issuance  dates.  When the draft permit is  signed by the
appropriate  officials  for approval,  the permit is considered
valid at  that time.   Therefore, the date of  approval  and date  of
issuance  are the  same and there is no time gap.
                               2-110

-------
 2.9  INTERFACE WITH SECTION 112(g)  MODIFICATIONS
 2.9.1  Comment:  In addition to allowing excess offsets under
 Section 173(a)(1)(A) to be creditable for the Early Reductions
 Program,  the coramenter would like the EPA to address "offset"
 provisions under Section 112(g) of the CAA (IV-D-24).
      Response;  The EPA recognizes that the availability of Early
 Reductions credits for use as offsets for emission increases
 under Section 112(g) is an important issue.  The EPA is
 developing regulations to implement Section 112(g)  of the CAA.
 The issue of offsetting emission increases with Early Reductions
 credits has not been resolved and will be addressed in the
 upcoming regulations governing Section 112(g)  modifications.
 However,  the EPA recognizes that there are disadvantages in
 allowing Early Reductions credits to be used to offset
 Section 112(g) emission increases.
      Section 112(i)(5) was designed to bring about a substantial
 reduction in air toxic emissions from sources prior to the actual
 issuance of standards governing such sources.   In exchange, those
 sources would receive a limited compliance extension from the
 emission limitations established pursuant to Section 112 of the
 CAA.  Section 112(g) is a mechanism to allow owners or operators
 of facilities to make'physical or operational changes to their
 facilities without triggering MACT requirements.  A "major
 source" is considered not to have undergone a modification if any
 increase (greater than a de minimis increase)  as a result of
 physical or operational changes is offset "by an equal or greater
• decrease in the quantity of emissions of another hazardous air
 pollutant (or pollutants) from such source which is deemed more
 hazardous..."  The EPA believes that the intent of
 Section 112(i)(5) would be undermined by allowing the Early
 Reductions credits to be used under Section 112(g).
      This is best illustrated by example.  If Company A submits
 an enforceable commitment to reduce its emissions by 90 percent
 at one-half of its facility (and that portion of the facility
 meets the definition of source in Section 63.73) it will achieve
                               2-111

-------
those reductions (for example, from 100 tpy to 10 tpy)  in
exchange for an alternative emission limit for the first 6 years
beyond the otherwise applicable compliance date.  Later, the
company wants to undertake a physical change at the other half of
the facility that would result in an increase in emissions of 90
tpy of a HAP.  If, under Section 112(g), the owner or operator is
allowed to use as an offset credit those 90 tpy of reductions
from the Early Reductions Program, the major source overall will
emit the exact same level of HAP's as it did prior to making
early reductions (i.e., 100 tpy); Section 112(d) requirements
will be avoided on one-half the facility, and the other half of
the facility will have received a 6-year extension from an
otherwise applicable Section 112(d) standard.
2.10 INTERFACE WITH TITLE I PROVISIONS
2.10.1  Comment;  The commenters believe that emission reductions
under the Early Reductions Program should be able to be used as
credits for offsets and netting.  The Program is voluntary and,
therefore, the emission reductions are  not  "required"  (IV-D-27,
IV-D-45, IV-F-1).
     Response;  Most of the HAP's emitted by sources
participating in the Early Reductions Program are either VOC or
particulate matter  and, therefore, subject  to other requirements
under the CAA.  For such  situations, questions  have surfaced over
the  interaction between the Early Reductions Program and the
other CAA requirements.   Specifically,  a frequent question  from
interested  companies concerns whether  reductions made  under the
Early Reductions  Program  can  be considered  creditable  reductions
for  purposes  of New Source  Review (NSR) permitting under Parts C
and  D  of Title  I  of the CAA.   The preamble  to  the proposed  Early
Reductions  rule  described the way early reductions would be
treated in situations  involving offsets under  Part D NSR
 requirements but did not  discuss  "netting"  transactions.  The
 promulgation preamble reiterates  the proposal  preamble discussion
 regarding use of early reductions as offsets and adds  the EPA's
 policy regarding netting transactions;
                               2-112

-------
a. Offsets.  Emission reductions of HAP's for the purpose of
obtaining an alternative emission limitation under
Section 112(i)(5) of the CAA are not creditable for the purpose
of meeting an offset requirement under Section 173(a)(1) of the
CAA.  A source in a nonattainment area (an area where a national
ambient air quality standard is exceeded) or an ozone transport
region may need to obtain offsets for emission increases from
planned new construction or modification of existing facilities.
The HAP reductions are not allowed as offsets in this instance
because Section 173(c)(2) of the CAA states: "Emission reductions
otherwise required by this Act shall not be creditable as
emissions reductions for purposes of any such offset
requirement."  A source successfully participating in the Early
Reductions Program will be granted an alternative emissions
limitation (for the duration of the compliance extension period)
in lieu of a Section 112(d) emission standard.  Therefore, the
reduction of HAP emissions under the Early Reductions Program is
a substitute for the reduction of HAP emissions that would
otherwise be required of the source under Section 112(d).
     However, a source owner or operator may use as offsets any
reductions in HAP emissions in excess of those required to
qualify for an extension under the Early Reductions Program or
reductions in nonHAP emissions that are coincidentally obtained
through use of the HAP reduction measures, if such reductions are
not required by any other provision of the CAA and meet any other
requirements for offsets under NSR rules.  These reductions are
allowed as offsets pursuant to Section 173(c)(2) of the CAA,
which further states: "Incidental emission reductions which are
not otherwise required by the CAA shall be creditable as emission
reductions for such purposes...."
b. Netting.  In general, an owner or operator considering a
physical or operational change at a major  stationary source (as
defined in the NSR rules) will be subject  to  (1) the requirements
of Section 173(a)  [e.g., offsets, application of LAER] in
nonattainment areas or ozone transport regions or  (2) the
                              2-113

-------
requirements for Prevention of Significant Deterioration (PSD)
[e.g., application of BACT] in attainment or unclassifiable
areas, unless the changes will not cause a "significant net
emissions increase" in pollutants subject to NSR.  To determine
the net emissions increase for NSR purposes, the owner or
operator is allowed to sum the emissions increase from the
proposed change with any creditable increases and decreases
elsewhere at the plant.
     The NSR rules and the EPA's "Emissions Trading Policy
Statement"  (ETPS)  (51 FR 43823? December 4, 1986)52 limit the
creditability of some decreases in emissions for this "netting"
procedure.  For example, the NSR rules for nonattainment areas
state that  a decrease in emissions is creditable only to the
extent that "...the reviewing authority has not relied on it in
issuing any permit under regulations approved pursuant to 40 CFR
Part 51 Subpart I or the State has not relied on it in
demonstrating attainment or reasonable further progress;..."  [40
CFR 51.165(a)(1)(vi)(E)].  The PSD rules contain similar
language.   Essentially what this restriction does is prevent
sources from obtaining two credits for one reduction, where the
credits are related to the same air quality objective  (which  in
this case  is the attainment/maintenance of NAAQS).  Thus, as  an
example, a source  cannot use an emissions reduction to meet
reasonable further progress requirements and as  a reduction
credit in  netting  calculations.
     However, under  the ETPS, HAP decreases credited under the
Early Reductions Program also may be credited for purposes of
determining -the  net  emissions increase for  a plant change
proposed at a later  time for NSR purposes,  provided of course
that the reduced HAP's  also are pollutants  subject to  the NSR
rules  and  that the decreases meet all other requirements for
netting.   In  such  situations, the HAP decreases  produce benefits
for two different  air quality objectives  and one credit can be
given toward each; the HAP credit is associated  with the air
toxics reduction objectives  of  Section  112  of the CAA  and the NSR
                               2-114

-------
credit is associated with the attainment/maintenance of NAAQS and
PSD.  The amount of the HAP reduction creditable in these
situations will be limited if the netting calculations involve
HAP emissions increases.  Specifically, the creditable HAP
reductions from the Early Reductions Program will be reduced by
the amount of any increase in HAP emissions involved in the
netting calculations.  If no HAP increases are involved, the
entire HAP reduction is creditable.
     The principle behind this policy limitation is similar to
the one behind the netting restriction in the NSR rule mentioned
above, namely that a reduction should not receive two benefits or
credits (double counting) for the same air pollution control
objective.  The objective of the Early Reductions Program under
Section 112 of the CAA is to achieve significant reductions of
HAP's at existing facilities.  Sources that achieve such HAP
reductions in accordance with the rules promulgated today receive
credit for the reductions in the form of a 6-year compliance
extension for applicable Section 112(d) standards.  If the
reductions also were allowed to be used as netting reduction
credits for physical or operational changes involving increases
in HAP's, then the reductions in effect would be promoting HAP
increases elsewhere at the plant site by helping such facilities
net out of NSR control requirements.  Under such a scenario, an
owner or operator could receive a 6-year compliance extension to
Section 112(d) standards for some portion of the plant, net out
of NSR control requirements, and have overall HAP emissions equal
to preexisting levels.  Clearly this is not a result consistent
with the objectives of the CAA.
     To illustrate the effect of this policy, consider a plant
site in which a portion of the facility (e.g., a process unit)
participates in the Early Reductions Program and achieves a 50
tpy reduction of particulate HAP's.  Later, the owner or operator
proposes a physical or operational change at another section of
the plant that would increase particulate emissions by 75 tpy,
but none of the emissions increase would be HAP's.  In this case,
                              2-115

-------
the owner or.operator can use all of the HAP reductions from the
Early Reductions Program to net against the particulate emissions
increase because -no HAP increases are involved.  However, if 30
tpy of the proposed 75 ton increase are HAP's, then only 20 tpy
of the HAP reductions under the Early Reductions Program could be
used as reduction credits in any netting calculations (20 tpy is
the amount by which the HAP reduction exceeds the proposed HAP
increase).  Finally, if 50 tpy or more of the proposed 75 ton
increase would be HAP emissions, then none of the HAP reductions
from the Early Reductions Program could be used to net against
the particulate emissions increase  (because the HAP increase from
the proposed  modification is equal to or greater than the HAP
reductions from the Early Reductions Program).
     It should be noted that this netting policy for HAP
reductions is applicable only for NSR programs.  Under
Section 112(g) of the CAAf the EPA must promulgate separate
requirements for modification of HAP sources.  The provisions to
implement Section 112(g) are under development but will not
become effective in a State until the State has obtained approval
of a Title V permitting program.
2.10.2  Comment;  The definition of contiguous facility should
not mirror the PSD program, where emission points under common
ownership within 25 miles have been considered part of the
contiguous facility  (IV-D-47).
     Response;  The Early Reductions Program  differs from the PSD
program in that  a source must be part of a contiguous area under
common ownership or control.  The PSD program considers both
contiguous areas and  adjacent areas in  the source definition.
Therefore, the Early  Reductions Program is more limited when
attempting to combine emission points in the  source definition.
2.10.3  Comment;  Sources that achieve  a 90-percent reduction
from base year  emissions  should be  granted the  same 6-year
extension from complying with reasonably available control
technology  (RACT) requirements under Title  I  as received from
Section 112(d) -standards  under .Title III  (IV-D-45).
                               2-116

-------
     Response:  Section 112(i)(5) of the CAA authorizes the
Administrator (or a State operating pursuant to an authorized  .
Title V permit program) to grant a 6-year extension from
compliance with an emission standard promulgated under
Section 112(d) that would otherwise be applicable to that source,
provided the source satisfies certain conditions.  Nothing in
Section 112(i)(5) or any other provision of the CAA authorizes
the Administrator to grant extension of any RACT requirement that
might be required under other sections of the CAA.  As a
practical matter, however, the EPA anticipates that many of the
methods or technologies adopted to reduce a source's HAP by 90
(95) percent for purposes of obtaining a compliance extension
from a Section 112(d) standard may also constitute compliance
with RACT requirements under other provisions of the CAA.
     Although sources successfully participating in the Early
Reductions Program will be granted an alternative emission
limitation in lieu of meeting an application Section 112(d)
standard, the source still is responsible for complying with
other applicable requirements of the CAA.  For example, sources
located in areas designated nonattainment for the criteria
pollutant ozone are subject to other emission reduction
requirements.  State implementation plans (SIP's) for these
nonattainment areas (plans for attaining the criteria pollutant
ambient air quality standards) must contain requirements for
application of RACT requirements for stationary sources of VOC.
The RACT is required for sources (1) for which the EPA has
published—or will publish—a control techniques guideline
document, and (2) that are "major"', as defined in the CAA.
     State implementation plans also must provide for other
emission reductions sufficient to demonstrate attainment by
specified deadlines and to meet interim reasonable further
progress requirements.  States may obtain reductions from any VOC
emission source in the inventory, which means that they may
eventually require additional emission reductions from sources
participating in the Early Reductions Program.  (It should be
                              2-117

-------
noted that VOC emission reductions resulting from compliance with
HAP rules, including those under the Early Reductions Program,
are creditable toward the Program requirements to the extent that
they were not required prior to enactment of the CAA Amendments
of 1990 and are consistent with creditability requirements under
the CAA.)
     The interaction between the Early Reductions Program and
other requirements for the attainment of NAAQS_ causes concern in
that the prospect of later application of additional requirements
to sources that make early reductions would effectively limit the
attractiveness of, and therefore participation in, the Program.
Therefore, with respect to the percentage reduction requirements,
reasonable further progress, and attainment demonstration
requirements, the EPA has established a policy to reduce the
amount  of uncertainty for sources that choose to participate in
the Early Reductions Program.  This guidance should further
encourage possible applicants to participate in the Early
Reductions Program.  Briefly, the policy regarding RACT and
reasonable further progress  requirements for sources making early
reductions states that:
      •    The source must meet any applicable  existing RACT
          requirements  (including RACT that is required but has
          not been adopted by the State, i.e., RACT  fixups);
      •    The EPA encourages States to favorably  consider  the
          reductions made under  the Early  Reductions Program  in
          their analysis of  any  future RACT requirements  for  the
           source;  and
      •     The EPA encourages States to  seek any  additional
           reductions (beyond those  under the  Early Reductions
           Program)  needed to make reasonable  further progress
           first from sources not participating in the Early
           Reductions Program.
      The intent of the policy is to require compliance with
 existing RACT requirements,  but also potentially give sources
 some breathing room with respect to future RACT or  reasonable
                               2-118

-------
further progress requirements, in recognition of the fact that
substantial (90 [95] percent) reductions already have been made
at the source.  This policy is appropriate because, in most
cases, a 90 (95) percent reduction of HAP's from the sources will
produce comparable reductions in criteria pollutants or criteria
pollutant precursors (i.e., VOC).  For example, nearly all HAP's
emitted as gases also are VOC.
     However, it is possible that an Early Reductions
demonstration will not produce a criteria pollutant reduction.  A
creditable HAP emission reduction may be achieved by substituting
a nonHAP compound for the HAP.  If, for example, both compounds
are VOC, then there is no VOC reduction, although the HAP has
been eliminated.  In such cases, it would not be appropriate to
apply this policy.
2.11  TEST METHODS AND PROCEDURES
2.11.1.  General
2.11.1.1  Comment;  The EPA should establish de minimis
concentrations for all HAP's for Method 301.  This would be the
lowest concentration for which the test methods would have to be
validated (IV-D-21, IV-D-31, IV-D-50).
     Response:  It is beyond the scope of Method 301 to specify
de minimis levels.  Method 301 does provide specific procedures
for the determination of the limit of quantitation.  In addition,
Method 301 requires that the data and the proposed method be
validated for the concentration(s) of HAP's in the waste
stream(s).
2.11.1.2  Comment
     Method 301 should contain a procedure for calculating the
concentration range for which a method would be considered
validated (IV-D-21).
     Response;  Method 301 advocates that the proposed test
methodology be validated over the range of expected HAP
concentrations either in the field or through laboratory testing
utilizing the ruggedness test procedures.  In addition, a
procedure is offered for determining the limit of quantitation.
                              2-119

-------
Such procedures would  establish the concentration range.
2.11.1.3   Comment;  Method  301 does not specify under which
process conditions the validation testing should be conducted
 (IV-D-33).
     Response;  The choice  of operating conditions depends upon
the end use(s) of the  data. Considerations in structuring the
validation test should include;  sample matrix complexity in
regard to future  applicability of the  proposed method,  sample
analyte concentration(s)  in the  sample waste  stream and of the
applicable emission  standard, and representativeness of the
process conditions in  regard to  establishing  actual annual
 emissions.
 2.11.1.4   Comment;   If applicants choose to use the weighted
 average approach  to  determine the required percent reduction,
 they  should be required to state the temperature  and pressure
 ranges used to determine the mass emission rates  (because gaseous
 emissions may condense into liquid  droplets under atmospheric
 conditions) (IV-D-59).
      Response;  The EPA agrees  that temperature and pressure
 information is necessary to determine  the  emission rate.
 Expressing emissions at standard conditions  is not the same  as
 changing the sampling conditions.
 2.11.1.5  Comment;   The final rule should include a  definition  of
. "particulate" (IV-D-50).
      Response;  The definition of particulate matter is dependent
 upon the  test method  involved.   The requirements  of  the
 applicable regulation, SIP, or Federal regulation should be
 reviewed  first for applicable definition(s).   The problem exists
 for emissions that can be  in either a particulate or gaseous
 phase.  If there is a question, the applicant should consult with
 the EPA to arrive at  a resolution.
 2.11.1.6  Comment;  Method 25A should be accepted for the
 determination of base year and post-reduction emissions.  The
 commenter suggests a  circumstance that would allow for  such use
 of a  total hydrocarbon analysis on a  continuous basis;  when the
                               2-120

-------
gas stream is principally comprised of the HAP of, concern and the
HAP contribution to the total hydrocarbon concentration can be
documented.  The commenter also points out that continuous
measurements would be more appropriate for batch processes
(IV-D-33).
     Response:  The EPA believes that application of Method 25A
to a single HAP in a sample matrix containing no other
hydrocarbons would .prove acceptable, assuming the_detector is
calibrated using that HAP.  Total emissions of the HAP would then
be determined using the emission value integrated for the entire
test run.  Application of Method 25A to more complex gas streams
would need to be examined on a case-by-case basis.   Use of
continuous mass spectrometric techniques where the HAP could be
speciated may be more likely to succeed.
     Method 301 addresses composite sampling techniques and the
EPA believes that procedures similar to those specified for a
composite sampling test method could be applied to. the
measurement of emissions from a batch process.
2.11.1.7  Comment;  Use of the validation protocol in Method 301
will increase the cost of testing and the time required to
complete a test.  The commenter requests that the EPA not be
required to pre-review the validation report before the test
method can be used (IV-D-33, IV-D-47).
     Response;  The EPA agrees that the validation of data
obtained by the proposed method can be conducted concurrently
with other testing in order to combine the validation step with
the source test.  This is encouraged, where appropriate, and.was
an intent of Method 301.  Note that the source owner or operator
incurs some risk in that the proposed method may not meet the
Method 301 validation criteria and the emissions data may not be
applicable for compliance determination.  Well-substantiated
validation reporting should keep this risk to a minimum.
2.11.1.8  Comment;  The cost of validating the data obtained by a
proposed method may be prohibitive for many sources (IV-D-47).
                              2-121

-------
     Response:  The EPA agrees that the development and testing
of a method, even the limited validation approach of Method 301,
can be expensive; however, the cost is necessary for ensuring
data of quality sufficient for the intended purposes.
2.11.1.9  Comment;  The optional procedures of Method 301
(Sections 7, 8, and 9) will increase the cost,of validating a
method (IV-D-47).
     Response;  Although these procedures are not required in all
situations, the EPA agrees that there will be additional costs
incurred when these optional procedures are conducted.
2.11.1.10  Comment;  The commenter urges the EPA to publish a
guidance document outlining source testing methods as soon as
possible (IV-D-58).
     Response;  The EPA has prepared an initial list of validated
methods.  This list may be obtained from the Emission Measurement
Technical Information Center  (EMTIC), U. S. Environmental
Protection Agency  (MD-19)/ Research Triangle Park, North Carolina
27711.  Updates to the list of validated methods will be made
available on  a periodic basis.  The EMTIC may be contacted at any
time to review the current list at  (919) 541-0200.
2.11.1.11   Comment;   The  commenters identified  several
typographical errors, as  well as  grammatical  inconsistencies
 (IV-D-21,  IV-D-31, IV-D-47).
     Response;   These have been corrected.
2.11.2 Method Applicability
2.11.2.1   Comment;   A hierarchy should be  established for
identifying the  appropriate validation procedure to .use
 (IV-D-21).
      Response;   The EPA agrees that such guidance would be
 appropriate in Method 301 and has reworded Sections 5.1 through
 5.3 to define this hierarchy.
 2.11.2.2  Comment;  Method 301 should allow proposed methods to
 be validated using any of the three procedures.  We do not agree
 that a proposed method should be demonstrated against a validated
 method, if a validated method for a particular application is
                               2-122

-------
recognized (IV-D-33).
     Response;  The EPA does not agree.  See response to Comment
2.11.2.1.
2.11.2.3  Comment;  Method 301 should provide a less rigorous
validation procedure for the early stages of method development
(IV-D-31).
     Response:  The EPA agrees that an overall test method
development program will be more extensive than a single
application of Method 301.  Method 301 is specifically a field
validation protocol.  Responsible method development and
evaluation should provide an indication that the proposed method
has a sufficient probability of performing within the criteria of
Method 301 during field validation testing.
2.11.2.4  Comment:  Method 301 may adversely affect State toxic
screening programs that have incorporated less precise emission
measurement techniques (IV-D-31).
     Response;  Method 301 is applicable to source owners or  •
operators wishing to comply with a Federal requirement using a
test method that has not yet been validated.  Because a source
emissions screening and ranking process for use in the
development of a State toxics strategy is not subject to specific
Federal requirements, Method 301 should not affect this type of
program.
2.11.2.5  Comment;  The commenter raises an important
consideration in preparing for a field validation involving
comparison of the proposed method against a validated method
(Section 5.2).  If the emission concentrations are highly
variable over time, as in the case of a batch process such as
coking operations, the method precision calculated using
Equation 301-7 can be disproportionately influenced by large
absolute differences  (IV-D-31).
     Response;  If paired rather than quadruplet samples are used
for procedures under Section 5.2, each test run of the validation
testing should be conducted over the entire process cycle.
                              2-123

-------
2.11.2.6  Comment;  Method 301 does not address the direct
interface or dilution interface sampling options of Method 18
(IV-D-33).
     Response;  Method 301 may be applied to the direct and
dilution interface sampling options of Method 18.  However, the
validation cannot be conducted without quadruplicate sampling and
analytical systems, which the EPA considered excessively
burdensome.  For the other sampling options of Method 18, such, as
container and absorption tube sampling, validation field testing
should be conducted using Method 301.
2.11.3  Bias Criteria and Calculation
2.11.3.1  Comment;  The bias criterion for validation of the data
obtained by a proposed method compared to a validated method is
too restrictive  (IV-D-31).
   '  Response;  For pollutants and sources for which validated
method(s) are available, Method 301 requires alternative methods
to have a similar precision and a reasonable bias  (±10 percent)
compared to the validated method(s).  The EPA believes that these
criteria are not too restrictive and allow for innovative method
development.
2.11.3.2  Comment;  Method 301 does not explain  the bias
correction  limits  (IV-D-33).
     Response;   References in Method 301 that document that the
levels  selected  are typical for those demonstrated in the process
of emission test method development.
2.11.3.3  Comment;  Method 301 requires an excessive number of
samples for the  determination of bias and precision.  The
commenter suggests that reducing the number  of test runs by half
would only result in  a difference  of less than  11  percent  of  the
t-statistic at an 80  percent  confidence  level, but would cut  the
time,  effort,  and cost in half.  (IV-D-33).
      Response;  The  decision  level specified in  the comment  is
 incorrect and should be 0.05.   The difference in the
 t-statistic resulting from reducing the  number of samples  by 50
                               2-124

-------
percent is 17 percent rather than 11 percent.  This difference is
not acceptable to the EPA.
2.11.3.4  Commentt  Section 6.1.5 does not indicate the
acceptance criteria for the correction factor calculated in
Equation 301-5.  Further, -this correction factor may be
inappropriate for methods that employ isotopic spiking (e.g., EPA
Method 23 surrogate standards) where values are not corrected as
long as .the recovery is within 30 percent (IV-D-31).
     Response:  The acceptance criteria for the correction factor
calculated in Equation 301-5 is specified in Section 1.2.1 of the
Method.  Isotopic spiking with surrogate standards as specified
in Method 23 is conducted as a quality assurance check of
breakthrough of the analytes on the sorbent.  The isotopic
spiking specified in Method 301 is used to assess method bias.
The two are not mutually exclusive.  Where a significant bias is
identified (Section 6.1.4) using the spike, the correction factor
(Section 6.1.5) is calculated.  The correction factor is
evaluated by the criteria in Section 1.2.1 and the data are
either corrected or the data and procedure to obtain the data are
rejected.
2.11.3.5  Comment;  The commenter expressed concerns about the
reasonableness of requiring reference materials at concentrations
below a proposed method's detection limit or in concentrations in
multiple ranges representative of several emission points at the
same facility  (IV-D-21).
     Response:  Validation of a proposed method requires that the
method be capable of measuring in the applicable range of
emission concentrations.  A method with a detection limit above
the emission concentration at a particular emission source could
not be validated for that source.  Method 301 does provide for
ruggedness testing, which allows a proposed method to be
validated for  testing at multiple concentrations without
requiring multiple field validation tests.  Whether ruggedness
testing or field validation is used to validate the method for
the appropriate concentrations, it will be necessary to obtain
                              2-125

-------
 reference materials in the applicable  ranges.
 2.11.3.6  Comment;   The commenter suggests  rephrasing
 Sections 5.1.2.1 and 5.1.2.2  to require  sampling the reference
 material at a concentration close to the "expected" concentration
 in the source rather than equal to the emission standard
 (IV-D-21).
      Response;  The EPA agrees and has revised these sections
• accordingly.     .                                  _ .          .  .
 2.11.4  Precision Criteria and Calculation
 2.11.4.1  Comment;   The criterion for  precision specified in
 Method 301 of 50 percent relative standard  deviation was  too
 high, even though nine sampling runs are needed to show
 compliance.  The commenter suggested that the relative  standard
 deviation should be reduced to 15 percent or less  so that only
 three sampling periods are needed to show compliance  (IV-D-47).
      Response;  The method criteria for  precision  and  followup
 testing provide maximum flexibility in method development and
 application while achieving high-quality results.   They also
 serve indirectly as an incentive in method  development; that is,
 the better the method precision, the less effort  is required
 during followup testing using the validated method.
 2.11.4.2  Comment;  The commenter recommends that more explicit
 terminology be used in Sections 6.2.1.3 and 6.2.2.3 regarding the
 F-test, which tests the significance of the proposed method
 variance  to that of the validated method (IV-D-31).
      Response;  The intent of these sections is to perform a
 comparison  of the variances of the two methods.   Method 301 has
 been revised  to include both the F-test and simple comparison of
 variances,  demonstrating  that the variance of the proposed method
 is less than  or equal  to  that of the validated method, as
 suggested by  the commenter.
 2.11.4.3  Comment;  Sections  6.1.2.4  and 6.2.2.4  should be
 revised to  calculate  the  "standard  deviation of the differences"
 rather than the "standard deviation of  the mean of the
 differences"  (IV-D-31).
                                2-126

-------
     Response;  The EPA agrees with the concerns regarding
Sections 6.2.1.4 and 6.2.2.4 and has made the appropriate
changes.
2.11.4.4  Comment;  Section 6.3.2 does not specify calculation of
the relative standard deviation for the spiked samples (as is
specified for the unspiked samples in Section 6.3.6).  It is not
clear which relative standard deviation must meet the +50 percent
criterion in Section 1.2.2 (IV-D-31).
     Response:  Section 6.3.2 has been revised to parallel
Section 6.3.6.  The method has also been revised to specify the
50 percent relative standard deviation criterion for both the
spiked and unspiked samples.
2.11.4.5  Comment;  Equation 301-13 should incorporate the
standard deviation of the unspiked samples as well as that of the
spiked samples (IV-D-31).
     Response:  The EPA agrees with the commenter's suggested
revisions to the calculation and has incorporated them into the
Method.
2.11.4.6.  Comment;  An error was identified in Equation 301-7
(IV-D-31).
     Response;  This has been corrected.
2.11.5  Sample Stability
2.11.5.1  Comment;  The commenter requests additional guidance in
Section 8 on appropriate data analysis techniques (IV-D-31).
     Response;  The procedures provided are sufficient.  The
analyst has flexibility to determine the applicable data analysis
techniques for the specific sample stability study.
2.11.5.2  Comment;  The commenter requests additional guidance in
Section 8.2.1 regarding sample stability studies for impinger
sampling systems that require sample extraction or digestion
(IV-D-31).
     Response;  Section 8.2.1 has been revised to include
guidance for samples requiring extraction or digestion.
2.11.5.3  Comment;  Sections 8.2.2 and 8.2.3 should be clarified
to specify whether spiked or unspiked samples (Section 5.3)
                              2-127

-------
should be used to study sample stability (IV-D-31).
     Response;  The study should include equal numbers of spiked
and unspiked samples.  Specific stability interpretations should
be made with the spiked samples.  The unspiked samples are
included to allow the analyst to assess the application of the
spiked sample findings.  Sections 8..2.2 and 8.2.3 have been
revised to clarify this point.
2.11.5.4  Comment:  The commenter requested additional guidance
and clarification on appropriate techniques.  For example,
Section 8.2.4 should specify that half the samples analyzed at
the minimum and maximum storage times should be from the
validated method and half from the proposed method.  It should
also address procedures for samples that require extraction or
digestion  (IV-D-31).
     Response;  Several wording clarifications have been made.
Section 8.2.4 has been revised to specify that equal numbers of
samples from the proposed and validated methods be analyzed.  The
procedures for samples requiring extraction or digestion should
parallel those described in Section 8.2.2.
2.11.6  Practical Limit of Quantification
2.11.6.1   Comment;   Section 9 on calculating the practical limit
of quantitation  should provide procedures for cases where there
is a non-zero intercept on the calibration curve  (IV-D-31).
     Response;   The procedure provided  is applicable to this
case.
2.11.6.2   Comment;   Section  9.3.5  should provide the calculation
method for SQ at zero concentration (IV-D-31).
     Response;   The procedure provided is applicable  to  this
case.
 2.11.7  Followup Testing
 2.11.7.1  Comment;   Method 301 should include additional quality
 assurance requirements for follow-up testing (IV-D-33).
      Response;  Quality assurance is the goal of applying Method
 301.  Quality control is a general requirement of any compliance
 test,  but is beyond the scope of Method 301.  We suggest that the
                               2-128

-------
general procedures of the Quality Assurance Handbook for Air
Pollution Measurement Systems, Volume III, Stationary Source
Specific Methods, EPA-600/4-77-027b53, be followed.
2.11.8  Waivers/Extending Method Applicability
2.11.8.1  Comment:  A State wishes to have its future methods
approved as either, "documented" or "conditional" (IV-D-31).
     Response;  State methods, approved through a Federal approval
process would have "conditional" status.
2.11.8.2  Comment:  "Documented" methods (Section 12.1.2 of
Method 301), such as National Institute of Occupational Safety
and Health, American Standards and Testing Materials, and
Occupational Safety and Health Administration methods should be
considered validated methods (IV-D-47).
     Response;  The.application for a waiver under
Section 12.2 should demonstrate the applicability of the
"documented" method to the particular source to be measured.  The
validation procedure and documentation for that particular source
would be reviewed with respect to Method 301 requirements.
2.11.8.3  Comment;  The Method 301 requirements in Section 12.1.1
for extending applicability of a method to "similar" sources are
burdensome.  There should be a more reasonable requirement than
conducting a ruggedness test and applying to the EPA for a waiver
(IV-D-33).
    . Response;  Section 12.1.1 states applicability to other
sources may be demonstrated by conducting ruggedness testing.
The EPA agrees with the commenter that ruggedness testing is not
always necessary/ depending upon the similarity of the sample
matrices.  In addition, validation of the proposed method on a
complex sample matrix may be sufficient to demonstrate that the
method is applicable to a similar source with an emission matrix
of less complexity.  In every case, the requester must apply for
a waiver to extend the applicability.
2.11.8.4  Comment;  Method 301 does not specify acceptable
deviations in method procedures after validation (IV-D-33).
                              2-129

-------
     Response;  Once validation of a method has been completed
and the data used to meet a Federal or State requirement,
modifications to the method cannot be made without approval of
the responsible agency or validation through application of
Method 301.  Any modification must be fully explained and
documented with supporting data such as that from a ruggedness
test.  The primary consideration in acceptance of a modification
will be its effect on the results obtained by the previously
validated method.
2.11.8.5 Comment:  A State is concerned that the validation of a
method at one source will be inappropriately interpreted as a
validation for all applications  (IV-D-31).
     Response;  The field validation procedure of Method 301 is
appropriate for validating emissions data only from the source
and at the levels for which the method was validated.  Optional
procedures are provided  and must be applied to extend
applicability of the methodology beyond the specific emission
source and concentration levels  at which the validation was
conducted.
2.12  MISCELLANEOUS COMMENTS
2.12.1   Comment;  The  commenter  believes that  the Early
Reductions Program offers an  incentive to  sources only if  the
Section  112(d)  standard is greater than  90  (95)  percent.   To
increase the incentive to make early reductions, the commenter
 suggests reducing emission fees  from sources participating in  the
Program based on the difference between  the required
 Section 112(d)  reduction and the 90 (95)  percent reduction.   For
 example, if emission fees would have been $100,000  per year and a
 Section 112(d)  standard of 98 percent is promulgated then the
 source would have to pay $100,000.  If a 112(d)  standard of 90
 (95) percent is promulgated,  the source would have to pay $0.   If
 a Section 112(d) standard somewhere in between were promulgated,
 the fee would be linearly proportional (IV-D-02).
      Response;  There is nothing in the CAA that allows Federal
 regulations to tie emissions reductions to a fee-based incentive
                               2-130

-------
program.  Fees are only allowed as,part of a SIP in order to
allow States to recover the cost of a Title V permitting program,
and are based on residual emissions.  Specific State regulations
governing fee schedules would have to be changed to address this
comment.
2.12.2  Comment:  The EPA should provide other incentives such as
preferred government procurement status, tax incentives, .and a
greater certainty of future control requirements to entice
industry to make voluntary reductions (IV-D-07).
     Response:  Incentives such as those suggested by the
commenter go beyond the authority of the CAA.  The EPA believes
that regulatory relief from compliance with a Section 112(d)
standard provides the necessary incentives for voluntary
reductionc.  The EPA is making every possible effort to
communicate the benefits of early reductions.  In addition, the
EPA will keep industry informed regarding the development of
control requirements under Section 112(d) standards.
2.12.3  Comment:  Two commenters suggest that the EPA establish a
question-and-answer hotline (IV-D-33, IV-D-45).  One commenter
further suggests developing an ongoing question-and-answer
document available that provides several seed questions
(IV-D-33).
     Response;  The EPA has developed a question-and-answer
document based on specific questions raised by Regions, State and
local agencies, and industry.  The document will continue to
evolve throughout the course of the Early Reductions Program and
should provide consistent answers to often-encountered problems.
The EPA is also available as necessary to provide support and
guidance to State and local agencies.
2.12.4  Comment;  One commenter was concerned that facilities
participating in the Early Reductions Program would have to: (1)
install controls to meet the 90 (95) percent requirement,  (2)
install additional control to meet Section 112(d) standards at
the end of the extension period (if the applicable Section  112(d)
standard is greater than 90 [95] percent), and (3) then install
                              2-131

-------
controls to meet any residual risk 112(f) standard.  In addition,
the State may require additional control at any time in the
process.
     The commenter does not believe this provides incentive to
participate in the Program and, therefore, recommends the
following options to increase the incentives  (1) allow the
control strategy used to achieve the 9Q  (95) percent reduction to
be defined as the Section 112(d) standard (i.e., rather than
extend compliance with the Section 112(d) standard, let 90 (95)
percent reduction replace the requirement of the standard), (2)
the source may determine what it believes the applicable
Section 112(d) standard will be and use  these controls to achieve
the 90  (95) percent or higher reduction; if the Section 112(d)
standard is determined to be something.else, the source will not
be punished for its "good faith" determination and will receive
an indefinite compliance variance from the applicable
Section 112(d) standard, or  (3) the EPA  will commit to complete
"residual risk analyses" 2 years prior to the delayed compliance
deadline  (this would allow facilities to directly  install
residual risk control strategies only rather than meet the
Section 112(d) standard and  then residual risk requirements
 (IV-D-07).
      Response;  The first two  suggestions offered  by the
commenter  are not allowed by the CAA.  Standards established by
Section 112(d) of the CAA will be based  on  technological
considerations.   Regarding the residual  risk  standards, the
provisions of the CAA will result in Section  112(f)  standards
being promulgated prior to the expiration of  the Early Reductions
 extensions or,  in the worst  case,  just as the extension expires.
 Facilities will not be  faced with the possibility  of installing
 unnecessary controls to meet Section 112(d)  and Section  112(f)
 standards sequentially.
 2.12.5  Comment;   Facilities that achieved 90 (95) percent
 reduction under the "Nine CEO's Voluntary Emission Reduction
 Program"  should be- automatically exempted for 6 years from
                               2-132

-------
compliance with the Section 112(d) standards without going
through the extension application process (IV-D-05).
     Response:  Facilities that participated in the "Nine CEO's
Voluntary Emission Reduction Program" must.submit an application
or enforceable commitment according to the same procedures as
facilities that did not participate in the.CEO program..  The
process will allow the owner or operator to describe which
portion of a facility will be defined as the "source."  The
application process also allows for review and public comment,
which the CEO program did not.  Recent activity and documentation
by the CEO participants should make an Early Reductions submittal
easier to develop.
2.12.6  Comment;  The rule should specify how relatively new
facilities may participate in the Program when actual base year
emissions are yet to be established (IV-D-30).
     Response;  It may be very difficult for sources built just
prior to proposal of an applicable Section 112(d) standard to
participate in the Program.  First, the regulations require that
actual and verifiable base year emissions be established.  This
means that a representative operating history is required prior
to proposal of Section 112(d) standards.  Additionally, most
States require a reasonably high level of control for HAP
emissions from new sources as part of new source permitting.
Although possible, it will be difficult to achieve 90 (95)
percent reduction above the level of control required.
2.12.7  Comment;  The EPA should reduce requirements for
application submission and supporting documentation, and agency
review times should be shortened (IV-D-27).
     Responses  The requirements for application submission,
supporting documentation, and allowances for review time are
judged to be the minimum requirements necessary to ensure that
the Early Reductions Program is successful.  This information is
needed to verify that emissions reductions are actual reductions,
and not "paper" reductions.  Any reduction of agency review time
would not ensure adequate review of the applications.
                              2-133

-------
2.12.8  Comment;  Sections 63.75(d) and 63.77(f)  should provide
consistent deadlines for submittal of late test data:  180 days.
The only exception should be enforceable commitments, which
should be provided by March 31, 1994 (IV-D-33).
     Response;  The test data referred to under Section 63.75(d)
supports base year data for enforceable commitments.  Sources
making enforceable commitments may need to act quickly to submit
the commitment prior to proposal of a Section 112(d) standard.
In this instance, sources will have very little time to obtain
accurate test data.  Therefore, the EPA has granted a reasonable
time period of 180 days after proposal of an applicable
Section 112(d) standard for sources to conduct necessary testing
to support enforceable commitments.
     The test data referred to under Section 63.77(f) supports
post-reduction emissions for permit applications.  In this case,
the EPA proposed  allowing only 90  days for submission of test
data.  Because sources planning to submit a permit application
will have sufficient lead time before applying, less additional
time is needed than for submittal  of supporting data for the
enforceable commitment to produce  necessary test data.
     The EPA  has  modified the Early Reductions regulation to make
consistent  the deadline for post-reduction emissions test data to
support the reduction demonstration for enforceable  commitments
and permit  applications.  Section  63.77(f) of  the proposed rule
allowed  sources with enforceable commitments  120 days  (i.e., -from
December 1,  1993  until March  31,  1994) after  the permit
application deadline to  submit test data  but  required  other
permit applicants to  submit the  same type data within  90  days.
The  90-day deadline has  now been changed  to  require test  data
within 120-days after  the deadline for submittal of the permit
 application.
 2.12.9  Comment;   The alternative emission limitation  should not
 be required for each emission point within a source but as  a
 single annual average,  source-wide emission limit  reflecting the
 required reductions (IV-D-27).
                               2-134

-------
     Response;  Any alternative emission limitation must be
enforceable.  The alternative emission limitation could be
written for a single emission point, a group of emission points,
or the entire source.  The specific requirements will depend on
the definition of source and what the permitting authority
     *
considers to be enforceable.  At a minimum, the alternative
emission limitation must reflect continuing compliance with 90
(95) reduction. _If an applicable Section 112(d) standard has
been proposed, the alternative emission limitation may parallel
the requirements of the standard.
2.12.10  Comment;  The commenter states that the CAA does not
specify that numerical limits be set for alternative limits.
Therefore, the commenter urges the EPA to specify in the final
rule that the permitting authority has discretion in defining
alternative emission limitations, and is not required to use
numerical limits whenever possible (IV-D-38).
     Response;  Section 112 of the CAA requires that numerical
standards be specified wherever possible.  However, any format is
acceptable as long as it is enforceable and ensures continued 90
(95) percent reduction as required by the Early Reductions
Program.
2.12.11  Comment;  Regarding Section 63.72, should the words
"permitting authority" be used instead of "state acting pursuant
to a permitting..."?  The current wording implies that only the
State has permitting authority (IV-D-52).
     Response;  The regulation has been changed to incorporate
this comment.  Section 63.72 of the regulation describes the
general provisions of the Early Reductions Program.  During the
course of the Program, the authority to review and issue permits
will shift from the EPA to the States.  The wording in the
proposed regulation limited the provisions to States.  The term
"permitting authority" refers to States or the EPA and is,
therefore, the more appropriate term in this context.
2.12.12  Comment;  The definition of responsible official in the
proposed rule is too complicated.  A corporate officer or
                              2-135

-------
delegated official should have the authority to sign.  The
"financial test" in the rule should be deleted (IV-D-29,
IV-D-38).  Another commenter suggested the definition of
responsible official be consistent with the definition used in
the Title V permit regulations (IV-D-45).
     Response;  The definition of responsible official is
consistent with the definition presented in the Part 70
regulations.
2.12.13  Comment:  The EPA should publish as quickly as possible
the intended regulation proposal dates.  In addition, a list of
source categories for which Section 112(d) standards are under
development should be published from time to time, with names and
telephone numbers of contacts with whom sources could discuss the
progress of the Section 112(d) standards  (IV-D-27, IV-D-34 -
should be published in the Federal Register at least 90 days
prior to proposal; IV-D-38 - at least  1 year prior to proposal).
     Response;  The EPA published its  source category list in the
Federal Register on July  16, 1992  (57  FR  31576)54, for which it
intends to develop Section  112(d) standards and a draft schedule
for proposing those standards  (57 FR  44147; September 24,
1992)55.  In addition, every 6 months, the EPA publishes  a
regulatory agenda that identifies the regulations currently under
development,  with the names and phone numbers  of the appropriate
contacts.  These published  lists  should  keep  industry and the
public adequately  informed  with regard to specific proposal
dates.
 2.12.14  Comment;   The EPA should allow  certain process
 information to be  submitted as confidential  on a  case-by-case
 basis.   Specifically, the following types of  data should be
 shielded;  (1) concentration of specific chemicals in process
 vents, (2)  boiler and process design capacities,  (3) hourly
 maximum design rates, (4). release temperatures,  (5)  release
 velocities,  and (6)  specific chemical emission rates (IV-D-28,
 IV-D-30, IV-D-33,  IV-Dr35,  IV-D-45).
                               2-136

-------
      Response;  According to the CAA, "emissions data" cannot be
 considered confidential.  A recent Federal Register notice
 specified categories of -data that qualify as "emissions data" and
 are therefore, not confidential (56 FR 7042, February 21,
 1991).5*>  Such data include but are not limited to
      "...identification of facility and emission points, emission
      types (type of release point and specific pollutants),
      emission rates, release heights, descriptions of terrain and
      surrounding structures, stack and vent diameters at the
      point of emission release, release velocities, release
      temperatures, frequency of releases, duration of releases,
      concentrations, densities of emission streams or average
      molecular weights, boiler or process design capacities,
      emission estimation methods, percent space heat, and hourly
      maximum design rates."
 The EPA will continue to follow these regulations while
 implementing the Early Reductions Program.
      Other information not classified as emission data may be
 determined to be confidential on a case-by-case basis at the
. request of the submitter.  The EPA has strict guidelines and
 procedures for handling confidential data to protect its
 confidentiality.
 2.12.15  Comment:  This commenter wants assurance that time
 periods can be extended when confidential data are in issue
 (IV-D-39).
      Response;  Data initially identified by the submitter as
 confidential but later determined nonconfidential will be
 afforded the full review time provided in the rule.
 2.12.16  Comment;  The EPA should clarify the rule so as to make
 the 1 year extension under Section 112(i)(3)(B) available to an
 Early Reductions source at the end of the 6-year compliance
 extension, if needed to install controls (IV-D-38).
      Response;  The extension provision given in the CAA
 [Section 112(i)(3)(B)] only applies to sources impacted by
 individual Section 112(d) standards.  The Early Reductions
 Program already provides a 6-year extension from complying with
 any Section 112(d) standard.  Sources participating in the Early
 Reductions Program should have ample time to plan and install
                               2-137

-------
necessary control technology to meet these standards.
2.12.17  Comment:  The commenter feels the regulation should
emphasize the importance of pollution reduction, not
participation by industry and gives specific examples where the
emphasis on participation may lead to increased emissions
(IV-D-39).
     Response;  The legislative history indicates that Congress
wanted the EPA to encourage participation in the Early Reductions
Program.  The EPA believes that greater participation will lead
to more early reductions and, hence, more overall benefit to the
environment.  The overall goal of the regulation is to reduce
pollution.  In order to achieve this goal, sources must
participate in the Program.
2.12.18  Comment;  The commenter proposed that the EPA allow a
6-month extension for companies to comply with Section 112(d)
standards if they make a genuine effort to participate in the
Early Reductions Program but are forced to withdraw prior to
December  1, 1993  (IV-D-48).
     Response;  Under Section  112(1)(3)(B) of the CAA, the
Administrator  (or a  State with a program approved under
Title V)  may issue a permit that allows a source up to 1
additional  year  to comply with a Section.112(d) standard if such
an additional  period is necessary  for  the installation of
controls.   If  the  source needs extra time to install  controls
because  of  participation in the Early  Reductions Program,  the
source may  apply for this  1 year extension.  An additional
extension is  not authorized by the CAA.
2.12.19   Comment;   The commenter  suggests clarifying the
regulation to state that  Section  112(d)  reductions  apply to base
year emissions and not to  emissions remaining  after reductions
made for the Early Reductions Program.  Clarification is also
 needed in how to define base year and the Section 112(d)  standard
 if the Section 112(d)  source and the early reductions defined
 source are not the same (IV-D-02,  IV-D-08).
                               2-138

-------
     Response:  Section 112(d) standards will be specific to each
source category and may assume several formats, including percent
reduction, concentration limits, mass limits, etc.  If a percent
reduction is selected, for example 98 percent for MACT, the 90
percent reduction for Early Reductions generally will count
toward the required Section 112(d) reduction.
2.12.20  Comment;  Clarify that Section 63.81 "Rule for special
situations," should be clarified to indicate that a company must
make an enforceable commitment or achieve reductions before the
first Section 112(d) standard is proposed for the "source" that
will receive the extension, not before the first Section 112(d)
standard applicable to the facility is proposed (VI-D-08,
IV-D-19).
     Response;  Section 63.81 already states that the
Section 112(d) standard must be applicable to a source as defined
under Section 63.73.  Thus, the regulation clearly indicates that.
only emission points included, in the Early Reductions source
determine' the appropriate submittal date.  To clarify the
provisions of this section of the regulation, a permit
application or enforceable commitment must be submitted prior to
proposal of the earliest Section 112(d) standard that applies to
any emission point in the Early Reductions source definition.
The extension for compliance, however, begins on the compliance
date of the Section 112(d) standard applicable to the emission
point.  This will lead to different compliance extension     "*
expiration dates for different emission points in a source
subject to more than one Section 112(d) standard proposed at
different times.  Emission points not included in the Early
Reductions source must comply with any applicable Section 112(d)
standard according to the schedule in Section 112(e).  At a later
date, the source owner or operator may reapply with a different
definition of source as long as no Section 112(d) standard has
been proposed that applies to any part of the "new" source.
2.12.21  Comment:  The emphasis on post-control emissions should
be removed (i.e, it should focus on post-reduction emissions)
                              2-139

-------
(IV-D-12, IV-F-lb).
     Response;  The EPA recognizes that there are many methods to
reduce emissions other than applying a control device.  Any
emphasis on using a control device versus other methods of
emission reduction was unintentional and, in fact, the EPA
encourages pollution prevention and recycling methods in
preference to add-on controls.  All references in the proposed
regulation to post-control emissions have been replaced with
post-reduction emissions.
2.12.22  Commentt  The situation in which more than one
Section  112(d) standard affects a source should be clarified.
The current regulation could be interpreted to mean that all of
the source must be in compliance with the applicable
Section  112(d) at the end of the 6-year extension, when in
reality  only the portion of the source affected by the standard
must comply  (IV-D-27).                             ,     .
     Response;  If several Section  112(d) standards apply to a
source under the Early Reductions Program, the source owner or
operator must  submit a permit  application or  enforceable
commitment  prior to the proposal of the  earliest  Section  112(d)
standard that  applies to any emission  point in the defined
source.  The  extension for compliance, however, begins on the
compliance  date  of the Section 112(d)  standard applicable to each
specific emission  point  or type emission point.   Thus, source
compliance  extensions  for various  emission points may* begin and
end at different times.  Emission  points not  included in  the
Early Reductions source  must  comply with any  applicable
Section 112(d) standard  according  to.the schedule in
Section 112(e).
 2.12.23  Comment;   In the  preamble to the  proposed rule,  the
 statement (on page 27360)  "If the application is denied,  the
 source will be subject to applicable MACT  standards" is
misleading.  The EPA should clarify the statement to reflect  that
 the source will be subject to Section 112(d)  standards on the
 timetable specified in Section 112(e) (IV-D-30).
                               2-140

-------
       Response:   The regulation in  Section  63.80(d)  states  that
  the  source must comply with an "applicable standard issued under
  Section.112(d)  of the CAA by the compliance date  specified in
  such standard."  Therefore, the regulation clearly  indicates that
  the  source must not immediately be in compliance  with the
  standard as soon as the permit is  disapproved but must  adhere to
  the  same compliance schedule as sources that did  not apply for  a
  compliance extension.
  2.12.24   Comment;  The commenter requests  clarification of the
  procedures and  requirements of the proposed rule  with respect to
  question 1 under paragraph F:   "Permit Application  and  Permits"
  of the preamble (page 27630)  (IV-D-30).
       Response:   The procedures and requirements for emission
  tests are stated in Sections 63.74(d)(3) and 63.74(i).   The
  questions in Section Fl of the proposal preamble  are reminders  to
  the  reviewer to ensure these requirements  are met.
  2.12.25   Comment:  The commenter suggests  that the  word "train"
  be substituted  for "unit" in the list of definitions.   The
  commenter feels that the term "unit" describes an independent
  piece of equipment, while a "train" better describes a  group of
  equipment linked together for the  purposes of producing a  product
  or performing a function.  The commenter also suggests  that the
  EPA  include a definition for "emission point" to  describe  the
  affected source where the enforceable emission limitations apply
"* (IV-D-39)            •
       Response;   Because it does not appear elsewhere in the
  regulation, the term "process or production unit" has been
  removed  from the -regulation under  the definition  of source and
  under Section 63.72, "Definitions."
  2.12.26   Comment;  There is no definition  of "facility"  in the
  regulation (IV-D-52).
       Response;   The term "facility" has been replaced throughout
  the  regulation  with "plant site,"  which is a "contiguous area
  under common ownership or control." The only reference to
 .facility that remains is related to Sections 63.73(a)(l) and  (3),
                                2-141

-------
both of which list "a building structure, facility, or
installation."  In both instances, the definition of facility
will remain the same as that intended by Section 111(a) of the
CAA.
2.12.27  Comment;  The proposed rule should emphasize that
companies that make good-faith efforts will not be subject to
enforcement action if errors or mistakes in the enforceable
commitment are discovered by the company or the EPA (IV-D-35).
     Response;  The EPA will use its enforcement discretion for
cases where companies make an honest mistake in their estimation
of the base year  emission data or supporting materials.  In such
instances, companies may be required to revise their base year
and make additional reductions in order to achieve 90  (95)
percent emission  reductions.  Alternatively, companies may
rescind their commitment without penalty prior to December 1,
1993.  Sources  found  submitting false  or fraudulent information
in their commitment for early reductions, even after initial
approval of  their base year  submission,  shall be subject to
enforcement  action under  Section  113 of  the CAA or other Federal
statutes.
2.12.28  Comment; The  commenter  recommends that the EPA allow
facilities that do not.meet  their enforceable commitments to  be
 subject to applicable Section 112(d)  standards only,  rather than
 face possible penalties (IV-D-44).
      Response;   Sources that make an enforceable  commitment to
 achieve early reductions can revoke the commitment without
 penalty up to December 1, 1993.   These sources would thus be
 subject only to the applicable Section 112(d)  standard.   This
 allows the source a considerable amount of  time to attempt  to
 achieve the promised reductions.   If the source is unable to
 achieve those reductions, it may terminate its participation in
 the Early Reductions Program without sanction,  provided it does
 so by the required deadline.  Other participants in the
 discussions  leading up to proposal of this rule expressed the
 concern that Congress intended.that commitments are to be.kept,
                               2-142

-------
and that once made, they could never be revoked.  To give any
meaning to the "enforceable" part of enforceable commitment,
there must be something that is enforceable.  .The EPA believes
that, if after given every opportunity to revoke the commitment
if necessary, the unrevoked commitment is enforceable.  The EPA
feels that the suggested compromise between the competing
positions is reasonable.
     Companies that achieve the committed reductions prior to
January 1, 1994, but subsequently increase emissions in violation
of their Title V permit conditions (or in violation of the
enforceable commitment if the permit has not yet been issued),
would not have to revert to meeting the Section 112(d) standard.
The companies would be subject to an appropriate penalty, but
would continue to maintain the considerable benefits of
continuing in the Early Reductions Program.  Likewise, the
companies that indicate that they are continuing in the Program
by not revoking their commitment prior to December 1, 1993, but
who fail in their performance by not achieving the reductions
committed to prior to January 1, 1994, will likewise be subject
to an appropriate penalty.  The EPA will use enforcement
discretion to determine the severity of penalties, based on the
efforts made to achieve the reductions.
2.13  REFERENCES
1.   126 Congressional Record, S16927, October 27, 1990.
2.   Federal Register, Volume 57, page 31576, July 16, 1992.
3.   Federal Register, Volume 56, page 9318, March 6, 1991.
4.   Federal Register, Volume 51, page 43814, December 4, 1986.
5.   House of Representatives Report Number 101-490, May 17,
     1990.
6.   Protocols for Generating Unit Specific Emission Estimates
     for Equipment Leaks of VOC and VHAP.  United States
     Environmental Protection Agency.  Research Triangle Park,
     NC.  Publication No. EPA-450/3-88-010.  October 1988.
7.   Federal Register, Volume 56, page 9315, March 6, 1991.
8.   Enabling Document for Regulations Governing Compliance for
     Early Reductions of Hazardous Air Pollutants.  United States
                              2-143

-------
     Environmental Protection Agency.  Research Triangle Park,
     NC.  Publication No. EPA-450/3-91-013.  July 1991.

9.   Memorandum from Hassett-Sippie, B., Pollutant Assessment
     Branch, Environmental Protection Agency, to Early Reductions
     Docket A-90-47.  May 30, 1991.  Revised Criteria for
     Selecting High Risk Pollutants for the.Purposes of Clean Air
     Act Section (i)(5)(E).
10.  Memorandum from Keating, M.H., Pollutant Assessment Branch,
     Environmental Protection Agency, to Early Reductions Docket
     A-90-47. May 8, 1991.  Background Information for
     Noncarcinogens Recommended for Inclusion of the High-Risk
     List Since Proposal.
11.  Federal Register, Volume 54, page 38044, September 14, 1989.

12.  Memorandum from Shoaf, C., Environmental Protection Agency
     Office of Criteria  and Assessment, to A. Vasu, EPA:ESD.
     January 13, 1992.   Need for Additional Weighting of Two
     Chemicals.
13.  Memorandum from Keating, M.H., Pollutant Assessment Branch,
     Environmental Protection Agency, to Early Reductions Docket
     A-90-47.  January 31, 1992.  Diisocyanates as High-Risk
     Pollutants- for -the  Early Reductions Program - Meeting
     between the Environmental Protection Agency and the Chemical
     Manufacturers Association  (CMA).
14.  Developmental Toxicity  Study of  Inhaled Toluene Diisocyanate
     Vapor  in  CD®  (Sprague-Dawley)  Rats.   R.W. Tyl.  Bushy Run
     Research  Center Project Report
     50-592.   1988.
15.  Two-generation Reproduction Study  of  Inhaled  Toluene
     Diisocyanate  in CD® (Sprague-Dawley)  Rats.  R.W. Tyl  and
     T.L. Neeper-Bradley.  Bushy Run Research  Center Pronect
     Report 51-576.   1989.
16.  Five-year Longitudinal  Study  of Wprkers Employed in a New
     Toluene Diisocyanate Manufacturing Plant.   J.E. Diem, et al.
     Am. Rev.  Respir.  Dis.,  1982.  pp. 420  thru 428.
17.  The Acute and Long-Term Respiratory Effects.of Aromatic
     Diisocyanates;  a Five-Year Longitudinal Study of
     Polyurethane Foam Workers.   J.'Bugler,  et al. Report
     distributed to the British Rubber Manufacturers'
     Association,  the International Isocyanate Institute,  and the
      Health and Safety Executive Council.   May 7,  1991.
 18.  Abnormal Lung Function in Polyurethane Foam Producers:   Weak
      Relationship to Measured TDI Exposures (abstract).  R.N.
      Jones et al.  American Review of Respiratory Diseases
      143:A440 (International Conference Supplement).   1991.
                               2-144

-------
19.



20.


21.




22.



23.




24.




25.



26.



27.




28.


29.
30.
Acute Respiratory Effects in Workers Exposed to Low Levels
of Toluene Diisocyanate (TDI).  J. M. Peters et al.  Arch
Environ Health—Vol. 16, May 1968.  pp 642 thru 647.
Studies of Isocyanate Toxicity.  J.M. Peters.  Proc. Roy.
Soc. Med. Volume 63, April 1970.  pp 372 thru 374.

The Relationship of Acute Pulmonary Effect of Organic
Materials to Chronic Pulmonary Effects.  J.M. Peters.
Reprinted from Annals of the New York Academy of Sciences,
Volume 221, February 28, 1974.  pp 44 thru 49.

A Dose-Response Relationship in TDI Workers.  D.H. Wegman,
et al.  Journal of Occupational Medicine. Vol. 16, No. 4,
April 1974.  pp 258 thru 260.

Epidemiology of Toluene Diiosocyanate (TDI)-Induced
Respiratory Disease.  J.M. Peters and D.H. Wegman.
Environmental Health Perspectives. Vol. 11, June 1975.  pp.
97 thru 100.

Chronic Pulmonary Function Loss from Exposure to Toluene
Diisocyanate.  D.H. Wegman, et al.  British Journal of
Industrial Medicine. Vol. 34, 1977.
pp. 196 thru 200.

Absence of Respiratory Effects in Subjects Exposed to Low
Concentrations of TDI and MDI:  A Reevaluation.  A.W. Musk,
et al.  Journal of Occupational Medicine, Vol. 27, No. 12,
December 1985.  pp. 917 thru 920.

Isocyanates and Respiratory Disease:  Current Status.  A.W.
Musk, et al.  American Journal of Industrial Medicine, 1988,
pp. 331 thru 349.

Health Assessment Document for Chromium.  Final Report.  D.
Basu et al.  EPA-600/8-83-014F, U. S. Environmental
Protection Agency, Research Triangle Park, North Carolina,
1984.  (NTIS PB85-115905).

U. S. Agency for Toxic Substances and Disease Registry.
Toxicological Profile for Chromium.  Oak Ridge National
Laboratory, 1989.  (NTIS PB89-236665).
Federal Register.  Volume 55, page 650, January 8, 1990.
National Toxicology Program Study (three reports).
a) Subchronic Inhalation Toxicity Study of Chloroprene in
Rats (Contract #NO1-ES-5515).
b) Subchronic Inhalation Toxicity Study of Chloroprene in
Mice (Contract #NO1-ES-5515).
c) Pathology Working Group Chairpersons Report.  1988.
13-week Subchronic Toxicity Test by Inhalation of
Chloroprene in Fischer 344 Rats and B6C3F1 Mice.
                              2-145

-------
31.  Relationship Between Lung Cancer and Distance of Residence
     from Nonferrous Smelter Stack Effluent.  W.M. Greaves, et
     al.  American Journal of Industrial Medicine 2:15-23.  1981.

32.  Lung Cancer Mortality Among Residents Living Near the El
     Paso Smelter.  W.N. Rom et al.  British Journal of
     Industrial Medicine  39:269-272.  1982.

33.  Chronic Toxicity and Oncogenicity of Inhaled Methyl
     Methacrylate and n-Butyl Acrylate in Sprague-Dawley Rats.
     Reiningnaus*, et al.  Fd. Chem Toxicol 29:329-339.  1991.

34.  Chronic Toxicity and Oncogenicity Bioassay of Inhaled Ethyl
     Acrylate in Fisher 344 Rats and B6C3F1 Mice.  R.R. Miller et
     al.  Drug. Chem. Toxicol. 8:1-42.  1985.

35.  Federal Register, Volume 51, No. 185, September 24,  1986,
     Guidelines for Carcinogen Risk Assessment.

36.  Health Assessment Document for VDC.  Final Report.    JjjPA-
     600/8-83-031F, U. S. Environmental Protection Agency, Office
     of Health and Environmental Assessment, Washington,  D.C.,
     1985.
37.  Carcinogenicity Bioassays of Vinylidene Chloride  - Research
     Plan and Early Results.  C. Maltoni  et al.   T.a Medicinia del
     Lavoro  68:240-262.  1977.
38.  Health  Assessment  Document for  Ethylene Oxide.  Final
     Report.  D.A. Gray et  al.  EPA-600/8-84-009F, U.  S.
     Environmental Protection Agency, Office of Health and
     Environmental Assessment, Washington,  D.C.,  1985.  (NTlb
     PB86-102597).
39.  Mortality Among Workers Exposed to Ethylene  Oxide.   *•
     Steenland et al.   Mew England Journal  of  Medicine 324:1402,
     May 16, 1991.
40.  A Mortality Study of Men Assigned  to Ethylene Oxide
     Production  or Other Ethylene Oxide Related Chemical
     Manufacturing.  'H.L. Greenberg et  al.  In press,  Brit. J.
      Ind. Med.  1989.
 41.  A Cohort Study of Industrial and Hospital Workers Exposed to
      Ethylene Oxide.   M.J. Gardner et al.  Presented at the Sixth
      International Symposium in Epidemiology in Occupational
      Health, Stockholm, Sweden.   August 15-18.  Brit.  J.  Ind.
      Med.  December,  1989.
 42.   Final Report on a Multicentral Epidemiological Study on
      Ethylene Oxide.   N. Kiesselbach et al.  Presented at the
      Sixth International Symposium in Epidemiology in
      Occupational Health, Stockholm, Sweden.  August  15-18, 1988.

 43.  Epidemiologic Support for Ethylene Oxide as a Cancer Causing
      Agent.  C. Hogstedt et al.  ^n^nal of tbp  American Medical
      Association 255:1575-8.  1986.
                               2-146

-------
44.  Epidemiologic Studies on Ethylene Oxide and Cancer:  An
     Update, In:  Bartsch, H., Hemminiki, K., O'Neill, I., eds,
     Methods for Detecting DNA Damaging Agents in Humans:
     Applications in Cancer Epidemiology and Prevention.  C.
     Hogstedt.  IARC Scientific Publication No. 89.  New York:
     Oxford University Press 1988:265-70.

     A Cohort Study on Cancer Mortality of Ethylene-Oxide
     Workers. L. Bisanti et al.  Presented at the Sixth
     International Symposium in Epidemiology in Occupational
     Health, Stockholm, Sweden.  August 15-18, 1988.

     A Preliminary Report of Cancer Incidence in a Group of
     Workers Potentially Exposed to Ethylene Oxide.
     P.O. Stolley, Clinical Epidemiology Unit, University of
     Pennsylvania Hospital.  Unpublished Report.  April 25, 1986.
     Mortality and Cancer Morbidity Among Workers in a Chemical
     Factory.  L. Hagmar et al. Scand. J. Environ. Health 12:
     545-551, 1986.

     Hazard Assessment of Ethylene Oxide.  L. Goldberg.  Ethylene
     Oxide Industry Council, CRC Press, Boca Raton, Florida.
     1986.

     U. S. National Toxicology Program.  NTP Technical Report on
     the Toxicology and Carcinogenesis Studies of Ethylene Oxide
     (CAS no. 75-21-8) in B6C3F1 Mice (inhalation studies).  NTP,
     U. S. Department of Health and Human Services, Public Health
     Service, National Institutes of Health, Research Triangle
     Park, North Carolina.  1987.

50.  Chronic Inhalation Toxicity of Hydrazine:  Oncogenic
     Effects.  J.D. MacEwen et al.  Report Number AFAMRL-TR-81-
     56, Air Force Aerospace Medical Research Laboratory, Wright-
     Patterson Air Force Base, Ohio.  1981.

51.  Federal Register.  Volume 56, page 27356, June 13, 1991.

52.  Federal Register.  Volume 51, page 43823, December 4, 1986.
53.  Quality Assurance Handbook for Air Pollution Measurement
     Systems, Volume III, Stationary Source Specific Methods,
     EPA-600/4-77-027b.

54.  Federal Register.  Volume 57, page 31576, July 16, 1992.
55.  Federal Register.  Volume 57, page 44147, September 24,
     1992.

56.  Federal Register.  Volume 56, page 7042, February 21, 1991.
45.
46.
47.
48.
49.
                              2-147

-------
1. REPORT NO.
       &50/3-92-006b
                                   TECHNICAL REPORT DATA    .
                            (Please read Instructions on the reverse before completing)
                              2.
                                                            3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
   National Emission Standards for Hazardous Air
   Pollutants-Compliance Extensions  for Early Reductions
   Background  Information for Promulgated Standards
              REPOR
               December 1992
            6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
                                                            8. PERFORMING ORGANIZATION REPORT NO.
B, PERFORMING ORGANIZATION NAME AND ADDRESS
   Office of  Air Quality Planning  and Standards
   Environmental Protection Agency
   Research Triangle Park, North Carolina  27711
            11. CONTRACT/GRANT NO.

                68D10117
12. SPONSORING AGENCY NAME AND ADDRESS
   Director,  Office of Air Quality Planning & Standards
   Office  of  Air & Radiation
   U.S. Environmental Protection  Agency-
   Research Triangle Park, North  Carolina  27711	
                                                            13. TYPE OF
            14. SPONSORING AGENCY CODE
                EPA/200/04
15. SUPPLEMENTARY NOTES
 16. ABSTRACT                                                                 ,
   Regulations govening compliance extensions  for  early reductions of hazardous
   air pollutants are promulgated under the authority of section  (612)(i)(5)  of the
   Clean Air Act (CAA)..   These rules would apply to sources that  seek compliance
   extensions from standards promulgated under section (112)(d) of the  CAA.   This
   document contains summaries of public comments  received on the proposed rule
    (56 FR 27338, June 13,  1991), EPA responses and a summary of changes since proposal.
                                 KEY WORDS AND DOCUMENT ANALYSIS
                   DESCRIPTORS
                                               b.lDENTIFIERS/OPEN ENDED TERMS
    Air Pollution
    Hazardous Air Pollutants
    National emission standards for
      hazardous  air pollutants
    Early Reductions
                                                 Air Pollution Control
                                  13B
 18. DISTRIBUTION STATEMENT

    Unlimited
      Unclassified
20. SECURITY CLASS (This page)
      Unclassified
                           22. PRICE
 EPA Form 2220-1 (Rev. 4-77)   PREVIOUS EDITION is OBSOLETE

-------